To return to the Kemper Faculty Home Page, click here.

To return to the ANTH 3311 Home Page, click here.

To return to the ANTH 3311 Syllabus and Schedule Page, click here.

  From the "Modern Tarascan Area" to the "Patria Purépecha": Changing Concepts of Ethnic and Regional Identity

  Robert V. Kemper and Julie Adkins

Hablar de la "región purépecha" (que algunos denominan tarasca) supone referirse a un espacio construido por múltiples actores. . . . Estudiar una región, en suma, no es describir una cosa, algo que está ahí: es abstraer un conjunto de procesos sociales que tienen una expresión territorial (de la Peña 1987:9-10).

In 21^st-century Mexico, the concept of "indigenous regions" continues to demand attention from anthropologists, government officials, planners, and the general public. Especially following the uprising of the EZLN in Chiapas on January 1, 1994, no one doubts the significance of the claims made by indigenous groups for greater control of their lives and their localities in the face of powerful external forces. The forms of regional domination that came into being during the Spanish colonial period have continued to evolve in step with the economic and political transformations in the Mexican nation state and modern capitalist enterprises. Current concerns with globalization and transnationalism represent yet another phase of long-established practices in which external powers confront local communities.

In the present article, we will consider the situation of what is commonly called the "Purépecha¹ Region" (or the "Tarascan Region"), located in the northwestern sector of the State of Michoacán. However, before we can focus our attention on the continuities and transformations in this specific social and territorial space, it is necessary to consider the concept of "indigenous region" in contemporary Mexico. After dealing with the recent advances in ethnographic research and census data analysis, we will turn our attention to the challenge of defining the Purépecha Region in the light of available ethnographic and census data.

Approaching Mexico’s Regions

The great diversity of Mexico has led social scientists and governmental agencies to create "regions" in order to reduce this diversity into more manageable units. For example, the geographer Angel Bassols Batalla (1979) elaborated a system of eight economic regions as a method of understanding Mexico’s historical development. These regions – which he identified as "Noroeste, Norte, Noreste, Centro-Occidente, Centro-Este, Este, Sur, and Peninsula de Yucatán" – were arbitrary constructs invented as heuristic devices for his analysis. Using data derived from the 1960 census, the National Commission on Minimum Salaries defined a much more elaborate scheme of 111 regions for the entire nation. These regions (one of which was called "Meseta Tarasca") were then compared according to a "development index" constructed by the sociologist Claudio Stern (1973).

An alternative approach to regionalization stressed the importance of "cultural" rather than geo-political characteristics. These regions are understood to be "real," since they are widely recognized by those peoples who dwell within (and beyond) their boundaries. Well-known classic examples of ethnographic approaches to such regions include Eric Wolf’s (1955) study of the Bajío, Guy Stresser-Péan’s (1952-1952) numerous publications on the Huasteca, Wolf’s (1976) ethnohistorical analysis of the Valley of Mexico, and Robert West’s (1948) treatment of the modern Tarascan area, to name only a few.

Clearly, all forms of "regions" must be recognized as cultural constructs. As Van Young has argued, "Regions . . . are hypotheses to be proven, rather than givens to be assumed" (1992:3). While regions often are constructed in terms of natural environmental features, ultimately they have a human dimension. Whether regions are defined in terms of economic elements or cultural practices, these human characteristics are not natural features of the landscape, but are imposed on the landscape by generations of humans acting within structures of power, ideology, and hierarchy. In the contemporary era, our interest in regions relates to the reality of the non-random distribution of human populations and economic resources across the landscape of territories controlled within nation states. Who lives in a given place? Why are they in that place and not in some other place? How are people in different places connected? And where is the locus of power for places and cultures? These are questions with practical meaning and theoretical implications (cf. Barragán López 2001; Boehm de Lameiras 1997; de la Peña 1988; Garza 1999; Smith 1976).

The characteristics of the population in any region depend not only on cultural traits (such as languages spoken), but also on the socioeconomic and political conditions present in that specific area, as well as competing conditions in all other places within (and beyond) the boundaries of the Mexican nation state. Applying this perspective to the regions in Mexico, Aguilar and Graizbord suggest three factors that influence the general settlement patterns for populations:

Approaching Indigenous Populations Through "Regions"

In focusing on the spatial distribution of specific indigenous populations in Mexico, we can see that each of the factors mentioned above by Aguilar and Graizbord is important in determining their present social locations. Their core populations continue to be located in areas historically important to their cultural customs and religious rituals. These territories are not merely of theoretical and academic interest. Indigenous regions have been disproportionately affected by man-made projects and natural disasters which – combined with the internationalization of labor markets – have, in some cases, forced individuals, families, or entire communities to resettle themselves into other areas (cf. Instituto Nacional Indigenista 2000).

An adequate approach to indigenous regions depends not only on a diversity of characteristics, but also requires appreciation of the perspective of the persons (or agencies) offering the definition. Thus, a governmental agency might employ one set of traits, while social scientists might prefer another. The members of a particular indigenous population might consider some elements more significant than others and, therefore, affirm yet another way of defining who they are in regional terms.

But it is not just a matter of distinctive understandings of what is an "indigenous region." Often, there are overlapping interests and identities. Just imagine, for example, that a governmental agency employs a social anthropologist who also is a member of the indigenous group being studied as part of a large-scale river or forest development project. What sort of definition might such a multi-cultural individual utilize to understand the territorial situation of her own people? Such an individual would need to find a way to integrate the bureaucratic definitions of a government agency and the professional definitions shared among anthropologists, while honoring the native definitions learned while growing up in her home community. Only in recent years are the experiences of such multi-cultural individuals beginning to penetrate the bureaucratic and professional worlds that claim power over and provide assistance to indigenous regions in Mexico.

Recent Projects Concerned with Indigenous Regions in Mexico

Since the early 1990s, the Mexican State has been especially interested in determining the extent of indigenous populations and their territories. Several agencies have worked, both independently and collaboratively, toward the goal of mapping the indigenous groups throughout the nation. These agencies – including INAH, INI (now the CDI), INEGI, CIESAS, and CONAPO – have carried out major projects, with substantial funding from the Mexican government, from foundations, and from international development agencies (e.g., Banco Mundial). These projects have employed diverse methodologies and theoretical frameworks, ranging from fieldwork in selected communities in different indigenous regions to demographic analysis of national census data. The substantial number of publications resulting from these projects is revolutionizing our understanding of indigenous groups in contemporary Mexican society. Relevant to our focus on the Región Purépecha is the fact that, without fail, each of these projects has elaborated a series of indigenous regions for the country as a whole – and in each project is manifested some territorial expression of the Purépecha people in the north-western sector of the state of Michoacán.

Ethnographic Approaches to Indigenous Regions

The ethnographic literature about indigenous communities and indigenous regions in Mexico is enormous and shows no signs of diminishing in importance. In an effort to provide anthropologists and other investigators with a resource guide to this literature, the INAH is in the process of issuing three volumes in a collection entitled "Las Regiones Indígenas en el Espejo Bibliográfico" (Barabas 2002), as part of a large project called "Etnografía de las Regiones Indígenas de México en el Nuevo Milenio."

In volume one, in his introduction to the section of the study dedicated to the State of Morelos, Miguel Morayta offers these important observations about the problem of understanding regions:

We share Morayta’s concern with how "region" articulates with categories like "culture" and "ethnicity." Conceptually, "region" is not a "natural" slate on which cultural facts are written and rewritten throughout history (cf. Aguirre Beltrán 1967; Dow and Kemper 1995; García Mora 1987-1988; Kirchhoff 1943; Kroeber 1939; Smith 1976; Vogt 1969). Nor is it necessary that our analysis of indigenous regions begin with the units of estados and municipios in which indigenous populations are embedded. Nonetheless, the political apparatus is powerful and difficult to ignore. Therefore, it is not surprising that, during the past century, anthropologists and other social scientists have found ways to combine "natural" (e.g., physiographic) and "cultural" (e.g., language use) traits with "political" (municipio and estado boundaries) to deal with the characteristics of indigenous regions in Mexico. At the same time, the processes of globalization at work in contemporary Mexico also raise the question of the "deterritorialization" of populations and communities (cf. Giménez 2000; Lomnitz-Adler 1995). As a result, we find it necessary to do ethnography in multiple sites, rather than in single communities, as we follow people in their circuits between one "home" and another "home" through their working careers and across the generations.

