The politics behind avocado production

Abstract

This article ethnographically examines the formation of autodefensas in the Mexican municipality of Tancítaro, and its ties to the avocado production industry. By conducting a political and processual analysis, this article pays close attention to the autodefensas and the class relations among them vis-à-vis the avocado production industry, how class tensions influenced their politics, and how wider claims for justice were dismissed in favour of security. The class analysis is linked to an examination of the historical process of liberalization of the agrarian sector and class fragmentation that took place in the nineties, by which the peasantry was overlooked in favour of a free-market-oriented project that enabled the avocado export market to be opened up. In doing so, the article presents a complicated social field which challenges the conventional romantic view of autodefensas in Mexico.

On February 24, 2013, groups of male residents from the agro-industrial valley of Apatzingán in Michoacán (south-west Mexico) took arms to liberate their towns from drug-cartel rule. They called themselves autodefensas (self-defense groups) and claimed to have been abandoned by the state. As the year progressed, they advanced into different parts of the territory and were featured on the regional, national and international news, often portrayed as heroes (Zamorano, 2018). Official records reported that autodefensas entered over thirty municipalities, including Tancítaro (33,453 inhabitants), the self-proclaimed avocado capital of the world (CNHD, 2016: 24).

Having emerged within a violent context related to drug-trafficking and the militarization of public security, the autodefensas phenomenon has mainly been discussed by academics as a collective effort to overturn criminal rule. While people in Michoacán lived everyday life in a violent context (Maldonado, 2018; Mendoza, 2022; Pansters, 2015), the fact that the autodefensas were largely led by agro-industrial producers or ranchers, whose livelihoods were connected to international export markets, was hardly discussed.

This article analyzes the case of Tancítaro, where avocado producers led the local autodefensas and, in the aftermath, consolidated a dominant position in local politics. Using a class analysis to explore their political strategy, I intend to show that, by claiming their right to command the armed defense of Tancítaro, they were able to establish security as the primary societal concern, while dismissing local calls for social justice and thereby safeguarding the export market. I argue that the autodefensas phenomenon is best grasped by considering three interlocking processes. The first was the introduction of Hass avocados in Michoacán in the 1960s, and how this novel crop became entangled with local politics and social relations of production. The second was the agrarian counter-reform of 1991, which ended the long-standing state policy of land redistribution, enabling the privatization of collective property via neoliberal reforms. ¹ These reforms preceded the opening up of the avocado export market within the framework of the North American Free Trade Agreement (NAFTA) in 1997. While avocado production had proved to be profitable, the chance to export to the US was what incentivized many ejidatarios ² to acquire land and turn to avocado production. The third process was the period of criminal rule starting in the mid-2000s, involving the atomization of drug cartels and the militarization of public security in Mexico. This created the conditions for the subsequent formation of autodefensas and their mass support from residents of all social classes. However, if these tumultuous years were not consolidated into a long-term common political project but rather, as I will show, into ongoing political disputes that fragmented the autodefensas, it was due to the inherent tensions in the relations of production (Smith, 2015: 74).

The 1991 agrarian counter-reform resulted in a process of class fragmentation (Bernstein, 2006; Kasmir and Narotzky, 2025). It atomized the peasantry into numerous “agricultural producers” and associations that competed against each other and were incapable of political unity (Macip Ríos, 2005). My use of the term “producer” is twofold: it is the emic category I encountered most often in the field, but it also references the dismantling of the agrarian structures of the 20th century. As a result of this, avocado producers emerged as independent producers aiming to sell their produce to export markets. I am interested in showing what a process of class fragmentation looks like politically in a context with a violent history (Gledhill, 2015; Stack, 2022).

In this article, I use the extended case method inspired by the work of Max Gluckman and developed by the Manchester School (see Evans and Handelman, 2006). This method traces events involving the same sets of actors over a long period of time, underlining the importance of the process (Mitchell, 2006: 29). My development of this case has been strongly informed by historical analysis (see Gledhill, 2015), in order to show the “manifestation in time of economic and political forces that shape human action” (Wolf, 1986: 326). The main event here is the assault by the Tierra Caliente autodefensas in Tancítaro, which enabled the formation of the local autodefensas. These were supported by people of all social classes, but quickly fragmented along class lines. In this article, I show the connections between the three above-mentioned processes, drawing on two main periods of ethnographic fieldwork corresponding to two different research projects: one focused on social responses to criminal violence (March 2017 to January 2018), and the other on regulatory frameworks for expanding avocado production (March to May 2022). I use pseudonyms for all my informants, except for public figures and those who were murdered.

