Skip to main content Accessibility help

Of myths and metallurgy: Archaeological and ethnological approaches to upland iron production in 9th century CE northwest Laos

  • Olivier Évrard, Thomas O. Pryce, Guido Sprenger and Chanthaphilith Chiemsisouraj


Our recent discovery and excavation of a series of iron smelting furnaces, dated to the eighth and ninth century CE, near upland Rmet villages in northwest Laos, potentially sheds new light on the role of regional upland groups during the immediate pre-Tai period. The oral tradition associated with these furnaces emphasises the role of an ancient population of metallurgists who left the area under pressure from the Rmet. These stories could refer to the actual arrival and departure (immigration and emigration) of a population of metallurgists in that area sometime during the second half of the first millennium CE or they can support the scenario of a dissimilation process. The latter would explain the existence of a Rmet subculture that the locals regard as ‘Chueang Lavae’ villages, a differentiation that Karl G. Izikowitz had labelled ‘Upper Lamet’ in the 1930s. Our finds show that archaeology and ethnology can both contribute to a much-needed reformulation of upland Lao history.


Corresponding author

Correspondence in connection with this article should be addressed to: (Évrard); (Pryce); (Sprenger); (Chanthapilith)


Hide All

1 Julie Van Den Bergh, ‘Safeguarding the Plain of Jars: An overview’, in Nouvelles recherches sur le Laos, ed. Yves Goudineau and Michel Lorillard (Paris: EFEO, 2008), pp. 65–80.

2 This is clear in northern Lao and northern Tai oral and written traditions for the Khmu and the Lawa respectively. In the case of the Khmu, see for instance, the translation of Lao royal chronicles in Auguste Pavie, Recherches sur la littérature du Cambodge, du Laos et du Siam, 2 vols. (Paris: Ernest Leroux, 1898). Other Western or Southeast Asian scholars (especially Marxists) have thereafter, in the course of the twentieth century, popularised these views of the lowlanders on the history of their highland neighbours.

3 Angchalee Konggrut, ‘Bones tell stories of Thai origin’, Bangkok Post, 5 June 2006, pp. 1–2.

4 Edmund Leach, Political systems of highland Burma: A study of Kachin social structure (London: Athlone, 1970 [1954]).

5 Georges Condominas, From Lawa to Mon, from Saa’ to Thai: Historical and anthropological aspects of Southeast Asian social spaces, trans. Stephanie Anderson et al.; ed. Gehan Wijewardene (Canberra: Dept. of Anthropology, RSPAS, Australian National University, 1990).

6 James C. Scott, The art of not being governed: An anarchist history of upland Southeast Asia (New Haven: Yale University Press, 2009).

7 Stephen, Rostain, ‘Amazonie: Une archéologie en attente de décolonisation’, Les nouvelles de l'archéologie 126 (2011): 41–5.

8 Oliver Pryce, ‘Sedentarity and metallurgy in Southeast Asia’, in Mobility and heritage in northern Thailand and Laos: Past and present, ed. Olivier Évrard, Dominique Guillaud and Chayan Vaddhanaphuti, Proceedings of the Chiang Mai Conference, 1–2 Dec. 2011 (Chiang Mai: IRD, 2013), pp. 27–45.

9 The Rmet are better known as Lamet, but this is the Lao version of their ethnonym, hence we have chosen to use the villagers’ own usage: rmet (long vowel). For the sake of simplicity, our transcription of vernacular names, both Rmet and Lao, does not take into account vowel length, consonant type or glottal stops.

10 Hideyoshi Kavasima, ‘Census of bronze drums in Udomxay and Luang Namtha provinces’, (Vientiane: National University of Laos, 2008) [in Lao].

11 Pryce, Thomas O., Chiemsisouraj, Chanthaphilith, Zeitoun, Valéry and Forestier, Hubert, ‘An 8th–9th century AD iron smelting workshop near Saprim village, NW Lao PDR’, Historical Metallurgy 45, 2 (2011): 81–9.

