Laos Nationhood

Asian Ethnicity, Volume 3, Number 2, September 2002

Nation/Representation: EthnicClassi� cation and Mapping Nationhoodin Contemporary Laos

VATTHANA PHOLSENA

(Centre for South-East Asian Studies, Department of Politics andAsian Studies, Hull University, United Kingdom)

In this article, I analyse ethnic classi� cations in contemporary Laos, starting with a briefreview of previous policies. I � rst look at the ideologies that have in� uenced the Lao ethnicclassi� cation, namely those of the former Soviet Union, China and Vietnam. Through ananalysis of the construction of the latest of� cial census (August 2000), I suggest a closerelationship between ethnic categorisation and the government’s nationalist discourse, stillstrongly in� uenced by guidelines on the concept of the nation disseminated by KaysonePhomvihane—the � rst President of the Lao PDR, now deceased but celebrated at presentin Laos as the inspirational � gure of the regime.

My intention is thus to demonstrate how one technology of power in particular—the Laopopulation census based on ethnic criteria—attempts to map the nation’s ‘invisible’ethnicity through a dual process, namely the objecti� cation of the Other ethnic groups’arbitrarily de� ned cultural features on the one hand, and the erasure of the dominant ethnicgroup’s ethnicity (the ethnicity of the ethnic Lao) on the other. But this technology of poweris limited, as the names and the lists of ethnic groups have remained subject to alterationevery few years. Twenty years after Kaysone Phomvihane urged for a change in ethnicterminology and classi� cation, the regime has yet to legitimise de� nitively the total numberof ethnic groups in the Lao PDR.

Introduction

The modern State, in the Foucauldian sense, is that hegemonic apparatus whose raisond’etre is to control and to administer the body of the population through a series ofdiscourses and technologies of power that together form the ‘regime of truth’, whichFoucault de� ned as follows:

Each society has its regime of truth, its ‘general politics’ of truth: that is, the types of discoursewhich it accepts and makes function as true; the mechanisms and instances which enable oneto distinguish true and false statements, the means by which each is sanctioned; the techniquesand procedures accorded value in the acquisition of truth; the status of those who are chargedwith saying what counts as true.1

Truth is not transcendental, ‘out there’: it is produced here and now. Indeed, modernstates, among other agents, participate in a continuous and uninterrupted process ofgenerating ‘truth’ through the use of ‘technologies of power’ in order to legitimate andnaturalise their authority. They transform innovations into everyday practice ‘by constant

1 Michel Foucault, Micro� sica del Potere (Turin, 1977), in Colin Gordon (ed.), Power/Knowledge, SelectedInterviews and Other Writings 1972–1977 (The Harvester Press, Brighton, Sussex, 1980), p. 131.

ISSN 1463-1369 print; 1469-2953 online/02/020175-23 Ó 2002 Taylor & Francis LtdDOI: 10.1080/1463136022013272 7

176 Vatthana Pholsena

reiteration of [their] power through what have become accepted as natural (rational andnormal) state functions, of certifying, counting, reporting, registering, classifying, andidentifying’.2 Likewise, Giddens’ theory of the modern state has pointed out the state’smonopoly over the means of violence, both as regards surveillance and internal paci� cation(police) and in the form of military power.3 Directly inspired by Foucault, Giddens employsthe concept of surveillance, a capacity strongly developed by the modern state. The entirepopulation is now subordinated to the control of the state, and the state has the capacity topenetrate, monitor and control the population’s everyday life.

My intention in this article is thus to show the determinant role that administration ingeneral, and population censuses in particular, play in the modelling of Lao society in theimage of a national community. In other words, the State in modern Laos has operated thecensus as a vector of ethnicity (through the manipulation of ethnic boundaries) in order tofashion an imaginary nationhood out of real heterogeneity.

1. Description and Interpretation of the Early Censuses

Insightful works have shown the long-lasting impact of the knowledge produced by colonialadministrations concerning the independent states they once governed.4 For instance, Cohn,in an article that analyses the conduct of the censuses in India under British rule, argues thatthe censuses that classi� ed the Indian population into castes signi� cantly in� uenced‘scholars’ and scientists’ views on the nature, structure and functioning of the Indian castesystem’.5 Indeed, the colonial State contributed to shaping the ruled population accordingto a new conceptual framework by operating instruments of domination, or technologies ofpower, which included the census, the map and the museum.6

The French Administration

The French censuses of the population of colonial Laos were basic. The data reported inthe 1911, 1921, 1931, 1936 and 1942 censuses (see Table 1) were mainly based on generaladministrative reports, such as the Annuaire Statistique du Laos and the Annuaire Statis-tique de l’Indochine.7 Only major groups were listed. There were nine of them, namely, the‘Lao’, ‘Tai’, ‘Kha’, ‘Meo-Yao’, ‘Vietnamese’, ‘Chinese’, ‘European’, ‘Cambodian’, ‘Indianand Pakistani’ (though, for the two last groups, the data are patchy and are completelymissing from the 1942 survey). The table gives the ‘Ethnic Composition of the Populationof Laos’ as a general title, and the very neutral term of ‘group’ is used to head the columnof ethnic classi� cations. Therefore, presumably, the nine statistical categories were con-

2 Bernard S.Cohn and Nicholas. B. Dirks, ‘Beyond the Fringe: The Nation State, Colonialism, and The Technologiesof Power’, Journal of Historical Sociology, vol. 1, no.2 (1988), p. 225.

3 Anthony Giddens, The Nation-State and Violence. Volume Two of A Contemporary Critique of HistoricalMaterialism (Polity Press, Cambridge, 1985).

4 See, for instance, Charles Hirschman, ‘The Making of Race in Colonial Malaysia: Political Economy and RacialIdeology’, Sociological Forum, vol. 1&2 (1986), pp. 330–62; and ‘The Meaning and Measurement of Ethnicityin Malaysia: An Analysis of Census Classi� cations’, The Journal of Asian Studies, vol. 46, no. 3 (1987),pp. 555–82.

5 Bernard S. Cohn, ‘The Census, Social Structure and Objecti� cation in South Asia’, in Bernard S. Cohn, AnAnthropologis t among the Historians and Other Essays (Oxford University Press, Oxford, New York, 1987),p. 242.

6 Benedict Anderson, Imagined Communities: Re� ections on the Origins and Spread of Nationalism (Verso,London, 1991), p. 167.

7 For the 1911 census � gures, the data came from the Bulletin de l’Of� ce colonial (No. 62, February 1913).

Nation/Representation 177

sidered as ‘ethnic groups’.8 Yet, more detailed statistics were available from other sources,such as military reports, ethnographic studies (conducted by both French and foreignresearchers), and provincial statistical records. But the subcategories were simply not listedin the ethnic composition of the population related to the whole territory. There seem to beseveral related reasons for the use of the basic classi� cation.

The French administration classi� ed the population into categories that followed raciallines; hence, inevitably, the use of rough methods of categorisation. That could also explainwhy the ‘Lao’ and ‘Tai’ groups were invariably placed at the top of the list, despite the factthat the ‘Kha’ outnumbered the ‘Tai’ population in general (see Table 1). The classi� cationwas indeed based on the assumption that the ‘Kha’ were not situated at the same level ofcivilisation as the ‘Lao’ and ‘Tai’.9 But then, one could equally well ask why the‘Europeans’ were usually put at the bottom of the list. Economically as well as strategically,French authorities saw Laos as a mere extension of Vietnam—especially after the failureof the mise en valeur policy.10 Consequently, marking the domination of the ruling Whitesover the subordinated Asians was probably not such an issue, in a possession that has beencalled a ‘colonial backwater’.11 The poor system of data-collection, aggravated by poorcommunications, especially in the upland areas, was certainly another factor that con-tributed to the lacunae within, and inaccuracy of, the statistics.12 However, the non-utilityof a detailed ethnic classi� cation was probably the primary reason for keeping the censusat such a basic level. The taxation administration needed only simple distinctions toadminister the population: the Lao and the non-Lao (the Europeans (French and others) andthe Asian foreigners (including Chinese) were subject to a different tax system13). Later on,the system became slightly more complex, but the classi� cation still followed the samepattern as that of the census,14 or may, alternatively, have simply been following theadministration’s interests? In any case, rough, i.e. racial, categories were apparentlysuf� cient for the functioning of the colonial tax system which was emphatically thebackbone of French Laos.

8 However, in 1953 the term ‘nationalities’ was used for the heading of a census referring to ‘European, Vietnamese,Chinese and Cambodian’ populations. These appeared to be Lao citizens, since there is another table referringto the same population based on the same source but which gives the list of the ‘proportion of each nationalityin the foreign populations’ (Joel M. Halpern, Population Statistics and Associated Data, Laos Project Papers No.3 [Mimeographed studies, Department of Anthropology, University of California, Los Angeles, 1961], p. 45,quoted, in turn, from Annuaire des Etats Associes, 1953). But, the distinction between ‘ethnic groups’ and‘nationalities’ appeared rather fuzzy and did not seem to follow a coherent pattern (for example, in the 1955 census,the Vietnamese, Chinese and European populations were still included in the ethnic composition of the populationas ‘group’) (Halpern, Population Statistics, p. 19, quoted, in turn, from Annuaire Statistique du Laos, Lao Ministryof Interior).

