The Culture of Migration in Southern Mexico

[ Anthropology ]

The Culture of Migration in Southern Mexico

By Jeffrey H. Cohen

This book explores the complex constellation of factors that cause rural Oaxacans to migrate, the historical and contemporary patterns of their migration, the effects of migration on families and communities, and the economic, cultural, and social reasons why many Oaxacans choose not to migrate.



33% website discount price


6 x 9 | 207 pp. | 20 illustrations, 17 tables

ISBN: 978-0-292-70592-0

Migration is a way of life for many individuals and even families in the Mexican state of Oaxaca. Some who leave their rural communities go only as far as the state capital, while others migrate to other parts of Mexico and to the United States. Most send money back to their communities, and many return to their homes after a few years. Migration offers Oaxacans economic opportunities that are not always available locally—but it also creates burdens for those who stay behind.

This book explores the complex constellation of factors that cause rural Oaxacans to migrate, the historical and contemporary patterns of their migration, the effects of migration on families and communities, and the economic, cultural, and social reasons why many Oaxacans choose not to migrate. Jeffrey Cohen draws on fieldwork and survey data from twelve communities in the central valleys of Oaxaca to give an encompassing view of the factors that drive migration and determine its outcomes. He demonstrates conclusively that, while migration is an effective way to make a living, no single model can explain the patterns of migration in southern Mexico.

  • Acknowledgments
  • Introduction: Studying Migration in Oaxaca's Central Valleys
  • Chapter 1: The Household and Migration
  • Chapter 2: History, Trajectory, and Process in Oaxacan Migration
  • Chapter 3: Contemporary Migration
  • Chapter 4: Migration, Socioeconomic Change, and Development
  • Chapter 5: Nonmigrant Households
  • Conclusion: Migration in Oaxaca's Central Valleys and Anthropology
  • Appendix A: Characteristics of the Population by Community
  • Appendix B: Household Survey
  • Appendix C: Cultural Consensus
  • Notes
  • References Cited
  • Index

There are many ways to approach the study of migration. In this ethnography of migration in the central valleys of Oaxaca, Mexico (the intermontane region surrounding the state's capital), I will argue that a cultural model—that is, a model in which the decision to migrate is rooted in the everyday experiences of rural Oaxacans—is most useful. However, before I describe that model, I want to begin by offering two views of migration that come from two Mexican folk songs. The first is "Llegan los norteños" ("The Norteños Arrive"), by Guillermo Velázquez y los Leones de la Sierra de Xichú (García de León 2003). The song tells the story of migrants from the north of Mexico who move from their hometowns to cities in the United States in an effort to find wage labor. The refrain, from which the title is taken, establishes the daunting nature of the Norteño's lot:

Llegan los norteños masticando inglés, vuelvan a la fiesta, vuelvan a su tierra.
Se acaba la fiesta y a sigue la guerra, en busca del dólar se van otra vez.

[Here come the Norteños, chewing up English, they return for the fiesta, they return to their homeland.
Once the fiesta is finished, the war continues, looking for dollars, they go again.]

The Norteño's life is a never-ending process of movement across borders that are both geographic and social. He (the Norteño is typically a young man) does not learn English well enough to integrate fully into the U.S. system. At the same time, the urge to find well-paid work means he cannot stay in Mexico. The Norteño lives by moving between Mexico and the United States. He returns to Mexico, but only for fiestas and brief visits with family. Then he must again cross the border to continue the "battle," seeking out dollars.

The second example contrasts with Velázquez's song and comes from "Canción mixteca," by José López Alvaréz. Described as an "achingly beautiful anthem of the lonely Mixtec farmworker" (Magagnini 2002), "Song of the Mixtec" tells of the longing that the Mixtecos, a minority population in the state of Oaxaca, feel for their homeland when they are abroad.

¡Que lejos estoy del suelo donde he nacido!
Inmensa nostalgia invade mi pensamiento;
y al ver me tan solo y triste cual hoja al viento,
quisiera llorar, quisiera morir de sentimiento.
\!O tierra del sol! Suspiro por verte,
ahora que lejos yo vivo sin luz, sin amor;
y al verme tan solo y triste cual hoja al viento,
quisiera llorar, quisiera morir de sentimiento.
[How far I am from the land where I was born!

An intense sadness invades my thought;
I am so alone and sad, like a leaf shaking in the wind,
I want to cry, I want to die from these feelings.
Oh, land of the sun! I yearn to see you
now that I live so far away—without your light, without love;
I am so alone and sad, like a leaf shaking in the wind,
I want to cry, I want to die from these feelings.]

"Canción mixteca" is ubiquitous in Oaxaca. It is sung at parties and performances throughout the state and by indigenous as well as mestizo Oaxacans. The song describes a pull exerted by the homeland that is so strong, a migrant will die of loneliness and heartache if he cannot return. Rather than describing a chase for money, the song tells of the hold that geography and, by extension, traditional culture has on migrants as they move.

The images created in these songs are powerful and profound, but they are also unrealistic. We should not assume that Velázquez described a nation's experiences in his song of the Norteños or that López Alvaréz was any more accurate about Oaxacans in his ode to the Mixtecos. Nevertheless, the images in these songs are powerful, and they do reference certain kinds of experiences that characterize at least some of the outcomes that migrants talk about.

The term "Norteño" describes a kind of migrant who originates in northern Mexico and is drawn to the United States by the combined pull of a labor market that promises wealth and the push of local economies that promise little. Caught between failed local systems and the seduction of the United States, the Norteño fills a middle world that transcends borders but at the same time lacks roots in either Mexico or the United States.

The image found in "Canción mixteca" contrasts with that of the Norteño. For Mixtecos, and for all rural Oaxacans by association, the ties to homeland are more than important; they are a force that centers the migrant and gives him hope even when he is away. Unlike Norteños, Mixtecos are not chasing dollars; rather they are looking homeward, with a nostalgia that keeps them connected.

I have a reason for bringing up these two images as a way to begin this study. Often in migration studies, the analysis focuses on a specific and singular cause of movement. A Norteño-like model describes migrants as laborers searching for relatively high wages. The story finds its counterpart in the pull-push models of the migration economists that date to early research on the subject by Ravenstein (1889). In contrast, the essentialist framework provided in "Canción mixteca" suggests that tradition and geography are the critical determinative forces in a migrant's decision to leave.

