Artículos y ensayos Estudios Fronterizos, vol. 2, núm. 4, 2001, 49-64

Transnationalism and Borderlands: Concepts of Space on the US-Mexico Border and Beyond


Anne J. Goldberg*


* Department of Anthropology. Arizona State University. E-mail:



Transnationalism and Borderlands: Concepts of Space on the US-Mexico Border and Beyond. This paper problematizes the linkage between studies of international borders and studies of transnationalism. While obvious connections exist, the conceptualization of space by both actors and anthropologists creates a division in Nogales, Arizona, a town on the US-Mexico border and the site of an applied study of education. Implications of this theory for the stated research and for anthropology will be discussed.



Este documento es sobre la relación entre los estudios realizados acerca de las fronteras internacionales y aquéllos sobre el trasnacionalismo. Aunque hay similitudes obvias, la conceptualización de espacio tanto por parte de los participantes en estos estudios como por los antropólogos, crea una división en Nogales, Arizona, un poblado en la frontera entre los Estados Unidos y México, y el cual se eligió para aplicar un estudio sobre educación. Las implicaciones de esta teoría en la investigación existente y en la antropología, serán discutidas.



"A major finding to emerge [...] is that minority students do better in school when they feel strongly anchored in the identities of their families, communities, and peers and when they feel supported in pursuing a strategy of selective or additive acculturation" (Gibson, 1997:445-446).

Identity has become a central concept in current anthropological research. Recent social theory emphasizes the constructed, enacted nature of social structure (Giddens, 1984), and identity, as a concept, embodies this trend. Identity also makes the agent or actor the focus of study, thus, it has become a very attractive subject of study for researchers who align themselves with agent-based theory. Current research in education, as the quote from Gibson cited above suggests, has taken the idea of identity to heart and illustrates the complex issues embedded in minority education.

Nogales, Arizona, a small city that abuts (or blurs into) its larger sister city, Nogales, Sonora, on the US-Mexico border, provides an ideal site to examine identity and its effects on education. Using one middle school, Buena Vista, as a focal point, I consider how students, teachers, and parents express identity, and create and reproduce societal divisions in schools. However, this study extends beyond the school into the larger community on both sides of the international boundary.

I have looked for identity as it is enacted and conceived of on a border, and, more specifically, in Nogales. What does identity mean here, and how can we use that information to help students succeed in school? I propose that language use, length of residence on the border, and how people conceptualize place serve as indicators of ethnic identification in Nogales. I then apply this information to education programs.

In many ways, Nogales and the US-Mexico border provide a unique site for testing theory on how ethnicity and class identities affect performance in the classroom. The border itself can act as a metaphor for ethnicity: permeable, dependent on a contrasting group for definition, and changeable. Continuous migration and settlement produce and define fluid and mutable ethnic groups, providing a laboratory for examining the connections between ethnicity, class, and power. For example, in a place where almost everyone is "Mexican-American", ethnic boundaries may form according to the length of time spent in the US or in the border city of Nogales. Generational differences, particularly of families in the area, may prove to be a critical element in understanding school success.1

In this paper I address one of several problems in anthropology raised by the focus on identity. Margaret A. Gibson and others are correct in calling attention to identity in education, however, this attention is merely a beginning, since without knowing what identity is, we cannot enact educational programs that would profit from taking identity into account. The many variables composing identity, such as class, ethnicity, and gender, intersect, especially on the US-Mexico border, in ways that illuminate the meaning of identity. For example, class and ethnicity are often conflated in studies of education. On the US-Mexico border, where the majority of people use the same ethnic category for self-description, class has a dramatically different meaning. I use Gibson's position as a theoretical starting point from which to ask: 1) What do anthropologists mean by identity?; 2) Is there a border identity?; and 3) How is it enacted? Once these issues are discussed, we may then concern ourselves with the implications of border identity for education.

Education in Nogales, Arizona raises questions about the role of ethnic identity and minority status on school performance in the borderlands. High schools in Nogales, Arizona experience a graduation rate of only 69.5% (White, 1996). Ethnicity and class are frequently cited as crucial variables for understanding differences between minority students' performance in schools and the performance of children of dominant groups. However, the term minority is somewhat problematic when applied to Mexican and Mexican-American students in Nogales, since they make up 83% of the community and a larger percentage of the school. Lotty Eldering suggests that the term minority refers to a low social position of a group over several generations, not simply to demographic status (Eldering, 1997:336). In Nogales, though, the "low social position" must be contextualized beyond the community level, extending it into national and international spheres. By relating position to the concept of ethnicity, students' minority status becomes more understandable. indeed, the connection between ethnicity and social position becomes even clearer when class is included in the picture. Differential access to power and resources link ethnicity, class, and social position as conceptual tools.

Gibson (1997) suggests that attending to issues of minority student identity will reveal much about their academic achievements and failures. The concept of identity, however, contains a complex package of ideas about gender, ethnicity, class, and power. These deas may intersect differently on the border than they do in the interior of the US. Then, what does identity mean in Nogales, and what impact does identity have on education? This paper examines how identities are enacted on the US-Mexico border, particularly ethnic and class identities, and how they affect education.

In order to address these questions, I begin by examining the concept of community, which has been problematized by transnational and border theory. The way people in Nogales conceive of space and place has bearing on the ways they identify who they are. Once identity as it relates to community is illustrated, a consideration of identity with regard to education can be explored. I focus in particular on two aspects of identity: ethnicity and class. In the bilingual cities of ambos Nogales,2 language is central to identity as well, and must be included in any picture of identity in schools.


The research

My first trip to Buena Vista in Nogales took place in October 1997. I continued to travel there regularly for the rest of the 1997-1998 school year. My trips included observations in classrooms at each grade level and in a variety of subjects. Semistructured, informal, and openended interviews involved solely adults, including teachers, aides, administrative assistants, the principal, and the school counselor. I spoke informally with current students, but did no formal interviews.

I also explored other sources of information for this project. The school counselor provided me with 114 questionnaires administered to parents of children at the school. Some of the questions specifically addressed the value parents placed on language use and bilingualism in schools. In an effort to identify trends in the public discourse about language in schools, I also searched the electronic index of The Arizona Republic for newspaper articles on bilingualism and education, or on the English-only movement, for all of 1998. Further research relied on the local paper in Nogales, Arizona.

Beginning in June 1998, I intensified my research by living in Nogales for six weeks. I regularly sat in on the summer school classes, consisting of around 50 students and taught by two teachers. Using snowball sampling, I was able to meet and interview people involved in education in Nogales at many levels and at several schools, including high schools and grade schools.

Additionally, I met and interviewed several Nogales residents who were unconnected to the school system. I spent a day visiting maquiladoras in Nogales, Sonora with a sales representative, for example. I talked with staff and volunteers at a local art museum and met their families and friends. I participated in many aspects of border living, crossing the border with informants —who rapidly became friends— to shop, buy medicine, and relax. These six weeks of fieldwork provide the primary data used for this paper.

My project begins with an acknowledgment that variables outside of schools impact academic performance. Thus, the study extends geographically into the community of Nogales, on both sides of the US-Mexico border, and raises conceptual issues concerning global processes and local agency. Much of this research has been performed with the help of funds from the Office of Youth Preparation (OYP), which works with schools throughout the state of Arizona. My research ultimately aims to provide information useful for the design and implementation of programs to improve education in Arizona. This project also strives for one of the goals of applied anthropology; that is, it is directed toward problems identified by the people under study and performed with their cooperation.


Nogales: the community defined

Defining a community has been significantly complicated by border research, and this study can not, and should not, avoid this dilemma. To some extent, one can rely on statistics and definitions provided by government agencies. These data must be considered a beginning, though, and not the complete picture of the research site.

The following description of the community begins with a statistical overview, but will also include a theoretical framework to consider the nature of communities.

Nogales, Arizona has a population of about 20 000 people and is closely tied to Nogales, Sonora, a community of 300 000 habitants across the border in Mexico. In Nogales, Arizona, 83% of the community is identified as Hispanic, 16% as white, and 1% as other, according to the 1990 census (Pffice of Youth Preparation, 1996). Nogales is the largest port of entry for winter vegetables in the United States, and the predominant source of income for the community is in sales (although what percentage is involved in produce is not stated). The produce business is active mainly in the winter months and invigorates the city. Life is noticeably slower-paced in the summer. An estimated of 60 000 Mexicans cross the border daily to shop in stores such as Wal-Mart and JC Penney, which can be reached by bus from the border. Despite the fact that Nogales is a center for international commerce, 27% of the population is below poverty level. This disproportionately affects children under 18 years old, 35% of whom fall below poverty level.