Geo-Political Approaches: Estados y Municipios

The geo-political space of contemporary Mexico provides the basic framework within which, in the absence of designated "reservations" or "territories," more than sixty indigenous groups are recognized by the State. The Mexico nation state currently is divided legislatively into 32 entities (31 estados plus the Distrito Federal). Each of these estados is, in turn, divided into numerous municipios, each containing many localidades. This is the geo-political context within which indigenous populations and their cultures are recognized – and regionalized – by the State. Of course, the boundaries of such political units (estados, municipios, and localidades) often have very little to do with the ways in which indigenous peoples themselves have understood their own historical, political, and cultural relationships within and beyond their home territories.

Governmental agencies need a relatively simple and direct way to determine where to invest their limited resources to improve the lives of those living on the margins of society. In the policy arena, complex definitions create confusion; simple indicators are much easier to use to determine where programs should be implemented. Therefore, it is not surprising that the Mexican State has built its programs on two principal criteria – the first being geo-political and the second linguistic-cultural.

The first geo-political framework for determining the social location of indigenous populations is the estado. States have the advantage of having stable boundaries over long periods of time. Unfortunately, states usually are too large to be useful for designing programs and development projects directed toward indigenous populations. Moreover, there is little sense of political or cultural membership between indigenous groups and their states of residence, especially in those circumstances in which members of the same indigenous group live in different states (cf. Sandoval Forero 2002). For example, based on the distribution of speakers of indigenous languages, CIESAS (2003) divided the entities of Mexico into four levels ranging from "high" to "moderate" to "low" to "none."

Therefore, government agencies have turned to the municipio as the most appropriate geo-political unit for assessing the level of "indigenous" population in a region. The number of municipios in the nation is relatively large (for the 2000 census, the number of municipios was 2,443), thus permitting regions to be created by combining municipios within states and even across state boundaries. Moreover, municipios constitute the first line of local government throughout the nation and thus represent a key mechanism for political control and economic development. Although the INEGI provides a report on localities for each decennial census, it is much less comprehensive and less widely available than information provided for states and municipios.²

Determining the Characteristics of Indigenous Municipios

The problem remains of how to determine "indigenous" versus "non-indigenous" for any specific municipio. Beginning with the 1930 census, the only consistently available information for all municipios in all states has been based on a single item in the decennial censuses.³ This item – the population of persons five years and older who spoke an indigenous language – provides a minimal definition of cultural participation in indigenous populations. From census to census, the question about speaking an indigenous language has varied somewhat, especially in terms of the treatment of monolingualism and bilingualism. Despite its obvious limitations,

Consequently, the category "speakers of an indigenous language" (HLI = hablantes de lengua indígena) has continued to serve as the first step for identifying indigenous population. In this way, estados, municipios, and localidades can be categorized according to the distribution of this cultural trait in a given census year and over a period of censuses.

By the time of the 1990 census, the limitations of the HLI caused the INEGI to include a new question in the national census: the population from 0 to 4 years old whose "jefe de familia hablaba lengua indígena." By adding this number of children to the number of persons 5+ years old who reported themselves to be speakers of an indigenous language, a more complete picture of the indigenous population could be obtained than in prior censuses. Nevertheless, based on the growing indigenous population reflected in the 1970 and 1980 censuses, the resultant total did not meet governmental agency expectations.

Therefore, the INI (1993) made its own estimates of the indigenous populations, based on the census data and observations of the populations served through its more than 100 centros coordinadores y indigenistas (CCI). In the process of moving beyond the census data on speakers of indigenous languages, the INI

Redefining the Indigenous Population: Localidades y Hogares

This pursuit of new census variables led analysts at the INI to move beyond municipios to localidades, based on the proportion of HLI. Using data from the 1990 census, the INI identified three categories of localidades:

Not satisfied with the results achieved through using the HLI in the context of localidades, the INI analysts took further steps to calculate more accurately the indigenous population in Mexico. First, they considered that all residents in the "localidades eminentemente indígenas" should be considered as indigenous; second, they considered "indigenous" all the inhabitants of any other localidad if INI reported that the Instituto was providing resources to them; and, finally, for the "localidades medianamente indígenas" and "localidades con población indígena dispersa," they included in the "indigenous" population the children from 0-4 years old whose household head spoke an indigenous language. In this manner, the INI created the new and important concept of "Población Indígena Estimada" (PIE).

No longer limited to HLI, the indigenous population reported in the 2000 national census came to be estimated in several different ways, not only at the level of localidades but even at the level of hogares. In conjunction with representatives of the Consejo Nacional de Población (CONAPO), the INI adopted the hypothesis that "todas las personas que habitan en una vivienda en la que el jefe de familia o su cónyuge habla lengua indígena, todas son consideradas indígenas (Embriz Osorio y Ruiz Mondragón 2003: 93). This view is much more in line with ethnographic practices than was the previous definition based solely on HLI or even PIE.

The estimates for 2000 also made use of a new census question to which persons would respond with an "auto-adscripción indígena" – in which persons would be asked whether "¿Se considera indígena?" o "¿No se considera indígena?" Ultimately, because the household head tended to be the person to respond to the question on behalf of all household members, this self-identification actually was more accurately defined as "pertenencia indígena" – and, as such, was added to the growing set of variables that formed the PIE. Through such estimations, the indigenous population grew to more than 12.7 million persons for the year 2000 and surpassed 13% of the national population – more than double the number of speakers of indigenous languages (HLI) reported in the census.

This new way of defining the PIE also would have an impact on the ways in which municipios were defined as indigenous. Thus, the new standard for defining an entire municipio as "indigenous" would be to have a PIE of 30% or more of its total population. By this definition, using data provided through the XII Censo de Población y Vivienda 2000, the Comisión Nacional para el Desarrollo de los Pueblos Indígenas (CDI, formerly INI) calculated that "Para el año 2000, de los 2,443 municipios que existen en nuestro país, 803 pueden ser calificados como indígenas en la medida que concentran 30% y más de Población Indígena Estimada (PIE)" (CDI 2004; http://www.cdi.gob.mx/conadepi).

In effect, from the time of the 1990 census to that of the year 2000, governmental agencies revolutionized their approach to defining the indigenous population and indigenous regions. From the use of the single, simple census variable of speaking an indigenous language (HLI), a complex set of variables was developed to provide an estimate of the indigenous population (PIE). The creation of the concept of PIE, and its application to municipios, resulted in almost one-third of the nation’s municipios being defined as "indigenous" for official purposes.

The result has been a doubling in the number of persons defined as indigenous, whether they spoke an indigenous language or not. The cultural concept of the hogar emerged as a new way to approach census data, as did the cultural concept of auto-adscripción. These new ways of estimating the indigenous population and, thereby, defining indigenous regions open up possibilities for new concepts of ethnic identity, as these might be used by ethnographers in their field research and also as these might be used by ethnic groups that want to advance their own political programs (cf. Reina 2000). Having invested in new ways of defining the indigenous population, the government now faces the responsibility and enjoys the opportunity to relate these "regiones indígenas" to endangered natural resources and the patrimony of future generations.

La Región Purépecha

The Región Purépecha will be examined in what follows as an example of how a specific indigenous region has come to be understood by multiple actors, including academic anthropologists and social scientists, as well as governmental agencies and their representatives. To date, the major way of determining the boundaries of the Purépecha Region has depended on analysis of the number of speakers of the Purépecha language – in other words, the criterion of HLI. The new ideas about estimating the indigenous population (PIE) have not yet been implemented in studies of the Purépecha area, although we assume that future studies will move in this direction within a few years. Given the importance of HLI for studies of the region, we begin by reviewing the census data for 1940, 1970, and 2000.