The economist's model argues that the supply of underpaid labor in Mexico and the presence of higher wages in the United States drives migration between the two countries. Hometowns in Mexico lack infrastructure and opportunity, while in the United States, good-paying jobs go wanting. In response, rural Mexicans join the ranks of migrants entering the United States by the millions.

The successes of U.S.-bound migrants bring more individuals. The pool of potential migrants expands further as successful migrants return to hometowns with money and new ideas. Migration becomes a self-reinforcing process as more individuals join the flow of labor across the border, and so on (Massey 1990).

Nevertheless, people do not blindly follow migrants as they leave. A large percentage of any community remains "immobile" (at home) even as migration rates increase (see chapter 5). Thus, although a push-pull model based in labor market demand tells us something about one force behind migration, it cannot explain the variations encountered among migrants, their households, and their communities (see discussion in Massey et al. 1998, 45-50).

A model that argues that geography and tradition are critical forces in determining outcomes is no more satisfying (see critiques of essentialism by Mitchell 1995 and Watanabe 1992). To suggest that traditions, culture, and place drive people in their decision making maintains the fiction that we are at the mercy of superorganic forces beyond our control.

In such a system, migrants do not decide where they will go, what they will do, or how they will get where they are going. Instead, they respond to cultural influences and other forces that cannot be defined in economic or political terms. In other words, migrants follow certain patterns as they move, because they are rural, traditional folk, and that is what rural, traditional folk do.

Of course, migrants are not cultural automatons. Traditions do not drive migrants to make certain decisions, nor do inequalities in the labor market always determine migrants' final destinations. Migrants are individuals, and they bring certain qualities—personal strengths and weaknesses—to their decisions that include education, experience, and expectations for the future. Migrants are also members of households. They are embedded in social networks that are rooted in kinship and friendship, that connect households locally and beyond, and that are maintained through cooperative and reciprocal ties (see Cohen 1999). Finally, they are also members of communities, which further influence outcomes. And although regional patterns in Oaxaca are apparent, communities are unique. Each community is geographically different, with its own specific history, economy, and ties to external forces. Thus migrants make their decisions in response not only to their individual strengths but also to the strengths of their households and communities (Faist 1997; Fischer et al. 1997; Hammar and Tamas 1997).

Some migrants choose to embrace their families and households—remitting to support the group. Some choose to leave and sever ties with their households and communities. Some Oaxacans take the role of the Norteño and are forever chasing a pot of dollars and the promise of economic success. Others migrate, but always with the goal of return. Some migrants succeed, whiles others fail and disappear.

Mutersbaugh (2002) argues that Oaxacan communities will exile migrants and set up serious sanctions for sojourners who have traveled for too many years. Nevertheless, the majority of central valley Oaxacans who do migrate (and a surprising number never leave their communities of origin) elect to travel for no more than about a year total, and throughout that time they send money home to support their families and by extension their community.

For the typical Oaxacan migrant the decision to move is not uncommon or exceptional, whether he or she elects to travel to a destination in Mexico or to the United States. Rather, migration in the central valleys is pervasive and commonplace. Moore (1988, 96) describes migration as "part of a strategy for coping with economic change, and opportunity which depends on multiplex links being established between rural and urban areas." It is part of everyday life and perhaps best understood in terms of what is a "culture of migration."

By "culture of migration," I mean to argue, first, that migration is pervasive—it occurs throughout the region and has a historical presence that dates to the first half of the twentieth century. Second, the decision to migrate is one that people make as part of their everyday experiences. Third and finally, the decision to migrate is accepted by most Oaxacans as one path toward economic well-being.

A culture of migration captures how I understand migration to work in Oaxaca. The choice comes from the interplay of individuals, their households, and their communities, as well as national and international socioeconomic forces. To call migration in Oaxaca "cultural" is not to say it is some kind of hard-wired response—or an automatic reaction to a set of specific outcomes. Instead, migration is one response among many to patterns and processes that link households and rural communities to global labor markets, flows of goods, and personal demands. In other words, migration in Oaxaca is "deeply ingrained into the repertoire of people's behavior, and values associated with migration become part of the community's values" (Massey et al. 1998, 47).

Some Background

Over the last six years, I have investigated migration patterns in twelve communities in rural Oaxaca, Mexico. Oaxaca is a poor state in the south of the country that faces serious economic, environmental, and social challenges. It is home to a large pool of migrants, and some days it seems as though every town has lost a substantial number of its able-bodied young men and women to the seductive pull of the United States. In towns like San Juan Guelavia, about 60% of the community's households have sent members to the United States, and on average about 40% of the households in central valley communities include U.S.-bound migrants.

Nevertheless, according to the Instituto Nacional de Estadísticas Geografía e Informática (INEGI), Oaxacan migration to the United States is a small percentage of the overall flow of migrants from Mexico to the United States. The state was ranked sixteenth of thirty-one Mexican states in terms of migration. Furthermore, INEGI estimated that movements of rural Oaxacans remained relatively low through the year 2000, given national patterns (INEGI 2002a, 2002b).

INEGI found that 96.8% of the state's population of just over 3 million individuals above the age of five were in their natal hometowns in 1995. A relatively small group of Oaxacans (2.5%) was moving internally within the state, and 2.8% were moving either within Mexico or to another country. For those Oaxacans moving out of the state, 91.2% remained within Mexico's national border, and 8.8% (7,439 individuals) crossed into another country (INEGI 2002a, 2002b).

To understand where Oaxaca fits—to make sense of why migration matters even if the state ranks rather low relative to other states—I focused on understanding the history of movement, the impact of migration, and the importance of migrant remittances (the moneys that migrants return to their families) for rural Oaxacans from the central valleys region. I began this work in 1996, with three goals in mind: to contextualize Oaxacan migration, to define the place of ethnicity in migration outcomes, and to develop a model that will explain Oaxacan migration in relation to broader national patterns.

