Costume is a language. It is not more misleading than the graphs drawn by demographers and price historians. In fact [at the mid-millennium] the future belonged to societies which were trifling enough, but also rich and inventive enough to bother about changing colours, material and style of costume, and also the division of the social classes and the map of the world. Everything is connected.

Capitalism and Material Life (Braudel 1973: 235-236)

24 January 1981; a village southwest of Tecpán. Though the night was cold enough to leave hoarfrost on the cornstalks drying in the fields, the day dawned sunny, with a promise of warmth. This made the early-morning trip bearable as I rode the several kilometers out into the country on the back of a Peace Corps volunteer's motorcycle. I was near the end of my year doing fieldwork in Tecpán Guatemala (or Tecpán for short), and a Maya family with whom I was close had invited me to attend the inauguration of a new Catholic oratorio (chapel) in Xejabí, the rural community where some of their relatives lived.The old chapel had been destroyed by the 1976 earthquake, and it had taken almost exactly five years for the people, many working through the local cooperative, to raise enough money from local and foreign sources to rebuild the structure (Photo I). For the occasion, Tecpán's new priest was invited to say mass. Unlike the last priest in town, this one seemed to be well respected by the local population—Maya and ladino alike. He had roots in the area and, though ladino himself, thought enough of indigenous culture to have tackled the Kaqchikel Maya language.

The service itself was packed; and, for those not fortunate enough to find a space inside, a microphone hooked up to a car battery ensured that the word was carried outdoors. Afterward, people gathered in front of the chapel to socialize, and I took photographs of my friends, their family members and neighbors, and the priest. As I was about to leave the chapel yard, I noticed a frail, elderly Maya woman sitting to the side of the crowd.

She was wearing a frayed and faded huipil (the handwoven blouse worn by Maya women) that I recognized immediately as a xilon blouse.Women back in the town center with whom I was working had described this style to me on a number of occasions and said that it was popular when their grandmothers were young. Up to that point, however, I had never seen one worn and neither had some of my younger friends.

15 September 1983; the streets in the center of Tecpán. Two and a half years after leaving Tecpán, I returned for a visit. During the time I was gone, la situación (the situation, as people euphemistically referred to the violence) had intensified, altering life in Tecpán in numerous ways. The new priest had been murdered less than four months after he helped inaugurate the oratorio; his position was still vacant. The municipal hall, rebuilt after the earthquake, had been blown up by "unknowns" in November 1981 and had not been replaced. By the time of my visit, the army was well established in town, with a military stockade and scores of soldiers stationed across from the post office.

The fifteenth of September is Independence Day in Guatemala. In 1983, in what was meant to be a public show of nationalistic zeal and local unity, soldiers from the stockade and children from the schools in town dressed up in their respective uniforms and marched through the streets to the rhythm of beating drums (Photos 2, 3, and 4). They passed through the plaza area, near the stockade and the empty lot where the municipal building used to stand, and ended up at the municipal soccer field with a round of speeches and gymnastic demonstrations. Maya schoolgirls from all but the Protestant school wore indigenous dress, with their cortes (Maya skirts) coordinated with the jumpers and skirts of the ladino girls in their classes. In many cases, their huipiles were new for the occasion, and, for some, they were woven in the latest local Maya fashion: a version of the xilon-style huipil that I had seen two and a half years before in Xejabí.

June-July 1990, 1991, 1992; the large Thursday market in Tecpán. After a five-year absence, I returned to Tecpán in the summer of 1990 and then again in 1991 and 1992. Each time I marveled at how large the Thursday market had grown and how vibrant it was, despite the hard economic times. Market stalls were crushed together, temporarily filling all the open spaces in the plaza, church square, and nearby streets, and pressed against the buildings that lined them. One of these buildings was the new municipal hall, constructed in the colonial style of its pre-earthquake predecessor; another was a new three-story hotel—the tallest structure in town—built where the military stockade once stood. By the early 1990s, the army troops had moved to the soccer field on the outskirts of town where they had dug in, quite literally, to protect themselves from attacks: their armed presence continued to be visible on the streets of Tecpán.

A survey of the market vendors selling traje showed a vast array of offerings (Photo 5). In the stalls specializing in huipiles, xilon blouses were but one of many styles available, and not a very prominently featured style at that. Joining them were other, more elaborate and expensive blouses, some incorporating color combinations and materials that I had not seen on earlier trips. The customers passing by also wore a range of huipil styles, including some xilon blouses, faded and comfortable from months of wear and repeated washings.

An Ethnography of traje

The subject of this book is traje, or Maya dress, and its central and active role in the lives of people in the town of Tecpán Guatemala. More than mere cloth, traje is a powerful and densely meaningful expression of social identity and a vital element of life in the highlands. I still believe in the basic truth of this statement, despite the fact that, in the twenty years since I first articulated my ideas on the subject, many things have changed—in Guatemala, in the discipline of anthropology, and in some of my own thinking.

Over the past two decades, the "eternal tyranny" (Simon 1987) of life in Guatemala has taken natural and human forms. The earthquake that devastated the central highlands in 1976 left between 20,000 and 25,000 people dead and whole towns destroyed. Many of these same communities had barely started to recover from nature's violence when waves of human violence overcame the highlands, leaving more than 100,000 persons dead or "missing" and over one million people displaced (Jonas 1991: 149). A whole generation has been educated in what people call the reality of life in Guatemala by seeing young soldiers toting large guns in the streets of their towns, hearing gunfire in the mountains at night, learning of family members and neighbors killed or abducted, and witnessing the horrors of violent death and destruction firsthand. While the Maya have not been the only people affected by the violence, they have suffered disproportionately and remain subordinate to the non-Maya in the Guatemalan state (see Smith 1990b).

On a more positive note, many people who were small children when I first met them have gone on in school, surpassing their parents' educational levels and graduating into jobs as teachers, social workers, agricultural specialists, church workers, and administrators. In fact, the battle for indigenous rights and for the articulation of issues pertinent to the indigenous community in Guatemala is now being fought with notable success by skilled and articulate Maya. These are women and men who are becoming increasingly savvy in the internal workings of the national and regional governments and who are reaping the benefits of expanded ties to groups (both Maya and non-Maya) throughout Guatemala and to an international community sympathetic with indigenous projects. A subset of these individuals also forms the core of what is being called the "Maya revitalization movement," a general label that refers to people interested in "the reconstruction of Mayan identity" and efforts "to resuscitate indigenous customs such as the calendar, traditional clothing, religious practices; to systematize the alphabets of the indigenous languages; and to take control of development projects and research in their communities" (Nelson 1991: 6). The "traditional clothing," or traje, of current interest to Maya activists has long been valued by members of Maya communities as well as tourists, collectors, retailers, and anthropologists and other researchers. The different valuations of indigenous dress and the ways in which traje has endured and changed over the years provide a rich material commentary on life in late-twentieth-century Guatemala; its active presence needs to be accommodated in an anthropological exploration of the highlands.

Exactly how traje is explained in an anthropological frame depends on the theoretical perspectives available as well as the training and disposition of the researcher. In the past twenty years my own thinking has been shaped and reshaped by my formal schooling, recent developments in the discipline, shifting events in Guatemala, and ongoing conversations with friends and colleagues in the United States and Central America. I first articulated my thoughts on traje based on observations made during a year of volunteer work in Guatemala prior to beginning my training in anthropology. I went on to produce a dissertation heavily influenced by symbolic and structuralist approaches and have since been persuaded of the need to problematize the material in terms of a multiplicity of voices, situated practices, and negotiated identities (to use the terms that are fashionable).When working with earlier drafts, I take on a reflexive mode and worry that I am dealing with an archaeological artifact—a would-be palimpsest but for the delete function of my computer. I also worry that my work will look dated unless I change each draft in all the right places, but then think I may be cheating because I reframe my arguments at home while the data from field trips long past remain "the same."

