Savagery has become their character and nature. They enjoy it, because it means freedom from authority and no subservience to leadership. Such a natural disposition is the negation and antithesis of civilization.
—Ibn Khaldun on nomads
As quaint customs and exotic hill tribes are celebrated in museums, the media and tourism, the populace—or perhaps only the urban middle class—comes to know itself by what they once were and who they are not.
—Richard O’Connor
The world I have sought to describe and understand here is fast disappearing. For virtually all my readers it will seem a very far cry indeed from the world they inhabit. In the contemporary world, the future of our freedom lies in the daunting task of taming Leviathan, not evading it. Living in a fully occupied world, one with increasingly standardized institutional modules, the two most hegemonic of which are the North Atlantic modules of individual freehold property and the nation-state, we struggle against the enormous disparities in wealth and power spawned by the former and the ever more intrusive regulation of our interdependent lives by the latter. Populations have never, as John Dunn tellingly puts it, depended “more abjectly for their security and prosperity on the skills and good intentions of those who rule them.”1 And, he adds, the only frail instrument we have for taming Leviathan is another North Atlantic module—via Greece: representative democracy.
The world evoked here is, by contrast, one in which the state has not come so close, as it now has, to sweeping all before it. That world, on a long view, was the world most of mankind inhabited until quite recently. Simplifying greatly, we might identify four eras: 1) a stateless era (by far the longest), 2) an era of small-scale states encircled by vast and easily reached stateless peripheries, 3) a period in which such peripheries are shrunken and beleaguered by the expansion of state power, and finally, 4) an era in which virtually the entire globe is “administered space” and the periphery is not much more than a folkloric remnant. The progression from one era to the next has been very uneven geographically (China and Europe being more precocious than, say, Southeast Asia and Africa) and temporally (with peripheries growing and shrinking depending on the vagaries of state-making). But about the long-run trend there can be not a shred of doubt.
It just so happens that the upland border area we have chosen to call Zomia represents one of the world’s longest-standing and largest refuges of populations who live in the shadow of states but who have not yet been fully incorporated. In the past half-century or so, however, the combination of technological prowess and sovereign ambitions has so compromised even the relative autonomy of Zomian populations that my analysis here has far less applicability to the situation after the Second World War. Since then, throughout Zomia, there has been a truly massive transfer, both planned and spontaneous, of lowland populations of Han, Kinh, Thai, and Burmese to the hills. There they serve the dual purpose of peopling the frontiers with a presumably loyal population and producing cash crops for export, while relieving population pressure in the valleys. Demographically, it represents a conscious strategy of engulfment and eventual absorption.2
Until very recently, however, the massif has signified the basic political choice confronting much of mankind before the hegemony of the nation-state. The choice was not how to tame the inevitable Leviathan but rather how to position oneself vis-à-vis valley states. The options ranged from remote, egalitarian, ridge-top swiddening and foraging—staying as far from state centers as possible—to settling in more hierarchical groups close to valley states to take advantage of the tributary, trading, and raiding possibilities. None of these choices was irreversible. A group could adjust its distance from the state by altering its location, social structure, customs, or subsistence patterns. Even if it altered none of its practices or customs, its distance from an adjacent state could shift under its feet, so to speak, by the collapse or rise of a dynasty, a war, or demographic pressure.
Who were the Zomians? Initially, of course, the entire population of mainland Southeast Asia, whether in the highlands or lowlands, were Zomians in the sense of not being the subjects of any state. Once the first, small, Hinduizing mandala states were formed, the vast majority of those not yet incorporated as subjects became, ipso facto, the first self-governing peoples in an environment that now included (small) states. As it happens, we know something about these nonstate populations on the basis of archeological research. These findings suggest widespread craft specialization and complexity, but in a context that appears politically decentralized and relatively egalitarian (suggested by a rough equality in “grave goods”). The findings are consistent with what some archeologists have called “heterarchy”: social and economic complexity without unified, hierarchical ranking.3 What evidence we have indicates that the hills were sparsely populated and that the bulk of these nonstate populations lived on arable plateaus or in the lowlands, though rarely on the vulnerable flood plains.