Over the years, I have met Oaxacan migrants who were home to attend fiestas, to celebrate weddings, or to mourn at funerals. I watched as these migrants quickly left to return to the United States (el otro lado, the other side) to jobs, new families, and new challenges. I sat in plazas with native-born men and women who looked as out of place in their hometowns as any foreigner. I even met the occasional child born to migrant parents and sent home to Oaxaca for a summer with grandparents. Some of these children spoke very little Spanish (communicating instead in English), had never lived in a rural setting, and had little sense of or interest in peasant life in Oaxaca.

However, I met few migrants who behaved like Norteños. In general, the men and women I encountered over the years were fathers and mothers with small children whom they had to leave behind in Oaxaca. These migrants boarded buses in Oaxaca City's second-class bus station, with tears streaming down their faces. Clutching small bags with a few changes of clothes and the telephone number and address of a relative, they began their trek north, seeking jobs in the United States. These were not solitary migrants hiding in the shadows; rather they were mothers and fathers, brothers and sisters, sons and daughters, struggling to balance the demands of their families against their own wishes for the future.

Oaxacan migrants work hard to balance their temporary sojourns to other parts of Mexico and the United States with the demands of family, home, and community. They make their trips to the United States as members of resilient social networks that develop from ties of kinship and friendship. In fact, few migrants travel to the United States without a destination in mind and a friend or relative to meet (see chapter 3).

The migrants I know are deeply committed to their families. They migrate to support children, siblings, and parents. They risk their health and their lives for the good of their families and households. Usually they do not leave for long. Instead, they return to their hometowns after a year or two (sometimes three) to farm, to serve in local government, and to regain or perhaps renew their self-image as valuable, honest, and hardworking citizens of their communities.

With so much migration in the news and with such an emphasis in anthropology on the study of migration, it is easy to forget that many Oaxacans never leave their hometowns. Over the years, I have encountered many nonmigrants, people who have not and will not migrate for one reason or another (see chapter 5). So yes, I find the typical, dislocated migrant, but more often I meet gente humilde (humble folk)—fathers, mothers, sons, and daughters who face a changing world with grace and dignity as they manage the need for migration against other demands in an effort to maintain family, home, and community.

In this book, I introduce you to rural Oaxacans from communities in the state's central valleys. I let these people tell you about their experiences in their own words. First, I want to introduce you to the state and to describe some of the theoretical models that are used in the study of migration, including the household model I have developed. I also share with you why I believe anthropology is well positioned to interpret migration outcomes and remittance use.

The Central Valleys of Oaxaca, Mexico

Oaxaca is a unique place. Along with Chiapas, Oaxaca is one of Mexico's most ethnically diverse states. One reason I chose to do this work in Oaxaca was to discover whether there were differences between the migration experiences of indigenous and mestizo (nonindigenous) Mexicans. The communities that make my sample include Zapotec, Chinantec, and mestizo villages.

Given Oaxaca's status as one of Mexico's most "indigenous" places, it should come as little surprise that it is a poor, rural state. Oaxaca holds a distinctive spot at the bottom of most economic indicators such as per capita income, industrialization, and employment. The infrastructures of most rural communities are fragmentary at best, and basic services like running water, sewerage, and phone service remain difficult to find outside the state's capital. Oaxaca also ranks poorly in terms of health and social status indicators, with a high infant mortality rate and a low literacy rate, for example. Nevertheless, Oaxaca is not an isolated place. It is not a timeless world where Indians follow ancient rituals and calendars. Rather Oaxaca is part of the growing global capitalist system linked through tourism, development, education, entertainment, and migration.

The central valleys comprise the intermontane region that surrounds Oaxaca City (the state's capital) and includes the Centro, Etla, Ocotlán, Tlacolula, and Zimatlán districts. Communities in the central valleys are relatively well off when compared with those in the rest of the state. In general, communities in the central valleys are linked to Oaxaca City through bus and taxi service. The local economies of these communities are tied to the city, which attracts thousands of day laborers (accounting for at least 10% of the city's workforce—INEGI 1999). The city offers opportunities for education and is an important tourist destination for foreign travels.

I chose to work in the central valleys for several reasons. First, I am most familiar with the central valleys region of the state—particularly the eastern arm of the valley where Santa Ana del Valle is situated (see Cohen 1999). Second, the presence of mestizo and indigenous communities in the valley meant that I could examine the role of ethnicity in migration outcomes. Third, the differences that separate central valley communities are not so great as to make their comparison difficult.

In general, these are rural villages where small-scale agricultural production is found alongside craft production and limited wage labor. Communities from more isolated regions of the state are much different and lack access to the city and its opportunities (see Kearney 2000).

I chose eleven central valley communities for this study from a randomized list that Dr. Martha Rees of Agnes Scott College created for the central valleys region (see table 0.1 and map). The communities are located in one of the three branches of Oaxaca's central valleys (the Etla, Tlacolula, and Ocotlán/Zimatlán valleys) with the exception of San Pedro Ixtlahuaca, which is in the Centro district, 10 kilometers west of Oaxaca City.

The Etla valley extends from Oaxaca City on an axis from west to northwest and includes the communities of San Pablo Huitzo (31 km from the capital), San Juan del Estado (27 km), and Guadalupe Etla (19 km). South of the city the communities of San Martín Tilcajete (23 km) and Santa Inés Yatzeche (40 km) are in the Ocotlán/Zimatlán valleys. Finally, to the east and in the Tlacolula valley are Santa María Guelacé (23 km), San Juan Guelavia (37 km), Santa Ana del Valle (34 km), Villa Díaz Ordaz (40 km), San Lorenzo Albarradas (68 km), and San Juan del Río (80 km).

From afar, these dozen communities look very much alike. Brick, adobe, and cement-block homes of one or two stories with red tile roofs fill villages that radiate in standard block grids from the central plaza. The plazas are constructed around central churches, governmental buildings, and small market areas. Often the plaza includes at least one basketball court where competitions are held nightly, a band shelter, and other public spaces. Circling the communities are dusty farmlands that look tired and overused to the untrained eye. Nevertheless, land in the central valleys is fertile, and households in our survey produced enough maize to support themselves for six months to one year from nonirrigated fields (described as temporal, or rain-fed, lands) that averaged about 1 hectare in size.