The angst subsides as I reconsider my motivations for this project (or succession of projects). Without thinking that anthropology is purposely developing in ways that will allow me better to articulate my concerns, I nonetheless feel fortunate to be working in the field at this point in history. I see anthropologists attempting to find new ways to give voices to those whose words and actions have been limited and showing their commitment to issues of power, resistance, identity, and change in ways that do not reduce these to mere theoretical matters. As far as the study of material objects is concerned, anthropology has recently expanded and embraced the subject outside of narrow frames such as market networks and production technologies. With cloth, attention is owed, in part, to the growing influence of feminist theories, which have helped increase the value and voice of a doubly damned subject: a "craft" and what is often seen as a product of "women's work."

For the most part, the ethnographic present described in this book is 1980-1981. While I also write about the years and events that have followed, I do so in relation to this earlier period and note any shifts in the time frame. By writing this ethnography of traje, I hope to make a modest contribution to ongoing conversations within anthropology and, more important, to an understanding of issues of importance to Maya and non-Maya alike.

The Ethnographic Setting

I decided to live in Tecpán after the highland town in which I had proposed to work became a center of guerrilla and army activities. In October 1979, when I made this switch, Tecpán was a relatively peaceful place, and I judged (or, rather, hoped) that it would remain so for the foreseeable future. Located eighty-eight kilometers west of Guatemala City at a point where high, fertile plateau lands give way to yet higher mountains, the town of Tecpán has easy access to the Pan American Highway and, for my purposes in the early 1980s, to a quick exit from the region (Figure 1).

Traveling from the capital to Tecpán along the Pan American Highway, one passes through fields of wheat, for which the municipality is renowned, as well as corn, potatoes, beans, and, more recently, such export crops as broccoli and sugar snap peas. Lying at seventy-five hundred feet, Tecpán is in a moderately cold zone that favors the presence of pine and oak in the forested mountainsides, but the weather is not so cold as to deter the growth of the occasional domesticated palm. Another noteworthy feature of the Tecpán landscape is the deep ravines that cut the mountains and adjacent plateau. So formidable an obstacle are these to individuals intent on crossing or controlling the area that they were used by the ancient Kaqchikel Maya as natural protective devices to safeguard their fortified capital of Iximche'.

In late July 1524, the Spanish did, however, succeed in gaining access to the area, penetrating Iximche' and setting up the first capital of Spanish Central America. Santiago, as the site was renamed, served as headquarters of civil and ecclesiastical administration (Lutz 1994: 6). This status lasted only a matter of months, however, as the Kaqchikel people rebelled and the capital was moved to a politically more accommodating area. Today the buildings of Iximche' are partially restored and complemented by a museum and picnic area, which are visited regularly by schoolchildren, foreign tourists, and Guatemalan families on weekend outings (Photo 7).

Along with frequent service to the capital, bus routes link Tecpán directly to some of its nearest neighbors: San José Poaquil, Santa Apolonia, Patzicia, and Chimaltenango. Not only are these communities geographically close but their residents have strong economic, educational, and familial ties with Tecpanecos (the people of Tecpán). What is more, because Tecpán is an important market town, special buses for local buyers and sellers connect it with Santa Cruz Balanyá, Patzún, and Comalapa on Thursdays, the biggest market day of the week. This lively exchange of goods also attracts vendors from more distant places such as Sololá, Chichicastenango, Chimaltenango, Totonicapán, and Quezaltenango, and they join the hundreds of other people selling items in the central plaza, church park, and surrounding streets (Photo 8). However, despite the size of the market and its popularity among people in the area, it attracts few tourists.What visitors do arrive usually pass quickly through on their way to the Iximche' archaeological site. Consequently, there is almost no market for tourist items in town, a fact that relates to the nature of local traje production.

Administratively, Guatemala is divided into twenty-two departamentos (departments). Departments, in turn, are divided into municipios (municipalities), which are often described—by local residents and foreign academics alike—as ethnically distinct and defined by differences in costumbre (custom), traje, and speech. A municipio is further divided into a cabecera (town center) and outlying aldeas (villages) as well as other small residential units.

Geographically and demographically Tecpán is considered large, at least in comparison with its neighbors. It shares boundaries with eight other municipalities in three different departments (Figure 2). In the 1981 census, the population of the municipio numbered 29,564, with 5,977 residents listed as living in the center of town (Dirección General de Estadistica1984: 121). The remaining people live in the scattered rural villages, in caseríos (small hamlets) or on fincas (large farms), and along the secondary dirt roads that snake through the mountainous terrain. Population figures, while never completely accurate, were particularly rough between 1976 and 198o as a result of the earthquake, which killed some 510 people in Tecpán's cabecera, according to one report (Diagnóstico de salud y nutrición 1979: 8). While that figure does not distinguish between deaths in the "urban area" (as the census figures describe the cabecera) and deaths in the country, I noted widespread agreement that the town suffered a disproportionately larger number of fatalities than the rural areas.

Besides the loss of lives from the quake, Tecpanecos suffered almost complete destruction of their property. I repeatedly heard how only two buildings in town were left standing and salvageable after the earthquake. The old, two-story, colonial-era municipalidad (municipal hall), remembered nostalgically by many, was in ruins and soon replaced by a plain, one-story structure of cinder block and corrugated metal.The large Catholic church near the park, one of the oldest in Latin America, was severely damaged but, owing to its historic significance, has benefited from years of restoration efforts (Photo 10). In the meantime, a functional Catholic church was built to the east of the old one and new structures for the health clinic, post office, police station, and central businesses once again line the plaza and central park.

While the streets of Tecpán adhere to the grid pattern common to colonial Latin American cities, its central plaza area is something of a variation on a theme (Photo 1 1; Figure 3). Instead of a single central plaza and park, with the Catholic church, municipal hall, and other public offices bordering the square, Tecpán's plaza and park are divided into separate but contiguous spaces. The municipal hall faces the plaza (as do the post office, health clinic, and police station along with a few shops); one block diagonally to the east, the Catholic church and additional shops face out onto the central park. Four barrios (neighborhoods)—Asuncion, Poromá, Patacabaj, and San Antonio—divide the town center into quarters along lines running through the center. To the four barrios has been added a new residential area known as Colonia Iximché and lying at the north corner of Barrio Poromá. The colonia (colony) was built with funds from the Salvation Army for Tecpán residents displaced by the earthquake. Far from simply giving money from afar, the funding body maintained an active presence in the colonia after the earthquake relief houses were built. Along with religious activities, the colonia had a community center that offered day care, crafts classes for women, and other services. In 1980, a missionary family (a Hispanic couple from California) employed by the Salvation Army was also residing in the colonia.

The Salvation Army is but one of many local institutions with a religious mission.While the presence of the Catholic church has been central (literally, but also in many ways figuratively) since the days of the conquest, Tecpanecos estimate that the establishment of evangelical and other crístiano groups in town dates back to the 1940s. Scattered throughout the town are permanent meeting places for the congregations of the Church of Christ,Assembly of God, Prince of Peace Church, Church of Jehovah's Witnesses, and Church of Christ of the Latter-Day Saints, to name but a few. In addition, adherents to smaller, less established groups meet to worship in homes or rented rooms and to make plans to increase membership as well as find permanent church facilities. In contrast to the variety of cristianos in town, Tecpán Catholics are more or less unified. Certain members of the congregation lean more toward involvement in the longstanding cofradías (religious brotherhoods) while others have taken up the work of Acción Católica, a fairly recent movement that Annis (1987: 91) labels as "modern Catholicism"; however, it is not uncommon for people to be involved with both groups.This contrasts with the situations in neighboring Comalapa and San José Poaquil, where there have been violent splits between los tradicionales (the traditional members who support the cofradía system) and adherents to Acción Católica. In both those towns, there are two different Catholic churches, each with its own congregation.