A class-based analysis of the autodefensas

“We were divided from the start,” explained Victor, a Tancítaro autodefensa leader in his thirties and, at the time of my fieldwork, a senior member of the unity government representing the left-wing Partido de la Revolución Democrática (PRD). We were talking about “the movement” near a barricade. “Those who took up arms,” he said as he traced two circles on the ground using a branch, each split in half, “divided between them, who allied with Papa Smurf, and us, who support Dr. Mireles [two of the regional autodefensa spokesmen].” One circle represented the autodefensas, and the other represented “the rich, they own the empaques [packing houses] and controlled the Board and everything related to avocado [production]”. The circle representing the rich was also divided in half: “some of them decided to support us [the autodefensas], they thought that way they could control us, but no, they couldn’t” (Fieldwork diary, May 27, 2017).

The term autodefensas evokes a situation of vulnerability caused by a lack of protection. In the case of the Michoacán autodefensas, it points to the need for self-protection created by the perceived retreat of the state in the context of criminal violence and neoliberalism (Zamorano, 2018). The autodefensas were formed by heterogeneous groups of residents, including farmers, ranchers, manual laborers, schoolteachers (Le Cour Grandmaison, 2016: 3) and even members of the military (Maldonado, 2018: 167–9). Using the name autodefensas diluted their heterogeneity and class differences. Instead, a more homogeneous group appeared. ³ The autodefensas emerged within the context of the militarization of public security, along with increasing numbers of desaparecidos (missing people) and homicides. This occurred during what was initially called the “war on drugs”, launched by President Felipe Calderón in 2006, the first such security strategy of the 21st century in Mexico to mandate deployment of the armed forces and federal police to combat drug cartels. These conditions are likely to have imbued the autodefensas with a sense of legitimacy, sparking great interest among academics, journalists and the wider public. ⁴

Political scientists, sociologists and anthropologists took an interest in the phenomenon. While acknowledging the heterogeneity of the autodefensas, some of this research focused on their collective aims of protecting “the moral economy of drug production”, defending their territory or restoring the rule of law (Álvarez, 2020; Fuentes-Díaz and Paleta, 2015; Le Cour Grandmaison, 2016). In the case of Tancítaro, some scholars conceptualized the struggle as an act of “collective vigilantism” or identified the “driving force” in their collective identities, forged through past experiences of oppression (Moncada, 2021: 122; Wolff, 2020: 37). Other authors explored the collective process of citizenship-building while acknowledging internal conflict among the autodefensas (Curry, 2019). While the analytical approaches taken were diverse and summarizing them exceeds the scope of this article, most authors coincided on the autodefensas’ collective goals and the existence of a mutual enemy, that is, the drug cartels extorting local populations.

In Tancítaro, the local autodefensa groups included avocado producers, agricultural laborers, avocado cutters, ⁵ shopkeepers, seasonal migrants, and construction workers. Moncada (2021) argued that the local residents, who were tired of living under criminal rule, joined forces to liberate their towns and villages, regardless of their class or political differences. While Moncada (2021: 132) contended that residents were successful in expelling Los Caballeros Templarios (the Knights Templar drugs cartel), “the movement” failed to consolidate further changes precisely because of its class differences. I suggest that an analysis focusing on the autodefensas’ class relations rather than their group unity may unveil the hidden complexities of the case, and might explain their rapid collapse and even their fragmentary nature (Wolff, 2020: 44).

Class is best grasped in terms of relations. For Marx, class relations can be framed as oppositions between producers and non-producers (Roseberry, 1997: 31). Here, I am interested in the broad opposition between avocado producers and non-avocado producers. This more material or concrete form of class contrasts with the potential (hence, political) form of class put forward by Smith (2015), which he termed “praxis”. With praxis, he referred to the “willed intervention of agency into the making of history to modify its unfolding” (2015: 80–82). Building upon Smith, we may understand praxis as a potentially collective agency to destroy the social configuration enabling the relations of production from where such agency emerged. I contend that, although the autodefensas had massive popular support, resulting from the rage and powerlessness experienced while living under criminal rule, it was not an emancipatory movement. Residents of different social classes united to overcome criminal rule by a drug cartel, that is, the mutual enemy. By expelling Los Caballeros Templarios and participating in organized patrolling of the municipality, autodefensas of all social classes ended up safeguarding avocado production and producers’ access to the export market.