12 Personal communication, Dr Jean-Luc Schwenninger, Oxford University.

13 Radiocarbon dates the age of death of the fuel, which could have been reused/stored for years, whereas thermoluminescence dates the last heating of the furnace.

14 All the Rmet (as well as Khmu) villages of this region are located on hilltops and surrounded by a screen of old forest where one finds the springs used for daily water consumption and, usually in the southwest, the graveyards.

15 That is indeed what the Rmet themselves say, but they usually refer to a not-so-distant past.

16 Dufumier, Marc, ‘Slash-and-burn, intensification of rice production, migratory movements, and pioneer front agriculture in Southeast Asia’, Moussons 9–10 (2006): 14.

17 Five to six persons are generally considered necessary to operate a bloomery furnace, plus people to mine and transport the ore, if not done by the smelters, as well as those needed to forge the iron produced. We should add here the number of people required to cut and transport the wood necessary to prepare the substantial quantities of charcoal needed. See further Peter Crew and Susan Crew, eds., Early ironworking in Europe: Archaeology and experiment; Abstracts, International Conference, Plas Tan y Bwleh, 19–25 Sept. 1997 (Maentwrog: Plas Tan y Bwleh Occasional Paper 3, 1997); and The world of iron, ed. Jane Humphris and Thilo Rehren (London: Archetype, 2013).

18 Karl Gustav Izikowitz, Lamet: Hill peasants of French Indochina (New York: AMS Press, 1979 [1951]), p. 79.

19 See Barbara Wall, Les Nya Hön: Étude ethnographique d'une population du plateau des Boloven (Vientiane: Vitthagna, 1975), p. 18.

20 This area overlaps the intersection of two provinces (Luang Namtha and Bokeo) and three districts (Nalae, Viengphukha and Pha Udom). It is also the eastern limit of the Rmet presence in Laos.

21 Lundström, Håkan and Tayanin, Damrong, ‘Kammu gongs and drums (I): The kettlegong, gongs and cymbals’, Asian Folklore Studies 40, 1 (1981): 6586.

22 While the flood myth is a common tale of the origin of ethnic diversity shared in variations with numerous upland groups, the money tree story tells about the origin of the wealth difference between Rmet and lowlanders. See Van, Dang Nghiem, ‘The flood myth and the origin of ethnic groups in Southeast Asia’, Journal of American Folklore 106, 421 (1993): 304–37; Guido Sprenger, ‘Political periphery, cosmological center: The reproduction of Rmeet socio-cosmic order and the Laos–Thailand border’, in Centering the margin: Agency and narrative in Southeast Asian borderlands, ed. Alexander Horstmann and Reed Wadley (New York/Oxford: Berghahn, 2006), pp. 65–92; Guido Sprenger, Die Männer, die den Geldbaum fällten: Konzepte von Austausch und Gesellschaft bei den Rmeet von Takheung, Laos (Berlin: LitVerlag, 2006).

23 Izikowitz, Lamet, p. 24.

24 The term could possibly refer to the Mon. In Chinese, the word man refers generally to ‘Southern barbarians’, with no distinctions.

25 Currently: Takhong, Tako, Kayae, Talouy, Samin, Saprim, Takhueng, Lahang and Kanueng villages. Informants sometimes add villages that used to be located in the same area but have now been partly or entirely resettled: Katoy, Takchak, Muksuk, Mokkala and Chomsy.

26 Rmet genealogies rarely extend more than three generations beyond the oldest living generation.

27 While the Katuep stream actually belongs to the Nam Tha basin, its source is separated from the source of the Nam Ngao only by a small ridge.

28 Two informants, a Rmet man in Ban Takrong and a Hmong informant who used to live in that area, also referred to a gendered opposition between ‘male’ and ‘female’ objects.

