Journal of Archaeological Research, VoL 4, No. 1, 1996
A Review of Archaeology in Mainland Southeast Asia C. F. W. Higham 1
This article outlines recent advances in establishing and understanding the prehistoric sequence in mainland Southeast Asia. Research has been unevenly distributed, varying from virtually none in Cambodia to a marked intensity in Hong Kong and the adjoining mainland. A new pattern is becoming apparent, due in no small part to the new findings in southern China. It is argued that despite its long history of occupation, beginning with H o m o erectus almost a million years ago, tropical Southeast Asia was occupied by scattered inland groups and larger, possibly sedentary coastal foragers until exposed to intrusive agricultural societies during the third millennium BC. These communities, which originated ultimately in the Yangzi Valley, brought with them rice cultivation and were responsible for the wide distribution of Austroasiatic languages. The three divisions of this language superfamily are now distributed from eastern India to southern China. Following a relatively brief Neolithic period, small and autonomous communities, particularly in Lingnan and Bac Bo (Guangdong and Guangxi provinces of China and the area of the Red River Delta), were exposed to exchange contact with the Shang state of the Chinese Central Plains, and this brought exotic jades and bronzes. Within this context, a local bronze industry was established between 1500 and 1000 BC over much of Southeast Asia, but without any obvious social developments until the middle of the first millennium BC, when several major changes occurred. These incorporated iron smelting and forging and exposure in the southern parts of the region to Indian mercantile contact and along the northern margins to the expanding Chinese empire. Adaptations varied regionally, from the establishment of warrior chiefdoms to counter the Chinese to the construction of water control systems in the arid heart of Southeast Asia to alleviate environmental unpredictability. It is within these regional changes
1Department of Anthropology, University of Otago, Dunedin, New Zealand. 3 1059-0161/96/0300-(D03509.50/0 © 1996 Plenum Publishing Corporation
4
Higham
that we can identify trends toward the establishment of states, some of which persisted in an unbroken lineage to the present.
KEYWORDS:Southeast Asia; prehistory; BronzeAge; Austroasiaticlanguages; socialchange; state formation; rice cultivation.
INTRODUCTION Southeast Asia lies uneasily between China and India, in part colonized by the former and its early civilizations influenced by the latter. But it has a distinct character of its own, seen linguistically in the Austroasiatic and Austro-Tai languages, economically by wet rice cultivation and environmentally by the monsoon. The region broadly defined sustains a sixth of humanity, a weight of numbers that demands careful consideration by the prehistorian. The cultivation of rice turns on the dominant force in the environment, the monsoon. While the climate varies from region to region, the May wind change, which originates in the Indian Ocean, introduces hot, moist conditions in place of the cooler weather seen from December until April, the dry season. A group of the world's great rivers, from the Brahmaputra to the Yangzi, radiates like the spokes of a wheel from the hub in the Eastern Himalayas, and it is along their middle and lower reaches, and up their many tributaries, that a huge area has been converted by human endeavor into rice fields (Fig. 1). In the most widespread technique, low bunds are raised to retain rainwater that percolates through the fields, for rice is a marsh plant that absorbs energy from water. Progressively more complex systems rely on the construction of dams and reservoirs to provide water when the monsoon falters. One of the most complex of these underpinned the agricultural system of Angkor from the ninth century AD (Groslier, 1979). To understand the personality of Southeast Asia, one must appreciate the pattern of the river systems, arteries of communication that sustain dense human settlement and provide the conditions for rice agriculture. In this context, the boundaries of Southeast Asia are flexible. If we follow climatic variables, it could stretch from eastern India to the Yangzi Delta. A mixture of subsistence and linguistic evidence would see a rather smaller area, but still stretching from the Xijiang to at least the Salween. Archaeological research in this considerable area has been influenced by political constraints, for Southeast Asia has been only partially colonized The dominant early force was the French Union of Indo China, which involved the lower Red and Mekong valleys and the coast of Vietnam. Thailand and the southern provinces of China, a region known as Lingnan,
The Archaeology of Mainland Southeast Asia
Shandong 35*-
30 ° .
)
25 °
PLATEAU
LINGNAN GUANGXI
Delta
GUANGDONG
20 °
10° 0
I
95 °
I
1
100 °
Land above 180 m.
105* ~
1000 k m
I
110*
I
115~
I
120*
Land above 2 4 7 0 m.
Fig. 1. Southeast Asia, showing the principal rivers and sites mentioned in the text.
escaped Western dominion, although the latter area was incorporated into the Han Empire late in the first millennium BC. The French colonial regime saw the introduction of archaeological research, particularly following the foundation of the ]~cole Franqaise d'Extr~me Orient in 1898. In Lingnan, however, archaeology came later, and was concentrated in Hong Kong and the adjacent coast of Guangdong. Thailand has seen consistent pre-
Higham
6
historic research only since the early 1960s, but liberal regulations have allowed both Western scholars and Thai professionals to obtain a rich and growing body of data. Archaeological research began early in Cambodia but has now virtually ceased, while in Vietnam, fieldwork has been pursued with vigor, and the results of excavations are promptly published. Synthesizing available information, however, is exacerbated by many languages and contrasting theoretical approaches. This is reflected in the varied and often conflicting systems of nomenclature. Shell middens located on raised beaches in Vietnam and southern China have been labeled "coastal Neolithic," a name that recognizes the presence of pottery and polished stone implements but that lies uncomfortably with the absence of evidence for agriculture. Bayard (1984) suggested the terms General Periods A-D to describe the sequence from the first agricultural villages to the formation of states in Northeast Thailand, a system that I have applied more widely (Higham, 1989). The Vietnamese use the term culture to describe like sites in the same region, but the general lack of an agreed-upon order is seen in the terms used for the late Prehistoric period, which include General Period C (Bayard, 1984), Mode 2 (Higham, 1983), the Muang period (Bayard, 1992), the High Bronze Age (Hutterer, 1991), the Late Bronze Age (von Dewall, 1979), the Iron Age (Charoenwongsa, 1988; Penny, 1984), the High Metal Age (Ho, 1992), the Late Metal Age (Bronson, 1992), the late Prehistoric (Glover, 1991), and the Formative (Welch, 1985). I employ the three-age system below, without implying that it follows similar patterns in any other region. Developing an agreed-upon structure of Southeast Asian prehistory has not been straightforward. Where there are few excavated sites, the interpretation of one can have a major impact on the whole. Over the past two decades, controversy has attended the chronology of certain key sites, and the absence of final site reports has made evaluating data difficult or impossible. Interpretations have lurched between extremes. In the early sixties, Southeast Asia was seen as a prehistoric backwater (Clark, 1961), but the next decade saw claims for the earliest agriculture and bronze- and ironworking in the world (Bayard, 1972; Gorman and Charoenwongsa, 1976; Solheim, 1968). Many colleagues still mention Spirit Cave, Non Nok Tha, and Ban Chiang as if they are sites of outstanding significance. But much research has intervened, and a more judicious review is both timely and desirable.
THE STRUCTURE OF THE PREHISTORIC SEQUENCE
Before proceeding to a review of recent developments, it is necessary to outline the main chronological and cultural variables. Following few and
The Archaeology of Mainland Southeast Asia
patchy indications of the presence of Homo erectus during the middle Pleistocene, we find a series of sites dating from about 40,000 years ago that indicates occupation of the area by anatomically modem humans, although sites are few and data thin. The Holocene saw two principal forms of adaptation by hunter-gatherers, inland and coastal. Southeast Asia was greatly affected by the Holocene rise in sea level, when much land was lost to the sea (Fig. 2). As a consequence, we gain an appreciation of the wealth of the coastal adaptation when the sea level stabilized at a level higher than that at present, then fell back, leaving stranded prehistoric sites. These coastal communities provide evidence for sedentary settlements that were able to tap the rich, self-renewing food resources characteristic of the tropical shore. There are extensive inland river floodplains, but no evidence to suggest occupation by hunter-gatherer communities. This might reflect extensive recent modification to create rice fields, but equally, such areas might have been avoided for their low-lying, swampy terrain, dense ground cover that flourished under the deciduous forest and extensive seasonal flooding. Attempts to isolate and explain the transition to agriculture have been given impetus by research in the Yangzi Valley, where a long Neolithic sequence is unfolding. By combining new archaeological evidence with that of linguistics and human biology, it has proved possible to suggest a major expansionary movement of agriculturalists, which probably originated in the Yangzi Valley. In its widest context, it ultimately saw expansion into Southeast Asia and eastern India and the settlement of the islands of Southeast Asia and the Pacific and Indian Oceans by those speaking languages in the Austric phylum (Betlwood, 1985, 1989, 1992; Blust, 1976, 1985, 1993a; Higham and Thosarat, 1994; Reid, 1993; Spriggs, 1989). There is an extensive literature on the dating of the Bronze Age. A critical scrutiny of available radiocarbon determinations has failed to find convincing evidence for copper smelting or alloying before about 1500 BC. From that juncture there is much evidence for a rapid spread of bronze casting technology along the coastal and river networks. Bronze and copper objects joined exotic stone, shell, and ceramics in mortuary contexts, though metal was never abundant before the mid-first millennium BC. By that period, iron was being smelted and forged in the Chao Phraya and middle to lower Mekong valleys, an industry that probably developed in the milieu of copper smelting, where hematite was used as a flux. The available evidence indicates that iron came rather later in the Red River Valley and the Yunnan Plateau sites, along with the importation of goods from the Chu state and, later, the expanding Qin and Han empire. In Yunnan and Lingnan, there was a predilection for casting bronze hilts onto iron sword and spear blades. This period also saw the development of ex-
Higham
i!ii!iiiiiiii!iiiiii!!iiiiiiii!i!!ii!iiiiiiii~;iiiii~:i~'~iiii~i!!~:i~i;~ iiiii!!!!ii;i!i~i!!i?!i!;i~i!!!iiii!iiii!ii~ii!!!~i~" )~;';"'~'~iii!~.~
iiii!i!i!!!i!i~!ii!iii!!iiii!!iiii!!!~!!!!!ii~!ili.. :i:::;!!i!!~!!:i~!!iiii!!i!iii!iiiiiiii!iiiii!ii
~ !).,~7~j
iiiiiiiiiiiiii!ii!iiiiiiiiiiii!iiiiii!!iiiiiiiiiiiiii~
"
s~ o Japan
~ i ~ ~ i
iiiiii!iiiiiiiiiii!iliiiiiiiiili!i!iiiiiiiiiiii!!i!iiiii~
:!iiiiiii!!!!ii!ii!~!~i{
(Z~!!!: A ::::~
~,
~!
~ i~:ii~!!!:i::::::!i3i~
, .:
,~c,~:L:~; ~'
~
~;~
D
o
Arafura shelf
Sahul shelf ~.t.~ ~i!i!!!!ii!!i;, (.! // ~, :!i i ~'Tr~!! !i~i!!i ii ~ . /!! ::::::::::::::::::::::::: i!?~.~:~i ,I / ,.~r.~...:::':: ::::::::::': ::: "'~ Rowley shelf j~::::: :::::::::::::171:2:::::::: ............
2000km
0 i
................
,,,
Land a b o v e 200 metres
Fig. 2. S o u t h e a s t Asia, showing the likely location of the early Pleistocene shore.
change relationships with India, which are most clearly evident in the Chao Phraya and lower Mekong valleys. This exchange of exotic goods and ideas, together with the foundations of an iron industry and increasing expertise
The Archaeology of Mainland Southeast Asia
in bronze casting, occurred within the context of developing social hierarchies. Several sources indicate that most languages spoken from at least 300 AD were Austroasiatic. Early inscriptions in the Mekong Valley sites were written in Sanskrit and included passages in archaic Khmer. In the Chao Phraya Valley, we encounter the Mort language, and in Bac Bo, early Vietnamese.
INITIAL HUMAN SETTLEMENT Southeast Asia is well known for the discovery of Homo erectus, now dated as early as 1.3 million years ago in the Sangiran formation on Java. Although virtually all the skeletal remains have been found on that island, some evidence suggests very early human settlement on the mainland. Maw (1993) has reported a fragment of human maxilla from Nwe Gwe in Myanrnar (Burma), which he has ascribed to a later Pleistocene context, in the vicinity of 500,000 years ago. At Ban Mae Tha and Ban Don Mun, Pope has reported pebble tools below a basalt flow dated between 600,000 and 800,000 years ago on the basis of palaeomagnetism (Pope, 1985; Pope et aI., 1987). In Vietnam, teeth and jaw fragments ascribed to Homo erectus, and dated to about 250,000 years ago, have been recovered from Tham Khuyen (Ciochon and Olsen, 1987), and flaked stone tools have been found at Nui Do (Shutler, 1995). As Bellwood (1992) and Bowdler (1993) have pointed out, there is no undoubted association between cultural material and Pleistocene human remains in either mainland or island Southeast Asia, largely because the fossil bones are usually found in redeposited contexts. It is, however, now possible to recognize a series of later Pleistocene sites across mainland and island Southeast Asia and into greater Australia that display related stone tool industries. These, Bowdler has suggested, probably reflect the expansion into the area of anatomically modern humans who possessed a maritime technology adapted for crossing substantial bodies of water. In mainland Southeast Asia, documentation of this broad horizon is best seen at Lang Rongrien, a rock shelter in peninsular Thailand, where Anderson (1990) has uncovered occupation contexts dating back to at least 37,000 BE To these must be added the growing body of data for the late Pleistocene Son Vi tradition of the Red River Valley (Bac Bo). The occupation of these sites occurred at a time of major environmental change, seen in Southeast Asia not so much by temperature fluctuations, although these occurred, but by the changing sea level. Nowhere else was such an extensive area inundated. Given the maritime technology in place by 40,000
10
Higham
years ago, which surely owes much to the properties of bamboo, there must have been a long tradition of coastal adaptation now lost to the Holocene rise in sea level (Fig. 2). Former coastlines as the sea rose above its present level and then fell back, however, are marked by prehistoric settlements (Bui Vinh, t991; Ha Van Tan, 1980; Higham and Thosarat, 1994). Although often labeled "coastal Neolithic," no convincing evidence has yet been offered for agriculture or stock raising. Given the lesson of the Jomon culture, where evidence for agriculture is tenuous and late, there is no reason to expect that food production in such rich coastal habitats should accompany a ceramic industry, the polishing of stone tools, or the development of complex social forms. In contrast to these coastal sites, we find that inland settlement was confined to, or has survived best, in inland rock shelters. Broadly defined as Hoabinhian, these sites evidence small groups of mobile foragers adept at exploiting the more scattered food resources of the canopied forest. Spirit Cave is one of many such sites (Gorman, 1971, 1972). In a recent approach centered on the lower Khwae Noi Valley of western Thailand, Shoocondej (1994) has considered in detail the seasonal variations in resource availability and outlined likely cultural adaptations. These involve the possibility of small group exploitation of the wide diversity of wet season resources and concentration in bigger social groups during the dry season, when resources could be located with greater predictability. Results from her testing of this hypothesis in the field will be most interesting, for hitherto, our knowledge of the Hoabinhian has turned on the occupation in rock shelters, surely a highly biased sample of sites.
