The Mexico-United States Border in Anthropology : A Critique and Reformulation

The Mexico-United States border has recently been used in anthropology as a metonym for the study of inequality, power, global economics, and connections among cultures and societies. This use occurs not only in studies that literally describe MexicoU.S. border (or near-border) locales (e.g., R. Alvarez 1987; R. Alvarez and G. Collier 1994; L. Chávez 1992; M. P. Fernández-Kelly 1983; J. Greenberg 1987; J. Heyman 1991; M. Kearney 1991; C. Vélez-Ibáñez n.d.) but also in works that are either theoretical (e.g., A. Gupta and J.Ferguson 1992; R. Rosaldo 1988) or works that concern relationships between Mexicans and the United States conceived broadly (e.g., R. Behar 1993; R. Rouse 1991). The Mexico-U.S. border (simply "the border" henceforth) contains many well-publicized developments--immigration law enforcement, maquiladoras (and thus, NAFTA), and cultural interchange--that make it appear to be relevant and happening for intellectuals. American anthropology has, of course, emerged into the search for relevance from an era that largely emphasized the romantic search for cultural distance. In this change, anthropologists have mixed sense with nonsense, arch rhetoric with penetrating rethinkings of flawed social science concepts. The problem is, can the border withstand being a buzzword for theories of power, struggle, and connection? I propose that a single-image representing grand theoretical assertions is too general for the political and economic environment of the border. I propose that we specify our analytical tools for the border: that is, that we respect the concretely located nature of the Mexico-U.S. border. In so doing, I will propose a model combining the territorial nature of state activities and the partly deterritorialized activity of capital, both partaking of bureaucratic forms of action by contrast with border populace network action. If the border is to contribute to rethinking the social sciences, it will do so by careful exposition of state and capital actions and limitations, not through momentarily satisfying but paperthin imagery.

at the University of California, San Diego; the immigration project of the RAND Corporation and the Urban Institute; the Udall Center at the University of Arizona; the U.S.-Mexico Border Program at the University of Texas; and, as a radical counterpart, the Inter-Hemispheric Education Resource Center in Albuquerque.Undoubtedly many others can be listed.
The peculiar setting of the international border does raise so-called policy problems.But does the policy approach adequately get at generative processes?For a process or consequence to be identified as policy it must be posed as an issue by either a state agent or a similar polity-level actor.Furthermore, not just any state-organized action comes into question, only those that state-level actors perceive as problems.Inevitably, then, the policy-level approach rarely reflects on its own political motivation.It can discover much, but it is inevitably limited.For example, the quick and easy "voluntary departure" of apprehended undocumented immigrants from the U.S. to Mexico makes it possible for would-be immigrants to attempt entry repeatedly until they finally reach the U.S. At a collective level, voluntary departure makes possible continuous labor migration from Mexico.This has never been raised as a political issue (J.Heyman 1994).Instead the Mexico-to-U.S. immigration policy literature focuses on whether or not it is desirable to limit immigration by controlling access to employment (a policy of great importance) and by physical law enforcement at the border (an almost completely irrelevant policy as long as voluntary departure is in force.)Thepolicy approach also tends to be inhumane, which is both an ethical pitfall and a theoretical flaw.I will argue below that state-populace interactions are important to the border.If one accepts the state point of view, then tinkering with the state patterns of action (policy recommendations) appears more efficacious than understanding the emergent phenomena arising from state-populace interaction.The latter are simply too complex and too uncontrolled.
There are, of course, works in the policy vein that transcend these criticisms (e.g., the awareness in the immigration policy field that states rarely halt "chain" migration [W.Cornelius 1981; M. Piore 1979]).There is nothing wrong, per se, to begin with policy debates, for state action in many ways "makes" the border.If the mark of policy studies, however, is non-self-awareness of their origins in polity or state processes, then a more adequate approach must transcend these origins and penetrate from policy to polity and power.

The Border As Image
One result of making border processes into policies is media publicity about the border.Those who sympathize with the underdog and favor dramatic alternatives to U.S. policies gravitate toward the border as an image that inverts the U.S. "state idea" (P.Abrams 1988).The border image suggests conflict, change, and interest in relatively poor and powerless Mexicans.Academics then blur the inversion of policy with the inversion of a preceding generation's theories.
The problem of this image of the border is not that science must be separated from politics; we are engaged, in part, in the study of politics.It is that when the border is condensed to an image, and when this image symbolizes wide-ranging political or theoretical stances, understanding of the border becomes reductive and delocalized. 1Roger Rouse (1991; also see 1992), for example, opens his article with three images, the longest and most sustained of which is the U.S. southern boundary fence and the corresponding words "the border."In fact, Rouse addresses long-term cyclical migration between interior western Mexico (Michoacan) and the San Francisco bay region, and he argues that migrants live within two different class relations in each place, actively deploying rural Mexican class-culture to limit their inclusion in U.S. working class discipline.An interesting point, but one lacking analytical reference to Mexico-U.S. boundary processes.Because the people happen to cross the boundary, Rouse (1991:15) uses the border as a metonym for the juxtaposition of two experiences in one life.
Rouse engages tough questions of delocalizing communities in international migration with this apparently profound image, yet he does not search for strong analytical links to state-border processes such as smuggling and immigration law enforcement.(To be fair, Rouse, in his 1992 piece touches on the extended presence of the Mexico-U.S. border in northern California.He describes how undocumented immigrants fear local police arrests that might result in being turned over to the Immigration and Naturalization Service; he proposes that his teaches lessons of class discipline in the United States capitalism.)The simple use of the border as an image in conjunction with seemingly relevant, but unlinked material, is easiest to criticize.First we must ask: is the border image decoration?Second we must ask: when we rely on reader reception of Mexico-U.S. border images (e.g., ironies that work because they invert policy) do we not presume that readers know what the border implies; do we not fail to reflect on and inquire into the construction of the known border and the propagation of its meanings?