The Distribution of Tarascan Speakers in Michoacán: 1940-2000

Tables 1, 2, and 3 provide data for monolingual, bilingual and total speakers of Tarascan for 1940 and 1970 and of Purépecha speakers for 2000 for 31 municipios in the state of Michoacán in the context of the totals for the rest of the state. These 31 municipios represent all of the municipios cited by at least one of the twelve sources reviewed here; the range runs from a low of 6 to a high of 23 municipios for a particular source. In the tables, we have divided these municipios into three groups: (1) the core municipios; (2) urban destinations; and (3) outlying municipios. We recognize that, in a few of the outlying cases, certain localidades have a good number of Purépecha speakers while the rest of the municipio has very few or none.

[insert Tables 1, 2, 3 here]

In Table 1, which deals with the year 1940, we see that eighteen municipios in the northwestern part of Michoacán contained 100% of the monolingual Tarascan speakers in the state and the nation. These same municipios also contained about 97% of all Tarascan speakers, including bilingual Tarascan-Spanish speakers, who were living in Michoacán at that time. In relation to the population of these 18 municipios, the proportion of Tarascan speakers in 1940 averaged about 23% of the total population, ranging from a low of 4% in the municipio of Uruapan to a high of 70% in the municipio of Cherán. None of these eighteen municipios had fewer than 1,000 Tarascan speakers in 1940.

In Table 2, which deals with the year 1970, the core region of Tarascan speech had shrunk to 16 municipios, even if we reduce to 500 speakers the minimum for inclusion. In addition, as urban destinations, Morelia reported 556 Tarascan speakers and Jacona had 153. Nuevo Parangaricutiro and Puruándiro had suffered substantial losses in their Tarascan-speaking populations and became peripheral to the core linguistic area. Compared to 1940, the proportion of Tarascan speakers in the regional population declined considerably, because the total number of Tarascan speakers in the 16 municipios changed little in the 1940-1970 period, while the total regional population more than doubled. The number of monolingual speakers declined by half, to less than 10,000 persons, although the number of bilingual speakers increased by about an equivalent amount (from about 31,000 to almost 42,000). These 16 municipios contained more than 99% of all monolingual Tarascans and 95% of all bilingual Tarascans in Michoacán.

In Table 3, which deals with the year 2000, the core region (of what the INEGI now officially called "Purépecha" speech) consisted of the same 16 municipios as in 1970, again using the cut-off point of a minimum of 500 Purépecha speakers. All of the municipios in the core region -- except Tzintzuntzan, located on the eastern edge of Lake Pátzcuaro and most distant from the Sierra core region -- witnessed a gain in the absolute numbers of Purépecha speakers in the period from 1970-2000. In addition, Morelia (with 2,430 Purépecha speakers), Zamora (with 1,003), and Jacona (with 570) represented significant urban enclaves peripheral to the core region.⁴

For 2000, the number of monolingual Purépecha speakers had increased to more than 15,000, a gain for the period 1970-2000 that made up the "decline" reported in 1970. Nevertheless, from 1940-2000, the number of monolinguals did decline by more than 20%. If the monolingual speakers had matched the overall growth rate of the state of Michoacán for the period from 1940-2000, the number of monolingual Purépecha speakers would have increased to about 65,000. So, the imputed "net loss" has been about 50,000 speakers in sixty years. Meanwhile, the number of bilingual speakers in 2000 rose to more than 91,000 in the state; of these, about 90% were to be found in the 16 muncipios of the core speech region. If the number of bilingual speakers of Purépecha in the core linguistic region had maintained the overall growth rate of the state of Michoacán, their number in the year 2000 would have been 105,000. So, the imputed "net loss" of bilingual speakers has been about 13,000 in sixty years.

[insert Table 4 here]

As the data in Table 4 show, in 1970, the impact of emigration beyond the state of Michoacán was already becoming significant, with the result that only 83% of all Tarascan speakers in the nation still resided in the core territory. For instance, in the Distrito Federal, there were 2,148 Tarascan speakers (27.4% of all Tarascan speakers outside of Michoacán), while Nayarit reported 725 (9.3%), Estado de México had 555 (7.1%), and Oaxaca had 537 (6.8%). The other half of the Tarascan speakers reported by the 1970 census as living outside of Michoacán were distributed among all of the rest of the states. In addition, an unknown number of Tarascan speakers were working and living in the United States at the time of the census.

By the year 2000, emigration from the state of Michoacán continued to be common for Purépecha households in the core region. The proportion of all Purépecha speakers in the nation still residing in the 16 municipios remained essentially stable at 82%. According to the national census, there were 3,074 Purépecha speakers in the state of Jalisco – including 1,071 in the municipio of Zapopan and another 620 in the municipio of Guadalajara; there were 1,754 Purépecha speakers in Estado de México and 1,724 in the Distrito Federal; and, reflecting the movement of individuals and families to the north, there were 2,097 Purépecha speakers in Baja California – including 1,309 in the municipio of Tijuana. The only other states with more than 500 Purépecha speakers were Sinaloa with 515 and Colima with 506.

Substantial emigration – whether temporary or permanent – to the United States from the Purépecha homeland suggest that thousands of individuals (and their families) are now beyond the scope of the Mexican national census, but are not being identified by the U.S. census. This expansion into nearby as well as distant areas reinforces the view that the geo-political boundaries of the municipios in the north-western area of Michoacán no longer are real limits to the continuity of Purépecha culture. We shall return to this issue in the final section of this article.

Approaching the Purépecha Region: Ethnographic Studies since the 1940s

In the more than seventy years that have passed since Moisés Sáenz’s (1936) pioneering study of Carapan, the Purépecha region has been the subject of hundreds of ethnographic studies, nearly all of which have focused on single communities. The community case studies by Ralph L. Beals (1946) in Cherán and George M. Foster (1948) in Tzintzuntzan soon came to be seen as "classic" models for nearly all ethnographies for the following three decades. In a similar fashion, the regional study by Robert C. West (1948) of the modern Tarascan area was recognized as a valuable contribution, but decades passed before this regional approach took on new life. With the exception of a few studies (e.g., Moone 1973; Stanislawski 1950), comparative and multi-sited studies were rare until very recently. Save a single early exception (Brand 1951), these studies paid virtually no attention to municipios as political units within which Tarascans practiced their culture. Nor did ethnographers make use of municipios as constituting the political, economic, and social space for the Tarascan region.

In contrast, beginning in the 1980s, some ethnographers rediscovered West and became more interested in the region rather than merely in specific communities. Even where fieldwork remained limited to a single locality, the research had a higher goal -- i.e., to understand the larger regional context for cultural transformations. As a consequence, ethnographers began to move beyond localities as speech communities toward an approach which examined speech (and other cultural characteristics) in the context of municipio units. This move was connected to the increasing use of Mexican census information to evaluate changes in local populations. Since census data were more widely available for municipios than for localities, it is not surprising that municipios emerged as useful units for regional analysis (cf. Dehouve 2001: 12ff.).

Now, we proceed to the review of the twelve studies, the first eleven of which use municipios as a mechanism for defining the region. These regional studies include ones carried out by independent academic scholars dedicated to particular problems (e.g., forestry, urban development), as well as general studies carried out by ethnographers working within the framework of national projects concerned with indigenous populations throughout the nation.

We begin by considering the first comprehensive regional study of the Purépecha – carried out by the cultural geographer Robert C. West (1948) – and then continue with other major regional studies, including the most recently published, that is, the project coordinated by Aída Castilleja et al. (2003) within the national project called "Etnografía de las Regiones Indígenas de México en el Nuevo Milenio." The last regional study to be examined, authored by Andrew Roth-Seneff and Robert V. Kemper (1995), does not use municipios as a central concept for defining the region, but instead emphasizes the more ambiguous concept of territorial homeland as a way to comprehend the Purépecha Region. In this review of the published works on the region, the diversity of approaches taken will become abundantly clear and the theoretical, methodological, and practical implications also will become apparent.