THE TRANSITION TO AGRICULTURE The transition to cultivation has long been identified as a key issue in Southeast Asian prehistory. At present, there are two principal regions where this may have occurred: the Yangzi Valley and the coastal tropics. The two are not incompatible, but both must be considered in the context of the relationships between archaeological and linguistic evidence proposed in Renfrew's (1987) synthesis to account for the distribution of IndoEuropean languages. He suggested that the expansion of agriculturalists from Anatolia saw the gradual establishment and local differentiation of languages in the Indo-European phylum. Such an expansion has been advanced to account for the distribution of languages in the Austric phylum (Bellwood, 1989, 1992; Blust, 1985, 1993; Higham and Thosarat, 1994). Austric incorporates Austroasiatic (AA) and Austronesian (AN) languages. The former are distributed from eastern India to coastal Vietnam
The Archaeology of Mainland Southeast Asia
11
and from Yunnan to peninsular Malaysia and the Nicobar Islands (Fig. 3). The best-known are Mon, Khmer, and Vietnamese, but there are well over a hundred others. AN languages are found from Taiwan through much of island Southeast Asia and then in a broad arc stretching from Malagasy to Easter Island. The third major grouping has been named Austro-Tai (AT). Benedict (1942, 1975) has suggested that AT and AN languages are related, but this is not universally accepted. They are found across southern China and extend into large tracts of Southeast Asia including Thailand and Laos (Fig. 4). Blust (1993) has placed the homeland of the Austric phylum in the upper Yangzi Valley. After the establishment of rice cultivation in settled village communities there, he has suggested an expansionary series of movements took place. These saw AA-speaking agriculturalists infiltrating down the river system into Southeast Asia and eastern India. There is a primary division between Mundaic languages of India and those to the east. So the ancestors of the Munda moved down the Brahmaputra River. The ancestors of Mon and Khmer speakers would have expanded via the Mek-
~ Munda Mon-Khmer #
N
South China Sea
0
1000km
Fig. 3. The distribution of Austroasiatic languages.
Higham
12
~]UD:c°rig-Mien]
@ 'llh ,.
~tli ~lh
• ,IIWI
#
N
Bay of Bengal
South China Sea
0
lO00km
Fig. 4. The distribution of Hmong-Mien and Daic (Austro Tai) languages.
ong system, while speakers of Vietic languages would have moved down the Red River (Fig. 5). How does the archaeological evidence relate to this hypothesis? This is not the place to consider in detail the timing and context of the origins of rice and millet cultivation in China. The broad pattern, however, presents intriguing similarities with the contemporaneous changes noted by Bar-Yosef (1995) for the Levant, where he traced an increasing reliance on cereal crops in conjunction with climatic change. In Southeast and East Asia, climatic fluctuations had an increasingly significant effect as one moves northward. The climate in central and northern China, identified on the basis of pollen cores, glacial activity, loess deposition, and fluctuating take levels, shows that from about 13,150 to 12,400 BP, the temperature rose by about 7°C. From 11,000 to 10,000 BP, there was a sharp deterioration, known as the Younger Dryas, which involved increasing cold (Zhou et al., 1991). This in turn was followed by a marked warming from 10,000 to 8500 BP, then a succession of fluctuations between warm and cold conditions. During the warm phase dating from 10,000 BP, a number
13
The Archaeology of Mainland Southeast Asia
4 0 * --
3 5 * -[1
+ill
30° -
+
25* -
4"+
20 ° -
°
l
/
I
I
I
I
I
95 °
100*
105*
110"
115 °
L a n d a b o v e 180" m .
o-9000-6500BP
~
•-5500BP
~
I 120*
Land above 2470 m,
o-4500BP
o-3500BP +-1500BP
A. Proto Munda, B. Proto Mon, C. Proto Khmer, D. Proto Viet
Fig. 5. Possible routes for the introduction of Austroasiastic languages into Southeast Asia.
of open and cave sites in southern China were occupied. Xianrendong, for example, represents a riverine site from which aquatic resources were exploited (Fig. 6). The occupants also made cord-marked pottery, although
14
Higham
~*
100"
+i++i++i++!i+i+:+ a vo 1 , 0 .
105 +
lib
110"
+ov.
115"
120"
4,o.
Fig. 6. The location of the sites mentioned in the text. Where sites are located close to each other, they are given the Same location number. (1) Spirit Cave; (2) Ban Na Di, Ban Chiang; (3) Nwe Gwe; (4) Ban Mae Tha, Ban Don Mnn; (5) Tham Nguyen; (6) Nui Do; (7) Lang Rongrien; (8) Xianrendong; (9) Liyuzui; (10)Zengpiyan; (11) Pengtoushan; (12) Hemudu; (13) Daxi; (14) Majiabaog; (15) Songze; (16) Shixia, Niling; (17) Nianyuzhuan; (18) Hedang, Sbijiacun; (19) Ytmg Long, Slao Lo Wan, Tung Kwu, Man Kok Tsui, Tung Wan Tsai, Lamma Island, Tai Wan, Sham Wan, Sbek Pik; (20) Zuoxuangongshan, Wugongshan, Houshan; (21) Futoubu; (22) Tazijinshan, Dingdapushan, Wanglang; (23) Phung Nguyen, Dong Dau, Lung Hoa, Xom Ren, Thanh Den; (24) Trang Kenh, Viet Khe; (25) Sanxingdui; (26) Erlitou; (27) Anyang; (28) Baiyangcun; (29) Dadunzu; (30) Non Pa Wai, Khao Wong Praehan, Nit Kham Haeng, Lopburi, Noen Ma Kok; (31) Khok Phanom Di; (32) Nong Nor; (33) Xing'an; (34) Xinjie, Zhongshan, Tangiiahun; (35) Matitang; (36) Wushui; (37) Yuanlongpo; (38) Tung Wan Tsai, Doumen, Apowan, Tangxiahun; (39) Ban Kao; (40) Non Nok Tha, Non Praw, Non Pa Kluay; (41) Ban Prasat, Noen U-LOke; (42) Ban Don Ta Phet; (43) Ongbah; (44) Ban Lure Kao; (45) Ban Wang Hi; (46) Non Dua; (47) Ban Chiang Hian, Muang Fa Daet; (48) Nong Yang; (49) Shizhaishan, Dian; (50) Samrong Sen; (51) Chau Can, Xuan I.a; (52) Sa H u ~ (53) Dong Son; (54) Mi Son, Dong Duong; (55) Yinshanling; (56) Doe Chua; (57) Angkor.
The Archaeology of Mainland Southeast Asia
15
no one has suggested that agriculture was involved. At Liyuzui, a number of human burials have been found within the occupied cave. Eighteen flexed burials have been examined at Zengpiyan. Chang (1986) has suggested that these sites represent broad spectrum foragers, who took advantage of the warm conditions and would have encountered wild rice growing as far north as the Yangzi Valley. In both the Huanghe and the Yangzi valleys, the ensuing cold phase saw the establishment of the first village communities, millet being cultivated in the former area and rice in the latter. This transition may represent an adaptive response to a climatic deterioration (Higham, 1995a). There followed a long sequence of Neolithic cultures spanning at least five millennia to the first evidence of bronze working. Pengtoushan, on the northwestern shore of Lake Dongting in the fiat, marshy reaches of the middle Yangzi, has provided the earliest evidence for a settled village community and rice remains are relatively abundant as a ceramic temper (HAI, 1990; He Jiejun, 1986; Yan Wenming, 1991). This site, which covers about 1 ha and rises to a height of 4 m, was excavated in 1988 over an area of 400 m2. The foundations of four houses were traced, and 19 human burials were found, some with pottery vessels as grave goods. The inhabitants also used pendants of siltstone, but polished stone adzes were rather rare. The radiocarbon dates indicate settlement between about 8500 and 7800 BP. The rice remains have not yet been ascribed to a wild or domesticated plant. In the marshy habitat reconstructed for the site, rice could well have been collected from wild stands. Nevertheless, Pengtoushan and other similar settlements in the middle Yangzi area evidence a novel settlement form, the permanently occupied village and associated cemetery. Such communities were the basis of lowland prehistoric settlement and remain the backbone of rural life. An increasing number are known not only from the middle, but also from the lower Yangzi, where Hemudu has yielded abundant evidence for life in the low-lying swamp margins 7000 years ago (ZPM, 1978). The foundations of houses raised above the swampy substrate on wooden piles have been found, as well as a rich material culture. There are numerous stone adzes, as well as the wooden sleeves into which they were halted. The remains of the houses show considerable carpentry skill, with neat mortice and tenon joints being used in building construction. Wood was also used for spades, spears, and spindle whorls. Among the more unusual artifacts are the bone spades. These shoulder blades were perforated twice to allow them to be lashed to a wooden handle, and their working surfaces bear a polish resulting from use in the soft lake muds. Pottery vessels were made in a wide range of forms, and clay was also molded into animal and human figures.
16
Higham
Forty-seven animal species were represented by the bones, many indicating lacustrine conditions. There are remains of David's deer (Elaphurus davidianus) and water buffalo (Bubalus bubalis), both of which are adapted to swamplands. The latter might even have been domesticated. Bones of the domestic pig and dog were abundant. The range of species indicates a climate rather warmer and wetter than today, a conclusion that is supported by some of the plants represented only by pollen. Fish and water birds, such as the cormorant, crane, egret, and duck, were also found. The waterlogged lower levels favored the preservation of plant materials ( Z P M , 1978). These included oak leaves, the remains of gourds (Lagenaria), and water chestnuts (Trapa). There was also a layer of cultivated rice remains which had accumulated to a depth of between 20 and 50 cm (You Rujie, 1976; Zhou Jiwei, 1981). Hemudu appears to have been occupied for at least a millennium. With the passage of time, the number, size, and complexity of these agricultural communities increased. In the middle reaches of the valley, we encounter the Daxi (6500-5000 BP) and Qujialing (5300-4600 BP) cultures. The former villages contained houses, storage pits, and cemeteries (Li Wenjie, 1986). Cultivation of rice was undertaken with stone spades, harvesting with stone or shell reaping knives. The Daxi cemetery has been excavated, and 208 burials uncovered. Pottery vessels, jade, ivory, and shell ornaments were associated with the dead (Li Wenjie, 1986). North of Lake Taihu in the lower Yangzi region, there are village sites belonging to the Majiabang and later Songze cultures (SMRPC, 1980). The increasing wealth of the burials with time suggests that there was a trend toward more intense social ranking. On the basis of pottery styles and other similarities in material culture, it is highly likely that the expansion of these agricultural groups proceeded in a southerly direction toward Lingnan. Shixia and Niling, which date to the early third millennium BC, illustrate this trend. Guangren and Wangping (1981) have shown that there are unusual forms of pottery vessels, such as the gui tripod, that have a wide distribution in sites stretching from Shixia in the south to Shandong province in the north. This phenomenon has encouraged Chang (1986) to suggest the existence of a Chinese Interaction Sphere, in which a common origin and later communication between regional Neolithic cultures resulted in a veneer of like artifacts. While the characteristic items peter out south of such sites as Shixia, there are distinctive regional Neolithic groups in Lingnan and the lowlands that flank the Red, Mekong, and Chao Phraya rivers. The sequel to this intrusive settlement in the upper Beijiang catchment is seen in the sites of the Nianyuzhuan culture, represented at several settlements including the middle layer of Shixia itself. Here we find modifi-
The Archaeology of Mainland Southeast Asia
17
cations in the mortuary ritual, but also an element of continuity. The 32 burials of this second phase are still orientated on an east-west axis, and extended inhumation remained the norm. Grave goods included fine ceramic vessels, decorated with a great range of impressed designs, but the exotic stone ornaments are lacking. One of the important points about this culture, represented at several sites such as Nianyuzhuan itself, Pushaoshan, and Zoumagang, is the recovery of oval or round kilns for firing the fine, thin-walled pottery found as grave offerings. The stone industry, too, shows considerable refinement in the manufacture of projectile points, adzes, chisels, and spearheads. The recovery of saddle querns is one possible pointer to the successful establishment of rice cultivation in this area during the period 2000-1500 BC. This intrusive impetus did not, as far as is known, reach the coastal tract, where we find strong strands of continuity meshed with a number of new features. Hedang, which gives its name to a group of sites in the Zhujiang Delta region, saw a continuation of the exploitation of marine resources in association with pottery impressed with geometric designs. Firing temperatures were in excess of 1000°C. The site includes thick deposits of shell and dates to the late third millennium BC. It has yielded 77 inhumation graves, in which males were interred with polished stone adzes, and females with spindle whorls. Tooth evulsion was common. Some very fine ornaments were also recovered, made from ivory, as well as exotic stone bracelets and slit rings. There was also a vigorous bone industry, which produced hairpins, needles, awls, and weaving shuttles. A fortuitous find at Shijiaocun included the remains of domestic wooden structures nearly 15 m in length, which were rectangular and raised on wooden piles. The degree to which these people incorporated rice agriculture into their subsistence activities, however, is not known, and Zhu Feisu (1984) has adroitly pointed out that this innovation appears to have been slow to develop in such coastal, estuarine contexts. Li Guo (1994) has further suggested that there may well have been an increasing trend toward agriculture, but not at the expense of exploiting the marine habitat. He also suggested that rice cultivation would in all likelihood have been derived from a northerly source, presumably incorporating such sites as Shixia. Radiocarbon determinations place this culture in the period 3000-2000 BC. This late Neolithic period clearly encompassed a series of regional traditions, none of which was more tightly distributed or singular in its characteristics than the "stone shovel" sites of southern Guangxi (Allard, 1994). These large and enigmatic artifacts, the largest of which stand up to 70 cm in height, were often made of stone unsuited to any possible agricultural or industrial function. Groups of shovels were commonly found interred in
Higham
18
pits, positioned vertically. No associated settlements have been found, and ritual activity of some form is the most likely explanation. In Hong Kong, the late Neolithic (LN) has been subdivided into two phases. LNI, named after the site of Yung Long, is again characterized by geometric decoration on the pottery, imparted by a carved paddle and a decorated anvil. It has been dated from about 2650 BC (Meacham, 1993). Excavations have uncovered graves on a north-south orientation that contain pottery vessels and slit stone rings. Jadeite ceremonial axes suggest, to Chiu (1993), the development of a strong ritual element by this juncture. The ceramic assemblage developed into the classic soft geometric or LNII, which belongs to the first half of the second millennium BC. Excavations at Sha Lo Wan, a promontory commanding views across the Zhujiang Estuary to the mainland, have revealed an occupation site of this period (Chiu, 1993). As in the preceding phases, there are ovens, postholes, and much pottery. Spindle whorls and net weights attest a textile industry and fishing, and there were numerous stone adzes and projectile points. Excavations at Tung Kwu have also exposed a LN occupation containing much soft geometric pottery and utilized stone but, again, insufficient organic material to illuminate the subsistence base. The Hanjiang Estuary and its riverine hinterland in eastern Guangdong saw a cultural sequence not dissimilar to that recognized in the Xijiang catchment to the west. Over the period from approximately 3000 to 1500 BC, Zhu Feisu (1984) has described three phases, named after the sites of Zuoxuangongshan, Wugongshan, and Houshan. We again encounter settlements located on the coastal fringe and on raised ground adjacent to the main rivers and the local firing of pottery impressed with geometric designs. The middle phase at Futoubu has also yielded sophisticated kilns used over a lengthy period of time for firing pottery vessels, and the local stone industry flourished, providing polished adzes, arrowheads, knives, and awls. The last phase saw a locally distinctive pottery form, a jug in the shape of a chicken. The Fubin Culture
The sites ascribed to this culture are pivotal to our understanding of the late Neolithic in Lingnan. The presence in graves of exotic stone artifacts allows us to relate the sites to the late Shang and early Western Zhou, that is, a couple of centuries either side of 1100 BC. They follow the later neolithic sites just described, clustering in the valley of the Hanjiang and the adjacent coastal plain, and into southem Fujian Province, with a preferred location on low hills. Pottery vessels predominate among the grave
The Archaeology of Mainland Southeast Asia
19
goods, and there is also a new range of highly sophisticated stone weapons and ornaments. Among the former, the ge halberd is most informative, because it provides parallels with related weapons from the zhongyuan. The presence of glazed pottery bearing incised symbols or numbers also indicates relations with communities in Jiangxi Province (Allard, 1994). Jade rings are also found, and to judge from the variation in the size of the grave and the quantity of offerings among the 22 interments at Wanglang, these late Neolithic communities displayed an element of social ranking. This same tendency was noted at Tazijinshan, where the richest and largest grave also occupied the summit of the hill. This particular grave, number 1 among the 16 excavated, involved considerable energy: it measures 4.2 x 2.9 m, and the base is 3.6 m deep. It is also one of eight equipped with a ledge running round the basal part. Ceramic vessels predominate among the 36 grave goods, and there are also three ge halberds, but while contemporary with the vigorous and long-established Shang bronze tradition, no grave at the site has yielded any bronze artifacts. There is some doubt over the context of a bronze ge halberd from the nearby site of Dingdapushan, found in association with a Fubin-styte stone adze, but in the other Fubin sites, we find the same situation: ceramics and fine stone artifacts that indicate some form of relationship with complex societies in the zhongyuan, but there is no metal.