9 Hirschman, ‘The Meaning and Measurement of Ethnicity in Malaysia’, p. 568.10 When the French annexed the Lao territories east of the Mekong River at the expense of Siam in 1893, they still

held high expectations as to the pro� t they would be able to make from the exploitation of the country’s naturalresources and from the trade between colonial Laos and the other components of French Indochina. However,the country neither met its own administration’s expenses nor bene� ted France. Commercial hopes wereexaggerated and the country’s budget always showed a de� cit.

11 Geoffrey C. Gunn, Rebellion in Laos: Peasant and Politics in a Colonial Backwater (Westview Press, Boulder,San Francisco, Oxford, 1990).

12 Eric Pietrantoni, ‘La population du Laos de 1912 a 1945’, Bulletin de la Societe des Etudes Indochinoises, vol.27, no.1 (1953), pp. 25–38.

13 Pietrantoni, ‘La population du Laos de 1912 a 1945’, p. 28.14 By 1940, a complex system was implied with � ve different categories differentiated along class, land ownership,

professional, and ethnic lines. The last two categories set apart, on the one hand, the ‘Lao, Vietnamese and“evolved” montagnards such as the Hmong, Man, Lu, Yao and Kha Loven who paid 2.5 piasters a year personaltax’; and, on the other hand, ‘those montagnards such as Kha and Phoutheng who paid 1.5 piasters per annumpersonal tax’. See Gunn, Rebellion in Laos, p. 52, quoted, in turn, from the colonial document dated 1940 andentitled, Devoirs en matiere � scale des autorites provinciales francaises et laotiennes et des autorites cantonaleset communales laotiennes (Residence Superieure au Laos, Vientiane, 1940), p. 19.

178 Vatthana PholsenaT

able

1.E

thni

cco

mpo

sitio

nof

the

popu

latio

nof

Lao

s,19

11–5

5

1911

1921

1931

1936

1942

1955

a

(%)

(%)

(%)

(%)

(%)

(%)

Lao

276,

801

4542

9,00

052

485,

000

5056

5,00

056

441,

450

43.6

856,

000–

865,

000

77–7

460

.9(L

ao-T

ai)

Tai

124,

238

1912

2,00

015

113,

000

1210

0,00

010

175,

170

17,3

Kha

195,

996

3222

1,00

027

268,

000

2824

7,00

024

300,

138

29.6

210,

000–

258,

000b

19–2

2M

eo-Y

ao15

,205

225

,000

339

,000

447

,000

549

,240

4.9

52,3

00–5

2,90

04

Vie

tnam

ese

4,10

90.

79,

000

19,0

002

27,0

002.

739

,470

3.9

8,00

0c

Chi

nese

486

3,00

00.

36,

100

0.6

32,3

50c

Eur

opea

n22

6d35

31,

000

0.1

900

0.1

8,00

016

3de8f

Cam

bodi

an1,

270

1,30

02,

000

Indi

anan

d6

500

Paki

stan

i

Tot

al61

8,50

098

.780

7,65

397

964,

000

96.4

988,

000

97.7

1,01

2468

100

1,29

1,95

1–1,

320,

402

100

aE

xclu

des

Phon

gSa

lyan

dS

amN

eua;

pres

ent

calc

ulat

ions

base

don

proj

ectio

nof

avai

labl

ees

timat

es.

bK

haun

dere

stim

ated

due

toth

eex

clus

ion

ofpr

ovin

ces

that

cont

ain

few

Lao

1959

estim

ates

;41,

121

non-

Lao

Asi

ans

wer

ere

gist

ered

inL

aos

in19

59.

cIb

id.

dFr

ench

.eM

ixed

-rac

e.fO

ther

Eur

opea

ns.

Sour

ces:

Lao

Min

istr

yof

the

Inte

rior

,19

55�

gure

s;A

nnua

ire

Stat

isti

que

duL

aos

1913

–22;

Ann

uair

eSt

atis

tiqu

ede

l’In

doch

ine

1930

–31,

1936

–37,

Gou

vern

emen

tG

ener

alde

l’In

doch

ine;

Bul

leti

nde

l’O

f�ce

Col

onia

l19

11,N

o.62

,Fe

vrie

r19

13;

Dir

ectio

nde

sA

ffai

res

econ

omiq

ues,

Serv

ice

dela

Stat

istiq

ueG

ener

al(H

anoi

),19

27;P

lan

dede

velo

ppem

ente

cono

miq

ueet

soci

al,R

oyau

me

duL

aos,

Mar

s19

59(H

alpe

rn,P

opul

atio

nSt

atis

tics

,p.1

9).


Table 2. Ethnic composition of Laos, 1954–55, by percentage of ethnic groups in eachprovincea

Lao & Taib Kha Meo & YaoTotal Total

Province population (%) (%) (%) (%)

Nam Tha 46,809 17,104 36.54 26,798 57.25 2,907 6.21 100Luang Prabang 136,821 66,687 48.74 63,416 46.35 6,718 4.91 100Sayaboury 98,516 86,389 87.69 8,167 8.29 3,960 4.02 100Xieng Khouang 93,609 44,090 47.10 12,178 13.01 37,341 39.89 100Vientiane 186,269 183,978 98.77 317 0.17 1,974 1.06 100Khammouane 108,603 99,785 91.88 8,818 8.12 100Savannakhet 214,974 171,743 79.89 42,231 19.64 99.5Saravane 125,957 65,498 52.0 60,459 48.0 100Champassak 122,078 117,769 96.47 4,309 3.53 100Attopeu 43,315 12,865 29.70 30,450 70.30 100Phong Salyc 50,000Sam Neuac 65,000Total 1,291,951 865,908 258,143 52,900

a Exclusive of Chinese, Vietnamese and European population.b Includes Lu, Tai Dam and other tribal Tai groups; does not include Thai.c Data incomplete since most of the area was under Pathet Lao Control during this period.Sources: Unpublished records of Ministry of Interior of the Government of Laos, Vientiane(Halpern, Population Statistics, p. 18).

The Royal Lao Government Administration

The � rst post-colonial census under the Royal Lao Government appears to have beenconducted in 1955, while the last colonial countrywide census seems to be dated as far backas 1943.15 But the of� cial departure of the French from Indochina in 1954 did not removethe colonial imprint. Their classi� cation inscribed their image of the Lao population on thepost-colonial censuses. Indeed, the latter kept the same pattern, i.e. the naming and thecategories (see Table 1, the year 1955), which meant that the racial lines were insidiouslyperpetuated among the natives. However, the Lao authorities brought about one signi� cantchange that re� ected the imperative of forming a self-conscious Lao national community.They gathered the ‘Lao’ and ‘Tai’ groups together under a single category (see Table 2),resulting in an increase in the � gures in favour of the ethnic Lao, at the expense of the‘Kha’ population. In addition, there were suspicions on the part of some individualsinvolved in the census that the ‘Kha’ � gures were underestimated in the censuses (Tables1 and 2).16 But, on the other hand, there is no concrete evidence that the underestimationwas systematic and politically orientated, and it could simply have been due to the dif� cultyof listing populations living in remote areas.

15 Eric Pietrantoni, ‘La population du Laos de 1912 a 1945’ and ‘La population du Laos en 1943 dans son milieugeographique’, Bulletin de la Societe des Etudes Indochinoises, vol. 32, no.3 (1957), pp. 222–43.

16 Curiously enough, a hand-written note was added below the original table suggesting that the non-Lao groupswere indeed underestimated .


In the 1950s, the use of the seminal terms ‘Lao Lum’ (valley Lao), ‘Lao Theung’ (Laoof the mountain slopes) and ‘Lao Sung’ (Lao of the mountaintops), was initiated under theRoyal Lao Government. It was a stroke of genius. That classi� cation is still being used inpresent-day Laos and is widely applied, even among academic works. And yet from thebeginning, it was clearly a political attempt to emphasise the unity of the country bysuppressing the pejorative nature and the racial connotations attached to the previousnaming system, and by denying the reality of the cultural differences among the peoples.17

The mapping of Lao society was � rst shaped by a racial theory in the form of scienti� cdiscourse engendered and developed by the colonial authorities, with the support of acertain type of ethnographic study and in response to a demand from the administration.The racial discourse was then naturalised through a ‘repertoire of rituals and routines ofrule’.18 In the aftermath of colonial rule, the discourse was perpetuated and reproducedunder a new terminology by the newly independent state. The ‘Lao Lum’, ‘Lao Theung’and ‘Lao Sung’ categories referred exactly to the same major ethnic groups (‘Lao and Tai’,‘Kha’ and ‘Meo-Yao’, respectively) de� ned by the colonial administration. The categorisa-tion was as arbitrary as the previous one, but it endured because it coincided with the ‘truth’that had been produced and already legitimised during the French period.