Studies Focused on the Región Purépecha

In 1948, the Institute of Social Anthropology of the Smithsonian Institution published West's Cultural Geography of the Modern Tarascan Area, which became influential in defining the "Región Purépecha" for many subsequent scholars. Basing his analysis on the changes in the boundaries of Tarascan influence over five centuries, West began his study with the simple statement, "The present territory of Tarascan speech consists of a relatively small area (about 3,500 sq. km.) in northwestern Michoacán" (West 1948: 1). His emphasis on Tarascan speech was fundamental for his analysis of Tarascan culture. In this regard, West was following the paradigm then dominant in cultural anthropology -- i.e., to assume that the cultural trait of speech was the key to unlocking the distribution of native groups across the natural landscape. Before analyzing in detail the physiographic areas of northwestern Michoacán, including information on climate, vegetation, and soils, West defined the "Modern Tarascan Area," as follows:

According to West, four geographical regions comprised the modern Tarascan area: The Sierra, The Lake Pátzcuaro area; La Cañada; and The Northern Plateau zone. In addition, West mentioned the existence of isolated concentrations of Tarascan speakers in surrounding areas (e.g., Cuanajo, to the southeast of Pátzcuaro), as well as a small number of Tarascan-speaking folk living in some of larger mestizo towns adjacent to the main Tarascan area, including Uruapan, Pátzcuaro, Coeneo, Zacapu, Zamora, and even Morelia (West 1948: 1).

West made a personal visit to virtually every pueblo and rancho in northwestern Michoacán. In addition, he made use of data available in the 1940 Mexican census. Nevertheless, he ignored municipios as significant spaces within which Tarascans spoke their language and practiced their culture. Although previously West had divided the Tarascan area into four reggions, in his analysis by municipios and localidades, he employed five regions: The Sierra: Charapan, Cherán, Chilchota, Erongarícuaro, Los Reyes, Nahuatzen, Paracho, Parangaricutiro, Pátzcuaro, Tancítaro, Tangamandapio, Tangancícuaro, Tingambato, Tingüindín, and Uruapan; The Lake Area: Erongarícuaro, Pátzcuaro, Quiroga, and Tzintzuntzan; The Northern Plateau Region: Coeneo, Zacapu, and Morelia; La Cañada: Chilchota, Tangancícuaro; and The Tierra Templada: Ziracuaritiro, Tingambato, Uruapan, and Ario de Rosales (1948: 18-20, Table 1). Yet , it is striking that none of these municipio boundaries and municipio names are included in any of the 21detailed maps in his monograph. It also is noteworthy that the set of municipios is geographically discontinuous, extending beyond the core linguistic area, in order to include isolated pockets of Tarascan speakers in distant localities, including the state capital, Morelia.

His analysis of the "present distribution of Tarascan speech" was that it was "confined mainly to the tierra fría – Lake Pátzcuaro and the Sierra. These regions, together with La Cañada, appear to be a resistant core in which the Tarascans are making a last stand against complete loss of their native tongue" (West 1948: 17).

West also was concerned about future trends in the population of Tarascan speakers. He stated, "Judging from the long-range trend in both areal recession and actual numerical decrease in Tarascan-speaking folk, it would appear that the Tarascan language is headed toward exitinction" (1948: 23). West argued that

A Municipal Framework for the Tarascan Region: Kemper

One of the first efforts to analyze the Tarascan region in a municipal framework was a paper presented by Robert V. Kemper at the Second Symposium on Regional History and Anthropology at the Colegio de Michoacán, held in Zamora, on 14-16 August 1980. In that paper (which was revised and published both in English [Kemper 1981] and in Spanish [Kemper 1987]), Kemper argued that

Approaching the question of Tarascan speech in this municipal framework, Kemper examined the census data available for 1940, 1950, 1960, and 1970 (since the 1980 data were not available at the time of the presentation of his conference paper). He found that, for 1940, 18 municipios contained a significant number (i.e., more than 1,000) of Tarascan speakers, whereas by 1970 only 16 muncipios had at least 500 Tarascan speakers.

After examining the changes in population in the Tarascan Region between 1940-1970, Kemper showed how urbanization, industrialization, tourism, artesanal production, and emigration contributed to the transformation of Tarascan culture. Using census data, he found that Tarascan speakers were located in every federal state. Examining ethnographic studies, he discovered that the socioeconomic universe of communities had ceased to correspond to their traditional boundaries in northwestern Michoacán. Kemper concluded his analysis by observing that

By the early 1990s, one author -- Luis Vázquez León -- moved from the cultural trait of language to a more complex set of criteria for defining the Purépecha Region. To develop his idea of the "sistema social regional" for La Meseta Tarasca, he first examined the Meseta Tarasca as a "región geográfica y económica," in which he considered and rejected studies by COPLAMAR that the Meseta Tarasca consisted of some 17 municipios -- located in the Sierra, in the Lake Pátzcuaro area, and in the Zacapu area -- with a very high level of "marginación." He also examined and rejected the idea promulgated by the SARH that the Meseta Tarasca consisted of some 14 municipios that formed "una región campesina en la que predomina la agricultura de subsistencia" (Vázquez León 1992: 49). He then reviewed and rejected studies by the Plan Lerma that the Tarascan area consisted of some 24 municipios that emphasized "cultural" traits, especially language, as the primary trait for defining the Tarascan region. In the same manner, he rejected the proposal by Kemper (1987) that the Tarascan region had consisted of 18 municipios in 1940 and only 16 municipios in 1970. Then, Vázquez León shifted his focus to consider the Meseta Tarasca as a "región forestal," on the grounds that -- by the mid-1980s -- ethnic communities in the Sierra had begun to organize themselves cooperatively and collectively to regain control of the region's forestry resources from outside forces. In the process,

In this manner, Vázquez León came to view the Meseta Tarasca as becoming a "región étnica" which was "producto de la articulación de niveles distintos de poder social" (1992: 99). Moreover, he argued that "la viabilidad de la organización étnica en un nivel regional depende de su capacidad para ejercer el control sobre los recursos naturales en su interior" (1992: 102-103).

Without a doubt, Vázquez León advanced the project of approaching the Purépecha Region from one that emphasizes the cultural trait of language embedded in municipal contexts to one that emphasizes social power and ethnic organization to control resources in the region. His analysis was focused on the Sierra and ignored completely the rest of the Tarascan Region -- including the Lake Pátzcuaro area and the Zacapu region. In fact, even within the Sierra, he moved from a regional emphasis to a communitarian focus on the town of Tanaco and its involvement in political and economic problems related to forestry resources.

Within her focus on water and environment, Patricia Ávila García felt the need to provide a "delimitación regional de la Meseta Purépecha" in order to provide a context for the problems of water supply for localities in the highlands. Therefore, she limited her field studies to the Sierra and excluded the Lake Pátzcuaro area, presumably because scarcity of water is less severe at the lower altitudes around the lake. Ávila García stated that

She continued by noting that

Thereafter, Ávila García emphasized just the four core municipios -- Charapan, Cherán, Nahuatzen and Paracho -- as she presented information on the "evolución de la población hablante de lengua indígena en la Meseta Purépecha" for the period 1930-1990. Then, she analyzed the Economically Active Population for the same period and traced the "evolución de la población de la Meseta Purépecha por localidad y municipio" for the period 1822-1990. Ávila García demonstrated the importance of linking specific localities -- in the style of West (1948) -- to the data available for municipios. She was the first to state clearly that only some parts of "peripheral" municipios were relevant to the "core" municipios within the Sierra.

