From Yue To Tai...The second theory began among Chinese historians, especially Jiang Ng Liang whose Tai ju su (On the Thai nation, 1983) identified the Yue in early sub-Yangzi China

peakers of languages which are nowadaysidentified as part of a Tai linguistic family

are found along a thousand kilometre arc fromGuanxi to Assam. Earlier attempts to explainthis spread were bound up with the search forthe “origins” of a T(h)ai race, and worked fromthe model of an originally united and pure racewhich subsequently fragmented in theprocess of migration. More recent searches foran original Tai culture use a similar model.

Anthropologists and historians have attackedsuch models on the basis of both theory and ob-servation of Tai societies over the past century.Ethnic groups are not blood-denominated groupsbut constructed social identities which evolveover time. Tai communities have a spectacularability to absorb other peoples. The culturalvariety among different Tai–speaking groupsmay exceed the differences between them andother neighbouring groups (Evans 1999; Turton2000).

Yet the process by which this region becamepopulated by communities speaking variantsof this linguistic family is still of historicalinterest. Since the 1980s, the story of Tai migra-tions southward from the Altai mountains has

been discredited. Linguists identify the originof the Tai language family in southwesternChina. A school of historians in both China andThailand links the Tai to the peoples in sub-Yangzi China which early Chinese texts generi-cally call Yue (Ye, Yueh, Yüeh, etc). In thisessay, I attempt to reconstruct the dispersionof Tai–speaking communities. I argue that theprocess must be seen as more complex thanwarrior migrations or cultural diffusion.

I begin by looking briefly at the historiogra-phy of Tai “origins”. I then summarise somelinguistic theory on language spread, and use itto re-examine the evidence for the dispersion ofTai languages. In the remaining sections, I useChinese texts, Tai chronicles and legends tosuggest the pattern of movement, settlement, andcontact with other peoples.

Tai Historiography

In the late nineteenth century, westerntravellers like A. R. Colquhoun noted the largenumber of Tai–speakers in southern China. In apreface written for Colquhoun’s book, Amongstthe Shans (1885), the Sinologist Terrien de

Chris Baker1

From Yue To Tai

S

Journal of the Siam Society 90.1 & 2 (2002) 1

01P1-26 8/8/05, 14:381

2 CHRIS BAKER

Journal of the Siam Society 90.1 & 2 (2002)

Lacouperie (1885) speculated that the“Ngai–Lao” and “Ta Mung” found in Chinesetexts of the second to fourth centuries AD mightbe forerunners of these peoples. In The Tai Racepublished posthumously in 1923, W. C. Doddcited Lacouperie as his main source for anaccount that went far beyond Lacouperie’stentative speculation. Dodd (1923, 4) wrote thatthe Tai had been “fixed residents in China,living under stable governments which enduredfor millenniums” before the Han arrival; werefirst recorded in Chinese texts in northwesternSichuan in 2200 BC; subsequently moved southin seven “great migrations” ; and arrived in Siamafter the overthrow of their previous state ofNanchao. In Lak Thai (Origins of the Thai,1928), Khun Wichitmatra adopted Dodd’s listof migrations (without attribution), and em-broidered them with other speculations – includ-ing the thought that, before Dodd’s history, theThai must have come from the Altai mountainsbecause “that was the birthplace of the Mongols”(9-10). The rest is “history”. The story of theT(h)ais’ migration from the Altai mountainssouth through Nanchao to find a manifestdestiny in the Chaophraya basin was written intoschool texts and academic histories.2

Dodd’s identification of Nanchao as a Taistate was soon challenged, particularly by Mote(1964), and finally demolished in the 1980s(Backus 1981; Vinai 1990). Two rival theoriesof T(h)ai origins then appeared to fill the gap.

Inspired by the discovery of Ban Chiang andother early archaeological sites within modernThailand, Suchit Wongthet (1986) declared that “theThais were [always] here”, meaning that even if somewarriors had in-migrated with a new language, themajority of modern Thailand’s “ancestors” hadbeen resident since the Bronze Age. The SinlapaWattanatham school (named after a magazine whichSuchit edits) has constructed a history of Thailandwhich emphasises events which fall within themodern boundaries of Thailand – Ban Chiang,Dvaravati, Ayutthaya, Bangkok. Srisak Walipodom(Srisakra Vallibhotama) (1978, 88) wrote:

I think the study of Thai history shouldfocus on the present–day territory of Thai-

land since ancient times-how severalpeoples gradually joined together as poli-ties, as kingdoms, and finally as a countrywith a common culture called Thai culture.In the early 1990s, Suchit and colleagues went

on an enjoyable visit to the Zhuang in Guangxi.They were impressed first, that the Zhuang andThai languages were so close that they couldcommunicate easily, and second that the Zhuang’scliff paintings, because they looked rather likeEuropean cave paintings, were proof of theZhuang’s great antiquity. On return, Suchit(1994) modified his headline to “the Thais werealways here in Southeast Asia”, and speculatedthat the Tai migration away from the Zhuangcousins must have occurred as early as theBronze Age so that all subsequent Thai historywas still contained within the modern nationalboundaries.

The second theory began among Chinesehistorians, especially Jiang Ng Liang whose Taiju su (On the Thai nation, 1983) identified theYue in early sub-Yangzi China as the forerun-ners of the Tai.3 Jiang’s theories were acceptedby mainstream Chinese historiography, andexpanded by other historians including WangHui Kun and Chen Lufan. Jiang’s book wastranslated into Thai in 1986 and embroidered bya “Yue school” of Thai historians. This schoolappeared as China emerged from Maoism,diplomatic and familial links between Chinaand Thailand were restored, and Thailand’sSino-Thai-dominated urban economy began toexpand. Chen Lufan (1990) explained that thenew theory superceded the story of successivemigrations impelled by Chinese aggression,and was proof of the two countries’ eternalfriendship. Thai historians with Chinese familyorigins were undoubtedly warmed by the thoughtthat “Thais” and “Sino-Thais” enjoyed similarorigins and had only timed their migration ratherdifferently.

Language

The dominant model in historical linguisticshas been the family tree. Members of a particu-

01P1-26 8/8/05, 14:382

3From Yue To Tai


lar language family are assumed to havedeveloped from a single origin – the protolanguage. Linguists endeavour to reconstructthe proto language, and trace how it developedthrough a series of splits. In this vein, Li FangKuei published a “tentative classification” of Tailanguages in 1960, and then a fuller reconstruc-tion of Proto-Tai in 1977 (Li 1960, 1977). Bycomparing the complexity of developmentto the much better attested development ofEuropean languages, linguists agreed that thefragmentation of Proto-Tai began “less than twomillennia” ago (Diller 2000, 14-5). On groundsthat language families show the greatestintensity of variation around their point oforigin, William Gedney located the origin ofProto-Tai around the intersection betweenYunnan, Vietnam, and Guangxi (Gedney 1964;Hartmann 1980, 73). Several historical linguistshave traced the subsequent development ofthe Tai languages through a series of splits(see below). Historical anthropologists havespeculated that the languages were spread bywarrior-led migrant groups (e.g., Condominas1976, 1990).

However, the ramification of language fami-lies from a single origin is not the whole storyof linguistic development. Dixon (1997), forexample, proposed a division into two basicmodels. First, in periods of equilibrium, lan-guages in adjacent areas tend to exchangefeatures and converge. These periods may lasta long time. Second, at times of “punctuation”(meaning disruption), languages may split andspread in quite sudden and complex ways, andthe family tree model applies. This “punctua-tion” may be caused by natural causes, newtechnologies, or political changes which promptmigration movements.

Evans (1999, 12-3) suggests that work suchas Dixon’s questions the use of the family-treemodel to trace the Tai language family. In fact,the Tai family seems to fit Dixon’s “punctuatedequilibrium” rather well. Yet Dixon suggestssome qualifications to the family-tree modelwhich may be relevant. First, he notes that alanguage family “may have emanated not from

a single language, but from a small areal groupof distinct languages, with similar structures andforms” (Dixon 1997, 98). In a similar way, theThai linguist Anthony Diller (1990, 28), usingthe image of a piece of rope, proposed thatthe Proto-Tai culture was first formed by theconvergence of different strands into a singlerope, and then frayed again into several newstrands.4 Second, Dixon dismisses the idea thatthe time-depth can be estimated by comparingto the European case because the naturalenvironment and historical context might bequite different.

In addition, other linguists note that thepattern of successive branching in the visualrepresentation of the family-tree model tends toobscure the fact that languages may spread insuccessive waves, and that in the process theremay be exchange and convergence, both withother strands of the same language family, andwith other languages in the same area.

Linguists have also looked beyond militarycolonisation for explanations why certainlanguage families become dominant over a widearea. Renfrew (1997) argues that the bigdispersions of language in early human historywere associated mainly with agriculturalchanges.5 Groups with better ability to producefood may have higher population growthand thus incentives to migrate; or they may beable to support higher population densities andhence have greater military might; or they mayabsorb other language-speakers in peacefulways. Political factors may be alternative orsupplementary reasons to agricultural techno-logy. Groups which face hardship (famine,oppression) in their original homeland maydevelop military capacity or social mechanismswhich give them advantages over the local popu-lation when they migrate into a new area withmore benign conditions (e.g., Nichols 1997, 1998).

These theoretical findings have severalimplications for reconstructing the spread ofTai languages. The origin may not be a singleProto-Tai language and its speakers, but a group.The time-depth of the fragmentation need notbe two thousand years. The various dialects

01P1-26 8/8/05, 14:383

4 CHRIS BAKER


may be created not solely by a succession ofsplits, but also by joins and exchanges betweendifferent groups. While warriors might pioneerthe process of dispersion, technology and socialorganisation might be the underlying motiveforces.

Yue

Since Gedney’s first identification, Tai lin-guists have generally agreed on the languagefamily’s point of origin in the region around theYunnan-Guangxi-Vietnam trisection. Some alsolink the Tai family with other language groupswhich survive further to the west in sub-YangziChina.

Early Chinese records used the term Yue todescribe the non-Chinese people south of theYangzi. In the Spring and Autumn period (770-475 BC) the term was applied to a state on thesouth-eastern coast which was destroyed in 334BC as the Han Chinese moved across the Yangziinto the south. Subsequently, the term “hundredYue” was applied generically to the subjugatedpeoples in the south, with modifiers to denotegroups in different locations or with some otherdistinguishing characteristic (Phornphan 1988).The term Yue fades from usage around 0 ADas the Chinese gained more knowledge ofthe southern peoples and began using otherdescriptors (Barlow 2001, chs. 1-2; Taylor 1983,41-4). None of the modern terms used for Taigroups can be detected in these descriptorsexcept Lao or Ailao which was applied to avariety of groups, mostly hill-dwellers (Taylor1983, 172; Cholthira 2001, 22-4).

Historians of the Yue school draw attentionto some characteristics of the exotic southernersas noted in the early Chinese texts and some-times confirmed by archaeology: they cultivatedrice, built stilt houses without nails, tattooed theirbodies, cut their hair short, decorated their teeth,were expert weavers, cultivated silkworms, usedbronze drums in rituals, adopted snakes, frogsand birds as totems, practised spirit possession,divined with chicken bones, liked antiphonalsinging, put women to work, and resided with

the wife’s family. On the basis of the similarityto contemporary practices among Tai groups,scholars of the “Yue school” select sub-groupsof the Yue and claim them as “ancestors”(banphaburut) of the Tai (i.e., as the Proto-Tai).6

But many of these characteristics survive notonly in Tai societies, but also in other societiesof the region (Barlow 2001, chs. 2-4; Cholthira2001, 225-36; Somphong 2001, 234; Turton2000). It makes better sense to conceive of“Yue” as a broad and very varied cultural zone,from which many different elements “precipi-tated out” as a result of the entry of the HanChinese into the zone over more than amillennium.