The Phung Nguyen culture
A thousand kilometers to the southwest, we meet the delta of the Red River and the Phung Nguyen culture. As with the Fubin sites and Shixia, this culture represents a marked departure from the earlier pattern of settlement. Sites are located on slightly elevated terrain commanding stream valleys above the confluence of the Red and Black rivers. There are three phases, based on changes in pottery typology, which Ha Van Tan (1991) has dated between the end of the third millennium BC and about 1500 BC. The earliest, the Go Bong phase, is characterized by pottery decorated with burnished areas interspersed with incised bands filled with fine impressions. Spirals and "S" motifs were popular. This technique was modified by the second phase, the decoration being more formally applied but still retaining the spirals and "S" motif. The infilled bands became less popular with the final phase and incised decoration took the form of straight or wavy lines. Only 11 late sites of the 52 examined contained bronze, and no recognizable metal artifacts have been found. The fragments were, however, made of a tin bronze.
20
Higham
Excavations at Phung Nguyen covered 3960 m z, and despite this extensive area opened, no bronze was encountered. The rich material culture included over a thousand adzes or adze fragments. Most were quadrangular in cross section and rectangular in form, but there were also four shouldered specimens, and stepped adzes have been found that recall South Chinese forms. Stone bracelets were particularly abundant at Phung Nguyen, the 540 specimens being divisible into eight types (Nguyen Ba Khoach, 1980). There are also a few stone arrowheads and a bone harpoon. The degree of skill associated with the manufacture of stone jewelry is particularly clear at Trang Kenh (Nguyen Kim Dung, 1990). The pottery relates to the late Phung Nguyen styles. At this juncture, it is recalled that some sites of this phase have revealed a few fragments of bronze, and Trang Kenh is interpreted here as one belonging culturally to the terminal Neolithic. Located near the coast at Hai Phong, excavations at the site have revealed a wide range of nephrite ornaments, including bracelets and beads, as well as the chisels, drill points, saws, and grinding stones used in their manufacture. Glover (personal communication) has noted that a small amount of bronze has been recovered. The radiocarbon dates accord well with the received dating of the Phung Nguyen culture; the pooled mean for the four dates is 1679-1514 BC. Although no burials were found at Phung Nguyen, Hoang Xuan Chinh (1968) has uncovered 12 at Lung Hoa. These had been excavated up to 5.2 m into the ground and were provided, as at the wealthy Fubin sites, with ledges. The offerings in two graves included stone bracelets, beads, earrings, adzes, and pottery vessels, but others contained only pots and adzes. The excavators have suggested that this may reflect differential social ranking, although a larger sample would be necessary to examine this issue further. A stone ge halberd from burial 9 is a form that parallels those in the Fubin sites, such as Tazijinshan and Yuanguang, as well as at Sanxingdui in Sichuan (Nguyen Phuc Long, 1975). The occupants of Sanxingdui were in contact with the later Shang sites of the Huanghe Valley, where there are jade and bronze examples of this type of halberd (Bagley, 1988). Murowchick (1989) has argued that the presence of such similar artifacts at the same period is most unlikely to result from independent development and favors exchange contact to explain the presence of the Lung Hoa example. This is not the only example of contact between the Phung Nguyen culture and Shang China. Ha Van Tan (1993) has described a series of jade yazhang, or ceremonial knives, of a singular form from Phung Nguyen and Xom Ren. These are precisely matched in the Zhujiang Delta area, at Sanxingdui, Erlitou, and later Shang sites, and must surely represent imports from the Huanghe Valley (EATIA, 1976; Tang Chung, 1994). The dating of the latter sites fits well with the available radiocarbon dates from Bac
The Archaeology of Mainland Southeast Asia
21
Bo. Such exchange with the contemporary neolithic sites of Lingnan, which ultimately reached to Sanxingdui and the Shang capital at Anyang, could also have introduced knowledge of the properties of copper and tin. The Phung Nguyen culture probably became established within the period 2500-2000 BC, and available radiocarbon dates suggest that its late phase was developing into its successor, the Dong Dau phase, by about 1500 BC. In Yunnan, there are further sites that incorporate inhumation cemeteries with dentate stamped and burnished pottery. Baiyangcun is located only 60 km east of a tributary of the Lancang (Upper Mekong) River, close to the headwaters of the Red River. Excavations have revealed a stratigraphic sequence 4.35 m deep, divided into two phases (YPM, 1981). The foundations of 11 houses have been identified, and a cemetery of at least 34 burials, with heads oriented to the north or east. Their layout suggests the presence of two clusters, in which there is at least one double burial. Unusually, there are no grave goods, and many of the skeletons lack a cranium. The pottery was decorated with parallel incised lines infilled with impressions, a technique with parallels in Phung Nguyen and many other sites to the south, and the single radiocarbon date indicates contemporaneity with early Phung Nguyen and Non Pa Wai. The closest parallels to the pottery decoration strongly suggest links with communities down the Red and Mekong rivers. The same may be said of Dadunzu (YPM, 1977). This settlement covers 0.5 ha, and excavations in an area of nearly 500 m 2 have revealed 15 house plans and 27 burials in the cemetery. Houses were orientated on a north-south or an east-west axis, and superpositions indicate some length of settlement, the subsistence base of which included rice cultivation and the raising of domestic stock. Adults were buried in an extended position, and infants were interred in jars. There was no preferred grave orientation, and the infant jar burials were not regularly placed in association with adult burials. Once again, the pottery was decorated with infilled incised bands, and the single radiocarbon date falls within the chronological range for this tradition. Few Neolithic sites are known in the Mekong and Chao Phraya valleys to the south, so knowledge of the timing of early rice agriculture within settled village communities in this extensive area is sparse. This issue is being approached from two directions. The first involves excavation, principally in Central Thailand, integrated with the analysis of pollen, charcoal, and phytolith remains from sediment cores. If there was a major expansion of cultivators, using the river valleys as a conduit, then their activities should be reflected in changes in the vegetation cover. This is the case in Central Thailand where Kealhofer (1992) has recognized, in preliminary results
22
Higham
from the cores taken in the Lopburi region, changes compatible with burning and clearance of the natural vegetation that took place at or a few centuries before 2000 BC. As it happens, this date matches that for the initial settlement of two sites, Khok Phanom Di and Non Pa Wai, both of which contain Neolithic cemeteries (Higham and Thosarat, 1994; Pigott, 1992). A further series of cores taken recently in Northeast Thailand is under analysis. This pleasing harmony is, however, broken by cores from near Khok Phanom Di, an estuarine settlement and cemetery. A series of burning episodes and vegetation changes has been identified on the basis of charcoal, pollen, and phytoliths (Kealhofer and Piperno, 1995; Maloney, 1991; Maloney et al., 1989), These have been AMS dated to about 5500 BC, 53005000 BC, and 4000 BC. More than one factor, including natural conflagrations, hunter-gatherer burnoffs, and, possibly, agriculture, could have been responsible. We now have some archaeological evidence to accompany the data from the sediment cores. Two coastal sites have been examined: Nong Nor and Khok Phanom Di. The former reveals two periods, the first an occupation phase when the site was located on a broad marine embayment close to the sea (Boyd, 1994), dated to the mid-third millennium BC. This small settlement (0.07-0.1 ha) was later disturbed by a Bronze Age cemetery. It lies 14 km south of Khok Phanom Di, and phase 1 antedates it by about 500 years. Deposits comprised a shell midden that incorporated hearths, lenses of burnt material rich in charcoal, and a material culture which indicates a range of activities. Subsistence involved the collection of marine and mangrove shellfish, fishing, and the hunting of marine mammals. No evidence for rice or domestic animals has been encountered. In his analysis of the pottery, O'Reilly (1994) has noted similarities in form and decoration with Khok Phanom Di, including burnished and incised dentate stamped ware. There are many other parallels: both sites have yielded similar burnishing pebbles, bone fishhooks, awls, sandstone grinders, and polished stone adzes. At the present stage of analysis, Nong Nor appears to have been a base for marine hunting, fishing, and collecting, where pottery was manufactured, and activity areas included the conversion of large mammalian bone into fishhooks, points, and awls. It was probably occupied for months rather than years. While the material culture at early Khok Phanom Di reveals continuity with Nong Nor, there were significant changes in subsistence, material culture and social behavior. Thompson (1992) has shown that the rice remains were from a cultivated variety, possibly grown in freshwater swamps behind the mangrove fringe. The inhabitants employed shell knives, the wear on which is compatible with rice harvesting (T Higham, 1993), and we also
23
The Archaeology of Mainland Southeast Asia
find stone hoes (Pisnupong, 1993). The dog, exotic to tropical Southeast Asia, appears in the faunal spectrum (Grant and Higham, 1991), but subsistence remained heavily reliant on marine resources. Khok Phanom Di was also settled permanently for about five centuries, and the dead were interred in collective wooden mortuary structures (Fig. 7). The ritual involved inhumation with the head oriented to the east. The corpse was placed on a wooden bier, wrapped in a fabric shroud with a range of grave goods, including dentate stamped pottery vessels, shell and fishbone jewelry, and, on occasion, the anvils and burnishing stones used in pottery manufacture. The latter items provide a key to appreciating the social importance of the ceramic industry, for as time passed, anvils were
O
oo g )
Jl°Cluster D"
©
o
Oo
oo°
o
,
,,
ooo
o
0% o 00
~
°
O
0
•J
c~°
@
©
o
o O
o
0 oO 0
0
o..,m. O o 0
o
(.~]~@%
o
o
o
~'~ o jo
0
o
(~o
-co
0~
o 0
o o9 ~,o
o
o
0
6
0
Cluster E
o~
oo
o
0
.-
0
L
I
I
H
i
G
3o 0
0o
o o
~\o
o
F
= r~
o
o
o \,qC)~.!
i
['~I'~~ ~ ClusterA
'
D
I
C
J
Im
Fig. 7. Khok Phanom Di during Mortuary Phase 4.
°°° B
I
--
Odf
C)X3oO ©
Ooi ~\\k~
oo(9o
~
-
0
o nO d
-7
oo°
0
0o
Oo
0
o
0o0 ° o o O%Oo0oOo+ Ao 0
8 o
°o° o
o
0
o
oF
~
o
o~o
o
Oo
0 8 o
oOO
;9oOo
°2 O or 9
°O
ooo °°°%°°oOo OOo
10
o
A
4
24
Higham
included in the graves of some women and young individuals, while men were often interred with elaborate turtle carapace ornaments. Some women were associated with extraordinary wealth, in one case involving a garment embroidered with over 120,000 shell beads that would have dazzled in the sun. It has been suggested that such mortuary finery indicates the high social standing of woman potters, for it would have been important to retain the next generation of young women in the community. This might help us appreciate why some infants were also buried with considerable wealth (Higham and Thosarat, 1994). Several alternative explanations are consistent with this sequence from Nong Nor to Khok Phanom Di. The former, with its evidence for much burning, might represent a series of coastal communities with deep local roots that were responsible for the charcoal peaks and associated vegetation changes in the coastal sediment cores. If so, then there is no archaeological evidence there, as yet, for the cultivation of rice. From about 2000 BC, however, it is evident that the exotic dog, as well as rice cultivation, was established along the Gulf of Siam, a date comfortingly close to the evidence for widespread forest disturbances noted in the Lopburi area by Kealhofer (1992). Alternatively, Nong Nor might represent a brief, dry season occupation by a group accustomed to cultivating rice in a more suitable part of their annual territory during the rainy season. It is also possible that Khok Phanom Di was occupied by intrusive cultivators, who shared the coastal tract with indigenous coastal foragers, although the close similarities in virtually all aspects of material culture at Khok Phanom Di and Nong Nor weaken this alternative. While deferring to further research, the Nong Nor site points a clear message: we cannot be satisfied with any explanation of agricultural origins that is restricted to the intrusion of cultivators originating in the Yangzi Valley. Yet there remains the widespread distribution of AA languages, in particular, the number of widespread cognates incorporating terms for rice and rice cultivation (Zide and Zide, 1976). If we were to try to identify a common archaeological horizon for an intrusive Neolithic expansion, it would lie in the various groups of prehistoric sites identified in Lingnan, the lower valley of the Red River, central Thailand, and possibly south of the Mekong as it arcs round the Khorat Plateau (Fig. 5). This establishment of inland agricultural villages represents a novel settlement form dated, on available archaeological evidence, from the midthird millennium BC. The best-known group of such sites, known collectively as the Phung Nguyen culture, clusters above the confluence of the Red and Black rivers. Ha Van Tan (1980) has pointed out that the preferred method of decorating Phung Nguyen pottery involves incised curvilinear bands infilled with comb or shell impressions and has pointed to parallels
The Archaeology of Mainland Southeast Asia
25
with pottery from Samrong Sen in Cambodia and Ban Kao and Non Nok Tha in Thailand. We can now also add to this list the early incised pottery of Non Pa Wai and Ban Chiang and some of the mortuary wares from Khok Phanom Di. It is, therefore, possible that the widespread distribution of dentate stamped pottery, the new configuration of inland agricultural settlements and the inhumation burial rite in permanent cemeteries could reflect a Neolithic wave of advance into Southeast Asia via the main river systems. According to the present evidence, this took place during the third millennium BC, but many sites were not occupied until the second millennium, and large lowland tracts, such as the Khorat Basin, remained empty into the early first millennium. But the possibility that there was an independent transition in tropical Southeast Asia cannot be discounted.
