Drug trafficking, the informal order, and caciques. Reflections on the crime-governance nexus in Mexico

ABSTRACT While Mexico is widely considered as an example of consolidated statehood, the deepening of drug-related violence and insecurity has corroborated the existence and expansion of ‘dark spaces’ governed by coalitions of state and non-state actors driven by criminal and political interests. In contrast to the prevailing interpretations and public narratives, I will argue that it is historically and conceptually flawed to understand such expressions of limited statehood solely in terms of the proliferation of criminal organisations and the exacerbation of the so-called war on drugs only. Instead, I will examine the historical patterns in Mexican state-making, in which actors and practices of political ordering outside the state properly speaking exercise multiple forms of de facto sovereignty and governance. These arrangements, including caciquismo, accommodate distinct crime-governance manifestations. The article substantiates its claims by looking at the examples from different periods and regions such as Sinaloa, Sonora and Michoacán.


Introduction
In May 1984, shortly before his assassination, Mexican journalist Manuel Buendía wrote two columns in which he examined and commented on a pastoral letter written by nine bishops from the southern states of Oaxaca and Chiapas, issued in March that year, about the threats of drug production and trafficking in the region. The bishops expressed a sense of urgency and alarm about the development of the drugs business, and the consequences for local communities and society at large. Four things stood out: poverty was increasingly luring (indigenous) peasants into the production of marijuana and poppies, after which they became trapped in a production and trafficking system run by domestic and international mafias; money and violence have enabled these mafias to become powerful actors and take 'practical control of people's lives'; the capacity to exercise social, economic and coercive power with impunity 'is itself inexplicable unless one assumes the direct or indirect complicity of senior public officials at both the state and federal levels'; finally, the bishops were deeply concerned about clandestine trafficking networks exerting more and more political influence. 1 Buendía quoted the pastoral letter at length since for a while he had himself been investigating CONTACT Wil G. Pansters W.G.Pansters@uu.nl the ramifications of drug trafficking in the country´s security and political system, and was convinced that the latter had increased since the early 1980s. 2 Around the same time, Samuel del Villar, a former special advisor in matters of corruption to then president De la Madrid, found not only that Mexico´s drug industry had spectacularly grown but also that the relationship between the traffickers and the state security apparatus had been elevated´to a new level´. 3 Reports from rural Oaxaca and Chiapas and from the heart of the federal political and security system thus indicated that something was occurring that hinted at the qualitative (and quantitative) transformations of Mexico´s drug industry and its relations with the state actors and state governance. But what was happening? What changed? How were relations between crime and state actors before, and how did they develop thereafter? These substantive questions are explored in the remaining part of this article. This article looks back to the 1940s-1960s, examines important shifts during the 1970s and 1980s, and includes reflections on more recent developments. Although I follow this broad periodization, the changes and transformations took place gradually and in waves. They were not uniform or unidirectional when examined from particular regions.
A few thematic considerations about the connections among crime, violence and governance contextualise my historical approach. The first concerns the puzzling simultaneity of consolidated statehood and shocking violence and´unrule´of law. 4 Mexico is widely considered as an example of consolidated statehood: its economy is part of one the world's most prominent free trade agreements, it has a flourishing tourist industry (more than 30 million international visitors in 2015), a reputed foreign service, and an internationally respected, though hugely expensive, electoral system, to mention a few indicators. 5 At the same time, over the last fifteen years drug-related violence and insecurity have corroborated the existence of 'dark spaces' governed by criminal actors, often in coalition with state agents. Due to the proliferation of criminal organisations, violent competition among them, and the militarization of the war on drugs, such dark spaces (and territories) have been understood as expressions of limited statehood: i.e. when (central) authorities lack the control over the means of violence and the capacity to rule authoritatively. 6 Homicides rates have risen sharply: from 8.1 in 2007 to 24 in 2012, dropping in 2014 to 17.1, but climbing again to 21.3 in 2016. In the 2007-2016 period, approximately 211 thousand people were murdered in Mexico. 7 Certain professions have been particularly confronted with (lethal) violence, most notably journalists and municipal presidents. 8 Numerous governors have recently been arrested on suspicion of close ties to organised crime. 9 Despite indications of consolidated statehood, there is also much truth to the idea of dark spaces and limited statehood. A recent critical report about criminal rule, violence and impunity in the states of Tamauliupas and Coahuila was entitled´en el desamparo´(best translated as´unprotected´), a clear indication of limited statehood in security and rule of law. 10 In a controversial interview, former president Felipe Calderón argued that his head-on attack on organised crime was an answer to losing´territories and states where authority has been harmed´. 11 The state has thus limited control over particular parts of Mexico, especially when it comes to security, rule of law, but also the political system (e.g. party financing) and subnational and local economies.