Introducing Hass avocados in Michoacán

The avocado (persea americana) is a fruit that originated in Mesoamerica approximately 6,000 to 7,000 years ago (Galindo-Tovar et al., 2008: 442) but spread across the world, initially through the Spanish empire. The Hass avocado was developed in California in the 1920s as part of a regional campaign to produce improved varieties and was brought to Uruapan (Michoacán’s second largest city) in the 1960s because this location offered optimum weather, rainfall and soil for cultivating this crop (Hernández-Fernández, 2023: 131). Uruapan is located at a juncture between two regions: the fertile valley of Apatzingán, (also known as Tierra Caliente, or the Hot Land), and Sierra P'urhépecha. The valley of Apatzingán is a region historically used for agriculture and ranching, where development projects led to increased prosperity in the second half of the 20th century (see Barkin and King, 1970; Boyer, 2015). The new paved roads, heliports, and agricultural investments also benefited local drug production (Maldonado, 2013: 48). Sierra P'urhépecha is historically a marginal region with an uneven terrain, a scarcity of land suitable for agriculture, and dense conifer forests that were of no interest to the colonial and early Mexican states.

General Lazaro Cárdenas del Río was an important figure in the institutionalization of the post-revolutionary Mexican regime. He was state governor of Michoacán (1928–32) and later president of Mexico (1934–40). As governor, Cárdenas incentivized the formation of forestry ejidos and cooperatives in the highlands (Boyer, 2015). In the 1960s, entrepreneurs, civil servants and engineers invested in Hass avocados in Uruapan, and later expanded to neighboring municipalities like Tancítaro. The produce was sold in regional and national markets, due to a ban imposed by the US in 1904 on Hass avocados produced in Mexico (Herrera, 2017: 38).

Tancítaro is a municipality located on the western edge of Sierra P'urhépecha. The meandering and narrow road connecting it with Uruapan was only paved in 1997 to enable transportation of the first shipment of Hass avocados to the Mexico–US border. This same road exits Tancítaro in the north and meanders down the Sierra to reach Los Reyes, which was a sugar-cane production center in the 19th century, linked by rail to the main cities in the region. The municipality of Tancítaro is made up of a main town, four villages and more than eighty hamlets spread over the Sierra highlands and the lowlands, in the valley of Apatzingán. The main town is also called Tancítaro, which sits on the highlands and is called el pueblo (the town) by its residents, to differentiate it from the villages and hamlets, all of which are known as los ranchos.

Four families residing in el pueblo owned large agricultural estates in the valley and sugar-cane mills in el pueblo. These families amounted to a small, local bourgeoisie who were close to the Catholic Church, and controlled communications and commerce well into the 20th century (Estrada, 1998: 49ff). The sugar-cane mills were active until the 1960s, when the workers were forced to migrate to the valley of Apatzingán to seek work as agricultural laborers. In the 20th century, ten forestry ejidos and one agrarian community were established in the highlands. Like in other regions in Michoacán, ejidatarios in the highlands earned their living by combining seasonal migration to the United States with resin extraction, timber production and subsistence agriculture (see Gledhill, 1993).

In the 1960s, Hass avocados were slowly introduced by the old landed families, using forest plots to take advantage of the altitude, climate, and rainfall. “My grandfather planted the first avocado orchard in Tancítaro with his bare hands. He was a peón [day laborer], working for Don Pepe Arriaga. Have you seen the horse ranch near the sports center? It’s his,” explained Roberto, a man in his late forties and a popular shopkeeper at the local market. ⁶ These families were organized politically by means of the Partido Revolucionario Institucional (PRI), which held power at the national, regional and local levels until 1988. The Partido de Acción Nacional (PAN), a Christian democratic party, established a local committee in Tancítaro in the late forties; however, it was unable to offer serious opposition to PRI candidates until the 21st century.

Cuauhtémoc Cárdenas, a former governor of Michoacán for the PRI and the son of the late General Cárdenas, quit the PRI and ran in the election as leader of Frente Democrático Nacional (FDN), a coalition encompassing numerous left-wing political parties that seriously competed with the PRI for power for the first time in sixty years. Given the political legacy of General Cárdenas, the Cárdenas candidacy gained mass support from people in Michoacán. On the night of the election, the vote-count system failed, resulting in the triumph of the PRI candidate, Carlos Salinas, a technocrat educated at UNAM and Harvard University. In response, neocardenistas occupied town halls across Michoacán (Beltrán del Río, 1993) and, a year later, became founding members of the PRD. In 1989, the new party won local elections in numerous municipalities across the state, emerging as the leading left-wing political party in Mexico.

In Tancítaro, Cárdenas supporters came from los ranchos, like Vicente, a manual laborer and later a PRD member for whom the mobilization was a coming-of-age movement. As he put it: “It had to be done. These people [el pueblo residents] would look down on us, the people from los ranchos. We were not allowed into the town hall; this was our first victory. It changed things.” ⁷ The victory of the PRD in 1989 ushered in a period in which power alternated between the PRD and the PRI. Neocardenismo and the PRD structured a political opposition to the ruling classes in Tancítaro and, for the first time, the local government elections were being contested by the two main political parties. On one side was the PRI, which still represented the first avocado growers in Tancítaro, like Don Pepe Arriaga, and on the other were the PRD’s supporters and representatives, like Vicente, who were mainly from the villages. This was the political context preceding the agrarian counter-reform and the avocado export agreement.