Buena Vista was the first of two middle schools established in Nogales, Arizona; the second opened in 1996. The Buena Vista school district is bounded on the south by the fence marking the US-Mexico border and on the north by Mariposa Road. School enrollment forms show that 28% of students are born in Mexico.3 Of those students, 93% have attended less than two years in us schools. Taking the entire student body into consideration, 93% report living in a home where a language other than English is spoken (presumably Spanish), however, only 85% have tested as Limited English Proficiency Students, based on writing samples. Importantly, 61% of Buena Vista students come from a household below poverty level. Although this information does not provide a representative view of the population of Nogales, Arizona, it does allow a more detailed picture to emerge.

Border communities and the people who live in them, must continually construct and reconstruct their identities, due to the tides of migration that flow through the area. However, Nogales has been described to me as a very stable community, and thus it offers depth to the picture of a town affected by transnational processes. One man told me, "I'm a newcomer. I've only been here 40 years". People describe themselves as "American" or "Mexican" with little regard for their status in the eyes of the US or Mexican governments. Rather, social networks, cultural traits, and length of residence determine identity. While shifts in ethnic identity of individuals and groups of individuals are the norm in all communities, life on the border may exaggerate these processes. Political rhetoric surrounding the border, originating from local, state and national politicians, demands that people be aware of their identity as well as their nationality.

Statements about identity on the border require a definition of community in ways that transcend statistics. Without trying to define or essentialize this now slippery concept, anthropologists can put forward some concrete characteristics that define a community. One description states: "collections of individuals living or interacting within the same territory do not in themselves constitute communities —particularly if those individuals do not perceive them-selves as such. What binds a community is not its structure but a state of mind; a feeling of community" (Shore, 1994:98; italics in original). Chris Shore emphasizes that communities, and therefore place, are socially constructed by individuals. Gupta and Ferguson urge us beyond this point, saying: "the more urgent task would seem to be to politicize this uncontestable observation" [i.e., that place is made meaningful] (Gupta and Ferguson, 1997a:40). They ask who has the power to define a community, whose "feeling of community" is the "correct" one.

People who have the power to define identity and community in Nogales schools are widely varied. They include administrators from the state and district offices, state and national education policy makers (including politicians), teachers, parents, and students. Each of these groups plays some role in defining the community, although their power to do so is not equal.

People in Nogales communicated a feeling of community to me that included different visions of the actual US-Mexico boundary. People rarely referred to the national boundary as a "border"; rather, it was almost always called "the line". The line played a variety of parts in the lives of people in Nogales, but its presence was usually noted. The acts of crossing, the daily encounters with people from "across the line", and national attention focused on the line shaped the sense of community in Nogales. Although the line was not seen as a barrier for most people, it did play a part in the way people "imagined" their community (Anderson, 1983).

Two women, Maria and Luisa, illuminated contrasting conceptions of community in Nogales in our conversations. Both women grew up in Mexico and now live in Nogales, Arizona and work at the school. Maria came to Nogales as a teenager with her parents after moving around in southern and central Mexico. She felt that "the border is unique from both northern Mexico and the southern US. It is its own thing: the border". Maria described feeling a sense of camaraderie with a traveler from the border of Italy and Eastern Europe, whom she had recently met. They shared experiences common to borders, and she thought that their homes endured many of the same problems because they were borderlands. She thought that Nogales, Arizona was just an extension of Nogales, Sonora, stretching into the United States. Maria crosses the border frequently, visiting in-laws and friends on both sides of the line.

Maria's co-worker, Luisa, told a different story. Born in Nogales, Arizona, Luisa grew up in Nogales, Sonora until she was 14 years old. She lives on the US side now, and says the two sides are distinct. Luisa said, "You can just feel a difference when you cross the border". Suddenly, one is in Mexico or the US. They feel different. She also said that now that her family all lives on the US side, she hardly ever crosses back to Mexico. She feels unsafe there, fearing violence. "You can't risk it with children", she explained. Luisa saw many changes in Nogales, Sonora, which she felt made it different from her community. "It's not just people from Nogales, Sonora anymore. People from other countries are there now to look for jobs".

Maria and Luisa express differing views of their community, one drawing a line at the border and the other blurring the line. Maria explicitly said (without any prompting) that both she and her husband "identify toward the Mexican side. Well, we can't help it, we're Mexican!"

Nogalenses4 also defined a sense of place in contrast to other communities. Rio Rico, a small settlement just 10 miles north of Nogales, provides homes for many who work in Nogales, including several teachers. People considered Rio Rico "very American", unlike the "Mexican" city of Nogales, Arizona. Although many people who lived there worked on both sides of the border, this community was thought of as "white" and therefore separate from Nogales. Most Nogalenses acknowledged some integration there, mainly in the last decade, but stated that Spanish was rarely spoken in Rio Rico and that stores carried largely "American" goods. This "separateness" existed in contrast to other "Mexican" towns even farther north from the border. They were not a part of Nogales either, but were considered more similar because they shared cultural traits identified as "Mexican".

Another important contrast used to define the community are border control agents.

Gloria, a woman in her mid-twenties, discussed crossing the border, something she does frequently:

When there's a big line, we say, "Oh, there must be a new guy". The new guys make a big fuss when they don't believe you. My boyfriend gets stopped every time [...] I asked them once, why is it us? [...] You have to cross and, ah, the big lines. Sometimes I don't want to go across the line today because of that.

This distinction parallels Donna K. Flynn's description of the Benin-Nigeria borderland: "The physical and social distance maintained by guards and border residents resonates with signs of inequality between urban and rural, center and periphery, and educated and uneducated" (Flynn, 1997:317).

Obviously, power relations shape how people define community in Nogales, "on the border". Border crossing is shaped by power vested in border agents in ways that make border residents subordinate (in some circumstances) to the border control.

Communities are socially constructed by individuals, thus, the interplay between individual identity and perceptions of community is essential. Indeed, in Nogales, perceptions of community clearly shape people's identities as borderlanders. The relevance of this theoretical link between community and identity was made clear to me by several people in Nogales, who spoke to me about a connection between their personal identity and a sense of place.


Identity and community

Before a discussion of border identity can fruitfully proceed, a brief exploration of the concept of identity is essential.

The literature on the subject emerges out of two distinct approaches: psychological and sociological. Frable, writing in the Annual Review of Psychology, begins her article by declaring, "Identity is the individual's psychological relationship to social category systems" (1997:139). Sherry B. Ortner, in contrast, writes from a position influenced by practice theory, placing identity "back in the realm of discourse, that is, how people talk about themselves and others, and of the larger shape of the discursive field from which people draw their categories" (1998:7). My own theoretical perspective leans more towards the latter, since this perspective highlights the construction of social categories, rather than treats them as fixed entities. However, the two approaches are united in their acknowledgement of the complexity of identity, including gender, ethnicity, sexuality, and class as elements of identity in any person. Importantly, both authors use identity to link the individual to the social realm. Nonetheless, when scholars attempt to apply the concept of identity to a specific context, identity tends to become as elusive and naturalized as the concept of culture.

Border identity has been described in many different ways by borderlanders and researchers. For some on the border, identity lies somewhere between being a constant struggle and a harmonious integration. In Borderlands: The New Mestiza, Gloria Anzaldúa (1987) describes herself and others like her as both divided and unified on the border. "[W]e don't identify with the Anglo-American cultural values and we don't totally identify with the Mexican cultural values. We are a synergy of two cultures with various degrees of Mexicanness or Angloness" (Anzaldúa, 1987:63).

In contrast, Michael Kearney uses the term "polybian" to define people who "adapt their being to different modes of existence as they opportunistically move in and out of different life spaces" (Kearney, 1996:141).

The strategizing polybian differs strikingly from the emotional introspection we find in Anzaldúa: "I have so internalized the borderland conflict that sometimes I feel like one cancels out the other and we are zero, nothing, no one [...] Yet the struggle of identities continues, the struggle of borders is our reality still" (Anzaldúa, 1987:63).