From around the fourth century BC, the HanChinese began moving across the Yangzi. In221 BC, the emperor Shih Huang-ti rounded up“criminals, banished men, social parasites andmerchants” and sent five armies totallinghalf-a-million men to conquer the south (Wang1958, 10 quoting Shih Chi; Taylor 1983, 17-8).The main targets were the coastal ports whererich profits could be made by sending exoticgoods to the north. Around 200 BC, theChinese took Panyu, the main port on the PearlRiver estuary, from a Yue chief. Over the nextcentury they established nine ports and militarycentres stretching westwards along the coast tothe Red River delta, obliterating at least four Yuepolities in the process.

But consolidating control took a long time.The northern armies had low resistance totropical disease. On average around a quarteror a third died on campaign, and some armieswere totally obliterated.7 Emperors could notafford this too often. Hence the south wasabsorbed slowly over many centuries bymigration and acculturation, punctuated byoccasional armed invasions. Settlers, merchants,and administrators flowed into the south,particularly at times of dynastic decline andpolitical disorder in the north. In the early fourthcentury, for example, “vast throngs of Chineserefugees fled south” away from “maraudingnomadic peoples of the northern frontier”(Taylor 1983, 99). After the Tang collapse in

01P1-26 8/8/05, 14:384

5From Yue To Tai


the tenth century, noble families migrated southwith hordes of armed retainers (Ramsey 1987,33). Sometimes these migrants seized land.But they also tended to marry and blend into thelocal population, creating the hybrid languagesand cultures along the coastal strip.8

Meanwhile, in many cases as in 221 BC,“the Yüeh people fled into the depths of themountains and forests, and it was not possibleto fight them” (Taylor 1983, 18, quoting the Huainan tzu). As the Chinese consolidated controlalong the coast, those resisting integration movedaway into the interior, particularly into the highvalleys among the mountains in an arc runningwest and south from Nanning around the rear ofthe Red River delta. The Chinese gradually setup garrisons and provincial administrations inthe hills to control these peoples, but facedconstant resistance. In 722, for example, some400,000 “Lao” and “Nan Yue” rose in revolt,and the leader declared himself king of Nan Yue(Taylor 1983, 178, 192-3). From the sixthcentury onwards, the Chinese coastal settlementsoccasionally called on help from imperial armiesto conduct pacification campaigns on this hillfrontier. After the 722 revolt, some 60,000 werebeheaded. In a pacification campaign in theninth century, designed to secure the routethrough to the Nanchao kingdom in Yunnan, the

imperial armies beheaded 30,000 rebels in thecourse of a bloody campaign (Taylor 1983,239-49).

In sum, over more than a millennium, the oldYue of the south developed in (at least) twodirections. Some were absorbed into the hybridsocieties and states of the coast stretchingfrom the Yangzi estuary to the Red River delta.Others were pushed into the interior hills byarmed invasions and immigration waves, andoccasionally battered by frontier pacificationcampaigns. These latter were probably themechanism which pushed people westwards.

Little is known about the languages of theYue. The Chinese called them “bird speech”,twittering.9 One Yue song was transcribed intothe records in Chinese characters. This songremained unintelligible until a linguist managedto interpret it using reconstructed Proto-Tai(Zhengzhang 1991). Subsequently it was alsointerpreted using old Cantonese and Cham(Wade, personal communication). The Yue havemany descendants.

Dispersion

Chamberlain (1975, 1986, 1989, 1991, 1992,1998a, 1998b) has been the most ambitiousin attempting to construct a family tree of Tailanguages and dialects.10 Figure 1 is a simpli-

Northern Group (Zhuang)Central Group (Nung, Tho/Tay)

YuanKhunLuShanKhamtiAhomBlack TaiWhite TaiRed Tai

Southwestern GroupPhu TaiNeuaPhuan

Hua PhanVientiane

Luang Prabang IsanSouthern Lao

Sukhothai Southern Thai

01P1-26 8/8/05, 14:385

6 CHRIS BAKER


fied and synthesised version of his work. Cham-berlain has also tried to match the critical splitson this family tree with political events thatdivided peoples and spurred migration. Hesuggests that Proto-Tai was separated fromneighbouring languages by the Han militarypush around 330 BC which broke up thepolities just south of the Yangzi and propelledpeople south and west (Chamberlain 1998b,4-5). A division into three groups then took placebefore 0 BC. Both the “central group” of Tailanguages (including Nung, Tay/Tho) and“northern group” (including Zhuang) are foundin the mountain areas behind the early coastalstates. The third “southwestern” group is scat-tered westwards across the hill ranges dividingChina from mainland Southeast Asia, withsouthern spurs along the valleys of the Mekongand Chaophraya river systems. The division intodialects within this southwestern group is highlycomplex.

In the light of newer language theory,Chamberlain’s chronology should probably beloosened.11 Trade and cultural flows along theroute between the Red River delta and Yunnan

(either along the river or across the plain to itsnorth) are very old, attested by the distributionof distinctive corded pottery, “bean pots”,three-legged jars, and the Dongson bronze drums(Meacham 1983). Successive migrations andlanguage shifts may have occurred along thistrajectory over a long period. But it is likely, asChamberlain argues, that Han pressure createdthe “punctuation” for more dramatic change.

The dispersion of Tai languages across thehills took place over a millennium or more. Itwas not linear or regular. Although the overalltrend was westward, there were many sidetracksand reversals. Other peoples were movinginto the same region over the same era fromdifferent directions. Tibeto-Burman peoplesfrom the northwest were filtering into Burma andYunnan (Driem 1998). The Viets may haveexpanded northwards into the Red and Blackriver region (Chamberlain 1998a, 37-44).Chinese filtered south by many routes.

Map 1 shows the modern distribution of Tailanguages.12 The arrows show Chamberlain’s“family tree” overlain on the map. The arrowsdo not show migration routes (which were much

01P1-26 8/8/05, 14:386

7From Yue To Tai


more complex). But combining the map andthe “family tree” suggests that Chamberlain’sdivision of the Tai languages makes some sense,and hints at the main trends of movement. Inthe historical record – Chinese texts and locallegends – we get glimpses which help to fill inmore detail on routes and timing. Five streamsor phases of movement can be glimpsed in theserecords.

The earliest of these streams moved first tothe northwest. According to Chinese scholars,people moved along this route before 0 BC (Tan1994).13 They became part of the various king-doms - Dian, Nanchao, Piao - which appearedon the southwestern frontier of Han expansion.They were not the core people of these states(who were probably Tibeto-Burman) but morelikely settlers, slaves, and soldiers. Chinesescholars believe some arrived in Yunnan fromthe southeast; but others came from the north asrefugees from the warring states on the middleYangzi. Some of these groups seem later to havemoved south along the rivers, especially theIrawadi and Salween. Chinese scholars detectearly development of communities around theSalween area perhaps from the first century AD,including a “kingdom of riding elephants”.By the ninth century, the Chinese call thebarbarians around the upper Salween andMekong the “Mang Man”, and note that theirchief was called “mang-chao”, and that “theylive in pile-propped houses”. The Chineseclassified the various man barbarians of this areainto those with gold, silver, or black teeth, and thosewith tattooed legs or faces (Luce 1961, 42-3).

The Shan chronicles record the first migra-tions south from Yunnan into the Shweli Riverregion in a period estimated around the sixthcentury (Sai Aung Tun 2000). Foundationlegends of some Shan towns also relate west-ward movements through the Shan area aroundthe same era – from the Mekong westwardsacross the Salween, and then to the Shweli (Renu1998). The Yuan shi records that at the end ofthe eighth century, the Nanchao ruler attackedsouthern Yunnan, “defeated the hordes ofbarbarians (qunman), captured all of their people

and [moved them] to populate the south, east,and north of his territory”; but when the Nanchaokingdom disintegrated in the tenth century, thesepeople “regained their former lands” (Daniels2000, 69-70). The Mongol scribes whodocumented Yunnan in the thirteenth centuryidentified several groups including many “goldteeth” and the Bai-yi (or Pai-i, white barbariansor white clothes) who “flourish the most amongstthe barbarians of the south-west”. They wereespecially concentrated between the Salweenand Irawadi (the M. Mao area), but were foundalong an arc stretching from Tibet to the RedRiver delta and their “customs are generally thesame” (Daniels 2000, 71 from Yuan Shi, see alsoLuce 1959, 60).

In sum, there seems to have been amovement along the northern-most branch ofChamberlain’s tree-first up into lower Yunnan,then southward down the rivers-which beganfrom the period of Han expansion along thesouth Chinese coastal plain, and continuedsporadically over many centuries. By thethirteenth century, the Chinese are identifying adominant group in this area with a descriptor(Bai-yi) which is later applied to the Shan.14

A second stream may have begun from thesame disturbed period, and initially movedwestwards into the hills circling the Red Riverdelta. Petty polities appeared including one atCao Bang ruled by lords named Lac or Lo(Taylor 1983, ch. 1; Chamberlain 1998b, 5).These polities were occasionally pressured orabsorbed by the Han, but some rebelled andremained independent. This area was disturbedagain around the seventh or eighth century whenViet groups began to expand northwards into thedelta, and when the Han conducted frontierpacification campaigns. They seem to havemoved northwest towards Yunnan, and thendown into the valleys of the Red and Blackrivers. The name Lac or Lo is preserved as thename of the noble lineage among some groupsof Black Tai. Cam Trong (1998, 20-1) tracesthe Black and White Tai moving south into thearea of the Red, Black and Ma rivers over theeighth to thirteenth centuries.

01P1-26 8/8/05, 14:387

8 CHRIS BAKER


A third stream can be dated more exactlyaround the eleventh century. During the earlierHan expansion, some Yue groups took refuge inthe area between the You and Zuo rivers inGuanxi. They were attacked by the Han, theViet, and by raids from Nanchao, but wereprotected by the terrain described by Chineseofficials as “steep mountains and inaccessiblevalleys”. Their chiefs developed a strongmilitary system and the Chinese reportedrespectfully that the people “loved to fight andstruggle and regarded death lightly” (Barlow1987, 255). In the 1040s, one clan led by apowerful matriarch-shamaness (A Nong), herchiefly husband, and their son (Nong Zhi-gao)extended their rule to fourteen valleys, raised arevolt, and declared the founding of a stateusing Chinese political vocabulary and symbols.In 1052, the Nong rebels swept westwards, tookNanning, besieged Guangzhou for fifty-sevendays, and slew the commanders of five Chinesearmies sent against them. Finally they weredefeated, and many of the leaders killed (Barlow1987; He 1998; Hoang 1988; Anderson 2001).The clans who had not joined this ambitiousuprising were subjected to Chinese accultura-tion. Some of the rebels filtered back to theirearlier home. Others fled deeper into the hills.Clans which use the name Nong and claim descentfrom the rebels are settled around the Yunnan-Guangxi-Vietnam borderland. Others who retainNong as a clan name and who remember the re-bellion in oral legends, settled in Sipsongpanna,Lanna, and Dehong (He 1998, 93-4). The Lu,Yuan and Dehong dialects from these areas con-tain words and constructions found elsewhereonly among the Zhuang who continue to live inGuanxi (He 1998, 97; Pan 1990; Luo 2001).