Central valley communities share many demographic and socioeconomic attributes. These communities have experienced dramatic increases in population since the 1950s. The total population for the twelve communities in our data set grew from 19,254 in the 1950 census to 33,261 in the year 2000 (INEGI 2002b; SEN 1953). The increase in population has come with a rise in the demand for wage labor, schooling, services (electricity, running water), and medical care. Unfortunately, the infrastructures of these communities remains underdeveloped, and the market for labor is limited. In other words, there are few opportunities for wage labor, few doctors, poor schools, and limited access to market goods—all important motivations for migration.

A review of work and wages in the state illustrates the challenges facing most Oaxacans. The Mexican government defines a living wage as two times the daily minimum. In Oaxaca, the daily minimum has hovered around US$5 for the decade of the 1990s. Surveys by INEGI note that on average 80% of the households in these communities make no more than twice the minimum—in other words a living wage (INEGI 2001a). For specific communities, the percentage of households making less than twice the minimum ranges from 55% and 59% in San Pablo Huitzo and Guadalupe Etla, respectively, to 93% and 94% of the households in San Juan del Río and Santa Inés Yatzeche. In other words, in San Juan del Río and Santa Inés Yatzeche only 6 to 7% of local households make more than twice the minimum wage (table 0.2).

An average of 51% of the adults over the age of fifteen had not completed primeria (the first six years of primary school). Men had completed an average of about half a year more of school than women in the communities surveyed (on gender and education, see Kowalewski and Saindon 1992). Guadalupe Etla had the highest education rate, with 78% of its adults completing primeria. At the other extreme, 70% of the adults in Santa Inés Yatzeche had not completed the six years of compulsory education mandated by the government (table 0.3).

Educational opportunities are limited to primary school, although San Pablo Huitzo and San Pedro Ixtlahuaca are home to telesecundarias (closed-circuit high schools). Students interested in additional education must travel to nearby cities or the state capital. Health care is lacking throughout the state, and only 23% of the state's population have direct access to health care (INEGI 2002a, 2002b). Health care in the communities we surveyed included casas de salud (health clinics) that are part of the national health care system (Secretaría de Salubridad).

Infrastructure in the central valleys, including electrification, is problematic at best, and access to basic services like water and sewers continues to lag (table 0.4). Where improvements occur, they are largely self-funded or funded through a combination of local and state monies. To cover the costs of development, village leaders assess fees for households in their community. These funds, called cooperación, pay for projects and programs for which there is no, or only limited, state funding. Cooperación is one dimension of the traditional model of social organization and control that Oaxacans rely upon and that are found in most communities. The state describes this system as usos y costumbres (literally "uses and customs," but more accurately translated as "traditional practices"), and the system contrasts with the party politics of larger cities and Mexico in general (see Fox and Aranda 1996).

In addition to cooperación, traditional patterns of reciprocity, cooperation, and community participation are defined by tequio and by service in the cargo system. Tequio is communal labor organized by a community's leaders. Leaders can call for tequio at any time of the year to cover the labor needed for projects and programs in a village. Tequio depends upon households to contribute one worker each to projects that can range from the simple to the complex. Households typically send members to participate in tequio at least once a year, or they sometimes hire a replacement to cover their commitments. Community leaders can impose sanctions on households that fail to support tequio or that fail to send members to serve in labor brigades.

The cargo system is at the heart of village politics in rural Oaxaca. Cargos are the burdens that members of households must endure to maintain their household's status in their community (Cancian 1965). The cargo system contrasts with politics that are administered por partidos (party based). Households participate in the cargo system voluntarily. However, town leaders typically exert intense pressure upon household heads to send individuals to serve (see Cohen 1999). The response to noncompliant households includes sanctions that range from fines to expulsion.

Individuals volunteer or are nominated to positions in the system and serve terms of one to three years, depending on the nature of the cargo. In general, one adult member of any household in a community must serve in a committee every other year (although many exceptions are made to service). In the past, service was restricted to adult males. However, partly in response to the growing number of men who have migrated from their home communities, adult women are now serving in cargos.

The system of cargos includes a series of hierarchically arranged committees and positions that organize the political and civil life of the village. Three key committees typically occupy the highest positions in the hierarchy: the comite del pueblo, bienes comunales (common lands), and the comite del templo (church committee). The presidente municipal (village president) chairs the comite del pueblo and manages a board that usually includes seven suplentes (board members), a tesorero (treasurer), a síndico (organizer), and secretario (secretary). Bienes comunales is the second high-ranking committee that manages a community's natural resources. Finally, the comite del templo manages the spiritual life of the village, caring for the village's church and saints and planning rituals throughout the year. Dozens of committees follow these in descending rank and status, ranging from school committees (like the PTA in North America) to committees concerned with transportation, roads, water, utilities, and so forth.

Even with the common patterns and practices noted above, there are important differences among the various communities. The most obvious difference divides communities in terms of their ethnic makeup. San Juan Guelavia, Santa Ana del Valle, Villa Díaz Ordaz, San Juan del Río (all in the Tlacolula branch of the valley), and Santa Inés Yatzeche are indigenous communities with populations that continue to speak Zapotec in the home.

A recent survey was conducted by the Dirección General de Población de Oaxaca y el Consejo Nacional de Población (DIGEPO) to examine rural marginality, as measured by literacy rates, access to running water, access to health care, and employment patterns. The survey discovered that indigenous communities (including all of the indigenous villages in the present study) were marked by extreme socioeconomic marginality. Mestizo communities ranked low as well, but none scored as poorly as did indigenous communities. This is one indication that ethnicity correlates with poverty in rural Mexico, and indigenous communities are at an increased disadvantage.

Craft specialization is a second factor that divides these communities into different camps. San Juan Guelavia, Santa Ana del Valle, San Martín Tilcajete, San Lorenzo Albarradas, and Villa Díaz Ordaz are linked to the local craft market and the global tourist industry in profound ways, even though farming remains central to household survival.