That classi� cation was furthermore encouraged by the Royal Lao Government becauseit served its project of building Laos as a nation and, as such, was widely used. As a matterof fact, some of the RLG members saw the use of these gross ethnic categories as anindicator of the integration of the non-ethnic Lao peoples into the ethnic Lao culturalmainstream. Katay Don Sasorith thus wrote in 1953:

Indeed, the ethnic issue in Laos won’t be as complex as it is in Siam, because in Laos, the Laoelement clearly and undeniably predominates, in terms of numerical importance as well as byits degree of social and cultural development … . Some educated and developed Boloven orMeo tend to get closer to us, as much in their way of dressing and their lifestyle as in theirpatriotic ideal, to such an extent that they now want to be called under the name ‘LaoTheung’.19

That ‘truth’ was also legitimised by other agents, such as foreign scholars. Indeed, thosewho criticised the Lao Lum, Lao Theung, Lao Sung strati� cation also recognised the logicof it. For example, the American anthropologist, Joel Halpern, wrote in the 1960s:

Leaving the Mekong plain the land abruptly changes to rugged mountains cut by narrowvalleys. The observer looking closely at the settlement pattern below can discern almost atextbook illustration of ethnic strati� cation and economic-geographic adaptation to the landbased on varying degrees of altitude.20

A ‘truth’ had been produced and legitimised, naturalised through a series of discourses,which were operated, integrated and transmitted by diverse and multiple agents, i.e. thecolonial administration, the post-colonial state, scholars and the population—in other words,a political economy.21 The ‘truth’ was sometimes criticised and even denied, but it was thestarting point from which the debates over ethnic classi� cation were departing. It was this

17 Joel M. Halpern, Economy and Society of Laos: A Brief Survey (Yale University Monograph Series No. 5, 1964),p. 5.

18 Cohn and Dirks, ‘Beyond the Fringe’, p. 225.19 Katay Don Sasorith, Le Laos. Son evolution politique. Sa place dans l’Union francaise (Editions Berger-Levrault,

Paris, 1953), p. 21. Certes, le probleme ethnique au Laos ne sera pas aussi complexe qu’au Siam, car au Laos,l’element lao predomine nettement et indiscutablement , tant par son importance numerique que par son degre dedeveloppement social et culturel. […] des Bolovens ou des Meos instruits et evolues tendent a se rapprocher deplus en plus de nous, tant dans leur facon de s’habiller et dans leur maniere de vivre que dans leur ideal patriotique,au point de vouloir se faire appeler maintenant sous le nom de ‘Lao Theung’ …

20 Halpern, Economy and Society of Laos, p. 5.21 Michel Foucault, Micro� sica del Potere, p. 131.


‘regime of truth’ that the Pathet Lao, and later on the socialist regime, attempted to breakand to replace by their own discourse of the nation.

2. Socialist Regime: Breaking with the Past

The ‘Policy of National Equality’ and the Civilising Project

On 2 December 1975, the monarchy was abolished and replaced by the Lao People’sDemocratic Republic (the Lao PDR). The propaganda for socialist construction became anappeal for a united patriotic front. The ethnic minority members were called upon to jointhe struggle for the triumph of socialism against the ‘imperialists and the reactionaries’.Indeed, as Martin Stuart-Fox rightly observed, ‘[n]ational solidarity and defence had toproceed hand in hand. So long as ethnic and social divisions remained, these could beexploited by the “enemies” of the new regime’.22 The new regime explicitly recognised the‘Hill-Tribe Question’. Kaysone Phomvihane, then the leader of the Party, thus declared ina 1982 speech:

No tribe can be regarded as the majority as nearly 70 tribes with different levels of economic,cultural and social development live in Laos. The tribal question is one of the major problemsof the Lao revolution and socialist construction in Laos. Our Party is trying to resolve this issuewhile carrying out its overall cause of socialist transformation. The Party is striving to tightenunity among the nationalities and tribes, taking into consideration the special aspects of eachtribe and our harmonious interests in the struggle for a bright future.23

He called therefore for greater attention to be paid to promoting education among ethnicgroups, improving their living conditions and increasing production in remote minorityareas. Furthermore, he insisted on respect being paid to the ‘psychology, aspirations,customs, beliefs of each ethnic group’.24 On the tenth anniversary of the Lao PDR inDecember 1985, Kaysone announced that minority traditional culture had been preservedthrough schools of dancing, music and handicrafts. Lenin’s apprehension about the risk ofethnic awareness in the Soviet Union led him to initially promote the policy of ‘nationalequality’; so too did the Lao PDR, as had previously the People’s Republic of China (PRC)and the Democratic Republic of Vietnam (DRV).

Brief Account of Chinese and Vietnamese Communists’ Experiences

Although for Lenin nationalism was a secondary problem, it was essential to keep it undercontrol. His strategy for neutralising the national question was guided by his perception ofnationalism as the result of past discrimination and oppression. Consequently, nationalantagonisms and mistrust were to be dissipated by a period of national equality; this policycame to be known as ‘the � ourishing of the nations’. It was predicated upon the belief thatnations would naturally move closer together, a process described in the of� cial Marxistvocabulary as the ‘rapprochement’ or ‘coming together’ of nations.25

22 Martin Stuart-Fox, ‘Socialist Construction and National Security in Laos’, Bulletin of Concerned Asian Scholars,vol. 13, no. 1 (Jan.–Mar. 1981), p. 63.

23 Kaysone Phomvihane, ‘Strategy of Bypasssing to Socialism’, Vientiane Home service, 10–17 March 1982(Summary of World Broadcast (SWB) of BBC, 5 April, Part 3, The Far East, 1982); quoted, in turn, from HansLuther, Socialism in a Subsistence Economy: The Laotian Way (An Analysis of Development Patterns in Laosafter 1975) (Chulalongkorn University Social Research Institute, Bangkok, 1983), p. 44.

24 Kaysone Phomvihane, La Revolution Lao (Edition du Progres, Moscou, 1980), p. 233.25 Walker Connor, The National Question in Marxist–Leninist Theory and Strategy (Princeton University Press,

Princeton, NJ, 1984), p. 202. This policy of ‘national equality’ was therefore not merely a political device topreserve the unity of the Soviet state; it also re� ected the ideology of egalitarianism. It was based on the idea


This vision of the achievement of historical progress became the landmark of anyCommunist project. While the work of Marx and Engels centres on a critique of capitalismand includes analyses of societies characterised by slavery and feudalism (the stagesthought to be the immediate predecessors of capitalism on the evolutionary scale), theydrew heavily on the work of Lewis Henry Morgan when they turned to analyse ‘primitive’societies. Morgan’s theory of social evolution outlined three main stages: savagery,barbarism and civilisation.26 The programme of promoting ‘national equality’, althoughformulated differently, followed the same historicist and universalist vision; it was only aprerequisite for a higher stage in the movement towards assimilation27 that Lenin perceivedas progressive and inevitable. Later on, the Marxist–Leninist regimes’ attitudes in China,Vietnam and Laos vis-a-vis their minority populations were driven by that policy of‘national equality’. Subsequently, they all sent their cadres to the highland areas to list thepopulations and to collect data dealing with the material aspects of their lifestyle in orderto promote the ethnic groups under a so-called supra-national culture. Ethnographic studiesand censuses, indeed, re� ected the belief that cultural recognition would narrow the gapbetween peoples.

When the PRC was proclaimed in 1949, the traditional Han goal of forced assimilationwas rejected. The objective of the Chinese Communist Party (CCP) was to end theinequality between the ethnic groups through a programme of gradual cultural, economicand political equality. Stevan Harrell offers a concise de� nition of the work as ‘creatingautonomous regions, implementing educational and developmental plans, bringing leadersof the peripheral peoples into the ful� lling of the promise that all minzu,28 equal legally andmorally, would march together on the road to historical progress, that is, to socialism’.29

The slogan was then ‘Unity in Diversity’ or ‘Unity and Equality’. All the minoritieswere allowed to keep their cultural distinctiveness. For the Chinese Communists, the twogreat evils that had to be overcome now were the attitude of Han cultural superiority (‘Hanchauvinism’) and the fear the minority groups had of Han domination (‘local nationalism’).In the 1950s, party cadres were sent to the minorities’ areas to collect data on their customsand lifestyle, the objective also being to make the cadres’ minority propaganda and thetraining of a socialist proletariat among the minorities more effective.30

The Vietminh’s policy towards the minorities was very similar to the Chinese model.Article 3 of the amended 1960 Constitution of the Democratic Republic of Vietnamreiterated the policy of autonomous zones31 while asserting that such autonomy was to be

Footnote continued

that human nature would change radically in the homeland where revolution occurred and that a new man, strippedof all the prejudices of the former world, would emerge from the revolutionary process. See Helene Carrered’Encausse, ‘Determinants and Parameters of Soviet Nationality Policy’, in J.R. Azrael (ed.), Soviet NationalityPolicies and Practices (Praeger, New York, 1978), p. 47.26 Lewis H. Morgan, Ancient Society (Holt, New York, 1877).27 But the Soviet authorities were careful to avoid using the term ‘assimilation’, for they argued it conveyed a negative

meaning, as it was connected with capitalist societies and their coercive process of acculturation conducted bythe state’s dominant group towards the minorities. On the contrary, the Marxist–Leninist approach was claimedto be different: the process of merging together was doctrinally based upon absolute national equality and on thebasis of voluntary cooperation.