29 Nowadays, among the Khmu, during rituals for the spirits of a house, or during funerals, the blood of a buffalo is poured over a bronze drum.

30 Lundström and Tayanin, ‘Kammu gongs and drums (I)’: 65–8.

31 Kenny, E.C., ‘The Chinese gongs’, Man 27 (1927): 167; Harry Ignatius Marshall, The Karen people of Burma: A study in anthropology and ethnology (New York: AMS Press, 1980 [1922]): 118–19.

32 Marshall, ibid., p. 116. Among the Khmu, there are also mentions of ‘female drums’, but not of ‘male’ ones while ‘hot’ and ‘cold’ categories refer to playing styles.

33 Frank Proschan, ‘Chueang in Kmhmu folklore, history and memory’, in Tamnan keokap thao hung thao chuang: Miti thang prawattisat lae wattanatham [Proceedings of the First International Conference on the Literary, Historical, and Cultural Aspects of Thao Hung Thao Cheuang], ed. Sumitr Pitiphat (Bangkok: Thammasat University, Thai Khadi Research Institute, 1998), pp. 174–209.

34 The name finds its origin in rua, which is used as a collective pronoun (‘us’) in Lawa language (Michel Ferlus, personal communication). The Tai pronounce the ethnonym lua but the Lawa in Western Thailand call themselves lwa or lavuea (Ratanakul, Suriya, ‘The Lawa Lasom Lae poetry’, Journal of Siam Society, 73 [1981]: 183204). They should not be confused with the Lua of Nan province, who are also known as Htin or Mal in Thailand and as Pray in Laos.

35 Kerr, A.F.G., ‘Ethnologic notes: The Lawa of the Baw Luang Plateau’, Journal of Siam Society 18, 2 (1924): 135–46.

36 Holt S. Hallet, A thousand miles on an elephant in the Shan states (Bangkok: White Lotus, 2000 [1896]), p. 55; emphasis added.

37 Hutchinson, E.W., ‘The Lawa in northern Siam’, Journal of the Siam Society 27, 2 (1935): 153–83.

38 Seidenfaden, Erik, ‘The Lawa of Umphai and Middle Mae Ping’, Journal of Siam Society 32, 1 (1940): 2936.

39 Volker Grabowsky and Andrew Turton, The gold and silver road of trade and friendship: The McLeod and Richardson diplomatic missions to Tai States in 1837 (Chiang Mai: Silkworm, 2003), p. 232.

40 Ibid.

41 Hallet, A thousand miles on elephant, p. 143; emphasis added.

42 The linguistic differences between the two populations were confirmed in 1935 by Hutchinson in ‘The Lawa in northern Siam’, p. 73.

43 Fiskejö, Magnus, ‘Mining, history and the anti-state Wa: The politics of autonomy between Burma and China’, Journal of Global History 5, 2 (2000): 241–64.

44 Foon Ming Liew-Herres, Volker Grabowsky and Renoo Wichasin, Chronicle of Sipsong Panna: History and society of a Tai Lü kingdom, twelfth to twentieth century (Chiang Mai: Mekong Press, 2012), p. 17.

45 Évrard, Olivier and Chiemsisouraj, Chanthaphilith, ‘Les ruines, les sauvages et la princesse: Patrimoine et oralité à Vieng Phou Kha, Laos’, Aséanie 28 (2011): 67100.

46 Bennet Bronson, ‘Patterns in the early Southeast Asian metals trade’, in Early metallurgy, trade and urban centres in Thailand and Southeast Asia, ed. Ian Glover, Pronchai Suchitta and John Villiers (Bangkok: White Lotus, 1992), p. 75.

47 The Rmet say that their ancestors could never access the Chueang Lavae's mine for fear of the spirits guarding its entrance. It is even said that the mere sight of the mine kills. Observations done in the Bo Luang region in northern Thailand show that the ore was the exclusive property of a group of Lawa villages that owned the site in common (Hutchinson, ‘The Lawa’, p. 164) and that there were no settlements nearby.

