Quizzical notes on Zomia

State evasion is over

[Savage Minds welcomes guest bloggers Leif Jonsson, Masao Imamura, and Jacob Hickman, who offer individual takes on some issues raised by James Scott’s The Art of Not Being Governed: An Anarchist History of Upland Southeast Asia (Yale U P, 2009). Kerim’s previous post on the book is here. This post is by Leif.]

James C. Scott’s The Art of Not Being Governed: An Anarchist History of Upland Southeast Asia is an interesting read. If anthropology-readers are used to embarrassment regarding the gaze on tribal peoples, then here is a license to guilt-free gawking: These weren’t tribals but rather freedom-seeking secessionists from the lowlands. There were no real ethnic others, the book suggests. Instead, linguistic and cultural diversity and the profusion of ethnic labels are just markers of state-evading strategies. In my view this is all rather problematic, in that clueless western readers (people ignorant of, say, particular histories, cultures, societies, languages, peoples, or politics in Southeast Asia) are invited to feast on the identities and politics of the Southeast Asian hinterlands without any involvement.

The effect bears some resemblance to the fickle fascination with Indians of the Brazilian Amazon as natural allies of the rainforest, that evaporated once the noble Other was seen as somehow too modern. Scott draws explicitly on the work of Pierre Clastres regarding the Guayaki and other Indians of Latin America, that the Indians had run away from the state and hierarchy and all that. Clastres had been a student of Levi-Strauss, and his early tribalist work was deeply fatalistic regarding the looming disappearance of all indigenous peoples. Clastres’ shift in focus, from pre-contact- to ex-contact peoples does not remove the assumed purity of the tribal slot but instead relocates its source. The tribals aren’t pure because of their cultural- or other essence, but because they ran away from the source of all pollution (the state, with its inequality, taxation, sedentary lifestyles, and other contaminants).

If this holds, then it suggests that we have successfully abandoned the “savage slot” without changing much of anything. We can still gawk at peoples in the tribal zone, as long as we call attention to the “monster slot” of the state (or capitalism, or development discourse, and so on). I will not blame Scott for this tendency. Rather, I suggest that all of the western anthropology about mainland Southeast Asia had uncritically recycled colonial- and Asian nationalist projects of racial divides and stages of civilization. What the Zomians of Scott’s book share is twentieth century discrimination, of being denied sameness, relatedness, and/or equality in the modern states of Thailand, Burma, Laos, Cambodia, and Vietnam. They were brought together, analytically and descriptively, in various classificatory projects that denied them diversity, specificity, and any need for political negotiation: These were ethnic types, distinct from national society. The anti-defamation-effort of Scott’s analysis, to show that these were not traditional tribals but instead clever freedom-seekers, has a similar effect: These are not particular people in specific circumstances but the shadows of characters in a western academic drama about freedom and history.

Much twentieth century western anthropology was in the search for tribal peoples, and the credibility of researchers was partly based on their ethnic specialization (one scholar each for the Hmong, Lahu, Lisu, Mien, Lamet, Mnong Gar, Kachin, and so on). The structuring of research and expertise contributed to the notion of ethnic distinctness, similar to what went on regarding the Indian communities of the Amazon. Edmund Leach’s Political Systems of Highland Burma leaves the impression of people making strategic choices among rival “esteem systems” – stratified Shan, egalitarian Kachin (gumlao) and hierarchic Kachin (gumsa). This is one source for Scott’s ‘radical constructionist’ case for ethnogenesis. The areal or regional focus of Scott’s book is the best defense against the traps of tribalist specialization (the search to define the expression of Mien ways, say, and the general dismissal of their diversity, entanglements with various others, and of their historical specificity). But the regional focus arrives at predetermined conclusions: States are oppressive; sedentary agriculture invites hierarchic social organization; the only defense is voluntary assemblies of like-minded freedom seekers in the nooks and crannies outside the state’s reach.