San Martín Tilcajete is perhaps the best known of the craft-producing communities we surveyed. San Martín Tilcajete is the home of Alebrijes, brightly painted wooden animals and zoomorphic figures that are extremely popular tourist items (Chibnik 2001). The community is prosperous, and signs of the booming market for wood carvings are everywhere, from the refurbished village plaza that includes a covered basketball court to the numerous two- and three-story homes that are wired for satellite television. Wood carving involved 57% of the households we surveyed in San Martín Tilcajete. The income that households are able to earn from production and the sale of goods on the market is strong enough to slow out-migration. One carver described the situation to us in a matter-of-fact fashion:

I can earn plenty right here, and maybe in the U.S. I could earn ten times more, but it costs too much and there are too many risks. So why would I travel to California and spend 10,000 pesos, when I can stay right here?

Salvador Jimenez, San Martín Tilcajete, January 2002

Santa Ana del Valle and Villa Díaz Ordaz are in a different situation. Both communities participate in the production of woolen textiles for sale in local tourist markets and for export. However, unlike San Martín Tilcajete, these communities do not dominate production. Instead, the textile production in these towns follows a contract-labor model that the weavers described as mano de obra (piecework).

Buyers and intermediaries from the community of Teotitlán del Valle dominate the market and control prices, sales, and production (see Stephen 1991). In other words, craft producers from Santa Ana del Valle and Villa Díaz Ordaz do not control the production or sale of their goods. I found that 56% of Santa Ana del Valle's weavers work on contract with intermediaries from Teotitlán (Cohen 1999, 48). The percentage of artisan households in Villa Díaz Ordaz is much lower (13%). Households in Villa Díaz Ordaz lack access to the market, and problems with access likely limit production. Weavers in Santa Ana del Valle and Villa Díaz Ordaz do not earn enough from their work to make migration more of a choice and less of a necessity. In fact, weavers in both towns described craft production as little more than a means to an end.

San Juan Guelavia and San Lorenzo Albarradas also produce crafts, canastas (baskets) in the former and petates (reed mats) in the latter. However, these crafts are sold almost entirely on the local market. Except for the petates that tourists buy to use at the beach, the mat is an item that rural Oaxacans buy to use for a bed. San Juan Guelavia's baskets appeal to a local market as well. The townsfolk make baskets that are primarily used for hauling corn, groceries, and goods from one place to another. The baskets are not generally marketed to tourists, and they tend toward the utilitarian. Unfortunately, the market for handmade baskets has come under pressure from ready-made plastic containers and bags, and even though 38% of the households we surveyed produced canastas, no one earned a living wage from that work. Rather, basketmaking was something that Guelavians did outside of farm labor as a way to supplement income. The rising cost of supplies for the baskets is also a stress on business. Don Epiphanio Garcia described the situation for us:

We used to be able to go to the river [Río Salado] and just take all of the cane [caña] we wanted. But it has disappeared. Now we have to go to Pochutla or somewhere else to buy cane, and it is expensive. A load costs a peso per stem! Many of us are giving up—there is no market, and it is just too much money to produce.

Epiphanio Garcia, San Juan Guelavia, June 2000

San Lorenzo Albarradas has a second way of tapping into the tourist economy. The community is home to natural mineral springs called Hierve el Agua (the name literally means "boiling water"). Working with several nongovernmental organizations (NGOs), the town is involved in a series of programs to develop its natural resource.

Santa Inés Yatzeche and San Juan del Río are in much more precarious economic positions. Their local economies are defined by farming and little else. Neither town is home to a craft tradition, and the land in San Juan del Río in particular is of marginal quality for farming. It will come as no surprise that DIGEPO's survey found indicators of marginality high in both communities. Santa Inés Yatzeche benefits somewhat from its proximity to Zimatlán, and households can sell produce there. Nevertheless, few jobs are available in Zimatlán, and most of Santa Inés Yatzeche's populace must travel to Oaxaca City if they want to find local wage work.

Santa María Guelacé is also a town defined by farming, with no craft production present. However, unlike Santa Inés Yatzeche and San Juan del Río, Santa María Guelacé has a good portion of irrigated land, and its proximity to Oaxaca City makes it much easier for its population to find work. The community is also home to ajieros (garlic producers), who sell their goods to restaurants in the city.

Urbanism is also an important marker of difference in the valleys. Guadalupe Etla, San Pablo Huitzo, San Juan del Estado, and San Pedro Ixtlahuaca are larger, urban centers with more dynamic local economies, and they stand apart from the rest of the communities surveyed for this study. San Pablo Huitzo serves as a minor market center for villages that surround it, as does San Juan del Estado. San Juan del Estado's lands support a small lumber industry and a stone quarry. Finally, Guadalupe Etla has become a bedroom community for urban Oaxacans who are looking for a suburban lifestyle.

San Pedro Ixtlahuaca is the closest to Oaxaca City of the towns we surveyed. Only 10 kilometers due west of Oaxaca, the community is tucked below the archaeological site of Monte Albán and has rich agricultural lands. Like Guadalupe Etla, San Pedro Ixtlahuaca is home to a growing community of urban Oaxacans who are leaving the city. In San Pedro Ixtlahuaca this change has led to some tensions as newcomers and established families struggle over the value and importance of traditional political practices.

There are differences in the migration rates for each community. These differences are discussed in detail in chapter 1, but here it is important to understand, first, that migration is not uniform across the communities and, second, that it is not homogeneous across a community's households. Migration in rural Oaxaca on average involves about 47% of a community's households, but there is a great deal of variation from one community to the next. Similarly, each community has households that cannot or will not migrate.

Understanding Mexican Migration

"Migration" is a term that social scientists use to define and describe movement by human populations. Migration does not occur in a vacuum. People do not migrate because they must. We are not animals that have some deep-seated need to complete a circuit in response to some biological drive. Rather, humans migrate because they can. People make decisions to migrate in response to desires, lifestyles, resources, and needs.

To get an idea of where Oaxacan migration and Mexican migration in general fit into this process, let me review some basic facts about Mexican-U.S. migration. Millions of Mexicans live legally and illegally in the United States. Van Hook and Bean (1998) estimate that just over 7 million Mexicans live in the United States and that 2.35 million of those Mexicans were in the United States without authorization (see also Lozano Ascencio 1998, 1209). Oaxacans are a small group within this larger population. INEGI (2001b) estimates that Oaxacans account for no more than about 4% of the total migrant population currently in the United States—or nearly 100,000 individuals. In contrast, nearly 70% of Mexico's migrants come from just ten states: Michoacán, Guanajuato, and Jalisco (three traditional sending regions), along with Zacatecas, Durango, Mexico City, Chihuahua, Tamaulipas, Guerrero, and the state of Mexico (Bustamante et al. 1998, 116). There has been little overall change in the makeup of this population over the last thirty years, according to Marcelli and Cornelius (2001). Readers might ask, why should we be so concerned with Oaxacan migrants in the United States? The answer comes in at least two parts.