One might then reasonably reply that the border image is important because it suggests new theoretical perspectives in anthropology that reject the older focus on isolated and ahistorical culture.Rosaldo (1989) argues against "monumentalism," the idea that each "real" culture has a core that is isolated from other cultures.Rather, he proposes all cultures are multiple, incomplete, and contradictory.One culture does not correspond to one society; rather, society is a web of relations and so, likewise, are cultures.He likewise argues that cultures are not timeless or in the midst of loss, but rather constantly in flux. 2 Border zones occur everywhere because cultures are not homogeneous.People who actually live in border or transition zones are not culturally lost; indeed, border 1.I wish to acknowledge an analysis parallel to mine, contained in Victor Ortiz's abstract for the 1993 American Anthropological Association meetings (the paper, unfortunately, was not presented): The US/Mexican border has attracted widespread scholarly and artistic notice in the last decade, as specific economic and demographic changes (e.g., the maquiladora industry, the North American Free Trade Agreement) and official and unofficial flows of people, goods, drugs, and capital have drawn attention to the border area.Because of the perplexing nature of these dramatic interchanges and flows, the region has become a fashionable artistic/allegoric motif in certain US middle-class intellectual circles; specifically, those that favor postmodern stances.My presentation explores this fashionable attention by highlighting the sordid everyday dimensions of life in the region, which hyperdramatic depictions of the border area often overshadow.In these recent artistic and academic representations, particular aspects of the region are selectively emphasized while less spectacular but more pervasive conditions of ethnic animosity and subordination are overlooked.The aim of the paper is to underline the consequences of the lack of attention to incidents that bespeak a situation of low intensity conflict in the border area.I argue that the existence and detrimental impact of the conditions of violence in the border area have been selectively misrepresented, distorting analysis of a highly complex, volatile setting [V.Ortiz 1993].
peoples have unusually complicated and interesting cultures.Everyone has a culture, including the powerful (who Rosaldo [1989:198-99] perceptively notes, appear cultureless because they have citizenship in a nation-state instead).Correspondingly, the power system includes the apparently "cultural" peoples who are also less powerful.
I accept these propositions as part of a useful rethinking of cultural and social theory in anthropology.But if these propositions are widely or universally applicable, then we risk thinking that these observations suffice to analyze the Mexico-U.S. border.Because the images associated with a theoretical stance are drawn from a real, historical border, then that theoretical stance appears unproblematically to encompass that concrete situation without the hard labor of delineating linking propositions.We need more detailed analysis of a relational culture and power approach to the Mexico-u.S. border, e.g.how exactly does dual but unequal state power operate there, and how do cultural relations develop historically in this dual state power zone?
I take a regional particularist stance, then to question the new anthropological imagery of the border; I ask not only what the border image does for anthropology, but what analysis does for understanding and changing this border.A regional particularist approach plays several roles in critical anthropology.First, we should be skeptical about the rhetoric of deterritorialization in an era when the capitalist and state elites of the U.S. and Mexico see, through NAFTA, to deterritorialize in practice and in ideology, the peoples and resources of the two republics.Localism is, in such circumstances, a critical stance.Second, we need to locate some of the bitter realities of border life in the traceable actions and failures of powerholders--to point the finger--rather than simply use the life of the border as intellectual fodder.Finally, I note that the weight of the border image is on the cultural, the notion of two sides, two meanings, facing each other, rather than on the power, the idea of a forcible mode of territorial control.Without denying either aspect, I worry about emphasizing the former over the latter.
In the next several pages, I criticize from a regional particularist standpoint border image anthropology precisely to point out a series of requirements for stronger analytical models specifically about international boundary contexts and processes.I cover three clusters of claims about the border and new theory: one having to do with the nature of societies and other having to do with the concept of difference.
Akhil Gupta and James Ferguson (1992) follow Rosaldo in criticizing the single location, single-culture model prevalent in certain American and British anthropological traditions.They offer three principle contentions, two of which I list here (the third having to do with difference and identity, joins issues I address below).First, instead of being a series of bounded societies-with-cultures, the world is, and has long been, a series of spaces that are hierarchically interconnected (what old Leon Trotsky called combined and uneven development).The border is, of course, a salient image of combined and unequal development between Mexico and the United States.What, however, is the role of an international boundary, its multitudinous flows, and struggles over flows, in making spatial hierarchy?What is the role of people interacting with two state apparatuses in producing this global linkage on a daily basis?2. These arguments sound more novel to those awakening from the tradition/modernization and romanticization of exotic cultures mode of American anthropology than to those grounded in the work of Eric Wolf (1957).
Second for Gupta and Ferguson (and R. Rouse 1991 as well), flux and interconnection are, somewhat confusingly, very recent phenomena, products of postmodern capitalism.We are to suppose, that capital has become more global and flexible, both in its commodities and in its ideologies.The border, with its maquiladoras, provides a facile image of global capitalism.The idea that global capitalism at the Mexico-U.S. border (or anywhere else) is postmodern is historically untenable.The flexible and low-wage manufacturing at the Mexican border after 1965 stems from a long-term movement in the U.S. manufacturing (starting in the late 1880s) from New England to the Peidmont and Appalachia south to the Mexican border, as well as from the urban Midwest to small city Midwest to the border, in search of saturated and inexperience labor pools (NACLA 1975(NACLA , 1977)); J. Gaventa 1988).Likewise, labor and political migration from Mexico to the U.S., including the exchange of ideas and subjectivities (senses of identity), has occurred at least since the 1880s, if not before (L.Cardoso 1980; M. Garcia 1981; T. Sheridan 1986).Why capital uses concentrated bureaucratic edges as legal loci for transforming capital across nations and for realizing value in return is a truly hard and promising historical problem.We must consider it as building up gradually through a period we can identify as the modern or capitalist epoch.
Post-modern language likewise mystifies the border: The borderlands are just such a place of incommensurable contradictions.The term does not indicate a fixed topographical site between two other fixed locales (nations, societies, cultures) but an interstitial zone of displacement and deterritorialization that shapes the identity of the hybridized subject.Rather than dismissing them as insignificant, as marginal zones, thin slivers of land between stable places, we want to contend that the notion of borderlands is more adequate conceptualization of the "normal" locale of the postmodern subject [A.Gupta and J. Ferguson 1992:18].