People in Nogales discussed both the affective and the strategic elements of identity in interviews, as well as on questionnaires. Gloria clearly linked her ethnic identity to her community, saying:

Here, everybody is my race. I don't leave [Nogales] because I think people won't like me the way people like me here. Everybody knows everybody; it's kind of a small town. You could tell when somebody's not from Nogales. When you go to the mall in Tucson, you can tell when somebody's from Nogales even if you don't know them. The faces are familiar.

Another woman, Julia, raised in Nogales, Arizona, recalls traveling in Mexico with sentiments similar to Anzaldúa: "I went to Mexico but I did not feel Mexican there. Yet in the US, I don't feel American. I'm very patriotic, but I don't belong". Clara, a prominent administrator, told me, "It is exhausting to be Mexican-American. Nobody wants us". Another woman, born in Mexico but raised in ambos Nogales, said, "We're Nogalians, we're not Mexican!" Yet she also spoke of feeling like "strangers" since her family had only lived in Nogales for about 45 years.

People felt like strangers in their own community for reasons other than length of residence. Clara, whose family had lived in Nogales, Arizona for at least three generations and whose father had held a prominent public office for years, told me that she was "considered an outsider". She had left Nogales for school and had lived elsewhere, returning later in life for a job. Julia shared a similar history and felt similarly. Clara related the following to me in one conversation:

In our neighborhood, we were the first Mexicans to go in and the last gringos to leave [... ] I have wondered, when did we become gringos? I think when my grandma died. She only spoke Spanish. The ties to the Mexican part of the family faded [...].

However, her feeling of being an outsider seemed to be tied to leaving the community, rather than to her status as a gringa.

The term gringo was especially relevant in discussing identity. One young mother told me initially that gringo "means a white person". She said, "It sounds insulting, but it's not. It's like a nickname for blond, white people". However, when I asked her if I would be a gringa if I spoke Spanish very well, she said, "No".5 Others confirmed the connection between speaking English over spanish and the term gringo, as well. Since other research has linked bilingualism to a border identity, gringo as a descriptive term may be more important for understanding border identity than it seems to these informants. I discuss the importance of language choice and language use on the US-Mexico border in more detail below.

In Nogales, personal and ethnic identities closely relate to perceptions of the border as a community. Thus, a strong sense of place appears to be a fundamental element of Nogales border identity. Therefore, literature on place and space within the studies of transnationalism and borderlands, discussed below, takes on implications that transcend theoretical musings when placed in this context. However, a problem arises from a poor fit between transnational theories and the actual lives of border people; that is, varying conceptualizations of place and space. The case of Nogales suggests that border researchers should recognize that although border people live transnational lives, they may not conform to transnational theories.


Conceptions of place

In his review article on the anthropology of borderlands, Robert Alvarez describes a division within scholarly approaches to the border (Alvarez, 1995:449). One group of scholars, the literalists, focuses on life at national borders and study issues such as migration, settlement, and identity in this context. Alvarez labels the second group "a-literalists", although they could also be called "conceptualists". These researchers "focus on social boundaries on the geopolitical border and also on all behavior in general that involves contradictions, conflict, and the shifting of identity" (Alvarez, 1995:449). In anthropology, the conceptual contributions that have arisen due to consideration of borders have been tremendous. Research on concepts broadly applicable in anthropology, such as community and place (Gupta and Ferguson, 1997a), ethnicity (Bentley, 1987; Vélez-Ibáñez, 1996), and identity (Flores and Benmayor, 1997), has been inspired and influenced by theories developed in response to border scholarship.

The very fruitfulness of border metaphors, however, has led some border anthropologists to call for a return to the literal. Arguing that "concrete, physical borders" have been ignored (Donnan and Wilson, 1997:4), Donnan and Wilson make a case for "an anthropology of borders [that] is distinctive because of its subject matter, namely the body of scholarship that ethnographers bring to bear in local communities which live and work at international borders" (Donnan and Wilson, 1997:10; my italics). Indeed, those of us who study border communities repeatedly pick up volumes with intriguing titles, like the recent Borderless Borders (Bonilla, et al., 1998), only to find that none of the research was conducted at a literal border. Donnan and Wilson's focus on nation and state in their most recent book (Wilson and Donnan, 1998) can be seen as a direct response to the metaphorical paths followed by conceptual border scholars.

The split between literal and metaphorical border studies closely corresponds with another scholarly division: borderlands and transnationalism. The division discussed above forces us to ask why these ideas are united at all. The answer to this question, I propose, lies in the topics on which we focus when we study the border. Simply living at the border may not be enough to mark someone as a borderlander. Specific processes must be present to be classified as a borderlander, and to identify oneself and others as such. Oscar Martinez, for example, lists several traits as typical of his "borderland milieu": "transnational interaction, international conflict and accommodation, ethnic conflict and accommodation, and separateness" (Martinez, 1994:10). Tobias Wendl and Michael Rösler echo this sentiment when they write that borderlands "are more exposed to foreign, transborder influences and crossborder movements than are the heartlands" (Wendl and Rösler, n.d.:8). Clearly, transnational processes and activities provide a link between studies of borderlands and studies of transnational trade, migration, economics, and ideology.

I contend that scholars of transnationalism have been more likely to challenge existing anthropological ideas about community, space and place, and identity, than those studying literal border communities, due in large part to the nature and content of their data and the processes inherent in borderlands. However, many theories developed for transnational studies have been applied to border communities with only partial success.

Some aspects of border life have been illuminated by the deconstruction of notions of community, culture, and identity. Indeed, a study of kinship on the border is unthinkable without an end to the image of national boundaries as somehow "natural" and impermeable (see Familia [Alvarez, 1987] for an informed ethnography of transnational social networks). However, not all theory that results from transnational research questions has been appropriate for understanding border communities. One example of a problem that comes from a poor fit between transnational theory and border lives is conflicting ideas about place and space.

Many authors writing about the US-Mexico border, and borderlands in general, have emphasized processes of deterritorialization that accompany transnational and global processes (Gupta and Ferguson, 1997b; Kearney, 1995; Rouse, 1991). Gupta and Ferguson, in particular, emphasize the ways in which people construct place within contexts of power. They note that while this concept contains a challenge for anthropology as a whole, the nature of borderlands inherently questions traditional anthropological categories. "The fiction of cultures as discrete, objectlike phenomena occupying discrete spaces becomes implausible for those who inhabit the borderlands" (Gupta and Ferguson, 1997a:34). Thus, much recent work has used borders and transnational processes to destabilize an idea of community or culture that is bounded and isolated. Authors speak of "hybridized" subjects and "ambiguous" or "unstable" border identities.

In contrast, Donna K. Flynn uses a case on the Benin-Nigeria border to show that "a deeply placed stable identity" can be created by a borderland (Flynn, 1997:312-313). She uses the term "deep territorialization" to describe the sense of rootedness in the border that people on these borderlands express.6 Indeed, this concept appears to apply in Nogales, Arizona. Often, after I asked a question about education "on the border", I was met with a pointed response about what education was like in Nogales. People in most communities might wish to differentiate qualities of their home from a larger region. However, this reaction coupled with other data suggests that a sense of place, while not necessarily bounded, is particularly essential to any conception of identity in Nogales. These contrasting views about place, deterritorialization versus deep territorialization, illustrate a difference between transnational theory and border theory. The case of Nogales implies that Flynn's concepts, developed on a border, are more appropriate for understanding border identity, despite the prevalence of transnational processes in Nogales.

Statements about length of residence in Nogales reflect this sense of place. When people call themselves newcomers if their families have lived in a city for over 40 years, they are expressing a view of that place as stable and permanent. They also illustrate what is necessary to be a full participant in the community. Ideas about length of residence affect border identity and, thus, education on the border.


Length of residence

Margaret A. Gibson's contention that a secure ethnic identity promotes academic success, no matter what the identity may be, has interesting implications on the border. This theory implies that students who are recent immigrants are often more likely to do well in school than second-generation students who feel ambivalence toward their identity. I ask if there is a secure border identity, which reflects inherent conflict while being a stable category. If bilingualism is the ideal in Nogales, do students show their border identity through their use of language? Code switching, for example, may say more about identity than about solid skills in either spanish or English. Length of residence, and deep territorialization, at the border, appear to be crucial factors for understanding identity in this context (Flynn, 1997).