As the early states, especially the Han, expanded into the valley lands suitable for wet-rice cultivation, they created at least two kinds of “refugees” who came, over time, to dominate the population in the hills. The first were the hitherto stateless peoples of the plains (many of whom may well have been shifting cultivators), who lay in the path of the padi state’s horizontal expansion. It was from among such groups that the padi state’s original subjects had been gathered. Those who for whatever reason wished to evade incorporation as subjects had to place themselves out of range either on the plains at greater remove from the core or in the less accessible hills. There was, on this reading, a segment of the nonsubject population—those already in the hills and those avoiding the early states—who had never been directly incorporated into state structures. Over the long haul, however, it is clear that the hills were populated increasingly by pulses of migration by state subjects fleeing valley kingdoms for any one of several reasons—corvée labor, taxes, conscription, war, struggles over succession, religious dissent—all having directly to do with state-making. It could also happen that a subject population could find itself suddenly stateless when war, crop failure, or epidemic destroyed the state or impelled people to move to save their lives. On a time-lapse photograph, these pulses of migration might look like a maniacal game of bumper-cars, with each new pulse exerting its own jolt on earlier migrants and they, in turn, resisting or moving into the territory of still earlier migrants. It is this process that has created “shatter zones” and that goes a long way toward explaining the crazy-quilt pattern of constantly reformulated identities and locations in the hills.
Zomia was, in all these senses, a “state effect,” or, more precisely, an effect of state-making and state expansion. Shatter zones and regions of refuge are, then, the inescapable “dark twin” of state-making projects in the valleys. The state and its resulting shatter zone are mutually constituted in the full sense of that much-abused term; each stands in the other’s shadow and takes its cultural bearings from the other. The valley state’s elites define their status as a civilization by reference to those outside their grasp, while at the same time depending on them for trade and to replenish (by capture or inducements) their subject population. The hill peoples, in turn, are dependent on the valley state for vital trade goods and may position themselves cheek by jowl with valley kingdoms to take full advantage of the opportunities for profit and plunder, while generally remaining outside direct political control. Other hill peoples, more remote and/or egalitarian, appear to have structured themselves as something of an antithesis of valley hierarchy and authority. Valley and hill peoples represent two contrasting political spheres, one rather concentrated and homogeneous, the other dispersed and heterogeneous, but each unstable and each constituted of human material pulled, at one time or another, from the other.
Upland societies, far from being the original, primal “stuff” from which states and “civilizations” were crafted, are, rather, largely a reflexive product of state-making designed to be as unappealing as possible as a site of appropriation. Just as nomadic pastoralism is now generally recognized as a secondary adaptation by populations wishing both to leave the sedentary agrarian state and yet take advantage of the trading and raiding opportunities it afforded, so is swiddening largely a secondary adaptation. Like pastoralism, it disperses population and lacks the “nerve centers” that a state might seize. The fugitive nature of its production frustrates appropriation. Hill societies with their deliberate out-of-the-way locations, with their mixed portfolio of linguistic and cultural identities, with the variety of subsistence routines at their disposal, with their capacity to fission and disperse like the “jellyfish” tribes of the Middle East, and with their capacity, thanks in part to valley cosmologies, to form new resistant identities at the drop of a hat, are constituted as if they were intended to be a state-maker’s or colonial official’s worst nightmare. And indeed, they are largely so.
We are, analytically speaking, forced back to the terrain of the elementary units of hill society: the hamlet, the segmentary lineage, the nuclear family, the swiddening group. The uniqueness, plurality, and fungibility of identities and social units in the hills are poor raw material for state-making. Such elementary units may, from time to time, aggregate in small confederations and alliances for war and trade, and under the leadership of a charismatic prophet, but they are likely to lapse, just as soon, into their constituent units. If would-be state-makers found them unpromising, historians and anthropologists have found them equally frustrating. Noting this fluidity and, in particular, the chimerical nature of the major ethnic identities, François Robinne and Mandy Sadan have recently suggested that it would be more ethnographically correct to focus analysis on villages, families, and exchange networks and no longer privilege ethnicity as “a kind of superior artifact, covering other cultural markers; it would become as cultural marker among others.”4 Given the porosity of ethnic boundaries, the bewildering variation within any particular identity, and the historical vagaries of what it has meant to be a “Kachin” or a “Karen,” a healthy agnosticism about the category itself seems just the right move. If we follow Robinne and Sadan’s wise advice, I suspect that much of the flux and apparent disorder is resolved once we examine hill social order and reformulations of identity as strategic repositionings of various villages, groups, and networks vis-à-vis the gravitational force—political, economic, and symbolic—of the nearest valley state.