First, I am worried that rhetoric replaces concerned knowledge.the Mexico-U.S. borderlands, at least are not a "place of incommensurable contradictions."The contradictions of interest and the struggles of force and influence are quite commensurable--such as the numbers of INS arrests, rate of usage of smugglers (e.g.L. Chávez, E. Flores, and M. López-Garza 1990), and numbers of deaths of persons trying to enter the U.S. illegally.Second, a facile idea--at the border, two sides equal one hybrid -replaces analysis.By no means do we have sensitive enough ethnography or testimony at the border to declare that it is experienced through a hybrid subjectivity or identity; the limited evidence we do have (Jorge Bustamante's [1985] study of Mexican national identity and Americanisms in Spanish, also see O. Martinez 1988:121) indicates that the Mexican borderlander's subjectivity remains strongly Mexican.The experience of U.S. Mexicans may indeed be more ambivalent, and this demonstrates the most important point: we need historical analyses rather than deduction from dualism spun out of the word and image "border."Finally, I am concerned that the useful ideas here--that borderlands are not marginal zones between solid sociocultural cores but central historical zones in their own right--make the rest of the rhetoric appear more plausible.
The border image in anthropology also typifies the theoretical concept of difference.Difference, a term originating in French poststructuralism, basically conveys the following ideas: potentially equal persons are divided, labeled (that is, emphasizing languageexpressed conceptual divisions), and ranked in unequal relations of power and authority with an additional emphasis on the internalization of these linguistic contrasts into knowledge and self-identity.The border image conveys a series of linguistic contrasts: Mexico/U.S., illegal/legal, poor/rich, etc. Border images of flows and juxtapositions, likewise remind scholars of the linguistic play and maneuver of contrasts.The contrasting and juxtaposed images do convey some visual sense of border, but image naming or image writing does not explain the making of the contrasts themselves.The border literature on difference emphasizes either divisions experienced with the self, or linguistic determination of unequal social segments.Thus, as I look first at self-difference and then at other-difference, I will point out that the border-as-image approach raises important questions about power and identity that the approach cannot itself answer.
The word "border" represents ambivalence during self-contemplation in Ruth Behar's book Translated Woman: Crossing the Border with Esperanza's Story (1993).Behar grapples with the recording, editing, and production of books involving persons of highly unequal class and race relations.Behar (1993: 227-229, 241) represents this original sin of academic production through a metonym of crossing the border.She recalls her relatively facile ability to enter through Mexico's northern border, and she imagines the difficulty and exploitation involved were her friend Esperanza to cross into the U.S. as an undocumented domestic.Behar also relies quite heavily on the term border to express the multiple and conflicting authorized identities inside of Behar herself, such as Latina by contrast to Anglo.
Behar's discussion of ambivalence is illuminating, especially concerning her evolving relationship with Esperanza, her friend and informant.Furthermore, it is true that Behar has literally crossed the border between the United States and Mexico and brought back a book.However, the word "border" does not add to the analysis of either academic production (one may well grapple with deep inequalities between academics and informants inside the U.S.) or difference inside the self.
We may understand the self in any locality, as a mazeway assembled from disparate and unequal social and cultural relationships (A. Wallace 1961).This view of the self is, in fact, logically consistent with the abandonment of one society/one culture theory within which the self was an authentic or a marginalized copy of a single template.Given this new understanding of self and society, we expect that ambivalent identity is a regular experience in most spatial and historical settings (H.Rebel 1989).Therefore, as an image of complexity and relationship inside the self, the border is either too specific or it is without content.I am therefore skeptical about Behar's assertion that from their position straddling selfhood and otherhood, Spanish and English, Mexican identity and agringado identity, power and resistance, Chicano and Chicana writers have so radically shifted the terms of cultural analysis that it now seems impossible to imagine doing any kind of ethnography without a concept of the borderlands or of border crossings (1993:15; emphasis added).
When we examine this statement, we see that it dislocates Mexicano experience in the U.S.-side borderlands into the abstracted contemplation of ethnography.I wish to reemphasize the local experience.To make my point clear: we start with the perspective that the self is a partly contradictory mazeway inside most cultural situations.Then we specify personal mazeways and regional contexts that give rise to them.The Mexico-U.S. border cannot serve as a general metonym for mazeways.It provides one location for understanding arrangements of psychological complexity, including those dualities expressed by Chicana and Chicano writers.For example, we might consider that the Mexico-U.S. border zone selects certain economic and political relationships (e.g., those exemplified in school and labor systems).That context delineates a particular field in one or both nations for households and childhood, and consequently for the formation of interpersonal relationships and emotional constructs.
The border, one can argue, also authorizes the difference between U.S. citizens and Mexican others.When Mexicans migrate into the bounded U.S. space, a linguistic label ("immigrant" or "illegal alien") and therefore a new identity is imposed, and perhaps resisted.This argument by necessity transcends its origin in the border as image, however.It demands attention to political economy.Gupta and Ferguson indeed shift gears from the "politics of representation" to the "politics of immigration." In this perspective, power does not enter the anthropological picture only at the moment of representation, for the cultural distinctiveness that the anthropologist attempts to represent has always already been produced within a field of power relations.There is thus a politics of otherness that is not reducible to a politics of representation.Textual strategies can call attention to the politics of representation, but the issue of otherness itself is not really addressed by the devices of polyphonic textual construction or collaboration with informant-writers, as writers like Clifford and Crapanzano sometimes seem to suggest.
In addition to (not instead of!) textual experimentation, then, there is a need to address the issue of `the west' and its `others' in a way that acknowledges the extra-textual roots of the problem.For example, the area of immigration and immigration law is one practical area where the politics of space and the politics of otherness link up very directly.Indeed, if the separateness of separate places is not a natural given but an anthropological problem, it is remarkable how little anthropologists have had to say about the contemporary political issues connected with immigration in the United States [1992:17].