The agrarian counter-reform and its aftermath

In a critique of Wolf (1982), Roseberry (2014 [1989]:174) argued that it is necessary to pay close attention to the short cycles, junctures and trends that are shaped by structural change but also shape it. I argue that the emergence of avocado producers among smallholders exemplifies a change within a short cycle. While Hass avocados were initially cultivated by large estate holders, the neoliberal reforms and gradual opening up of the export market to all US states incentivized the expansion of avocado orchards by smallholders in many municipalities in Michoacán.

Mexican technocrat governments began to implement neoliberal reforms in the early 1980s. ⁸ Reforms to the agrarian sector started in the mid-1980s, like entry into the General Agreement on Tariffs and Trade, or measures to liberalize agricultural trade such as reduction of production subsidies, elimination of guaranteed prices for most crops and reconversion of credits (Assies, 2008:50). These changes preceded the reform of Article 27 of the Mexican Constitution, which was implemented in 1991, ending the agrarian policy of land redistribution in place since 1917. This was a counter-reform, as it prevented the state from intervening to expropriate large estates and redistribute the land in the form of ejidos and agrarian communities. In a broader and more tragic sense, the reform destroyed the peasantry (Gledhill, 1998: 286). With some ejidatarios selling their land rights and others acquiring them, the reform resulted in a process of fragmentation that enabled a range of agricultural producers of variable landholdings and resources to emerge, while forcing many to pursue their livelihoods in insecure conditions, many times without a wage (Bernstein, 2006: 455).

The PROCEDE (Programa de Certificación de Derechos Ejidales y Titulación de Solares [Ejido Rights Certification and Communal Lands Titling Program]) and PRODECOM (Programa de Certificación de Derechos Comunales [Communal Rights Certification Program]) programs were designed to implement the agrarian reform. For collective land rights to be converted into private property, 50% plus 1% of votes were required in an ejido or community assembly. When sufficient votes were gained, the land could be sold. Ejidatarios in Tancítaro began to vote in support of PROCEDE from 1995 onward, after which the land in their ejidos became available for purchase. Buyers came from el pueblo but also from los ranchos, like the Vélez family, who acquired land in their ejido and others to use for avocado production. Land in the forestry ejidos was sold at low prices because forestry activities were not profitable.

Julio, a man in his fifties, who served as representative on the Local Plant Health Board (see below) between 2017 and 2022 and was the son of one of the “pioneer avocado producers”, bought 20 hectares of land in a highland ejido after working in the US for eight years. While this land was cheap, it had to be cleared before an orchard could be planted. Julio denied that avocado producers caused deforestation: “no, there was no deforestation, the crops expanded, that’s different. [Once the market was open,] people preferred avocado production over wheat or maize.” ⁹ Nevertheless, in Tancítaro, deforestation associated with the expansion of avocado orchards is documented from the late 1990s onward (Hernández-Fernández, 2023: 156). Tancítaro residents remember the sound of chainsaws running night and day and, according to Roberto, whose family went from working as agricultural laborers to renting avocado orchards in the 2000s, nothing could be done as the mayor himself was involved in the avocado production business. ¹⁰

Following the reform of Article 27 and liberalization of the agrarian sector, the national institutions and associations that brought together peasants were substituted with independent associations of agricultural producers competing against each other to insert themselves into the free-market logic. Macip Ríos (2005) examined coffee production in Veracruz, Mexico, during the transition from protected production to the neoliberal model. His work showed that the new associations of agricultural producers were prone to clientelism and factional politics as they lacked the state guarantees of controlled prices and subsidies, while in textile cooperatives these associations were “ephemeral” as they varied in membership, structure and temporality (see Vargas-Cetina, 2005).

In Tancítaro, while NAFTA negotiations and the agrarian reform were under way, avocado producers founded the Local Plant Health Board (LPHB) in 1991. The Board, as it is known locally, is a private organization that assists the Secretary of Agriculture in completing health certification and plague control for national and international goods. In Tancítaro this means compliance with the export regulations imposed by importing countries, mainly the US. The Board brings together all avocado producers, regardless of the size of their landholdings, which range from 1 hectare to over 50. In addition, they elect one member to represent them in the Association of Avocado Producers and Exporters of Mexico (APEAM), a private association founded after the first avocado export agreement was established with the United States Department of Agriculture in 1997. Unlike ejidos, where landholdings tended to be more equal, there are significant differences in the landholdings, labor organization and resources of the avocado producers in the Board. While the household division of labor may be more significant for smallholders, large estate holders hire foremen and temporary laborers to perform all agricultural tasks, from pruning to cutting.