Anthropologists studying education have written about the interplay of ethnic identity and school performance in new ways. This literature often relates racist teachers and bad schools to poor minority student achievement (Gillborn, 1997; Rosenfeld, 1971). Gibson (1997), as cited above, goes a step further. She states that those most likely to fail in school "feel disenfranchised from their culture and at the same time experience racial conflict" (Deyhle, 1995:419-420, in Gibson, 1997). Other authors support the argument that, whatever the ethnic group is, a strong ethnic identity contributes to school success (Cummins, 1997; Vigil, 1997).

These observations have interesting implications for people in Nogales and in the borderlands generally. Schools in Nogales employ many non-Mexican-American teachers and staff members. Interactions between them and their students can be compared to those between Mexican-American teachers and students for evidence of ethnic boundary marking, using language and actions (Erickson, 1987). How does a stable identity manifest itself in a place "characterized by conflict and contradiction, material and ideational" (Alvarez, 1995:448)? The border offers a complex, mutable environment, that individuals must confront regularly.

The massive exchange of commodities, both human and material, dramatically affects life and behavior, as does the continuous shifting and reconfiguration of people, ethnicity, sexual orientation and identity, and economic hierarchy and subordination (Alvarez, 1995:451).

As such, the border is both a challenge to researchers and a rich laboratory. Thus, research on identity in Nogales contributes novel information to research on education. While many other studies have focused on immigration and education (Eldering, 1997; Gibson, 1987), research on education has not made the border itself a variable. Testing assumptions about ethnicity, class, and minority student achievement facilitates the design and implementation of programs to improve minority student success.

Much of the new work on minority education mentioned above has occurred in response to two competing theories about minority children in schools. John Ogbu (1978) proposed a distinction in minority students between voluntary immigrants and involuntary immigrants, or "castelike minorities". Involuntary immigrants included African-Americans and Mexican-Americans in the Southwest, while voluntary immigrants came from Asia, Europe or elsewhere in search of opportunity. Students who were part of castelike minorities adopted behaviors needed to survive their oppressed position, and these behaviors clashed with the expectations of teachers, who were mainly members of the dominant ethnic groups. Frederick Erickson (1978) and others wrote from a perspective labeled "cultural difference". This theory emphasizes the fact that minority students and their teachers often communicate in different ways, which prevents these children from excelling in school. Thus, cultural differences are the primary factors in academic success.

John D'Amato, commenting on the debate between Ogbu (1987) and Erickson (1987), wrote that both theories "rely on typification of minority children and more or less mechanical models of interaction" (D'Amato, 1987:358). How can our understanding of cultural behavior, as anthropologists, clarify the actions of students and teachers, without reducing either students or teachers to "mechanical models"? If we believe human action is produced by cultural and social structures, and remember that human action also reproduces and transforms those structures, then how does this knowledge help us analyze minority student failure? We must look at the interaction of two structures embodied in human agents (Sahlins, 1981). These structures contain information on gender, class, ethnicity, and generation, and so frequently conflict in students and teachers. As people strategize, resist, and practice their culture, they make and remake existing power relations. Analysis must be made of the interaction, not of each group as a separate, typified entity. If there is blame for minority student failure, we will find it, not in one individual (teacher or student), but in the interaction between the two. Each brings their own structures to the "conjuncture" to reproduce and transform the relationship between them.

Similarly, we gain little from a separate analysis of the roles of class and ethnicity in the schools. The interplay between class and ethnicity reveals the workings of both. These concepts are complex, holding variable meanings for anthropologists and lay people. Acknowledging power relations in ethnicity is as essential as considering the relation to the means of production when discussing class. Perceptions of ethnicity are intertwined with perceptions of class, and both affect individuals' identities.

I contend that we find identity such an intriguing and productive topic with respect to education because it calls attention to power differentials. Ethnicity and class always occur in a context of power relations, and pulling these elements apart may not help us to see power more clearly. For example, if a student identifies with an oppressed ethnic group and a household in poverty, those separate variables may be too dependent on one another to consider alone, however, that student's relationship to power and resources is clearly connected to his or her identity. Analyses of identity and power, rather than ethnicity or class, may provide a method with which to understand specific implications about a variety of social stratifications, including gender, class, and ethnicity.


Ethnicity and class

Ethnic identity always arises in relation to an other. Individuals conceive of themselves as a member of one group, and not as a member of the other. Thus, ethnicity is primarily an action, not a thing (Bentley, 1987:26; Hegmon, 1996). Indeed, it is more appropriate to use a phrase indicating an action, like ethnic identification, to express what is commonly called ethnicity, a noun. Outsiders also identify a person or group's ethnicity, though the perception of who makes up an ethnic group may not match that of the members. The ways people identify themselves can be based on many factors, including common history, cultural similarities, language, and physical traits, to name only a few possibilities.

Because ethnic identification is also extremely situational, "ethnicity" has interesting complications on the US-Mexico border. For some on the border, identity lies somewhere between a constant struggle and integration. Anzaldúa (1987) refers to herself and others like her as Spanish, Latin, mexicanas, chicanas, Mexican-Americans, and tejana (to name only a few), depending on context. Border identity is both divided and unified in her conception. Although not as extreme, other borderlanders face similarly complex choices when identifying.

Nogales is situated not just between two nations, but within the United States and Arizona as well. This position is important to consider, lest the border become a place unto itself. As constituents of larger societies, ethnic groups not only occur in relation to other groups, they are ranked (Aguilar, 1993). In fact, membership in an ethnic group can become the basis for inequality in access to power and resources (Cohen, 1978:391). Ethnic ranking can affect both the economic and political success of the group, as well as the self-esteem of individuals in the group. When examining the ramifications of ethnicity, an emphasis needs to be placed on understanding both of the effects of ranking, not, as so often happens, only on the more tangible economic and political consequences. As James C. Scott notes, "it appears that slights to one's dignity [...] loom at least as large in accounts of oppression as do narrower concerns of work and compensation" (Scott, 1990:23).

The results of ethnic ranking have led to a dichotomy in studies of ethnicity, characterized as primordialist and instrumentalist (Bentley, 1987: 25). Primordialists see people identifying traits that define and group themselves with similar people because of emotional needs. This theory is closely tied to genetic or racial relationships, as the traits cited are generally physical or linguistic. Instrumentalists see ethnic groups as people in special kinds of factions that have shared material interests. Cultural similarities can help to legitimize the unity of the group, transcending class and other barriers, as these people seek common political and economic goals (Brumfiel, 1994:93). Elizabeth M. Brumfiel, in particular, offers a compelling look at ethnic groups in ancient Mexico, in which she claims that "ethnicity was a tool, fashioned to the needs of political actors as defined by the existing political structure" (Brumfiel, 1994:102).

G. Carter Bentley, in critiquing both the primordialists and instrumentalists, points out that neither accounts for the individual's initial recognition of similarities and differences (Bentley, 1987:27). Bentley focuses on the concept of habitus to address this weakness (Bourdieu, 1990). Habitus is learned in the same way language is learned: practical skills are acquired without conscious awareness of a structure. Thus, a person is able to recognize speech patterns, body movements, and values in another as being "like" one's own, without being able to articulate why such recognition is possible. The identification takes place at the subconscious level of habitus.

Bentley goes on to suggest that simply because people use ethnic groups strategically for material goals does not mean they were formed for that purpose. Likewise, individuals may manipulate ethnic symbols to their emotional or material advantage in times of change and stress, but those symbols do not originate in order to serve that function (Bentley, 1987:48). Bentley theorizes that people do not form ethnic groups to serve a function, either emotional or political, but do so because they enact learned behavior. This idea does not deny that ethnicity can serve as a tool to be manipulated, but it challenges the idea of those functions creating identity.

Typically, anthropologists have described the use of ethnic symbols as tools to be manipulated, even suggesting that ethnic groups represent a political struggle along ideological lines (Williams, 1989). Recent writing about the US-Mexico border champions a different perspective that, while taking class and issues of power into consideration, puts the emphasis in new places (Vélez-Ibáñez and Greenberg, 1992).

The key to understanding the forces that shape U.S. Mexicans lies in the historical struggle of their households over control of their labor and resources, and for economic security [... ] We also will argue that, since the late-19th century, the combination of the historical forces of industrialization and their accompanying immigration policies has contributed binationally to the rise of U.S.-Mexican ethnicity (Vélez-Ibáñez and Greenberg, 1992:314).

In other words, a particular class position led to the rise of a new ethnicity on the border. From that assumption, actions that seem to be determined by ethnicity may actually originate in class-based responses.