Gupta and Ferguson go on to repeat this point, and gloss several possible themes: the enforced spatial segmentation of poverty; the political and organizing rights of immigrant workers; and the "appropriation of anthropological concepts of `culture' and `difference' into the repressive ideological apparatus of immigration law" (1992: 17).The problem with this impressive rhetoric about power is that it provides no path to get from here to there other than a comforting sense that these topics are relevant; Gupta and Ferguson offer no concrete model of how and why the recent capitalist nation-state uses borderline enforcement to differentiate movements of laboring peoples.Michael Kearney (1991) in fact offers such a model in conjunction with an authorial strategy that relies on border imagery.He argues that advanced capitalist nation-states enforce difference in two directions.They enforce the difference between core states and colonial or peripheral spaces, and they differentiate persons inside the state into unequal segments while unifying them as members of an imagined nation.Delineation and enforcement of an international boundary is central to both acts of difference-making, because it unifies nation-members against nonnationals, while the U.S. state (in Kearney's example) permits de facto labor immigration from Mexico as a divisive act for the internal stratification of the U.S. Having developed a model of state action, then, Kearney also delineates an argument about the character of social relations and culture in the immigrant-receiving region.In the border area, immigrant peoples are both boundary-defined foreigners and tacit, though bottom of the class structure, insiders.An encompassing and activating sense of ethnicity arises from narrower migrant and hometown networks, at least for Mixtec migrants to the U.S.There is much important in Kearney's arguments, as well as some ideas about the U.S. state that are overly mechanical and omnipotent (see below, and for more detail, see Heyman 1995b).The point here is simply that any productive effort to study and criticize the making of difference requires a model of state and populace actions specific to historically unequal boundaries; it is not inherent in the word/image "border" itself.
I have criticized the unclear mixing of the Mexico-U.S. border and a very abstract language of "border."But what of the use of anthropology of borders to describe social and cultural edges, real ones, but ones that diverge from the archetypical state-state boundary with large bureaucratic presence (border in my terms, and I hold no brief for centering it this way) I refer, of course, to Fredrik Barth's 1969 work on ethnic groups seen as boundaries.Part of the problem is that we use natural language to express analysis.The English natural language term "border" gives the illusion of parallelism when the interest lies in what aspects are similar and what diverge (Spanish is worse, for frontera means both frontier and bureaucratic border).Thus, the intriguing side of this problem is to ask if the state-state border shares some processes with other sociocultural boundaries in a manner that transcends similarity of images.
To explore this, I turn to an article by Robert Alvarez and George Collier (1994).Under the rubric of crossing borders Alvarez and Collier compare two sets of truckers: northern Mexicans who haul fruit into the U.S. across a state-state border, and indigenous Zinacantecos of Chiapas, Mexico who haul goods (such as cut flowers) across ethnoracial boundaries in southern Mexico.Alvarez and Collier show that when each group undertakes risks by penetrating political-economic boundaries, they use ethnicity, and within this their specific cultural idioms of interpersonal trust, to reinforce networks.The extended image of borders seems apt here, though I criticize it above.What do the two types of borders have in common?In each case, a relatively subordinate people, with weaker formal bureaucratic masters, succeeds by cultural creativity in working its way through an institutional structure (a web of political and organizational influence) controlled by an ethnically (and in one case nationally) different, superordinate group.This suggests that state-state borders, because the institutional structures tend to be rigid and obvious, highlight processes that occur in other contexts.Relevant contexts are not those where there is a sociocultural boundary but where that boundary involves differential access to formal channels of power.More generally, I am suggesting that anthropology will learn most when we extend fairly specific analytical insights from one "border" denotation to another.This requires that we first define a range of (natural language) border situations.
The regional particularist critique not only helps by asking for clarity and by pointing out local sets of details and dates.It also helps by raising intriguing analytical problems that do not permit easy closure, especially easy solution by means of metonymic language.The regional particulars of the Mexico-U.S. border require that we specify better the means of unequal power linkages, especially between immigrants, citizens, and states; that we specify better the role that crossing zones play in capitalist value creation; that we link better the structural portrayals of state and capital with complexity flowing and maneuvering peoples (that is to say, that we not rest with asserting that power goes along with border); and that we specify better the relationship of felt difference inside people's heads with political economy taken down to the level of personal and household life.Recognizing the immense challenge of such questions, I nevertheless try to encompass them within an analytical model of the Mexico-U.S. border.

An Analytical Model of the Mexico-U.S. Border
My starting observation about studying the border is not cultural juxtaposition, it is the massive state apparatuses of the boundary, and especially the overt and hidden force of the U.S. state.My general approach here is that the Mexico-U.S. borderlands are a specific local ecology for human action (J.Greenberg and T. Park 1994).The local ecology is largely politically organized. 3It is politically organized in the direct sense that major resources come from government agencies (and from the defiance of them), and in the indirect sense that private border economies flourish (in most locations) because of political rules and opportunities rather than advantages of physical geography.These references to politics are not to sociopolitical organization generally; they point us to located phenomena that only take place near or through boundaries.
I begin with a series of abstract assertions, and then observe how they are enacted at the border.States are aggregations of rules for social and economic action and the bureaucratic organizations required to implement these rules; for short, states are the rules of the game (M.Mann 1993:44-91).When I discuss rules, I generally mean two qualities: delineations of favored claims to resources; and the capacity to routinize behavior.Rules of the game through states are preeminently territorial. 4Capital (which I do not need at this point to define further) is relatively nonterritorial in the sense that value can be, and often is, realized by transactions across territorialized ecologies.Capital, however, needs territorial rules of the game: distinct capitalists seek specific rules, and more importantly, capitalists envisioning future accumulation desire predictability and manipulability provided by any (or at least many) sets of rules over territories. 5Likewise, states need capitalist activity, for states, even powerful ones, are remarkably constrained by their need to derive incomes from the external economy (J.O'Connor 1973).States and capital are both marked by high degrees of single-stranded specialization (what I call thought-work [J.Heyman 1995a]) that convey overwhelming technical capacity at any one moment but result in organized stupidity and inflexibility over the long run.
Finally I consider populaces as loose aggregations of persons and households, linked in flexible manners by network exchanges and ideals.Populaces are relatively weaker than states and capitalists because people have to live from day to day, they have to survive 3. The Mexico-U.S. border environment does have some socioecological characteristics related to the control of irrigation water (R. Fernández 1989:45-66).