Politics, like religion, is a serious concern for many Tecpanecos. National and international news reaches virtually every home via radio, television for many, and newspaper for a few. Because most information on the civil war has been unofficially censored or minimally reported, people rely heavily on informal channels of communication: conversations with relatives, neighbors, or fellow workers in town or away from home; gossip passed around in offices or at the public sinks; and comments overheard in the market or on buses. These are also the channels used for learning about local news—political and otherwise—that would probably never make it into the national media under any circumstance: the decision to repair the fountain in the plaza, the theft of a neighbor's cow or television set, and the appearance of antigovernment graffiti spray-painted on the walls of buildings leading into town.

Members of key Tecpán families wield political power in the mayor's office and through associated bureaucratic positions. According to a number of indigenous Tecpanecos, up until the 1970s the mayor had always been ladino, with certain individuals known for having taken particular advantage of the office for personal gain; the siphoning off of earthquake relief funds and the private sale of archaeological artifacts from Iximche' were repeatedly described. Maya involved in municipal government are not automatically exempt from criticism by other local Indians; they too can be faulted for using the office for selfish ends, for acting "ladino" Finally, people living in aldeas have ties to and become involved in local government via elected representatives and work obligations. While the large Maya population in these rural areas (95 percent versus about two-thirds in the cabecera) does not necessarily guarantee indigenous representation, that is usually the case. Given this fact, along with the higher status of the "urban" positions relative to the "rural" ones, conflicts between elected officials from the cabecera and the aldeas often get redefined in ethnic terms.

While employees of the local government are generally "Tecpanecos" in the sense that they are full-time residents of the town, own a home and agricultural land or have family members who do, and can trace their family roots back generations, this is not necessarily the case for the professionals working in local offices of the national government. These professionals are men and women who have completed educational programs known as carreras (careers) that lead to specialized degrees in such areas as primary education, accounting, bilingual secretarial work, or agricultural education.The younger government workers especially are often unmarried, have close family ties in Guatemala City or other, more urban communities, and yearn for the day when a new position opens that will take them closer to home and out of what they consider a remote area. They would not label themselves "Tecpaneco." In 1980, I knew a number of postal workers, agriculture specialists, and teachers in this category. They had families—parents, wives, and children—living in other communities and commuted regularly (daily or on weekends, at the least) to their more permanent homes.

This pattern of employment had (and continues to have) to do with who is educated and, hence, eligible for government positions. However, nowadays, as more and more students from "rural" communities like Tecpán continue their schooling, they too are contenders for these salaried jobs. For them, it is ideal to live in the highlands, with or near their families, and, in the cases of Indian professionals, to work with a largely indigenous population and perhaps even use their Maya language skills. These are the hopes of many, at least. In most cases it takes months or years of waiting for more or less the right job (or any job, in some cases) in an acceptable location. The person waiting also understands that assignments are political and subject to the whims of individuals higher up the bureaucratic ladder.

Among the self-employed in Tecpán, a large number of businesses are operated in spaces that adjoin the homes of their owners. General stores, pharmacies, bakeries, clothing boutiques, shoe shops, restaurants, and stores selling items of traje open directly onto the sidewalks and beckon people in, while the rear or upstairs areas of these properties serve the private, domestic functions of the owners' families and are off limits to customers. Some of these establishments are quite large and hire people to work fixed hours serving shoppers or preparing items for sale; other, smaller places are single-family operations. A common arrangement among this latter type is for the women in the house—often several generations—to engage in their normal domestic duties except when the occasional customer appears. This system also allows a mother of small children to work in a salaried job outside the home while leaving the children with her mother or mother-in-law and a hired muchacha (girl, but more specifically a young woman hired to work in the home or shop).

Businesses without retail outlets may be tucked away in the back or sides of properties. In Tecpán, items produced under this system include sweaters knitted on manually operated knitting machines, mercerized and dyed thread, corte cloth woven on treadle looms, and cinder blocks. Maya women with their backstrap looms who produce pieces for sale also have to be included in this category. Relatively speaking, their efforts do not pay well. However, the work is flexible, fits in with other domestic duties, and provides an additional source of income for the family. And the future prospects of the business seem solid, given the enthusiastic use of traje in Tecpán (cf. Ehlers 1990).

The market is a major commercial operation in Tecpán, with the largest market on Thursdays, a medium-size one on Sundays, and a few vendors all other days. The year I lived in town, there was no permanent market building (the old one had collapsed in the earthquake), so the whole enterprise was basically created and torn down from week to week. Since that time, the permanent market building has been rebuilt to the east of the Catholic church, just off the central park, but it almost does not count. It is what Goldin (1987) calls the "compartmentalized enclosed market," in contrast to the "quasi festive plaza" of the open-air market. The concrete walls are dark and inflexible, the presence of soldiers hanging around uninviting, and its size pathetically small compared with the enormity of the Thursday market. The people with whom I shop rarely go into the building and, instead, frequent vendors elsewhere in the market.

The transformation of the Tecpán plaza from Wednesday to Friday is fantastic to witness. On Wednesday morning the plaza is virtually empty, by late afternoon scores of people are constructing stalls of poles and tarps or displaying fruits and vegetables on squares of cloth, and by early Thursday morning everything is in place. While the market as a whole has its rhythm, particular categories of vendors come and go by their own clocks: the men who sell grinding stones appear only on Wednesday afternoons over by the health clinic, and people from Chichicastenango who have the sticks used for backstrap looms sit under the evergreen trees on the west corner of the park and usually sell out well before noon on Thursday. Unless the day is rainy, most vendors stay until mid- to late afternoon. As buyers begin to thin out, goods that have not sold are packed up and loaded onto buses or trucks or otherwise carted home. By Friday morning the last bits of trash are swept off the streets and traffic can easily pass through town once again.

Though the Thursday market might look chaotic to an outsider, it is, in fact, quite organized. To a certain extent, things are arranged by type: most of the traje sellers are grouped together, as are vendors of pottery or furniture or livestock. While there are exceptions to this rule, the one thing that can generally be relied on is the location of individual vendors over time. With few exceptions, I can visit the market in Tecpán after being away for several years and almost immediately locate a stall and vendor I remember from a decade ago.

The abundance and variety of fresh produce in the market reflects the importance of agriculture in Guatemala. In Tecpán the majority of families own or rent tillable land within an hour's walk from their homes and either farm this themselves or, in the case of the wealthier families, hire day laborers to do the work. Typically land is sown at the start of the rainy season in April or May to yield all or part of the domestic supply of corn by January. The dried ears of corn are harvested, stored in cribs at home, and prepared as needed for the family's supply of tortillas and other corn-based dishes. In the same fields, Tecpanecos often sow black beans and squash, which grow up and around the corn. Finally, depending on interests, energy, and the availability of additional land, fruit trees, grasses for work animals, garden vegetables, or a cash crop may also be planted. While the cultivation of major cash crops calls for a great deal of planning (e.g., arranging a contract for the sale of broccoli to exporters), excess produce from smaller efforts—a dozen or so squashes, an armful of gladiolas, some avocados or peaches—can be sold in the Thursday market. For most Tecpanecos, work in the milpa during the yearly agricultural cycle meshes with home life and the rhythm of daily chores, church and the cycles of religious celebrations, and the patterns of local commerce, including the weekly markets and the Monday-to-Friday workweek for students and salaried employees. To these I add my own irregular cycles of visits—relatively brief appearances and longer absences—that influence my relationship with the town and its people, the sorts of events I witness, and the types of information I can record and discuss here.