Undoubtedly, her analysis of inter-community conflicts -- some of which crossed municipal boundaries -- and the introduction of inter-community water pipeline projects (proposed by state and federal authorities) caused Ávila García to be aware of the mutual interaction of communities and municipios. Local politics, played out at the level of municipal elections, became manifestations of ethnic resistance on the part of Purépechas whose numerous petitions to the state and federal governments had previously failed to resolve the problem of water scarcity. The problem focus of her research showed the necessity for a broader perspective than could have been achieved within a "community studies" framework. The resultant regional perspective produced an excellent analysis of the interaction of natural resources and human settlements among the Purépechas in northwestern Michoacán.

Recently, Gunther Dietz published an impressive study of what he called "etnicidad, cultura y región en en movimiento indígena". After dealing with "desafios teóricos" and methodological questions, he proceeded to define the Purépecha region in the following terms:

In a significant discussion of "regionalización," Dietz recognized that the "poder de definir" a region such as that of the Purépecha has important consequences for the people who reside in the region. He discusses the ways in which the region has been defined by the Estado since the time of Lázaro Cárdenas (1934-1940) and the ways in which the region has been defined by academics since the pioneering studies of Moises Sáenz (1936), and the Proyecto Tarasco of the 1940s. In each case, Dietz sees that the state has played an important role in determining the ways in which the region came to be exploited by outsiders. In contrast, he argues that the "comuneros" whom he interviewed believe in a region that is connected through a series of social spaces: "la unidad doméstica;" "el barrio;" "la comunidad;" "la subregión en que se ubica la comunidad, que posee una toponimía propia: japonda (el lago), juáta (la sierra), eraxamani (la Cañada) y tsironda (la Ciénega);" "la región purhépecha, que abarca las cuatro subregiones frente a lo extrarregional, a menudo identificado con las ciudades mestizas colindantes;" y "por último, la nación mexicana, concepto que expresa una unidad compartida con los mestizos sobre todo frente a lo ‘gringo’" (Dietz 1999: 111).

Dietz does not deal directly with the rationale for including particular municipios in the region, but instead emphasizes the emic perspective of the Purépecha people whom he interviewed. In this context, he provides no complete listing of the 21 municipios that he says constitute the region, nor does he provide information about the distribution of speakers of the Purépecha language for the 21 municipios he includes in the region.⁵ Thus, it is clear that, for Dietz, municipios are less important for understanding the region than are comunidades, which serve as the basis for the collaborative enterprise known as la Nación Purépecha.

In 2003, Omar Masera, a scholar affiliated with UNAM, presented a comparative study of forestry practices in the Región Purépecha and the region of Calakamul, Campeche. In a brief PowerPoint presentation posted on the Internet, Masera included 19 municipios as defining the Región Purépecha. Perhaps because his focus is on deforestation, rather than on ethnicity or cultural issues, he does not give any explanation for why he included certain municipios while excluding others. While this brief presentation contributes little to our understanding of the Purépecha region, nonetheless it is noteworthy as one of the few intentionally comparative projects involving the Purépecha.

National Research Projects that Include the Región Purépecha

In 1996, as part of its program of publishing detailed reports resulting from the XI Censo General de Población y Vivienda, the Instituto Nacional de Estadística Geografía e Informática (INEGI) issued a sociodemographic profile of speakers of indigenous languages for Michoacán. The report offered quantitative data in 48 tables (plus an Anexo Estadístico with 11 more tables for municipios with 10% or more Hablantes de Lengua Indígena), without significant interpretations, on diverse characteristics of indigenous language populations in the state of Michoacán. Within the framework of the study, the INEGI defined

Following the guidelines used by INEGI at that time, the report emphasized that

This report included a brief summary of the characteristics of "ubicación territorial" for Purépecha speakers (persons 5+ years old). In seven municipios the indigenous persons represented more than 25% of the municipal population and in two – Charapan with 52.5% and Cherán with 52.3% -- more than half of the municipal population. Eight municipios in the Zona Purépecha had more than 5,000 speakers of the indigenous language, including Cherán, Chilchota, Nahuatzen, Paracho, Pátzcuaro, Quiroga, Los Reyes, Tangamandapio and Uruapan. In addition, the report showed that Coeneo, Charapan, Erongarícuaro, Morelia, Tangancicuaro, Tzintzuntzan, Zacapu, and Zamora had more than 1,000 speakers of Purépecha. In effect, the report found that the speakers of Purépecha were becoming an urban rather than rural population, in that more than 60% were living in places larger than 2,500 inhabitants and only 7% were living in places with fewer than 500 inhabitants. (INEGI 1996: 2).

This INEGI report, even though it depended only on HLI, provided a picture of a Purépecha population expanding beyond the rural homeland of the Sierra and lake region, with growing numbers of Purépecha speakers reported as living in the urban areas of Uruapan, Zamora, Zacapu, and even Morelia.

In the mid-1990s, the INI published an eight-volume collection entitled Etnografía Contemporánea de los Pueblos Indígenas de México (1995). In volume IV, dedicated to the "Región Centro," Arturo Argueta Villamar presented a summary monograph on Los Purépechas. In his chapter, he proclaimed that the Purépecha Region:

Having defined the Purépecha Region in this manner, Argueta Villamar continued with a description of the physiography, geology and soils, hydrology, climate, vegetation, and fauna. He then discussed the pattern of human settlement in the region, noting that "En los 22 municipios de mayor densidad de hablantes del p'urhé existen alrededor de 100 localidades que cuentan entre los 100 y los 5,000 habitantes" (1995: 230). This statement suggests that the principal rationale for the selection of the municipios included in the Purépecha Region was the density of speakers of Purépecha and the secondary rationale was the number of settlements with Purépecha speakers.

Starting from the premise that indigenous groups have been relegated to regions with extreme conditions and high risks, with few services and poor infrastructure, the INI published an impressive report entitled Riesgos y Desastres Naturales en Regiones Indígenas de México (2000) on some twenty indigenous regions in the nation. Listed as number four among these regions, the "Meseta Purépecha" is described in terms of the Población Indígena Estimada (PIE), using data from a baseline study done by INI in 1993 and a study of "división estatal y municipal" done by the INEGI in 1995. The geographical information provided for the Meseta Purépecha consisted of a large map listing twenty muncipios, with their official "claves" (as defined by INEGI): 16 Coeneo; 21 Charapan; 24 Cherán; 25 Chilchota; 32 Erongarícuaro; 39 Huiramba; 43 Jacona; 56 Nahuatzen; 65 Paracho; 66 Pátzcuaro; 70 Purépero; 73 Quiroga; 75 Los Reyes; 79 Salvador Escalante; 82 Tacámbaro; 84 Tangamandapio; 85 Tangancícuaro; 90 Tingambato; 100 Tzintzuntzan; y 111 Ziracuaretiro (INI 2000: 90). The map distinguished between the 12 municipios (numbers 21, 24, 25, 32, 56, 65, 73, 82, 84, 90, 100, and 111) with 30% or more PIE and the 8 municipios (numbers 16, 39, 43, 66, 70, 75, 79, 85) with less than 30% PIE.

Unusual about this regionalization was the inclusion of the municipios of Huiramba, Jacona, Purépero, and Tacámbaro – none of which were considered to belong to the Purépecha Region in any earlier studies. If these municipios, along with the somewhat problematic inclusion of Salvador Escalante, actually were showing significant numbers of PIE, it would be surprising in the light of their relatively low numbers of HLI as reported in both the 1990 and 2000 censuses.

Within its large-scale national project entitled "Perfil Indígena de México," begun in June 1997 through an agreement between the Departmento México del Banco Mundial and the Government of México, the Centro de Investigación y Estudios Superiores en Antropología Social (CIESAS) and several other governmental agencies developed a program of research in diverse indigenous regions of the nation. The first two phases of the research project were funded by the Banco Mundial, while the third phase was funded by CONACYT. The results of the project can be found on the Internet at http://www.ciesasistmo.edu.mx/ciesasweb/diagnosticoestatal.html.