In 1997, three years into NAFTA, the first avocado export agreement was signed. Four municipalities, including Tancítaro, then sent their produce on the first shipment from Michoacán. The mayor organized a huge event to celebrate and he was joined by the then state governor. Sixteen producers sent avocados from 364 hectares (Thiébaut, 2013: 163). In the early 2000s, Tancítaro was changing, with the first two avocado export companies, a paved road connecting it with Uruapan, and a thriving economy that was reshaping politics and social life.

A violent turning point in 2006

On December 1, 2006, Felipe Calderón, born in Michoacán’s state capital, took office as president of Mexico for the PAN party, after a contested election marked by fraud claims and political mobilization. A week later, joined by the Secretary of Defense, President Calderón launched Joint Operation Michoacán, the first special operation designed to target drug-trafficking and eradicate drug production in that state. In this first operation, Calderón deployed over 7,000 members of the military (Maldonado, 2013: 60) and it was later replicated in other Mexican states.

While drug production and trafficking in Michoacán can be traced back to the 19th century (Maldonado, 2013: 48), violence associated with clashes between drug traffickers emerged in the early 21st century. Broader changes, such as the neoliberal reforms in Mexico, the dismantling of the Caribbean drug-trafficking route by the US, anti-drug policies in Colombia, and the reduction of drug production in Peru and Bolivia all worked to strengthen Mexico’s position in drug-trafficking and production (Maldonado, 2013: 52). By the 1990s, the Hermanos Valencia (Valencia Brothers) cartel, also known as the Cártel del Milenio (Millennium Cartel), had gained control of marihuana and poppy production and distribution in Michoacán. Raised within a family of grocery shop owners, they established alliances with Colombian and Mexican cartels, as well as politicians for protection (Maldonado, 2013: 56–7). They became strategic allies with the northern Sinaloa Cartel in its dispute with the Gulf Cartel over other northern territories. In retaliation, the Gulf Cartel sent the Zetas (the Zs) – a mercenary group trained by the military – to Michoacán, while the Valencia Brothers brought in former Guatemalan Special Forces soldiers, giving these clashes a paramilitary nature (Gledhill, 2015: 167). In 2006, La Familia Michoacana (the Michoacán Family), whose leaders were trained by the Zetas, announced that they had formedin order to expel the Zetas and protect the michoacanos. They were the first to impose reestrictions on social life, such as curfews announced via SMS. La Familia Michoacana controlled pharmacies, restaurants, auto repair shops and convenience stores.

There were differences in how these criminal organizations operated (Mendoza, 2022). In Tancítaro, the Zetas and La Familia operated in a similar fashion: by extorting or kidnapping the children of wealthy, well-established avocado producers and relatively new producers. The Tancítaro PRD municipal government was also targeted: the mayor was shot, as he refused to collaborate with the cartel; two substitutes were appointed but, as the violence did not stop, all the civil servants resigned jointly in an unprecedented event. The state government responded by appointing a citizen council to conclude the term. Disputes within La Familia Michoacana ended in a schism in 2011, when Los Caballeros Templarios appeared. They had control over several municipalities in the Apatzingán valley and the highlands, particularly Uruapan and Tancítaro. In addition to extorting avocado producers and the local government, the new cartel imposed restrictions on everyday life. Social activities like meeting family or friends in the main square became nearly impossible, as people started receiving threats.

In 2011, Tancítaro residents elected a PAN candidate as mayor for the first time: a well-known avocado producer and one of the founding members of the Board, who amassed his fortune after the agrarian counter-reform. For a two-year period, Tancítaro became extremely violent. Unlike the previous cartels, Los Caballeros Templarios set up a racketeering structure anchored in the municipality’s territorial organization. All residents of Tancítaro, whether they lived in los ranchos or el pueblo, were forced to pay a fee (pagar piso). The administrative representative in each village or hamlet, who had no choice than to collaborate, would collect the fee each month, then deliver it to the jefe de plaza. Avocado producers paid per hectare owned, while export companies were also required to pay to keep operations going.

On January 24, 2013, groups of male residents in the valley of Apatzingán took up arms to expel Los Caballeros Templarios from their municipalities. They called themselves autodefensas and were mostly lime growers or ranchers. In response to the autodefensas, Los Caballeros Templarios used various strategies, like forcing people to show public support and leaking videos showing civil servants chatting with drug traffickers (Gledhill, 2015: 189). In Tancítaro, they demanded that export companies instruct their staff to hold public rallies showing support for the cartel; two of the four local export companies did so, while the other two refuse dto collaborate with the cartel and were burnt down in retaliation.