Fieldwork in Tecpán

Were I clairvoyant, I would have paid more attention in 1974 when I quickly passed through Tecpán on the back of a Peace Corps friend's motorcycle. I would have taken in the church, municipal building, and houses in what is now remembered as their pre-earthquake glory and made mental note of the current use of dress by local inhabitants. My future, however, did not flash before my eyes, and I left with few memories aside from the general one of having been there.

I was in Guatemala at that time working as a volunteer home economics teacher through a 4-H international program (4-S in Guatemala). I had recently finished college with a degree in mathematics and a lateblooming interest in the visual arts, and had volunteered to work abroad while I made up my mind whether to pursue a career in painting and printmaking. I left home with a lot of questions about what was going on in the art world in the United States. An academic understanding of various art movements and some practical experience selling my work had not dispelled a feeling that there was more to the subject. How, for example, did the energies and thoughts of the publicly proclaimed, preeminent creators of the visual in our culture fit in with the lives and ideas of everyone else? And what would our world look like if the appreciation, production, and ownership of art had a broader social and economic base?

In 1973-1974, I worked in Jutiapa, a large ladino town near the El Salvador border. It was during that year that I made my first visits to the highlands and, in June of 1974, enjoyed a month-long stay in the largely Maya town of Comalapa. Because of ties to the community through my Peace Corps friend, I was able to attend a local wedding, participate in a range of activities connected with the celebration of the town's patron saint, and spend most weekdays learning backstrap weaving with a family of women. Though these initial experiences in the highlands were limited, I was struck by the central role of weaving and handwoven clothing in Comalapa life. This came in sharp contrast to my experiences with local clothing production and use in Jutiapa, where there had been little or no community associated with the activity. It also contrasted with my earlier efforts in studio art. While I understood both fine arts in the United States and traje in Guatemala to be the preeminent visual expressions of a certain segment of each society, the latter seemed to have a wider, more all-encompassing audience and an immediacy of relevance that marked it as being very different from what was produced in my own society.Were these reasonable points of comparison, or was I seeing Guatemala through several layers of Western stereotypes of "non-Western" art? In an effort to move beyond these first impressions I applied to graduate schools in anthropology with a proposal to do research on traje and, four years after entering the University of Chicago, was ready to return to Guatemala and start my fieldwork.

Both Paul Rabinow (1977: 1) and Loring Danforth (1989: 300) write about vivid memories of key events that took place immediately before they left for the field: Robert Kennedy was assassinated two days before Rabinow left Chicago, and Nixon resigned five days before Danforth left for Greece. In both instances the upsetting national events became symbols for the disgust and alienation they felt toward their own cultural identity, and fieldwork was seen as a means by which each could become "a different person" (Danforth 1989: 300; see also Todorov 1988: 4). That, however, was not my experience. I left the States having just completed an internship in applied anthropological work on issues that I saw as having important human rights implications. After years as a graduate student, I had thrived in a work environment where I felt valued and able to make a concrete contribution. In contrast (and despite the celebration of the Sandinista victory in Nicaragua), reports of civil strife in Central America made me question why I was going to Guatemala—why I needed to challenge myself by doing work in what I knew could become a dangerous situation.

As it turned out, the memorable and upsetting events that marked the start of my fieldwork all took place in Guatemala. I arrived on October 22, 1979, and nine days later learned that two tourists had been killed in the area where I had hoped to work. I took that as a clear sign that I could not live in Nebaj, in the increasingly embattled northern El Quiché region, and chose to work in Tecpán instead. After a quick trip back to the States and a mad scramble to locate ethnographic materials on a different town, a different region of the highlands, and a different Maya language, I returned to Guatemala. On January 31, three days after my return, thirty-nine Maya and Maya supporters were massacred by the Guatemalan army at the Spanish Embassy in Guatemala City. I contemplated my own future in light of those events: I wondered how long I would be able to live in Tecpán, whether I would be forced to leave the country, and if, under those circumstances, I would have enough data to write a dissertation.

From February 198o to February 1981 I worked in Tecpin and returned for brief stays during the summers of 1983, 1985, 1990, 1991, and 1992. I chose Tecpán because, at that time, it was located in a relatively peaceful area, its weaving tradition was strong, and it had a local branch of the Ministry of Agriculture with an active home economics teacher, which meant I could volunteer to work with women's 4-S groups as I had in Jutiapa. Another factor that I had not explicitly considered but that greatly benefited my work was the progressive spirit of Tecpán—"el municipio vanguardista" (the avant-garde municipality), as one local writer put it (Ajozal Xuyá 1977b: 13). Tecpanecos have a knowledge of and participate in national and international trends, events, and policies to a notable degree. This characteristic distinguishes them from residents of other highland towns—at least, as they are portrayed in older ethnographies—who live in closed corporate communities seemingly without significant involvement in activities outside their own municipal boundaries. Moreover, there is a push for education and professional jobs among Tecpanecos (Maya and ladino alike) that is noteworthy; the display of material wealth is substantial (especially considering the destruction by the 1976 earthquake) among people engaged in such capital-intensive work as building comfortable homes, running trucking businesses, and operating modest-size stores and cooperatives; and outside advances in science and technology—categories generally treated as neutral in a society where so much is categorized as "ladino" or"Indian"—are sought out and embraced under select but widely occurring circumstances. At the same time, the strength of cultural expression by the indigenous segment of the community is evident.

Once settled in the Tecpán area, I began work with the local home economics teacher. I presented myself as a student doing her práctica (field practicum), a standard requirement at Guatemalan schools and universities. Tecpán, in fact, is a frequent field site for Guatemalan students completing professional degrees in home economics, health care, and nutrition. Although I explained that my field of study was anthropology, my interest in weaving and my activities with the 4-S teacher more often than not labeled me a student of home economics.

Some of my earliest contacts with townspeople were with members of 4-S clubs (principally one in the town center and another in a nearby aldea) and girls in the home economics classes at the local Protestant school where my co-worker also taught.The two groups presented an interesting contrast as the women and girls in the clubs tended to be from somewhat poorer homes, were largely uneducated beyond the primary-school level, and ranged from fifteen to twenty years of age. With one or two exceptions, all were Maya. In the school group, the girls ranged from ten to fifteen years of age, were approximately 80 percent ladino, and came from somewhat wealthier families who placed emphasis on school learning and discipline and who could afford the monthly tuition fee.

From this start, my connections grew to include other families and members of different organizations. Though I strove to work with a wide range of people, I spent much more time with women, especially Maya women—which, given my topic, made sense. Maya women in Tecpán wear traje much more than men and are much more likely to be involved in its production. They are also generally more involved in making clothes for children, dressing children, washing clothes, repairing them, and storing them. This is not to say that traje—or clothing, more broadly put—is a "women's topic." While there is a strong gender dimension to the subject, any study that focuses exclusively on women ignores the important ways in which men are—and are not—involved with traje and why. In this study I have tried to emphasize the importance of women's roles in the production and use of traje in Tecpán while also discussing men's more limited concerns.

As my contacts in Tecpán widened, I also started weaving (Photo 16). I found that this activity gave me an opportunity to be with women and to engage with them in conversations on clothing, cloth, and production processes. And as a result of my ability to knit, sew on a treadle sewing machine, read English and hence English-language repair manuals for the same treadle sewing machines, fashion children's parade costumes from crepe paper, make piñatas, and construct serviceable ovens from five-gallon lard tins, I found myself with invitations from homes in all parts of town and sufficient opportunities to ask research-oriented questions. In addition, I accompanied people on shopping trips for thread, cloth, and clothing; shopped for clothing and sewing materials for others (which necessitated their telling me in detail what they expected me to buy); acted as a photographer for special occasions (and, hence, of special outfits); and discussed photographs of traje as well as vestido (clothes) or ropa corriente (common clothes), as the clothing worn by ladinos and "gringos" is labeled, from a number of sources (e.g., books on indigenous dress, fashion or craft magazines, and snapshots of local people).