A fourth phase of movement seems to haveoccurred west of the Salween from the eleventhto fourteenth century. M. Mao15 emerged as anambitious new capital which rose to a peak inthe twelfth to fourteenth centuries (Zhou and Ke1990; Sao Saimong Mangrai 1965). Many newmigrants arrived, and many towns were foundedor refounded (Sao Hso Hom and Pu Loi Tun,1998; Somphong 2001, 25). These develop-

ments were then disrupted by the Mongolefforts to capture Pagan. Khublai Khan took Taliin 1253, and sent a first expedition south in thefollowing year. In the 1260s, the Mongolscreated a corridor through to the upper Irawadi.After Pagan’s failed pre-emptive northernexpedition in 1277, the Mongols forged south,recruiting the Pai-i as troops. One expeditionclaimed it “registered 110,200 households”(Luce 1958, 133). Luce argues that the Mongolinvasions of Pagan in 1277-8 and 1283-4precipitated Shan expansion westward. Thelegends of M. Mao and Mogaung tell of theirrulers extending their power eastwards tothe Mekong, northwards into Yunnan, andwestwards to Arakan, Manipur, and Assam (SaoHso Hom and Pu Loi Tun, 1998; Sao SaimongMangrai 1965). According to the Ahom Buranji(chronicle), one chief, Sukupha, moved his basefrom M. Mao via M. Mit, and finally over thehills into the Brahmaputra valley in thirteenyears of wandering. Gait (1994, 73) times thiscrossing very exactly to 1228, but a solar eclipsementioned in the Buranji suggests a date sometwo centuries later (Diller 1992, 10). In the Mingrecords, there is a flurry of reports in 1406 aboutan expansive state in the southwest named DaGu-la; this might mark the Ahom migration andsettlement in the Brahmaputra valley (Wade1994, 301-2, 4567). Another group fromMogaung/M. Mao moved west to the Chindwin.The name Khamti, subsequently applied to thedialect, had been a clan name in M. Mao (SaoHso Hom and Pu Loi Tun. 1998). Luce (1958,14, 174 fn. 10) says this settlement does notappear in the Chinese surveys until 1400.

A fifth phase of disruption and movementoccurred around the Mekong. It may havebegun during the Nanchao-Tali wars in thetwelfth century, and climaxed after the Mongolsturned their attention to this area at the end ofthe thirteenth century. The legends of the Lu,Khun, and Yuan relate movements to found newsettlements in the few generations before theseaccounts break into historical time in the latethirteenth century. The dating is far from pre-cise, but the freshness of the stories suggest they

01P1-26 8/8/05, 14:388

9From Yue To Tai


relate to quite recent events, perhaps from thetenth or eleventh centuries onwards.16 Thesemovements seem to be south from Yunnan alongthe Mekong River, and then southwards againalong the valleys of the Kok, Ping, and otherrivers. A Mongol survey of Yunnan’s southernborder announced “there were about 200,000people longing for civilisation and anxious tosubmit” (Luce 1958, 147). In 1292, Mongoltroops attacked Pa-pai-hsi-fu (Lanna) anddemolished the city gates (Luce 1959, 72).Ch’eli (Chiang Rung) was taken in 1296 andestablished as a military outpost. But the arearemained rebellious and difficult to control. TheMongol court ordered armies south in 1297,1301, 1303, 1309, 1311, and 1312, but not allthese expeditions took place. The 1301-2campaign resulted in a mutiny, the “ruin of thearmy”, and subsequent execution of the Chinesemilitary commander (Luce 1958, 165-6; 1959,77-80). The 1312 effort collapsed because of“pestilence and other hardships” (Chen 1949, 9).Mongol officials argued that the mountains,forests, malaria, and supply difficulties meantthe military losses outweighed any gains.17

After 1312, Che’li and Pa-pai-hsi-fu senttribute (tame elephants) in return for which theMongols left them largely alone (Luce 1958,164-71; Hsieh 1995, 308-9). In these fewdecades of warfare, peoples settled along theMekong and its northern tributaries were pushedsouth.18 The Yonok rulers at Chiang Saen movedsouth along the Kok and Ping, and relocated theircapital to Chiang Rai and then Chiang Mai(1296). The Kaeo rulers moved their capitalsouthwards down the Nan River in the earlyfourteenth century (Sratsawadi 1996, 86-91).The infant Fa Ngum went south to Cambodia inthe 1310s for mysterious reasons, and laterreturned to found the Lanchang capital at LuangPrabang (Stuart-Fox 1998, 37-8). The Taymoved south-eastwards from Yunnan back intoGuangxi (Barlow 2001, ch. 7).

These sketches suggest waves of migrationout from southern China, propelled by succes-sive phases of Han and Viet aggression, over aperiod running from at least the third century

BC to the eleventh century AD. Mostly thesemovements seem to have followed the north-westerly route into lower Yunnan, and thensouthwards along the river systems into the hills.Movements on the more southerly routes intothe Annamite hills and onwards to the middleMekong are much less clear.19 Further, thesesketches suggest a phase, possibly begun by theexpansion of Piao Pagan, and certainly intensi-fied by the Mongol intrusion after 1270, whenpeople were splayed to the west and the south.20

Passages

This dispersion was not a simple, linearprocess. It was made up of thousands ofshort-distance moves by relatively small groupsof people. Some of the legends talk of chiefsleading bands of a few thousand. The AhomBuranji says that Sukapha was accompanied byeight nobles, nine thousand people, twoelephants, and three hundred horses (Gait 1994,73). Some talk of a handful of families. Somerelate the story of only a single pioneer, couple,or pair of brothers. In some places, the legendsrecount a single time of foundation andsettlement. Elsewhere the picture is morecomplex. The M. Mao/Mogaung legends tellsof three successive influxes of Tai peoples(Sao Hso Hom and Pu Loi Tun, 1998; SaoSaimong Mangrai 1965). Some of the legendsportray several moves interrupted by pauses.The Black Tai move south to settle around M.Boum on the Black River, and then some twocenturies later moved on to M. Thaen (CamTrong 1998, 20-1; Chamberlain 1991; Phatthiya2001, 147-53). In one Chinese text, the migrantTai “move in large numbers to the far southerntropical forest along the mountain chains andrivers”. Occasionally they settle but “it was notlong before they began a new movement everyfive or ten years” (Wang 1990, 5).

Luo (2001, 185-6) finds that the Tai inDehong, Sipsongpanna use many commonwords from the “northern” Tai dialects, butpronunciation similar to the “southwestern”group. He hypothesises that refugees from the

01P1-26 8/8/05, 14:389

10 CHRIS BAKER


Nong revolt moved first into the Red-Blackrivers area where their pronunciation changedtowards that of earlier settlers, and then latermoved on northwards to Dehong.

The “family tree” portrayal of the languagedevelopment and the migratory pattern ismisleading in one important way. It shows thesplits. It does not show the mergers and joins.As in Luo’s example, both the dialects and thedialect groups were formed by mingling ofvarious migrant streams. Chinese scholars’efforts to trace the origins of Tai communitiesin Yunnan identify migrations coming south-wards from the Yangzi valley and “cold regions”beyond, as well as peoples arriving from thesouth-east along the Red and Black rivers (Wang1990; Liu 1999). Similarly, some legends ofLuang Prabang describe the early inhabitantsarriving down the Nam Ou from the Plain ofJars, while another version records the founda-tion of M. Sua in the Luang Prabang area by abetelnut merchant moving up the Mekong(Stuart-Fox 1998, 19-20). The Shan legendshave movements shuttling up and down the riversystems to and from Yunnan, as well as streamsarriving from the west. The foundation legendsof the Yuan have groups coming both up theMekong from the south-east, and down theMekong from Yunnan, as well as westwardsfrom the Shan States probably propelled by Piao,Nanchao, and Mongol raids. Throughout themigration, groups from the south-west weremerging with other migrants which may havehad very different origins.

Search

Did Tai-speakers come to dominate throughthis area because they were warriors, or did theycarry with them some agricultural advanceor social assets? The evidence is inevitablyindirect. But the dominant motif in the founda-tion legends of many Tai muang is a migrationin search of good riceland.21

The Tai migrations probably did not intro-duce rice to the hill tier. Indeed, according tosome scholars, rice was first domesticated in this

region (Chang 1993; Watabe 1998). Theearliest evidence of intense rice cultivation,however, comes from the lower Yangzi valleyaround the seventh millennium BC. From there,the technology seems to have spread slowly(though this may simply reflect the limitedextent of archaeological research). The earliestdates for cultivated rice in the Mekong basin fallin the late third millennium BC (Glover andHigham 1996; Bellwood 1996).22

The Tai migrants may have brought not riceitself but better ways to grow it through watercontrol. Early Chinese visitors to the Yue areasouth of the Yangzi were fascinated by theintricacy of rice-growing. One poem by ChengHeng (78-139 AD) about the Tung Yueh regiondescribes water management for paddy andintense garden cultivation:

From the streamsTunnels have been bored that lead therushing currentFlowing into these ricefields,Where channels and ditches link likearteries,Dikes and embankments web with oneanother;Dawn clouds need not rise upThe stored waters find their way alone,And when the sluices are opened, theydrain away,So that the fields are now flooded, nowdry again,And the winter rice, the summer wheatRipens each in its proper season.In the broad meadowsAre mulberry, lacquer trees, hemp, andramie,Beans, wheat, millet, and panicled millet,A hundred grains, thick and luxuriant,Burgeoning, ripening.In garden plotsGrow smartweed, fragrant grasses,turmeric,Sugar cane, ginger, garlic,Shepherd’s purse, taro, and melons(Daniels 1996, 182).

01P1-26 8/8/05, 14:3810

11From Yue To Tai


The groups pushed into the mountain valleysof Guanxi seem to have developed specialmethods of water management. They grew ricearound karst outcrops which absorbed rain-water and slowly released it as run-off. Theyused dikes, retention dams, bamboo pipes, stone-lined ditches, and lifting devices for managingwater. They chose settlement sites on flat landbetween mountain slope and river (Barlow 2001,ch. 2.1).23

The heroic foundation legend of many Taisettlements tells of a leader moving through thehills in search of a flat well-watered plain - theideal spot to support a large population of rice-growers. One of the most eloquent legends isthe Black Tai Khwam to muang. In the openingsection, the earth is separated from the heavens,and the sky god cleanses the earth by droughtand flood. The god then populates the worldwith people and animals (buffalo, ox, pig, horse,and deer). Amongst the people are two “ances-tors” who have several children. Six sons gooff to rule existing muang, leaving the seventhson.

This last child had no ricefieldThis latter child had no muangAnd so accompanied by his family, KhunLo, Khun Leuang, Khun KwangKhun Tong, and Khun LeoHe set forth to establish a territory of hisown.

This takes a long time. They raft acrossrivers, climb mountains, and scale cliffs. Thefirst place is too small. The next has a sacredcave which is attractive, but again the spacefor growing rice is too limited. The next iswell-defended by mountains, but still too small.The next is on a slope and the water is bad.“It was high and cold, a good place but not forTais”. In each case “the ancestor would notaccept it”. Then they hear of a “wide andspacious valley, with great fields all adjacent, agood place, with fertile paddies on either side”.Eventually,

Swinging their armsThey were not yet tiredAnd their destination was nearThey descended on Muang Theng.Muang Theng was round like a winnow-ing basket,A valley gently curved as a buffalo horn.It was a good place, wide, with ricefieldson either side,A desirable place where thousands couldlive.The Ancestors founded Muang Theng(Chamberlain 1992).