In October 2013, a girl was kidnapped by Los Caballeros Templarios, who asked for over USD 600,000 to return her alive. Her father, Don Jesús, an avocado producer, offered to pay half the sum in cash and the rest in avocado orchards. Despite the money being handed over to Los Caballeros Templarios, the girl was brutally murdered. According to Curry (2019: 61), several groups had been secretly planning retaliation before the girl was murdered, while Moncada contends that the Board provided a structure for organizing people and resources into a movement (2021: 138). The narrative that I encountered in the field pinpointed this murder as a turning point. Don Jesús had served in the army, so he knew about military tactics, techniques, and how to handle weapons. But what was most important was his proximity to the Church. “Had this atrocity happened to any other family, there would not have been an uprising,” ¹¹ stated Mauricio, a man in his 40s who joined the Tancítaro autodefensas and remained a member for nearly four years. But the support from the Board and other avocado producers was due to kinship relations. As Julio put it:

Don Jesús and the Vélez son parientes [are related via kinship]; [after his daughter was murdered] he contacted them [the Tierra Caliente autodefensas] but Vélez, the packing-house owner, collaborated with the Templars. He hid them inside his packing house [when the autodefensas entered Tancítaro] but supported his cousin. ¹²

Nearly ten months after the original assault, on November 16, 2013, the autodefensas entered Tancítaro via the village of Pareo, which neighbors the Tierra Caliente valley. People were scared, as rumor had it that the autodefensas might be another cartel. Mireles, a military serviceman trained as a surgeon, leader and autodefensa spokesman used to approach the local priest after entering a new town. Little by little people started to join the armed men from Tierra Caliente. The Chief, an avocado producer in his mid-forties, fled Tancítaro when he heard the rumors about a new cartel entering, but came back after a few weeks.

Some weeks later, Mireles spoke at a town-square meeting about the need to organize local autodefensas into a sort of council. This council would need to be independent from the mayor, would govern and conduct operations, and would be led by Don Jesús. According to Vicente, “Nobody voted for them, they just appointed themselves as council.” ¹³ It should be noted that the Oversight Council (consejo de vigilancia) was not actually a council, but a set of leadership positions held by Don Jesús and his men and respected by everyone. Today, its members are still recognized as a council and as leaders. Some of their nicknames, like The Chief and The Prosecutor, evoke a type of authority, while others, like the Tasmanian Devil, allude to a strong personality. Don Jesús and the Oversight Council were avocado producers or suppliers for the avocado industry, giving them the resources to acquire high-caliber arms. Others, like Vicente, who were unskilled laborers of various kinds, had to raise funds to buy their equipment and arms. An organic class distinction was consolidated among the autodefensas, which can be interpreted politically as the formation of a faction; this was the same distinction that Victor explained to me outside the barricade when he said “We were divided from the start.” The Oversight Council, which was made up of avocado producers, conducted operations, while other autodefensas, including Victor, a political leader member of the PRD, and unskilled laborers like Vicente, followed the orders of the council.

The autodefensas were organized on a village-by-village basis. Residents built checkpoints at the entrance of every village, which had different structures – some resembling bunkers, others watch towers – and received different names, from “barricade” to “caseta de vigilancia.” Their purpose was to secure the municipality by controlling the circulation of people. Each village would have its own system of organization but, essentially, men would rotate and take shifts at their local checkpoints, making it a collective practice. In the event of a threat, they communicated via walkie talkies. This system facilitated organization but also enabled other practices, like money collection, and a sort of division of labor in security provision (Curry, 2019: 64). Avocado producers often paid manual or agricultural laborers to complete a shift at their local checkpoints, creating a stronger distinction in security between those able to command and those mobilized as privates in an irregular army.

The Chief explained that “It took us the rest of the year to convince people to join in and to start cleansing.” ¹⁴ I often heard the expression “to cleanse” in the valley of Apatzingán and Sierra P’urhépecha when people were talking about security and control. Curry notes that the term belongs to the military jargon introduced into the region as part of special operations in the 20th century (2019: 53), while Gledhill notes its resemblance to usage in paramilitary contexts like Colombia (2015: 172; see also Taussig, 2003). Other expressions like peinar (to comb) are part of the local parlance.

