Revista Europea de Estudios Latinoamericanos y del Caribe 75, octubre de 2003 | 47 Reconceptualizing the Post-peasantry: Household Strategies in Mexican Ejidos

In his widely discussed book Micheal Kearney  (1996) questions the peasant concept in anthropology. Current global conditions, he argues, do  not favour the perpetuation of the peasantry.  Therefore, this category should be abandoned. Instead, Kearney proposes to call the globalised  postpeasant individuals polybians. Because they  carry out different economic strategies they are  slippery creatures, defy social bounds, and have  multiple social identities. Kearney is right in criticising the essentialism of earlier approaches.  However, his concept runs the risk of establishing  a new essentialisation. I will argue against the  point forwarded by Kearney and many others that  recent trends in globalisation have resulted in radical new household strategies of rural people.  Historical and ethnographical data of rural communities (ejidos) the Chenes region on the central  Yucatan peninsula, Mexico illustrate that the rural  economy has changed constantly – sometimes  slowly, sometimes radically – including processes of occupational multiplicity, wage labour and  migration. People adopted new or revived former  economic strategies in response to changing local,  regional, national and global processes. There was  no unilinear sequence from peasant to migrant.  During the ‘golden years’ of chicle exploitation in the 1930s and the early 1940s, for example, many  peasants became wage labourers in the chewing  gum industry. Later, a peasantisation process  developed due to the bust of the chicle economy  and increasing prices of corn. Resumen: Reconceptualizar lo post-campesino: estrategias domesticas en los ejidos  mexicanos En su libro ampliamente discutido, Micheal  Kearney (1996) cuestiona el concepto ‘campesino’ en la antropologia. Las condiciones  globales de hoy en dia, argumenta Kearney, dificultan la perpetuacion de ‘lo campesino’. Por lo  tanto, esta categoria debe entrar en desuso. En su  lugar, Kearney propone llamar polybians a los  post-campesinos globalizados. Ya que ellos son  criaturas inaprensibles al desarrollar diferentes  estrategias economicas, desafian limites sociales  y tienen multiples identidades sociales. Kearney  esta en lo cierto al criticar el esencialismo presente en perspectivas anteriores. Sin embargo, su  concepto corre el riesgo de convertirse a su vez  en un esencialismo. Yo discuto sobre la perspectiva que Kearny y otros autores proponen al decir  que el rumbo reciente de la globalizacion ha  producido estrategias domesticas radicalmente  nuevas entre la gente rural. Datos historicos y  etnograficos de las comunidades rurales (ejidos)  de la region chen en el centro de la peninsula de  Yucatan, Mexico, ponen de manifiesto que la  economia rural ha estado en constante cambio – algunas veces lentamente, otras mas intensamente-, visible a traves de procesos de ocupacion multiple, trabajo asalariado y migracion.  La gente ha revitalizado o creado estrategias economicas en respuesta a cambiantes procesos  locales, regionales, nacionales y globales. No ha  habido una trayectoria lineal en la transicion de  campesino a migrante. Por ejemplo, durante la  llamada ‘epoca dorada’ de la extraccion de chicle  de 1930 hasta comienzo de la decada de 1940,  muchos campesinos se convirtieron en trabajadores asalariados en la industria de la goma de mascar.  Tiempo despues, los trabajadores se volvieron  campesinos, debido a la quiebra de la economia del  chicle y el incremento en los precios del maiz.

In social anthropology, the economic strategies of rural populations were explored within the framework of peasant studies.For a long time 'peasants' were defined as rural cultivators whose most important means of livelihood was subsistence agriculture (understood as production for self-consumption).Their activities were seen as centred around the rural village/community, which was perceived as a relatively homogeneous universe with a distinct culture.Thus, peasants and peasant communities were representative of a specific socio-cultural type.'Community studies' was regarded as the relevant approach for the investigation of units of this kind.Authors like Robert Redfield (see e.g., 1947;1956) andGeorge Foster (1962;1965) assumed that for the most part peasant culture served the purpose of maintaining traditional structures in the allegedly homogeneous rural villages. 2 In the mid-1950s, Eric Wolf and Sidney Mintz, among others, had already begun to criticize this approach for not sufficiently taking into account the integration of peasants into national and international social, political and economic structures. 3They employed a historical perspective and concentrated on the interaction between rural populations and the dominant sectors of society.Internal peasant community structures were interpreted as a defence mechanism against threats produced by the unequal power relations between both spheres.The relationship between rural villages and the state, the impact of macro-economic processes, as well as the effects of development and modernization policies on rural communities and the internal differentiation of peasant societies increasingly became topics of anthropological research.Beyond this, there was recognition of the multiplicity of economic strategies pursued in the countryside (petty commerce, transportation, commodity production, rural wage labour and migration). 4 These developments in the discipline challenged traditional definitions of the 'peasant' as a rural producer who mainly worked his own piece of land (e.g., Wolf 1955, 453).Scholars began to classify the rural population into new categories such as 'smallholder' 5 , 'petty commodity producer ', 6 'peasant worker', 'semiproletarian' (e.g., De Janvry 1981) and 'peasant artisan' (e.g., Cook and Binford 1990), all of which reflected their research interests.However, 'peasant' is still the most widely-used term to designate the rural lower classes.Although the share of agriculture (mostly understood as tillage and livestock production) in the total of economic activities can be considered a crucial criterion for the choice of category, it is generally not specified.Thus, for example, the distinction between a wage labourer with a garden and an agricultural producer who temporarily sells his labour remains obscure. 7 All categories just mentioned suffer from the same limitation: the identity of a social actor (individual, household, group) is reduced to one or two economic activities or status.They disregard the existence of multiple economic strategies, the heterogeneous internal structure of the units described (e.g., age and gender distinctions) and processes of socio-economic differentiation within and between the respective sectors of the populations (see e.g., Brunt 1992, 11f, 15f).