First, even though Oaxacans account for a small percentage of the total migrant population living in the United States, the economic effects of their moves are profound for Oaxaca, amounting to at least US$11 million returned to the state during the 1990s alone (Lozano Ascencio 1993). Second, there are positive and negative social costs of movement, and the debate continues over just what migration means for rural Mexican in general. The history and outcomes of migration for rural Oaxacans from the state's central valleys offer an important comparison to better-known sending regions.

Migration and Remittances

Migration is not just about moving across the landscape. Migration is also about sending money home, or remitting. A second theme of this book focuses on what remittances mean for rural Oaxacans, and the positive and negative impacts that remittances can have on rural society. Mexican migrants (including migrants from Oaxaca) remit as much as US$3 billion annually to families in Mexico (Lozano Ascencio 1998, 1192), an amount that equals or exceeds the income generated by agricultural exports and tourism (see Lozano Ascencio 1993, 64, figure 5; Russell 1992; Taylor et al. 1996b).

Remittances can be used to positive ends (see Orozco 2002). Richard Jones (1998, 4) notes that remittances in Mexico are "safety nets for poor regions left behind by the agglomerative behavior of international capital, by the preoccupation of the international community with other matters, and by the indifference of their own government" (see also Bustamante et al. 1998 and Taylor 1999). However, migration is more than a process that leads to economic change for rural communities. It is also a process that has social and cultural costs and benefits that individuals initiate as members of households and communities. Thus it is probably not surprising that the debate continues over how best to explain outcomes of migration and the use of remittances for rural peasant households and their communities.

Although social scientists use many models to explore migration (see, for example, Brettell and Hollifield 2000), two competing ideas—development and dependency—dominate much of the debate. Those who favor dependency models focus on the socioeconomic costs of migration, whereas proponents of development models point toward the economic growth that comes from remittance use. Dependency models argue that migration exacerbates local socioeconomic inequalities and drives unproductive consumption within migrant households while creating pools of cheap labor waiting to be exploited (Reichert 1981). The result is a place where the population, in its quest to find the money necessary to purchase the goods it now wants, becomes addicted to migration. A population directs its energies not toward internal balance and progress but toward external markets, because the community lacks any kind of infrastructure that can support local labor and the creation of local market outlets. In other words, rural communities are dependent on distant centers of power for jobs and goods. Rural communities caught in this kind of a web become little more than nurseries for the young (future migrants) and "homes" for the elderly (those no longer able to migrate). The outcome of this process is the social disintegration of sending communities: the able-bodied residents are siphoned away by the pull of job opportunities and the disruption of local practices, as remittances are wasted (see Brana-Shute and Brana-Shute 1982; Diaz Briquets 1991; Guidi 1993; Martín 1991; Papademetriou 1991; and Rubenstein 1992).

Researchers who argue for development models emphasize the benefits of migration and the potential positive outcomes as remittances flow back to rural hometowns where there are few opportunities and even fewer wage-based jobs (Taylor 1999, 73). There is strong evidence that "migradollars" (the dollars generated through transnational migration) can foster economic growth nationally and locally (Durand et al. 1996a, 1996b; Smith 1998; Taylor et al. 1996a, 1996b).

Remittances are critical to national economies, and they are an important source of foreign exchange. The funds sent home by Mexicans living and working in the United States become the hard currency that the Mexican state needs to balance its deficits (Massey et al. 1998, 232). These funds also drive the expansion of the national economy; as a 1990 report indicated, "each migradollar entering Mexico ultimately produced a $2.90 increase in Mexico's Gross Domestic Product and raised output by a total of $3.20" (Durand et al. 1996a).

Nevertheless, these are national outcomes. At a local level, the impact of remittances is more varied. The majority of the remittances that are returned to central valley households go to covering the costs of household maintenance (see de la Garza and Orozco 2002: 37). Investments and savings, though they do occur, come only after households meet their basic expenses. This means that although remittances can become the basis for what informants describe as the self-advancement of their households, families, and villages, in general they do not. When remittances are invested, they underwrite the expansion of services, such as water and electricity, and the support and revival of community rituals. The money returned also helps families cover the costs of participating in the political life of their village (see Orozco 2002 and Smith 1998).

Remittances also carry costs for households and communities. Remittances can increase social inequalities that are rooted in local socioeconomic differences. As some households choose to migrate, others will not or cannot migrate. The result is that wealthy households that can afford migration's costs grow wealthier as they succeed. Households that cannot afford migration's costs are relatively impoverished in response. Furthermore, remittances will tend to decline over time and as migrants are away from their homes longer. Lowell and de la Garza (2002, 20) note that although 60% of all temporary Mexican migrants in the United States remit, the total dollars returned decline as migrants age and as the total number of years spent away from home increases.

A Household Model of Migration

To understand migration decisions and the use of remittances in Oaxaca, I use a three-part, household-based approach. This approach contrasts with macroeconomic models that focus on regional or sometimes national patterns in an effort to understand broadly based patterns of movement. A household model also contrasts with psychological models that focus on the migrant and explore migration and remittance outcomes from the perspective of the individual actor. Although both approaches have their strengths (macroeconomic models help us understand global patterns of movement, for example, while psychological models help us to define what qualities make for a successful migrant), neither model adequately addresses the social universe that defines migration for rural Oaxacans. An emphasis on the individual also causes trouble in the analysis of Oaxacan movement and misrepresents the ways in which households and communities inform how migrants define their social world.