At other times I had occasion to function in more "elite" roles. During the year, I helped evaluate school and fairtime competitions having to do with cake decorating, flower arranging, traditional Maya dress, and what was labeled folk dances and costumes. In each of these instances it seemed that my eligibility to serve as a judge was based on my status as a university-educated, home economics teacher from the United States rather than any concrete knowledge of the topic at hand.

I also developed a small survey and hired a local student to administer it. Its purpose was not to assemble broad-based statistics on the town, for by and large I have relied on existing materials for this type of quantitative data. Rather, I used the survey to check some of my own findings against those collected by a local Kaqchikel speaker. The results, which I discuss later, generally supported what I had already learned.

While the focus of this work is on contemporary Tecpán, I also researched the historical record of the town and early written references to municipal dress. Archival material in Tecpán itself was minimal since everything from before 1976 was lost when the municipal building, which reportedly housed a small library, collapsed in the earthquake.To date, the materials I have found in the General Archive of Central America in Guatemala City do not contribute greatly to this project, since they pertain largely to topics such as land claims, milling rights, and archaeological sites. Mention of dress is found in travel accounts from the late colonial period onward and, more recently, in the field reports of government employees, university students, and textile specialists. Though very early material on Tecpán is limited (see, for example, Cortés y Larraz 1958 [17691770]: 172-173; Stephens 1969 [1841]: 147), a field report written in the 1940s by Rosalio Saquic (1948) for the Instituto Indigenista Nacional provides an account of local clothing use in an era still within memory of many Tecpanecos.

Nearly all my field research has been done in Spanish. As soon as I started work in Tecpán, I began lessons in Kaqchikel Maya, but I did not pursue these for more than a month as it became apparent that I could manage well in Spanish and that I should get on with my research as quickly as possible in light of the political situation in Guatemala. The vast majority of the indigenous population living in the center of Tecpán is bilingual, with Spanish the preferred language among many. To communicate with some of the older, monolingual Kaqchikel speakers and to understand the occasional event en lengua (in language—i.e., in the Kaqchikel language), I asked friends to translate for me. During trips to Guatemala in 1990 and 1991, I started studying Kaqchikel again, this time in Antigua as part of the Tijonik Oxlajuj Aj language and culture program organized by Tulane University and the University of Texas at Austin. Our teachers were from throughout the Kaqchikel region and saw themselves as part of the larger Maya language revival movement in Guatemala. This recent study of the language has allowed me to check some of the basic terms associated with clothing collected a decade earlier and to move forward slowly with research in both Spanish and Kaqchikel.

With hindsight, I can see that I went to the field with a set of unspoken, perhaps even unrealized, assumptions about the nature of data collection and "good" anthropological data. While I knew that I wanted to go beyond simple word lists and explanations of techniques associated with traje, I found myself hoping to find a "text"—a story, myth, or history about clothing—that I could record, transcribe, translate, and then describe (thickly, of course). This general formula appeared in many pieces of anthropology writing that I admired from my graduate course work, and I longed to try my hand. I was also surprised by my interest in finding something that could count as a "ritual," or rather a ritual as a metaphor for culture. I now see this latter interest as an expanded version of my search for a "text," but it had its own appeal. On one occasion—as I sat watching a Mother's Day skit—I saw my dissertation flash before my eyes: Was this the "ritual" I was looking for? In the end, however, pages of field notes written enthusiastically after the event were reduced to a sentence or two in this book. I slowly came to realize that, on the subject of traje at least, information does not exist in an institutionalized, codified form and does not present itself as neat, prepackaged "facts." Rather, it is embedded in "'attitudes,'... individual comments or opinions or judgments about what so-and-so did [or wore] or should have done [or worn], about what ought to be, [and] about what does or does not 'make sense'" (Urciuoli 1983: 2; see also Urciuoli forthcoming), as well as nonverbal actions and activities that complement, contradict, clarify, confuse, or otherwise take place along with the verbal messages.What this means is that such timehonored field methods as the survey and the formal interview are likely to fail or misinform unless the person is extremely familiar with the ethnographic context (see Briggs 1989). Indeed, many of the most important observations derive from being in a place over a long period of time and becoming increasingly familiar with and able to comment on the content and interconnectedness of various forms of knowledge in a community.

Part of the problem with placing so much importance on "things that go without saying" (Comaroff and Comaroff 1991: 23, paraphrasing Bourdieu 1977: 167) is that the very information that is most desired by the anthropologist—what is presupposed by the local population—is hardest for an outsider to grasp and disentangle from its social context, especially an outsider intent on turning everything into written words, photographs, maps, and diagrams. In addition, the process of converting field notes to formal anthropological writing turns up numerous holes in the data—the very things one forgets to ask about or look for because they go without saying. In my own work, I was lucky enough to be able to return to Guatemala several times before I finished the dissertation and, again, before writing this book. Each time I went with lists of questions for people and mental notes of different activities I hoped to witness. My intent was always to fill in the holes, an often difficult proposition for someone who has stepped outside of the sweep of life as I had.

When I began writing the thesis on which this book is based, my efforts were helped immeasurably by a talk that Paul Friedrich gave to graduate students at the University of Chicago. In the course of about one hour he outlined what later appeared as "Experience and Methods" in The Princes of Naranjo (1986). I was especially struck by his discussion of "the personal experience of writing up the results of fieldwork" and, in particular, his "principles of composition" (ibid.: xv, 226). Inspired by Friedrich's ideas and other recent attempts at innovations in ethnographic writing, I have tried to weave a text that is complex but not obscuring and that gives multiple voices (Friedrich's partial holography [ibid.: 228]) to the issues of concern to both Tecpanecos and anthropologists.

Classifying Ethnicity in Guatemala

Up to this point I have used terms to refer to different ethnic groups without much comment. The labels, however, do much more than divide the more than nine million inhabitants of Guatemala into two groups of roughly equal size. In Tecpán, as is more or less true throughout the highlands, people refer to themselves, their families, or neighbors as belonging to one group or the other.The terms appear regularly in casual, everyday speech—in describing a stranger, gossiping about a neighbor, or commenting on a new vendor in the market—and largely go unnoticed. The terms, however, have the potential to inflame emotions, as when a member of one group makes a coarse generalization about the other group at, say, a political rally or a bar.

Of the two principal categories, the group designated with the Spanish terms indígena (indigenous), natural (native), or Maya or with the Kaqchikel terms qawinaqi' (our people) or Mayab' is the least general one. Unless an institution or object is labeled "Indian" or qualified by a term that serves the equivalent function, it can generally be assumed that it has little or nothing to do with the indigenous population per se. Thus, the name of the Comité Indigena Pro-fiestas Franciscanas (Indigenous Franciscan Fair Committee) specifically labels it as a Tecpán Indian fair group whereas the companion organization run by ladinos is more simply named the Comité Pro-fiestas Franciscanas (Franciscan Fair Committee). One of the effects of the labels is to make the former group look like an offshoot of the latter, and in this case, from a historical perspective at least, it is. But in addition, the difference in specificity of the titles hints at an asymmetry in relations as well as the differential power of the committees and committee members. The people involved in the former group are not members of simply the Fair Committee but rather the Indigenous Fair Committee.