28 Chinese term, of Japanese origin, that can be translated as ‘nation’, ‘people’, ‘nationality’ or ‘ethnic group’, DruC. Gladney, ‘Clash Civilizations? Muslim and Chinese Identities in the PRC’, in Dru C. Gladney (ed.), MakingMajorities. Constituting the Nation in Japan, Korea, China, Malaysia, Fiji, Turkey, and the United States (StanfordUniversity, Stanford, California, 1998), p. 117.

29 Stevan Harrell, ‘Introduction: Civilizing Projects and the Reaction to Them’, in Stevan Harrell (ed.), CulturalEncounters On China’s Ethnic Frontiers (University of Washington Press, Seattle and London, 1995), p. 24.

30 June T. Dreyer, ‘China’s Quest for a Socialist Solution’, Problems of Communism, vol. 24, no. 5 (1975), p. 52.31 After their victory over the French, the Vietminh rewarded its minority supporters by creating in 1955 and 1956

two Autonomous Regions–respectively, the Tai-Meo zone in the Northwest and the Viet Bac zone in the Northeast.However, the reuni� cation of the country in 1975 heralded a return to the minority policy of the 1946 Constitution,


within ‘the territory of Vietnam [which] is a single, indivisible whole from the North to theSouth’ (Article 1).32 In reality, the right of self-determination was subordinated to socialistideals. As within China’s autonomous zones, the minorities were expected to follow thepath to ‘progress’ by going through all the evolutionist stages—from primitive communismto feudalism, then to capitalism and � nally to socialism. As Jean Michaud observes:

In a new country where the collective project has to be popular, national and scienti� c, therewas little room left for the ways of the past. Following this frame of mind, and despite anopenly egalitarian state rhetoric, montagnards were considered culturally and economicallybackward unless they accepted the cultural supremacy of the lowland majority.33

As regards the Lao PDR, the ethnographic research methodology employed is probablybest exposed in a working manual, entitled ‘References and criteria for conducting researchon ethnic groups across the country for statistical purposes’, and written by SisoukChonmaly,34 a high-ranking of� cial of the Lao Front for National Construction’s (LFNC)Research Department on Ethnic Groups.35 The document is also worth mentioning as it wasused as the of� cial guide during the data-collection campaign for the 2000 census.

The applied methodology is directly based on Kaysone’s de� nition of the nation that hegave during a ‘Conference on Ethnicity’ in 1981, in which he declared: ‘The nationalquestion has four criteria or characteristics, which are: common language, commonterritory, common socio-economic organisation, and common psychology’.36 This de� nitionis itself clearly inspired by Stalin, who de� ned a ‘nationality’ by � ve similar criteria: astable community of people, a language, a territory, an economic life and a psychologicalmake-up or ‘national character’.37 But only two of the four criteria are to be applied forethnographic research in the Lao PDR, i.e. language and ‘material and spiritual ways oflife’; a third criterion is also added, namely, the origins and migrations of the group. Laosbeing a predominantly rural country, Sisouk argues that there is not enough variation insocio-economic organisation for this to be a factor of differentiation. As for the territory,there is simply no autonomy granted to ethnic groups in Laos.

Upon closer scrutiny of the document, however, it rapidly becomes clear that theethnographic element is accompanied, if not dominated, by the objective of controllingethnicity and producing � xed and ‘correct’ identities. Sisouk even admits that ‘spiritual’ aswell as ‘material ways of life’ are not such reliable criteria, especially the former because‘some ethnic groups mingle with others’ culture. Sometimes, some ethnic groups adoptothers’ culture as theirs. As a consequence, we must be careful to collect an adequateamount of clear data for comparison’.38 The data-collection seems in effect to be guided bytwo political aims: (1) to contribute to the national culture and (2) to censor ‘bad’ whilepromoting ‘good’ culture. Indeed, as Sisouk bluntly explains:

in general, all the ethnic groups’ psychological and cultural features mentioned above do notcon� ict with the overall psychology, with the national community’s culture. On the contrary,

Footnote continued

which had made no mention of self-determination. Thus, ignoring the provisions of the 1960 Constitution, thegovernment announced its decision to dissolve the Autonomous Regions on 29 December 1975.32 The Constitution of Vietnam: 1946–1959–1980–1992 (The Gioi Publishers, Hanoi, 1995), pp. 42–3.33 Jean Michaud, ‘The Montagnards and the State in Northern Vietnam from 1802 to 1975: A Historical Overview’,

Ethnohistory, vol. 47, no. 2 (2000), p. 357.34 I use a pseudonym here.35 I have the 1999 updated version, but apparently Sisouk has been using these guidelines since the late 1970s when

he started his ethnographic research in the Lao PDR–personal communication .36 Sisouk Chonmaly, ‘Criteria for Conducting Research on Ethnic Groups for Statistical Collection throughout the

Country. Comprehensive Scienti� c Research Methodology’ (Xerox copy, 1999), p. 1.37 Joseph Stalin, Marxism and the National Question. A Collection of Articles and Speeches (Lawrence & Wishart,

London, 1936).38 Sisouk Chonmaly, ‘Criteria for Conducting Research on Ethnic Groups’, p. 8.


they enhance the psychology and the culture of our Lao nation. They also provide us withinformation to conduct research for clari� cation as well as for supporting the Party-Governmentin its policies and socio-economic development plans that lead all the ethnic groups along thepath of development and prosperity. [A few pages later, he speci� es the aim:] Once the datais collected, we will write an assessment report and submit it to the appropriate organisationin order to set up short and long-term plans to view the backward practices and to promote thecorrect ones.39

Ethnographic works produced by Marxist–Leninist regimes have always been stronglyidenti� ed with a civilising project vis-a-vis ethnic minorities.40 So-called scienti� c studythrives on an ideology strongly in� uenced by evolutionist theories, the objective of whichis to ‘classify the ethnic groups according to their degree of cultural development’. In effect,criteria for distinction or grouping are thought of as ‘criteria of backwardness’.41 An ‘ethnicgroup’ in this sense is probably better de� ned as a ‘tribe’, i.e. a group of individuals, seenas being clear-cut and isolated, around which are traced arti� cial boundaries for administrat-ive and political reasons. The concept of ‘tribe’ originates in an illusion: at the ideologicallevel, ‘tribes’ are a colonial concept that re� ected the conception held by Western cultureof the rest of the world at that time, key themes of which were imperialist expansionismand the dichotomisation of humankind into the ‘civilised’ and the ‘uncivilised’.42 There isclearly a double agenda attached to ethnographic studies in the Lao PDR (on the one hand,promotion of cultural diversity and on the other, political control of ethnicity) and theemphasis is most likely to be on the political aspect. Consequently, an ‘ethnic group’appears as a � xed entity, upon which is imposed a set of characteristics that have beenaccepted as ‘correct’ and suf� ciently distinctive. That seems to be in effect the basis for an‘ethnic group’ to appear on the census list.

3. Case Study of the 2000 Population Census

Following the Chinese and Vietnamese Communists, the Lao authorities have launched datacampaigns in areas inhabited by minority populations, � rst in the ‘Liberated Zones’ duringthe war, and then, after their victory, throughout the country from the late 1970s. Asmentioned above, the ethnographic works were guided by similar Marxist–Leninist theories,i.e. ‘policy of national equality’ and ‘evolutionist theory’. In 1981, Kaysone was clear inthat the terminology founded upon the three large ‘national categories’, ‘Lao Lum’, ‘LaoTheung’ and ‘Lao Sung’, was to be replaced by a new ethnic classi� cation. He wrote thus:

[Each ethnic group] has … its own characteristics. As the revolution developed, the variousethnic groups became integral parts of the nation of their own free will, under the then politicaldenominations: Lao Lum (of the plains), Lao Theung (of the slopes) and Lao Sung (of highaltitudes). The Central Committee of Ethnic Groups must co-operate with the various branchesand with our brothers, the workers from the various ethnic groups, to conduct together researchand discussions with regard to the names and the lists of ethnic groups, in order to establishof� cial rules.43

39 Sisouk Chonmaly, ‘Criteria for Conducting Research on Ethnic Groups for Statistical Collection throughout theCountry’, pp. 3 and 18.

40 I use here the term ‘minorities’ instead of ‘groups’ for it is evident that the civilising project was not applied tothe political majority that is also an ‘ethnic group’ according to the Stalinist de� nition.