I have come to see this study of Zomia, or the massif, not so much as a study of hill peoples per se but as a fragment of what might properly be considered a global history of populations trying to avoid, or having been extruded by, the state. Such a task is clearly beyond me and, ideally, it would be a collaborative undertaking by a great many scholars. In the Southeast Asian context alone, it would encompass far more than I have able to examine here. It would, minimally, cover the history of the sea gypsies (orang laut), whose nonstate option was to take to their boats. Dispersed on the water, they could evade slavers and states amid the complex waterways of the archipelago while raiding, slaving, and occasionally serving as mercenaries themselves. They were, for a time, to the Malay Sultanate of Melaka, a watery version of what the Cossacks were to the tsarist armed forces. Their history would intertwine with that of the inhabitants of the mangrove coasts and of the constantly shifting deltas of the great rivers of Southeast Asia. Each of these locations has presented daunting obstacles to state administration and has therefore served as a zone of refuge.
Other peoples and other geographies that might belong to such a global history of extrastate spaces have been mentioned in passing by way of illustration. The Gypsies, the Cossacks, the Berbers, the Mongols, and other pastoral nomads would be essential to a broad history of state peripheries. Maroon communities wherever unfree labor was an integral part of state-building—as it was in most of the New World, Russia, and the Roman and Islamic worlds—would form another part of that global history, not to mention Africans like the Dogon, who evaded capture in the first place. And of course, all those areas of colonial conquest where indigenous peoples were menaced with extermination or were run out of their previous habitats to new locations would form a large chapter of this story.5 The comparative study of such zones of refuge would, despite their geographical, cultural, and temporal dispersion, share a few common, diagnostic characteristics. If they were of any historical depth, they would, like most shatter zones to which various groups have repaired over time, display something of the ethnic and linguistic complexity and fluidity we have found in Zomia. Aside from being located in remote, marginal areas that are difficult of access, such peoples are also likely to have developed subsistence routines that maximize dispersion, mobility, and resistance to appropriation. Their social structure as well is likely to favor dispersion, fission, and reformulation and to present to the outside world a kind of formlessness that offers no obvious institutional point of entry for would-be projects of unified rule. Finally, many, but by no means all, groups in extrastate space appear to have strong, even fierce, traditions of egalitarianism and autonomy both at the village and familial level that represent an effective barrier to tyranny and permanent hierarchy.
Most of the peoples dwelling in the massif seem to have assembled a fairly comprehensive cultural portfolio of techniques for evading state incorporation while availing themselves of the economic and cultural opportunities its proximity presented. Part of this portfolio is the very flux and ambiguity of the identities they may assume over time. So striking is this characteristic—and so vexing to state administrators—that Richard O’Connor has suggested that while we usually start with the assumption that a group has an ethnic identity, in Southeast Asia, “where people change ethnicity and locality rather frequently, we might better say that an ethnicity has a people.”6 It is perhaps one of the features of shatter zones located at the interstices of unstable state systems that there is a premium on the adaptability of identities. Most hill cultures have, as it were, their bags already packed for travel across space, across identities, or across both. Their broad repertoires of languages and ethnic affiliations, their capacity for prophetic reinvention, their short and/or oral genealogies, and their talent for fragmentation all form elements in their formidable travel kit.