As a component of this project, a team coordinated by Antropológo Luis Vázquez did research among the Purépecha, Otomí, and Nahuatl populations in different zones of the state of Michoacán. With regard to the Purépecha population, the CIESAS team defined the region as "Centro-Norte" and, based on the number of HLI in 1995, included in it the following 22 municipios: Charapan, Cherán, Chilchota, Coeneo, Erongarícuaro, Los Reyes, Nahuatzen, Nuevo Parangaricutiro, Paracho, Pátzcuaro, Periban, Quiroga, Salvador Escalante, Tancítaro, Tangamandapio, Tangancícuaro, Tingambato, Tingüindín, Tzintzuntzan, Uruapan, Zacapu, and Ziracuaretiro – although the researchers also note that in some cases only part of certain municipios (e.g., Los Reyes) should be considered to belong to the Purépecha region.

This report examined the "comunidad de lengua" as the first of several components of "identidad" for the Purépecha people. Other components included: comunidad de terruño, with emphasis on barrio identity within their pueblos; comunidad parental, with emphasis on kinship as well as padrinazgo and compadrazgo; comunidad de creencias, with emphasis on participation in the annual fiesta cycle honoring "los santos del catolicismo barroco popular purhepecha"; comunidad de signos culturales, with emphasis on clothing, music, dances, and celebrations (e.g., Año Nuevo); and comunidad territorial, with emphasis on the multi-dimensional connections between the people and the earth, including the judicial and political functions of the land as these relate to Purépechan ethnic organizations.

As part of the national project "Etnografía de las regiones indígenas en el nuevo milenio," conducted by investigators associated with the INAH, recently was published volume 3, entitled La comunidad sin límites: Estructura social y organización comunitaria en las regiones indígenas de México (Millán and Valle 2003). In this volume, the anthropologist Aida Castilleja and her research team provided a thorough análisis of "La comunidad y el costumbre en la región purépecha" (Castilleja et al. 2003). Their extensive article covered the history of the Purépecha region and their pueblos, the spatial and cultural characteristics of the people and their communities, and "la casa y el ciclo de vida" as well as "el gobierno local."

The authors recognized that

Castilleja and her colleagues gave special emphasis to "la comunidad" as the key feature for the social and cultural reproduction of the Purépechan ethnic group. Based on the distribution of the number of speakers of the Purépechan language (HLI), they delimited the area in which these communities are concentrated as the territory represented by all or parts of 23 municipios: Coeneo, Charapan, Cherán, Chilchota, Erongarícuaro, Jacona, Nahuatzen, Nuevo Parangaricutiro, Paracho, Pátzcuaro, Periban, Quiroga, Los Reyes, Salvador Escalante, Tancítaro, Tangamandapio, Tangancícuaro, Tingambato, Tingüindín, Tzintzuntzan, Uruapan, Zacapu, and Ziracuaretiro. They also appreciated that several thousand Purépecha speakers live outside the region defined by these 23 municipios, particularly in Mexico City, Guadalajara, and Tijuana, as well as in the United States. (2003: 41-42, Figura 1).

The researchers explicitly recognized that their regionalization is based on the one developed more than fifty years earlier by West (1948). They also followed his model of the four sub-regions (Lacustre, Meseta o Sierra, Cañada de los Once Pueblos, and Ciénega de Zacapu). Castilleja and her colleagues observed that these zones coincide with

Using data from the 1995 Conteo de Población y Vivienda, the authors observed that the greatest concentration of "pueblos indígenas" in both absolute and relative terms was to be found in the Meseta Tarasca, followed by the area of the Lago de Pátzcuaro, the Cañada de los Once Pueblos, and then the Ciénega de Zacapu. Again making use of the concept of HLI, they pointed out the uneven spatial distribution of the Purépecha population within the municipios. According to the authors, since the 1940s, the causes for this uneven distribution of the HLI in the region have included:

Castilleja and her colleagues recognized that

A Non-Municipal Approach to the Region

Writing within the framework provided by the editors of the Encyclopedia of World Cultures, Andrew Roth-Seneff and Robert V. Kemper focused on the concept of a "territorial homeland" for the Tarascans (or Phurhépechas):

In going beyond language use to ethnic identity, Roth-Seneff and Kemper did not employ the municipio as an analytical unit for defining the region nor did they provide a list of municipios that constituted the Purépecha Region. Instead, they gave attention to municipios as the setting for political and ethnic revitalization of the Purépecha heritage. For example, they observed that, since 1982, the celebration of the Phurhépecha New Year (Púrhépechaerhi Jimpanhi Wéxurhini)

Implications of the Review of the Regional Studies

Our review of the twelve studies of the Región Purépecha demonstrates that a variety of approaches has been taken over the past half-century to regionalize the area and its communities. To provide a framework for analyzing these studies, we offer Table 5, which arranges the municipios and the sources in a particular manner – namely, the municipios are ranked according to the number of times they have been included in the region, while the sources are ranked according to the number of municipios they included in their definition of the region. The result of this double sorting allows us to see that some municipios are included by all of the sources, while other municipios are included rarely.

[insert Table 5 here]

Furthermore, this data set can be analyzed using the well-known Guttman Scale procedure to determine the ways in which the set of municipios are perceived by the sources. The municipios fall into six categories. First, there are five municipios – Charapan, Cherán, Nahuatzen, Paracho, and Tingambato (in alphabetical order) – that were included by all twelve sources. These five municipios are clustered together in the Sierra. Second, there are five municipios – Chilchota, Erongarícuaro, Pátzcuaro, Los Reyes, and Tangancícuaro – that were included in all but one source, and this source (Vázquez León) focused his study on the Sierra to the exclusion of other sub-regions. These five municipios are located around the Sierra area, to the west, north, and east. Third, two municipios – Quiroga and Tzintzuntzan – fall together with ten of the twelve sources including them. The two sources that excluded them (Ávila García and Vázquez León) both focused their attention on the highlands sub-region and excluded the lake sub-region in which these two muncipios are situated. Fourth, four municipios – Coeneo, Tangamandapio, Uruapan, and Zacapu – are included completely by seven sources and incompletely by four others. These municipios are located on the four edges of the core Sierra area.

Taken together, these sixteen municipios represent the consensus of all the sources as constituting the core area of the Región Purépecha. The order in which the four sub-groups are arranged, in terms of their frequency of inclusion, also reflects their perceived relative importance with respect to the core area. In effect, the four sub-regions described by West (1948) and reiterated by several recent sources, are implicated in this ordering of the municipios and the sub-groups. For these sixteen muncipios, the Guttman Coefficient of Reproducibility (defined as 1- [number of errors/number of entries]) equals 0.96, where a perfect score is 1.00.

Beyond the first sixteen municipios, we find another group of six municipios: Nuevo Parangaricutiro, Tancítaro, Ziracuaretiro, Peribán, Salvador Escalante, and Tingüindín. About half of the sources reviewed favor the inclusion of all or some of these municipios in the Región Purépecha. In most cases, it appears that some localidades in these municipios have important "Purépecha" characteristics, even though the number of Purépecha speakers is significantly lower than in the sixteen municipios that we consider to be the core area. If we include this group with the first sixteen municipios, thus expanding the Región Purépecha to 22 muncipiios, the Guttman Coefficient of Reproducibility equals 0.92. This is an acceptable coefficient but not as significant as that for the first sixteen municipios alone. This suggests that more careful analysis, at the level of localidades, might help to determine which of this second group of municipios properly should be added to the Región Purépecha. Ultimately, we might reach the conclusion that sixteen municipios plus parts of these six other municipios represent the best model for the Región Purépecha.

Finally, there is a group of nine municipios which are included in the region only rarely by the sources. This group includes Jacona, mentioned both by Castilleja et al. (2003) and INI (2000); Ario de Rosales, included only by West (1948); Huiramba, listed only by INI (2000); Morelia, mentioned only by West (1948), Purépero and Tacámbaro, both listed only by INI (2000); Taretán, listed only by Masera (2003); Tocumo, listed only by Argueta Villamar (1995); and Puruándiro, listed by Kemper (1987) only for 1940 and then excluded for 1970. Clearly, this collection of municipios does not have consistent support from the sources. Their marginality to the region is demonstrated by the effect it would have on the Guttman Coefficient of Reproducibility, which would fall to 0.73, an unacceptable result for claiming a consensus among the sources.