According to Mauricio, in the case of Tancítaro, when the Tierra Caliente autodefensas entered the town, Los Caballeros Templarios had already fled, but the cleansing was conducted regardless. When a suspect was captured, he could either be expelled, pardoned, or killed. According to some accounts, the cleansing was conducted by the Oversight Council and decisions were made arbitrarily. When the Oversight Council members were acquainted with the suspect he could be pardoned, although only after switching sides. However, if the suspect had no connection to the Oversight Council or no way of defending himself, he could be killed and buried in the mountains, as was the fate of many. Continuous, low-profile cleansing was conducted by the autodefensas throughout the period. At the time of my fieldwork, people rarely talked about this, but it came out during moments of religious ritual with informants with whom I had a relationship of trust. One night, after having joined a large group of people to say the rosary for the disappeared, Miguel, the manual laborer who had invited me, said: “Well, you have seen it. This is what moves us.”

President Enrique Peña Nieto of the PRI succeeded President Calderón in 2012. In 2014, he appointed a Special Commissioner for Security and Integral Development in Michoacán to conduct negotiations with the autodefensas’ spokesmen and institutionalize these groups into the new Fuerza Rural (Rural Force) police corporation, which autodefensa leaders like Papa Smurf joined, while others like Mireles did not (Maldonado, 2018). Another outcome was the creation of a unified chain of command for the new Michoacán State Police that would centralize all the municipal police bodies into a single police force to avoid corruption. Other areas of intervention included establishing a new Public Security Law (Maldonado, 2018: 193), which supported the creation of local security councils. The Tancítaro autodefensas refused to sign the agreement, arguing that the unified command elements could be selected from all over the state, jeopardizing security. They argued that local people would be more loyal and committed to defending their own territory and people. Ratifying the unified chain of command was not mandatory and Tancítaro went through its own process of training and appointing a new police force.

The Board, then controlled by the Vélez family, was particularly interested in safeguarding the transportation of avocados. Together with the local government, it provided funding of nearly USD 1 million for the new police force. The federal police and the military trained the recruits for a three-month period, focusing on prevention of organized crime. The new police force was called the Cuerpo de Seguridad Pública de Tancítaro or CUSEPT (Public Security Body of Tancítaro), and its 36 officers were all Tancítaro locals. The Board’s contribution to CUSEPT’s funding caused controversy among residents. While civil servants argued that CUSEPT’s commander-in-chief responded not to the Board but to the mayor, it was public knowledge that the CUSEPT collaborated with the avocado producers in the Oversight Council.

The upcoming 2015 local election offered an opportunity to secure further changes. Through a series of closed meetings at the public high school, political party and autodefensa leaders agreed to run together in a joint candidacy for mayor to shield the local government from criminal organizations. Don Jesús was the obvious choice for candidate, but he rejected any role related to electoral or institutional politics. Instead, he proposed a schoolteacher and former PRI mayor, who was in good standing among residents, was close to the Church, and had been approved by most autodefensa leaders. The candidate went on to win the June 2015 local election. A couple of months later, Don Jesús was killed in an ambush. Media reports linked his murder to allegations he had made about the Templars continuing to extort avocado producers. Without Don Jesús, unity among the autodefensas was weakened, reinforcing a division into two factions along class lines. While some autodefensa leaders tied to the PRD, like Victor, took senior positions within the local government, the Oversight Council retained its leadership position.

The unity government formed on January 1, 2016 tried to advance further social and political changes. To satisfy the people’s call for social justice, the new government hired Jesuits for Peace, a non-governmental orgnization (NGO) founded by the Society of Jesus to run the “Rebuilding the Social Fabric Project.” The project was designed to target the deeper causes of criminal violence and its consequences among the people in Tancítaro, including, environmental issues, drug-trafficking and drug consumption among young people. The project leaders organized a range of activities, including religious retreats, rituals and masses where the victims of criminal violence were remembered. ¹⁵

The unity government also founded the Municipal Security Council, supported by the State of Michoacán Public Security Law of 2014. Autodefensa leaders from both factions, Board representatives, and priests were invited to join. Vicente and Miguel, who supported and collaborated with the Jesuit project initiatives, were active security council members, the Board representatives and the priests rarely attended these meetings, while some Oversight Council members like the Chief did.

Tensions between the autodefensa factions inherent to their class relations emerged at the security council meetings. When discussing threats to security, the Oversight Council were keen to reclaim the leadership for themselves and accuse the security council of political quarrels. As most Oversight Council members were avocado producers, other issues, such as the expansion of new avocado orchards and other environmental matters, were contested. When discussing deforestation, they acted defensively and sabotaged any decision-making regarding local regulations on new orchards or sanctions. But, most importantly, they disapproved of outsiders being so familiar with local political affairs and insisted that they shouldn’t be trusted, as they could be spying on the local government. While autodefensas like Vicente were committed to the Jesuit project and supported its initiatives, they did not dare confront the Oversight Council and remained silent, almost conceding to their critiques. More than once, I was asked to leave the meeting when patrolling strategies were discussed.