An important and obvious contradiction exists between Bentley and Vélez-Ibáñez. Bentley describes ethnic identity as learned from early childhood, without functional goals. Vélez-Ibáñez, on the other hand, calls attention to what is learned by using the concept of "funds of knowledge", which originate in specific political and class contexts, preserved and transformed over generations in households (Vélez-Ibáñez, 1996:162-163). Thus, even a middle-class urban Mexican-American family's ethnicity reflects their ancestors' rural, manual labor-oriented lives. Bentley paints ethnic identity as politically neutral, but able to be used for instrumental ends. Vélez-Ibáñez states that ethnicity, at least Mexican-American ethnicity, is generated by political and class-based circumstances. He reminds us that there is always a political context, shaped by power, and suggests that this context is embedded in identity. Bentley might reply that while the context may be present, ethnic identity still does not arise to serve any particular function within society.


Power and race

Regardless of the origins of ethnic groups, ethnicity and class are firmly entwined in the minds of anthropologists. A reemphasis on the importance of power has led to a resurgence in the use of the term race. Roger Sanjek (1994) discusses the distinction between race and ethnicity, defining one in terms of repressive exclusion, and the other as inclusive cultural identification (Sanjek, 1994:8). He suggests that terms like "multicultural" or "ethnic diversity" are euphemisms that hide a stratified racial ordering in society. By claiming equal status for all groups, and by denouncing race as nonexistent, anthropologists make it difficult to confront racism in their work. This idea supports the notion that without recognizing the relations of power and politics in our work, anthropologists are doomed to enable repression in the cultures that we study.

Power relations certainly must be addressed in discussions of identity. As Vélez-Ibáñez (1996) links US-Mexican identity to working class conditions, others have shown how upper and middle class conditions are connected to ethnic identity as well. Ruth Frankenberg (1994) demonstrates how conceptions of white ethnicity are related to dominance and power, which align all whites with the upper class. Sherry B. Ortner and Karen Brodkin Sacks have individually suggested that Jewishness is strongly linked to middle class values (Ortner, 1996; Ortner, 1998; Sacks, 1994). Ortner describes conversations with Jewish men and women for whom it was unthinkable that they could be other than middle class, even when their income belied that claim. Likewise, Vélez-Ibáñez suggests that Latino values are based in knowledge gained from experiences in manual labor (1996). All of these authors imply that, regardless of what a person's actual economic status is, he or she may identify with a class based solely on their ethnicity.

The secure bonds holding conceptions of ethnicity and class together exist in the minds of anthropologists and the minds of lay people. Why, then, is there greater attention paid to ethnicity as a factor in education than to class? Programs to combat poor test scores and statistics on school failure are consistently designed with categories reflecting ethnicity, not class. Considering the fuzziness of both categories, is it simply easier to identify an ethnic group than to define the boundaries of a class? Or does this uneven emphasis illuminate an aspect of our society and our schools to which we are largely blinded?

Many authors have suggested that schools reproduce class divisions and, indeed, are designed to perform this function (Bowles, 1972; Carnoy, 1972; Willis, 1977). Samuel Bowles writes that US schools "have evolved [... ] to meet the needs of capitalist employers for a disciplined and skilled labor force, and to provide a mechanism for social control in the interests of political stability" (1972:36). Paul Willis (1977) has provided ethnographic support demonstrating reproduction of divisions of labor in schools. (Division of labor is only one aspect of class, though, and this fact may be part of the reason why class is often downplayed).

Many studies that consider class are reduced to using income level as a guide (Vigil, 1997). In Personas Mexicanas, James Diego Vigil (1997) and —I assume— most of his informants conceive of class position as a combination of annual income and hierarchical social status. This categorization is inadequate in the eyes of political economists, for whom wage laborers are distinguished from people with access to capital. These uses of class, while related, are certainly distinct.7

Nogales provides a unique site for testing theory on how ethnicity and class affect performance in the classroom. For example, in a place where almost everyone is "Mexican-American", ethnic boundaries may form according to the length of time spent in the US or in the border city of Nogales. Furthermore, the example of Nogales allows us to examine potential changes in ethnicity. Sacks (1994) described the whitening of Jews and Eastern Europeans as they became middle class. Will this occur in Nogales as established, middle class residents increasingly differ from new immigrants? And will this perceived whitening affect children's performance in schools? All of these questions are singularly suited to the border.

Let me turn now to identity as it relates to school performance, which creates its own challenges. Here, the practical importance of the debates about identity becomes crucial. In this paper, I use the school's own standards to assess students' academic achievement. These standards include both the stated and unstated goals of schools. Standardized testing on course content appraises only one educational goal. Social success, as expressed by teachers' desires to see students participate in sports or other activities, is harder to quantify. Parental standards, as well as those of the students themselves, must also be considered. The parent questionnaire asked, "What is acceptable student academic performance?"8 The answers to this question lead us to a consideration of language as an essential element of border identity.


Language and identity on the US-Mexico border

Anything can become symbolic of ethnicity (whether food, dress, shelter, land tenure, artifacts, work, patterns of worship), but since language is the prime symbol system to begin with and since it is commonly relied upon so heavily (even if not exclusively) to enact, celebrate and "call forth" all ethnic activity, the likelihood that it will be recognized and singled out as symbolic of ethnicity is great indeed (Fishman, 1989b:32).

People commonly use language as an indicator of ethnic identity, in order to identify others' membership and to self-identify. Anthropologists have used language as a gauge to place border people in categories along a continuum, such as "Mexico-oriented" or "U.S.-oriented" (Vigil, 1997); these terms also appear as emic categories utilized by people in Nogales, Arizona. Furthermore, interviews with border residents suggest that language preferences can be a way of asserting allegiance with an ethnic group. In a school setting in which knowledge of English is demanded, poor English skills may be a sign of concern over identity rather than a resistance to learning. The situation is further complicated by the fact that Nogales is a dominantly bilingual community, where the most successful individuals speak Spanish and English.

Nogales, as a border town, exhibits a continually shifting population due to steady migration. Therefore, ethnic alliances will also shift as people alter their perceptions of their position in the community. As one woman told me, "We were the first Mexicans to go in [to that neighborhood] and the last gringos to leave".

Here I examine the relationship between language and ethnicity, and consider the consequences of this relationship on academic achievement. I contend that language has been deliberately used as a badge of membership in the borderland. Children are aware of and follow this pattern in schools, employing language as a means of expressing ethnic, and perhaps class, identity. As the interplay between class and ethnicity reveals more about students' achievement in schools than either factor by itself, so too may the interplay between language and group identity. Language, as used on the border, is a tool to express ethnic and class identity.

Many authors have found the relation of language and ethnicity to be fertile ground for debate. Epstein (1977) casts the movement behind bilingual education to be a political one, designed to bolster Chicano claims to power. Joshua A. Fishman, on the other hand, uses the issue of bilingualism to encourage a multiethnic, multilingual community in this country (Fishman, 1989a:652). John J. Attinasi extends the discussion into a look at dominant discourses, and the use of language to oppress minorities (Attinasi, 1994).

The introduction of Spanish as a learning tool for students has sparked a dramatic response from people calling for "English-only" in US schools. This movement demonstrably has issues of identity at its heart: it represents a battle over what it means to be American. Although several authors writing about bilingual education ignore the implications for students' identities (McGroarty, 1991), this aspect sheds light on students' academic performance in Nogales and elsewhere.

A recent comparison of Spanish/Nahuatl and the relation of language and power is especially relevant to the argument presented here. Norbert Francis and Phyllis M.Ryan (1998) compared students in two settings in Mexico: one in an urban, Spanish-dominant area, the other in a rural, Nahuatl-dominant town. The relative successes of both groups in learning English were compared, as were student attitudes toward this endeavor. The students in the Nahuatl town exhibited much more enthusiasm for English than those in the urban area. Indeed, English proved easier to learn in this context than spanish, since the Nahuatl children felt embarrassed to speak Spanish (not English) in public (Francis and Ryan, 1998:35). In the Spanish-speaking, urban setting, students consciously linked English with the United States. They expressed reluctance and even hostility toward learning the language due to their feelings about the country with which they associated English. Francis and Ryan claim that attitudes about language embody cultural conflict: Spanish against English is Mexico versus the United States, Nahuatl against Spanish represents the indigenous versus the dominant power (Francis and Ryan, 1998:27).