Scott does not offer his readers the “same old” shadow puppet of traditional tribals. He denies the relevance of language, ritual, social entanglements such as kinship, or strategies of livelihood, beyond how these situate people within the binary of freedom and subjugation. The anthropologists that Scott draws on were interested in various ethnic and cultural others. Their notion of history and identity assumed destruction or dissolution with “contact,” so much of the ethnographic work was distrustful of any process of national integration or political negotiation. This bias in the ethnographic record is entrenched through the notion that the highland peoples were running away from the state – no matter how much Scott argues against tribalist expectations. That is, the sources contain ontological and epistemological elements that can be unwittingly reproduced for as long as scholars and others assume that Southeast Asian highland peoples are a particular social type.

The case for strategies of state-avoidance assumes a historical break soon after the Second World War. This replicates how tribal anthropology had assumed tradition (and adaptation to the environment) that came to an end by the 1950s or ‘60s. These distinct analytical approaches agree on the authenticity of the past and on modernity as a force of homogenization and loss. This may be compelling to an audience that has no personal stake in the social situation, but implies to the highland people that they were all doing some great stuff until the 1950s, which we appreciate, but that it’s all been over for a long time.

But what is the historical predicament of hinterland peoples in mainland Southeast Asia? Since the 1950s, many became engulfed in war and related conflict. In Burma, it was ethnicized civil war, while in Laos and Vietnam it was national conflict that was blown out of proportion by US involvement after French colonial authorities had left. In both Laos and Vietnam, highland peoples were recruited to fight, for the leftist nationalists as much as for their US-supported rivals. If the story came to an end by 1950, we can enjoy the spectacle of Southeast Asian history and ethnology in ways that brush aside the realities and repercussions of those wars. It is a little bit spooky to be left with a post-Cold-War image of pre-Cold-War freedom fighters, and no sign of weapons, war, or of the specifics of the Cold War context in Laos, Vietnam, Cambodia, Thailand, or Burma.

If I did not feel implicated in the fashioning of this imagery then I could position myself somehow against Scott’s case. But I think that we are all in trouble, and that it is reasonable to aim for different ethnologies and politics. What does it suggest about the world of academics and their audiences that we easily imagine an essential tension between the state and the people, that this tension sets history and society in motion, and that we are keener on identifying with virtue than on undoing the conflict?

17 thoughts on “Quizzical notes on Zomia

  1. I wrote my own one-page demolition of the Zomia hypothesis some time ago…


    …and this isn’t the first time I’ve seen H.J. comment on the controversy (I’m thinking of some conference paper or another?).

    What Scott built atop the assumptions of Edmund Leach & Tambiah is both (1) wrong in itself, and also (2) wrong in that it recapitulates and makes more absurd those same tenets from Leach and Tambiah.

    The ethnic differences between these groups is real; the history of them going to war with one-another, enslaving one-another, etc., is real; the linguistic and religious differences between them are also very real –and indicate differences that date back along a much longer historical timeline. By contrast, Leach’s “ethnogenesis by altitude” hypothesis is completely unreal –it was never supported by facts, and continues to be contradicted by them. Tambiah’s very impressionistic remarks on some Dvaravati artwork (!) that were made into “the Mandala hypothesis” should have remained at the level of mere remarks, and never deserved to be fetishized into a prescriptive theory by anthropologists (nor anyone else).

    “Either this wallpaper goes, or I do”.

  2. It has been a long, long time since I read Leach’s Political Systems of Highland Burma for a course on Mainland Southeast Asian ethnography taught by Cold Warrior Lauriston Sharp at Cornell in 1969. My memory suggests, however, that reading Leach as advocating an “ethnogenesis by attitude” approach is, perhaps, a misreading. As I recall the argument, Leach assumes that the people he is writing about have two models, gumlao and gumsa for political organization and that actual groups on the ground tend to oscillate back and forth between them. If conditions on the ground favor stability and population growth, social stratification emerges justified by gumsa principles. That, in turn, may lead to class conflict that leads to the formation of new communities initially organized on gumlao principles. Right or wrong, what Leach describes is not people freely choosing one model or the other but a recurring social process. And, in any case, the models are givens, not supposed to be the result of the process, in Leach’s analysis.