4. When I assert that states are aggregates of rules of the game, I do not assert that states have a single preeminent function even under capitalism.Numerous arenas of rules arise in the history of specific capitalist polities.Given this somewhat consistent historical context, the rules are not completely particularistic, so that we can analyze rule clusters at state boundaries (e.g., bureaucratic education, citizenship rights) without a reified and mechanistic view of "an economy" controlling "a state" (P.Abrams 1988; M. Mann 1993).Wolf (1982:419) remarked that: I know of no fully satisfactory argument that can explain why a particular form of 'thick' state is essential or even functional for capitalist accumulation during strategic phases of its growth, or why the historical articulation of classes requires the development of just this kind of political-economic apparatus.I would not claim this to be fully satisfactory, but I do hold that thick states are sets of knowable rules within territories very much desired by capitalists.and this means that people capitulate to states and capitalists on a daily basis.Populaces must locate themselves, sometimes double and triple locate themselves, into niches emerging from the enormous and complicated activities of states and capitalists.This political ecological framework means that I see localized communities and cultural activities as emergent and changeable over time; cultural crossing and delocation of community is no novelty in this framework, but likewise the settling out of stable local groups in boundary areas in also expected.I tend, in this paper, to delineate the niches created by border processes of states and capitalists, and I will admit to painting a firstorder portrait of populaces as recipients of action.Yet I do posit, based on the different modes of flexibility and action, an ironic reversal in which populaces maneuver within and at cross purposes to state and capital rules.I especially envision this process of initiative and defiance occurring when populaces are defined into one niche (for definition is a key state process) but they must and do occupy an alternative, defiant niche.

Eric
International boundaries are edges of the rules of the game in two important ways.First, they are places where state bureaucracies regulate who and what can enter the territorialized rules: that is, the entrance into the game.An obvious example is export/ import regulations and duties.Enforcement is particularly important in this action, because it is at the border where the state first imposes conformity with rules that will hold in interior territories.Second, boundaries are places where territorial outsiders first encounter (and can access) rules of the game that have been routinized to the stateterritory as a whole.An example is access at the Mexican border to social security and other resource redistributions within the U.S.These institutions are not peculiar to the border but they take a particular cast there.For both instances, an effective boundary ecology only occurs when the state routinely influences nonstate behavior; influence may consist of either accepting state rules or explicitly avoiding them (P.Sahlins 1989).
Areas of disparity and of transition between rules offer unusual advantages for the creation of capitalist value.I specifically have in mind the Mexican side Border Industrialization Program (BIP or maquiladora program).The maquiladora program began in 1965 when Mexico's government formulated special sets of rules for foreign (read, U.S.) investors in assembly plants on Mexico's northern border.This was a successful effort to increase Mexico's state control over poorly integrated, rapidly growing border cities (M.P. Fernández-Kelly 1983:26-27).Yet at the same time, the success of the maquiladora program increased the fiscal dependence of the Mexican state on U.S. capital; it is a model that the Mexican government hopes NAFTA will reproduce.More important in value creation than the specific rules of the BIP (and of the U.S. tariff laws that facilitate it) is the fundamental disparity between Mexico and the U.S. in the rules about working people's claims on the social product and governmental redistribution of it.Simply put, the maquiladora program works because of the vast inequality between the total wage and benefits bill of U.S. workers versus Mexican workers (Fernández-Kelly 1983:27-28).One line of thought, then, about the extraordinary creation of value at borders is that it emerges from or can be perpetuated through the "radical disjuncture between power and politics" (J Heyman 1991:203), the gap between arenas where longterm capital investments are made and arenas where the distribution of private and socialized products can be struggled over.It is true that value creation of this nature does not take place exclusively at international boundaries; there are export processing zones equivalent to the Mexican border throughout the world.In each case, however, the dependent state does zone for special rules on behalf of capital, and thus we might, in my framework, think of all such locations as created boundary phenomena. 6 I prefer to apply these concepts to a historical border rather than an abstract model of all borders.The modern Mexico-U.S. border was formed during 1845-1853 by conquest, along with a small purchase, of Mexican territory by the U.S. Anglo-Americans destroyed the rules and power of Mexicano U.S. citizens in the new territories.The Anglo-American state at first instituted routines only in larger settlements and commercial or mine boom areas; they more or less ignored the boundary itself.The U.S. borderlands interior gradually filled in with the agencies of routine internal governance, such as Englishlanguage schools and police.The boundary emerged as a distinct zone when U.S. capital leapt into northern Mexico during the last two decades of the nineteenth century.The U.S. state then placed state agencies, such as customs, at the border line in order to control the flows of goods consequent to the flows of capital.The movement of capital into northern Mexico also reshaped northern Mexico from a frontier (a state-facing-nonstate process) into a borderlands, an area (or processual geography, better said) linked to another state (the U.S.) by capital and the cross-boundary movement of people.As the result of complicated political events (e.g., the Mexican revolution), Mexico gradually increased its routine state capacity in most of its northern borderlands, as well as the rudiments of state rules agencies at the international boundary.Unequal state capacities between Mexico and the U.S., however, will be a major feature of my arguments about the political ecology of the border, for this asymmetry shapes the movement and settlement of people.The net result of these processes is routine state institutionalization in the U.S. borderlands, relatively strong boundary-imposing conduct by both states, and U.S.-side domination of capitalist production in both the U.S. and Mexican borderlands (R. Fernández 1977).The Mexico-U.S. border ecology, as a result of these historical flows, supports several niches, and consequently settled or intermittent populations. 7Capital moved from the eastern U.S. west and south across the U.S.-side borderlands.Along with this capital has come a vast but invisible Anglo-American migration into the U.S. west (16.2 million whites and 1.2 million blacks from the east to California between 1930 and 1979 [The Times Atlas of World History 1989:289]).Not all these people were capitalists, but the capital was Anglo-American in terms of its organizing rules and culture (e.g., the business triumph of English and the dollar, which was by no means assured in the early southwest [J. Park 1961]).Therefore, Anglo-American culture became the regional cultural capital.
6. Capital needs brokers to handle value-creating and transferring transactions across boundaries; brokers understand and manipulate the complicated transitions from one rule set to another.Unfortunately, we lack an adequate study of border brokers as a sociocultural class, despite their importance as a (maybe the) dominant local elite in border communities.The Mexican and U.S. side border cities are centered on ports of entry and exit.In the borderlands before the BIP, brokers in the U.S. handled northern Mexican goods such as fruits, vegetables, cattle, and ores either produced or funded and purchased by U.S. capitalists.(Elite families who do not appear on the surface to be brokers, such as cattle ranchers, also play this role through their control of cattle marketing associations.)Brokers in Mexico handled the importation of consumer and capital goods into the Mexican interior.A more recent role for border brokers has been handling the legal and physical establishment of maquiladoras in Mexico for inexperienced U.S. corporations; these brokers will likely expand their work with the advent of NAFTA.