The 2018 local election challenged the continuity of the political and social projects. While some senior staff members called for the renewal of the unity government, there was no agreement and the three main parties held primaries. The PRI selected a member who was then serving in office, while the PRD split, as its main leaders (including Victor) revealed that they were already registered with MORENA, the political party founded by Andrés Manuel López Obrador, which won the presidential elections in 2018 and 2024 and has become the main political party in Mexico, including Michoacán. The PAN selected a female candidate, a woman with a relatively unremarkable political career but with close ties to the party’s leadership, who then won the election.

The first female mayor in Tancítaro cut funding for the Jesuit project, restructured the security council and appointed new members, none of whom had ties to the Jesuit project. In response, Vicente and other autodefensa leaders not related to the Oversight Council withdrew from institutional politics. The Oversight Council retreated to the shadows. However, they are still acknowledged as leaders and maintain close ties to the PAN leadership, the local government and CUSEPT, whose commander-in-chief works in collaboration with the Oversight Council. Tancítaro has been governed by the PAN since 2018; men belonging to the local autodefensas are armed, continue to patrol their villages using the same barricade system, and still respond to the Oversight Council’s command, but encountering armed civilians in daylight is now less common.

Conclusion

In this article I have offered a class-based analysis of the emergence of the autodefensas of Tancítaro, where a heterogenous group of avocado producers and non-producers joined forces to combat criminal rule but their differences and inherent class tensions contributed to their eventual fragmentation. Unlike other scholars, who analyzed the phenomenon by focusing on the collective efforts and goals of the autodefensa groups, I have argued that paying attention to class relations can help elucidate different political agendas and explain their fragmentary nature.

I have presented my ethnographic argument as an extended case study, placing it within a wider historical and political context. The case was fundamentally shaped by three interlocking processes: first came the introduction of Hass avocados in the 1960s in a historically marginal region. The second process was the dismantling of agrarian structures and the peasantry through neoliberal reforms, which resulted in a process of class fragmentation. Some ejidatarios began avocado production on their small plots of land, some acquired land before turning to avocado production, while others sold their land rights and became dispossessed of their means of production. I have conceptualized avocado producers as anthropological subjects who emerged as a result of the privatization of collective property and the opening up of the avocado export market by NAFTA. Thus, avocado producers must be understood not as peasants but as independent agricultural producers who benefited from a process of dispossession, but whose landholdings, labor organization and resources vary. This variation led to even greater differentiation between avocado producers, in contrast to ejidatarios, whose landholdings tended to be more equal.

The third interlocking process was the atomization of drug cartels and the militarization of public security in Mexico that started with the violent turning point of 2006 but continued through the three subsequent terms of federal government, and three successive changes in the governing party. At the time of writing, President Claudia Sheinbaum of MORENA has officially launched a new special operation in Michoacán to tackle the root causes of violence in that state, after the tragic murder of the Uruapan mayor on November 1, 2025 by members of the Cartel Jalisco Nueva Generación (Jalisco New Generation Cartel). On February 22, 2026, the Mexican military and air forces, with the help of the US intelligence services, conducted a special operation in which they located and killed the leader of the Jalisco New Generation Cartel, provoking a violent response by the cartel across the country. The autodefensas contended that they formed to overcome a similar situation, expel criminal organizations, and take security into their own hands, claiming that they had been abandoned by the state.

Analyzing these three interlocking processes in conjunction has the potential to shed light on the emergence and fragmentary nature of the autodefensas and, in the case of Tancítaro, may help explain why security became the primary societal concern. Furthermore, it illustrates what class fragmentation looks like politically after decades of neoliberal policies in a violent context where various armed actors have been pursuing distinct political agendas. Class relations between the avocado producers who controlled the security strategy and non-avocado producers who followed their commands illustrate the tensions inherent to such relations. My ethnographic research shows that security prevailed over other calls for social justice for the murdered, the disappeared and the displaced: in the words of Miguel, “what moves us”. But this occurred not because the Oversight Council forced their views on the other autodefensas and the population, but because there was consensus on the importance of security, even among critics of the Oversight Council. As residents of all social classes had experienced everyday life marked by murders, kidnappings, prohibitions on accessing public space and extortion, they valued the extra-legal security provided by the autodefensas. Even if these groups were commanded by a “council” that could not be held accountable, this was a type of authority that most residents respected. This translated to a prioritizing of the protection of avocado production and the export market, a private industry that continues to destroy the environment and to transform class relations, all to meet the international demand for avocados.