    From a broader perspective, I do not imagine Leach disagreeing that,

    The ethnic differences between these groups is real; the history of them going to war with one-another, enslaving one-another, etc., is real; the linguistic and religious differences between them are also very real –and indicate differences that date back along a much longer historical timeline.. But I can imagine his arguing that these facts are epiphenomenal, part of what needs to be explained, and suggesting that anthropologists should look deeper than this surface detail and construct models like the one he offers.

  3. i read Scott’s book last year, and think that he would agree that highland groups, etc., fought among each other. However, he wold also I think argue that the number one dominating fact in their push up into the hills was the growth of densely populated lowlands in China, Myanmar, Vietnam, and SIam. In doing so he makes the point that what anthropologists think of traditionally as “ethnicity” is indeed epiphenomenal, as John McCreery mentions above. So yes, ethnic/linguistic differences are real, but they are also a product of relations between the highland groups, as well as the continuing pressure from the lowland tax collectors that Scott claims such groups “resist.”

    I will try to find a copy of Scott to take a second look in the context of what you write here.

  4. A quizzical note on the reply of John McCreery:

    Where I wrote *altitude,* McCreery has mis-read *attitude.*

    That is a distinction with a difference.

    Leach’s assertion that the altitude of the culture (in three strata, still followed by the Lao government as official categories: highland, midland, lowland) was more important than very palpable differences of language, ethnicity and religion is demonstrably false (both in contemporary ethnography, and in local history).

    Leach was both misleading and misled in this regard, and he was so far off the mark in what he says about the differences between the languages of that region that (I surmise/infer) he must have had extremely good translators provided to him by the British military (he was on military assignment at the time, you know) in order to have even been able to overlook the evidence that was so starkly to the contrary of his hypothesis.

    This is not some minor detail that can be run roughshod over with such statements as, “Right or wrong, what Leach describes is…”. No no, in establishing facts through the methods of ethnography, it really does matter whether these things are right or wrong. Is it right or wrong? Let’s go out into the field, talk to villagers, and then start with that –we cannot start from an abstract theory, to then force the facts to fit.

    The facts as perceived and presented by the natives themselves (from the native perspective) are much, much more important than the top-down synoptic view of an outsider –especially if that outside view contradicts (and is contradicted by) real evidence.

    In Northernmost Laos, the fact that the Tai Lue and the Tai Neua regard themselves as completely separate ethnoi (and have a long history of warfare against one another, enslaving one-another, etc. etc.) is a very important fact –regardless of any outside theory that would put them into the same category. If they regard their own languages as totally distinctive, too, this is an important fact, even if a linguist can indicate how much they have in common (etc. etc.) despite this cultural (and even political) perception.

    My disagreement with Leach’s thesis is indeed fundamental: the difference between (e.g.) Mon-Khmer languages and Tai-Kadai languages is NOT an “epiphenomenon” –and the relationships between these ethnic groups (historical or current) cannot be explained through a so-called “model” that is constructed out of false facts.

    Leach’s work remains a laughing-stock to anyone who has actually lived and worked in the peculiar region described (on the edges of Burma, Laos, and Yunnan) –but it is taken terribly seriously by Anth. undergraduate students who know nothing but abstractions, and who are taught to worship theory for the sake of theory (the more abstract the better!).

  5. @Eisel

    “Altitude” not “attitude”: My bust. Thank you for the correction. Re the thinness of Leach’s ethnography: I’ve heard this before. We have no disagreement there. Where we may still differ is on the relation of ethnography to theory. I worship neither and maintain what I would describe as a skeptical and generous stance toward both.