We might, in this light, want to consider Fernand Braudel’s assertion about mountain peoples: namely that their “history is to have none, to remain always on the fringes of the great waves of civilization.”7 For Zomia, at least, one would want to recast this argument radically. Better put, they have multiple histories they can deploy singly or in combination depending on the circumstances. They can, as in the case of the Akha and the Kachin, create long, elaborate genealogies or, as in the case of the Lisu and Karen, have minimally short genealogies and migration histories. If they appear to be without a definite history, it is because they have learned to travel light, not knowing what their next destination might be. They are not outside of time nor historyless. Rather like tramp steamers and Gypsies working the seams of the great trade routes and states, respectively, their success depends on maximizing their agility. It is in their interest to keep as many of their options open as possible, and what kind of history to have is one of those options. They have just as much history as they require.
These cultural positionings, along with geographical remoteness, mobility, choice of crops and cultivation techniques, and, frequently, a “no-handles” acephalous social structure, are, to be sure, measures of state evasion. But it is crucial to understand that what is being evaded is not a relationship per se with the state but an evasion of subject status. What hill peoples on the periphery of states have been evading is the hard power of the fiscal state, its capacity to extract direct taxes and labor from a subject population. They have, however, actually sought, sometimes quite eagerly, relationships with valley states that are compatible with a large degree of political autonomy. In particular, a tremendous amount of political conflict has been devoted to the jockeying for advantage as the favored trading partner of one lowland emporium or another. Hills and valleys were, as we have seen, complementary as agro-ecological niches. This meant in effect that adjacent valley states typically competed with one another to acquire hill products and populations.
A favored relationship might, once secured, be formalized by a tributary relationship that, however asymmetrical it might appear ceremonially or in the valley records, might in practice give the hill partner the upper hand. The point is that we must not take valley representations at face value. Beyond the narrow arc of hard tax-and-corvée power, then, lay a very much larger penumbra of economic exchange often expressed in the idiom of tribute. This zone represented a durable link of mutually advantageous trade that, with very few exceptions, did not imply anything in the way of permanent political subordination. The greater the value of the commodities and the smaller their size and weight, the greater the circumference of this penumbra; in the case of gems, rare medicinals, and opium, for example, it could be enormous.8
When it comes to the symbolic and cosmological reach of the great valley states, their influence is both vast and, at the same time, shallow. Whether Sinitic or Indic or, in some cases, an exotic hybrid, virtually all the ideas that might legitimate authority beyond the level of a single village are on loan from the lowlands. Such ideas are, however, cast loose from their lowland moorings and are reformulated in the hills to serve local purposes. The term bricolage is particularly apt for this process, inasmuch as lowland fragments of cosmology, regalia, dress, architecture, and titles are rearranged and assembled into unique amalgams by prophets, healers, and ambitious chiefs. The fact that the symbolic raw materials may be imported from the lowlands does not prevent them from being confected by highland prophets into millenarian expectations that may be used to oppose lowland cultural and political hegemony.9
The role of lowland cosmology in facilitating collective action and overcoming what some social scientists would call the transaction costs associated with social fragmentation may, more speculatively, be related to the overall argument about state evasion. The very features of hill societies that help them evade incorporation—dispersal, mobility, ethnic complexity, small swiddening groups, and egalitarianism—encourage disunity and place enormous obstacles in the way of corporate organization and collective action. The only social resource for such cooperation came, ironically, from the lowland, where social hierarchy and the cosmology that goes with it are taken for granted.
Virtually all hill societies exhibit a range of state-evading behavior. For some, such characteristics are compatible with a degree of internal hierarchy and, from time to time, imitative state-making. For other groups, however, state evasion is coupled with practices that might be termed the prevention of internal state-making. Relatively acephalous groups with strong traditions of equality and sanctions against permanent hierarchy, such as the Akha, Lahu, Lisu, and Wa, seem to belong to this category. State-preventing societies share some common characteristics. They are likely to prevent the emergence of any permanent ranking of lineages through marriage alliances; they are more likely to have cautionary legends about the assassination or expulsion of overreaching headmen; and, finally, their villages and lineages are likely to divide into smaller and more egalitarian fragments when inequalities do threaten to become permanent.