This problem has recently been addressed by Michael Kearney in his comprehensive essay entitled 'Reconceptualizing the Peasantry' (Kearney 1996).For Kearney, the peasant term is an anachronism. 8In the current (globalized) world, the boundaries between rural and urban spheres are blurred because the rural population combines different sources of income and develops complex forms of reproduction: These varied ways of making a living are tactics in more complex economic strategies that nullify attempts to classify subjects and their communities on the basis of the standard productionist types (Kearney 1996, 147).
According to Kearney, parts of Mexico's rural population have to develop multiple identities as they move back and forth between different transnational spheres looking for employment opportunities e.g., as rural wage labourers or in the urban service sector. 9 The economy and many of the traditions of 'peasant-like' individuals still living in the rural villages are heavily subsidized by remittances from migrant relatives (Kearney 1996, 21, cf. also p. 147).Thus, agriculture has lost much of its importance for rural households.
In today's era of globalization, the existence of multiple economic strategies and identities does not permit the use of essentialist categorizations (Kearney 1996, 124f).Research must focus on the individual responding to the internal heterogeneity of rural households and communities.Individual actors are characterized as 'polybians' (Kearney 1996, 133, 141-143, 164-167) who, like chameleons, adjust to changing social, cultural and economic conditions: 10   To describe a polybian locally and ahistorically he or she might indeed appear in one context -perhaps at one moment in his or her life -as a peasant, in an-other as a plantation worker, and in others as a petty merchant or an urban slum dweller.Each of these identities is reified as the polybian's essential nature by the theories of external differentiation in concert with methodologies that are ahistorical and local in their scope.But these slippery creatures defy constructed social bounds; they cross out of their 'proper' places and enter into marginal spaces.And by populating these border areas they threaten normal social categories that the state has the responsibility to maintain (Kearney 1996, 141f).
I basically subscribe to Kearney's rejection of essentialist categories and to his claim for a historical approach in anthropology.This is also true for his emphasis on the flexibility of rural actors.Kearney deserves merit for calling attention to the multiplicity of economic strategies of rural people.Indeed, an ahistorical perspective or the exclusive concentration on local processes should be avoided.Such a narrow view runs the risk of mistaking a certain point in the life cycles of the actors investigated for their essential identity or way of life.However, Kearney's critique does not go far enough.
1) His description of transnationally migrating polybians applies to only part of the rural populations of the world.His approach was largely determined by his study of Mixtec and Zapotec migrants from Oaxaca in Mexico and California.Even in Oaxaca, large segments of the rural population do not migrate.The importance of migration varies tremendously between regions and villages.Not all migrants move in transnational spheres.Kearney unduly generalizes the situation of Mixtec and Zapotec migrants as applying to that of the rural lower class (e.g., in Mexico) on the whole.2) Kearney's primary concern is to explore the social identities arising from the multiplicity of economic strategies.He perceives them as the basis for political mobilization (Kearney 1996, 11, 146f).In his essay, Kearney shows little interest in discussing the actual conditions and relations of production or the reasons why rural actors pursue specific economic strategies.He does not consider the question of land tenure, for example, or make a distinction between shanty town dwellers, rural wage labourers or agricultural producers who till their own fields.Thus, Kearney omits fundamental differences in the living conditions of these people.In contrast to rural wage labourers or landless employees in the urban informal sector, those with access to land can at least partly supply their households with their own agricultural activity.Therefore, his approach remains of only limited value for the analysis of rural household economic strategies.3) Kearney dismisses the term 'peasant' and promotes a new era of 'postpeasant anthropology' inspired by the chaos theory.However, in the last analysis, his model remains unsatisfactory.In his effort to explain the formation of political movements (mainly new social movements), he merely shifts essentialism to another level: the sphere of identity.By employing the term polybian, he replaces one vague category (peasant) with another similarly fuzzy one.4) In addition, Kearney maintains the essentialist peasant term to describe former rural populations, notwithstanding his own critique of the concept.He thus rejects the peasant concept solely as a description of the people currently living in the countryside. 11He implicitly postulates a unidirectional development from peasant to polybian that is highly questionable.As will be shown in the following discussion, former economic strategies of rural populations were often heterogeneous and agriculture was rarely the sole means of livelihood.The problem of essentializing cannot be solved by inventing a new term.The term peasant or indeed any other term should not be employed to characterize the identity or personality of an actor but purely to describe a specific activity.