Second, scholarship often sees violence and the influence of criminal organisations in society as incompatible with (democratic) state governance. 12 In fact, this assessment often informs war-on-drugs discourses and mano dura policies of governments and international organisations. The narratives and epistemic perspectives underpinning these policies feature notions of opposition, dichotomy or antagonism: government rhetoric and policies portray drug traffickers, gangs and other criminals as bodies external to social, political and legal orders. They identify threats, and see states and state elites obliged to fight these destabilising forces. Although such policies have been politically expedient, they have not remained uncontested. 13 A belligerent president Calderón provoked a sharp condemnation from a mother whose son had been arbitrarily detained:´The President still thinks that he is the only one who has confronted drug trafficking without fear. . . while in the streets a lost war takes place with police officers and soldiers bought off by the traffickers and where thousands of disposable youngsters are cannon fodder that end up in prisons and graveyards´. 14 Ordinary people have long distrusted government bravery, known about political protection, and experienced the mediated application of the law.
Third, as Müller suggests in the introduction, normative assumptions and discursive constructions find their equivalent in analytical portrayals of crime and violence as antithetical to political, economic and social governance. This article is critical of the interpretive logic of opposition, dichotomy and antagonism, and, instead, uses a conceptual approach that concentrates on connections, entanglements and the nexus between crime, violence, governance and state-making.

Manifestations of the crime, violence and governance nexus
Weighing in on these considerations, I propose to examine the main features of the crime, violence and governance nexus in Mexico by looking at the three basic forms or manifestations. First, violence and criminal organisations can play productive and ordering roles and hence constitute a mode of governance. In the last part of this article I use the concept of criminal sovereignty and an example to further elaborate on this point. Second, illicit and criminal activities (drug cultivation and trafficking) appear deeply embedded in particular local and regional social relations and economies, through which they play a functional role in governance systems. Third, and more generally, the crime-governance/state nexus appears inherent to what I call the informal order. It is my contention that the logic of the informal order enables the accommodation of criminal activities within wider hierarchies and power relations. Crime and violence have been part of modes and practices of governance through personalistic loyalties, networks of corruption and impunity. In this context, I will also examine the phenomenon of caciquismo (bossism). This approach to the crime, violence and governance nexus requires drawing upon bodies of anthropological and historical literatures concerning political culture, power, caciquismo and state-making. I hope to show that engaging this scholarly work broadens our conceptual and empirical understanding of the crime-governance nexus.
I will deal with these three manifestations in reverse order, first because the last argument is of a more systemic and generic nature, but, more importantly, because I hypothesise that in recent decades drug-related criminal organisations, that were long nested within informal and legitimate social and political networks and loyalties, and hence constituted a particular variant of the crime, violence, governance nexus, have experienced profound changes. This has significantly affected the nature of that nexus. I suggest that the first signs of that transformation became visible in the 1980s. This article argues that during much of Mexico´s 20th century the drug-related crime, violence, and governance nexus was part of and regulated by informal and/or cacique orders, and embedded in the socio-economic arrangements and governance systems of local communities (manifestations 3 and 2). However, the accumulated effects of significant transformations in the constituent parts of the nexus disrupted socially rooted and politically tolerated networks of loyalty, protection and impunity. In some cases these embedded arrangements morphed into full-blown criminal sovereignties (corresponding to manifestation 1).
This article aims to understand the transformation of the crime, violence and governance nexus, and obtain a first grip of its distinctive modalities through time and space. I substantiate my argument with examples from different periods and regions. The next section starts with a brief conceptual exploration of the notion of the informal order, and examines how illegality and crime were absorbed in broader social and political networks of loyalty and complicity. I then look at the (drug) crimegovernance nexus as an integral part of the social and moral arrangements and institutions of the informal order, and, in particular, how drug production and trafficking were woven into the fabric local communities. I, then, briefly discuss the main forces and factors responsible for changes in the crime, violence and governance nexus. In the final sections, I study the ways the latter played out in particular communities and regions. I look at developments in a small community in northern Sonora close to the US border, and at the emergence of criminal sovereigns in central-western Michoacán, that subsequently provoked an armed popular response. For the latter, I build off recently published cutting edge research.