There is a paradox in trying to describe a shatter zone or region of refuge such as Zomia. In order to portray the flux and plasticity of hill societies one has, necessarily, to stand somewhere, even if that “somewhere” is itself in motion. I have surely done this by talking about “the Karen,” “the Shan,” and “the Hmong” as if they were solid, static units of social organization. They are not, particularly when observed over any considerable period of time. So at the risk of further dizzying the reader and myself, we ought to recall just how radical this flux is. Valley runaways have been replenishing the hill population for as long as we can tell. Hill people have been assimilating into valley-state societies also for as long as we can tell. The essentialized “line” between hill and valley peoples remains in place despite quite massive traffic back and forth in each direction. Hill societies themselves are porous; gradients of identity make any firm “identity frontier” quite arbitrary. While hill societies reformulate themselves, so do individuals, kin groups, and whole communities. And while hill societies are positioning themselves vis-à-vis state projects in the valley, they are also positioning themselves vis-à-vis their hilly neighbors in this complex constellation of peoples.10 There is nothing particularly unusual about this; the process of positioning and mutual adaptation is, to a large degree, the leitmotif of hill politics. If it makes our heads swim, it is some consolation that while it perplexes colonizers and state officials, the actors themselves are neither confused nor mystified about who they are and what they are doing.
Adaptation to the dangers or temptations of neighboring polities is hardly a practice confined to peoples at the periphery of states. Core peasantries as well have developed routines to take advantage of favorable developments at the political center and to shield themselves from the worst effects of turmoil. The repertoires deployed by the Chinese peasantry during the Ming and Qing dynasties to cope with dynastic collapse and with order and prosperity have been elaborated in some detail by G. William Skinner.11 What is distinctive about these repertoires, for our purposes, is that they represent defensive measures by a peasantry that remains where it is and continues to practice sedentary agriculture. It illustrates a pattern of self-defense under very constrained circumstances. Seeing how core peasantries adapt in this fashion will help us appreciate how the far wider options open to peoples on the state periphery operate.
In periods of dynastic consolidation, peace, and buoyant trade, Skinner explains, the local community opens and adapts to the opportunities these conditions afford. Economic specialization, trade, and administrative and political links flourish as the community takes advantage of the opportunities in the wider world. By contrast, in periods of dynastic collapse, economic depression, and civil strife and banditry, the local community withdraws increasingly into its own shell as a self-protective measure. The withdrawal was patterned, according to Skinner: first a normative withdrawal, then an economic closure, and, finally, a defensive military closure. Specialists and traders returned home, economic specialization diminished, the local food supply was guarded, outsiders were expelled, crop-watching societies were formed, stockades built, and local militias created.12 When flight and rebellion apparently were not available options, what the local community did in the face of a threatening external environment was to secede normatively, economically, and militarily. It tried, without budging, to create an autonomous, autarkic space—in effect declaring its independence from the larger society while the danger lasted. And when the threat subsided, the local community reopened in the reverse order: first militarily, then economically, and finally, normatively.
Zomia’s hill societies have, in a comparable but far more expansive fashion, a large bandwidth of configurations among which they can move to integrate themselves more closely with the neighboring polities or, alternatively, to keep them at a distance. Unlike Skinner’s stuck-in-the-rice-padimud Chinese peasantry, uplanders are physically mobile, capable of moving considerable distances, and in possession of a range of subsistence techniques they can deploy singly or in combination as the circumstances dictate. Upland society itself, after all, was created largely by a series of secessionists who, however, are capable of adjusting or modulating the degree of their secession in one direction or another. Such adjustments can take place along one or several dimensions not easily available to the core peasantry. The first of these dimensions is location; the higher and more remote their dwelling, the farther they generally are from state centers, slave raids, and taxes. A second dimension is scale and dispersal; the smaller and more dispersed their settlements the less tempting a target they represent for raiders and states. Finally, they can and do modulate their subsistence techniques, each of which embodies a position vis-à-vis states, hierarchy, and political incorporation.