Economic strategies as roles
The combination of different economic strategies by rural households 12 has often been interpreted in the anthropological debate as a device for maintaining a 'peasant' -i.e., agricultural -lifestyle and traditional culture.It was also assumed that peasants lack all orientation for commercial gain (e.g., Redfield 1947, 305f).However, the tremendous heterogeneity observed among rural actors categorized as peasants makes such an assumption of motivations problematic. 13The following analysis of the development of rural household economic strategies in the Chenes region and the findings of research on rural-urban migration show that agriculture was far from being considered a constant goal in life.Therefore, instead of developing a cultural type, it seems much more useful to focus on the flexibility and heterogeneity of the actors.The following categorizations: peasant, woodcutter, bee-keeper, chewing-gum collector, wage labourer, artisan, migrant etc. will be employed in a purely descriptive sense.They describe certain economic activities practised by rural actors at a given moment and are by no means exclusive.Depending on the context, an individual can be a wage labourer in the forest industry and at the same time or at another period of his life be characterized as a peasant who tills his own land as an independent producer.
While larger enterprises are characterized as farms, people who use less than five hectares of land are defined as peasants.In the Chenes region, enterprises with more than five hectares frequently employ farm hands on a regular basis and are therefore capitalist units of production.Various activities in agriculture can be specified: e.g., milpa cultivation (swidden cultivation), mechanized cultivation (cultivation using agricultural machinery), irrigation agriculture (agriculture employing different systems of artificial irrigation).The cultivation of kitchen gardens in house lots (solares) will be excluded from the category of agricultural activities but will be treated separately under economic activities in the solar.Animal husbandry inside or outside the solares will be categorized as livestock breeding.Beekeeping will count as a distinct activity.Cultivation (tillage), livestock breeding and bee-keeping will be summarized as agricultural activities in the broadest sense.
The role approach helps to avoid essentialist assumptions.I follow Anthony Leeds (1977) who proposed considering 'peasant' as an economic role rather than a cultural type.The role concept 14 does not assume an essential identity of the rural actor but rather asks what activities are pursued in given circumstances and different contexts.Actors may play various roles simultaneously or at different periods in their lives.Beyond this, the role approach makes it possible to distinguish between expected behaviour (e.g., government agent expectations of rural actors' behaviour) and real behaviour (performance) of individuals or groups.
Thus, in contrast to the polybian concept proposed by Kearney, the role approach allows for both an understanding of the inner logic of single roles as well as for a coherent conceptualization of the often contradictory constraints and interests that impinge on individual actors and can result in role conflict.Actors frequently have to cope with role conflicts, as when a person playing two or more roles at one time may experience incompatible demands.Role assignment and interpretation is not necessarily based on social consensus but determined more often than not by power relations.Beyond this, roles are not static but continuously created and modified interactively, which gives them a structural and historical dimension (Abercrombie et al. 1988, 210).

Economic strategies of rural households in the Chenes region 15
The municipality of Hopelchen (also known as the Chenes region) is situated in the northeast of the Mexican state of Campeche on the Yucatan peninsula. 16A large section of the municipality's rural population is bilingual and speaks Yucatec Maya, apart from Spanish.Most Maya speakers (mayeros) live in the more than thirty villages with ejido organization, 17 while in the regional centre Hopelchen and three other larger settlements (Bolonchen, Dzibalchen, Iturbide), a Spanishspeaking 'ladino' population controls most commercial activities, works in the service sector or in local government agencies and conducts some farming. 18In addition, low German-speaking Mennonite colonists, who began to migrate to the region in the 1980s, make up an increasing part of the population.They live in isolated villages and concentrate on farming (Schüren 2001a).