48 Donald K. Swearer and Sommai Premchit, The legend of Queen Cama (Albany: State University of New York Press, 1998), p. 18; emphasis added.

49 Condominas, From Lawa to Mon, p. 17.

50 Gérard Diffloth, personal communication.

51 Pryce et al., ‘An 8th–9th century AD iron smelting workshop’, p. 86.

52 Leach, Political systems, p. 251: ‘For the Shan, silversmithing is a profession proper and peculiar to the nobility; iron-working is a task for slaves’, and a few lines later: ‘Blacksmithing is not respectable for aristocratic gumsa Kachins because it is not respectable for Shan aristocrats either’.

53 Louis Harmand, Voyage au Cambodge (Paris: Société de Géographie, 1876), p. 6.

54 During his 2007–2008 survey in Udomxay and Luang Namtha with the provincial authorities and the Lao National University of Vientiane, the Japanese archaeologist Hidoyeshi Kavashima found and registered 37 bronze drums in Khmu and Rmet villages, all of them of so-called Heger III type. Out of the 37 bronze drums found, 23 had been inherited, 13 were from neighbouring villages, and one was in Chiang Mai. See Kavasima, ‘Census of bronze drums’.

55 Yunnan appears historically as an important centre for bronze drum production, with Karen myths indicating they were received first from so-called K'wa and S'wa peoples, which may refer to the Wa. They were also produced later on by Shan and evolved in their design to suit the taste of the Karen. New Dang, a Shan settlement in a Karen-ni region near Loikaw is their last known place of production. Bronze drums were cast there during the 19th century by Shan craftsmen and many tribal groups from Thailand and Laos, including Rmet and Khmu, came to buy them until production declined after the 1880s due to repeated periods of conflict. See Richard M. Cooler, The Karen bronze drums of Burma: Types, iconography, manufacture, and use (Leiden: Brill, 1995), pp. 51–5.

56 Richard M. Cooler, ‘The use of Karen bronze drums in the royal courts and Buddhist temples of Burma and Thailand: A continuing Mon tradition?’, in Papers from a Conference on Thai Studies in honor of William J. Gedney, ed. Robert J. Bickner, Thomas J. Hudak and Patcharin Peuasantiwong (Ann Arbor: Michigan Papers on Southeast Asia 25, Center for South and Southeast Asian Studies, University of Michigan, 1995), p. 109.

57 Karlsson, Klemens, ‘The Songkran festival in Chiang Tung: A symbolic performance of domination and subordination between Lowland Tai and Hill Tai’, Tai Culture 23 (2013): 5062.

58 Cooler, The Karen bronze drums of Burma, p. 11.

59 Guido Sprenger, ‘From kettledrums to coins: Social transformation and the flow of valuables in northern Laos', in Social dynamics in the highlands of Southeast Asia: Reconsidering political systems of highland Burma by E.R. Leach, ed. François Robinne and Mandy Sadan (Leiden: Brill, 2007), pp. 160–85. For the Khmu, see Lundström and Tayanin, ‘Kammu gongs and drums (I)’.

60 Svantesson, Jan-Olof, ‘U’, Linguistics of the Tibeto-Burman Area 11, 1 (1988): 114. The word for copper/bronze, léad, is also found among Khmu (lad).

61 Aymonier, Etienne, ‘Notes sur le Laos’, Excursions et reconnaissances 9, 22 (1885): 255347; Louis Harmand, ‘Voyage au Cambodge’, p. 5.

62 Pottery could have been another specialisation in some Lawa villages. E.G. Kauffman mentions three Lawa villages known as Changmo Manod, Changmo Noi (small Changmo) and Changmo Luang (main Changmo) — changmo in northern Thai meaning potter. See Kauffman, E.G., ‘Some social and religious institutions from the Lawa (N.W. Thailand). Part I’, Journal of the Siam Society 60, 1 (1972): 249.