41 Yves Goudineau, ‘Ethnicite et deterritorialisation dans la peninsule indochinoise : considerations a partir du Laos’,in M-J. Jolivet (ed.), Logiques identitaires, logiques territoriales (Editions de l’Aube, IRD [Institut de recherchepour le developpement , ex-Orstom], Autrepart, Cahier des Sciences Humaines, 14, 2000), p. 23.

42 Cohen Ronald, ‘Ethnicity: Problem and Focus in Anthropology ’, Annual Review of Anthropology, no. 7 (1978),p. 384.

43 Kaysone Phomvihane, Enhancing Solidarity’s Roots Between the Diverse Ethnic Groups within the Lao NationalCommunity, United and Determined to Durably Preserve the Nation and to Successfully Build up Socialism(Vientiane, 15 June 1981), p. 47, my stress.


As a consequence, the new regime started using a new classi� cation for the populationcensus. Although the Pathet Lao consciously used the classi� cation to recruit non-ethnicLao for the revolutionary cause—the promise was that they would thereby be grantedrecognition on a par with the ethnic Lao44—the three-fold categorisation with the ‘Lao’pre� x was deemed to be anti-revolutionary after the Communists took power and its usewas abandoned in political documents. Nevertheless, this terminology is still widely usedin present-day Laos. However, I suggest that the pattern of the Lao PDR population censusnot only re� ects Marxist–Leninist ideology, but also parallels the shaping of the nation. Assuch, an analysis of the construction process and the structure of the latest populationcensus is most helpful.

Analysis of the Construction of the 2000 Census

Data collection for the 2000 census was once more driven by the constant imperative to listand to identify the exact number of ethnic groups, an obsession that Lao of� cials commonlyshare with their Chinese and Vietnamese counterparts. But in Laos the issue is particularlyacute, as in contrast to China and Vietnam the � gure has been subjected to severalrevisions. An of� cial document mentions the successive � gures of 200, 177, 150, 131, andeven up to 820 and 850 ethnic groups!45 Unfortunately, it does not specify any dates, butaccording to Grant Evans, the number of 820 self-named ethnic groups was the result ofthe 1983–85 census.46 It was not until 1985 that the Party approved an of� cial estimate of47 ethnic groups; and yet, even after the publication of the 1985 population census, thedeliberations continued.47 The latest of� cial population census published in the Lao Census1995 retained the � gure of 47 ethnic groups.

The 1999 data-collection campaign once again aimed to clarify this forest of anarchy,uncertainties and confusion. And, after � ve years of apparent stability, new � guresemerged: 49, instead of 47, ethnic groups, which were distributed between four, instead ofthe previous six, ethno-linguistic categories (see Appendix 1). The issue is not, of course,to debate the � gures’ accuracy, since the census is anyhow arbitrarily constructed. Rather,I attempt here to analyse how and why the changes occurred or did not occur. Byaddressing these questions, the answers may, in turn, help to interpret the broader picture,i.e. the nationalist discourse.

Teams, composed of of� cials from the LFNC Research Department on EthnicGroups, were sent out across the country in spring 1999 to collect data on the ethnic groups.The campaign lasted nearly four months until September of that year. Each team stayedfor about one month in the provinces, which had been divided into � ve majorgeographical areas.48 Except for two members, the teams had little, if any, knowledge oracademic background relating to ethnographic methodology. Before they left for the� eld, they received only a few days’ training at a seminar taught by Sisouk, whosupervised the census. To be fair, their task was not to conduct an exhaustive investigationas had been done in the past; in fact, it merely consisted of collecting the population

44 Martin Stuart-Fox, ‘On the Writing of Lao History: Continuities and Discontinuities’, Journal of Southeast AsianStudies, vol.24, no. 1 (March 1993), p. 115.

45 Sisouk Chonmaly, Meeting’s Report on the Research and Study of the Ethnic Groups’ Names in the Lao PDR(Vientiane, 16 November 2000), p. 1.

46 Grant Evans, ‘Apprentice Ethnographers : Vietnam and the Study of Lao Minorities’, in Grant Evans (ed.), Laos:Culture and Society (Silkworm, Chiang Mai, 1999), p. 178.

47 Yves Goudineau, ‘Ethnicite et deterritorialisation dans la peninsule indochinoise’, p. 22.48 The � ves areas were the South (Champassak, Saravane, Sekong and Attapeu); the Centre (Savannakhet ,

Khammouane and Bolikhamxay); the Northwest (Oudomxay, Phongsaly, Luang Namtha and Bokeo); theNortheast (Luang Prabang, Houaphan, Xieng Khouang and Saysomboun) ; and an area including Vientiane, itsprefecture and the adjacent province of Saygnabouly.


censuses from the provincial LFNC organisations. Some ethnographic studies were never-theless carried out to ful� l the imperative of clari� cation, but these works were performedby an exclusive few, including Sisouk himself.

In reality, the changes in the census went far beyond the addition of two new ethnicgroups. In fact, the � rst assessment of the provincial censuses gave a list of 55 ethnic groups(see the ‘55 list’ in Appendix 2), which was whittled down to 49. Ironically, the mainobjective of the census coordinators was not to check information in the � eld but insteadthe data provided by the provincial LFNC organisations. Indeed, the authorities, alarmed bythe confusion that was reigning throughout the country and among the of� cials themselves,placed a high priority on keeping the proliferation under control. Sisouk, thus, recounted inhis report:

According to the data of the Committee on Population Census, the population amounts to4,574,848 inhabitants, among whom 24,084 have not speci� ed their ethnic af� liation and10,201 do not appear on the list of 47 ethnic groups. Many problems, however, occurred afterthe population census’ [of 1995] documents were sent to the Party-State’s provincial organisa-tions. The Statistical Department has received letters and phone calls from the CentralCommittee’s of� ces and ethnic groups’ representatives in the provinces, pointing out theabsence of ethnic groups’ names as well as names that did not satisfy ethnic groups.49

Finally, 13 new ethnic groups, which were not listed on the national census, appearedin the provincial data.50 In total, � ve new ethnic groups were of� cially recognised forinclusion in the national census at the LFNC meeting held on 13–14 August 2000, thoughonly two of these had been listed in the provincial data (these were the ‘Then’ (out of the‘Khmmu’ group) and the ‘Thai Neua’ (out of the ‘Phuthai’ group), the latter being listedin no less than three provinces51). Two of the other groups, the ‘Tai’ and the ‘Idou’, seemto have been proposed at the central level, as neither of them appeared in the provincial dataat all. The � fth ethnic group, the ‘Lahu’, was a new name in the of� cial list.52 Itencompassed the ‘Kui’ and ‘Mousseu’ peoples, which disappeared from the census bybecoming subcategories.53 The construction of the 2000 census was also probably inspiredin some ways by the Vietnamese census’ names. Indeed, in total 11 names (some of themattributed (supposedly) to the French colonisers and/or being perceived as derogatory in theauthorities’ eyes) were replaced by ‘correct’ ones, among which four were already listed inthe Vietnamese census.54 That similarity could also possibly explain the apparition of a‘Tai’ group (extracted from the ‘Phutai’), which is found among the 54 Vietnamese of� cialethnic groups.

This cultural objecti� cation is most unlikely, however, to have any short-term impact onthe population, all the more so as the names still remain subject to possible revisions,depending on whether they will � t the evolving standard of ‘correctness’.55 During a tripto a village in 1999 in the province of Sekong, I was intrigued when an of� cial kept

49 Sisouk Chonmaly, Meeting’s Report on the Research and Study of the Ethnic Groups’ Names in the Lao PDR,pp. 4 and 5.

50 Kado, Kanay, Tong, Ine, Yang, Meuang, Kayong, Thai Rat, Khoumma, Bri/Labri, Kinh, Thai Neua, Then.51 Phongsaly, Luang Namtha and Bokeo. As for the Then (separated from the Khmmu), I can only make suppositions.

As Sisouk disagrees with the idea of granting them the status of ‘ethnic group’, the other reason could be thatthey are in� uential enough to have gained themselves their visibility.

52 The Lahu ethnic group is also listed in the Thai, Vietnamese and Chinese population censuses.53 A third group, the Khir, was simply suppressed for an unknown reason.54 The ‘correct’ ethnic groups are as follows, with their former names in parentheses: Khmmu (Khammu), Y’ru

(Laven), Triang (Talieng), Brao (Lave), Kriang (Nge), Arak (Alak), Iumien (Yao), Akha (Kor), Singsili (Phounoy),Lahu (Kui and Mousseu), Sila (Sida). The four Vietnamese ethnic groups are: Kho-mu (Khmmu), Brau (Brao),La Hu (Lahu), Si La (Sila).

55 Sisouk Chonmaly, Meeting’s Report on the Research and Study of the Ethnic Groups’ Names in the Lao PDR,p. 8.


repeating to the villagers that they should no longer call themselves ‘Lao Theung’ or ‘Nge’,but ‘Kriang’. However, a month later, when I came back to the village, the answers werestill ‘politically incorrect’. I then asked the village’s Party Representative the ways in whichhe usually described himself:

Question: What is your national group (sonsat)?Answer: ‘Lao Theung!’ the man replied at once. He then started enumerating the different

national groups: ‘There are the Lao Theung, the Lao Lum, the Lao Khong, …’ Hestopped, looking hesitant, and then mumbled a few more words I was unable tounderstand.