Although often called Indians or indígenas by government agencies and the urban ladino middle class, the Maya-speaking inhabitants of the ejido villages have a regional identity as cheneros (people from the Chenes region) or campechanos (inhabitants of the state of Campeche) and classify themselves either as campesinos (peasants), ejidatarios (ejido members) or comuneros (local residents without ejido membership) of a specific village.However, while in some communities the majority of rural households has already given up agriculture, most households in other ejidos practise various forms of agriculture (swidden and/or mechanized rainfed, rarely irrigated agriculture) primarily to produce corn and squash.As a rule, agriculture is combined with several other economic strategies (e.g., bee-keeping, forest exploitation, production of handicrafts, kitchen gardening, animal husbandry, petty commerce, transportation, and temporary or permanent wage labour as farm workers or domestic servants in the region or migrant wage labour in the urban centres of the peninsula).Although government subsidies and credit have been radically reduced in recent years, access to direct aid for agricultural producers via PROCAMPO 19 is still important for many households (Table 1).
The relative importance of economic strategies varies between different villages (Table 1) as well as in different households from the same community.There are several determining factors, such as access to capital and land, distance to wage labour opportunities, access to local transportation, social networks, household structure and organization, consumption needs, risk and drudgery of a specific economic activity, government programmes, price policies and market demand.The multiplicity of rural household economic strategies has frequently been interpreted as a recent phenomenon and linked to neoliberal policies in the agrarian sector.Economic differentiation and land concentration have become more pronounced, particularly since the early 1990s, due to the liberalization of corn prices and a rise in production costs.Thus, many of the poorer ejidatarios have begun to sell or at least rent out some of their plots.The programme introduced to individualize and privatize ejido lands and reduce government support for rural producers has greatly facilitated this process. 21Many households are compelled to engage in several economic activities since agriculture alone is insufficient to guarantee their reproduction. 22Wealthier ejido households also combine several economic strategies, especially large-scale bee-keeping, commerce, transportation, and subsistence and commercial agriculture.They tend to buy or rent farmland offered in the ejidos and continue to invest capital in agriculture quite successfully.On the whole, time schedules of different economic activities do not come into conflict, as these households can afford to employ their fellow ejido members on a regular basis.This shows that multiple household activities are not always the result of economic necessity but can be a means of economic/capitalist expansion (Schüren 2003, 492ff).However dramatic the current processes may be, changes in the choice of rural household economic strategies in the Chenes region are far from new.

Economic booms and busts
Far into the twentieth century, the Chenes region was a frontier zone close to a sparsely inhabited 'no man's land' of tropical forest.It was not until the early 1940s that the region was connected by road to the central cities of the Yucatan peninsula (Campeche and Mérida), as well as to other parts of Mexico (Aranda González 1985, 184).Even the Mexican Revolution reached there with considerable delay as a result of its geographical isolation (Schüren 2001b).Thus, in the first half of the twentieth century, the Chenes region might appear to have been a perfect example of Redfield's folk society: isolated Maya villages were increasingly influenced by urban society ('folk-urban continuum', here in a diachronical sense).The economy of the region, however, had been incorporated into international -today we would say 'global' -markets since the late nineteenth century. 23 During the dictatorship of Porfirio Díaz (1884-1911) a large part of the Maya-speaking rural lower class in the Chenes region was forced into a relationship of dependency with the local oligarchy, which controlled access to land and water.They were exploited as farm workers and domestic servants by a system of debt peonage.While the regional economy was based on commercial production of sugarcane, corn, cattle and henequen, which was mainly exported to the US, the labour shortage posed a constant threat.Labour legislation was therefore extremely repressive and sanctioned the total control of farm workers by landowners and administrators.The latter had the right to administer justice and inflict sanctions.
Workers who wanted to leave the farm for short periods of time had to ask for permission.Corporal punishment was common.Farm workers additionally tilled swidden fields (milpas) for the subsistence production of corn, squash and beans, mostly on land controlled by the hacienda owners.
When debt peonage was abolished in 1914, hacienda owners -now deprived of their farm hands -were obliged to change their economic strategies.Whereas many of them intensified their activities in trade and transportation, others became entrepreneurs in the growing chicle industry.In the 1920s, the southern part of the Chenes region became one of Mexico's major chicle production zones.The 'white gold' extracted from the chicozapote tree (Achras zapota) was used as raw material for the production of chewing gum.Regional entrepreneurs from the Chenes region obtained concessions from the government of Campeche to exploit large forest areas but were dependent on loans from US firms, which primarily exported the chicle to the United States.
Without land of their own, only two viable options were open to the freed hacienda labourers.They could either rent land and become sharecroppers or work as natural resin gatherers in the growing chicle industry.Both cases implied a continuation of dependency relations with their former bosses.A large section of the rural population lived as wage labourers or sharecroppers during this period and could thus hardly be defined as peasants.