THE BRONZE AGE To judge from the evidence from Lingnan, Khok Phanom Di, and Phung Nguyen, the Neolithic inhabitants of Southeast Asia deployed considerable craft skill, and exchange of exotic goods played an important role in the determination of social status. These trends were maintained during the Bronze Age, part of the prehistoric sequence that has attracted considerable confusion. Muhly (1988, p. 16), for example, noted that In all other comers of the Bronze Age world--China, Mesopotamia, Anatolia, the Aegean and central Europe--we find the introduction of bronze technology associated with a complexof social, political and economicdevelopmentsthat mark "the rise of the state." Only in Southeast Asia, especiallyin Thailand and Vietnam, do these developmentsseem to be missing. This situation has been clarified over the past years, although the last pieces of the jigsaw have yet to be found. The new information comes from a better chronological definition of the Neolithic, the examination of copper mines and associated smelting and casting areas, the application of A_MS dating to one of the two sites that has been cited as evidence for bronze working in the third or even fourth millennium BC, and the exposure of Bronze Age cemeteries that illuminate social organization. To this list, it is also now possible to appreciate more clearly the Chinese Bronze Age, in terms of its chronology, distribution, and contacts with Southeast Asia. By the middle of the second millennium BC, we can recognize a series of regional settlement concentrations from the Hanjiang to the Red River deltas. They have a number of features in common. Although there remained a distinct coastal orientation, there was also a preference for the establishment of small villages in the inland river valleys. These communi-
26
Higham
ties cultivated rice and maintained domestic stock. They also included skilled workers of clay and stone. The former employed enclosed kilns, and their fine wares were fired under controlled conditions at high temperatures. The latter made tools and ornaments of high quality, some of which were used as mortuary offerings. The burial technique, extended inhumation, saw a considerable expenditure of energy in the provision of deep graves that were equipped with ledges and contained impressive sets of grave goods. These sites have in common a further variable of critical importance. They include jade artifacts that have their closest parallels in later Shang contexts to the north. There can be no doubt that coastal and riverine exchange placed these late Neolithic communities in touch with one of the most sophisticated bronze traditions in the ancient world. That bronzes traveled the same routes is clearly evident in the recovery of stray finds, the distribution of which again stresses the importance of riverine communication. Perhaps significantly, one of the earliest specimens, a you vessel ascribed on the basis of its dragon design to a later Shang context, has been found at Xing'an, almost literally on the watershed between the Xiangjiang, which flows north to Lake Dongting, and the Fuyishui, which flows south to the Xijiang (Liang Jingjin, 1978). A similarly exotic dragon-phoenix design was identified on a halberd from Xinjie, located in the same part of northeastern Guangxi (GXBWG, 1984), while Huang Zhanyue (1986) has noted that the nao bell from nearby Zhongshan has close parallels in Hunan. Both of the latter finds date to the Western Zhou period, as does a probable copy of a yong bell from Mei'ershan. Further downstream, at Matitang, a stray lei vessel incorporating dragon designs has been recovered, a vessel virtually identical to one from Wushui (GXBWG, 1984). The location of Xing'an and Xinjie indicates the most likely exchange route for a marked concentration of early exotic bronzes northeast of Nanning. It may not be coincidental that the bronzes are found in the same general area as the late Neolithic caches of stone spades. The pivotal location of Shixia, however, should not be overlooked, and indeed the upper layers, likewise belonging to the Western Zhou and Spring and Autumn periods, have yielded a significant range of bronzes including a short sword or dagger, an ax, awls, and scrapers. These stray finds and the upper context at Shixia point unequivocally to a vigorous exchange network linking Lingnan with the Yangzi Valley and ultimately, with the zhongyuan. Yuanlongpo, a most important cemetery near the Wuming River Valley northeast of Nanning, provides us with a glimpse of the role bronze played in mortuary rituals in the early centuries of the first millennium BC (Allard, 1994; GXBWG, 1988). Unfortunately, the excavation report does not provide the detailed information from each of the 350 burials uncovered, but it is still possible to obtain some valuable results. The mortuary ritual
The Archaeology of Mainland Southeast Asia
27
involved inhumation in individual graves, some of which were provided with a ledge or a side chamber. A wide variety of grave goods was encountered, about 10% being bronzes. Already these provide a portent of the warfare that was to dominate later bronze assemblages in this region: the items include spearheads, axes, arrowheads, and daggers or short swords. A ritual or festive element is also seen in the vessels, at least two of which were probably exotic, and there are also knives and a fragment of bell. The division between imported and local bronzes is facilitated by the recovery of some 12 stone bivalve molds, some of which were broken, probably as part of the mortuary ritual. These were intended for casting yue and fu axes, dun (the tubular cover at the end of a spear), knives, and arrowheads. Many jade ornaments were also found as mortuary offerings, and fragments of lacquer indicate some elaboration in the manufacture of coffins. Allard (1994), in summarizing this site, has stressed the likelihood that there was some form of social hierarchy, for burial 147 not only was 1 of only 16 equipped with a ledge, but also included a probably exotic you bronze vessel, was particularly large, and contained more than the usual number of grave goods. It is not, possible, however, to probe further and seek evidence for or against the presence of ascribed rank rather than achieved status through personal endeavor. Coincident with the arrival of exotic bronzes from the zhongyuan (the central plains of the Huanghe Valley) and the middle reaches of the Yangzi, in the context of the late neolithic in Hong Kong, was the beginnings of a local tradition in casting that involved the production in bronze of a limited range of artifacts long since rendered in stone or bone. These comprise arrowheads, axes, fishhooks, and spearheads. The conjunction between hard geometric pottery and bronze in Hong Kong sites was noted during the 1930s, for Finn (1958) discovered six axes during his excavations on Lamina Island, four of which were socketed and cast in a bivalve mold. He stressed their affinities with axes found in Vietnam, Laos, and Cambodia, noting, in particular, similarities with those from Samrong Sen. Fishhooks, leaf-shaped knives, and arrowheads matching those from Sham Wan have also been found at Man Kok Tsui on the southeast coast of Lantao Island (Watt, 1968), but one of the best-known bronze assemblages comes from Tai Wan, about 1.5 km north of Sham Wan. Finn (1936) has described two socketed spearheads from this site, as well as a socketed axe, both of which he compared to similar examples from Bac Bo. One of the spearheads, for example, had the same two slots on the socket that are regular features on examples found to the south. In 1937, Schofield excavated at Shek Pik on Lantao Island (Meacham, 1975). Although not in stratigraphic contexts, he found six bivalve sandstone molds for casting socketed axes, three having clear parallels with
28
Higham
those from Vietnam and Thailand. The hard geometric layer at the site also furnished a few bronze items. The 1971 excavation at Sham Wan encountered bronze fishhooks and arrowheads, the alloy including about 10% of tin. More recent excavations have clarified the chronology of the Bronze Age in this coastal region (Meacham, 1993). There are three determinations from Kwo Lo Wan and one each from Lung Kwu Sheung Tan and Sha Po Tsuen. They suggest the establishment of bronze casting by 13001000 BC. Research at Kwo Lo Wan has also added considerably to our knowledge; three radiocarbon dates from burials place the site between 1300 and 1000 BC (Meacham, 1994). Two of the eight burials, which were orientated on a north-to-south axis, contained bivalve sandstone molds for casting socketed axes. Other offerings include hard and soft geometric vessels, slotted stone rings of marble and agate, and two bronze projectile points. Further examples of sandstone ax molds have been found in the Zhujiang Estuary sites of Tung Wan Tsai, Doumen, Apowan, and Tangxiahun; the first site having also produced three radiocarbon dates between 1700 and 927 BC (Li Wan, 1995; Rogers et al., 1995). A similar sequence and chronological framework has been identified in the lower Red River valley. The first bronzes were found in contexts that received exotic northern jades. A few pieces of bronze have been found in some late Phung Nguyen sites, but the following phase, named after the site of Dong Dau, saw a virtually identical range of bronzes and the same casting technology as found in the Zhujiang Delta area. There is a reasonable corpus of radiocarbon dates for the Dong Dau phase. Most come from Thanh Den, a site with a relatively shallow stratigraphy and evidence for casting in the form of molds and melting furnaces. Two of the dates from Thanh Den seem aberrant (Anon, 1990), particularly compared with those available from the later Phung Nguyen contexts. The remaining 11 suggest that bronze working was established within the period 1500-1000 BC, a context that corresponds well with the available dates for virtually the same industry in Hong Kong. Dong Dau is located just north of the Red River, 35 km east of Phung Nguyen. It covers about 3 ha and has a cultural stratigraphy between 5 and 6 m deep. Its basal layer contains late Phung Nguyen pottery, but thereafter the assemblage developed into the Dong Dau culture. Sites are distributed in the same general area as Phung Nguyen settlements. While the pottery continued to be incised with a series of curvilinear lines originating in the Phung Nguyen repertoire and the stone adzes and points continued from local prototypes, there was a flowering of the local bronze industry. Dong Dau and Thanh Den have provided sandstone molds. Artifacts made from a tin-copper alloy included axes, chisels and arrowheads, socketed spears,
The Archaeology of Mainland Southeast Asia
29
and fishhooks. The analysis of a sample of 22 Dong Dau bronzes has revealed an alloy (with no lead) similar to that in use in Northeast Thailand at the same time. Tin levels, however, appear to have been rather higher, with values varying between 6.8 and 28% and averaging 11%. The same alloy was used for the axes, spearheads, points, fishhooks, and bracelet analyzed, but three arrowheads were made from a most unusual alloy comprised of copper and between 2.9 and 6.5% antimony, with no tin (Trinh Sinh, 1990). Small clay-lined furnaces, which were probably used for melting copper and tin before casting, have been found at Dong Dau and Thanh Den. Ha Van Phung (1993) has identified three phases of the succeeding Go Mun culture, largely on the basis of rim typology; the earliest is best represented in the upper layers at Dong Dau. Go Mun itself, where the second and third phases are present, is located just above the strategic confluence of the Red and Black rivers (Ha Van Phung and Nguyen Duy Ti, 1982). Excavated between 1961 and 1971, its particular interest lies in the unusual recovery of bronzes in nonmortuary contexts. Moreover, the excavations covered various parts of the site and, therefore, provide a spatial dimension to the location of the bronzes. The excavations covered 1500 m 2 in the central and southern parts of the site. But all yielded a similar variety of bronzes, and objects from one of the 1965 squares revealed a considerable density of bronze finds within a stratigraphic buildup that barely exceeded a meter in depth. Although stone adzes and bracelets remained abundant, the assemblage from Go Mun reveals a proliferation in the range and function of bronzes. These can be considered in four distinct categories: decorative, utilitarian, ritual, and for use in conflict. Only three bracelets were encountered, a marked contrast to their predominance in bronze assemblages from the mortuary contexts west of the Truong Son Range. More significant, though, was finding bronze employed in agriculture and industry. The Go Mun sample includes a sickle and five socketed hammers. There are also fishhooks, awls, chisels, axes, and knives. Two fragments of bowls of some sort and a human figurine may reflect a nascent interest in applying bronze to ritual activity, while the presence of a socketed bronze spearhead and arrowheads suggests the presence of conflict. Trinh Sinh (1990) has reported on the results of a spectrographic analysis of five Go Mun bronzes and has found that, as with the Dong Dau material, a tin bronze was used in casting axes and spearheads. While one arrowhead lacked tin, it included 2.1% of antimony. Slit stone earrings of a form paralleled in earlier Phung Nguyen and Dong Dau sites (Ha Van Phung, 1993) are also found in Go Mun contexts.
Higham
30
Central Thailand
In Central Thailand, there are three extensively excavated Neolithic sites, Ban Kao, the first mortuary phase at Non Pa Wai and Khok Phanom Di, all of which date between 2400 and 1500 BC (Higham and Thosarat, 1994; Pigott, 1992; Scrensen and Hatting, 1967). Of these, Non Pa Wai is the most significant, because it lies in the shadow of a major copper deposit in the Khao Wong Prachan Valey, the locus of many ore outcrops, associated mines, and processing sites. Above the Neolithic cemetery there, Pigott and his co-workers have encountered a second phase, again incorporating a cemetery, but in the main comprising the remains of copper smelting and casting into ingots. Three radiocarbon dates from this phase are 1690-1225, 1450-1136 and 1270-800 BC, corrected at 2cr. These indicate that copper smelting and casting were under way at about the same juncture as in Lingnan and the Red River Valley. In the sequel, copper processing grew to such an intensity that smelting and casting debris covered 5 ha (Natapintu, 1988a, 1991; Pigott, 1992; Pigott and Natapintu, 1988). The copper ore deposits of the Khao Wong Prachan Valley are associated with a series of sites in which the ore was first crushed and sorted, then smelted in clay crucibles or shallow bowls set in the ground (Pigott, 1992). Bennett's (1988, 1989) analysis of the slag has shown that some prills contain a significant amount of arsenic. To judge from the clay molds, most metal was cast into ingots. But the Bronze Age cemetery of Non Pa Wai also includes a burial containing the two halves of a bivalve clay mold for casting a socketed axe and a second interment associated with such an axe (Pigott, 1992). It was formerly thought that all Southeast Asian bronzes were cast from a copper-tin alloy, but there is an almost-complete lack of the latter in the Khao Wong Prachan metal industry. Moreover, according to Weiss (1992), who has directed excavations at the large ore processing and casting site of Nil Kham Haeng, working metal was probably a dry season activity, involving task-oriented groups of people with particular skills and experience. Such an organizational framework echoes the way in which pottery manufacture was probably undertaken at Khok Phanom Di. Northeast Thailand
Non Nok Tha, excavated in 1966 and 1968, has, in common with many other sites, an association of bronze artifacts and casting molds with the dead. Dating such sites has proved difficult because of the lack of well provenanced charcoal. In his earlier publications, Bayard (1972) offered alternative chronological frameworks, due to conflicting radiocarbon deter-
The Archaeology of Mainland Southeast Asia
31
minations. The development of AMS dating has now cut this Gordian Knot by allowing dates to be obtained from the rice chaff used as a ceramic temper. The results suggest that the Bronze Age graves belong within the period 1500-1000 BC (Higham, 1995b). Bayard, however, has stressed that the dated sherds do not all come from secure contexts. Nevertheless, there is a consistent body of evidence from more recently excavated sites in the same region of northeast Thailand, including Non Pa Ktuay and Non Praw, that supports a later second millennium BC date for bronze casting (Buranrak, 1994; Wilen, 1989). Ban Chiang remains controversial. The application of the principles of "chronometric hygiene" to the dates, most of which come from grave fill, suggests that the Bronze Age graves there fall within the latter half of the second millennium BC (Higham, 1983, 1994). However, AMS dates are currently being obtained from the rice chaff in mortuary vessels, and it may well be necessary to revise this chronological context further back into the second millennium. It is, for example, intriguing that the copper mines at Phu Lon, 150 km northwest of Ban Chiang, have supplied a radiocarbon date of 1750-1425 BC (2~) from a basal context (Pigott and Natapintu, 1988a). Ingots were widely exchanged, for we find evidence for casting in village communities hundreds of kilometers from the nearest mines. At Ban Na Di a small furnace for raising bronze to melting point was encountered in the Bronze Age levels; this furnace was ringed by crucibles and stone molds for casting axes and arrowheads (Higham, 1988). In this case, the examination of the metal residues within the crucibles as well as the artifacts indicates that copper was mixed with tin as part of the casting procedure (Maddin and Weng, 1984). Whereas clay molds were preferred in Central Thailand, sandstone was employed widely in the Mekong Valley sites for casting axes, spearheads, and projectile points. Clay was used, however, in the cite perdue process for casting bracelets. The skill of the local metal casters may also be judged by the wire-like ties cast through holes in stone bracelets to repair breaks. Doc Chua, located northeast of the Mekong Delta, attests to intensive casting activity that employed sandstone molds for axes, spearheads, harpoons, and arrowheads (Dao Lin Con and Nguyen Duy Ty, 1993). The few extensive exposures of cemeteries have provided some insight into the range of bronzes cast and the social organization of the Bronze Age communities. Bangles predominate, but we also find socketed axes and spearheads, several forms of arrowhead, small chisel-like socketed implements of an undefined purpose, and, very rarely, sickles. Some bangles imitate the form of earlier examples in shell and stone. At Non Nok Tha, Ban Na Di, Ban Prasat, and Nong Nor, it is possible to identify spatial patterning in the cemeteries, seen in rows and clusters of graves (Bayard, 1984;
Higham
32
Higham and Kijngam, 1984; Higham and Thosarat, 1994; Monkhonkamnuanket, 1992; Phommanodch, 1991). Bronzes were never abundant, and there is no clear evidence for a concentration of wealth, expressed in bronzes or indeed any other category of mortuary offering, in a particular social grouping. Only at Ban Na Di does one cluster of burials reveal consistently more exotic grave goods, including bronzes, than another. But the exposure there was constrained, and further excavation is needed to confirm the situation. What does emerge, however, is that no Bronze Age cemetery provides evidence for more mortuary wealth or social ranking than at late Neolithic Khok Phanom Di. The last word has not been spoken on this period, but by degrees, we are now able to identify some interesting possibilities. During the late Neolithic in Lingnan and Bac Bo, there was some exchange for jades and bronzes with states in the Yangzi and Huanghe valleys. We cannot yet pinpoint this date, but the Chinese source material is consistent with a period a few centuries before or after 1500 BC. It is possible that such exchange brought the knowledge of copper and tin smelting into Southeast Asia, but proponents of an indigenous origin have an equally valid case that receives support from the recognition of copper smelting and casting, without recourse to alloying, in Central Thailand from between 1500 and 1000 BC. This bronze working tradition has been identified through the bivalve casting technique, and the characteristic products are found from Central Thailand to coastal Guangdong and down the course of the Mekong River from Yunnan to the sea. It was a distinct tradition to that which developed in Central and northern China.