    I am perfectly willing to consider the likelihood that Leach’s theory is wrong. Theories often are. But what I would like to hear more of, particularly from those with local knowledge like yourself, is precisely why it is wrong and what, if anything, has replaced it.

    I would also like to hear more about the Tai Lue and Tai Neua, about whom I know nothing. Are their languages as different as Mon-Khmer and Tai-Kadai? If so, why are both referred to with names that share the initial syllable “Tai”? There must be an interesting story here.

    I would certainly agree that if members of the two groups see themselves as their languages as totally distinct, this is a social fact that must be taken into account. But how long has that “fact” been around? How far back does the enmity go? Are there no members of the two groups who recognize similarities between them? In whose interest is the difference asserted? Whose ox is gored if the difference is denied? How would this difference compare with, for example, that between the Nuer and the Dinka, Israelis and Palestinians, or Protestant and Catholic Irishmen? Do intermarriage, adoption, trading partnerships or friendships cross-cut the difference in question?

    Anyway, lots to talk about here.

  6. Well, feel free to get in touch, McCreery, because the number of human beings with any interest in these things is dangerously close to zero.

    In my reply to you, above, I was making an implicit contrast between ethnoi that are members of totally separate language-families (Mon-Khmer vs. Tai-Kadai) and ethnoi that are very much of the same language-family (Tai Neua vs. Tai Lue) and yet will themselves explain to you how stark the difference is between themselves and the others who share a category with them –i.e., a category imposed by outsiders.

    The categories imposed by Leach are, however, especially absurd. In their further application by James Scott they become exponentially more absurd, as they are made to pull the cart of a supposed ideology of anarchism, that misrepresents intensely hierarchical, slave-trading societies into supposed exemplars of “anarchy”, on such flimsy grounds as the occasional quote referring to such peoples as “anarchic” (in a desultory sense) by the representatives of the British Empire who rather failed to “pacify” (i.e., conquer) the region, in the explorations that were presumed to set the stage for a (doomed) railroad to Yunnan through that part of Burma.

    I can honestly say that I have read much more useful descriptions of the traditional relations between highlanders and lowlanders written by Lao officials –both because their propaganda purpose was much less perverse than Scott’s, and because they were built on real ethnography, and real historical research, conducted by fluent speakers of the languages in question.

    Please take a look at my one-page article against Zomia (linked to above) and you can see also my longer “Laos in 1893” on the subject of that doomed railroad (a project that indirectly illuminates much of the British misinformation on the region, i.e., the peculiar alignment of interests –such as the opium trade and the fear of the French presence on the Mekong– that resulted in any of this information being created by these peculiar military expeditions, that have just lately been dragged from obscurity by this absurd “fad” in anarchist anthropology).

  7. @Eisel

    How shall I get in touch? I have been browsing the various sites to which clicking on your name here leads me, but the same aging eyes that mistook “altitude” for “attitude” are not discovering any straightforward way to send you email. The search, however, has been interesting. I have just briefly scanned the Oxford Buddhist Society lecture Language Hierarchy, Buddhism and Worldly Authority in Yunnan, Laos, Etc.and found it very rich, indeed. Your remarks about law, lawyers and legal language are particularly thought-provoking.

  8. P.P.S. I would be delighted to have your collection of materials on Hokkien. I was, oddly enough, the first U.S. anthropologist to conduct fieldwork in Taiwan in Hokkien instead of Mandarin.

    Re your despair over linguistic variations, one of my most vivid grad school memories was a short course in Hokkien taught by Nicholas Cleveland Bodman, who had written the textbooks used by the British Forces in Malaya while fighting the Communist/Chinese insurgency there. Very curious textbooks they were, indeed. Learned how to say “pistol” and “hand grenade” before I learned how to say “thank you.” But, back to the point at hand, our native informant was a Taiwanese graduate student. When Bodman asked him how far he had to travel to hear dialectical differences, he answered, “The next village.”