To ignore the important role of the household is to misunderstand how rural Oaxacans create their social universe. The point is not to suggest that individuals are not decision makers, nor is it to argue that all migrants make their decisions in consultation with their household or in deference to communal concerns. In fact, in each community I have visited there are examples of individuals who turn their backs on families and communities and sever ties with their hometowns against the will of their parents and spouses. There are also examples of entire families or households who have left their communities. Nevertheless, the outcomes for the migrants who decide to sever ties and ignore their household's wishes are serious for those members of the household who are left behind. A household approach helps us capture this process and gain a better understanding of variation in outcomes. Finally, a household model reminds us that migration is not solely a process that pulls individuals to new labor markets so that they can improve incomes in relation to some abstract, distant social standard. Rather, as Massey et al. argue (1993, 438), migrants wish "to increase income relative to other households, and hence, to reduce their relative deprivation compared with some [well known and local] reference group."


My approach, which defines migration as a decision rooted in the household and seeks to describe and predict patterns and outcomes for a region, has important ramifications for fieldwork and analysis. First, as pointed out above, it means we are not concerned so much with individual variation as with how that variation is rooted in the overall survival and status of the household or domestic group. Second, because we are interested in explaining variation, we must consider more than a single community's experiences. A focus on one community or population would make it difficult to define variation and to discover the various factors that influence and predict outcomes across space and time.

Therefore this study began by randomly selecting eleven communities from throughout the central valleys (with Santa Ana as a twelfth site). Because I was working in so many communities and because it was crucial to define large samples in each community, I could not do this work independently, in the traditional anthropological manner of the lone ethnographer. Instead I used a team of fieldworkers whom I helped train and who were coordinated through the Instituto Tecnológico de Oaxaca. Team-based research was a challenge for me, but it became an effective model that allowed for the definition of a large data set supporting both ethnographic and statistical analysis of migration outcomes.

Once I had selected communities for the project (with the help of a randomized list created by Martha Rees), the team began work in earnest. Over two summers we collected surveys in a randomly selected sample of 590 households, or about 15% of the households in each community. A household approach meant that we collected data on all the members of the domestic unit. In rural Oaxaca, the household could be difficult to define. In general, most rural households (63%) were nuclear units living in independent compounds, quite like their U.S. counterparts. In other words, the household included members of two generations living in a single homestead and pooling the resources and skills of its members. The senior generation consisted of a legally married couple; the junior generation included the offspring of the seniors. However, some households in the central valleys included more than two generations and were better thought of as extended units (35%). Typically, we found extended units organized around a married couple and their children, with the addition of a grandparent. Sometimes a household appeared to be an extended unit but was in fact a series of independent nuclear units that shared a common area or patio. Finally, some households defied classification (2%) and included odd mixtures of members. One memorable "other" was a household in San Juan del Estado that included three brothers in their late sixties. These brothers made their living by selling carbon (charcoal) and doing limited farmwork. They had no relatives in the community and in many ways existed as hermits.

We used maps from INEGI and plotted locations of potential households in each block within a community, ignoring blocks with no homes. We rotated the selection of households, moving from the northwest to the southwest, southeast, and northeast corners of consecutive blocks. Once we identified a location, a fieldworker received his or her map and began in the northwest corner, selected the first household counterclockwise from that corner, and knocked on the door to ask if he or she could conduct a survey, explaining our project and why it was important. If members of a household refused to participate, the fieldworker proceeded to the next available house, going in a clockwise direction, and began the process again. In the event that no households in a block participated, the fieldworker moved to one in a series of "safe blocks" that were set aside for just this problem.

The survey included several sections and identified members of the domestic group, as well as their work, migration experience, land use, household organization, consumption patterns, and community participation. The first sections of the survey focused on household membership and organization (part 1), work (part 2), and migration experience (part 3). We assigned each household a unique code and described its members according to age, gender, civil status, place of birth and current residence, languages spoken, and education. We recorded work histories for all members involved in household maintenance, with attention to nonwage and informal labor (particularly among women) that is typically central and crucial to the domestic group's survival. We asked individuals to recount as many labor activities as they could remember and to identify how they combined various activities (farming and wage labor, for example) to meet the needs of the domestic group.

We identified migrants as we created inventories of a household's members and their activities in parts 1 and 2. Part 3 of the survey focused specifically on international and transnational (that is, back and forth) migration. In this section of the survey, we endeavored to gain a clear count of the total number of migrants in a household, as well as the number of trips members had taken. This was typically where we identified individuals who left their households and who no longer actively participated as members. Migrants described their experiences, and we asked them to note their destinations; with whom they traveled; how they organized money to cover the expenses of border crossing; their work in the United States; with whom they stayed once they were settled; and their remittance history.

Next, in parts 4, 5, and 6, we asked about agriculture, household expenses, and housing. We created an inventory that included animals, goods and appliances, construction materials, and access to water and utilities (parts 4 and 5). We asked about weekly expenses for food, utilities, transportation, education, entertainment, and health care and how members covered those expenses (part 6). The last part of the survey focused on household members' participation in the social life of their village (part 7). We noted their political service, their participation and sponsorship of local rituals, and the reciprocal relationships they held with other families. Finally, we asked about their investment of time, money, and effort in village projects and programs.

We elicited detailed, personal responses on local social life, migration, and the structure of community by using open-ended questions built into the survey. After completing the surveys, we conducted follow-up interviews with community leaders and key informants to further document unique experiences. We also collected oral histories in each community. Fieldworkers identified informants for extended interviews and oral histories. In the final observation section of the survey, we noted if interviewees were "good talkers" and interested in sharing more of their experiences with us. We combined the surveys, interviews, oral histories, and additional archival work in Oaxaca City to create an ethnographically rich and dynamic picture of migration in the central valleys and to help illustrate how migration emerges from and interacts with local socioeconomic processes.

A second implication of a household approach is that it allows me to define migration as a stage-specific and predictable process that is influenced by the structure of the household (its members, their ages, their resources), local practices (sociocultural norms), regional economic trends, and macroeconomic forces. If migration follows specific stages, it means that outcomes are patterned and identifiable over time. In the case of Oaxacans from the central valleys, migration has peaked with each of Mexico's economic crises, and overall, sojourns to the United States have dramatically increased over the last two decades (see chapter 1).

The resources that are available to a household and its members can greatly influence the decision to migrate. Beyond the household, a community and its leadership also have some bearing on migration decisions and remittance use. Rural Oaxacan households maintain their status and standing in their communities through participation in a series of community-defined activities that include tequio, servicio, and cooperación. Households must also respect the demands that a community's leaders place on the population—the demand for participation in leadership, the demand for funds to support development projects, the demand that a migrant return home annually to avoid expulsion of his or her family (see Mutersbaugh 2002).