In Tecpán and throughout Guatemala, "ladino" and "Indian" are considered labels of one's raza (race) or grupo étnico (ethnic group).As they are most commonly used, the terms point to a division of people according to (1) certain overt markers (dress and language are regularly mentioned); (2) actions that have ethnic significance, with costumbre used to refer to Maya customs; (3) one's blood, heritage, or historical roots (as in ties to Maya ancestors or European stock); and (4) a history of relations between the conquering and conquered that still holds true today. However, how different people define themselves, even how the same person presents himself or herself under different circumstances, is not set. Hence, different Maya women have told me that they would feel positively naked if they had to wear vestido; they would be scorned by their family and friends and feel great alienation. At the same time, the husbands of these very same women do not wear traje, and most never have. In fact, the vast majority of the men in the Kaqchikel region do not wear indigenous dress, although males who are working to revitalize Maya culture are actively exploring ways to express their ethnicity through traje (see Otzoy 1992a).

Ethnicity also provides a rationale for action, at least in the case of the Maya population.An indigenous person might consciously and explicitly speak out, act a particular way, or criticize another human being in accordance with his or her perceived ideal of what it is to be Indian and how this ideal should find expression in the world. Maya people and the Maya way of life are seen to be threatened by nonindigenous forces that are powerful, nearly omnipresent, and often opposed to indigenous values. At the local level, ladinos are representatives of these forces, as are the national government and the United States within certain "higher-level" contexts.

Judged by its breadth of use and its power to explain the widest range of social phenomena, ethnicity must be considered one of the fundamental classifiers of the person in Guatemala. Other systems of labeling the person exist, but, in the eyes of many Indians at least, a number of the more sociopolitically oriented labels need to be understood in terms of basic ethnic distinctions. Thus, Maya will charge ladinos with prejudicial actions based on ethnic considerations (the low level of government farm credit to Indians is one example), while non-Maya, talking about the same situation, will disclaim indigenous charges and contend that the division is based on contextually appropriate and racially unbiased criteria such as wealth, the size of land holdings, or the number of votes. Maya, commenting on the latter claims, will agree that criteria other than ethnicity can be used to divide the population. However, as they see it, it is because they are Indian that they are confined to the lower socioeconomic rungs and therefore can always be separated out by classificatory schemes that divide according to elements of power.

While pan-Maya awareness is growing in Guatemala and has meaning in many contexts, the boundaries of the community implied by the terms for "Indian" are often much smaller than the nation or the highlands. As I mentioned earlier, the municipality is recognized locally as a key unit for Maya identity, with municipal styles of traje important visual signs of local ethnic unity. In 1937 Sol Tax first articulated this idea in an anthropological context when he claimed that municipalities were the basic units for defining Indian communities and Indian identity in Guatemala. Municipalities also became the basic unit for anthropologists doing fieldwork in the highlands (recent examples of the resulting ethnographies include Warren 1978, Brintnall 1979, Hawkins 1984, Annis 1987, Ehlers 1990, and Watanabe 1992). My work in Tecpán is no exception to this tradition. However, in contrast to earlier works in which highland residents often come across as blinkered and oblivious to events beyond their cornfields, recent writings take a more broadly integrated approach: local ideas and interactions are described in multiple forms and linked to social phenomena inside and outside of the community.

"Ladino," as I noted, is the usual complement to "indígena," but this term, too, has its complicating factors. Pansini (1977: 138-177) further distinguishes "European Guatemalans"—those with strong, direct ties to families and life abroad—and "Guatemaltecos"—a label that goes beyond referring to the person's nationality and indicates group status akin to the criollos of Guatemala's past (that is, native-born descendants of foreigners with upper-class socioeconomic ties). Non-Indians in both these categories generally do not refer to themselves as "ladino," in part, he suggests, because the class status that this term implies is lower than what they consider to be their own. In fact, along with the markedness factor, the status element might be another reason why I heard non-Indians of Tecpán use an ethnic label to refer to themselves much less often than indigenous residents.While no non-Indians in Tecpán have European connections in any immediate sense, a number of the children from ladino homes attend school in the capital and, hence, are likely to come in contact with more "European" or "Guatemalteco" types. A feeling for the hierarchical system into which all non-Indians fall (vis-à-vis other non-Indians, not to mention in comparison with Maya) can only be highlighted by choosing one label over the other.

Finally, there are people who are, as one Maya woman put it, "in the middle." In a 1979 statistical survey of Tecpán completed by a team of students from Guatemala's national university (Diagnóstico de salud y nutrición 1979), male and female heads of households from one-seventh of the town's homes were asked to define themselves using one of the following categories: "ladino," "natural," "otro" (other), or "no aplica" (does not apply). Of the 180 homes surveyed, 109 people (male and female) or 30 percent of the sample population classified themselves as "ladino," 230 people or 64 percent classified themselves as "natural," no one selected "otro," and 21 people (all men) or 6 percent selected "no aplica." When I asked one of the survey team members what kind of people tended to choose the last of these categories, he commented that they were largely "naturales" but generally those who as youths had forsaken the use of Kaqchikel and other public markers of Indian ethnicity, gone to the capital, married non-Indians, and then returned to live in Tecpán. While these individuals take what are seen as steps away from being Indian, at least for a while or on certain occasions (some also reclaim their Indian status), neither are they accepted as members of the ladino community. Others' perceptions of them plus their own self-definition may put them "in between."

In my own experience each quandary, hesitation, or shift in ethnic classification is surrounded by a complex of stories and judgments. As examples from Tecpán will show, evaluations of character, actions, relationships, ambition, social status, moral worth, as well as dress and other physical dimensions can come into play in an assessment of one's own or someone else's relative ethnicity. I say "relative" because it is not so much a matter of an Indian becoming a ladino as a person trying to blur the lines or to act like one in certain contexts and for particular reasons. What is more, the categories applied to people are not stagnant. The defining factors associated with the ethnic terms shift over time, with people offering explanatory criteria selectively and changing their stories (not to mention their clothes) according to the situation. A group of people discussing the same person, object, or event might vary considerably in their ethnic assessment.

The Basic Vocabulary of Traje Indigena

Maya clothing is referred to as traje, traje or vestido típico (typical traje or dress), traje or vestido traditional (traditional traje or dress), traje indígena, or traje Maya. As a preface to the ethnography of clothing that follows, I list the Kaqchikel and Spanish terms for the different parts of Maya women's and men's traje and give a simple description of each.This vocabulary list allows me to present traje as a whole—a complete outfit—since, with some exceptions, a person wearing only select pieces of traje is not considered to be wearing traje in an unqualified sense.

What exactly constitutes traje differs from person to person and town to town. The terms listed below are ones given in an illustrated Kaqchikel-Spanish dictionary produced by PRONEBI (the National Bilingual Education Program) and used in some of the primary schools in Tecpán (PRONEBI 1989:26-27). In some cases I have noted variations particular to that community, since the standardized vocabulary of the dictionary cannot accommodate all dialect forms. The entries in the dictionary are listed under ri tzyäq/la ropa (the clothes) and are accompanied by illustrations (Figure 4).