This legend records the foundation of M.Thaen (Dien Bien Phu). The same theme of theitinerant seventh son appears in the foundationlegends of many other places. The details vary.In another version of the same Black Tailegend, the sky god populates the world not onlywith humans and animals, but also with“everything indispensable for life” including 330types of rice, and books teaching about divina-tion and magic (Roux 1954, 377). The journeyincludes boat trips along rivers, and treksthrough deep forests. One of the places foundhas beautiful women, but is still too small to besuitable. At another, the water courses are toofull of rocks. At another, the local people aretoo tormented. Another is plagued by viciousyellow-winged bees. Eventually they find abroad plain suitable for rice.

In the Sinhanavati story from Lanna, the rulerhas thirty sons and despatches twenty-nine tofound new muang. Sinhanavati travels south tothe Mekong and rejoices in a location “withmany water courses of all dimensions” (Notton1926, Sinhanavati, 143). In the SuvannaKhomkham story from Lanna, a banished princefounds a new settlement along the Mekongwhere “Heavenly rain made the countryflourish. The ricefields produced abundantharvests and nothing was lacking.” (Notton,1926, Suvanna K’om Kham, 116).

In another popular theme, the Tai hero isout hunting, and is led on a long chase by agolden deer. The deer seems lame and easy to

01P1-26 8/8/05, 14:3811

12 CHRIS BAKER


catch but always evades the hunter, leading himeventually to a beautiful place. In someversions, the hunter then has lunch, and the seedsfrom his garlic and onions fall on the groundand start growing. In other versions he meets arishi who convinces him to build a city, andindicates clumps of lotus which will produceenough rice to feed a big population (SaoSaimong Mangrai 1981, 224-8; Notton, 1926,Yonok, 1-13).

The key plot elements about seven sons, aflighty deer, a helpful rishi, and plants whichstart growing of their own accord, recur indifferent combinations with different localdetails through many of these legends. Com-mon to all versions are the themes of arduoustravel, the search for broad well-watered plainssuitable for rice production, and success markedby spontaneous or spectacular fertility.

The places chosen as the major settlementshave a similar topography. They are largebasins, often at a confluence, where the rivershave created a broad alluvial plain. Thesurrounding mountains provide defense.Streams flowing from the hills down to the mainriver can be diverted for irrigation. The riveroffers a secondary reason why such a site isattractive: trade. One Tai poem lists severalmuang, some which have good rice yield, somewhich have “fruit orchards and coconut groves”,and another where the ruler “can collect a greatdeal of tribute because it is a trading centre withmany markets” (Chamberlain 1992, 18-19,29-30). The Sipsongpanna chronicles describesan exchange of tribute between two relatedcentres. One seems to have a craft industry andsends golden and silver howdahs, a goldenwater-carrier, embroidered mattresses, andblankets. The other appears less developed andreciprocates with horses, cattle, mules, steelswords, and salt (Ratanaporn 1998, 118). Afterthe Tai chase the Akha into the hills, the Akhastill return to the valleys to trade. In their texts,“The lowland scene is represented as a largeriver which has to be crossed, a valley, and ...markets where the Akha go to buy and sell”(Alting 2000, 138).

Across the hill ranges from the Red River tothe Brahmaputra, there were several placeswhere the ecological conditions were right,and where the settlements became large andpolitically central: M. Lai (Lai-chau) and M.Boum on the Black River; M. Thaeng (Dien BienPhu); Luang Phrabang, Chiang Saen, and ChiangRung (Jinghong) on the Mekong; M. Pan andM. Nai on tributaries of the Salween; Hsipawand Hsenwi on the Myitnge; M. Mao on theShweli; Mogaung on the upper Irawadi; Khamtion the upper Chindwin; and Charaideo on theBrahmaputra. Even the settlements whichbranched south into the plains retain somememory of this ideal landscape. Sukhothai,Kamphaeng Phet, and Champassak are situatedon flat ground between hill and river, though ifthe eyes turn south there is no longer hills butonly a flat expanse.

Originals

Another constant theme of these legends isthat the Tai are not the first pioneers: others werethere before them.

Throughout these stories the itinerant land-seeking Tai come across peoples that arevarious called Kha, Khamu, Thamin, Milaku, Sa,Wa, Lawa, Laha, and Meng. They are oftendescribed as dark-skinned. Most of these wereprobably Mon-Khmer speakers, as small groupsusing the same names still exist, usually livingon the hills or in the forests throughout main-land Southeast Asia.24 Daniels (2000, 62) notesthat several Mon-Khmer polities survived inYunnan until the seventeenth century.

In the Tai legends, sometimes these earlysettlers are described as living on the hill slopesand in the forests. The first contacts with theTai are about exchange of products. In thePhayao and Doi Tung tamnan, the first ruler is a“Tamila” called Lawa Cok. He is an expertsilversmith and metalworker. His people grow“rice, beans, peas, sesame, galangal, ginger,gourds, marrows, cucumber, water-gourd,pumpkins, and maize” on the hill slopes. Theybring these products down “to the foot of the

01P1-26 8/8/05, 14:3812

13From Yue To Tai


northern mountains” to exchange with the Taifor “meat, fish, betel nuts, leaves and salt”(Aroonrut 2000; Cholthira 1997). Izikowitz(1981, 104) notes that many Tai centres alongthe Mekong are situated at the confluence oftributaries (like the Ou) which facilitate tradewith the people deeper in the forests and hills.

But in many cases these earlier peoplesoccupy the same broad valleys which attract theTai. Indeed, one thing which makes a site a goodplace to end their migration is that it is “wellpeopled” and already has “paddy fields on bothsides of the river” (Roux 1954, 379 from a BlackTai legend). The site chosen for Chiang Mai,for example, was earlier occupied by three orfour Lawa settlements (Sratsawadi 1996;Aroonrut 2000; Daniels 2000, 59). In someversions of the Doi Tung story, the Lawa chiefbrings his people down to the riverine plain tofound Ngoen Yang and later confronts the Tai(Ratanaporn 1998; Sratsawadi 1996, 37-40).In another Lanna legend, a Lawa mountain chiefannounces, “We are too many and we occupythe whole mountain. There is no more space tocultivate. We must go down into the dry regionand the thick forest.” They clear forest, build amoated settlement fed by a natural stream, andlive well on rice and fish. The Chinese attackwithout success. In old age, the Lawa chiefreturns to the hill and the settlement is aban-doned. Later the Tai come and reoccupy the site(Notton 1926, Mahathera Fa Bot).

Contact I: War

In these stories, the encounters between theTai and the earlier inhabitants are resolved intwo ways. In the first, there is conflict and theearlier inhabitants are dispersed into the hills.

In the Kengtung legend, the first settlers areChinese, but the local spirits announce, “We donot like the Chinese and are opposed to theconstruction of the city by them,” so the rice willnot grow properly and the Chinese leave. Then80,000 Palaung Wa arrive under eight chiefs,start growing roots and yams, and build a citywith a moat and twelve gates. The Chinese now

demand tribute but are sent packing. A splendidPalaung ruler emerges who “constructed villagesand towns, demarcated boundaries, appointedministers and officials to rule over provinces,small and big... ruling over one million paddyfields”. Eventually a Tai chief arrives. Hecannot defeat the Palaung by force, but catchesthem out with a trick, kills their ruler, and chasesthem away. The Palaung move off to Mongkhaand build a new town. But the Tai come againwhile they are still digging the moat, and thistime the Palaung are scattered into the hills (SaoSaimong Mangrai 1981, 201-3, 214-8). In theChiang Saen legend, the Lawa chief accepts tenthousand pieces of gold to hand over his peopleto the Tai. He retires to live on the hill of DoiTung from where his ancestors had come(Notton 1926, Sinhanavati, 171-5).

In another story, two brothers lead the BlackTai down the Red River to settle on the broadplain at M. Lo. Two generations later, Cheung,the youngest of seven sons, sets out in the usualseventh-son way. First he crosses the mountainsto the Red River, attacks a Kha settlementunsuccessfully, then returns with a bigger armyand succeeds. He and his followers colonisealong the river, eventually arriving at a fine butalready occupied valley. They try to claimpossession by force but fail because the exist-ing Laha inhabitants are fierce. They ask thelocal chief’s permission to take up residence sothey can observe the local habits. Later Cheungasks to marry the chief’s daughter. At thewedding feast, the Laha chief brings fiftybodyguards, displaying a conspicuous lack oftrust. But these guards are persuaded to stacktheir weapons and proceed to get drunk. TheTai seize the opportunity to kill the local chief.His daughter is rendered speechless by thistreachery, and the other Laha run away in fright.After more colonising along the Black River,Cheung crosses the hills into the Ma valley, andthen branches out southwest across the hills tothe Ou and other tributaries of the Mekong.Finally he retraces his route to M. Thaen whichis already a thriving settlement of Black Tai andKha, where he becomes chief. This story traces

01P1-26 8/8/05, 14:3813

14 CHRIS BAKER


a movement first southeast down from lowerYunnan, then southwest across the ridgesdividing the Black, Red and Ma rivers, andfinally beyond into the watershed of the Mekong.All along the route, the Tai create new muangalong the river valleys, often encounteringresistance from earlier settlers. Cam Trongestimates Cheung’s journeying took twentyyears, and places the events in the eleventhor twelfth century (Phatthiya 2001, 147-52;Chamberlain 1992, 67-8).

The most dramatic of these stories ofconflict is found in an epic poem. The nameCheung appears to be a generic name forlegendary heroes which turns up in several closevariants (Hung, Rung, Ruang). Some of theCheung stories are versions of the Tai tales ofheroic pioneering, and often the name Cheungis used for the legendary first ancestor of amuang. But the most elaborate version of theCheung epic is reckoned to be based on a storyof Lawa defiance against the Tai.

The first part of the tale is a romance. Cheungfalls in love with the daughter of a muang ruler,but is rejected because he cannot pay the hugebride price demanded. But then when anothersuitor - probably Tai - begins to pursue the girl,Cheung is called on to help, becomes a greatwarrior, and gets the girl. The epic then movesinto a second act in which Cheung leads theLawa warriors in a more general defiance of theTai. Cheung is splendidly successful until theTai unfairly ask for help from the gods andterrestrial allies. Against such heavy odds,Cheung is killed on the field of battle. But all isnot lost. He is immediately reborn as a spirit,and resumes the battle in heaven. He defeatsmany gods including Indra and the ruling spir-its of several muang. This sublime defiance hasechoed back to earth. Revolts by Lawa, Khmu,and other subordinate groups right down to thetwentieth century have invoked Cheung asmillenarian inspiration (Chamberlain 1986,1992, 1998b; Smalley 1965; Proschan 1998).

Another tale of conflict is recorded by theAkha, a Tibeto-Burman people who preservetheir history through oral tradition in great

detail (Alting 2000). From around the thirdcentury AD, the Akha lived in southern Yunnanon the Red and Black rivers where they grewrice in the lowlands and became subjects of theNanchao kingdom. The Tai arrived in the areaonly around the thirteenth century. The Akhabriefly founded a kingdom at Mojiang to resistboth Tai and Chinese inroads. But Tai warriorsbesieged the capital, cut off water supplies, andforced them to flee. The Akha first retreatedsouth-westwards and crossed the Mekongaround Chiang Rung. But the Tai continued topress them, and the Akha broke into groups andscattered into the hills. The Akha are now foundas mid-level swidden farmers in scattered areasacross the hills. Their songs and oral texts recordhow they lost the irrigated lowlands to the Taiwho came “upstream” from the south. They saythat “the Tai invented war” (Alting 2000, 137).