Although the revolution in Campeche had led to some important social and economic changes, local power relations were not seriously challenged.Isolated as it was, the Chenes region remained under the political control of a few prominent families, who remained the biggest land owners, the most notable merchants, carriers and chicle contractors.As a result of the mostly reactionary policies of Campeche's government, agrarian reform was introduced with vast delay.Established property relations were hardly affected during its first phase (from 1926 until about 1935) and the land distributed almost entirely national territory of poor quality.Thus, landowners had little reason to oppose land distribution.They accepted the limits of the ejido territories and were even present as guests of honour at the ceremonies to convey land to local ejidos.The plots distributed to the ejidos were far too small for the swidden system practised in the region, with the result that many ejidatarios had to persevere with previous sharecropping practices.
Only during the presidency of Lázaro Cárdenas (1934)(1935)(1936)(1937)(1938)(1939)(1940) did the Chenes region experience a cycle of social upheaval.In the course of the Cárdenas administration, pressure from the Campeche cooperative movement, the ejidal bank and the official party contributed to forming a gradually more progressive political climate and the acceleration of agrarian reform.The ejidal bank, which had been operating at national level since 1936, emerged as a new actor in local politics.It or-ganized peasants and workers and acted as an advocate against the arbitrary conduct of tradesmen and entrepreneurs by breaking their credit monopoly.Many ejidos sent petitions for more and better lands to the governor of Campeche with the result that land distribution increased.Even some of the private haciendas were affected by agrarian reform although most landowners were able to avoid huge losses by dividing their properties among family members, friends and straw men.At the same time, the chicle industry had already become the primary source of income for a large part of the rural population (including numerous women, who worked as cooks in the forest camps).Wage labour as a chiclero was far more consequential than agriculture, even for many of the ejidatarios, a fact that had profound effects on the development of agrarian reform in the region.In his report to the Agrarian Department in July 1938, a surveyor responsible for surveying future ejido land in Hopelchen wrote: [It] is impossible to fulfil the task, the reason being that we can neither count on the material nor on the moral support of the interested party.There were never more than four or five individuals available to carry out the survey.This is by no means enough to continue work on the boundary of the ejido territory.There is a prevailing state of utter disorder and cleavage within the working class in this place, which is the result of constant political quarrels and the complete lack of organization in the ejidos.This lack of participation in the agrarian reform by the rural working class was not confined to Hopelchen and was primarily due to the attraction of working as a chiclero.The chicle boom reigned from the mid-1930s to the late 1940s and most male village inhabitants were out in the forests extracting chicle for more than six months at a time during the rainy season.Some chicleros continued to work their fields during the rest of the year, but many households preferred to buy corn rather than produce it themselves.As Moisés de la Peña remarks in his significant study of the economy in Campeche in 1942: When the chicleros arrived back from the forests after a six-month absence, life returned to the towns and villages and trade flourished for a few months.After their arduous, dangerous and isolated work in the forests, the chicleros wasted most of their income on drinking bouts and other forms of entertainment.When there wasn't a peso left, they asked for advance payments on the next chicle season.[...]Under these circumstances, it is obvious that Campeche peasants have little interest in practising agriculture.
Even today, many older ejidatarios can work up enthusiasm for the golden age of chicle and the relatively high cash income earned from chicle gathering (Schüren 2001b).
It was only when the chicle industry experienced a severe crisis at the end of the 1940s, following a cutback on US imports and a dramatic fall in chicle prices, that the milpa system of agriculture was to enjoy a revival as the most important economic activity for much of the ejido population in the centre and south of the Chenes.As a consequence, the latter began to appreciate ejidal lands.A process of 'peasantization' was fostered in the 1950s since no alternative, comparable sources of income had come to the fore after the decline of the chicle business.Although some employment opportunities developed in the construction and lumber industries, the end of the chicle boom had left a gap that could only be filled by increasing agricultural activities.The growth of agriculture was also stimulated by government programmes to develop corn production.Corn prices soared 24 and cultivated milpa areas expanded considerably.Although it was mainly large-scale commercial producers and tradesmen who benefited from the new guaranteed prices and official warehouses, many ejido members and former chicleros took up commercial agriculture in large swidden areas, some of whom even employed labourers (Schüren 2001c).
Contradicting trends have been observed in the Chenes region since the late 1960s.While the government endeavoured to promote ejidal agriculture and keep the rural poor in the countryside, environmental degradation and access to fresh sources of income was fostering depeasantization.As a result of deforestation, the agricultural potential for milpa production in the Chenes region changed radically.Yields declined and the increasing growth of weeds made milpa production more arduous and less profitable.An adequate supply of land is an essential requirement for sustainable milpa production, allowing producers to cope with the fallow periods when vegetation has to recover its agricultural potential.The quality of the forest felled has a decisive influence on milpa production yields.The higher and thicker the secondary growth, the more nutrients are left for crops after a successful milpa burning.High milpa yields in the first year depend on the relative age of the forest, which implies, 1) a reduction of weeds due to the shade, and 2) a larger amount of biomass converted to fertilizing ashes through the burning.Therefore, milpa productivity is higher in high forest areas and labour costs lower because weeding can be reduced to a minimum (cf.Rosales González 1988, 120ff;Ewell and Merrill-Sands 1985, 107f).