  9. @Eisel: (and John)
    Could one of the problems of understanding ethnoi be that we correlate a single language too closely with a single label? Many of the highlanders from northern Thailand and Laos I met years ago spoke several languages, while maintaining more or less exploitative relationships with the lowland Lao or northern Thai. Identity roughly seemed to follow parentage–but only roughly so. There were also adoptions, marriages, and slave-like relationships that crossed linguistic boundaries the HTin of Nan/Xayaboury who presumably had a Mon-Khmer background, but spoke Tai dialects at home come to mind. What are your thoughts about multi-lingualism in such contexts?


  10. My disagreement with Leach’s thesis is indeed fundamental: the difference between (e.g.) Mon-Khmer languages and Tai-Kadai languages is NOT an “epiphenomenon” –and the relationships between these ethnic groups (historical or current) cannot be explained through a so-called “model” that is constructed out of false facts.

    This was almost exactly what I was going to post before I had to leave for work this morning and as soon as I saw the title of the SM post I knew Eisel would be here. Good to see.

    I don’t think Leach’s model is ‘wrong’, exactly. I find it reasonably convincing when it comes to looking at cycling and achievement-based societies around the world – Flannery and Marcus use Political Systems… as an example of such phenomena. It feels to me as if Leach thought up the model and then went to war in Burma rather than the other way around – it applies beautifully to plenty of other situations, after all (and I honestly wouldn’t lay much importance on the ethnic aspect of it, which Eisel objects to most; it’s more about competition, aspiration, and mutually incompatible attempts to grab power).

    At the same time, Eisel has a very important point: while it may be true that many highland southeast Asian societies are multi-lingual, the different languages in the area are not relics of people fleeing into the hills from states, and they aren’t epiphenomena of state avoidance or ‘Zomian’ anarchism. Three separate language families with very different origins can be found in Leach’s work: Austroasiatic (perhaps the oldest in the region), Sino-Tibetan (likely more recent), and Tai-Kadai (introduced to what is now Thailand and Laos, and elsewhere, under a thousand years ago, with the first Tai inscription in Thailand dated to 1167 CE (Inscription K966 from Dong Mae Nang Muang, in case you’re interested)).

    Shan is Tai-Kadai (the Shan endonym is tai, = Thai [effectively], and Shan is related to ‘Siam’). Burmese is Sino-Tibetan (or Tibeto-Burman). Mon is Austroasiatic, as are Wa and Palaung (if I remember correctly). Austroasiatic has an Urheimat along the Mekong in mainland SE Asia c.2000 BCE (according to Paul Sidwell). The information I have available puts Sino-Tibetan on the Tibetan Plateau about 5,000 years ago, although I’m not sure if that’s the current consensus. Tai-Kadai languages are most diverse in southern China and almost certainly originated there (the Mongol conquests are often claimed as the impetus behind the expansion further into SE Asia, although it appears that Tai-speakers were present in the area for at least a century before this).

    This is not the pattern you get if the linguistic diversity is an epiphenomenon of avoiding the lowland states. These are not arbitrarily and deliberately diversified versions of lowland languages, but languages with very different origins showing descent from three separate proto-languages fairly far back in the past.

    Also, John – Eisel has a really cool website that you can access here, if you’re interested.

  11. Al, I like this comment a lot. What a pleasure it is to have someone allude to facts on the ground. I also like your take on Leach. Physicists don’t kvetch about Copernicus because he thought the Earth’s orbit was a perfect circle around the Sun. Why are we anthropologists so intent on killing our ancestors instead of building on what they have done?

    But, enough of that meta stuff. I have in my mind an image of Han civilization expanding southward from North China, either assimilating or pushing further south the peoples in its way. The mountains around the rim of what is now China are a fracture zone, in which the highly dissected creates a wealth of niches in which survivors of original inhabitants or earlier invaders can survive in relative isolation while others spill out of the mountains and onto the plains in what is now mainland Southeast Asia.