THE IRON AGE
Chinese imperial expansion and contact (through exchange) with India were factors that contributed to a sea change in Southeast Asian prehistory in the mid-first millennium BC. Other variables included the concentration of population in larger settlements, the institution of methods of water control, and the development of iron technology. During this period, we can recognize related regional developments that underlay the rise of states. Ironworking may have developed through local initiatives, the introduction of the idea, or a combination of both. The use of hematite as a flux in copper smelting in Central Thailand, along with the early dates there for forged iron artifacts, could evidence local origins, but the rapid rise of ironworking during the period of Warring States (475-221 BC) in China also saw the introduction of the new technology into Yunnan and Lingnan. As with bronze, both possibilities might be valid. In any event, the forging
The Archaeology of Mainland Southeast Asia
33
of iron weapons and agricultural implements took hold rapidly in the lower Mekong and Chao Phraya valley communities, as has been shown at the cemetery of Ban Don Ta Phet (Glover, 1989, 1990). This site lies at the eastern gate of the Three Pagodas Pass and was exposed early to Indian contact. Graves contain exotic glass, carnelian, agate, and jade jewelry and locally manufactured bronze bowls that express a considerable advance in casting technology. The difficulty of casting and decorating these bowls, some of which have been found in Indian sites, suggests specialization. Ban Don Ta Phet, dated to the early fourth century BC, is one of many Central Thai sites that evidence participation in a growing international exchange network (Glover, 1989). At Ongbah, for example, Sorensen (1979) has described wooden coffins associated with a set of bronze drums that almost certainly represent exchange with Yunnan. Mankong (1989) has recovered exotic beads of agate, carnelian, and glass at Noen Ma Kok, while iron implements include socketed hoes. At Ban Lum Khao and Ban Wang Hi, Natapintu (1988b) and Chiawanwongsa (1987) have found the same range of exotic jewelry associated with inhumation graves. The settlement aspect of Iron Age culture has been clarified by a number of site surveys. Ho (1992), for example, has suggested that there was a hierarchic pattern in which regional centers were ringed by moats. The Mun and Chi valleys of the Khorat Plateau witnessed a proliferation of moated settlements. Although no firm dates have yet been obtained, the ubiquity of Iron Age occupation within, and the fact that some sites have no occupation after the Iron Age, suggests that moat construction was under way during the later first millennium BC. These sites concentrate in the most arid part of Southeast Asia, with a dry season stretching from November until May. Where basal deposits have been exposed, as at Non Dua, Ban Chiang Hian, Ban Prasat, Noen U-Loke, Muang Fa Daet, and Non Yang, we find that initial settlement took place some time after 1000 BC (Chantaratiyakarn, 1984; Higham, 1977; Indrawooth et al., 1991; Nitta, 1991, 1992; Phommanodch, 1991; Wichakana, 1991). Indeed, no Neolithic settlement has yet been identified for the lower two-thirds of the plateau. Some of these sites, such as Noen U-Loke and Ban Prasat, incorporate a transition from the Bronze into the Iron Age, and we can trace how bronzes were increasingly adapted for display, while iron was employed for personal ornaments, billhooks, and hoes. Many of the exotic goods seen at Ban Don Ta Phet also penetrated the plateau, but never in such quantities. The purpose of the moats, which ringed settlements far larger than their Neolithic and Bronze Age predecessors, has been debated widely, with defense, agriculture, and the provision of domestic water prominent as possible explanations. More recently, Parry (1992) has examined landsat images and demonstrated how some moats were meshed into a system of
34
Higham
canals and reservoirs. Surface surveys, as, for example, at Noen U-Loke, have traced such canals from local streams into the moats and then out to the surrounding countryside. Again, there is no evidence that the junction between the innermost moat and the settlement was fortified. It is highly likely that the moats were a device to retain and reticulate water, perhaps to supplement rainfall during the agricultural season. Any water left in the moats during the dry season would have helped sustain the populace within. These sites were also foci for the intensive exploitation of salt, and iron smelting furnaces are often encountered (Nitta, 1991, 1992). It is not difficult to hypothesize that moated sites were central places for an increasingly hierarchic social system in which emerging chiefs controlled salt and iron production and exchange in a new range of exotic ornaments. This can be tested by examining the mortuary record. Hitherto, few graves have been recovered, and no sample is big enough for testing the hypothesis. One point, however, is consistent: there is, as yet, little evidence for the application of iron to weapons of war. Moreover, site surveys in the Phimai region have shown that moated sites are very common and are distributed only a few kilometers apart along the margins of the Mun floodplain and up the main tributaries (Fig. 8) (McNeill and Welch, 1991). There is no compelling evidence in the upper Mun Valley for central places ringed by dependent settlements, although the situation might be different in the Chi Valley to the north, where sites like Muang Fa Daet and Ban Chiang Hian do seem to follow such a hierarchic pattern (Chantaratiyakarn, 1984; Indrawooth et aL, 1991). If militarism is not evidenced on the Khorat Plateau, it was of central importance on the Yunnan Plateau, Lingnan, and Bac Bo. Here we encounter societies not only directly exposed to contact with Chinese states, but also occupying constrained lacustrine or riverine floodplains. Chinese histories refer to these "southern barbarians" as being adept at guerilla war and skilled at casting bronze drums. Archaeologically, the best-documented groups are known as Dian, after Lake Dian on the Yunnan Plateau, and Dong Son, after the eponymous site on the southern bank of the Ma River in Bac Bo. The former has been documented on the basis of cemeteries located on low hills thrusting up from the lacustrine plain, of which Shizhaishan (Stone Fortress Hill) is best known (Murowchick, 1989; Pirazzoli-t'Serstevens, 1974; YPM, 1956). It incorporated a royal necropolis, with graves up to 5 m long and nearly 3 m deep that contained lacquered and painted wooden coffins resting on similarly ornamented wooden planks. Burial 6 included a gold seat inscribed "seal of the King of Dian." The male interment was accompanied by numerous beads of turquoise, agate, and gold, a bronze mirror, jade rings, an iron-bladed sword with a bronze hilt and golden scabberd, and many bronze vessels and armaments.
The Archaeology of MainLand Southeast Asia
35
E
E
0
D
~
36
Higham
Cowrie containers, some in the form of drums, are a particular feature of Dian grave goods, and they were embellished with scenes of ritual and warfare rendered in tiny cast bronze figures. These provide one of the clearest sources for prehistoric chiefly activities: we can watch a vicious skirmish involving a gilded aristocrat on his horse, in the act of decapitating an enemy. His followers are armed with crossbows and swords. We also witness scenes of feasting, torture, hunting, agricultural ceremonies, gift giving to the elite, even high status women being carried aloft on palanquins (Huang Ti and Wang Dadao, 1983). Weapons not only were cast in bronze, for iron was never abundant in Dian contexts, but also were given special treatment: male elite graves include superbly decorated bronze armor and horse trappings, while those for women, as has been shown by Yun Kuen Lee (1994), included bronze weaving implements. One of the scenes cast onto the top of a cowrie container shows a meeting of aristocrats on a raised dais ringed with bronze drums. These are a recurrent feature of many chiefdoms that developed across southern China and in Vietnam during the last few centuries BC, and they are particularly abundant in Dong Son contexts. Bac Bo ranks with the most productive rice growing regions, but it is constrained by uplands and the sea. The Red River flows a virtually straight course from its headwaters in Yunnan, providing a direct route for the exchange of goods, ideas, and the ingress of enemies (Taylor, 1991). The later prehistory of the area is marked by political instability, culminating in the Han seizure of 111 BC. Such threats might well account for the development of a warrior aristocracy, reflected in the wide range of weaponry found in graves and scenes of ritual and conflict on the great bronze drums (Pham Huy Thong et al., 1990). Several cemeteries have been investigated, and we find that the dead were interred in coffins fashioned from tree trunks. Burial 2 at Viet Khe is the most impressive yet identified, 4.76 m long and containing a rich assemblage of bronzes. These include sumptuary vessels, axes, spearheads, daggers, a sword, and knives of Chinese origin (VMH, 1965). No iron remains, however, were found, and the radiocarbon dates from coffin wood, which will probably have a reasonable inbuilt age, indicate a context from about 400 to 500 BC. Some of the bronzes are paralleled at the Warring States cemetery of Yinshanling in Guangxi (475-221 BC), where iron artifacts are abundant (GZAR, 1978; Murowchick, 1989). Similar iron implements reached the Dong Son area in due course, for examples have been recovered from the cemetery of Xuan La (Pham Quoc Quan and Trinh Can, 1982). The conditions for preservation within the wooden coffins has led to the recovery of an unusual set of grave goods, including wooden
The Archaeology of Mainland Southeast Asia
37
carvings of human figures, ax hafts, wooden trays, a lacquered box, and the remains of clothing. The chiefly society of Dong Son, along with that of Dian, saw the apogee of Southeast Asian bronze casting, involving specialist production to fulfill elite demands. These groups, and many others in Lingnan (AUard, 1994), operated in a volatile political environment and were, in due course, incorporated as commanderies of the Han empire. To the south, however, along the coastal plains of Vietnam, we encounter a further and quite distinct iron age culture, that of Sa Huynh. Blust (1993b) has demonstrated beyond reasonable doubt that linguistically, the Austronesian-speaking Chams of this region represent an intrusion on the mainland from Borneo. Few would deny that the Cham civilization rested on Sa Huynh foundations. The latter is recognized on the basis of extensive urnfield cemeteries, a signal departure from the widespread inhumation technique, and the grave goods include iron weaponry and exotic jewelry of glass, carnelian, zircon, agate, gold, and jade. One variety of ornament, the slit earring (linfling-o), and a double-headed animal pendant have distributions stretching from Central Thailand to the Philippines and Hong Kong, indicating the burgeoning maritime exchange contacts of the later first millennium BC. These, and the growing contribution of linguistics, link with the archaeological record in illuminating the later prehistoric foundations of early states in Southeast Asia. There is a deep division between the groups absorbed within the Han empire and those that developed in the context of increasing contact with Indian culture to the south. For eight centuries, the descendants of Dong Son underwent Sinicisation without losing their Austroasiatic language. The Chams adopted Hinduism and Sanskrit within a series of coastal polities where political centralization was focused on cult centers such as Mi Son and Dong Duong. The Mon-Khmer Austroasiatic speakers of the Mekong and Chao Phraya valleys also adopted, to their own ends, aspects of the religion, legal system, language, and architecture of India. There is a strong undercurrent of opportunism, rather than slavish copying, and Wolters (1979) among others, has noted how the emerging overlords were particularly enamored of the Hindu rituals that allowed them to claim divine status. As the cult centers of the Khmer and Dvaravati polities developed, so we encounter the same environmental constraints that were resolved in the Iron Age. The massive reservoirs of Angkor might have symbolized the oceans that ringed Mount Meru, home of the Hindu gods, but they also fulfilled the same purpose as the earlier moats that ringed Noen U-Loke and the myriad of other such sites distributed across the broad lands that flank the Mekong.
Higham
38
SUMMARY Recent research has clarified some issues and posed new ones. We now have the first remains of Homo erectus from mainland contexts and some examples of contemporary material culture. Further details have been added to the widespread inland forager tradition, widely known under the term Hoabinhian, and a long sequence beginning in the late Pleistocene has been given more precise definition at Lang Rongrien in southern Thailand and in the Son Vi industries of Bac Bo. The rapid rise in sea level and extensive inundation have obliterated corresponding coastal sites occupied before the establishment of raised beaches. From about 4000 BC, however, we know of many settlements that took advantage of the rich marine habitat. At least semisedentary, the occupants fashioned pottery vessels and polished stone implements, and although sometimes described in the literature as "coastal neolithic," no convincing evidence has yet been advanced for agriculture or the raising of domestic stock. The transition to rice cultivation has attracted much attention, and at present there are two principal alternatives. The first sees a wave of advance, originating in the Yangzi Valley. Linguistically, this may be reflected in the widespread distribution of Austroasiatic languages and cognates for rice and aspects of its cultivation. In terms of archaeology, there is no evidence that this expansion reached tropical Southeast Asia before the midthird millennium BC, although the analysis of pollen and phytoliths from sediment cores in Thailand has provided tantalizing evidence for an earlier presence of forest clearance that might indicate agricultural activity. The second hypothesis, which is not an alternative, but rather complementary, is that there was also a transition among coastal sedentary communities. This is currently best evidenced by comparing Nong Nor with Khok Phanom Di. Only 14 km apart, the former was a small base for coastal foraging, while the latter was occupied for half a millennium, during which time cultivated rice was part of the diet. Yet the material culture of Nong Nor clearly anticipates that of Khok Phanom Di. If, as seems possible at present, agricultural communities were spreading into Southeast Asia only by the early third millennium BC, then little time was involved before copper and tin came to be exploited. Again, there are two alternatives: exchange contact, which ultimately reached Sichuan and the zhongyuan might have brought a knowledge of the properties of metal ore to Southeast Asia. On the other hand, it is possible, given the presence of a copper industry without any evidence for alloying in Central Thailand, that there was an indigenous development of metallurgy. The consensus of available evidence is compatible with an inception of the Bronze Age between 1500 and 1000 BC, possibly a century or two earlier.