7. Another work delineating settings and borderlands "types" is O. Martinez (1994).U.S. citizenship is by law independent of national origin, but in the west, citizenship became racialized (M.Barrera 1979).U.S. Mexicans have been defined--according to the effective rules of the game--as outsiders, though capable (as in the Chicano civil rights struggle) of reclaiming citizenship rights.However, the fact that persons of Mexican origin residing in either nation were expropriated from the means of power ironically brought more people across the boundary, for these people had become relatively inexpensive and coercible labor for expansive capital.If Anglo-Americans are the invisible migrants, so Mexicanos are the visible migrants.As excluded rather than included migrants, Mexicanos learn much more about the double-sided niches and cultural repertoire of the borderlands; Anglo-Americans are, with individual exceptions, remarkably ignorant of the border.
As we have seen, U.S. capital did not halt at the international boundary, and boundary agencies were brought to manage its transactions.Fairly large cities grew on Mexico's northern border; they were founded in the 1900s and thrived in the 1920s, but the Mexican border cities grew rapidly after 1940.These cities are supported directly and indirectly by state activities at the international boundary itself, so the Mexican-side border city (and the smaller U.S.-side border city) represents a sui generis set of niches and sociocultural practices.Above all, Mexican border cities grew because of changes in the U.S. rules of the game in conjunction with Mexican population growth and disequilibrating development.After 1940, migration from Mexico to the U.S. increased, but legal opportunities to settle in the U.S. decreased (they did not disappear).The border city grew in the period after 1940 because it contained numbers of would-be immigrants, such as daily border commuters, and their families (J.Heyman 1991:203-04).The border was, and is, a sticking point when capital is more mobile than residential rights. 8U.S. capital directly or indirectly transformed, and continues to reorganize Mexico south of the U.S. boundary.We may speak of a Mexican borderlands as a zone mixing Mexican rules of the game with U.S. capital dominance.Many scholars (including Behar and Rouse) recognize that the Mexican borderlands cover the vast majority of the nation.To take but one example, Alejandro Portes and John Walton (1981)  Recently urban working class areas, including Mexico City, have joined the migrant stream.Southern, relatively "Indian" Mexico, such as Oaxaca, now also exports migrants (W.Cornelius 1989;M. Garcia y Griego 1990;D. Massey, R. Alarcon, J. Durand, and H. Gonzalez 1987;and C. Nagengast and M. Kearney 1990).Each particular chain migration network "learns" about the several loci outlined herein.
Two topics, immigration smuggling and senses of self identity, demonstrate that this reformulation addresses issues raised within the border image literature.The Mexico-U.S. border is, of course, a classic setting for smuggling.Guns are restricted in Mexico, and thus they are smuggled from the U.S.; there is a high demand for psychoactive drugs in the U.S. (and, before that, for prohibited alcohol) and thus illegal drugs are smuggled from Mexico.The list certainly could be lengthened.Smuggling (of people or of goods) 8.One might be tempted to attribute Mexican-side border city growth to the maquiladora program.The maquiladoras, however, started in 1965 and became a very large source of employment in the 1980s.Their expansion was, in fact, premised on the preexisting labor pool on the border.requires that the activity be legally prohibited; the defiance of this prohibition adds value above and beyond normal, unbounded commerce.There are two sets of fundamental questions about smuggling.One is why a given historical state makes certain goods and activities illegal, that is, elevates them from unregulated commerce to an activity subject to explicit rule-based control and prohibition.This is less a border than a broad societal topic, and I shall not pursue it further.The other is what the characteristic modes of state and antistate actions are once the conditions for smuggling do occur, and what the political ecological consequences of these sets of actions are.
Smugglers characteristically operate on the basis of an interpersonal multistranded network, containing extreme inequalities of wealth and power, but nevertheless still flexible and informally organized.The state responds with bureaucracies: single-stranded hierarchical organizations for the transmission and implementation of control over persons (J.Heyman 1995a).The networked and flexible organization of smugglers possesses significant advantages over bureaucratic action, especially in a boundary situation where smugglers have a safe zone inside the U.S. or Mexico; in fact, I know of no situation where state action has ever defeated extensive smuggling without the help of a broader societal change that reduces demand for the smuggled good.The low rate of success gives rise, however, at least in the short term, to increased size of the bureaucracy with resources flooding in accompanying a succession of policy studies and "new" initiatives.
Border control over human migration is not inherent.During the initial expansion of capital, states permitted the active recruitment of new migrants, in keeping with the deterritorializing character of capital.The U.S. clearly did this for Mexican workers in its western territories (L.Cardoso 1980).Advanced capitalist states, however, have shifted toward a more contradictory migration system during this century.These states establish a regime of rules to put a ceiling on the number and types of persons who move into their territories (A. Zolberg 1983;1992), while simultaneously accommodating capitalist demands for labor migrants, whether they enter in the gambit of rules or in an improvised, extralegal manner. 9 On the U.S. border with Mexico, this has taken the form of symbolic policing close to the boundary itself, with no or ineffective penalties on capitalist employers (K.Calavita 1982;1990).Political symbols--real arrests--cater to a media-image that the government is trying to defend the border against hordes of immigrants.. Image production responds to the narrowed perspective of Anglo-Americans thoroughly imbued with ignorance of border processes by their near-monopoly of U.S. Cultural capital and citizenship tokens (J. 9.The reason for this shift remains unclear.Aristide Zolberg (1983) argues that it occurs because the world has maldistributed resources and relative overpopulation in poor nations; rich nations therefore try to restrict the balancing of resources that would occur with equilibrating population shifts.I have argued (J.Heyman 1994, drawing on K. Calavita 1984;W. Cornelius 1982;R. Thomas 1985) a more internal, political history for immigration restriction.In a nutshell, citizenship claims on the state are a regular mode of political action in capitalist societies; as capitalist polities become more encompassing and powerful, citizenship politics does likewise.Citizenship politics favors restrictive, though perhaps not necessarily closed, immigration policies--at least, policies in the overt sense of official law and publicized police action, though capital bends actual state action to the extent that some labor migration continues through legal loopholes or ineffective patterns of policing.Heyman 1994).This bizarre combination of ineffectuality and force at the border determines the niches that undocumented immigrants occupy, in particular the pattern of smuggling.Kearney (1991), as above, is correct in envisioning two processes: the de facto toleration of labor immigration from Mexico to the U.S., and the stratifying discourse among U.S. citizens themselves of "illegal aliens" as usable but rights-less outsiders.Discursive difference spoken largely by Anglos to Anglos, however, does not automatically shape immigrants' behavior given their great confidence in their covert social networks and lack of fear of the INS (H.Delgado 1993).The application of power requires a specifiable mechanism.Undocumented workers are victims of the stateantistate struggle, and of the relative success of smugglers over the state; persons who are smuggled or who transport themselves northward become enmeshed in networks of extralegal conspiracy that feed into workplaces as a means of exploitation and control above and beyond any other weaknesses of new immigrants.The significance of this argument is that the mode of policing and of flexible smuggling counteraction, and not a linguistically delineated, abstractly powerful discourse of subordination, concretely causes undocumented residents and workers to be vulnerable.Discourse enters the system of subordination by affecting the political process among insulated citizens (J.Heyman 1994).