63 Ibid.: 244.

64 Sprenger, Guido, ‘From power to value: Ranked titles in an egalitarian society, Laos’, Journal of Asian Studies 69, 2 (2010): 412.

65 Izikowitz, The Lamet, pp. 106–8.

66 The village is not precisely located in his monograph, but more details are given in Karl Gustav Izikowitz, Over the misty mountain: A journey from Tonkin to the Lamet in Laos, trans. Helena Berngrim (Bangkok: White Lotus, 2004 [1994]), p. 103. The Swedish ethnographer indicates that the village is ‘far off Southeast of Tafa’ and that it takes more than one day's walk from there to reach the village. It could therefore have been relatively close to the area under study here. West of Ban Tako, there is currently a village named Ban Satun, but none of us visited it.

67 Kanda Keosphha, ‘Standing stones in northern Lao PDR’, in Uncovering Southeast Asia's past: Selected papers from the 10th International conference of the European Association of Southeast Asian archaeologists, The British Museum, London, 14–17th Sept. 2004, ed. Elisabeth A. Bacus, Ian C. Glover and Vincent C. Pigott (Singapore: NUS Press, 2006), pp. 148–53.

68 Taluy villagers say that there are even more standing stones on the old site of Satun village, a few hours' walk northwest of Taluy.

69 See Guido Sprenger, ‘Do the Rmeet have clans?’, in Recherches nouvelles sur le Laos, ed. Michel Lorrillard and Yves Goudineau (Paris: Ècole française de l'Extrême-Orient, 2008), pp. 568–9, for a discussion of these categories.

70 The duration of these cycles can vary, but they are always in a combination of an odd number of years between one and nine; typically it was a combination of three and nine year cycles, but sometimes also one and three, one and five, three and seven. Similar religious practices related to a so-called mbled ritual are found among the Khmu Khuaen, in Viang Phu Kha.

71 This character seems to be more related to Khmu oral tradition and is found mostly in and around Viang Phu Kha and Nalae. See further Évrard and Chiemsisouraj, ‘Les ruines, les ‘sauvages’ et la princesse’, pp. 80–87, and a shorter English version by the same authors, ‘The ruins, the “savages” and the princess: Myths, migrations and belonging in Viang Phu Kha, Laos’, in Évrard et al., Mobility and heritage in northern Thailand and Laos, pp. 55–73.

72 Kauffman, Hans E., ‘Stone memorials of the Lawa’, Journal of the Siam Society 59, 1 (1971): 129–51.

73 In the graveyard at old Taluy Noy (small Talouy) village, we found a stone with at least two layers of inscriptions. The most recent one was dated from 1997: it comprised the name of a man, his date of birth (1910), his age at the moment of his death (at 87), an encircled tree-headed elephant, and a caption (‘period of the French colonisation’) written in Lao script. This settlement moved near the track ten years ago and in 2009, the villagers were willing to move the stone of the spirit of the border closer to their new location.

74 The details of the competitions go as follows:

These big men first competed over the quality of their rice jar alcohol. Chaia Cham Choy won because instead of having water poured in the jar, he had pure alcohol, so people continued to drink from his jar for seven days and seven nights while the other jars gradually lost their alcoholic content. Then, the chiefs competed over the number of their followers. Once again, Chaia Cham Choy cheated. He had his followers entering and exiting the collective house repeatedly so that their number appeared much higher than those of the other chiefs. Pae Meng tried to do the same thing in his village of Takhueng but the other chiefs noticed his trickery. He lost face and committed suicide. Finally, the chiefs decided to mark their respective territory. They agreed that each of them would leave at cockcrow and walk toward each other's village. The place where they would meet would be the border of their territory. Chaia Cham Choy left at bat crow [e.g. much sooner than the others] and that's why today Saprim territory is still much bigger than the other neighbouring Rmet villages.