Corn prices dropped by 33 per cent between 1963 and 1972.Many rural producers abstained from expanding their increasingly labour-intensive but less profitable commercial corn production on milpas exposed to environmental risk.In fact, both small and large-scale commercial producers in the Chenes region reduced the size of their milpa plots while simultaneously diversifying and adopting alternative and more profitable economic strategies.For many households, milpa corn production ceased to be the most important economic activity.The decrease in commercial milpa production in the Chenes region was accompanied by a change in the significance of agriculture in general (Schüren 2001c;2003).
As emigration to urban centres (mainly Campeche and Mérida, later also Cancún and the Caribbean coast) increased, the government inaugurated several development programmes.Small irrigation projects in a few ejidos in the Chenes region, mainly directed at developing citrus production in collectively organized ejidos, were set up in the second half of the 1960s.However, these projects were not successful (cf. Messmacher 1967, 147;Gates 1993, 143-151;Schüren 2003).Within the framework of President Echeverría's large-scale investment programme for rural development (PIDER) in the 1970s, further government measures were introduced, including the distribution of credit and equipment for commercial beekeeping and the mechanization of ejido lands.In addition, village infrastructure (the construction or extension of potable water and electricity systems) was improved and several new roads were built (Quintal Avilés 1976).
Large-scale government project clearings carried out since the 1980s proved to be even more detrimental to forest areas than the earlier commercial milpa production.Due to inadequate planning many of the clearings were useless. 25More than 10,000 hectares of bottomlands (of which only about 5,000 were actually used) were cleared in the 1980s in Chunchintoc and 500 hectares in Xmaben for the production of dry rice.While bee-keeping had become an attractive economic activity in the rural villages, most other programmes constituted what Gates (1993, 133ff) has termed an 'industry of disasters'.As a result of poor planning they failed, leaving ecologically impoverished grasslands in their wake.Large-scale cattle ranching programmes that never functioned had the same effect.The mechanization of agriculture in many of the ejidos did not have the desired effect of improving standards of living permanently but fostered instead the producers' dependence on the government and on expensive means of production (pesticides, hybrid seeds, fertilisers, gasoline, tractors etc.).From the very beginning, the amount of mechanized land per producer, and especially the technology supplied to the ejidos, was inadequate for profitable production.Ejidatarios were subject to control by government agencies including the Rural Credit Bank.Their agricultural knowledge was disregarded.Crop failures were frequent and the producers, increasingly confronted with fraud and politicking, lost much of their motivation.Loans could not be repaid and the majority of ejidatarios fell into debt. 26Although many of these projects failed and there was no evidence of a significant increase in agricultural output, ejidatarios grew dependent on financial support from the government and 'farming the programmes' (Barlett 1989, 263) became an important economic strategy in its own right.
As mentioned earlier, government aid for the ejido sector is decreasing under the current neoliberal policy, while the process of depeasantization has been accelerated.Since opportunities for commercial production and wage labour are becoming scarce at local and regional level, migration is now an option for many of the younger generation.Nevertheless, permanent labour migration and other forms of employment implying a high degree of mobility (seasonal employment outside the region or commuting) are mainly restricted to the Yucatan peninsula (see also Gates 1993, 235).The urban centres of the peninsula, Mérida, Campeche and especially the tourist enclaves along the Caribbean coast, absorb a large part of the labour force from the rural areas.Thus, transnational (permanent or temporary) migration to the US, which is so prominent in other Mexican states, has little bearing in the Chenes region.This is more a consequence of inadequate migrant networks in the US than of the cheneros' notorious love for their home country.

Conclusion
I have done almost every kind of job in my life.I was a milpero and worked as a lumberjack.I cut mahogany and harvested sugarcane.I was a farm hand, a builder's assistant and I worked on road construction.In the past, there were no machines to build roads.We built the highway to Campeche with our own hands.It took years.Sometimes I was away from home for six months.Life was not easy but my six children got a good education.They have all become teachers (Interview with a former ejidatario in Hopelchen, March 1996).
Authors following Robert Redfield have characterized the rural population of socalled marginal regions of the Yucatan peninsula in the first half of the twentieth century as mainly traditional subsistence producers.Their integration into the commercial economy, wage labour and the spread of 'modern urban culture' has been described as a continuous process of diffusion (e.g., Redfield 1941;1956;Press 1975).However, there has been no consistent trend in agricultural change in the Chenes region.While the hacienda economy was once characterized by forced labour, most rural households became engaged in the chicle industry during the chicle boom in the 1930s and 1940s.Although agrarian reform was promoted after the Revolution, peasantization only became widespread after the chicle boom drew to a close in the late 1940s.As a result of government price policies and a lack of alternative sources of income, many rural households fell back on agriculture.Later on, environmental degradation caused by extensive commercial milpa farming led to a serious crisis in this form of agriculture.Forest exploitation in the Chenes region enjoyed a short boom in the 1970s and 1980s, after which it declined rapidly as a result of overexploitation and a dramatic decrease in prices when Mexico joined GATT.