The informal order: a brief exploration
Scholarly interpretations of the history of Latin America have long grappled with the persistent influence of informal institutions and practices in increasingly complex and modern societies and polities. In recent years, there has been a wave of studies about informal institutions in post-Cold War political contexts. 15 O'Donnell, for example, observed that the´excess´of informal institutions such as clientelism in Latin America seemed to contradict formal liberal-democratic rules and organizations. 16 Examples of informal organisations are camarillas, cacicazgos, clans and cliques, but also mafia´s and drug cartels. I put forward the concept of the informal order to speak in a comprehensive manner about these realities, practices and phenomena in politics, society and the economy. 17 The concept is relevant for understanding the crime-governance nexus that concerns us here. Several specifications are in order.
First, my approach is critical of equating the´informal´with particular socio-economic groups and classes (common in much of the informality scholarship), and instead focuses primarily on how and why all social and political actors strategically navigate between formal and informal institutions and orders. 18 In analytical terms, the focus is thus on agency and practices. Latin Americans permanently negotiate 'at the juncture of a system composed of ideal norms and a legitimizing system of social practices based on personal relations'. 19 Second, the informal order refers to the universe of informal institutions, organisations, and practices, as well as the corresponding cultural repertoires that underpin them. The notion of the informal order connects well to clientelism, personalism, and, in Mexico, caciquismo. A key feature of the political and social functions of caciquismo is its complex and ambivalent relationship to formal orders of power, government, authority and law. Quintessential caciques may operate behind the scenes of formal institutions, but just as camarilla networks they also establish informal connections between public administration, the ruling party, elected office, business interests, the legal system and coercive state institutions. As paradigmatic actors in the informal order, which over time shaped and was shaped by the longevity of the one-party regime, they were instrumental in connecting political clans and their social bases to modern governance institutions. 20 However, since it thrives on personalistic mediation, caciquismo also tends to undermine state-making and the rule of law. Corruption is a key ingredient of this way of organising power and order. Shielded by political protection and impunity, it easily blends into gray areas of illegal transactions. Cacicazgos generally develop out of a capacity to extend a position of leadership or power in a particular domain (administration, army, business, union) towards others. This can generate a´drive-wheelé ffect, which articulates different interests, relations and networks, all centered on the cacique. 21 An effective articulation of different societal domains constitutes the core of cacicazgo-building, because it enhances the capacity of intermediation, constructs protective networks and accesses means of violence. 22 Third, the core of the informal order is the exchange of myriad goods and services (favours) between members of networks of friendship and/or (extended) kinship, that are not or only difficult to attain through formal systems, such as loans, water rights, contracts, preferential administrative treatment (licit) and drugs (illicit). Personal relationships called upon in informal exchanges persist if there is a reasonable chance favours will be eventually 'returned'. Reciprocity is a key element of the informal order. The willingness to engage in the obligations of informal exchange mechanisms is ultimately based on´focalized´trust (confianza). 23 This form of trust is articulated in moral terms (loyalty, solidarity), and expressed in rituals, such as drinking or establishing fictive kinships (compadrazgos). 24 When personalised relations of trust and exchange colonise formal organisations as state bureaucracies, the latter are pervaded by ambiguity.