75 Izikowitz, The Lamet, p. 116.

76 In the Lawa case, the relation is even more explicit. Several of the megaliths studied by Kauffman in two locations had clearly been erected for persons involved in the iron economy. In the first spot, the names were those of village founders who had moved from Bo Luang to settle nearer to the ore mine. In the second spot, the inscription refers to the ‘Lord of the Mine to the Right’ and ‘Lord of the Mine to the Left’ (Kauffman, ‘Stone memorials of the Lawa’, pp. 145–6).

77 The Rmet borrowed these titles from the Tai lowland societies, but they then adjusted them to their own categories and perceptions of hierarchy, related mainly to the distinction between wife-givers and wife-takers. Such a process of the ‘indigenisation’ of a cultural borrowing led after several generations to important variations between villages and informants: titles known in one location may be unknown in another or their order can be different. See Sprenger, Guido, ‘From power to value: Ranked titles in an egalitarian society, Laos’, Journal of Asian Studies 69, 2 (2010): 413–14. Tao Set's tale is a good example of this absence of a unified system of ranks: some titles do match the list given by Sprenger, but are given different values (saen, pae, chaia); some are combined together in a single word (paecha); finally, other titles given here (khun, long) do not appear in Sprenger's list.

78 Olivier Évrard, ‘Oral histories of livelihoods and migration under socialism and post socialism in northern Laos', in Moving mountains: Ethnicity and livelihoods in highland China, Vietnam, and Laos, ed. Jean Michaud and Tim Forsyth (Vancouver: University of British Columbia Press, 2011), p. 94.

79 Leach, Political systems, p. 251: ‘(…) it was the trade in iron more than everything else which gave the early Jinghpaw power and which enabled the gumsa to become feudal satellites of the Shan rather than their serfs'.

80 Scott, The art of not being governed.

The authors would like to thank the ANR-funded programme ‘Sedentism in the Mekong region: Identities, techniques, territories and environment’ (2008–2011), led by Dr Dominique Guillaud (IRD, UMR 208 Patrimoines Locaux et Gouvernance), during which the first and third authors initiated the study of Rmet archaeology and mythology, and the second and fourth authors conducted an excavation near Ban Saprim in 2010; the IRD, which funded part of the fieldwork and excavations done by the lead and second author in 2012 and 2013 with senior postdoctoral fellowships; LASS, which coordinated the administrative aspects of the study and the organisation of fieldwork; Dr Viengkeo Souksavatdy, Head of the Dept. of Archaeology at the Ministry of Culture, Information and Tourism (MCCT), who gave the authorisation for excavation; Mr Mien Lovankha and Mr Phomphone Chanthala, of the Bureau of Culture, Information and Tourism in Luang Namtha province and in Nalae district, who greatly facilitated access to the field and collaboration with the villagers. We would also like to thank the Lao participants in the surveys and excavations undertaken in the course of the last five years, including: Mr Thammalay Chanthamongkhon (LASS) and Mr Sengphet Norkhamsomphu (MCCT, Vientiane); the villagers of Ban Takhueng, Ban Saprim, Ban Taluy, Ban Tako, Ban Chomsy, Ban Raye and Ban Takrong for their warm welcome, their help during the surveys and excavations, and their willingness to share their mythology and knowledge with us. Unless otherwise indicated, all illustrations in this article were prepared by the authors.

Of myths and metallurgy: Archaeological and ethnological approaches to upland iron production in 9th century CE northwest Laos

  • Olivier Évrard, Thomas O. Pryce, Guido Sprenger and Chanthaphilith Chiemsisouraj


Altmetric attention score

Full text views

Total number of HTML views: 0
Total number of PDF views: 0 *
Loading metrics...

Abstract views

Total abstract views: 0 *
Loading metrics...

* Views captured on Cambridge Core between <date>. This data will be updated every 24 hours.

Usage data cannot currently be displayed