The Archaeology of Mainland Southeast Asia
39
The tradition of metalworking is in sharp contrast, in terms of techniques and finished products, with that of central China. For the millennium in question, there is no evidence to suggest that bronze was used by social elites to advance their standing: the relevant cemeteries, rather, show that bronze took its place alongside other exotic substances, such as slate, marble, and marine shell, in small segmentary communities lacking a tradition of ascribed rank. This situation, which is similar to that documented in the Near East, Iberia, Greece, and temperate Europe, encountered a major dislocation with the combined effects of the availability of iron, the opening of exchange with India and the islands of Southeast Asia, and the growing exposure in the northern marches to Chinese expansion. Each region provides its own course of development: marked militarism associated with the rise of a chiefly aristocracy in Yunnan and Bac Bo, a growth in community size and the imperative to cope with aridity in the Khorat Plateau and Tonle Sap plains, and the rush to exploit the new products of India in the Chao Phraya Valley. Meanwhile, groups of Austronesian speakers at last found a foothold on the mainland, infiltrating and settling the coastal plains of Vietnam and the Malaysian peninsula. The next few years of research will provide more information, and hopefully some alternatives will be illuminated. What we need more than anything at present is an accumulating body of accurately dated and fully published sites. The rarity of final reports contrasts sadly with the plethora of test excavations and interim claims. Hutterer (1982, p. 567) rightly encouraged his colleagues in Southeast Asia to seek broader horizons, to adopt a comparative approach that, in his words, would "move Southeast Asia into a meaningful and important place within the history and development of mankind." Some steps have been taken in this direction, but it is a long and winding road.
REFERENCES CITED z Allard, F. (1994). Interaction and social complexity in Lingnan during the first millennium B.C. Asian Perspectives 33(2): 309-326. Anderson, D.D. (1990). Lang Rongrien Rockshelter: A Pleistocene, Early Hotocene Archaeological Site from Krab~ Southwestern Thailand, University Museum, Philadelphia. 2Abbreviations used: EATIA, Erlitou Archaeological Team, Institute of Archaeology; CASS, I~FEO, F;cole Franqaise d'Extr6me Orient; GXBWG, Museum of the Guangxi Zhuang Autonomous Region; GZAR, Guangxi Zhuangsu Autonomous Region; HAI, Hunan Archaeological Institute; SMRPC, Shanghai Municipal Relics Preservation Committee; VMH, Vietnam Museum of History; YPM, Yunnan Provincial Museum; ZPM, Zhejiang Provincial Museum.
40
Higham
Anonymous (1991). List of radiocarbon dates. Khao Co Hoc 75:78-79 (in Vietnamese). Bagley, R. W. (1988). Sacrificial pits of the Shang period at Sanxingdui in Guanghan county, Sichuan. Arts Asiatiques XLIII: 78-86. Bar-Yosef, O. (1995). The transition to agriculture in the Levant. In Price, T. D., and Gebauer, A. B. (eds.), Last Hunters-First Farmers: New Perspectives on the Prehistoric Transition to Agriculture, American School of Prehistoric Research, Santa Fe, NM (in press). Bayard, D. T. (1972). Early Thai bronze: Analysis and new dates. Science 176: 1411-1412. Bayard, D. T. (1984). A tentative regional phase chronology for Northeast Thailand. In Bayard, D. T. (ed.), SoutheastAsian Archaeology at the XV Pacific Science Congress, Otago University Studies in Prehistoric Anthropology 16, Dunedin, pp. 161-168. Bayard, D. T. (1992). Models, scenarios, variables and suppositions: Approaches to the rise of social complexity in Mainland Southeast Asia, 700 BC-500 AD. In Glover, I. C., Suchitta, P., and Villiers, J. (eds.), Early Metallurgy, Trade and Urban Centers in Thailand and Southeast Asia, White Lotus, Bangkok, pp. 13-38. Bellwood, P. (1985). Prehistory of the Indo-Malyasian Archipelago, Academic Press, New York. Bellwood, P. (1989). The colonisation of the Pacific: Some current hypotheses. In Hill, A. V. S., and Serjeantson, S. W. (eds.), The Cotonisation of the Pacific. A Genetic Trail, Clarendon Press, Oxford, pp. 1-59. Bellwood, P. (1992). Southeast Asia before history. In Tarling, N. (ed.), The Cambridge History of Southeast Asia, Cambridge University Press, Cambridge, pp. 55-136. Benedict, P. K. (1942). Thai, Kadai and Indonesian: A new alignment in Southeastern Asia. American Anthropologist 44: 576-601. Benedict, P. K. (1975). Austro-Thai Language and Culture, HRAF Press, New Haven, CT. Bennett, A. (1988). Prehistoric copper smelting in Central Thailand. In Charoenwongsa, P., and Bronson, B. (eds.), Prehistoric Studies: The Stone and Metal Ages in Thailand, Thai Antiquity Working Group, Bangkok, pp. 125-135. Bennett, A. (1989). The contribution of metallurgical studies to Southeast Asian archaeology. World Archaeology 20: 329-351. Blust, R. (1976). Austronesian culture history: Some linguistic inferences and their relationships to the archaeological record. WorldArchaeology 8: 19--43. Blust, R. (1985). The Austronesian homeland. A linguistic perspective. Asian Perspectives 26: 107-117. Blust, R. (1993a). The Austronesian settlement of Mainland Southeast Asia. Unpublished paper read at the second annual meeting of the Southeast Asian Linguistic Society. Blust, R. (1993b). Beyond the Austronesian homeland: The Austric hypothesis and its implications for archaeology. Paper read at a meeting on Austroasiatic languages, Philadelphia. Bowdler, S. (1993). The earliest voyages: The Pleistocene colonisation of Southeast Asia and Australasia. Unpublished paper read at the 34th Congress of Asian and African Studies, Hong Kong, Aug. 1993. Boyd, W. (1994). The palaeoenvironment of the archaeological site Nong Nor, and the implications of a palaeoenvironmental model for the regional archaeology. In Higham, C. F. W., Thosarat, R., Boyd, W., Chang, N., Debreceny, J., Higham, T. F. G., Hogg, A., Hunt, V., Mason, G. M., O'Reilty, D., Pailtes, C., and Tayles, N. (eds.), The Excavation of the Prehistoric Site of Nong Nor, Central Thailand, Report to the National Research Council of Thailand. Bronson, B. (1992). Patterns in early Southeast Asian metals trade. In Glover, I. C., Suchitta, P., and Villiers, J. (eds.), Early Metallurgy, Trade and Urban Centers in Thailand and Southeast Asia, White Lotus, Bangkok, pp. 63-114. Bui Vinh (1991). The Da But culture in the Stone Age of Viet Nam. Bulletin of the Indo-Pacific Prehistory Association 10: 127-131. Buranrak and Co. (1994). Report on the Excavation of Non Praw, Fine Arts Department, Bangkok (in Thai). Chang, K.-C. (1986). The Archaeology of Ancient China, Yale University Press, New Haven, CT.
The Archaeology of Mainland Southeast Asia
41
Chantaratiyakarn, P. (1984). The research programme in the Middle Chi. In Higham, C. F. W., and Kijngam, A. (eds.), Prehistoric Investigations in Northeast Thailand, British Archaeological Reports (International Series), Oxford, pp. 63-114. Charoenwongsa, P. (1988). The current status of prehistoric research in Thailand. In Charoenwongsa, P., and Bronson, B. (eds.), Prehistoric Studies: The Stone and Metal Ages in Thailand, Thai Antiquity Working Group, Bangkok, pp. 17-42. Chiawanwongsa, P. (1987). The Archaeological Site of Ban Wang Hi, Fine Arts Department, Bangkok (in Thai). Chiu, S. T. (1993). Recent excavations at Yung Long. Unpublished paper read at the 34th International Conference of Asian and North African Studies, Hong Kong, Sept. 1993. Ciochon, R. L., and Olsen, J. W. (1987). Expedition to Vietnam. The search for ancient cultures, early humans and giant apes in the Asian Pleistocene record. Anthroquest 28: 12-18. Clark, J. G. D. (1961). World Prehistory--An Outline, Cambridge University Press, Cambridge. Dao Linh Con and Nguyen Duy Ty (1993). Dia Dien Khao Co Hoc Doc Chua, Nha Xuat Ban Khoa Hoe Xo Hoi, Ha Noi. EATIA (1976). The bronzes and jades recently discovered at Erlitou in Yanshi County, Henan. Kaogu 1976:259-263 (in Chinese). Finn, D. J. (1936). Archaeological finds on Lamma Island near Hong Kong, Part XI. The Hong Kong Naturalist XI: 37-60. Finn, D. J. (1958). Archaeological Finds on Lamina Island near Hong Kong Ricci, Hong Kong. Glover, I. C. (1989). Early Trade Between lndia and Southeast Asia: A Link in the Development of a World Trading System, University of Hutl Center for Southeast Asian Studies Occasional Paper No. 16, Hull. Glover, I. C. (1990). Ban Don Ta Phet: The 1984-5 excavation. In Glover, I. C., and Glover, E. (eds.), Southeast Asian Archaeology 1986, Proceedings of the first conference of the Association of Southeast Asian Archaeologists in Western Europe, British Archaeological Reports, International Series 561, Oxford, pp. 139-183. Glover, I. C. (1991). The Bronze-Iron Age in Western Thailand. Paper presented at the conference The High Bronze Age of Southeast Asia and South China, Hua Hin, Thailand, Jan. Gorman, C. F. (1971). The Hoabinhian and after: Subsistence patterns in Southeast Asia during the late Pleistocene and early recent periods. Worm Archaeology 2(3): 300-320. Gorman, C. F. (1972). Excavations at Spirit Cave, North Thailand: Some interim impressions. Asian Perspectives 13: 79-107. Gorman, C. F., and Charoenwongsa, P. (1976). Ban Chiang: A mosaic of impressions from the first two years. Expedition 8(4): 14-26. Grant, A., and Higham, C. F. W. (1991). The large mammalian fauna. In Higham, C. F. W., and Bannanurag, R. (eds.), The Excavation of Khok Phanom Di, VoL 2. The Biological Remains (Part 1), Society of Antiquaries of London, Research Report No. XLVIII, London, pp. 147-191. Groslier, B. P. (1979). La cit6 hydraulique Angkorienne. Exploitation ou surexploitation du sol? Bulletin de Franfaise d'Extr~me Orient 66: 161-202. Guangren, G., and Wangping, S. (1981). A preliminary study of the pottery gui tripod of the prehistoric period. Kaogu Xuebao 1981(4): 427-460 (in Chinese). GXBWG (1984). Pre-Qin bronzes recovered from Guangxi in recent years. Kaogu 1984: 798-806 (in Chinese). GXBWG (1988). Short excavation report on the burials of Yuanlongpo, Matou, Wuming County, Guangxi. Wenwu 1988:1-13 (in Chinese). GZAR (1978). Warring states burials at Yinshanling, Pingle. Kaogu Xuebao 1978(2): 211-258 (in Chinese). HAI (1990). Excavation of an early Neolithic site at Pengtoushan in Lixian, Hunan. Wenwu 8:17-32 (in Chinese). Ha Van Phung (1993). Stages of development of Go Mun culture. Khao Co Hoc 85:48-63 (in Vietnamese).
42
Higham
Ha Van Phung and Nguyen Duy Ti (1982). Di Chi Khao Co Hoc Go Mun, Nha Xuat Ban Khoa Hoe Xa Hoi, Hanoi. Ha Van Tan (1980). Nouvelles recherches pr6historiques et protohistoriques au Vietnam. Bulletin de l'Ecole Frangaise d'Extr~me Orient 68: 113-154. Ha Van Tan (1991). From pre-Dong Son to Dong Son: Sociocultural changes. Unpublished paper read at the Conference on the High Bronze Age, Hua Hin, Thailand. Ha Van Tan (1993). Yazhang plaques in the Phung Nguyen culture sites. Khao Co Hoc 86: 16-27 (in Vietnamese). He Jiejun (1986). The Neolithic culture of the Lake Dongting area. Kaogu Xuebao 1986(4): 385-408 (in Chinese). Higham, C. F. W. (1977). The prehistory of the southern Khorat Plateau, with particular reference to Roi Et Province. Modem Quaternary Research in Southeast Asia 3: 103-142. Higham, C. F. W. (1983). The Ban Chiang culture in wider perspective. Proceedings of the British Academy LXLX: 229-261. Higham, C. F. W. (1988). Prehistoric metallurgy in Southeast Asia: Some new information from the excavation of Ban Na Di. In Maddin, R., (ed.), The Beginning of the Use of Metals and Alloys, Massachusetts Institute of Technology Press, Cambridge, Mass. pp. 130-155. Higham, C. F. W. (1989). The Archaeology of Mainland Southeast Asia, Cambridge University Press, Cambridge. Higham, C. F. W. (1994). Chronometric hygiene and the Bronze Age of Southeast Asia. Unpublished paper read at the Congress of Indo-Pacific Prehistory Association, Chiang Mai, Jan. Higham, C. F. W. (1995a). The social and chronological contexts of early bronze working in Southeast Asia. In Bulbeck, F. D. (ed.), Ancient Chinese and Southeast Asian Bronze Age Cultures, Australian National University Press, Canberra (in press). Higham, C. F. W. (1995b). The transition to rice cultivation in Southeast Asia. In Price, T. D., and Gebauer, A. B. (eds.), Last Hunters-First Farmers: New Perspecives on the Prehistoric Transition to Agriculture, American School of Prehistoric Research, Santa Fe, NM (in press). Higham, C. F. W., and Kijngam, A. (eds.) (1984). Prehistoric Investigations in Northeast Thailand, British Archaeological Reports (International Series) 231(1-3), Oxford. Higham, C. F. W., and Thosarat, R. (1994). Khok Phanom Di. Prehistoric Adaptation to the World's Richest Habitat, Harcourt Brace, Fort Worth, TX. Higham, T. F. G. (1993). The shell knives. In Higham, C. F. W., and Thosarat, R. (eds.), The Excavation of Khok Phanom Di. Vol. 3. The Material Culture (Part 1), Society of Antiquaries of London, Research Report No. L, London, pp. 179-212. Ho, C.-M. (1992). An analysis of settlement patterns in the Lopburi area. In Glover, I. C., Suchitta, P., and Villiers, J. (eds.), Early Metallurgy, Trade and Urban Centers in Thailand and Southeast Asia, White Lotus, Bangkok, pp. 39-46. Hoang Xuan Chinh (1968). Bao Cao Khai Quat Dot Di Chi Lung Hoa, Nha Xuat Ban Khoa Hoc Xa Hoi, Ha Noi. Huang Ti and Wang Dadao (1983). Commentary on the plates. In Rawson, J. (ed.), The Chinese Bronzes of Yunnan, Sidgewick and Jackson and the Cultural Relics, London and Beijing, pp. 217-241. Huang Zhanyue (1986). Pre Qin bronzes unearthed in Guangdong and Guangxi. Kaogu Xuebao 1986:433 (in Chinese). Hutterer, K. L. (1982). Early Southeast Asia: Old wine in new skins? Journal of Asian Studies XLI: 559-570. Hutterer, K. (1991). The High Bronze Age in Southeast Asia: Introductory remarks. Unpublished paper read at the Conference on the High Bronze Age, Hua Hin, Thailand. Indrawooth, P., Krabuansang, S., and Narkwake, P. (1991). Muang Fa Daed Song Yang: New archaeological discoveries. In Universit6 Silpakon (ed.), Recentes Recherches en Archdologie en Thailande: Duexieme Symposium Franco-Thai, Silpakon University, Bangkok, pp. 98-111.