Question: What is your nationality (sansat).Answer: He replied without hesitation: ‘Lao’.Question: What is your ethnic group (sonphaw)?Answer: Nge. He then speci� ed: ‘We belong to the 68 ethnic groups like the Lao Sung, the

Meo, …’ He stopped and mumbled inaudibly again.

During our whole stay in the village, I never heard anybody spontaneously introducinghimself or herself as ‘Kriang’.

Accompanying Sisouk Chonmaly on another short trip to a Nge village in SekongProvince the same year, I witnessed an intriguing scene involving the LFNC of� cial and thevillagers. Sisouk declared that their ethnic group actually belonged to a larger category, the‘Bru’. He argued that both ethnic groups indeed shared some traditions, such as offering tospecial guests the boiled feet, head and giblets of a chicken. The villagers looked puzzled,but silently submitted before the central authority.

Sisouk is often depicted as a pro-Makong (the of� cial ethnic group to which he belongs,and which is referred to as the ‘Bru Makong’ in the census’ sub-categories) chauvinist. Iwould suggest, rather, that this high-ranking of� cial is lobbying for the cause of the ‘Bru’as an of� cial category. In effect, as Cohn similarly observed in India:

The implied argument is that the census was one of the situations in which Indians wereconfronted with the questions of who they were and what their social and cultural systemswere. I don’t think that the act of a census enumerator asking a question of a peasantcontributed too much to the process … If there was a direct effect of the census [on the massof the Indian population], it was on the enumerators.56

In addition to the list of 55 ethnic groups resulting from the Lao Front provincialcensuses, Sisouk proposed a much-condensed list of only 34 groups (see Appendix 3).Among the causes of that dramatic reduction was the grouping of eight ethnic groups listedseparately in the 1995 census within a single one, the ‘Bru’.57 During our trip through theSouthern provinces in August–September 1999, I observed Sisouk’s remarkable lobbyingwork for the cause of the ‘Bru’, an ethnic group of� cially classi� ed under the Austroasiaticcategory in the Vietnamese census, but invisible in the Lao version. The desirability ofbeing recorded as ‘Bru’ might come from the material bene� ts such a denomination wouldbring. As he cunningly explained to some villagers in Sekong during the trip: ‘We must nolonger name the Makong, Pacoh, Tri, Katang, Nge, Katu, Ta-Oy, Souay, one by one. Eachof them, singled out, makes too low a number, whereas the “Bru” would amount to morethan 300,000 persons. That makes a bigger population, more in� uential!’

His cause was echoed recently in a project � nanced by the International LabourOrganisation (ILO). The latter became involved in Laos in 1996 through its Project for thePromotion of ILO Policy on Indigenous and Tribal Peoples. In 1999, it started a study onthe ‘Government’s policies regarding the ethnic groups in rural development’. In the

56 Bernard S. Cohn, ‘The Census, Social Structure and Objecti� cation in South Asia’, p. 248.57 These eight ethnic groups are: Katang, Makong, T’ri, Taoy, Pako, Souay, Katu, Triang. Sisouk also includes the

sub-categories , Kanay, In, Tong, Kadu.


following year, upon completion of that study, a two-year pilot project (with a budgetestimated at $140,000) was initiated in the province of Khammouane. According to theProject Document, the � ve target villages were mainly populated by the ‘Bru’, with thePhutai occupying second position (these two ethnic groups accounted for 78 per cent and22 per cent of the population, respectively). This ethnic distribution apparently justi� ed theinclusion of ‘Bru’ and Phutai as the project’s languages, along with English and ‘Laotian’.Similarly, the candidate for the post of Project Manager was required to possess knowledgeof the ‘Bru’ language.58

I have no evidence that conclusively proves Sisouk’s direct in� uence on the choice ofarea or on the outline of the project; however, there are some disconcerting elements thatmake coincidence appear to be rather a weak alternative explanation. First, the Lao Frontfor National Construction, along with the Central Leading Committee for Rural Develop-ment, is the Government’s agency for the project. Consequently, Sisouk is directlyinvolved, since he is responsible for all the foreign projects asking for the LFNC’scooperation. Although he has limited executive power, no project can be pursued withouthim as supervisor or coordinator. In addition, the project’s target area is his very own nativedistrict, with which he still keeps in regular contact, notably via relatives and businesspartners. Yet, Sisouk is not the only scholar to argue for the existence of a ‘Bru’ group inLaos. James Chamberlain, a linguist, also lists in his census of the ethnic groups of the LaoPDR the ‘Bru’ as an Austroasiatic group, which corresponds, according to his classi� cation,to the Makong and So peoples within the Lao nomenclature.59 It should also be noted thatChamberlain was the main author of the 1999 ILO consultants’ report. One may concludethat Sisouk and Chamberlain’s concordant positions, in addition to Sisouk’s unique positionwithin the project, have played a signi� cant, if not decisive, role in the selection of the‘Bru’ as the main target ethnic group. The ‘Bru’ became a living, distinct entity as far asthe ILO project was concerned.

Could Sisouk’s ethnicity be de� ned as strategic, then? His involvement in the ILOproject seems to support that analysis. Ethnic solidarity is reinforced as the reaction of aculturally distinct periphery against the centre. Under these circumstances, ethnic differ-ences do not disappear and indeed may form the basis for collective action by members ofthe peripheral communities against the central community, because ethnic identity cannotbe detached from one’s economic and political interests within the system. Sisouk, throughhis project of creating a ‘Bru’ ethnic identity, is his own cultural agent. As a Laoanthropologist, he has been, and still is being, strongly in� uenced by the Soviet andVietnamese ethnographic traditions. His study of minorities, like his conception of the ‘Bru’cultural identity, is therefore very much guided by taxonomic principles, whereby culturesare conceived of as bound, static and objective. For instance, he explained to me in a vaguefashion that the eight ethnic groups mentioned above60 shared the same language—or, atleast, they could understand each other—as well as a few traditions, though he did notspecify which ones.

However, during a discussion with him on criteria for ethnic classi� cation, he himselfadmitted that even the villagers from his native area were not self-conscious of their ‘Bru’identity. He also acknowledged that language could be a tricky criterion for drawing

58 Sustainable Development of Ethnic Minorities in Lao PDR under the INDISCO Approach, Project Document, ILO(2000).

59 James C. Chamberlain, Charles Alton and Arthur G. Cris� eld, Indigenous Peoples Pro� le. Lao People’sDemocratic Republic (CARE International, Vientiane, 15 December 1995).

60 These eight ethnic groups are: Katang, Makong, T’ri, Taoy, Pako, Souay, Katu, Triang. Sisouk also includes thesub-categories , Kanay, In, Tong, Kadu.


distinctions, as two ethnic groups could share the same language. He gave the example ofhis own father’s village, Namtok, in Savannakhet, where nobody spoke ‘Bru’ any longer.In addition, they were all Buddhist. I then asked him how they introduced themselves. ‘Lao’he replied, a hint of disappointment in his voice. ‘But it doesn’t matter!’ he added. ‘Theyknow in their heart that they’re ethnic people (pen khon sonphaw)!’ Yet, there was noevidence of a higher degree of ethnic incorporation, except through their self-proclaimedadvocate, Sisouk. Thus, the anthropologist Stephen Sparkes, who carried out a study for theNam Theun 2 dam project in Khammouane Province in 1997, reported that the peopleliving in the area were embarrassed about his (Sparkes’) use of the term ‘Bru’, as it hadderogatory connotations. Worse, another anthropologist who visited a few ‘Makong’villages in a district in Savannakhet in March 2001 revealed that the people dismissed thename ‘Bru’ as a colonial term!61

In essence, the process of identi� cation seemed to remain limited to a very restrictednumber of persons, namely, Sisouk and his followers. Some of them were themselvesLFNC of� cials in Khammouane and Savannakhet Provinces, Sisouk’s territorial strongholds(his mother and father’s native provinces, respectively). These two provinces, withBolikhamxay in central Laos, were the only ones to list the ‘Bru’ as an ethnic group inSisouk’s version of the 2000 census, the most extreme case being Savannakhet where therewere only three groups listed—Lao, Phutai and ‘Bru’—despite this being the most populousprovince of the country.

The reason for his ethnic chauvinism cannot be reduced to a mere competition forresources. Sisouk cleverly understands the politics of culture and the desire of theinternational community to protect so-called ‘indigenous people’. However, at the presenttime the bene� ts from such a strategy of ethnic incorporation remain largely uncertain.The newly released 2001 Lao census still does not recognise the ‘Bru’ as an of� cialethnic group. Moreover, Sisouk lacks the support of those he seeks to categorise as ‘Bru’.To sum up, his ethnicity is socially irrelevant outside his group of followers. In suchuncertain conditions, Sisouk’s campaign looks like a lost battle. He may count on thelong-term effect of the labelling process, through which the name becomes the identity. But,as long as these people disregard the name ‘Bru’ as a colonial legacy, it is dif� cult to seea process of self-identi� cation occurring again as a result of the input of an externalcategorisation.