Their study raises pertinent questions for the border. Will new immigrants, particularly children, exhibit resistance to English as a way of fighting for their identity? Francis and Ryan suggest that there may be regional differences in attitudes toward English and Spanish. Other authors add class into the picture (Urciuoli, 1995). Bonnie Urciuoli writes:

How different Latin groups view and retain Spanish-English use varies with ethnic and class location [...] The class and race differences that are mapped onto language are reproduced in the practices and performances that make up students' experiences (Urciuoli, 1995:537).

The connection between language and identity is abundantly documented in the literature. Evidence from Nogales supports this connection, but in ways specific to the context of the border. Below I illustrate this connection by reviewing conclusions from three types of data from my research in Nogales: 1) ethnographic interviews; 2) a parent questionnaire; and 3) newspaper articles.


Ethnographic data

My first inkling that I should look into the connection between language and identity came from a conversation with Maria, who works with special education students. Maria began her life in southern Mexico, moving north to Mexico City. From there, her family continued to Nogales, Sonora. Later, she attended college in Tucson, and found the Spanish there very different from what she knew in the interior. Indeed, the whole way of life was different. Maria also found Tucson completely different from Nogales. She emphasized several times that "the border is its own place, not Mexico or America. On either side, you are a border person".

The process of learning English was frustrating for Maria at times. She said she thought, "Why should I have to work so hard to read a book when it is so easy for me in spanish?" She links her own experiences with those of her students at Buena Vista, who come to her with problems in English and math. She thinks that language acquisition is "a cultural issue". She herself thought and describes her students as thinking:

I don't wanna be a gringo, so I don't want to learn English! They are afraid of losing their identity. I see now that this is wrong, but I used to feel that way. It is possible to be Mexican and speak English.

Maria went on to tell me about the two or three kids who only speak English. They have a very hard time, socially; they are not well accepted. Interestingly, Maria's belief that students who have the most confidence are the ones with strong skills in both languages calls attention to the role of language on the border, where being bilingual is part of the border identity. Observations of classrooms and student conversations reveal the social importance of bilingualism as well. Although Spanish is clearly dominant in casual conversation, English is dominant during class time.

In a review of my field notes, I noticed other situations in which teachers had volunteered information about their experiences with language (even without my asking). Clearly, language and its role in identity play a significant part in border life. One teacher recalled growing up in a small town on the New Mexico-Arizona border. Ten to twenty percent of the town was Mexican-American, yet he and other students were punished for speaking Spanish in school. Today, he uses both Spanish and English to teach his eighth grade math students. Another teacher, Trini, who grew up in Nogales, told me without prompting that she did not remember being scolded for speaking in Spanish in school; however, Trini thought she had an easier time in school because she knew English. Her mother, who was raised in Nogales, taught her children both languages, and taught her Mexican husband English as well.

These comments reinforce the validity of the assumption that bilingualism is associated with a border identity. The parent questionnaires also reflect these attitudes.


The parent questionnaire

In 1998, the school sent parents questionnaires in both English and Spanish (table 1). The questionnaires contained 13 queries about education generally and at Buena Vista specifically. A total of 114 parents responded: 60% (68) used Spanish, 38% (43) used English, and the remaining 2% (3) responded bilingually.

The use of language on the questionnaires raises several questions. The school reports that 93% of their students come from Spanish speaking households. Does the relatively high number of English responses reflect a large number of fully bilingual parents? Does it reflect a proportionally greater involvement of English-speaking parents in their children's schooling? Does it reflect the perceived prestige value of English in academic settings, or simply a situational choice?9

For the purposes of this paper, I examined the responses to question ten: "How important is it for you that your child be fluent in two or more languages?"10 I also scanned the rest of the questionnaires for any other comments about the need to learn English and/or Spanish in school. For example, questions one and two ask, "What is acceptable academic performance?" and, "What should our children be learning in school?"11 However, I did not take comments about reading and writing skills into account, since those did not illustrate the need for either language.

In answer to question ten, 92% (105) wrote that bilingual fluency was "very important" or "muy importante" for all or part of their response (table 1). Indeed, one paper was entirely blank except for this question, to which the parent replied, "Very important!" Breaking this response down by language, the Spanish replies stated "very important" on 100% of the forms, while the English replies of "very important" were on 86%. This discrepancy was compounded when similar statements mentioned elsewhere on the questionnaire are taken into account. Nineteen percent (13) of the parents writing in Spanish commented elsewhere that learning English, or English and Spanish, should be an essential part of their children's education. In comparison, only 7% (3) of the English-using parents did the same.

Many parents elaborated on their response to question ten. I attempted to break their comments into meaningful categories, using their word choice as a guide. For example, one parent wrote: "Es muy importante por el área donde vivimos. Si se hablan 2 idiomas se pueden obtener mejores trabajos". I marked them as making a reference to the border and work, as well as falling in the "very important" category.

In another case, a parent responded: "Very important. That is about the only advantage they might have over other high skilled countries". In this instance, I marked "competition" since the writer referred to competing with others for jobs. While these categories would benefit from further refinement, I was able to see differences between the two dominant groups. Tables 2a and 2b show the frequency of several categories. The percentiles are in relation to the total number of elaborations for the respective language. Some elaborations were not tallied (due to ambiguity or legibility), and were therefore excluded from the total. No statistical analysis has been applied to this data to determine the relative significance of the responses, but some patterns do emerge. Both sets of parents commonly saw bilingual fluency as a way for their children to access economic opportunity in the future, however, the parents using English made comments about competitiveness more frequently. The same group felt that learning English should be a priority over being bilingual, while the parents who responded in Spanish did not express this sentiment. Both sets of parents often discussed the importance of the future, often without qualifying statements.

The responses having to do with the border and ethnicity especially illuminate a border identity, unique from a continuum model running from "Mexican" to "American". 10% of the Spanish replies and 19% of the English ones mentioned the border as a reason for the importance of bilingualism. This result reinforces ideas about identity and bilingualism in Nogales. Furthermore, two parents, both writing in English, specifically addressed issues of identity. One wrote, "It is very important to learn how to read and write in Spanish because after all we are hispanics [sic]. The same with English". The other wrote, simply, "As a hispanic [sic] family and living in a border town". Although these answers do not represent a majority, they do directly link bilingualism, the border, and ethnic identity in the minds of parents. These questionnaires offer clues about the attitudes that parents of Buena Vista students have about language; they lend support to the thesis that bilingualism is an essential part of a border identity.


Newspaper articles

In January and February of 1998, the Arizona Republic published about one article a week on bilingual education. The eight articles reviewed here can be sorted in several ways. Three of these discussed bilingual education in nonpolitical terms, focusing on the efforts of schools, teachers, and students to improve bilingualism in local schools (Anderson, 1998; Jones, 1998; Santos, 1998). Two other articles focused on efforts in California to ban bilingual programs as they stand; one of these was an editorial, the other masqueraded as a piece of journalism (Irvine, 1998; lasken, 1998).

Another editorial posited that bilingual education is "an expensive failure" in Arizona and should be eliminated (Steele, 1998).

One article described the efforts of local bilingual educators preparing for battles in this state similar to those in California (Navarrette, 1998).

Finally, an article reported on Arizona's new high school exit exam, which will only be offered in English (Baker and Barrett, 1998).

This variety of articles illuminates the politically charged nature of this discourse. The articles and editorials are filled with references to the power of language. "English is, after all, the language that binds us together", writes Douglas Lasken (1998). One man is quoted by Martha Irvine as saying, "he has no worries that [his sons] do not speak Chinese [... ] Says Pon, now vice chairman of the state's Republican Party, 'They are Americans, not Chinese'" (Irvine, 1998). These statements and others demonstrate a connection between American identity and English in the minds of the speakers. These are not messages about the economic necessity of knowing English in this country, nor about academic excellence (although those statements exist too). Such statements send a message about who can be American and what Americans do.

Concern about language has been manifested in political initiatives on both sides of the border, not only in the US. The Mexican government in the late 1970s established the Comisión Nacional para la Defensa del Idioma Español. The mission of the commission was to safeguard Spanish from English encroachment (Martinez, 1988:121). The program poured its efforts into major Mexican cities, tourist hubs, and the US-Mexico border region. Language was stressed in order to invigorate "a dormant national consciousness and infirm ethnic identity" (Hidalgo in Martinez, 1988:121). The United States and Mexican governments have each used language to emphasize national, and ethnic, identity. I contend that these messages are not lost on people at the border, especially children. The media discourses could have a profound impact on the decisions people make about using English or Spanish. Indeed, the statements made by children ("I don't wanna be a gringo") about their reluctance to speak English clearly show their concern over their identity.