In addition to a community's resource base, access to regional markets, education, entertainment, and—perhaps most important—jobs has a great effect on migration outcomes and possibilities. Some of this effect is intuitive. Access to jobs in the state's capital, Oaxaca City, means there are local opportunities that Oaxacans living in isolated mountain communities lack. But resources also mean land and ways in which land can be used, and one point that will become clear is that a household's resource base is sometimes so strong as to make migration irrelevant—or, alternatively, the lack of resources can put migration out of reach for poorer households.

Regional demand for labor works both with and against international demand. The pull of jobs in the United States is a strong motivating force for Oaxacans seeking the money they need to feed their families, to buy consumer and luxury goods, and perhaps to start a new business. Through the early 1980s the seductive pull of higher wages came largely from Mexico City and other boomtowns within the nation. Now that pull is largely from the United States; nevertheless, the promise of wages is not in and of itself a force that will build migration. Rather, migrants and potential migrants also think about the dangers of the border and their reception once they have arrived. For some rural Oaxacans, fear of the border is more than enough to limit their desire to migrate, no matter the promise of wealth.

The third and last part of my approach defines the decisions to migrate and to use remittances as progressive (that is, made with the goal of satisfying the needs of the household), even though the outcomes of migration and remittances remain hard to predict. In other words, the decision to migrate is based in the experiences and strengths of the decision maker who "classifies the various alternatives in his [or her] subjective environment as to their expected outcomes, whether satisfactory or unsatisfactory" (Wolpert 1964, 544). Thus the migrant is an active agent who calculates within his or her abilities how to respond to opportunities and challenges. Migration becomes an option, not a given, and it is certainly not a process entered into blithely and with little prior knowledge.

These three points of investigation (the household, the stages of migration, and the progressive nature of decision making) resolve some of the contradictions that dependency and development models create. First, they allow us to move beyond the kinds of moralistic arguments that tend to dominate much of the debate on migration. The question of whether migration is good or bad is left behind, and we can instead focus on the patterns (whether local, regional, or global) that define and predict migration and remittance use outcomes. Second, the household model allows us to place individuals into their social milieu in a way that builds upon ethnographic analysis. Finally, because we define migrants as largely rational, we can better understand how they organize their resources and strategize to succeed.

Structure of the Book

Chapter 1 introduces the rural Oaxacan household and details why it is important to understand migration as part of a household's overall strategy for survival. Chapter 2 begins with a discussion of the history and geography of Oaxaca and the central valleys region. The history of movement for the area is reviewed to reinforce the point that although migration has increased rapidly over the last two decades, it is not a new process. I also review how Oaxacan migration to the United States links with local circuit moves (commutes from central valley communities to Oaxaca City) and movements within Mexico. Finally, the chapter details many of the factors that specialists believe are important predictors of migration. This list becomes important in chapter 3, where contemporary migration in the central valleys is discussed.

Chapter 3 focuses on contemporary migration outcomes. I share some of the false starts that plagued my team as we tried to understand the data we collected. The discussion also shows how we developed working models that bridge ethnographic and quantitative data to explain the outcomes of migration in the central valleys. I begin with a discussion of migration in general and why it is hard to aggregate the data. Three sections follow that focus on various kinds of moves, with examples from specific communities. My goal is not only to describe Oaxacan migration in detail but also to point out why we cannot focus solely on migration to the United States if we hope to understand local patterns of movement.

Chapter 4 examines the socioeconomic and cultural outcomes of migration and remittance use and the costs and benefits of movement. I focus on three areas: the revival of traditional celebrations that are paid for by remittances from migrants; the continued importance of traditional practices for the organization and maintenance of central valley communities; and strains that migration places on local communities.

Chapter 5 describes rural Oaxacans who do not migrate. One group of households does not migrate because the costs and risks are just too high and too great to make migration an option. The second group of households does not migrate because they are effectively able to maintain their households and cover any other kinds of expenses in a way that makes migration pointless. Land-poor, socially isolated households in San Juan del Estado illustrate those rural Oaxacans who cannot afford the risks of migrating. Dairy producers in Guadalupe Etla and craftspeople in San Martín Tilcajete illustrate the other extreme—Oaxacans who do not need to migrate, because they are doing well.

The conclusion returns to the question of how best to study migration and anthropology's role in migration studies. I argue that a household approach focused on domestic groups and their organization over time allows for a powerful analysis of migration and remittance use outcomes by articulating micro (ethnographic and detailed) data with macro (more generalized) data and models of migration outcomes (Brettell 2000). A second goal is to answer the question of whether a "culture of migration" exists in the central valleys. The evidence presented indicates that Oaxacan migration is embedded in a series of sociocultural patterns. Rather than destabilizing or undermining local cultural patterns and social processes, the decision to migrate can often support and even sometimes invigorate those patterns and processes. The outcome, then, is a culture of migration, a system in which migration is integrated and integral to ongoing sociocultural development.

Oaxacan migration is also embedded in global socioeconomic processes that include migration, tourism, education, market expansion, entertainment, health care, and so forth. There is no reason to think about Oaxaca as isolated—or as a home to native peoples who cannot cope with a changing world. The example of rural Oaxacans and their responses to global capitalism shows that there are better ways to describe natives and rural peoples. Studying the ways in which they respond to globalization accomplishes two important goals. First, it shows us that they are not victims of modernity but rather are people responding to the world and its inequalities in the best way they can. Second, the analysis continues to push anthropology beyond its romantic roots and toward a future where it can matter as a field.


By Jeffrey H. Cohen

Jeffrey H. Cohen is Associate Professor of Cultural Anthropology at Ohio State University.

"This book is important to migration studies for a number of reasons. First, I really like the presentation of the complexity of motives for migrants and households.... Second, the emphasis on the anthropological approach, focusing on a holistic view—community, individual, migrant destination—is really important."

—Martha W. Rees, Associate Professor of Anthropology, University of Cincinnati

Also by Author

Cultures of Migration MORE +

Cooperation and Community MORE +