I would therefore predict that statistical stability will remain the authorities’ imperative.In Richard Handler’s acute observation, nationalist discourses are ‘attempts to constructbounded cultural objects’.62 Consequently, too dramatic a change (in one way or another)of the census would indubitably disrupt the whole picture of the nation. As Eriksen notes:‘[N]ationalism rei� es culture in the sense that it enables people to talk about their cultureas though it were constant’.63 The Lao PDR’s authorities are still in search of the symbolof national identity in the form of an almost sacred � gure, i.e. that of the total of ethnicgroups. In order to keep the number of 47 ethnic groups they could have turned their backon the claims discussed above. I suspect, however, that pressures for clarity and order weretoo strong to be ignored. In the � nal section below, I furthermore argue that the census’pattern has been in� uenced by Kaysone Phomvihane’s 1981 guidelines on ethnicity andnationalism in Laos.

61 I thank an anonymous referee for these telling accounts.62 Richard Handler, Nationalism and the Politics of Culture in Quebec (The University of Wisconsin Press, Madison,

1988), p. 27.63 Thomas H. Eriksen, Ethnicity and Nationalism: Anthropologica l Perspectives (Pluto Press, London, 1993), p. 103.


4. Kaysone’s Theory of Nationhood

Kaysone’s guidelines still strongly in� uence the actual nationalist discourse. In particular,his booklet entitled, Expanding Roots of Solidarity between Various Ethnic Groups withinthe Lao National Community, written in 1981, is still referred to as the ideological bedrockfor works on ethnic groups, including the censuses and the ethnographic studies. In onesection, Kaysone explained the different processes that led to the emergence of nations inthe world. He developed his views in a lengthy argument, of which large extracts are wellworth citing in order to understand the idea of nationhood that still permeates the nationalistideology in the Lao PDR:

Clans (kok)] and tribes (law) … are communities of individuals living only in a primitivesubsistence society that lacks stability and relational consistency. In the present day, in someunderdeveloped countries, there are still remains of this type of society of clans and tribes:based on kinship, ancient strati� cation, psychology, lineage, animist practices and weddingrituals, etc.

Then, along with the development of production, the society divides into classes (sonsan). Theclans and tribes, in general, disintegrate due to the broadening of relationships between peoplesfrom different clans and tribes, and to the apparition of new economic bases. A new humancommunity has been created. That is the ethnic group (phaw) (peuplade).64

Ethnic group: Each ethnic group has its own language. But there is as yet no language unityin each area. There are even real differences, due to the fact that individuals from the sameethnic group went to seek refuge and mixed with those from another ethnic group. The ethnicgroups’ economy is based on self-suf� ciency, which does not promote the development ofeconomic and cultural exchanges, and which also explains the non-homogenous character ofethnic groups’ languages.

The ethnic group formed with the slavery system and subsisted until the feudal period in thehistory of human society. It can be the basis for the formation of a nation, as in the case ofAustria, Hungary, Russia, Georgia, etc.

An ethnic group can be the constituent basis for a certain number of nations, as is the case ofthe Russian ethnic group, which is the common origin for Russian, Ukrainian, Byelorussianethnic groups, which later on became Russia, Ukraine and Byelorussia.

A certain number of ethnic groups in a country can blend into a nation, as is the case withFrance, Germany, Italy, England, etc.

Some ethnic groups in certain countries can progressively get along with one another in the� elds of language, psychology, as well in economic and territorial areas so as to becomenations. This is the case with ethnic groups that united throughout historical periods to struggleagainst external aggression and to protect the nation subjected to the feudal system, such asVietnam, for example.

As a matter of fact, this is the case with the ethnic groups of all colonies, which unite to resistimperialism, to achieve independence, freedom and to become a nation … as well as an unitedcountry … , such as some Asian and African countries in the history of the contemporaryworld.

Thus, the word ‘nationality’ (sat) (nation)65 … is not the outcome of man’s will, neither of theadministrative power’s imaginative efforts. It is the consequence of the impact of economic andsocial laws. Consequently, a nationality is a community of individuals that normally emergesin history, on the basis of common language, territory, economic lifestyle and psychology thatre� ects a cultural community. This is what appears in Western European countries, such asEngland, France, Italy, while feudalism collapses and capitalism develops. As far as eastern

64 Ethnic group in French.65 Kaysone translated the word saat as ‘nation’. However, the term ‘nationality’, in its cultural sense, seems to be

a more appropriate translation in the light of his de� nition that follows; all the more so as, in the next paragraph,he refers again to the concept of nation, but this time using the term pathet saat, pathet meaning ‘country’.


countries are concerned, such as Laos, Vietnam and Cambodia, etc., they emerged duringdifferent historical periods. Despite their different ways of forming, the national communities,once created, are characterised internally by four characteristics, i.e. language, territory,economy and culture. An ethnic group also possesses these four features, but they are notconsistent. Dealing with the improvement of national unity simply means dealing with theimprovement of these four features … .

A nation (pathet sat) … : is clearly distinguished from ‘race’ (seuasat) [.. . ] and ‘nationality’(sat) [.. . ]. A nation is the whole community or a group of several communities, each of themhaving different features, but united within the same historical destiny, with the will of livingon the same territory, under the same administration, constitution and laws. A nation canconsist of only one nationality (sat) … (for example, Korea, Eastern Germany, Japan, etc.), orof several nationalities (sat) or several ethnic groups (paw) (for example, several Asian andAfrican countries).66

In their book on Asian Forms of the Nation, Tønnesson and Antlov classi� ed thenationalist ‘route’ that the Lao PDR, like China, Vietnam, and to a lesser extent, Cambodia,followed, as class struggle or social revolution, during which the population seized powerfrom the colonial state to form a ‘state based on an ideology of class struggle’.67 During the� rst years of Communist rule, the Socialist Revolution indeed planned to create a loyaltyto the new state greater than the loyalties to particular ethnic identities. In other words, thepriority was to overwhelm the ethnic identities with ‘the principle of the primacy ofpolitics’,68 which claimed to be able to outline an ideal social and political order on thebasis of universal ideas and then to act politically to realise it. The regime’s real objectivewas not to build a society based on national consciousness; rather, the concept of class wasthought to be the new society’s main axis of identi� cation. The ultimate goal for the LaoCommunists, as it had been for their Soviet, Chinese and Vietnamese counterparts, was toeradicate the ‘old’ identities and replace them by a ‘socialist’ one. Kaysone asserted in1976:

The building of socialism does not only consist of creating new relations of production and newproductive forces, but also in contributing a new superstructure. If one wants to create newrelations of production and new productive forces, there must be new, socialist men.69

But, by the time Kaysone wrote his booklet, the second phase of the regime had beenunder way for two years. The Seventh Resolution of the Supreme People’s Assembly wasendorsed in December 1979. This document proposed a number of important changes ineconomic policy in order to improve the disastrous economic performance that character-ised the � rst four years of the new administration. It admitted that, while Laos might be inthe process of ‘by-passing capitalism’, it was going to take time to construct a socialisteconomy. Restrictions on private production and internal trade were liberalised; pricecontrols were abolished for goods sold on the free market. Economic management of stateenterprises was reformed by adopting pro� t as the criterion of ef� ciency.70

In 1981, Kaysone certainly believed in the construction of a socialist society, but

66 Kaysone Phomvihane, Expanding Roots of Solidarity Between Various Ethnic Groups Within the Lao NationalCommunity, pp. 19–25.

67 Stein Tønnesson and Hans Antlov, Asian Forms of the Nation (Curzon Press, London, 1996), p. 22, emphasisin original.

68 Geoffrey C. Gunn, ‘People’s War in Laos: a New Guerrilla Model ?’, Journal of Contemporary Asia, vol. 21,no. 4 (1991), p. 530.

69 Kaysone Phomvihane, Rapport sur la situation de l’an dernier, les orientations et les taches revolutionaires dansla nouvelle etape et les orientations pour 1977, p. 106; quoted, in turn, from Amphay Dore, ‘The Three Revolutionsin Laos’, in Martin Stuart-Fox (ed.), Contemporary Laos. Studies in the Politics and Society of the Lao People’sDemocratic Republic (University of Queensland Press, St Lucia, London, 1982), pp. 101–15.