Some of the Tai migrants may have beenwarriors. Condominas (1990, 45) assumed allof them were. Barlow (2001) describes how thegroups in the Guanxi hills became steadily moremilitarised from around the fifth century BConwards. The typical political unit was a singlevalley (dong). Even before the Han incursions,warrior chiefs had begun to displace earliershaman-rulers and extended control overseveral valleys. In the face of the Han, theysometimes revolted and sometimes hiredthemselves out as mercenary armies. Theirarmies were based on a three-man unit whichfarmed together and also fought as a team: “Oneman carried the shield which covered the bodyand the other two threw spears from behind....They came on like a southern fire” (Barlow1987, 253, 257-8). The Nong group whichrebelled in the eleventh century bred horses forcavalry, fought with swords and crossbows, builthill-top forts for defense, had female warriors,tattooed their bodies, and used chicken bones todivine their prospects for victory (Barlow 1987;Schafer 1967, 50, 56).

01P1-26 8/8/05, 14:3814

15From Yue To Tai


Contact II: Compromise

But not all stories of contact are stories ofbattle. The second form of encounter results inmore subtle stories of compromise.

In the Suwanna Khamdaeng story fromLanna, the Tai pioneer who is led to the newcity by a golden deer announces, “it is neces-sary to obey the people who were born and liveat that place”. He raises the Lawa chief to ahigh official rank and marries two of his daugh-ters. The ranking Lawa chief then teaches theTai people how to live without theft, lying,adultery, drunkenness, or drugs. Both Lawa andTai are happy. The rains are good. One year ofcultivation yields seven harvests. People multi-ply and new settlements proliferate (Notton1926, Yonok, 7-11).

Several tales specify how important it is forthe Tai migrants to honour local customs andsocial practices. A little later in the SuwannaKhamdaeng story, three of the Tai ruling familydie of fever. The Lawa tell them that if theyhave a fever, stomach ache, sore eyes or otherproblems, it means the local phi (spirits) arehungry and the problem can be overcome byfeeding them. “Ever since this epoch,” thelegend notes, “the Tai have taken up the exampleof the Lawa.” When drought and famine occurlater, they decide to prevent the bad spiritsdistinguishing between peoples: the Tai adoptthe Lawa hairstyle, and the Lawa put onTai-style dress (Notton 1926, Yonok, 19-21).In another version, the Lawa and Tai swap dressand hair-styles to confuse an enemy (Aroonrut2002, 5).

When the Tai chief comes upon the Lawa’sabandoned settlement in Kengtung, his soldiersare about to destroy an idol left behind. Thechief stops them, and sends a mission to theLawa. He has the Lawa chronicle translated intoTai so he can understand the local background.He takes instruction from the Lawa chronicleron how properly to propitiate the local spirits(Notton 1926, Mahathera Fa Bot, 44-5). As aresult, he prospers. Other tales tell how the Taiwho mistreat the Lawa’s cult images come to grief.

At the end of the Mahathera Fa Bot tale, themoral of cooperation is heavily emphasised. Ifthe town is attacked, offerings have to be madeto the Lawa gods, the Tai ancestors, various Indicdeities, other local spirits including that of thePing River, and the spirit of the Lawa chief. Thetown is then ritually surrounded by a rope ofplaited grass, symbolising the “perfect unity” ofthe fibres plaited in an emerald cord (Notton1926, Mahathera Fa Bot, 60-3).

In several of the Tai legends, the Lawa areportrayed as yak (monsters) who terrorise thelocal people until the Tai come. Then the yakare either subdued or chased away, but their roleis not totally erased; often the yak are thenincorporated into the spirit pantheon asprotective deities. At Chiang Mai for example,the yak are transformed into Pu Sae and Na Sae,the ancestor spirits of the Lawa and protectorspirits of the city (Tanabe 2000, 297). In otherversions, the Lawa chief who compromises andcoexists with the Tai is transformed into aprotective ancestral spirit (Ratanaporn 1998).The Chiang Saen Lawa who retires back to DoiTung is ultimately reborn as such a spirit.

Several town legends include a story in whichthe Lawa come down from the hills, enter thepalace just before the Tai ruler’s coronation, starteating a meal, and have to be chased out so theceremony can proceed. This bit of horse-playwas re-enacted in the coronation ceremonies atChiang Mai, and other Tai capitals down torecent times. Similarly, in the Chiang Mai story,Mangrai has to enter his capital following a Lawacarrying a basket and leading a dog (Aroonrut2002). In Kengtung, the Khun ruler ascends thethrone only after striking a bargain with the priorLawa ruler. At Luang Prabang the Kasak, whoclaim to be descendants of the originalMon-Khmer rulers of the place, shoot arrowsat the palace until the Tai king disperses themby throwing a rice ball (Tanabe 2000, 298).Chamberlain (1986, 63) summarises themeaning of these stories:

The subjugated race is placed in aposition of quasi servitude, but because

01P1-26 8/8/05, 14:3815

16 CHRIS BAKER


they are the older population they hold thekey to the control of the local spirits andthey have a more powerful magic. Thus,they must receive a modicum of respectfrom the conquerors and will always takepart in spirit ceremonies.

The readiness to adopt local knowledgeand practice, and the ability to develop ritualdramas which stabilised relations of subjection,suggest that the Tai shared a cultural backgroundwith the people whom they dominated. Theymay also have carried some practical socialtechnology. We have little knowledge ofsociopolitical organisation among the Yue otherthan that land was controlled by the chieftainand granted to households in return for militaryservice (Barlow, 2001, ch. 2.4). The earliestdescription of a Tai polity is found in the Bai-yizhuan, the report of two envoys sent to Luchuan(M. Mao) in 1396 (Wade, 1996). This politicalstructure is elaborate. The ruler (zhao) isassisted by a chief administrator (zhao-meng)and other officials who control units of peopleand are rewarded by corvée and taxes from adesignated territory. “When assembled thepeople are troops, when dispersed they arecivilians.” Levies of one in three or five areraised by official recruiting agents. Entry intoofficial ranks is so prized that, “When a youngman is given an official rank, his father andelder brother kneel and offer obeisance to him.”Nobles wear distinctive and luxurious clothes,are accorded great public respect, and travelaround with a thousand attendants such that their“their elephants, horses and servants fill theroad”. A courier service transmits messages overlong distances. With this system, the Bai-yi ruleover several different peoples, some of whichare described as “extremely black”, and someof which live in the mountains.

Many later studies have described thecapacity of Tai social systems to absorb otherpeoples by emulation. This appears to alreadybe the case in fourteenth century Luchuan:“While the language and customs of the variousyi differ, the Greater Bai-yi are the chiefs, and

thus the various yi sometimes imitate theirbehaviour” (Wade 1996, 7). Some of thelegends suggest that Mangrai, the greatest Taichief of all, is descended from the Chiang Saenlineage whose names suggest strongly they areLawa. He may have “become Tai” throughmarriage to the daughter of the Tai Lu ruler ofChiang Rung.

Terwiel (1980, 1981, 1983) compared socialhabits and cultural practices across differentgroups of Southwestern Tai from the Red Riverto the Brahmaputra. Besides the Tai linguisticstructure and affinity for rice-growing, themajor common elements were a cosmology witha sky god who cleanses the world throughdrought and flood before populating it withpeople, animals and knowledge; a physiologybased on the khwan (essence or life-force) ofeach body part; an affinity for various forms ofdivination using animals; the political structureof the muang; and the importance of villagespirits, usually associated with a jai ban (centralvillage pillar) or similar ritual objects, andanimal sacrifice. But in other respects, Terwielfound, Tai local cultures are very varied. Whilethe cosmologies share a sky god and humandescent into the world, the details of thedescent vary greatly, probably through theaccretion of local legends, histories and herostories. In many places, the mass of humanityis born out of a gourd which the gods place inthe world. Elsewhere, creation uses otherdevices including eggs, vines, and seeds. Eventhe gourd legend varies to suit differentsociologies. In some places, all humanityemerges from a single aperture. In others, thegods make a second hole with a burnt stickthrough which the Lawa emerge, becomingblackened in the process. In yet others, thereare several gourds from which different peopleemerge. The hero figures who make thetransition from divine to earthly existence -Borom/Bulom, Cheung, Lo - recur in legendsand ancestries found in different places, but therelations between them differ between versions(siblings, successors, allies). Similarly, manyrituals seem to have been borrowed or devel-

01P1-26 8/8/05, 14:3816

17From Yue To Tai


oped in the locality, particularly fertility ritualswhich require a strong relevance to the localecology. Along the middle Mekong, majorannual fertility festivals are focused around boatraces and processions. Further south, rocketfestivals are popular. Techniques of divinationsimilarly vary greatly from place to place,variously using pig’s livers, eggs, or chickens’feet, tongues or thighs.

Chinese

There are also stories of interchange withChinese on the frontier. The heroic Tai princeSaengto defends Kengtung against Chineseattacks and terrifies the Chinese emperor byfilling his cushions with wasps. As a result, theemperor gives him a daughter in marriage.Saengto lives in China but sends his sons backsouth to found new muang and new rulinglineages (Sao Saimong Mangrai 1981, 232-3).In the foundation legend of Mogaung, a daugh-ter of the Chinese emperor floats down theIrawadi on a raft, meets a “white tiger” who hadbeen her husband in a previous existence, andtogether they produce the four sons who ruleover Mogaung and its outlying muang (Sao HsoHom and Pu Loi Tun 1998, 244-5).

Tai were drafted into Chinese armies andtaken north. Tai warriors conducted raidingexpeditions into Yunnan and Sichuan (Wade1997). Chinese frontier officials and soldierstook local wives, or “went native” and faded intothe local population. According to a Chineserecord from 1499, “many criminals on-the-runfrom Jiangxi, Yunnan and Dali” took refuge inthe Shan regions (Daniels 2000, 87). All alongthis long frontier between Chinese expansion andTai migration, genes were being exchanged.

Li Fang Kwei suggests that the distinctive60-year calendar used by most Tai groupsoriginates from China before the sixth century.Cheah Yanchong (1988, 1996) argues that theTai adopted much Chinese political structure andvocabulary, including words such as khun, jao,and chiang; the idea of kin muang (eating thecity or state, a literal translation from the

Chinese, shi yi), and possibly also the moreformalised structure of the muang. He alsodraws attention to the similarity between Tai andChinese concepts of land/ancestor spirits, andthe probable Chinese origin of important wordssuch as thaen, the most common Tai name forthe sky god, and khwan, the body essence.Terwiel (1981, 151) also points out severalborrowings from Chinese kinship practice,including the use of kin numeratives. Maspero(1981) notes possible Chinese origins for Taimarriage rituals, the use of city and villagepillars, house spirits, and stories of descent intothe world. He also draws attention to possibleTai origins for some Chinese words and culturalpractices. Exchange went both ways.

The organisation of Tai armies in decimalunits (nai sip, nai roi) may have been adaptedfrom Chinese (and more especially Mongol)practice. Lu chronicles describe how theMongols used this ho sip system in the suppres-sion campaigns in southern Yunnan in thethirteenth century (Lemoine 1987, 131). TheMogaung chronicle describes how the fourruling brothers (produced by the liaison of aChinese princess with a local “white tiger”)travelled to the imperial court to learn statecraftand military strategy (Sao Hso Hom and PuLoi Tun 1998, 245). Some believe that ChiangMai’s nearly-square city plan was modelled onthe Chinese frontier garrison settlements.Vickery (1996, 180-2) suspects that thestructure of state administration adopted in Thaistates originated from these early Chinese contacts.

Conclusion

The early history of the Tai – whether in theNanchao story, the “Yue school”, or the counternarrative of “the Tai were always here” – ispursued as a search for “ancestors”, implying ablood line stretching back into the past. Thelinguists’ concept of a Proto-Tai language iseasily assimilated into this historical model toimagine the development of a Tai linguistic,cultural and ethnic diaspora from a single pointof origin.