Conclusion and Implications

In sum, our analysis of the literature reviewed above suggests that – despite the diversity of approaches taken to define the Purépecha Region – there is a consensus that sixteen municipios ought to be included in defining the core area. In addition, about half of the sources would find acceptable the expansion of the area to include six more muncipios, thus bringing the total to 22 muncipios. This is not an altogether unsatisfactory solution to the problem of what is the best way to approach the definition of the Región Purépecha.

Nevertheless, we are left with the challenge of going beyond the numbers and distribution of Purépecha speakers (HLI) or even the estimated indigenous population (PIE) for the muncipios of the region. To meet this challenge, we need to move beyond single variables – no matter how satisfying the results that they may yield – to recognize the cultural complexity of the region and its people.

Our review of the regional literature suggests that an alternative path might be followed in future studies. In fact, this path is suggested by looking at what is happening in the region at the present time. Increasing numbers of individuals and families have been emigrating, temporarily or permanently, from the Purépecha Region during the past thirty years. They have settled in nearby urban centers – such as Zamora, Jacona, Morelia, and Guadalajara – as well as in more distant cities, including Mexico City (and the adjacent areas of Estado de México), and Tijuana to the north. In addition, important enclaves of emigrants from the region have settled in diverse urban and rural areas of the United States, especially in California, Arizona, Oregon, and Washington in the west, as well as in Illinois, Indiana, Texas, and other states. Indeed, the distribution of emigrants from the region is constantly expanding in the North, as people follow networks of family and friends to new locations (Kemper 1994, 1995).

One impact of the significant emigration on the Purépecha Region has been to extend the sense of community far beyond the territorial homeland, as described by Roth-Seneff and Kemper (1995). In the process, the geo-political framework demarcated by municipios – whether 16 or 22 really does not matter – is being expanded as well. What West (1948) called the "Modern Tarascan Area" is no longer limited to the highlands of north-western Michoacán. Instead, even as the number of speakers of the language is failing to keep pace with the overall rate of population growth in the region, the culture parameters of Purépecha life are expanding beyond the core of about 100 communities traditionally associated with the region. Such a transformation ultimately calls into question the utility of defining "indigenous regions" in terms of geo-political boundaries such as municipios, rather than in terms of cultural affiliations. When cultural traditions such as patron saint fiestas are being replicated among emigrant enclaves in the North, and when the comisionarios of local fiestas now include men working and living in the North, we social scientists need to recognize what the people themselves have done (cf. Brandes 1988; Cahn 2003: 1-26, Chamorro 1987). They have extended the boundaries of their communities, and in the process, have expanded the Purépecha Region far beyond anything that the emperors of the pre-hispanic Tarascan empire might have imagined.

On the other hand, as Dietz (1999) pointed out, the power to define regions is a real power, and has important political implications. Thus, as social scientists, we must be aware that our definitions of the region can have an impact on governmental decisions about programs and planning for the people we study. In the case of Michoacán, the Secretaría de Planeación y Desarrollo Estatal (SEPLADE) of the state of Michoacán carries out its programming within a regional system for the entire state. Among their ten planning regions, one is called the "Meseta Purépecha," and encompasses 13 municipios. Eight of these fall into our consensus set of 16 municipios, and four others fall into our broader set of 22 municipios. Only one (Taretán) seems an inappropriate choice for the region. The other Purépechan muncipios identified in our review of the literature find themselves divided among four other planning regions. The existence of these multiple programmatic regions means that, for political and funding purposes, the government is dividing the potential for communitarian efforts within the Región Purépecha. For instance, the people of Santa Fe de la Laguna (in the municipio of Quiroga, and the SEPLADE region of Pátzcuaro-Zirahuén) are separated administratively from other communities with whom they work collaboratively, but which happen to be located in the Sierra (and thus fall into the Meseta Purépecha region).

In his Segundo Informe as Governor of the Estado de Michoacán, Antropológo Lázaro Cárdenas Batel offered a plan to revise the regionalization of the state (available on line at http://www.michoacan.gob.mx/gobierno/2doinforme.htm). The new regionalization would still encompass ten regions to cover the 113 municipios in the state. Under the proposed plan, Region VI would come to be known as "Región Purépecha", but would include only eleven municipios. The rest of the municipios within our consensus model of the region would be spread among four other regions. In effect, the municipios which we, as social scientists, associate with the Región Purépecha would still be situated among five of the ten regions proposed for the state of Michoacán.

The potentially divisive impact of these governmental regionalizations runs counter to the expressed desires of many organizations working toward enhancing the political power of indigenous groups, including those at work in the Región Purépecha.. For example, when the Congreso Nacional Indigenista held its third national meeting in March 2001 at Nurio, Michoacán, the fourth of its several demands was:

In closing, we can only imagine, as do many of the leaders of the Nación Purépecha, that in some future moment the government might come to recognize the communitarian solidarity of the people residing in the Región Purépecha by creating a more culturally appropriate regionalization of the state of Michoacán. In the meantime, the Purépecha people themselves are continuing to extend their communities into the rest of the nation and beyond the borders into the North. They may not reach the point of becoming what Castilleja et al. (2003) called "comunidades sin límites," but there is little doubt that they are transforming the traditional Región Purépecha into a significant territorial homeland (patria).

Notas

1. There are many variations in the orthography of Purépecha. In this article, we will use the spelling as it is found in the sources cited. Also, we use the term "Tarascan" interchangeably with Purépecha. In general, the term Tarascan was in common usage before the 1990s and since the 1990s has been replaced by Purépecha.

2. INEGI now provides complete information for the year 2000 census through the internet at no cost at the level of estados and municipios, whereas information on localidades is minimal. In the same manner, most states – including Michoacán – offer their own web sites with information at the level of municipios rather than localidades.

3. According to Valdés (2003: 133-134), in the census of 1921, there was a question about "pertenencia étnica o la llamada autoadscripción" for specific groups. The question was asked in this way, "¿siente usted pertenecer a la raza blanca, mestiza o india? The results revealed that one-quarter of the national population self-identified as "raza india." This created great alarm in government circles, because political leaders did not want to apportion 25% of the national budget to meet the needs of the indigenous population. So, it was decided that all persons would be treated the same, regardless of ethnic identity, and the question was subsequently eliminated from the census for 1930 and thereafter.

4. Curiously, none of the twelve sources reviewed for this article included the municipio of Zamora in the Purépecha region, even though two did include the neighboring municipio of Jacona.

5. The listing that we provide in Table 5 represents our best effort to interpret a small sketch map on which the numbers of the municipios are illegible and Dietz has shaded over all or part of those municipios which form part of the 21 municipios in the region.

Referencias Bibliográficas

Aguirre Beltrán, G., Regiones de refugio: El desarrollo de la comunidad y el proceso dominical en Mestizoamérica, México, Instituto Nacional Indigenista, 1967.

Argueta Villamar, A., "Purépechas", en INI, Etnografía contemporánea de los pueblos indígenas de México (8 vols.), México, 1995: 215-288.

Ávila García, P., Escasez de Agua en una Región Indígena: El caso de la Meseta Purépecha, Zamora, El Colegio de Michoacán, 1996.

Barabas, A. M. coord., Las regiones indígenas en el espejo bibliográfico, México, INAH, 2002.

Barragán López, E., "Formas espaciales y procesos sociales en la Sierra del Tigre," Relaciones 85: 105-129, 2001.

Bassols Batalla, A., México: Formación de Regiones Económicas, México, UNAM, 1979.

Beals, R. L., Cherán: A Sierra Tarascan Village, Washington, DC, Smithsonian Institution, Institute of Social Anthropology (Publication No. 2), 1946.