    The inspiration of this model is Europe, with barbarians pushing westward out of Central Asia, in successive waves that, in the type case of Britain, wind up with smaller, gaelic speaking populations in Cornwall, Wales and the Scottish Highlands, with descendants of Danes, Angles and Saxons and later Norman French ancestors filling what is now England. One also thinks of Italy, Germany and the Balkans before the 19th century unification of nation-states and, one particularly successful example of highland survival, the French, German, Italian and Romansh-speaking Cantons that make up Switzerland.

    Care to comment from the perspective of someone who clearly knows a lot more about recent research in Southeast Asia than I do?

  12. The linguistic/cultural diversity of the highlands has been maintained by the fact that no state ever conquered the entire region (even the British failed in this task, although they came closest), and that is probably down to the geography of the place. But it wasn’t pressure from China that was responsible for the origins of the linguistic diversity in the region. There were probably lots of push and pull factors at work drawing different groups into southeast Asia, and some of these migrations were truly prehistoric, occurring before writing was invented in China and before the first Chinese states rose.

    Southeast Asia as a whole certainly shouldn’t be seen as a backwater only attractive as a refuge from China, either. There were plenty of ‘pull’ factors in addition to the ‘push’ from Chinese states, including the fertility of the land, the presence of navigable rivers, and the ease of trade connections with the Indian Ocean and South China Sea. The trade by sea was immensely profitable, as was the trade through Myanmar-Yunnan (ancient Pagan and Nanzhao), and many of the native states were powerful and important in their own right. It’s not just, or even primarily, a refuge.

    The linguistic situation is also a little problematic. Austroasiatic (aka ‘Mon-Khmer’) seems to be on quite firm ground, but even the name of Sino-Tibetan is contested (specifically by George van Driem, who prefers ‘Trans-Himalayan’). Tai-Kadai, or ‘Daic’, is well-established linguistically and the reason for the diffusion into southeast Asia may have been the spread of the Chinese state and people into southern China in a big way during the Song and Yuan eras. In any case, the highlands of southeast Asia proper are nothing in terms of linguistic diversity to Arunachal Pradesh, which seems to be excessively diverse. Most of its languages are classified as Sino-Tibetan, but this is apparently not the case, with many of the languages probably present before Sino-Tibetan arrived.

    It seems like the mountains exert a trapping effect, preserving linguistic diversity. I feel obliged to point out that this is much the same ‘geographical determinism’ as Diamond supposedly espouses! 😛

  13. Al, thanks for continuing the conversation. Sorry I gave the impression that Southeast Asia was a big empty space until peoples pushed south by Han expansion filled it. What I had in mind is more like the history of Britain, in which Celtic tribes exterminate or assimilate earlier inhabitants and are then, in turn, exterminated or assimilated by Danes, Angles and Saxons, who in turn are conquered by the Normans, with remnants finding refuge in the hills (whether they were there and simply remained untouched or actively fled there is an open question). Here, too, we may see the trapping effect you attribute to the mountains. The Southeast Asian case is, of course, different because, depending on historical period, there may be different expansionary states on both sides of the mountains, so that the East to West movement in terms of European history is recounted is almost certainly a too simplistic model.

    I wonder if anyone here knows what happened in the Pyrenees?

  14. I am the ignorant westerner of which the author of this blog post speaks. I find it interesting however reading about how the independent tribes of Zomia have been able to retain their independence from the nation states. However it seems that Scott portrays the nation-state as monogamous groups that only desire that being to bring in the tribal people under their control. In doing so he misses some key historical factors. In describing the Karen he states that the atrocities the Burmese army army commits against them while they must flee for their lives, yet all the while Scott describes them as a “free people” that fight for autonomy. Yet it seems it might be better stated a autonomous group that fight for freedom. Later on he states they are a historical group that are “historical[ly] fear slavery”. But it seems that they are justified in their fear and that their fight is much more then just a fight for autonomy.
    That being stated I also think that he doesn’t fully capture the nation-states nor their struggles or motives.

Comments are closed.