In the 1960s, the expansion of the construction industry (with a particular focus on road construction and public buildings), industrialization and the oil boom in Campeche encouraged migration to urban centres.The dismantling of the ejido in the 1990s, a cut in government credit and a dramatic drop in crop prices (following the creation of NAFTA) still pose a threat to Mexican agricultural producers.The general economic trends outlined here for the Chenes region by no means affected all households in the same manner.The choice and combination of multiple strategies used has varied tremendously, even among households in the same village.
Michael Kearney and many others argue that the current process of 'globalization' necessarily leads to radical new household strategies and social identities.From my historical study of the municipality of Hopelchen, it is evident that people have always had to adopt new economic strategies or readopt old ones in response to changing local, regional, national and global processes.The impact of the chicle economy (tightly linked to US economic interests) was as significant at the time as labour migration to Cancún is today.The combinations of different economic strategies that Kearney and others characterize as 'depeasantization' are not really new.As the example of Campeche clearly illustrates, the relationship between agriculture and wage labour and its relative importance for the reproduction of rural households has changed in the course of time in accordance with the overall economic situation.The interplay of environmental, cultural, economic, social and political factors has influenced the choice of household strategies employed by rural people, forcing them to become either chicleros, artisans, woodcutters, swidden farmers, bee-keepers and farmers on mechanised land, tourist guides, wage labourers, government aid recipients or all of them at once.There have also been processes that could be called 'repeasantization'.The combination of different economic strategies is in no way a recent phenomenon. 27Kearney rejects the peasant category for present-day social actors because it reduces a large part of the rural population to mere agriculturalists and a certain cultural type.Nevertheless, he retains this category to denote rural actors in the past.
In this paper, I alternatively proposed a role model to describe rural household economic strategies following Leeds (1977) and others, who perceive economic actors as role players and employ the terms peasant and chiclero, etc., merely to describe a concrete and well-defined activity.Each role must be investigated in its economic, social, cultural and political context in order to understand why it has been chosen.In contrast to the still frequent use of the peasant concept (and other restricted categorizations), the role approach conceptualizes the heterogeneity of economic strategies and the flexibility of the actors concerned as they adapt to changing economic surroundings.Furthermore, unlike e.g., Kearney's 'postpeasant' or 'polybian' term, this Chambers and Conway 1991;Scoones 1998;Bebbington 1999;Ellis 1998, 1f;2000;Zoomers 1999;2001. 2. Rural ways of life and their attendant economies were, of course, studied by other disciplines as well.The topic was first treated profoundly in the 1950s with the introduction of the peasant type, described as rural producers distinct from capitalist farmers, on the one hand, and from the socalled 'primitives' on the other (see e.g., Wolf 1955;Redfield 1956).While the latter were perceived as not being integrated into the surrounding state, peasants were defined as part of the dominant society.3. See e.g., Wolf 1955;Mintz 1956;Wolf 1957;Wolf and Mintz 1957. 4. For discussions of the many different theories and directions in peasant studies, see e.g., Dalton 1971;Ennew, Hirst and Tribe 1977;Long 1977;Silverman 1979;Hewitt de Alcántara 1984;Ellis 1988;Roseberry 1989;Cancian 1989;Blum 1989, 9-136;Wimmer 1995;Kearney 1996;Bryceson 2000. 5. See e.g., Netting 1989;1993;cf. Bryceson's discussion of the usage of this term (2000, 19-28). 6. See e.g., Friedman 1978;Smith 1984;Smith 1985;Cook and Binford 1990. 7.Many authors do not even apply the classic criteria suggested by Wolf (1955, 543-455;1957, 1;1966, 9) for the definition of peasants that include control of the land, production as a means of subsistence (and not of reinvestment or profit), and subordination to a dominant economic and so-cial system.8.According to Kearney the peasant term introduced after the Second World War created a new subject for investigation following the decline of the 'primitive', and thus secured anthropology its bread and butter (1996,(30)(31)(32)(33)(34)(35)(36)(37)(38)(39).9.The actors follow communication networks (reticula) that have neither a formal structure nor a beginning or end: 'Thus, the social morphology of networks is like an amoeba, a creature with complex internal differentiation but without distinct cells or organs that correspond to the social components of corporate communities.Also, unlike official communities, which are bound to delimited spaces, amoebas can extend themselves in any and all directions' (Kearney 1996, 125).10. 