01P1-26 8/8/05, 14:3817

18 CHRIS BAKER


But pushing ethnicities back into the past israther like pushing nation-states back into thepast. It imagines things which were not there.It obscures the changes which bring ethnicitiesand nation-states into being. Ethnonyms arealways historico-political constructs.

Linguists identify the origins of the Tailanguage family in sub-Yangzi China. The HanChinese described the varied inhabitants of thisarea with the generic term “Yue”. Recenttextual and archaeological study has stressedthat these societies were more complex than the“barbarian” tag implies. Over a long period fromaround 400 BC onwards, the Han progress intothe sub-Yangzi region caused major changes insociety, culture, and language. These changesevolved gradually because the armed invasionsand population inflows were obstructed bydisease and armed resistance. Eventually,however, there were two main results. SomeYue were absorbed into the mixed Han/localcultures which developed along the southerncoast. Others retreated into the mountains andvalleys of the hinterland. Both developmentscreated new languages–the hybridised dialects(Wu, Gan, Yue, Min) of the sinified coastalpolitics on the one hand, and the Tai family onthe other.

As linguists such as Diller and Dixon argue,the Tai language may have evolved fromseveral earlier tongues rather than from aspecific, identifiable sub-set of Yue. The Tailanguages, and their language communities,were formed by the processes or migration,resettlement, and contact in this tumultuousperiod begun from the Han intrusions. The arcof hills stretching from Guangxi to Assam wasalready an established trade route and a zone ofcultural diffusion. The spread process was nota singular, even flow, but successive waves withboth splits and joins. Probably there were twomajor stages in this development. In the first,lowland refugees and more established hill-dwellers (Ailao?) developed new societies in thehills behind the coastal plain. In the second,peoples moved westwards, precipitated by suc-cessive Chinese frontier pacification campaigns.

Projecting the family-tree of Tai languagesonto the map of the region suggests the overallpattern of this second phase. One major routeseems to have gone up into Yunnan; then downthe Mekong, Salween, and Irawadi riversystems; and finally further west to the Chindwinand the Brahmaputra. Another route seems tohave moved across the lower arc of the Annamitehills, then down into the valleys of the Mekongand Chaophraya. There may have been anothernorthern arc running west along the Yangzisystem, and then south through Yunnan to inter-sect with the others. Each of these trends wascomposed of many smaller short-term moves,with forks, dead-ends, and u-turns. There seemto have been certain periods of special activity -during the Han inroads from the third centuryBC; again in the seventh and eighth centuriesAD during the border pacfication campaigns;again in the eleventh century following the Nongrebellion; and finally in the twelfth to fourteenthcenturies with the Nanchao-Tali wars andMongol pressure.

This westwards movement brought thesemigrants into contact with many other peoples.This was not a pioneer flow into vacant territory.There were pre-existing settlements of Mon-Khmer speakers. Tibeto-Burmans were movinginto the same region from the northwest andViets from the south over the same era. Therewas constant contact and exchange with Chinese.

Earlier attempts to imagine this process havestressed the Tai success in defeating and subju-gating the earlier populations. Condominas(1990, 73) argued that the Tai diffusion was“the work of small groups of warriors led by anaristocracy which succeeded in imposingitself on numerous and varied groups of peoplecovering vast territories... These Tay chiefsimposed their own system of relations on thoseconquered, who henceforth became the produc-tive base of servile manpower.”25 Certainlyconquest and domination seems to have been amajor theme. These are the stories which findtheir way into legends, songs, and chronicles.But warrior domination should probably not beimagined as the sole or central theme.

01P1-26 8/8/05, 14:3818

19From Yue To Tai


Recent studies of the big movements oflanguage diffusion in the world have tended tostress that association with agricultural innova-tion or social innovation are what enables acertain language to spread and become domi-nant. The Tai languages may have spread alongthe hill tier not in association with rice, whichwas already present, but with better water-management techniques to increase and stabiliseyields.26 The evidence is indirect but intrigu-ing. Many Tai foundation legends stress thatthe migrants are searching for a particular kindof landscape - a flat plain between mountain andriver ideal for growing rice with water from hillstreams. This landscape echoes that of the moun-tain valleys in the area of language origin. It isechoed too in most of the Tai settlements acrossthe spread zone.

In addition, the Tai may have had (ordeveloped in the course of migration) someeffective social techniques. There is now a largeliterature which explains the muang as astructure for managing a varied population in agraded hierarchy, and which stresses the abilityof the Tai to absorb other peoples. But thisliterature is based on evidence from the lastcentury. We cannot simply project the findingsback into the past. Yet again there is someevidence that these techniques were present orevolving. Many of the foundation legends stressthe Tai newcomers’ readiness to adopt pre-ex-isting local practices, and their skill in drawingother groups into subordinate relationships. Theearliest account of political structure, in theBai-yi zhuan, suggests a complex of military,administrative, legal and cultural techniques tomanage a varied population. The Bai-yi zhuanalso hints at the famous capacity of the Taimuang to absorb others by emulation.

Finally, much of what became the features ofTai languages, cultures, and societies developedin the process of diffusion itself. Thesesocieties were highly mobile. Migrationscontinued over later centuries, resulting in acomplex mosaic of local sub-communities,constantly rearranged. In addition, Tai societieswere synthetic, with a high capacity to absorb

both peoples and cultural practices, and to adaptthemselves to new environments. Much culturalpractice was adopted locally from earlierinhabitants, including local spirits, divinationtechniques and fertility festivals. Muchsystematic learning was adopted from theChinese, especially software such as calendarsystems, political terminology, and perhaps thestructure of muang polities.

The Yue disappeared. New societies andcultures took shape out of the experience ofmigration, contact, and exchange. By thethirteenth century, the term “Tai” was being usedas a descriptor for some of these, principallyaround the upper Mekong. Only much later, andmainly as a result of work by western historiansand linguists, was the term given a widermeaning as a descriptor for a whole family oflanguages and peoples. Then the term becameavailable for various political purposes.

References

Alting von Geusau, Leo. 2000. Akha InternalHistory: Marginalization and the EthnicAlliance System. In: Andrew Turton ed.Civility and Savagery: Social Identity in TaiStates. London: Curzon.

Anderson, James Adams. 2001. MonumentalPride: Sino-Vietnamese Cross-borderCommemorations of Nùng Trí Cao. PAHSeminar Paper, 12 June.

Aroonrut Wichienkeeo. 2000. Lawa (Lua): AStudy from Palm-Leaf Manuscripts and StoneInscriptions. In Yukio Hayashi and GuangyanYan eds. Dynamics of Ethnic Cultures AcrossNational Boundaries in Southwestern Chinaand Mainland Southeast Asia. Chiang Mai:Ming Muang.

Aroonrut Wichienkeeo. 2002. “ ‘Lua LeadingDogs, Toting Chaek, Carrying Chickens’Some Comments.” In Yukio Hayashi andAroonrut Wichienkeeo eds. Inter-EthnicRelations in the Making of Mainalnd South-east Asia and Southwestern China. Bangkok:Amarin.

Backus, Charles. 1981. The Nanchao Kingdom

01P1-26 8/8/05, 14:3819

20 CHRIS BAKER


and T’ang China’s Southwestern Frontier.Cambridge: Cambridge University Press.

Bai Chun. 2002. Tai seuksa nai chin (TaiStudies in China). Bangkok: 5 Area StudiesProject.

Ballard, W. L. 1981. Aspects of the linguistichistory of South China. Asian Perspectives,24 (2).

Barlow, Jeffrey G. 1987. The Zhuang MinorityPeoples of the Sino-Vietnamese Frontier inthe Song Period. Journal of Southeast AsianStudies 18 (2).

Barlow, Jeffrey. 2001. The Zhuang: A Longitu-dinal Study of Their History and TheirCulture. At: http://mcel.pacificu.edu/as/resources/zhuang/, version last updated:January 22, 2001.

Bauer, Robert S. 1996. The Cantonese-TaiRelationship within the Southeast AsianSprachbund. Bangkok: Institute of Languageand Culture for Rural Development, MahidolUniversity.

Bellwood, Peter. 1996. The origins and spreadof agriculture in the Indo-Pacific region:gradualism and diffusion or revolution andcolonization?. In: David R. Harris (ed.),The Origins and Spread of Agriculture andPastoralism in Eurasia, London: UCL Press.

Cam Trong. 1998. Baan-müang: A Character-istic Feature of the Tai Socio-political Sys-tem. Tai Studies 3 (2).

CeDRASEMI. 1985. Carte des Ethnies deLangues Thai. Map of Thai Speaking Groups.Also published in Journal of the SiamSociety, 76, 1988.

Chamberlain, James R. 1975. A New Look atthe History and Classification of the TaiLanguages. In: J. G. Harris and J. R.Chamberlain eds. Studies in Tai Linguisticsin Honour of William J. Gedney. Bangkok:Central Institute of English Language, Officeof State Universities.

Chamberlain, James R. 1986. Remarks on theOrigins of Thao Hung or Cheuang. In: R. J.Bickner, T. J. Hudak, and PatcharinPeyasantiwong eds. Papers from a Confer-ence in Honour of William Gedney. Ann

Arbor: Centre for Southeast Asian Studies,University of Michigan.

Chamberlain, James R. 1989. Thao Hung orCheuang: A Tai Epic Poem. Mon-KhmerStudies 18-19.

Chamberlain, James R. 1991. The Efficacy ofthe P/PH Distinction in Tai Languages. In: J.R. Chamberlain ed. The Ram KhamhaengControversy: Collected Papers. Bangkok:The Siam Society.

Chamberlain, James R. 1992. The Black TaiChronicle of Muang Mouay, Part I, Mythol-ogy. Mon-Khmer Studies 21.

Chamberlain, James R. 1998a. The Origin of theSek: Implications for Tai and VietnameseHistory. Journal of the Siam Society 86 (1 & 2).

Chamberlain, James R. 1998b. A Critical Frame-work for the Study of Thao Houng orCheuang. In: Tamnan kieo kap thao hung thaochuang: miti thang prawattisat laewatthanatham. Bangkok: Thai Khadi Seuksa.

Chang, Te-Tzu. 1993. The origin and earlycultures of the cereal grains and foodlegumes. In: D. N. Keightley (ed.), Originsof Chinese Civilisation. Berkeley, Los Ange-les, London: University of California Press.

Cheah Yanchong. 1988. The Ancient Culture ofthe Tai People: The Impacts of the Hua XiaCulture on it, and its Implication. Journal ofthe Siam Society 76.

Cheah Yanchong. 1996. More Thoughts on theAncient Culture of the Tai People. Journalof the Siam Society 84.

Chen Han-seng. 1949. Frontier Land Systemsin Southernmost China. New York: Instituteof Pacific Relations.

Chen Lufan. 1990. Whence Came the Thai Race:An Inquiry. Kunming.

Cholthira Satyawadhna. 1997. Ethnic Inter-Relations in the History of Lanna: Reconsid-ering the Lwa Role in the Lanna Scenario.Tai Culture 2 (2).

Cholthira Satyawathana. 2001. Pai yew: kanseuksa cheung manutwitya (The Bai Yue:An Anthropological Study), Bangkok:Khrongkan prawatisatsangkhom laewatthanatham chonchat tai.

01P1-26 8/8/05, 14:3820

21From Yue To Tai


Colquhoun, Archibald Ross. 1885. Amongst theShans. London: Field and Tuer etc.

Condominas, Georges. 1976. Essai sur l’evolu-tion des systèmes politiques thaïs. Ethnos(Stockholm) 41 (1-4).