Hjorleifur Jonsson, in this context, contrasts three subsistence strategies: 1) foraging–hunting and gathering, 2) swiddening, and 3) fixed-field agriculture.13 Foraging is virtually appropriation-proof and permits little in the way of social inequality. Swiddening is appropriation resistant, though it may generate a surplus and some, usually temporary, internal hierarchy.14 Fixed-field agriculture, especially of the wet-rice variety, is appropriable and raidable and is associated with large settlements and durable social hierarchies. These techniques could be combined in various proportions and adjusted over time, but it was quite clear to the Yao/Mien making such choices that any adjustment expressed a political option. Foraging and swiddening were both understood by those who practiced them as forms of political secession from the lowland state, with foraging the more radical, distancing choice.15
Upland groups have, then, a large bandwidth of possible locations as well as social and agro-ecological configurations available to them. They run the gamut all the way from, say, taking up padi cultivation on the plains and inviting incorporation as peasants into the valley state to the other extreme of foraging and swiddening in remote, fortified, ridge-top settlements while cultivating a reputation for killing intruders. Between these stark polar opposites lie a host of hybrid possibilities. Which of these options is actually exploited at any particular time will depend, in part, as with Skinner’s Chinese peasants, on external conditions. At times of peace, economic expansion, and state encouragement of settlement, hill groups are more likely to take up sedentary cultivation, move closer to state centers, seek tributary and trade relations, and drift ethnically and linguistically toward valley cultures. At times of war, turbulence, and rapacious taxes and slave raids, hill groups would drift in the opposite direction and, in all likelihood, be joined by refugees fleeing the state cores.
Any particular hill people will appear, at any moment, to have adopted a particular configuration, say, as hilltop swiddeners and opium growers. Their culture may even appear to confine them to that configuration. But over long stretches of time there is likely to have been considerable movement, often by various fragments of the same ethnic group who have found themselves in different situations. Nor is there any reason whatever why the movement should be in a single direction.16 On the contrary, on any long view, there is every reason to imagine a history of dozens of reformulations and adjustments toward and away from valley states, all assimilated successfully as “tradition” within a pliable oral culture.
Here it is worth recalling that most foragers and nomadic peoples—and perhaps swiddeners as well—were not aboriginal survivals but were rather adaptations created in the shadow of states. Just as Pierre Clastres supposed, the societies of many acephalous foragers and swiddeners are admirably designed to take advantage of agro-ecological niches in trading with nearby states yet manage to avoid subordination as subjects. If one were a social Darwinian, one might well see the mobility of hill peoples, their spare dispersed communities, their noninherited rankings, their oral culture, their large portfolio of subsistence and identity strategies, and perhaps even their prophetic inclinations as brilliantly suited to a tumultuous environment. They are better adapted to survival as nonsubjects in a political environment of states than to making states themselves.
British and French colonial administrators, justifying the novel tax burdens they were imposing on their subjects, often explained that taxes were the inevitable price one paid for living in a “civilized society.” By this discursive legerdemain they neatly managed three tricks: they described their subjects as effectively “precivilized,” they substituted imperial ideals for colonial reality, and above all, they confounded “civilization” with what was, in fact, state-making.
The “just-so” story of civilization always requires a wild untamed antagonist, usually just out of reach, to eventually be subdued and incorporated. The hypothetical civilization in question—whether French, Han, Burman, Kinh, British, or Siamese—is defined by this negation. This is largely why tribes and ethnicity begin, in practice, where sovereignty and taxes stop.