I consider language as a symbol and tool employed for expressing identity in the borderlands. Ethnographic interviews revealed the importance of language to a sense of self in schools. The parent questionnaires reflect the value of bilingualism on the border. Even articles in local papers consider language and identity unified.

At this point, we must ask ourselves what this interconnectedness means for students and other border dwellers.

First, language alone does not determine ethnic identity. Many other facets of behavior and attitude make up a total identity. Losing Spanish for English does not necessarily indicate a switch in allegiances. Yet, when educators ask Mexican children to learn English, they should be conscious of the internal conflicts that may arise for students. Sensitivity to the issue is a first step; curricular design, taking issues of ethnicity into account, should be an essential next step.

Second, further research should take class into account. In addition to ethnicity, class position may be enacted with the use of English. Research should not overlook the importance of power relations on the choice of when to use a given language. More study is needed to examine this idea as it relates to language use in Nogales schools.

Finally, bilingualism has emerged as a key component of border identity. In Nogales, Spanish is the dominant language, but the distribution of responses to the parent questionnaire points to the pervasiveness of bilingualism. If 93% of students at Buena Vista report coming from a home where Spanish is spoken, then a 40% response in English is notable, to say the least. The use of both languages in Nogales suggests that perhaps there is an identifiable, if fuzzy, category for border people. This category reflects bilingualism but transcends it, incorporating many aspects of life on the boundary between two nations. Neither one nor the other, they are both.

The results of this study highlight one aspect of a specific border identity. It is a gateway for addressing Margaret A. Gibson's (1997) suggestion that a secure identity promotes academic success. Bilingualism as a key to understanding the border identity may help educators better anchor their students in their community, promoting greater academic achievement on the border and in other educational sites.



Clearly, language use is not the only, or even the most important, variable for understanding identity on the US-Mexico border. Here, I illustrate that language use, attitudes about length of residence at the border, and conceptions of place, form important aspects of a "border identity". This category, although fuzzy,12 should be separated from a nation-based continuum model, in which people must identify somewhere between two poles, in this case the nations of Mexico and the United States. We need to ask instead, does this person identify along a continuum that runs from "more of a borderlander" to "less of a borderlander"? Perhaps we should contrast a borderlander from someone who lives in the heartland of either country, rather than apply a nation-based model to people whose lives transcend national boundaries.

Research in Nogales and elsewhere has shown identity to be a topic suffused with power relations. Rather than trying to pull apart class, ethnic identification, gender, or other variables, I use the concept of identity to access those power differentials as they exist in Nogales, Arizona. This place, specifically imagined by its residents, embodies processes particular to borders, but it also holds implications about identity for people in the heartlands. Border identity in Nogales is enacted in a variety of ways: in language use, in perceptions of place and space, in attitudes toward length of residence in the community. If we take identity to be critical to an understanding of minority academic achievement, we need to look for the ways it is embodied and enacted in academic contexts. The concept of identity can tie anthropologists to agent-based theory, but we should look for it in human actions, not see it as a fixed thing.

A benefit of using the concept of identity, rather than concentrating on one social category, lies in the freedom to consider factors normally excluded from research. For example, studies on education tend to emphasize ethnicity and class, which are certainly crucial variables for understanding differences in academic success between groups. However, this study found that border identity was characterized by a very stable sense of place. How would such a factor fit into the categories of ethnicity or class? Far more important than the kind of food one eats, a particular sense of place and space has relevance to education and the social divisions reproduced in schools. Identity allows us to move out of bounded ideas of what makes up social categories, letting us see the behaviors and beliefs that truly give meaning to people's lives.



Aguilar, John L. (1993). "Expressive ethnicity and ethnic identity in Mexico and Mexican America". Mexican American Identity. M. E. Bernal and P.C. Martinelli, eds. Floricanto Press: Encino. pp.55-66.

Alvarez, Robert R. (1987). Familia: Migration and Adaptation in Baja and Alta California, 1800-1975. University of California Press: Berkeley.

---------- (1995). "The Mexican-US Border: The making of an anthropology of borderlands". Annual Review of Anthropology, 24:447-470.

Anderson, Benedict (1983). Imagined Communities. Verso: London.

Anderson, E.J. (1998). "Bilingualism the best of both worlds". The Arizona Republic. February 20.

Anzaldúa, Gloria (1987). Borderlands/La Frontera: The New Mestiza. Aunt Lute: San Francisco.

Attinasi, John J. (1994). "Racism, language variety, and urban US minorities: Issues in bilingualism and bidialectalism". Race. S. Gregory and R. Sanjek, eds. Rutgers University Press: New Brunswick. pp. 319-347.

Baker, Lori, and Jennifer Barrett (1998). "Fail this test, get no State diploma: Foes say it's biased against Hispanics". The Arizona Republic, February 9.

Bentley, G. Carter (1987)."Ethnicity and practice". Comparative Studies in Society and History, 29:24-55.

Bonilla, Frank, Edwin Melendez, Rebecca Morales, and Maria de los Angeles Torres (1998). Borderless Borders: US Latinos, Latin Americans, and the Paradox of Interdependence. Temple University Press: Philadelphia.

Bourdieu, Pierre (1990). The Logic of Practice. Nice, R., transl. Stanford University Press: Stanford.

Bowles, Samuel (1972). "Unequal education and the reproduction of the social division of labor". Schooling in a Corporate Society. M. Carnoy, ed. David McKay: New York. pp. 36-64.

Brumfiel, Elizabeth M. (1994). "Ethnic groups and political development in ancient Mexico". Factional Competition and Political Development in the New World. E.M. Brumfiel and J.W. Fox, eds. Cambridge University Press: Cambridge, England. pp. 89-102.

Carnoy, Martin (1972). "Introduction". Schooling in a Corporate Society. M. Carnoy, ed. David McKay: New York. pp. 1-17.

Cohen, Ronald (1978). "Ethnicity: Problem and focus in anthropology". Annual Review of Anthropology, 7:379-403.

Cummins, Jim (1997). "Minority status and schooling in Canada". Anthropology and Education Quarterly, 28(3):411-430.

D'Amato, John (1987). "The belly of the beast: On cultural differences, Caste-like status, and the politics of schools", Anthropology and Education Quarterly, 18(4):357-360.

Deyhle, Donna (1995). "Navajo youth and Anglo racism: Cultural integrity and resistance". Harvard Educational Review, 65(3):23-67.

Donnan, Hastings, and Thomas M. Wilson (1997). "An anthropology of Frontiers". Border Approaches: Anthropological Perspectives on Frontiers. H. Donnan and T. M. Wilson, eds. University Press of America: Lanham. pp. 1-14.

Eldering, Lotty (1997). "Ethnic minority students in the Netherlands from a cultural-ecological perspective". Anthropology and Education Quarterly, 28(3):330-350.

Epstein, Noel (1977). Language, Ethnicity, and the Schools: Policy Alternatives for Bilingual-Bicultural Education. Institute for Educational Leadership: George Washington University.

Erickson, Frederick (1978). The Politics of Speaking: An Approach to Evaluating Bilingual-Bicultural Schools. National Dissemination and Assessment Center, California State University, Los Angeles: Los Angeles.

Erickson, Frederic (1987). "Transformation and school success: The politics and culture of educational achievement". Anthropology and Education Quarterly, 18(4):335-356.

Fishman, Joshua A. (1989a). "Bias and anti-intellectualism: The frenzied fiction of 'English only'". Language and Ethnicity in Minority Sociolinguistic Perspective. J.A. Fishman, ed. Multilingual Matters Ltd.: Clevedon. pp. 638-654.

---------- (1989b). "Language and ethnicity". Language and Ethnicity in Minority Sociolinguistic Perspective. J.A. Fishman, ed. Multilingual Matters Ltd.: Clevedon. pp. 23-65.

Flores, William V., and Rina Benmayor (1997). Latino Cultural Citizenship: Claiming Identity, Space, and Rights. Beacon Press: Boston.

Flynn, Donna K. (1997). "We Are the Border: Identity, exchange, and the state along the Benin-Nigeria border". American Ethnologist, 24(2):311-330.