Condominas, Georges. 1990. From Lawa to Mon,from Saa’ to Thai: Historical and Anthropolo-gical Aspects of Southeast Asian Social Spaces.Canberra: Australian National University.

Daniels, Christian. 1996. Agro-Industries:Sugarcane Technology. Vol VI, Part III of JosephNeedham, Science and Civilisation in China.Cambridge: Cambridge University Press.

Daniels, Christian. 2000. “The Formation of TaiPolities between the 13th to 16th Centuries:The Role of Technological Transfer.”Memoirs of the Research Department of theToyo Bunko, 58.

Diller, Anthony. 1990. Phasa trakun tai: boenglang naewkhit baep upalak [The Tai languagefamily: background to the comparativemethod]. In: B. J. Terwiel, A. Diller, CholthiraSatyawatthana, Khon tai (doem) mai dai yuthi ni [The (old) Tai were not here]. Bangkok:Muang Boran

Diller, Anthony. 1992. Tai Languages in Assam:Daughters or Ghosts? In: Carol J. Comptonand John F. Hartmann eds., Papers on TaiLanguage, Linguistics, and Literatures.University of Northern Illinois.

Diller, Anthony. 2000. The Tai LanguageFamily and the Comparative Method. In:Proceedings of the International Conferenceon Tai Studies, July 29-31, 1998. Bangkok:Institute of Language and Culture for RuralDevelopment, Mahidol University.

Dixon, R. M. W. 1997. The Rise and Fall ofLanguages. Cambridge: Cambridge Univer-sity Press.

Dodd, W. C. 1923. The Tai Race: Elder Brotherof the Chinese. Cedar Rapids: Torch Press.

Doré, Amphay. 1987. Jalons pour une Histoiredu Lanna avant le 13e Siècle: une approcheethno-historique. In: Ann Butler ed. Proceed-ings of the International Conference onThai Studies, Australia National University,Canberra, 3-6 July.

Driem, George van. 1998. Neolithic correlatesof ancient Tibeto-Burman migrations. In:Roger Blench and Matthew Spriggs (eds.)Archaeology and Language, Vol. II: Corre-lating Archaeological Data and LinguisticHypotheses, London and New York: Routledge.

Eberhard, Wolfram. 1968. The Local Culturesof South and East China. Leiden: E. J. Brill.

Evans, Grant. 1999. Introduction: What is LaoCulture and Society? In: Evans (ed.), Laos:Culture and Society, Chaing Mai: SilkwormBooks.

Gait, Edward. 1994 [1926]. A History of Assam.Guwahati: Lawyer’s Book Stall.

Gedney, William. 1964. A Comparative Sketchof White, Black and Red Tai. Bangkok:Social Science Review.

Glover, Ian C. and Charles F. W. Higham. 1996.New evidence for early rice cultivation inSouth, Southeast and East Asia. In: David R.Harris (ed.), The Origins and Spread ofAgriculture and Pastoralism in Eurasia,London: UCL Press.

Hartmann, John F. 1980. A Model for theAlignment of Dialects in Southwestern Tai.Journal of the Siam Society 68 (1).

He Zheng-ting. 1998. On the National War ledby Nong Zhi-gao. In: Tamnan kieo kap thaohung thao chuang: miti thang prawattisat laewatthanatham. Bangkok: Thai Khadi Seuksa.

Hoang Luong. 1998. Nung Zhi Cao in theHistory of Vietnam. In: Tamnan kieo kap thaohung thao chuang: miti thang prawattisat laewatthanatham. Bangkok: Thai Khadi Seuksa.

Hoshino, Tatsuo. 2002. Wen Dam and itsNeighbours: The Central Mekong Valley inthe Seventh and Eighth Centuris. In: MayouryNgaosrivathana and Kennon Breazeale(eds), Breaking New Ground in Lao History:Essays on the Seventh to Twentieth Cen-turies. Chiang Mai: Silkworm Books.

Hsieh, Shih-ching. 1995. On the dynamics ofTai/Dai-Lue ethnicity. An ethnohistoricalanalysis. In: Steven Harrell (ed.), CulturalEncounters on China’s Ethnic Frontiers.Seattle and London: University of Washing-ton Press.

01P1-26 8/8/05, 14:3821

22 CHRIS BAKER


Izikowitz, K. G. 1981. Propos sur les Thaïs.In: Orients pour Georges Condominas, Paris:Sudestasie Privat.

Lacouperie, Terrien de. 1885. “The Cradle of theShan Race”, Introduction to Colquhoun(1885).

Lebar, Frank et al. 1964. Ethnic Groups of Main-land Southeast Asia. New Haven: HumanRelations Area Files Press.

Lemoine, Jacques. 1987. Tai Lue HistoricalRelation with China and the Shaping of theSipsong Panna Political System. In: AnnButler ed. Proceedings of the InternationalConference on Thai Studies, AustralianNational University, Canberra, 3-6 July.

Li Fang Kuei. 1960. A Tentative Classificationof Tai Dialects. In: Culture in History: Essaysin Honour of Paul Radin, ed. S. Diamond.New York: Columbia University Press.

Li Fang Kuei. 1977. A Handbook of Compara-tive Tai. Honolulu, University of Hawaii Press.

Liu Yan. 1998. An On-the-spot Investigation intothe Southerly Emigration of the Dai Race.Kunming: Yunnan Nationality Press.

Luce, G. H. 1958. The Early Syam in Burma’sHistory. Journal of the Siam Society 46 (2).

Luce, G. H. 1959. The Early Syam in Burma’sHistory: A Supplement. Journal of the SiamSociety 47 (1).

Luce, G. H. 1961. Man Shu: Book of theSouthern Barbarians. Ed. G. P. Oey. CornellData Paper 44. Ithaca: Cornell University Press.

Luce, G. H. 1965. Rice and Religion. Journal ofthe Siam Society 53 (2).

Luo Yongxian. 2001. The hypothesis of a newbranch for the Tai languages. In: M.R. KalayaTingsabadh and Arthur S. Abramson eds.,Essays in Tai Linguistics. Bangkok:Chulalongkorn University Press.

Maspero, Henri. 1981. Taoism and ChineseReligion, tr. F. A. Kierman. Amherst: Univer-sity of Massachusettts.

Meacham, William. 1983. Origins and Devel-opment of the Yïeh coastal neolithic: amicrocosm of cultural change on the main-land of East Asia. In: D. N. Keightley (ed.),The Origins of Chinese Civilization. Berke-

ley, Los Angeles, London: University ofCalifornia Press.

Mote, Frederick W. 1964. Problems of ThaiPrehistory. Social Science Review 2 (2).

Nichols, Johanna. 1997. The epicentre of theIndo-European linguistic spread. In: RogerBlench and Matthew Spriggs (eds.) Archae-ology and Language, Vol. I: Theoretical andMethodological Orientations, London andNew York: Routledge.

Nichols, Johanna. 1998. The Eurasian spreadzone and the Indo-European dispersals. In:Roger Blench and Matthew Spriggs (eds.)Archaeology and Language, Vol. II: Corre-lating Archaeological Data and LinguisticHypotheses, London and New York:Routledge.

Norman, Jerry. Chinese. 1988. Cambridge: Cam-bridge University Press.

Notton, Camille tr. 1926. Annales du Siam,I: Chroniques de Suvanna Khamdaeng,Suvanna K’om Kham, Sinhanavati. Paris,Limoges, Nancy: Charles-Lavauzelle.

Notton, Camille tr. 1930. Annales du Siam, II:Chronique de La:p’un, Histoire de laDynastie Chamt’evi. Paris, Limoges, Nancy:Charles-Lavauzelle

O’Connor, Richard A. 1995. AgriculturalChange and Ethnic Succession in SoutheastAsian States: A Case for Regional Anthro-pology. Journal of Asian Studies 54 (4).

Pan Qixu. 1990. The Historical and CulturalRelations between the Zhuang, the Lao andthe Thai in the Light of Linguistics. In:Proceedings of the 4th International Confer-ence on Thai Studies, Vol. III, Kunming:Institute of Southeast Asian Studies.

Phatthiya Yimrewat. 2001. Prawattisat sipsongchuthai. Bangkok: Sangsan.

Phornphan Jantharonanan. 1988. Pai yew:banphurut khong chonchat tai? In: Khon tainok prathet phromdaen haeng khwam ru [Taioutside the country: borders of knowledge],papers of annual seminar of the HistoricalAssociation, 28-29 January.

Proschan, Frank. 1998. Cheuang in KhmuFolklore, History, and Memory. In: Tamnan

01P1-26 8/8/05, 14:3822

23From Yue To Tai


kieo kap thao hung thao chuang: miti thangprawattisat lae watthanatham. Bangkok:Thai khadi Seuksa.

Pulleyblank, E. G. 1983. The Chinese and theirNeighbours in Prehistoric and Early HistoricTimes. In: D. N. Keightley ed., The Originsof Chinese Civilization, Berkeley, Los Angeles,London: University of California Press.

Ramsey, S. Robert. 1987. The Languages ofChina. Princeton University Press.

Ratanaporn Sethakul. 1998. Lan Na History inthe Lü Chronicles. Tai Culture 3 (1).

Renfrew, Colin. 1997. World linguistic diversityand farming dispersals. In: Roger Blench andMatthew Spriggs (eds.) Archaeologyand Language, Vol. I: Theoretical andMethodological Orientations, London andNew York: Routledge.

Renu Wichasin. 1998. Sangkhep phumlangkhong chao tai yai nai rat chan (OutlineBackground of the Thai Yai in the ShanStates). In: Chalatchai Ramitanon, WiradaSomsawadi, and Renu Wichasin eds., Tai TaiTai, Bangkok: Toyota Foundation.

Roux, Henri. 1954. Quelques minorités ethniquesdu Nord-Indochine. France-Asie 92-3.

Sai Aung Tun. 2000. The Tai Ethnic Migrationand Settlement in Myanmar. In: YukioHayashi and Guangyan Yan eds. Dynamicsof Ethnic Cultures Across National Bounda-ries in Southwestern China and MainlandSoutheast Asia. Chiang Mai: Ming Muang.

Sao Hso Hom and Pu Loi Tun. 1998. TheFounding of Mong Kawng: A Legendary TaiState in Upper Myanmar. In: Tamnan kieokap thao hung thao chuang: miti thangprawattisat lae watthanatham. Bangkok:Thai Khadi Seuksa.

Sao Saimong Mangrai. 1965. The Shan Statesand the British Annexation. Ithaca: CornellUniversity Press.

Sao Saimong Mangrai. 1981. The PadaengChronicle and the Jengtung State ChronicleTranslated. Ann Arbor: University of Michigan.

Sawaeng Malasaem. 1997. Khon yong yaiphaendin. Bangkok: Thammasat UniversityPress.

Schafer, Edward H. 1967. The Vermilion Bird:T’ang Images of the South. Berkeley and LosAngleles: University of California Press.

Smalley, William. 1965. Cian: Khmu? CulturalHero. In: Felicitation Volumes of SoutheastAsian Studies Presented to His HighnessPrince Dhaninivat Kromamun BidyalabhBridhyakorn, I. Bangkok: Siam Society.

Somphong Withayasakdiphan. 2001. Prawatisattai yai. Bangkok: Krongkan prawatisatsangkhom lae watthanatham chonchat tai.

Sratsawadi Ongsakun. 1996. Prawattisat lanna.Bangkok: Amarin Printing.

Srisak Walipodom and Suchit Wongthet. 1991.Thai noi thai yai thai sayam [Big Thai, LittleThai, Siam Thai]. Bangkok.