'A first approximation to denote the identities of complex transforming types such as migrants, who move back and forth from 'peasant' to 'proletarian' life spaces, is to speak of them as amphibians (amphi, both, + bios, mode of life).But biological amphibians, which spend part of their lives in water and part on land, are but dual creatures, whereas the kinds of categorical migrants with which we are dealing move in and out of multiple niches and are therefore more correctly called polybians (poly, many).The polybian is rather like a chameleon, which can adjust its color to match that of its immediate environment.But unlike such protective coloration, which is but skin-deep, polybians adapt their being to different modes of existence as they opportunistically move in and out of different life spaces' (Kearney 1996, 141; his emphasis).11.Besides, Kearney is inconsistent when he continues to refer to parts of the present rural population as 'peasantlike' or 'postpeasant'.12.I follow David Preston's definition of households that takes the internal heterogeneity of such units into account, a fact that is often neglected by other approaches.He writes: 'Households are commonly defined as comprising those living and eating together and include other people, not biologically related, who are part of the sharing unit.(...) Households comprise individuals of different gender and age, and with distinct interests and personalities.What people do to sustain themselves is a function of the opportunities available, individual preference and group needs interpreted in the light of local custom.The range of activities of all members of the household makes up a collective household strategy, some elements of which may primarily satisfy individual needs but others of which may be undertaken primarily out of responsibility to the group as a whole.The sum of all activities, both individual and collective, represents the means by which individuals and groups ensure their continued existence -their livelihood strategy' (Preston 1994, 203, 206).13.See also Roseberry 1989, 123.The notion that peasants are conservative and traditionalistic and therefore pursue goals unlike those of capitalist enterprises had already been rejected in the 1960s by the economist T.W. Schultz (1964).He claimed that the peasant orientation was not a result of traditionality but a reflection of poverty (see also Harris Ed. 1982, 212-215).14.Roles are generally understood in anthropology and sociology as the total of behaviour expectations associated with a certain social position (e.g., Banton 1985, 714f; see also Sarbin 1968, 546;Turner 1968, 552f;Leeds 1977, 229f ). 15.The following presentation is based on the analysis of data from different archives in Campeche, Hopelchen and Merida, on interviews carried out during my field research since 1993 and on secondary sources.16.On 31st December 1996, the southern part of Hopelchen (which comprised 34 per cent of its former territory) was ceded to the newly-created municipality of Calakmul, which includes the Calakmul Biosphere Reserve (founded in 1989) (Diario de Yucatán 2/1/97).According to recent census data, the municipio has 31,214 inhabitants and a population density of 4.18 persons per square kilometre (own calculation based on INEGI 2002, 31f).17.The Mexican government introduced the ejido after the Revolution in 1910-1917 as a system of collective land tenure offered to landless petitioners in rural areas.While the land was kept under the domain of the nation, membership was limited to a group of (mostly male) officially recognized usufructuaries.These ejidatarios had a permanent right to work the land and participate in ejidal assembly (asamblea ejidal) decisions.The usufruct right to land could only be ceded to or inherited by a selected person.Ejido land grants were first given to inhabitants of villages, later to workers of former haciendas and to colonists of the so-called New Centres of Ejidal Population (NCPE).Generally, ejido territory consists of an urban zone (zona urbana) where houses are built, an area for collective use (área de uso común) and often individual plots worked by ejidatarios and their families.Up to 1992, when agrarian legislation changed, ejido members were obliged to work the land personally and stood to lose their derecho agrario in the case of long absence.Legally, ejido land 20

Table 1 .
Contribution of economic strategies to the gross household income in twoejidos, 1994 (percentage) approach avoids new essentialistic connotations.Ute Schüren is an anthropologist and teaches at the Lateinamerika-Institut, Freie Universität Berlin.Her main research interests are economic anthropology, and the social anthropology, ethnohistory and archaeology of Mexico and Central America.Recent publications include: Rationalität oder Irrationalität bäuerlichen Wirtschaftens im Kontext staatlicher Politik?Haushaltsstrategien in mexikanischen Ejidos; Das Beispiel der Chenes-Region-Campeche (The peasant economy -rational or irrational?Household strategies and the state in Mexican ejidos.The example of the Chenes region, Campeche) Freie Universität Berlin, 2003 (Ph.D. dissertation, published online in German with an English summary at http://darwin.inf.fu-berlin.de/2003/237/); and 'Economic Strategies of Rural Producers: A Comparison of Ejido and Mennonite Agriculture', in A. Zoomers (ed.) Land and Sustainable Livelihood, Royal Tropical Institute, Amsterdam, 2001.<schueren@zedat.fu-berlin.de>Notes 1. See e.g.,