Smuggling is experienced at the border through a very different discourse.Immigration is a source of bitter low-level violence and defiance on the border.First, the INS is an armed police force.Most arrests are nonviolent, because Mexicans and INS officers both know that quick voluntary departure usually ensues; but obviously arrest itself is a nonvoluntary event involving potential unequal force, and there are circumstances when INS officers do abuse human rights (J.Heyman 1995a).Second, smuggling networks are highly coercive, again involving either potential or actual force; the undocumented immigrant is physically and psychologically dependent on the smuggler, whereas smuggling organizations internally run on veiled threats and fear as well as rewards.Finally, Mexicans resent the inequality between the U.S. state and themselves as past or present undocumented migrants.They engage in low-level politics in the classic Scottian manner (J. Scott 1985).In Mexico I witnessed rude skits of defiance by young men (prime migrant material) aimed at imaginary Border Patrol officers (J.Heyman 1991:18); acts of real, physical resistance include throwing stones at ("rocking") INS officers by the boundary fence.
As the above example demonstrates, situational politics is heightened, if not literally created, by the people's ability to act in two different modes in two nations: rude to the INS while in or near Mexico; complacent when caught by the INS and angling for voluntary departure.Mexicans, including border city residents, therefore undertake active responses to U.S. power as embodied in the INS; they both use it (e.g., to get visas) and resist it.Smuggling and antismuggling likewise creates a social-relational/cultural arena near the boundary which we might gloss (in my terms above) as a borderlands mazeway.People in nonborder cultural situations play amongst relationships of confidence and conspiracy versus distrust and information withholding, but the Mexico-U.S. border specifically offers complex and perhaps more vital arena for these forms of social action (C. Vélez-Ibáñez 1983).If the Mexico-U.S. boundary was once an arena of conquest and overt resistance (A. Paredes 1958:149-50), it is now a zone of routine struggle between prosaic state force and networked defiance. 10 The resources emanating from the network migration/state bureaucracy interaction in turn support significant mobile and settled populations at or near the border: legal immigrant communities inside the U.S. in laboring/receiving areas, often holding relatively advantaged roles in enterprises and acting as de facto labor brokers; border commuters between Mexico and the U.S. who bend and manipulate immigration rules in order to gain a modicum of control over their lives in two different nations/ecologies; and beachhead communities in Mexican border cities that shelter and protect undocumented immigrants either leaving Mexico or momentarily expelled from the U.S. by voluntary departure, as explained earlier in the paper. 11Smuggling networks, likewise, support a variety of border people on both sides, ranging from the smuggling family cliques to various legitimate workers and entrepreneurs whose businesses are buoyed by smuggling profits and payrolls.Finally, the INS and other state control bureaucracies--the police system, to be blunt--are likely to prove the largest and "best employers in the U.S. border towns.
These brief observations suggest that niches and modes of action add strength to the rather general term transnational applied to immigration and border processes.I do not aim to replace transnational with a single alternative theory or term; rather I seek a theory and a field learning strategy for stronger analytical specificity.If we think about niches, for example, we discover that we still lack a good description of the resources in and transfers between each niche in this overall process.What do the Mexican border city settlers get (morally and materially) for aiding voluntarily returned hometown mates?Does U.S. immigration restrictionism, with a strong border barrier, shape a particular pattern of transnational niches?Anthropologists have paid greater attention to how people experience transnationalism, but we remain largely at the level of amazement that people can bridge and relocalize identities in two nations (e.g. A. Grupta and J. Ferguson 1992).An honorable exception is Leo Chávez's (1992) poignant discussion of how undocumented settlers cope with their niche, forming imagined communities in the U.S. despite their rejection from legally imagined (and thus legally real) immigration rights.How do different crystallizations of the migrant experience--those who dislike their 10.Peter Sahlins (1989:134-39) argues that local populations in border areas of the two nations strengthen the governmental presence at the border by calling for state support and intervention; local populations, that is, make strong borders rather than opposing or avoiding them.This is, I think, an empirical question.I venture to guess that brokers often do clamor for more state regulation and bureaucratization at borders.On the other hand, I shall argue below that some border networks exist precisely to defy states.Perhaps it is better said that state rules become a resource for both conforming and resisting actions-the denser the rules, the more the resources for both activities--and that borders by their "edge-of-the-rules" nature are particularly rich environment.I also argue that whether local populations invite the state in or unsuccessfully resist its extension, the manner in which a state responds is consistent.It sets up a bureaucratic work organization and a bureaucracy-petitioning routine for the populace (see P. Sahlins 1989:279-80).Therefore the direction of history, at least in the last two centuries, is unilinear; it favors state organized routines and specifically antistate networks.11.Richard Mines (1981) presents an exemplary delineation of all these niches in the migrant stream; on Mexican border city support communities, see Carole Nagengast and Michael Kearney (1990:78-79).imagined U.S. and those who settle into it--motivate persons and households in the enactment of these multiple niche networks?What do people make of niche choices over the long-run, as they crystallize historical traditions of national and political feelings?I have put much emphasis on smuggling, and by implication, on those state agencies specifically meant to impose rules at entry into the country.However, important border niches occur as a by-product of the border as the ultimate edge of routine interior state institutions (P.Sahlins 1989).These are quite unequal between Mexico and the U.S.. English is the language of U.S. (and world) capitalism, and even were conditions in Mexican and U.S. schools otherwise equal (and they are not), for this simple reason some Mexican residents would seek to place their children in U.S.-side schools.Likewise, the U.S. and its constituent institutions fund a significant number of households with subsidies such as Social Security, pensions, Food Stamps, etc. Subsidies to households in Mexico are extremely sparse (J.Greenberg 1987).Households in Mexican border cities support themselves, in part, through clever ways of gaining access to U.S. resources (e.g., by transferring dual residential-rights children from Mexico to the U.S.); furthermore, households in the U.S. subsidize elderly and other households in Mexico in lieu of state subsidies.These are quintessentially boundary activities (except the interhousehold transfers).I feel certain that a full economic-anthropological study will reveal that U.S.institution to Mexican-household transfers undergird a significant proportion of the population of Mexican border cities, perhaps as much as the maquiladoras.These strategies are, likewise, an important arena of low-level politics in border communities, and arranging for them involves the full panoply of interpersonal/interhousehold alliance, and much conflict as well.