The Archaeology of Mainland Southeast Asia
43
Kealhofer, L. (1992). Holocene environmental sequences in Central Thailand: Preliminary phytolith data. Paper read at the 4th International Conference of the European Association of Southeast Asian Archaeologists, Rome, Sept.-Oct. 4. Kealhofer, L., and Piperno, D. (1995). Two phytolith sequences from the Bang Pakong Valley. In Higham, C. F. W., and Thosarat, R. (eds.), The Excavation of Khok Phanom Di. Vol. 4. Subsistence and Environment: The Biological Evidence (Part II), Society of Antiquaries of London, Research Report, London, pp. 226-238. Liang Jingjin (1978). Bronze artifacts recovered in Guangxi. Wenwu 10:93-96 (in Chinese). Li Guo (1994). Neolithic sand bar sites round the Pearl River Estuary area and Hainan: A comparative study. Unpublished paper read at the Congress of Indo-Pacific Prehistory Association, Chiang Mai, Jan. Li Wenjie (1986). On the classification and periodization of the Daxi Culture. Kaogu Xuebao 1986(2): 131-152 (in Chinese). Li Yan (1995). A preliminary study on problems relating to early Bronze Age finds in Guangdong: Begin with stone molds unearthed from Tangxiahuan in Zhuhai. In Yueng Chung-tong and Li Wai-ling, W. (eds.), Archaeology in Southeast Asia, University Museum and Art Gallery and the University of Hong Kong, pp. 87-94. Maddin, R., and Weng, Y. Q. (1984). The analysis of bronze wire. In Higham, C. F. W., and Kijngam, A. (eds.), Prehistoric Investigations in Northeast Thailand, British Archaeological Reports (International Series), Oxford, pp. 112-116. Maloney, B. K. (1991). Palaeoenvironments of Khok Phanom Di: The pollen, pteridophyte spore and microscopic charcoal record. In Higham, C. F. W., and Bannanurag, R. (eds.), The Excavation of Khok Phanom Di, Vol. 2. The Biological Remains (Part 1), Society of Antiquaries of London, Research Report No. XLVIII, London, pp. 6-140. Maloney, B. tC, Higham, C. F. W., and Bannanurag, R. (1989). Early rice in Southeast Asia, palynological and archaeological evidence from Thailand. Antiquity 63: 363-370. Mankong, S. (1989). Noen Ma Kok, Fine Arts Department, Bangkok (in Thai). Maw, B. (1993). The first discovery of an early man's fossilized maxillar bone fragment in Myanmar paleoanthropology. The East Asian Tertiary~QuaternaryNewsletter 16: 72. McNeill, J. R., and Welch, D. J. (1991). Regional and interregional interaction on the Khorat Plateau. Bulletin of the lndo-Pacific Prehistory Association 10: 327-340. Meacham, W. (1975). Schofield and Shek Pik in retrospect. In Meacham, W. (ed.), An Archaeological Site at Shek Pik, Journal Monograph 1, Hong Kong Archaeological Society, Hong Kong, pp. v-ix. Meacham, W. (1993). New C14 dates and advances in establishing a precise chronology for Hong Kong's prehistory. Journal of the Hong Kong Archaeological Society XIII: (in press). Meacham, W. (1994). On the dating of painted pottery in Hong Kong. Unpublished paper read at the Conference on Painted Pottery, Chinese University of Hong Kong, Feb. Monkhonkamnuanket, N. (1992). Ban Prasat. An Archaeological Site, Fine Arts Department, Bangkok (in Thai). Muhly, J. (1988). The beginnings of metallurgy in the Old World. In Maddin, R. (ed.), The Beginnings of the Use of Metals and Alloys, MIT Press, Cambridge, MA, pp. 2-20. Murowchick, R. E. (1989). The Ancient Bronze Metallurgy of Yunnan and Its Environs: Development and Implications, Ph.D. thesis, Harvard University, Cambridge, MA. Natapintu, S. (1988a). Ban Lure Khao, Fine Arts Department, Bangkok (in Thai). Natapintu, S. (1988b). Current research on ancient copper-base metallurgy in Thailand. In Charoenwongsa, P., and Bronson, B., (eds.), Prehistoric Studies: The Stone and Metal Ages in Thailand, Thai Antiquity Working Group, Bangkok, pp. 107-124. Natapintu, S. (1991). Archaeometallurgical studies in the Khao Wong Prachan Valley, Central Thailand. Bulletin of the Indo-Pacific Prehistory Association 11: 153-159. Nguyen Ba Khoach (1980). Phung Nguyen. Asian Perspectives 23(1): 23-54. Nguyen Kim Dung (1990). The lithic workshop site at Trang Kenh. Khao Co Hoc 74:64--82 (in Vietnamese). Nguyen Phuc Long (1975). Les nouvetles recherches arch6ologiques au Vietnam. Arts Asiatiques XXXI: 1-154.
44
Higham
Nitta, E. (1991). Archaeological study on the ancient iron-smelting and salt-making industries in the northeast of Thailand. Preliminary report on the excavations of Non Yang and Ban Don Phlong. Journal of Southeast Asian Archaeology 11: 1--46. Nitta, E. (1992). Ancient industries, ecosystem and environment. Historical Science Reports, Kagoshima University 39: 61-80. O'Reilty, D. (1994). The pottery and bone artifacts from Nong Nor. In Higham, C. F. W., Thosarat, R., Boyd, W., Chang, N., Debreceny, J., Higham, T. F. G., Hogg, A., Hunt, V., Mason, G. M., O'Reilly, D., Pailles, C., and Tayles, N. (eds.), The Excavation of the Prehistoric Site of Nong Nor, Central Thailand, Report to the National Research Council of Thailand. Parry, J. T. (1992). The investigative role of Landsat-TM in the examination of pre- and proto-historic water management sites in Northeast Thailand. Geocarto International 4: 5-24. Penny, J. (1984). Fish in the water and rice in the paddy: Contributions to studies of the Southeast Asian Iron Age. In Bayard, D. T. (ed.), Southeast Asian Archaeology at the XV Pacific Science Congress, Otago University Studies in Prehistoric Anthropology 16, Dunedin, pp. 152-160. Pham Huy Thong, Pham Minh Huyen, and Lai Van Toi (eds.) (1990). Dong Son Drums in I.qet Nam, The Viet Nam Social Science Publishing House, Ha Noi. Pham Quoc Quan and Trinh Can (1982). The pirogue-coffins at Xuan La (Ha Son Binh Province). Khao Co Hoc 44:36-50 (in Vietnamese). Phommanodch, S. (1991). The past of Thung Samrit at Ban Prasat. Silapakon Journal 34: 6-21 (in Thai). Pigott, V. (1992). The archaeology of copper production at prehistoric Non Pa Wai and Nil Kham Haeng in Central Thailand. Paper read at the 4th International Conference of the European Association of Southeast Asian Archaeologists, Rome, Sept. 28-Oct. 4. Pigott, V. C., and Natapintu, S. (1988). Archaeological investigations into prehistoric copper production: The Thailand Archaeometallurgy Project 1984-6. In Maddin, R. (ed.), The Beginnings of the Use of Metals and Alloys, MIT Press, Cambridge, MA, pp. 156-162. Pirazzoli-t'Serstevens, M. (1974). La Civilization du Royaume de Dian ~ l'Epoque Han, ]~FEO, Paris. Pisnupong, P. (1993). The Industrial Stone Technology. In Higham, C. F. W., and Thosarat, R. (eds.), The Excavation of Khok Phanom Di, VoL 3. The Material Culture (Part 1), Society of Antiquaries of London, Research Report No. L, London, pp. 45-104. Pope, G. (1985). Evidence for early Pleistocene hominid activity from Lampang, northern Thailand. Bulletin of the lndo-Pacific PrehistoryAssociation 6: 2-9. Pope, G., Nakabanlang, S., and Pitragool, S. (1987). Le Pal6olithique du nord de la Thailande. D6couvertes et perspectives nouv611es.L'Anthropologie 91: 749-754. Reid, L. A. (1993). Morphological evidence for Austric. Paper presented at the Conference on Austronesian-Mainland Southeast Asian, Honolulu. Renfrew, C. (1987). Archaeology and Language: The Puzzle of Indo-European Origins, Jonathan Cape, London. Rogers, P. R., Leininger, N. W., Mirchandani, S., van den Bergh, J., and Widdowson, E. A. (1995). Tung Wan Tsai: A Bronze Age and Han Period Coastal Site, Antiquities and Monuments Office, Hong Kong, Occasional Paper No. 3. Shoocondej, R. (1994). Working toward an anthropological perspective on Thai prehistory: Current research on the post-Pleistocene. Unpublished paper read at the Congress of the Indo-Pacific Prehistory Association, Chiang Mai, January 1994. Shutler, R. (1995) Hominid cultural evolution as seen from the archaeological evidence in Southeast Asia. In Yueng Chung-tong and Li Wai-ling, W. (eds.), Archaeology in Southeast Asia, The University Museum and Art Gallery and the University of Hong Kong, pp. 49-56. SMRPC (1980). The second season of excavation at Songze site in Qingpu County, Shanghai. Kaogu Xuebao 1980(1): 29-58 (in Chinese). Sotheim, W. G., II (1968). Early bronze in Northeastern Thailand. CurrentAnthropology 9(1): 59-62.
The Archaeology of Mainland Southeast Asia
45
Sorensen, P. (1979). The Ongbah cave and its fifth drum. In Smith, R. B., and Watson, W., (eds.), Early South East Asia, Oxford University Press, Oxford, pp. 78-97. SCrensen, P., and Hatting, T. (1967). Archaeological Investigations in Thailand, Vol. 11. Ban Kao, Part 1: The Archaeological Materials from the Burials, Munksgard, Copenhagen. Spriggs, M. (1989). The dating of the island Southeast Asian Neolithic: An attempt at chronometric hygiene and linguistic correlation. Antiquity 63: 587-613. Tang Chung (ed.) (1994). Ancient Cultures of South China and Neighbouring Regions, Chinese University of Hong Kong, Hong Kong. Taylor, K. (1991). Encountering Southeast Asia in Shui Ching Chu. Unpublished paper read at the Conference on the High Bronze Age, Hua Hin, Thailand. Thompson, G. B. (1992).Archaeobotanical Investigations at Khok Phanom Di, Central Thailand, Ph.D. thesis, Australian National University. Trinh Sinh (1990). Spectrographic analysis of bronze artifacts in Go Mun and Dong Son cultures. Khao Co Hoc 75:49-59 (in Vietnamese). VMH (1965). Ngoi Mo Co l~et-Khe, Xuet Ban, Hanoi (in Vietnamese). von Dewall, M. (1979). Local workshop centers of the Late Bronze Age in highland South East Asia. In Smith, R. B., and Watson, W. (eds.), Early South East Asia, Oxford University Press, Oxford, pp. 31-49. Watt, J. C. Y. (1968). Archaeological finds in Hong Kong and their cultural implications. Journal of the Hong Kong Archaeological Society 1: 10-18. Weiss, A. D. (1992). The social context of copper production in Central Thailand: Evidence from mortuary and industrial data. Paper read at the 4th International Conference of the European Association of Southeast Asian Archaeologists, Rome, Sept. 28-Oct. 4. Welch, D. (1985). Adaptation to Environmental Unpredictability: Intensive Agriculture and Regional Exchange at Late Prehistoric Centers in the Phimai Region, Ph.D. thesis, University of Hawaii (in Thai). Wichakana, M. (1991). Prehistoric sacrifices at Noen U-Loke. Muang Boran 16(4): 69-79 (in Thai). Wilen, R. N. (1989). Excavations at Non Pa Kluay, Northeast Thailand, British Archaeological Reports, International Series 517, Oxford. Wolters, O. W. (1979). Khmer Hinduism in the seventh century. In Smith, R. B., and Watson, W. (eds.), Early South East Asia, Oxford University Press, Oxford, pp. 427-442. Yan Wenming (1991). China's earliest rice agriculture remains. Bulletin of the Indo-Pacific Prehistory Association I0: 118-126. You Rujie (1976). A few observations on the rice grains and bone 'si" unearthed from the fourth layer of the Hemudu site. Wenwu 8:20-23 (in Chinese). YPM (1956). The ancient site and cemetery at Shizhaishan, Jinning, Yunnan. Kaogu Xuebao 1956:43-63 (in Chinese). YPM (1977). The Neolithic site of Ta-Tun-Tzu in Yuan-Mo County, Yunnan Province. Kaogu Xuebao 1977:43-72 (in Chinese). YPM (1981). The Baiyangcun site at Binchuan County, Yunnan Province. Kaogu Xuebao 1981: 349-368 (in Chinese). Yun Kuen Lee (1994). Material representations of status in the Dian culture. Unpublished paper presented to the 15th Congress of the Indo-Pacific Prehistory Association, Chiang Mai, Jan. Zhou, S. Z., Chen, F., Pan, B. T., Cao, J. X., Li, J. J., and Derbyshire, E. (1991). Environmental change during the Holocene in Western China on a millennial timescale. The Holocene 1: 151-156. Zhou Jiwei (1981). Report of the examination of ancient rice from the middle and lower Yangzi. Yunnan Nongye Xuexi 1981(1): 1-6 (in Chinese). Zhu Feisu (1984). Several problems related to the archaeology of Neolithic Guangdong. In Lain, P. Y. K. (ed.), Archaeological Finds from Pre-Qin sites in Guangdon~ Hong Kong Chinese University Press, Hong Kong, pp. 30-42. Zide, A. R. K., and Zide, N. H. (1976). Proto-Munda cultural vocabulary: Evidence for early agriculture. In Jenner, P. N., Thompson, L. C., and Starosta, S. (eds.), Austro-Asiatic Studies Part II, Oceanic Linguistics Special Publication 13, Honolulu, pp. 1295-1334.
46
Higham
ZPM (1978a). Excavations (first season) at Ho-Mu-Tu in Yus-Yao County, Chekiang Province. Kaogu Xuebao 1978(1): 39-94 (in Chinese). ZPM (1978b). A study of the animal and plant remains unearthed at Ho-Mu-Tu. Kaogu Xuebao 1978:85-108 (in Chinese).