Boehm de Lameiras, B., "El enfoque regional y los estudios regionales en México: Geografía, historia y antropología," Relaciones 72: 15-46, 1997

Brand, D., Quiroga: A Mexican Municipio, Washington, DC, Smithsonian Institution, Institute of Social Anthropology (Publication No. 11), 1951.

Brandes, S., Power and Persuasión: Fiestas and Social Control in Rural México, Philadelphia, University of Pennsylvania Press, 1988.

Cahn, P., All Religions Are Good in Tzintzuntzan: Evangelicals in Catholic Mexico, Austin, University of Texas Press, 2003.

Castilleja, A., Gabriela Cervera, Carlos García Mora e Hilario Topete, "La comunidad y el costumbre en la región purépecha," en Saúl Millan y Julieta Valle, coords., La Comunidad sin Límites: Estructura Social y Organización Comunitaria en las Regiones Indígenas de México, México, INAH, 2003: 17-112.

CIESAS Unidad Istmo, "Ubicación en el territorio nacional," en CIESAS, Perfil indígena de México, México, CIESAS, 2003. (http://www.ciesasistmo.edu.mx/ciesasweb/perfilnacional.html)

Chamorro, A. "Fiestas oficiales y fiestas tradicionales en el municipio de Paracho, Michoacán", en Brigitte Boehm de Lameiras, coord.., El municipio en México, Zamora, El Colegio de Michoacán, 1987: 604-627.

de la Peña, G., "Presentación", en Guillermo de la Peña, ed., Antropología social de la región purépecha, Zamora, El Colegio de Michoacán y el Gobierno del Estado de Michoacán, 1987: 9-15.

de la Peña, G., "Los estudios regionales," en Carlos García Mora y Martín Villalobos Salgado, coords., La antropología en México: Panorama histórico 4. Las cuestiones medulares (etnología y antropología social), México, INAH, 1988: 629-674.

Dehouve, D., Ensayo de geopolítica indígena: Los municipios tlapanecos, México, 2001.

Dietz, G., La Comunidad Purhépecha es Nuestra Fuerza: Etnicidad, cultura y región en un movimiento indígena en México, Quito, Ediciones Abya-Yala, 1999.

Dow, J. and Kemper, R. V., eds., Middle America and the Caribbean (Encyclopedia of World Cultures, vol. 8), Boston, G. K. Hall & Co., 1995.

Embriz Osorio, A. y Laura Ruiz Mondragón, "Los indicadores socioeconómicos de los pueblos indígenas y la planeación de la política social en México", en François Lartigue y André Quesnel, coords., Las dinámicas de la población indígena: Cuestiones y debates actuales en México, México, CIESAS-IRD-Miguel Ángel Porrua, 2003: 85-114.

Foster, G. M. (with the assistance of Gabriel Ospina), Empire’s Children: The People of Tzintzuntzan, México, Imprenta Nuevo Mundo, Smithsonian Institution, Institute of Social Anthropology (Publication No. 6), 1948.

García Mora, C. coord., La antropología en México (Colección Biblioteca del INAH, 15 vols.), México, INAH, 1987-1988.

Garza, G., Cincuenta años de investigación urbana y regional en México, 1940-1991, México, El Colegio de México, 1999.

Giménez, G., "Territorio, cultura e identidades", en Rocío Rosales Ortega, coorda., Globalización y regiones en México, México, UNAM-Miguel Ángel Porrua, 2000: 19-52.

Instituto Nacional de Estadística Geografía e Informática (INEGI), Michoacán: Hablantes de Lengua Indígena – Perfil Sociodemográfico, Aguascalientes, AGS, 1996.

Instituto Nacional de Estadística Geografía e Informática (INEGI), "Características del censo," en Tabulados Básicos, Estados Unidos Mexicanos, Tomo I, XII Censo General de Población y Vivienda, Aguascalientes, INEGI, 2001: 1-27.

Instituto Nacional Indigenista, Indicadores socioeconómicos de los pueblos indígenas de México, México, 1993.

Instituto Nacional Indigenista, Riesgos y desastres naturales en regiones indígenas de México, México, 2000.

Kemper, R. V., "Urbanization and Development in the Tarascan Region Since 1940," Urban Anthropology 10(1): 89-119, 1981.

Kemper, R. V., "Urbanización y desarrollo en la región tarasca a partir de 1940," en Guillermo de la Peña, compilador, Antropología social de la región purépecha, Zamora, El Colegio de México y Gobierno del Estado de Michoacán, 1987: 67-96.

Kemper, R., "Extendiendo las fronteras de la comunidad en teoría y práctica: Tzintzuntzan México, 1970´1990", Estudios Michoacanos 5: 119-129, 1994.

Kemper, R. "Comunidad y migración: el caso del pueblo de Tzintzuntzan, Michoacán, 1988-1994", Relaciones 61/62: 133-148.

Kirchhoff, P., "Mesoamerica: sus límites geográficos, composición étnica y carácteres culturales", Acta Americana 1(1): 92-107, 1943.

Kroeber, A. L., Cultural and Natural Areas of Native North America, Berkeley and Los Angeles, University of California Press, 1939.

Lomnitz-Adler, C., Las salidas del laberinto, México, Joaquín Mortiz-Planeta, 1995.

Masera, O., "Modeling Deforestation in México and Implications for Carbon Sequestration Projects: Descripción de dos estudios de caso," México, UNAM, Centro de Investigación en Ecosistemas, 2003.

Moone, J., Desarrollo tarasco: Integración nacional en el occidente de México, México, INI, Ediciones Especiales No. 67, 1973.

Morayta, M., "Introducción: Señas y reseñas de los nahuas de Morelos", en A. M. Barabas coord., Las regiones indígenas en el espejo bibliográfico, México, INAH, 2002: 75-81.

Reina, L., coorda., Los retos de la etnicidad en los estados-nación del siglo XXI, México, CIESAS-INI-Miguel Ángel Porrua, 2000.

Roth-Seneff, A. y Robert V. Kemper, "Tarascans," en Melvin Ember y Carol Ember, coords., Encyclopedia of World Cultures, Vol. 8, 1995: 243-247.

Sáenz, M., Carapan: Bosquejo de una experiencia, Lima, Peru, Librería e Imprenta Gil, 1936.

Sandoval Forero, E. A., "Grupos etnolingüísticos en el México del siglo XXI," Papeles de Población 34, 2002: 219-235.

Smith, C., ed., Regional Analysis (Vol. II, Social Systems), New York, Academic Press, 1976.

Stanislawski, D., The Anatomy of Eleven Towns in Michoacán, Austin, University of Texas, Institute of Latin American Studies, Publication No. 10, 1950.

Stern, C. "Las regiones de México y sus niveles de desarrollo socioeconómico", Jornadas 72. México, D.F., El Colegio de México, 1973.

Stresser-Péan, G., "Les Indiens Huastèques", Revista Mexicana de Estudios Antropológicos 23(2-3), 1952-1953: 213-234.

Van Young, E., "Introduction: Are Regions Good to Think?" en E. Van Young, ed., Mexico’s Regions: Comparative History and Development, La Jolla, Center for U.S.-Mexican Studies, University of California, San Diego, 1992: 1-36.

Vázquez León, L., Ser indo otra vez: La purepechización de los tarascos serranos, México, Consejo Nacional para la Cultura y las Artes, 1992.

Vogt, E. Z. ed., Ethnology (vols. 7 and 8, Handbook of Middle American Indians), Austin, University of Texas Press, 1969.

West, R., Cultural Geography of the Modern Tarascan Area, Washington, DC, Smithsonian Institution (Institute of Social Anthropology, Publication No. 7), 1948.

Wolf, E., The Mexican Bajío in the 18^th Century, New Orleans, LA, Tulane University, Middle American Research Institute, Publication 17, 1955.

Wolf, E., coord., The Valley of Mexico: Studies in Pre-Hispanic Ecology and Society, Albuquerque, NM, University of New Mexico Press, 1976.