Srisak Walipodom. 1978. Nakhon Sithammaratkap prawatisat thai [Nakhon Sithammarat andThai history]. Muang Boran 4 (3), April-June.

Stuart-Fox, Martin. 1998. The Lao Kingdom ofLan Xang: Rise and Decline. Bangkok: WhiteLotus.

Suchit Wongthet. 1986. Khon thai yu thini [TheThai were here]. Muang Boran, special number.

Suchit Wongthet. 1994. Khon thai yu thini naiutsakane [The Thai were here in SoutheastAsia]. Bangkok: Silpakon University.

Tan, Leshan. 1994. The Tai before the ThirteenthCentury: A Perspective from ChineseRecords. Journal of the Siam Society 82 (1).

Tanabe, Shigeharu. 2000. Autochthony and theInthakhin Cult of Chiang Mai. In: AndrewTurton ed. Civility and Savagery: SocialIdentity in Tai States. London: Curzon.

Taylor, K. W. 1983. The Birth of Vietnam.Berkeley, Los Angeles and London: Univer-sity of California Press.

Terwiel, B. J. 1978. The Origins of the T’aiPeoples Reconsidered. Oriens Extremus 25(2).

Terwiel, B. J. 1980, 1981. The Tai of Assam andAncient Tai Ritual. 2 vols. Gaya: Centre forSouth East Asian Studies.

Terwiel, B. J. 1983. Ahom and the Study of EarlyThai Society. Journal of the Siam Society 71.

Terwiel, B. J. 1990. Kan khonha laeng kamnoetkhong chonchat tai: bot pramoen khwam hen

01P1-26 8/8/05, 14:3823

24 CHRIS BAKER


tangtang lae naewkhit mai [Searching for thebirthplace of the Tai peoples: review ofopinions and a new direction]. In: B. J.Terwiel, A. Diller, Cholthira Satyawatthana,Khon tai (doem) mai dai yu thi ni [The (old)Tai were not here]. Bangkok: Muang Boran.

Turton, Andrew. 2000. Introduction. In: Turton(ed.), Civility and Savagery: Social Identityin Tai States, Richmond: Curzon.

Vickery, Michael. 1996. The Constitution ofAyutthaya: The Three Seals Code. In:Andrew Huxley ed. Thai Law: BuddhistLaw. Essays on the Legal History ofThailand, Laos and Burma. Bangkok: WhiteOrchid.

Vinai Pongsripian. 1990. Nan-chao and the Birthof Sukhothai: Problems of the Twentieth-Century Thai Perception of the Past. In:Proceedings of the 4th International Confer-ence on Thai Studies, Vol. IV, Kunming:Institute of Southeast Asian Studies.

Wade, Geoff. 1996. The Bai-yi zhuan: AChinese Account of a Tai Society in the 14th

Century. Paper presented at the 14th IAHAConference, Bangkok.

Wade, Geoff. 1997. An Epilogue by theTranslator of this Text. Tai Culture 2 (1).

Wade, Geoffrey P. 1994. The Ming Shi-lu(veritable records of the Ming Dynasty) as asource for Southeast Asian history, 14th to 17th

centuries. Ph.D thesis, University of HongKong.

Wang Gungwu. 1958. The Nanhai Trade: AStudy of the Early History of Chinese Tradein the South China Sea. Journal of theMalaysian Branch of the Royal Asiatic Soci-ety 31(2).

Wang Yizhi. 1990. The Origin of the DaiNationality. In: Proceedings of the 4th

International Conference on Thai Studies,Vol. III, Kunming: Institute of SoutheastAsian Studies.

Watabe Takaya. 1998. Laeng kamnoet lae kanphrae krajai khong khao thi pluk nai thawipasiya [The birthplace and spread of rice-growing in Asia]. In: Charnvit Kasetsiri andHayao Fukui (eds) Yipun-thai-utsakhane

[Japan-Thai-Southeast Asia]. Bangkok:Thammasat-Kyoto Core University Program.

Wichitmatra, Khun. 1928. Lak Thai [Origins ofthe Thai]. Bangkok: Hanghunsuan.

Wiens, Herold J. 1967. Han Chinese Expansionin South China. Hamden Conn.: Shoe StringPress

Zhengzhang, Shangfang. 1991. Decipherment ofYue-Ren-Ge (Song of the Yue Boatman).Cahiers de Linguistique Asia Orientale 20(1).

Zhou Hongyun and Ke Yuanxiao. 1990. On therelationship between Dehong Dai and AhomDai. In: Proceedings of the 4th InternationalConference on Thai Studies, Vol. III, Kunming:Institute of Southeast Asian Studies.

Notes

1 With thanks to Acharn Chatthip and AcharnCheah who helped me see the landscape. Thanksalso to Geoff Wade, Michael Vickery, GrantEvans, Paul Sidwell and Thongchai Winichakulwho helped with sources and criticism but whoare in no way guilty of the result.

2 There were several variants of the sameplot, reviewed in Terwiel (1978, 1990).

3 I am greatly indebted to Cholthira (2001,3-31) who reviewed this literature. See also Bai(2002).

4 Diller’s rope image (28) is strikingly simi-lar to Dixon’s graphic portrayal of the same point(101). The processes underlying Diller’s andDixon’s observations are different, but here I’mnoting the similarity of the pattern.

5 This was the insight behind JaredDiamond’s best-seller, Guns, Germs and Steel.

6 Wiens (1967) and Eberhard (1968)prepared the way for this school by calling therice-growing communities in early sub-YangziChina “Tai”. In effect, they applied a modernethnonym to these early communities on groundsthat they exhibited certain cultural traits laterassociated with the Tai. In other words, Eberhardand Wiens identified and named the ancestorsby reference to their supposed descendants, andthen the Yue school repeated the process in

01P1-26 8/8/05, 14:3824

25From Yue To Tai


reverse, achieving a perfectly circular argument.Both groups assume that ethnicities areessential, and ignore when and how they arehistorically constructed.

7 The editor of the Han Shu said, speaking ofthe Yue region: “The southern regions are humid;as summer approaches it becomes terribly hot. Ifyou live near water, then poisonous snakes willcome. There are many plagues and pestilences;of ten troopers, two or three will die beforeblades are even bloodied” (Barlow 2001, ch. 3).

8 “The farther south one goes from theYangtze, the more tones one hears in theChinese dialects along the way”, also moretrailing modifiers and more everyday words thatare not Mandarin (Ramsey 1987, 36). See themaps in Norman (1988, 184) and Ramsey (1987,Figures 5 and 6). Below an oblique line runningroughly from the Vietnam-Yunnan-Guangxiintersection to Shanghai, the dialects spoken(including Cantonese, Hakka and Teochiew) arestructurally “Chinese”, but have absorbedelements from the older languages from the area.Ballard (1981, 176) concludes: “Cantonese(Yueh) shows the clear influence of Tai-likelanguages; some Min areas show similarinfluences. Min shows affinities with Yao, andWu/Chu even more so with Miao. We cansuggest, then, that the original populations inthese areas before sinicization were related tothese ethnic groups.” Bauer (1996) shows thatsome non-Chinese words in Cantonese probablycame from the same origins as Tai.

9 “The Ba, Shu, I, Liao, Qi, Li, Chu, Yue, alltwitter like birds and call like animals and theirlanguages are not the same; they are as differentfrom each other as are monkeys, snakes, fish,and tortoises.... The river valleys in which theylive extend for thousands of miles, they can begoverned with an intelligent loose rein, but theirpeople cannot be controlled. They have irrigatedfields and seldom farm dry fields, and engage infishing.” (Barlow 2001, ch. 4, from Wei Shu).

10 Chamberlain’s work is controversial formany reasons. First, he actively seeks connec-tions between linguistic events and other sortsof historical events. Purist linguists find this too

ambitious. Second, his work was used (byothers) in the controversy over the origins ofSukhothai Inscription I. Third, Chamberlain isoften abstruse and unclear. The point aboutChamberlain’s work is that he is looking forpatterns which have historical meaning, ratherthan conducting classification for classification’ssake. There are many different ways to classifydialects, resulting in very different groupings.See the original sketch by Li (1960), and thereview by Hartmann (1980).

11 The relatively high degree of mutualcomprehension among speakers of differentmembers of the Tai language family suggeststhat the dispersion was both recent and quick.But in the light of recent linguistic theory, it isunnecessary to stick to the timetable of 2,000 years.

12 The map is based on CeDRASEMI (1985),Lebar et al. (1964) and the map in theEncyclopaedia Britannica Languages of theWorld entry on the Tai languages (1995, 714).It omits the large Tai-speaking areas to the south.For simplicity, Chiang and its variants (Jiang,Xhiang) has been abbreviated as C., and Muangand its variants as M.

13 See the routes and map reconstructed froma ninth century Chinese source in Luce (1961).One route travelled up the Red River and thenacross the plains to Tali. A second went almostdirectly westwards from Tali to the Brahmaputra.A third passed southwest from Tali to the Irawadi(around the site of Mandalay) and then turnednorthwest to join the second route on theBrahmaputra (around the later Ahom area).

14 Note I am not arguing here, as some do,that “Tai” were migrating into this area from theTang era or earlier. Rather I am suggesting therewas movement into this area over a long period,at the end of which there are peoples who seemto be relatively identifiable as the Shan.

15 I am abbreviating muang/maung/mong as“M.” to simplify the problem of transliteration.

16 Doré (1987) times it much earlier.17 “It is a mean rustic place, of no use what-

ever. The people are all obstinate, stupid andignorant. If we get the land, it can hardly becounted as an asset” (Luce 1959, 79).

01P1-26 8/8/05, 14:3825

26 CHRIS BAKER


18 The Silpa Watthanatham historians arguethat the military pressure was not decisive, andthat these groups were pulled southwards bytrade routes extending upward from the coast.But their evidence is elusive.

19 There is one hint: Tatsuo Hoshino (2002)argues that Wen Dan and associated places foundin the seventh century Tang annals were Taisettlements around the middle Mekong. But hisevidence for the Tai in these places is based ondetecting “a very Daic ring” in the Chineserendering of place names and titles.

20 This pattern is similar to the assumptionof Srisak and Suchit (1991). But the timing isdifferent.

21 These legends have survived through oraltransmission and constant copying, and hencemay have changed over time. As evidence ofspecific events or dates, they are a difficult source.Here, however, I am using the story content.

22 Renfrew and others stress that agriculturalinnovations tend to spread east-west (rather thannorth-south) because temperature, day-length,and other climatic factors remain similar.Bellwood suggests rice may gradually have

moved southward from the Yangzi becauseprevailing temperatures were cooling. Thepassage across the hills is of course roughlywestward.

23 Pulleyblank (1983, 430) thinks that beforethe Ming era, the Chinese referred to the Zhuangas “Tung” which meant a mountain valley or“level ground between cliffs and beside astream”.

24 Luce (1965) speculates that, some centu-ries earlier, the Mon-Khmer went through aprocess of dispersion very similar to that of theTai, and following a similar path from the coastof southwestern China, along the Red River toYunnan, south down the Irawadi, and finally overthe hills to Assam and India, where isolatedMon-Khmer speakers (Munda) are still foundon the hills behind the northeastern coast.

25 See, however, Vickery’s review in the ThaiYunnan Project Newsletter, 13 June 1991.

26 See also O’Connor (1995) which uses asimilar argument to rationalize the dominationof Tai and other ethnicities from the north ofmainland Southeast Asia.

01P1-26 8/8/05, 14:3826

Documents

From Yue To Tai...The second theory began among Chinese historians, especially Jiang Ng Liang whose Tai ju su (On the Thai nation, 1983) identified the Yue in early sub-Yangzi China