Francis, Norbert, and Phyllis M. Ryan (1998). "English as an international language of prestige: Conflicting cultural perspectives and shifting ethnolinguistic loyalties". Anthropology and Education Quarterly, 29(1):25-43.

Frankenberg, Ruth (1994). "Whiteness and Americanness: Examining constructions of race, culture, and nation in white women's life narratives". Race. S. Gregory and R. Sanjek, eds. Rutgers University Press: New Brunswick. pp. 62-67.

Gibson, Margaret A. (1987). "Punjabi immigrants in an American high school". Interpretive Ethnography of Education. G. Spindler and L. Spindler, eds. Lawrence Erlbaum Associates: Hillsdale, New Jersey. pp. 281-310.

Gibson, Margaret A. (1997). "Complicating the immigrant/ involuntary minority typology". Anthropology and Education Quarterly, 28(3):431-454.

Giddens, Anthony (1984). The Constitution of Society. University of California Press: Berkeley.

Gillborn, David (1997). "Ethnicity and educational performance in the United Kingdom: Racism, ethnicity, and variability in achievement". Anthropology and Education Quarterly, 28(3):375-393.

Gupta, Akhil, and James Ferguson (1997a). "Beyond culture: Space, identity, and the politics of difference". Culture, Power, and Place. A. Gupta and J. Ferguson, eds. Duke University Press: Durham. (1997b). Culture, Power, and Place. Duke University Press: Durham.

Hegmon, Michelle (1996). "Technology, style and social practices: Archaeological approaches". Technological Choices and Social Boundaries in Material Culture Patterning. M. Stark, ed. Smithsonian Institution Press: Washington, D.C. pp. In Press.

Irvine, Martha (1998). "'English for the Children': Latino parents among foes of bilingual study". The Arizona Republic, January 25.

Jones, Melissa L. (1998). "2 Tempe schools seek grants for dual-language method". The Arizona Republic, January 10.

Kearney, Michael (1995)."The local and the global: The anthropology of globalization and transnationalism". Annual Review of Anthropology, 24:547-565.

Kearney, Michael (1996). Reconceptualizing the Peasantry. Westview Press.

Lasken, Douglas (1998). "Californians have chance to ban bilingual teaching: Voters should end costly failure, promote English". The Arizona Republic, January 16.

Malkki, Lisa (1997). "National geographic: The rooting of people and the territorialization of national identity among scholars and refugees". Culture, Power, and Place. A. Gupta and J. Ferguson, eds. Duke University Press: Durham. pp. 52-74.

Martinez, Oscar J. (1988). Troublesome Border. University of Arizona Press: Tucson.

Martinez, Oscar J. (1994). Border People: Life and Society in the US-Mexico Borderlands. University of Arizona Press: Tucson.

McGroarty, Mary (1991). "What do we know about effective second language teaching?" Languages in school and society. M. McGroarty and C. J. Faltis, eds. Contributions to the Sociology of Language. Mouton de Gruyter: Berlin. pp. 107-116.

McNeill, Daniel, and Paul Freiberger (1993). Fuzzy Logic: The Discovery of a Revolutionary Computer Technology and How It Is Changing Our World. Simon and Schuster: New York.

Navarrette, Ruben Jr. (1998). "Bilingual education, 'Ugly Political Mood' clash". The Arizona Republic, January 31.

Ogbu, John (1978). Minority Education and Caste: The American System in Cross-Cultural Perspective. Academic Press: New York.

Ogbu, John U. (1987). "Variability in minority school performance: A problem in search of an explanation". Anthropology and Education Quarterly, 18:312-334.

Ortner, Sherry B. (1996). Class and the Jewish Question. American Anthropological Association, San Francisco.

Ortner, Sherry B. (1998). "Identities: The hidden life of class". Journal of Anthropological Research, 54(1):1-17.

Rosenfeld, Gerry (1971). "Shut Those Thick Lips!": A Study of Slum School Failure. Holt, Rinehart, and Winston, Inc.: Prospect Heights, Illinois.

Rouse, Roger. (1991). "Mexican migration and the social space of postmodernism". Diaspora, 1(1):8-23.

Sacks, Karen Brodkin. (1994). "How did Jews become white folks?" Race. S. Gregory and R. Sanjek, eds. Rutgers University Press: New Brunswick: pp. 78-102.

Sahlins, Marshall (1981). Historical Metaphors and Mythical Realities: Structure in the Early History of the Sandwich Islands Kingdom. University of Michigan Press: Ann Arbor.

Sanjek, Roger (1994). "The enduring inequalities of race". In Race. S. Gregory and R. Sanjek, eds. Rutgers University Press: New Brunswick. pp. 1-17.

Santo, Beverly G. (1994). "Creating boundaries, crossing borders: The lives of professional women in the Mexico-US Borderlands". Dissertation. Union Institute.

Santos, Sheryl L. (1998). "14 going to Mexico to learn about bilingual education". The Arizona Republic, January 9.

Scott, James C. (1990). Domination and the Arts of Resistance: Hidden Transcripts. Yale University Press: New Haven.

Shore, Chris (1994). "Community". The Blackwell Dictionary of Twentieth-Century Social Thought. W. Outhwaite and T. Bottomore, eds. Blackwell Reference. Blackwell Publishers Ltd.: Oxford. pp. 98-99.

Steele, Mark B. (1998). "Bilingual education program an expensive failure: Viewpoint". The Arizona Republic, January 23.

Urciuoli, Bonnie (1995). "Language and borders". Annual Review of Anthropology, 24:525-546.

Vélez-Ibáñez, Carlos G. (1996). Border Visions: Mexican Cultures of the Southwest United States. University of Arizona Press: Tucson.

Vélez-Ibáñez, Carlos G., and James B. Greenberg (1992). "Formation and Transformation of Funds of Knowledge Among US-Mexican Households". Anthropology and Education Quarterly, 23(4):313-335.

Vigil, James Diego (1997). Personas Mexicanas. Harcourt Brace and Company: Fort Worth.

Wendl, Tobias, and Michael Rosler (n.d.) "Frontiers and borderlands: The rise and relevance of an anthropological research genre". Frontiers and Borderlands: Anthropological Perspectives. M. Rosler and T. Wendl, eds. Institute of Ethnology and African Studies, University of Munich: Munich.

White, Jonathan B. (1996). Graduation Rate Study: Class of 1994. Arizona Department of Education: Phoenix.

Williams, Brackette F. (1989). "A class act: Anthropology and the race to nation across ethnic terrain". Annual Review of Anthropology, 18:401-444.

Willis, Paul (1977). Learning to Labor. Columbia: New York.

Wilson, Thomas M., and Hastings Donnan (1998). Border Identities: Nation and State at International Frontiers. Cambridge University Press: Cambridge.



1 Gibson's model of accommodation can be tested along these lines (Gibson, 1987).

2 Ambos Nogales, literally "both Nogales", refers to Nogales on both sides of the international border (santo, 1994).

3 This number may be an underestimate due to the sensitive nature of the query. Those who cross without documentation may be reluctant to state their true place of birth on official forms.

4 People that lives in Nogales, Sonora.

5 I should note here that my relatively dark complexion may play a role in her response. Although I am not Mexican-American, many people responded to me as if I were.

6 See Lisa Malkki (1997) for a lucid discussion of roots.

7 Marx's idea of class could be productively used in studies of this kind. For example, Vigil cites examples of Mexican-American families that feel they have "made it" when their income and social status improves, even if only to the lower-middle class level. A comparison to see if this feeling arises primarily when the family has their own business (i.e., become capitalists) as opposed to when they remain wage laborers, albeit making more money, would be useful.

8 ¿Qué trabajo es aceptable como académico del alumno?

9 Further investigation will help clarify this issue. Nonetheless, I resist using the terms "Spanish-speaking" or "English-speaking" when referring to this population, since it is unclear how many parents are bilingual. Furthermore, I cannot assume that language choice alone indicates the ethnic identity of the parents.

10 ¿Qué tan importante es para usted que su hijo/hija hable correctamente dos o más idiomas?

11 ¿Qué trabajo es aceptable como académico del alumno? ¿Qué deberían aprender nuestro hijos en la escuela?

12 By "fuzzy" I refer to a logic in contrast to traditional set theory, which forces people to strictly demarcate groups. A fuzzy category cannot be definitively bounded, but can still be considered a "set" (McNeill and Freiberger, 1993).