The basic pieces of Maya women's traje are as follows:

1. pach'ub'dl wi'aj/listón (ribbon). A five-foot-long strip made of brocaded satin cloth in a pastel hue and braided into the hair.
2. pot/huipil (blouse). Made of fabric handwoven on a backstrap or treadle loom, with embroidery, velvet, or other cloth ribbons sewn around the neck and armholes, depending on current styles in a town. Rectangular in shape and formed by folding the woven cloth in two, huipiles have slots in the side seams for armholes and a hole in the top for the head. They are also the single most distinctive piece of women's attire, often elaborately and expensively created and signaling a particular municipal origin.
3. ichi'q'ab'aj/anillo (ring). At one time locally made from silver or gold, now more commonly produced outside the highlands and of unknown materials. Nupq'a' is another Kaqchikel word for this item.
4. ximb'äl/faja (belt). Generally two to three yards long and woven on a backstrap or treadle loom. It is worn wrapped around the waist, the end tucked in to secure the belt. Design motifs or color combinations often indicate a particular municipal or regional origin. Tecpán Indian men also use woven belts for their traje, securing the rodillera (wool "apron") at their waist with a belt that may be over five yards long. In some contexts, ximb'äl refers exclusively to a man's woven belt while pas refers to a woman's.
5. uq/corte (skirt). In Tecpán this article generally consists of six varas of cloth wrapped around the body. Two types of uq cloth are common in the central highlands—one made of acrylic threads with bands of cotton threads jaspeado (tie-dyed) to form intricate designs on the weft, and the other made entirely of cotton, with jaspe (ikat or tie-dyed designs) on the warp and weft threads (Photo 17). Both are made on treadle looms in standard eight-vara lengths. While these two types of uq are seen as "Maya," they do not link the wearer with a particular municipality.
   An older, more "traditional" style of corte, called a morga, is made from dark, indigo-dyed threads. This thick cotton cloth has patterns of white lines that form stripes and rectangular patterns distinctive of particular central highland municipalities. The Tecpán style is referred to as penecita (Photo 18).
6. peraj/perraje, rebozo (shawl). A long rectangular shape often woven so that bands of solid color alternate with bands of jaspe designs. Q'ub'äl, another Kaqchikel word meaning "shawl," is the preferred term for some indigenous speakers who are involved with the revitalization activities and who want to avoid using Maya terms derived from the Spanish.
7. su't/pahuelo, servilleta (general-purpose cloth, covering cloth). Handwoven, often in a style that is town-specific, these may range in size from a foot square to three or four feet square. They are used to cover or wrap a number of items, including tortillas and other foods in the kitchen, television sets and other electrical appliances, and the contents of baskets carried to market. Servilletas are also part of what people call el mero traje (the real traje) of Tecpán. Folded and placed on the head, women wear these to Catholic mass or on occasions when reference is made to the most traditional or authentic traje from Tecpán.
8. lantar/delantal (apron). Made of pleated uq fabric, usually adorned with rickrack or other decorative touches, and with a deep, zippered pocket for guarding money.
9. ichixkinaj/arete (earring). Like rings, earrings are made in the area from silver and gold, though many women prefer to wear "imported" styles. The term used in Tecpán is chixkin.
10. ichinaj/collar (necklace). Traditionally made from glass beads and, at times, old coins. These, more often than rings and earrings, have been replaced by pieces not marked as "Indian."

Not listed but always mentioned in Tecpán as an element of women's traje is the ri'j pot/sobre huipil, huipil degala (overblouse). Larger than the regular pot, which is also called the rupam pot (inside huipil), the sobre huipil is worn loose, something like a poncho, over the blouse and skirt. Though armholes exist, the wearer's arms generally hang down inside the garment and extend out under the bottom selvage.

Men's traje in the Kaqchikel region includes the following:

11. pawi'aj/sombrero (hat). Generally made from plaited straw.
12. kamixa'/camisa (shirt). Made of white or light blue cotton. The Kaqchikel term is derived from the Spanish.
13. wexaj/pantalón (pants). White, straight-cut through the leg, and rather loose. In almost all cases now, these are made from industrially manufactured cloth. In 1990 the major choice in the Tecpán market was between the cheaper, all-cotton pants made from Guatemalan fabric and the more expensive pairs made from a synthetic fabric produced in El Salvador.
14. xajab'/zapato, caite (shoe, sandal). The same terms are used for women's footwear.The Kaqchikel term b'ukut refers more specifically to enclosed shoes and boots, such as those shown on the right in the figure.
15. k'ul/poncho, chamarra (poncho). No longer worn as part of a man's traje in Tecpán. The only article I have seen that I would assign to this category is a piece in a private collection. Woven of dark blue wool with a white zigzag just above the bottom, fringed edge, it was said to have been used for cofradía functions years ago.
16. peqe's/morral, bolsa (bag). Knitted, crocheted, or woven and used to carry a range of things such as a lunch or school books. Girls and women also use these for school books and items related to their jobs, though morrales would probably never be listed as part of women's traje in Tecpán.
17. xerka/rodillera. A rectangle of black-and-white wool, woven in a checkered pattern, and measuring approximately four feet by one and three-quarters feet. A rodillera is worn folded in half so that the white fringed edge hangs down. It is secured at the front of the man's waist, over the front of his white pants, by a woven belt.

Not listed but, in my experience, always named as part of men's traje in Tecpán is the chaket/saco de hombre (men's jacket). Made from a dark blue or black wool, it has a notched collar, patch pockets, and a single row of buttons down the front. The ximb'äl/faja (woven belt) used by men is identical to one style used by women: it is warp-faced, with blue, pink, green, and white the predominant colors.

While this list provides basic vocabulary, it can only hint at the complexity of social meanings. The mero traje—real traje—of a municipality is assumed to be a unique form of dress that has been handed down from preconquest times to the present. Yet what counts as municipal traje—or even traje, more broadly put—is much more context-dependent. Questions having to do with the wearer, the occasion, and the other clothing options available all come into play.

Interpretive Frame

The academic framework for this book is drawn from two broad sources. One body of literature is largely nonanthropological in orientation and is written by a wide range of people, including art historians, professional weavers, and museum specialists; the other reflects my anthropology training. The former focuses on Guatemala in general and traje more specifically; the latter deals with cultures around the world and with subject matter ranging from language to ritual to revolution. The first has provided much background detail and technical insight; the second has provided more of the theoretical inspiration. Both are ancestors of this work.

While virtually every anthropologist writing on the Guatemalan highlands acknowledges the distinctive role of dress as a marker of ethnic identity (usually in the opening pages of the book and then rarely again), nonanthropologists produce the bulk of the published material on traje. Much of the research has adhered to models of data collection in the folklore or folklife tradition that focus sharply on the reality of the physical object and its visible, describable characteristics. These studies privilege the sorts of data that are most accessible to outside investigators, namely, lexical items that refer to and label the world "out there." Traje production, for example, can be described with an extensive technical vocabulary that is articulated fairly readily and translated into roughly parallel terms in Spanish or English (see, for example, Anderson 1978; Bjerregaard 1977; Maxwell Museum of Anthropology 1976; Sperlich and Sperlich 1980). Similarly, the formula that equates one municipality to one style of traje (or somewhat more complex variations on this) is at the root of another large portion of this literature, with photographs, drawings, or paintings complementing text (e.g., Atwater 1965; Bunch and Bunch 1977; Conte 1984; Delgado 1963; Dieterich, Erickson, and Younger 1979; O'Neale 1945; Pettersen 1976; Wood and Osborne, 1966). As something of a blend of these two approaches, works also focus on design motifs associated with different towns and provide detailed design outlines or thread counts (see, especially, Mendez Cifuentes 1967 and Morales Hidalgo 1990, but also parts of Atwater 1965; Delgado 1963; O'Neale 1945)

Other trends in writing about highland traje move more into the realm of ethnohistorical or ethnographic accounts of data. For example, a number of texts emphasize the history of the subject and examine the general chronology of weaving in an area or trace historical changes in the traje of particular municipalities (e.g., Anawalt 1981; Pancake 1977; Rowe 1981; Schevill 1985). Following the lead of Penny Capitalism (Tax 1971 [1953]) and other studies that emphasize the economic dimension of highland life (e.g., Nash 1958; Wagley 1941) are works concentrating on the economic aspects of traje production and marketing (Hagan 1970; Pancake and Annis 1982; Rosales Arenales de Klose 1978). Two recent, long-term ethnographic studies on textile entrepreneurship and economic change—Annis (1987) and Ehlers (1990)—relate weaving production to religion and gender, respectively. To my mind, these last two plus a series of municipality-based studies published through the Ixchel Museum of Indian Dress (e.g., Asturias de Barrios 1985; Asturias de Barrios, Mejia de Rodas, and Miralbés de Polanco 1989; Mayén de Castellanos 1986; Mejia de Rodas and Miralbés de Polanco 1987; and Miralbés de Polanco, Sáenz de Tejada, and Mejia de Rodas 1990) provide some of the most insightful ethnographies on traje, weaving, and related subjects. They are built on in-depth research in one community and bring current theoretical issues to bear on the topics. This combination has the effect of enabling the subjects to "speak" with greater vitality.