Luis Moll and James Greenberg (1990) and Carlos Vélez-Ibáñez (1994) argue that Mexicano households in the U.S.-side borderlands use multiple "funds of knowledge."These funds of knowledge clearly include knowledge about U.S. instituted rules and routines (both schooled knowledge and family-conveyed knowledge about how to maneuver around institutions).The schooled knowledge of state routines is, in the context of historical Anglo-American economic and cultural capital, more highly rewarded than family knowledge about border niches.Clearly, education seen as funds of knowledge applies to a wide variety of times and places, but I suggest that the boundary environment, with its multiple sets of knowledge that are relatively distinct and localized.
These observations suggest ways the political ecology perspective extends the inquiry termed difference beyond linguistic contrasts.Let me raise tentative arguments about three borderlands populations.First, many Anglo-Americans identify strongly with the border not only as a skin or edge of the state, but as a boundary that contains the stuff or self, threatened by penetration and damage (this is an issue suggested by Kearney's 1991 piece).What are the proximate causes of the state-self equation in citizenship anxiety of the 1980s-1990s?What is the interaction between instrumental greed and extrainstrumental fear?Does this fit my longer range historical hypothesis that Anglo culture capitalized (in the metaphorical sense) the Mexico-U.S. borderlands?
Second, Mexican borderlanders have relatively few identity worries about the U.S. (J.Bustamante 1985).For example, my study of imported U.S. goods shows that Mexican borderlanders are less concerned with their relations to the U.S. than with class relations and inequalities inside Mexico.They use U.S. goods to speak about Mexican issues (J.Heyman 1995b).To take another example, bitter and creative gender and class/race experiences, not U.S. domination of Mexico, are the emotional core of Esperanza's story in Translated Woman (R. Behar 1993).This dimension provides the stunning quality of Behar's book.We certainly need more realism in anthropology about the searing experience of inequality inside Mexico.We might return, then, to discuss Mexican visions of the U.S. not in a border centered manner but with attention to arguments over Mexico's national economic and ideological projects.
U.S. Mexicans occupy the most complicated place in the personal landscape of the borderlands.I have mentioned contradictions in the ecology of knowledge and the complicated U.S. Mexican roles in the niches of immigration, befitting diverse histories and multiple legal statuses among these people.More importantly, I have suggested that many U.S. Mexicans are most knowledgeable about the maneuvers that make up the border, but that this is a self-contradictory skill in a nation that utilizes Mexican capacities to navigate the border but gives them no value.What we are really only beginning to understand, in a fundamentally empathetic manner, is that emotional and cognitive complexity of the U.S. Mexican mazeway.Rosaldo's essay (1989:147-167) on changing Chicano narratives makes a first approximation of this subject, and I anticipate significant discussion in Vélez-Ibáñez' forthcoming book.I argue above, and I repeat with emphasis, that facile abstraction of a border condition does serious disservice to this U.S. Mexican arena of profound self-strength and self-doubt, and elides close examination of the "hidden injuries" of U.S. life on the exemplary model of Richard Sennett and Jonathan Cobb (1972).

Final Reflections
George Marcus writes in his inimitably pretentious style that: This second...position is no less radical than deconstruction, but it does recognize a space and function for such radical critique within the historic projects of scientific disciplines.Rather than being nihilistic, however, permanent metacritique of discourse and rhetoric confronts the mainstream's unmediated realism, taste for objectification, and belief in transparent language, in order to shock and argue with these styles.Playing the minor key of skepticism against the major key of so-called logocentrism in Western cultural cognition takes on a local discipline-specific purpose of generating self-critical consciousness.This position rationalizes a place for permanent, radical critique within disciplines, but with uncertain implications for modes of theory and practice.It in fact recognizes that the mainstream is glacial and may perhaps be untransformable (1994:41).

Marcus instead calls for experimentation with form and language:
The fourth position, experimentalism, is much more substantive and defensible.... The word `experiment' evokes either the central mode of the natural sciences or that of the avant-garde of historic aesthetic modernism in the West from the 1890s to the 1960s.These latter, through experiments with form, overturned developed, but naive, notions of realism in the arts.It is this latter sense of experiment I mean to evoke here (G.Marcus 1994:41).
I beg to disagree.When authors play with word-images of the Mexico-U.S. border, they reify and give causative force to language-based contrasts of the two sides of the border: poor/rich, Mexico/U.S., immigrant/citizen, etc. Perhaps, in fact, we most need the "minor key" as the majority of academic production is, and probably should be, careful objectification of events in the "major key." 12 Much of this paper's argument--that the Mexico-U.S. border requires specification of ecologies of state and capital--indeed plays the major key.But the minor key likewise rings out when we reflect on and reveal the hidden realms of public problems and when we critique the production of mystifying academic language.The Mexico-U.S. border is not a privileged place for critical social science; every located ecology deserves major and minor analysis.But the visibility of state and capital at the boundary does invite critical and reflective study.
demonstrate U.S.related unequal development has spurred the Mexican migration supply.Mexican migrants have departed the rural north, north central, and west Mexico for a century.