BIBLIOGRAPHY OF RECENT LITERATURE Aitken, J. J. (1992). Archaeological Sediments as Artifacts, Ph.D. thesis, University of Otago, Otago. Aranyanak, C. (1991). Archaeological Textiles in Thailand, Fine Arts Department, Bangkok (in Thai). Ardika, I. W. (1991). Archaeological Research in Northeastern Bali, Indonesia, Ph.D. thesis, Australian National University, Canberra. Bannanurag, R. (1988). Evidence for ancient woodworking: A microwear study of Hoabinhian tools. In Charoenwongsa, P., and Bronson, B. (eds.), Prehistoric Studies: The Stone and Metal Ages of Thailand, Thai Antiquity Working Group, Bangkok, pp. 61-79. Bannanurag, R., and Bamrungwongse, A. (1991). A site survey of Ban Chiang culture sites in the Sakon Nakhon Basin, Northeast Thailand. Silpakon Journal 34:38-60 (in Thai). Barnard, N. (1995). The entry of cire-perdue investment casting, and certain other metallurgical techniques, (mainly metal-working) into South China and their progress northwards. In Bulbeck, F. D. (ed.), Ancient Chinese and Southeast Asian Bronze Age Cultures, Australian National University Press, Canberra (in press). Bellwood, P. (1991). Foraging towards farming; A decisive transition or a millennial blur? The Review of Archaeology 11: 14-24. Bellwood, P. (1993). Cultural and biological differentiation in peninsular Malaysia: The last 10,000 years. Asian Perspectives 32: 37-60. Bellwood, P. (1995). Early agriculture, language history and the archaeological record in China and Southeast Asia. In Yeung, Chung-tong, and Li Wai-ling, W. (eds.), Archaeology in Southeast Asia, University Museum and Art Gallery Hong Kong and University of Hong Kong, pp. 11-22. BeUwood, P., Gillespie, R., Thompson, G. B., Vogel, J. S., Ardika, I. W., and Datan, I. (1992). New dates for prehistoric Asian rice. Asian Perspectives 31: 161-170. Bennett, A., and Gtover, I. (1992). Decorated high-tin bronze bowls from Thailand's prehistory. In Glover, I. C. (ed.), SoutheastAsian Archaeology, 1990, Center for South-East Asian Studies, University of Hull, Hull, UK, pp. 187-208. Bernet Kempers, A. J. (1988). The kettledrums of Southeast Asia. Modern Quaternary Studies in Southeast Asia 10: 1-599. Birth Da Manh (1993). Research summary. Southeast Asian Archaeology International Newsletter 3: 10-11. Boyd, W. (1994). The palaeoenvironment of the archaeological site Nong Nor, and the implications of a palaeoenvironmental model for the regional archaeology. In Higham, C. F. W., Thosarat, R., Boyd, W., Chang, N., Debreceny, J., Higham, T. F. G., Hogg, A., Hunt, V., Mason, G. M., O'Reilly, D., Paitles, C., and Tayles, N. (eds.), The Excavation of the Prehistoric Site of Nong Nor, Central Thailand, Report to the National Research Council of Thailand. Bronson, B. (1992). Patterns in early Southeast Asian metals trade. In Glover, I. C., Suchitta, P., and Villiers, J. (eds.), Early Metallurgy, Trade and Urban Centers in Thailand and Southeast Asia, White Lotus, Bangkok, pp. 63-114. Brouson, B., and Natapintu, S. (1988). Don Noi: A new flaked tool industry of the middle Holocene in Western Thailand. In Charoenwongsa, P., and Bronson, B. (eds.), Prehistoric Studies: The Stone and Metal Ages in Thailand, Thai Antiquity Working Group, Bangkok, pp. 91-106.
The Archaeology of Mainland Southeast Asia
47
Bronson, B., and White, J. (1992). Radiocarbon and chronology in Southeast Asia. In Ehrich, R. W. (ed.), Old Worm Archaeology, Part 1, Vol. 1, Chicago University Press, Chicago, pp. 491-5t5. Bui Van Liem and Pham Quoc Quan (1991). Boat coffins burial ground at Minh Duc (Son Binh Province). Khao Co Hoc 76:56-63 (in Vietnamese). Bui Vinh (1991). The Da But culture in the Stone Age of Viet Nam. Bulletin of the lndo-Pacific Prehistory Association 10: 127-131. Buranrak and Co. (1994). Report on the Excavation of Don I~2ang, Fine Arts Department, Bangkok (in Thai). Chao-Mei Lien (1991). The Neolithic archaeology of Taiwan and the Peinan excavations. Bulletin of the Indo-Pacific Prehistory Association 11: 339-352. Choosiri, P. (1992). The Human Remains from the Excavation of Khok Phanom D~ Fine Arts Department, Bangkok (in Thai). Ciarla, R. (1992). The Thai-Italian Lopburi regional archaeological project, a preliminary report. In Glover, I. C. (ed.), Southeast Asian Archaeology, 1990, Center for South-East Asian Studies, University of Hull, Hull, UK, pp. 11t-128. Coote, V. (1990). Ancient metal mines and mining in Southwest Thailand. In Glover, I. C., and Glover, E. (eds.), Southeast Asian Archaeology 1986, BAR International Series 561, Oxford, pp. 131-137. Cremaschi, M., Ciarla, R., and Pigott, V. C. (1992). Paleoenvironment and late prehistoric sites in the Lopburi region of Central Thailand. In Glover, I. C. (ed.), Southeast Asian Archaeology, 1990, Center for South-East Asian Studies, University of Hull, Hull, UK, pp. 129-142. Diffloth, G. (1991). Austro-Asiatic languages. Encyclopaedia Britannica Macropaedia 22: 719-721. Earle, T. (1991). The evolution of chiefdoms. In Earle, T. (ed.), Chiefdoms: Power, Economy and Ideology, Cambridge University Press, Cambridge, pp. 1-15. Fine Arts Department (1990). Archaeological Sites in Thailand: VoL 1-K. The Southern Provinces of the Northeast, Fine Arts Department, Bangkok (in Thai). Fine Arts Department (1992). Archaeology of Pak Mun, Fine Arts Department, Bangkok (in Thai). Glover, I. C. (1991). The late prehistoric period in west-central Thailand. Bulletin of the lndo-Pacific Prehistory Association 10: 349-356. Glover, I. C., Suchitta, P., and ViUiers, J. (eds.) (1992). Early Metallurgy, Trade and Urban Centers in Thailand and Southeast Asia, White Lotus, Bangkok. Higham, C. F. W. (1989). The later prehistory of mainland Southeast Asia. Journal of World Prehistory 3: 235-282. Higham, C. F. W. (1989). The prehistoric site of Khok Phanom Di. Arts Asiatique XLIV: 25-43. Higham, C. F. W. (1995). Lingnan and Southeast Asia in Prehistory. In Yeung, Chung-tong, and Li Wai-ling, W. (eds.), Archaeology in Southeast Asia, University Museum and Art Gallery Hong Kong and the University of Hong Kong, pp. 23-40. Higham, C. F. W., and Bannanurag, R. (1990). The Excavation of Khok Phanom Di, a Prehistoric Site in Central Thailand, Vol. 1. The Excavation, Chronology and Human Burials, Society of Antiquaries of London, Research Report No. XLVII, London. Higham, C. F. W., and Bannanurag, R. (1990). The princess and the pots. New Scientist 126(1718): 50-56. Higham, C. F. W., and Bannanurag, R. (eds.) (1991). The Excavation ofKhok Phanom D~ a Prehistoric Site in Central Thailand, Vol. 11. The Biological Remains, Part 1, Society of Antiquaries of London, Research Report no. XLVIII, London. Higham, C. F. W., and Thosarat, R. (eds.) (1993). The Excavation of Khok Phanom Di, a Prehistoric Site in Central Thailand, VoL 3. The Material Culture, Part 1, Society of Antiquaries of London, Research Report No. L, London. Higham, C. F. W., Bannanurag, R., Mason, G., and Tayles, N. (1992). Human biology, environment and ritual at Khok Phanom Di. World Archaeology 24: 35-54.
48
Higham
Indrawooth, P., Kruabuansang, S., and Narkwake, P. (1990). Archaeological study of Ban Krabuang Nok. Spaafa Digest XI: 12-20. Junker, L. L. (1993). Southeast Asian archaeology: A review. Asian Perspectives 32: 251-253. Leong, S. (1990). A tripod pottery complex in Peninsular Malaysia. In Glover, I. C., and Glover, E. (eds.), Southeast Asian Archaeology 1986. Proceedings of the First Conference of the Association of Southeast Asian Archaeologists in Western Europe, British Archaeological Reports, International Series 561, Oxford, pp. 65-75. Leong Sau Heng (1991). Jendaram Hilir and the Mid-Holocene prehistory of the west coast plain of Peninsular Malaysia. Bulletin of the Indo-Pacific Prehistory Association 10: 150-160. Mason, G. M. (1991). The molluscan remains. In Higham, C. F. W., and Bannanurag, R. (eds.), The Excavation of Khok Phanorn Di. Vol. 2. The Biological Remains (Part 1), Society of Antiquaries of London, Research Report No. XLVIII, London, pp. 249-314. McConnell, J., and Glover, I. C. (1988-1989). A newly found bronze drum from Bali, Indonesia: Some technical considerations. Modern Quaternary Research in Southeast Asia I1: 1-38. Moore, E. (1989). Water management in early Cambodia: Evidence from aerial photography~ The Geographical Journal 155: 204-214. Mudar, K. M. (1993). Prehistoric and Early Historic Settlements on the Central Plain: Analysis of Archaeological Survey in Lopburi Province, Thailand, Ph.D. thesis, University of Michigan. Ngo Si Hong and Tran Quy Thinh (1991). Jar burials at Hau Xa, Hoi An Tower (Quang Nam-Da Nang provinces), and new knowledge of the Sa Huynh culture. Khao Co Hoc 78:64-75 (in Vietnamese). Nguyen Truong Ky (1991). An outline on pre-Dongsonian sites in the Ca River valley. Khao Co Hoc 76:48-55 (in Vietnamese). Pilditch, J. S. (1993). The personal ornaments. In Higham, C. F. W., and Thosarat, R. (eds.), The Excavation of Khok Phanom Di. Vol. 3. The Material Culture (Part 1), Society of Antiquaries of London, Research Report No. L, London, pp. 119-176. Pirazzoli-t'Serstevens, M. (1988). Les cultures du Sichuan occidental h la fin de l'Age du Bronze et leurs rapports avec les steppes, tn Minist~re des Affaires Etrangrres (ed.), L'Asie Centrate et ses Rapports avec Ies Civilisations Orientales, des Origines a L'Age du Fer, Boccard, Paris, pp. 183-196. Pookajorn, S. (1990). Hoabinhian cave excavations in Ban Kao district, West Thailand. In Glover, I. C., and Gtover, E. (eds.), Proceedings of the First Conference of the Association of Southeast Asian Archaeologists in Western Europe, London, 8-10 Sept. 1986, British Archaeological Reports, International Series 561, Oxford, pp. 11-28. Prishanchit, S. (1988). A preliminary survey of lithic industries in Mae Hong Son, Nan and Uttaradit, Northern Thailand. In Charoenwongsa, P., and Bronson, B. (eds.), Prehistoric Studies: The Stone and Metal Ages in Thailand, Thai Antiquity Working Group, Bangkok, pp. 81-90. Rispoli, F. (1990). La Ceramica di Non Pa Wai (Lopburi, Thailandia Centrale), Ph.D. thesis, Universit~t di Roma, Rome. Rispoli, F. (1992). Preliminary report on the pottery from Tha Kae, Lopburi, Central Thailand. In Glover, I. C. (ed.), Southeast Asian Archaeology, 1990, Center for South-East Asian Studies, University of Hull, Hull, UK, pp. 129-142. Rogers, P. R. (1995). Subsistence continuity in prehistory in south coastal China. In Yeung, Chung-tong, and Li Wai-ling, W. (eds.), Archaeology in Southeast Asia, The University Museum and Art Gallery Hong Kong and University of Hong Kong, pp. 467-478. SCrensen, P. (1988). The kettledrums from Ongbah cave, Kanchanaburi Province. In SCrensen, P. (ed.), Archaeological Excavations in Thailand. Surface Finds and Minor Excavations, Scandinavian Institute of Asian Studies Occasional Paper No. 1, Copenhagen, pp. 95-156. Stech, T., and Maddin, R. (1988). Reflections on early metallurgy in Southeast Asia. In Maddin, R. (ed.), The Beginnings of the Use of Metals and Alloys, MIT Press, Cambridge, MA, pp. 163-174. Tayles, N. G. (1992). The People ofKhok Phanom D~ Ph.D. thesis, University of Otago, Otago.
The Archaeology of Mainland Southeast Asia
49
Taylor, IC W. (1992). The early kingdoms. In Tarling, N. (ed.), The Cambridge History of Southeast Asia, Cambridge University Press, Cambridge, pp. 137-81. Trinh Sinh (1992). The socio-economic effect of metallurgy. Khao Co Hoc 75:19-26 (in Vietnamese). Trinh Sinh (1992). Copper artifacts in the Dong Son culture. Khao Co Hoc 80:55-64 (in Vietnamese). Vernon, W. W. (1995). The crucible in copper-bronze production in prehistoric Phu Lon, Northeastern Thailand: Analysis and interpretation. In Bulbeck, F. D. (ed.), Ancient Chinese and Southeast Asian Bronze Age Cultures, Australian National University Press, Canberra (in press). Vincent, B. A. (1988). Prehistoric Ceramics of Northeast Thailand, British Archaeological Reports, International Series 461, Oxford. Welch, D. J., and McNeill, J. R. (1988-1989). Excavations at Ban Tamyae and Non Ban Kham, Phimai region, Northeast Thailand. Asian Perspectives28: 99-123. Welch, D. J., and McNeill, J. R. (1991). Settlement, agriculture and population changes in the Phimai region, Thailand. Bulletin of the lndo-Pacific PrehistoryAssociation 11: 210-228. White, J. C. (1988). The southern metallurgical tradition. In Maddin, R. (ed.), The Beginning of the Use of Metals and Alloys, Massachusetts Institute of Technology Press, Cambridge, MA, pp. 175-181. White, J. C., Vernon, W., Fleming, S., Glanzman, W., Hancock, R., and Pelcin, A. (1991). Preliminary cultural implications from initial studies of the ceramic technology of Ban Chiang. Bulletin of the Indo-Pacific PrehistoryAssociation 11: 188-203. Wilaikaeo, C. (1988). Archaeological evidence from the Bronze Age in the Kwae Yae River Valley. Muang Boran 14:73-80 (in Thai). Woods, M., and Parry, S. (1992). An archaeological assessment at Don Dong Muang, Northeast Thailand. Spafa Journal 1993: 10-17. Yan Wenming (1991). China's earliest rice agriculture remains. Bulletin of the lndo-Pacific Prehistory Association 10: 118-126. Yeung, Chung-tong, and Li Wai-ling, W. (eds.) (1995). Archaeology in Southeast Asia, University Museum and Art Gallery Hong Kong and University of Hong Kong. Zhao Baoquan, Hu Youyan, and Lu Benshan (1988). Ancient copper mining and smelting at Tonglushan, Daye. In Maddin, R. (ed.), The Beginning of the Use of Metals and Alloys, MIT Press, Cambridge, MA, pp. 125-129.