Theoretically, this book reflects my evolving interests in issues being debated within anthropology and related disciplines in the United States and Guatemala. If one were to think of this project as a geological site, the oldest layers—the symbolic and structuralist concerns—are now somewhat obscured, but they remain fundamental nonetheless. Motivating this work since its earliest days are the ideas that the material world must be understood as a cultural system, that objects reflect a wealth of cultural categories, and that meaningful patterns relate all "objects" within a cultural universe. I was formally introduced to these concepts in a first-year graduate course given by Marshall Sahlins as he was sending Culture and Practical Reason (1976) to press. In the course and in the book, as well as in Petr Bogatyrev's work on folk costume (1971 [1937]), I also found analyses of the structural-functional dimensions of two different clothing systems. From these rather sparse and formal treatments, I went on to read the more field-based (but nonetheless structurally informed) work of Marie Jeanne Adams (1969, 1971) on Sumba, Indonesia, and Terence Turner (1979) on the Kayapo of the Brazilian Amazon.

In those early days I was also inspired by work in the ethnography of speaking, as this was introduced to me in classes with Michael Silverstein. I was attracted to works that focused intensely on the contextual interpretations of a single linguistic "object" or set of "objects" (whether myths, songs, sermons, jokes, promises, or intimate conversations). From these studies I drew parallels to the material world, not in the sense of seeing textiles as "texts" in any literal sense, but rather considering them to be culturally bounded entities that must be understood as contextually situated and multiply meaningful within a society.

Since my original fieldwork, the writing of the dissertation, and now the rethinking and writing of this book, my work has remained within the historical trajectory of symbolic anthropology. However, along the way I have tried to update and refine my thinking on and sensitivity toward issues arising within Guatemala and anthropology—for example, issues having to do with gender, ethnicity, identity, and conflict as well as questions of agency, intention, and boundary. Written in such terms, my concerns may seem bloodless just words. However, what has motivated me to keep writing about traje is a desire to put to work theoretical perspectives that are alive within the social sciences in the service of understanding a topic that is very much alive in the highlands.

***

The general organization of this book was inspired by the works of Fernand Braudel, which I began reading after I returned from the field. At that point I had a year's worth of data that I had collected in Guatemala as well as an assortment of interpretive passages written in an attempt to prime the dissertation-writing process, but little idea of exactly how all the pieces were going to fit together. Back in a university setting, I took particular interest in looking at books to see how other authors crafted a whole from a series of parts, to see how the parts were ordered and the pieces made to add up.

Braudel's ability to combine details of the daily and the mundane along with the extraordinary of any epoch appealed to me. In my own work, I had been frustrated by my inability to represent the apparent discontinuity between the ordinary—the everyday recurrence of dressing, washing, and making clothes, for example—and the timely and politically loaded use of dress in public spectacles. In The Mediterranean World in the Age of Philip II (1976), Braudel achieves this integration of materials within an interpretive framework of three categories of time that I have appropriated and manipulated. While I will not try to pass these off as "Maya," it does seem appropriate to me to use time as the frame through which I organize and translate my field experiences in Mesoamerica into text for an English-language audience.

The first measure of time, which Braudel sees as tied to the geography of a place, is characterized by persistent, unchanging, virtually timeless patterns—la longue durée, as he labels it (1980 [1958]). This idea is also reflected in a culture's notion of "human nature"—that part of a person or group that goes beyond individual lives or particular historical moments and reflects some essentially timeless core. Chapter 2 examines the enduring relationship between traje and municipal identity and contrasts this with non-Indian dress and the geographic elements associated with it. Chapter 3 emphasizes the wearers of traje and the role of dress in the construction of images of the Maya that claim to reflect some essential, unchanging truth.

The second category, labeled "social time" or "social history" by Braudel, is characterized by regular patterns that repeat within human memory. I prefer to think of this as conventional or stereotypical change. Time in this sense occupies a unique, intermediate place in the scheme of Brandel's histories and my ethnographies. On the one hand, the stereotypically repeating patterns can be considered constants of sorts: changes occur in the short run, but the pattern taken as a whole holds a constant shape over time. On the other hand, emphasizing the change in "stereotypical change," ephemeral and transformative aspects of phenomena emerge, though these fall within the frame of repeated appearances."

While the social histories discussed in this book are labeled "life cycles," perhaps a better term would be "life span." Unlike the economic and political cycles examined by Braudel, the life cycles of highland Guatemalans and traje pieces are linear developments with specific starting points and end points (the former even more so than the latter). In a certain sense they are like Geertz's cockfight gatherings, namely, "a particulate process that reoccurs rather than a continuous one that endures" (1973: 424). However, the discreteness of these life spans or cycles does not erase the similarities among the instances nor does it preclude comparisons between different people or traje articles. Tecpanecos, for example, are more than willing to draw parallels or cite differences between, say, the quality of cortes produced today versus those of thirty years ago or the types of huipiles worn by their teenage daughters in the annual Independence Day parade versus those of neighbors' children.

The differential emphasis on the wearer and the worn that distinguishes Chapters 2 and 3 carries over to the life cycle in Chapters 4 and 5. Chapter 4 focuses on the process of socialization and the role of traje in defining and creating different stages of a person's life. Chapter 5 examines the stages of the "life" of indigenous dress, beginning with thread bought in the market and following the production and use of articles of clothing over time.

Braudel calls his third measure of time "traditional history," the history of events and individuals, who are passionately engaged in their own worlds but essentially blinded to the larger contours of social life (1976, 1:21). Filled with reports of the actions and people considered important in their day, this history is, for Braudel, "the most exciting of all, the richest in human interest, and also the most dangerous" (ibid.). The excitement lies in the possibility of locating the germ of long-term structural change within the particulars of daily life; the danger lies in the possibility of prematurely labeling something as "new" when, over the long run, it will turn out to be just "more of the same."

Chapter 6 is my attempt to locate possible points of "real change" in the contemporary production and use of traje. However, instead of focusing on the great men and momentous political and economic events that affect a vast region—Braudel's traditional history—the events that interest me are more local in scale, more female-centered, and seemingly insignificant. Here I examine several instances of the creative in Maya clothing from the early 1980s and discuss how these relate to changes in other parts of society. As if to bear witness to the truth of Braudel's warnings, the spirit of this chapter, much more than the others, has changed in the years since the data were first collected. What I and a segment of the Tecpán population saw as "new" in traje more than a decade ago now seems only mildly innovative, if not downright routine, in terms of current classifactory schemes.

When his book on the Mediterranean was first published, Braudel received a note from a colleague asking why he had not written his three histories in the reverse order. In answer, Braudel referred to "the metaphor of the hourglass, eternally reversible" (1976, 2:903). And so it is with the central, ethnographic chapters of this work, balanced as they are between discussions of change versus stasis, meaningful whether read in forward or reverse order.

Chapter 7, on the other hand, stands apart and relates to the preceding chapters as a whole. Here I consider the ideological identification of traje with the people who label it "our dress" (in a sense similar to the term "our mother tongue"). I examine how traje can be taken as an active, enduring cultural object—one that summarizes and symbolically maps some cultural core of the people with whom it is related—at the same time that it is actively manipulated by different individuals for changing audiences. In this instance the power of traje, seen as a quintessential Maya expression, reflects the totality, uniqueness, and vitality of "our culture."