One can see in a flash why these just-so stories, concocted largely to improve the self-confidence and cohesion of the rulers, might be less than convincing at the frontiers of empire. Imagine, for example, an education in the Confucian classics—filial piety, observance of the rituals, the obligations of rule, benevolent care for the well-being of the subjects, honorable conduct, rectitude—in the context of, say, the mid-nineteenth-century frontier in Yunnan or Guizhou. How could one not be struck by the chasm between these imperial imaginings on the one hand and the realities of the Ming and Qing frontier on the other? The “lived” frontier, as distinct from the discursive frontier, was rife with corrupt civil magistrates selling justice to the highest bidder, military adventurers and bandits, exiled officials and criminals, land grabbers, smugglers, and desperate Han settlers.17 Small wonder that the ideals of Han civilization had little traction on the ground. On the contrary, the contradiction between ideal and reality was sufficient reason both for local people and for reflective imperial officials to conclude that the civilizational discourse was mere humbug.18
The Han and Theravada polities of China and Southeast Asia had somewhat different perceptions of the ideal, “civilized” subject. In the Han case there was no religious test for civilization, although the patriarchal family, ancestral tablets, and knowledge of the characters presupposed ethnic assimilation. In the Burman or Thai case, Buddhism and the veneration of the sangha did constitute a religious test although, on the other hand, the manpower-starved states of mainland Southeast Asia could not afford to be ethnic snobs. The Indic-style classical kingdoms were hierarchical, like the Han, but ethnically quite inclusive.
All such states, however, were, for impelling fiscal and military reasons, padi states. In practice, therefore, the padi state did everything in its power to encourage densely concentrated settlement and the irrigated wet rice that fostered it. To the degree that its subjects grew the same grains in roughly the same way, in communities that were roughly homogeneous, the tasks of land valuation, taxation, and administration were that much easier. In the Han case, the codification of the patriarchal household as the basic unit of property and administration further facilitated social control. The ideal subject of the padi state also represented a vision of landscape and human settlement in which the cleared plains of irrigated rice fields and their human communities came to represent an ideal that was, at once, cultivated and cultured.
The padi state’s officials had, on the other hand, every incentive to discourage all form of settlement, subsistence, and social organization that represented an inappropriable landscape. They discouraged and, when they could, prohibited dispersed settlement, foraging, swiddening, and migration away from the core. If the padi fields had come to mean civilized landscape of properly organized subjects and their production, then by extension those who lived in remote places, in the hills or in the forests, who shifted their fields and often shifted themselves, who formed and re-formed small egalitarian hamlets were uncivilized. What is most striking here, of course, is how closely the ideal of a civilized landscape and demography coincides with a landscape and demography most suitable for state-making and how closely a landscape unsuitable for state appropriation, as well as the people who inhabit it, is understood as uncivilized and barbaric. The effective coordinates, from this perspective, for figuring out who is civilized and who is not, turn out to be not much more than an agro-ecological code for state appropriation.
The tight correlation is unmistakable between life at the margins of the state on the one hand and primitiveness and backwardness on the other, in the view of valley elites. One has only to list the most salient characteristics of landscapes and peoples beyond the state’s easy grasp to produce, simultaneously, a catalogue of primitiveness. Dwelling in inaccessible forests and on hilltops codes as uncivilized. Foraging, forest collecting—even for commercial gain—and swiddening also code as backward. Scattered living and small settlements are, by definition, archaic. Physical mobility and transient, negotiable identities are both primitive and dangerous. Not following the great valley religions or not being the tax- and tithe-bearing subjects of monarchs and clergy places one outside the pale of civilization.
In the valley imagination, all these characteristics are earlier stages in a process of social evolution at the apex of which elites perch. Hill peoples are an earlier stage: they are “pre-” just about everything: pre–padi cultivation, pre–towns, prereligion, preliterate, pre–valley subject. As we have seen at some length, however, the characteristics for which hill peoples are stigmatized are precisely those characteristics that a state-evading people would encourage and perfect in order to avoid surrendering autonomy. The valley imagination has its history wrong. Hill peoples are not pre- anything. In fact, they are better understood as post–irrigated rice, postsedentary, postsubject, and perhaps even postliterate. They represent, in the longue durée, a reactive and purposeful statelessness of peoples who have adapted to a world of states while remaining outside their firm grasp.
There’s nothing particularly wrong with the valley understanding of the agro-ecology, social organization, and mobility of the peoples who elude them. They’ve sorted these people, as it were, into the right bins. In addition to radically misunderstanding the historical sequence, however, they have got their labels wrong. If they merely substituted “state-subject” for “civilized” and “not-a-state-subject” for “uncivilized,” they’d have it just about right.