70 Martin Stuart-Fox, ‘The First Ten Years of Communist Rule in Laos’, Asia Paci� c Community, vol. 31, no. 1(1986), p. 61.


ethnicity persisted as a vertical phenomenon that cut across class and socio-economic strata,despite the land reform and collectivisation campaigns. Ethnicity did not fade away asexpected. Kaysone’s classi� cation of social entities suggests an ethnic character to histheory of nationhood: ‘one ethnic group can form the basis of a nation’. However, he clearlyrejected that ‘route’ in explaining the creation of Laos as a nation. Meanwhile, Kaysonecleverly proposed the idea of nationhood based on sentiments. Ethnic groups, thus, mayshare nothing but the will to pursue a life in common and thereby form a nation. They arejoined together, in Kaysone’s terms, ‘within the same historical destiny, with the will ofliving on the same territory, under the same administration, constitution and laws’. Theyunited to � ght and defeat the ‘imperialists’ together. The sense of solidarity that allowedthem to gain victory has endured since then. The emotional feelings that emerged underhardship have become the cement of the new nation. Kaysone’s theory of the creation ofthe Lao nation is arguably subjectivist, civic-orientated and modernist. ParaphrasingGellner’s famous words,71 the drive for ‘independence and freedom’ led to the emergenceof the Lao nation where it did not exist before. It also transcends linguistic and culturaldifferences, as the constitution of a nation is de� ned by the concomitance of territory,national community and the Modern State. In other words, in Kaysone’s theory, the peopleswho live within the national boundaries belong to the Lao nation-state. But this populationneeds controlling and subordinating under the State’s project of constructing a nation.

In his glossary of human organisations (1999), directly inspired by Kaysone’s 1981writing, Noychansamon Denchaleunsouk (pen name of an LFNC member) highlights thedistinction between ‘ethnic group’, ‘nationality’ and ‘nation’. He points out the fundamentaldistinction between the juridical/political and ethno-cultural concept of nationality. Like-wise, he re-applies Kaysone’s subjectivist and civic de� nition of ‘nation’. His de� nitionsare as follows:

Ethnic group [son phaw]: … community of individuals forged by history, on the basis ofcommon language, roots and naming, sharing thoughts and psyche that re� ect a culturalcommunity … .

Nationality [sansat]: community of individuals that belong to the same country, on a legalbasis, no matter what their physical and cultural differences;

Nation [pathet sat]: all the communities having different characteristics which join in the samehistorical destiny, and which are willing to live on the same territory, under the sameadministration, constitution and laws.

Nationality [sat] (National category or large ethnic group) [sonsat or son phaw nyai]: it is notformed by human aspirations, or by the will of the executive authority. It is the result of theimpacts of the whole socio-economic system. Therefore, a nationality truly means all thehuman communities that emerged in history, on the basis of common language, territory,livelihood, thoughts and diverse cultural features.72

According to the above de� nition, a ‘nationality’ (in the ethno-cultural sense)—or‘national category’ or ‘large ethnic group’—has similar characteristics to those of an ‘ethnicgroup’, but the difference is that a ‘nationality’ would arise from socio-economic inequal-ities: some ‘ethnic groups’ would dominate, and eventually absorb the others. On thecontrary, in Laos, the ‘ethnic groups’ have not reached the stage of ‘national category’ or‘large ethnic group’, and probably never will under the present ideology. The Lao State,applying the Stalinist-cultural de� nition of ethnicity, de� nes an ethnic group as a culturaltotality. An ethnic group is identi� ed by a set of taxonomic features, and according to the

71 Ernest Gellner, Thought and Change (Weidenfeld & Nicolson, London, 1964).72 Noychansamon Denchaleunsouk , ‘Fundamental Questions Concerning Ethnic Groups’ Issue’, Characteristics of

Some Ethnic Groups in Sekong Province, 3 October 1999, Xerox copy (personal communication) , pp. 1 and 2.


evolutionist theory used by Kaysone, it succeeds the ‘clan’ and the ‘tribe’ but is thepredecessor of the ‘national community’ or ‘nationality’. But Kaysone’s distinction between‘nationality’ and ‘ethnic group’ is not always clear. The terms sometimes appear to beinterchangeable in the text. Nonetheless, ‘nationality’—in the sense of cultural com-munity—seems to suggest a more advanced stage, whereas ‘ethnic groups’ are stilldescribed as ‘inconsistent’ and ‘non-homogenous ’ entities.

In effect, slogans such as ‘equality in diversity’ or ‘one people, many ethnic groups’ aresynonymous with the principle of levelling down; this is the reason why claims such asSisouk’s (for the admission of the ‘Bru’ as an ethnic category) have little chance of success.Anyhow, this LFNC of� cial and member of the Communist Party is certainly conscious ofthe risk of politicising his ethnicity, for which he would likely be accused of threateningthe equilibrium of the nation.73 For the purpose of controlling ethnicity, the state in Laosde� nes what is ‘correct’ in terms of language, locality and culture—regardless of a group’ssubjective belief in its existence as a people or in the legitimacy of these state-de� nedcultural traditions. The deviant peoples will not become full members of the ideologicallyde� ned nation unless they stop claiming their right to a self-de� ned identity.

This rule, however, does not seem to apply to the ethnic Lao group. In the Laolanguage, there is only one term (‘Lao’) to designate both the ideas of ethnicity andnationality. The ambiguity is therefore as much linguistic as conceptual. Used in thelegalistic sense, the term ‘Lao’ should be viewed as an a-ethnic and a-racial status,attributed to members of all ethnic groups. The term ‘Lao’, in other words, is used as asynecdoche for the whole population. Thus, ambiguity thrives on the equivocal use of theterm. Yet, textually, ‘Lao’ refers to an ethnic group, since it derives from a group to whichare attached speci� c characteristics. But this rule never strictly applies to the ethnic Lao,because the ‘Lao culture is shared by all the ethnic groups. It is composed of the best ofeach ethnic group’s culture. The ethnic Lao language is the vernacular language, and thescript is the national script for all the ethnic groups’.74 Again, Kaysone wrote:

Our country is among those that have several ethnic groups, of which the ethnic Lao group hasa greater population than the others, located in almost all the provinces and holding a superiordegree of economic and cultural development. Each ethnic group shares common characteris-tics with the Lao national community.75

The census, accordingly, perfectly depicts this form of the nation. Indeed, the ethnic Laoare always put at the top of the list followed by the small, scattered ‘ethnic groups’.Obviously, the principle of levelling down does not apply to the ethnic Lao.

Conclusion

The egalitarian policy that held all ethnic groups to be equal (economically, socially andpolitically) was initiated under wartime conditions as part of the Communists’ survivalstrategy. The goal was to gain highlanders’ loyalty and their support for the Pathet Laotroops. After 1975, the strategic plan was turned into an ideological programme ofpreservation and promotion of every ethnic group’s ‘culture’. The principle, copied from the

73 Sisouk told me about an intriguing conversation he had with his friend, who is himself (in Sisouk’s words) ahigh-ranking ‘Bru’ of� cial. His friend apparently told him that there was a ‘Bru’ association in France. Sisoukthen asked him jokingly if he wanted to set up a similar association in Laos as well, which prompted his friendto retort: ‘Are you mad? Do you want us to get arrested?’

74 Kaysone Phomvihane, Expanding Roots of Solidarity Between Various Ethnic Groups Within the Lao NationalCommunity, p. 49.

75 Ibid., p. 47.


Stalinist model, was to give every member of the ‘multi-ethnic’ state of� cial recognition onan equal footing. In consequence, the new regime created ‘ethnic groups’ as equals in theirmembership of the nation. In reality, the egalitarian ideology serves another purpose: theState, as a vector of ethnicity, actively manipulates, creates, suppresses (or maintains) ethnicboundaries, the ultimate objective being the formation of a homogenous national culture outof real heterogeneity.

Meanwhile, the concepts of ethnicity and nationality seem to be con� ated with regardto the ethnic Lao ethnicity. As Banks comments, ‘the nation’s de� ning group, the one thatclaims the national label as its own … , is not then simply another “ethnic group”, it is verydeliberately and self-consciously everything and nothing’.76 Two interdependent processesare involved in the construction of a nation, once state control is achieved: on the one handreifying diversity—through which ethnicity is enhanced and controlled outside the culturalmainstream but within the national paradigm—and on the other hand, homogenising—theblending-in process whereby ethnicity both as a process and as a category is suppressed.

But, in the case of Laos, these two processes of mapping nationhood are hampered byinconsistency, which reduces their effectiveness. As has been discussed, the 2000 popu-lation census, behind its ordered facade, is in reality the product of multiple negotiationsand arbitrary decisions that leave members of some ethnic groups confused and sometimesresentful. It is the lack of consistency that prevents this technology of power from becoming‘natural’ and from becoming integrated in individuals’ minds. The continued popular andof� cial use of the threefold category—Lao Lum, Lao Theung and Lao Sung—in contem-porary Laos, despite Kaysone’s 1981 call for its replacement, thus epitomises the Laostate’s de� cit of power. Sisouk’s lobbying for the visibility of the ‘Bru’ in the census maybe another example of the regime’s de� ciencies in controlling ethnicity. At the same time,the (ethnic Lao) Majority’s ethnicity is still a long way from establishing hegemony.

76 Marcus Banks, Ethnicity: Anthropologica l Constructions (Routledge, London and New York, 1997), p. 160.




Documents

Laos Nationhood