Fourth, the cultivation of friendship and kinship ties provides opportunities to achieve power, status and authority. Composed of groups (and identities) such as clans, camarillas and clientelist networks, the informal order constitutes the key domain upon which oppositions are constructed: political conflicts are often driven by factional and personalistic rivalries, rather than by formal partisan or ideological disputes. Widely used terms like bandas, grupos, facciones, clientelas, equipos, corrientes and tribus indicate that the Mexican political vernacular is a fascinating meaningful repository of the informal order. Nourishing friendship and kinship brings affection and allegiance, passion and trust, emotion and solidarity and drinking and partying to the core of informal institutions and organisations. Informal and personalistic power is inherently prone to the use of violence. It is interesting to note the parallels between the social and cultural rules and expressions of informal political arrangements and those of the drug world: the importance of blood ties, personalised trust and violence, and of rituals such as partying and drinking. Studies have shown how contemporary (urban) caciques, just as modern narcos, cherish and cultivate symbols of´traditional´ranchero identity (horses, gunplay, clothing styles, alcohol, etc.). 25 Fifth, in normative terms, the informal order should be understood as the antithesis of the formal rule of law, because its universalist basis contradicts the particularistic basis of the informal order. The informal order is in fact about 'interrupting' the universal application of the law. Writing about East Asia, Fukui once stated that against the rule of law, the rule of man is marshalled. 26 Informal or personalistic mediation of the law is behind one of Mexico's (and Latin America's) most insidious problems: impunity. Despite normative incommensurability, entanglements between informal and formal (state) institutions occur all the time. This is particularly but not exclusively true at the fringes of the informal order, where it transmutes into the illegal order. The latter can thus be conceived as a particular subdivision of the broader informal order. 27 In fact, criminal organisations as well as relationships between clandestine actors, criminal networks and state institutions often find their origins in the belly of the state. 28 This understanding of the state of the rule of law brings together insights from old-school cacique studies with new legal histories of Latin America. 29

Embedded drug trafficking
During the decades after the Second World War, drug cultivation and trafficking and their connections to governance were deeply embedded in and protected by informal and cacique orders. In the summer of 1968, two Americans asked a friend from Guadalajara for contacts in a marihuana producing area so they could go there and write an article about it for Life magazine. The magazine had advanced five thousand dollars for expenses. Since the Americans had themselves been involved in marihuana smuggling for years, they knew their way around. Their friend from Guadalajara was a middle-man, who subsidised around fifty marihuana-growing peasants with seeds, fertilizers and other materials, bought up their harvest to subsequently sell ton loads of it to an organisation based in Ciudad Juárez on the Texas border. The Americans took off in a Cessna from an airfield close to the capital of Guerrero and flew to the coastal mountains. Accompanied by local peasants, who had come to pick up the plane´s load, they moved deeper into the mountains on mule-back for two days. Although they gained the trust of the campesinos, they were not allowed to take their pictures while working in their marihuana fields. Instead, they proposed the Americans to purchase a field, where they would be able to do as they pleased. With half of Life magazine´s advance payment, they bought approximately four thousand square meters with five thousand plants. For two months, they worked the plants with local peasants, chronicled the process, shot pictures and ended up with prime quality Acapulco gold.
The American´Beat smugglers´entered a world in which local indigenous peasant communities in the Guerrero mountains depended heavily on the cash income from marihuana cultivation (around three dollars per kilo at the time). 30 Distrustful of the authorities, the men who accompanied the visitors into the mountains were all armed. While urban-based traffickers financed investments, and moved the merchandise towards the border, the peasants were clearly in charge of the region and the cultivation process: they decided to take the visitors along, accommodated them in their communities, prevented them from taking pictures, sold a marihuana field and worked with them. Traffickers and producers maintained an interdependent relationship, in which each controlled their part of the chain. The Americans also learned that local marijuanos did not fink on one another, and were no users themselves. If someone was arrested, friends and family would provide help. Local elites were involved in the illegal economy, and provided protection through their family connections in local security agencies. In a fascinating account, one of the American smugglers concluded that the marihuana business in Mexico was entirely´wrapped up with the general economic situation´. They found themselves inside a´whole substructure of illegal money-making activities that so many Mexicans deal in. . .´, a local society in which illegal activities were part of daily life, controlled and shielded by local social networks and governance structures. 31 It is difficult to imagine that their experience could be repeated in current Mexico with the same outcome. 32 Such locally regulated illegal economies and their links with formal and informal governance systems were not the exception during the 1940s through 1960s. The highlands of Sinaloa were already a key region for the cultivation of marihuana and poppies (for opium and heroin). Local peasants combined it with subsistence crops and started to cooperate in family and place-based networks and clans. Juan Antonio Fernández Velázquez has gathered unique oral histories from the Sinaloan highlands that provide insights into these processes. One man from Badiraguato told the author how the workings of the trade could become an incentive for cooperation:´Those of us involved in growing [poppies, WP] knew people who would buy from us. Sometimes harvests were already agreed upon in advance with an intermediary or a buyer. . . on other occasions somebody sent word from Culiacán that so and so wanted so many kilos for a particular date. If you had that quantity you would sell it, and if not we would put it together among several producers, we cooperated'. 33 Peasant producers also exchanged technical information about how to make heroin and improve its quality. Fernández shows that these cooperative efforts were deeply rooted in family ties, friendships and place-based relations, which developed into the clans that ran drugs cultivation in the Sinaloan highlands and trafficking routes in the lowlands and beyond. These were tightly knit communities in which everybody knew each other. Using the language of the informal order, Fernández argues that at the´core of the illegal business. . .family ties and friendship acquire a vital importance for the consolidation of clandestine business transactions´. After all, members of intimate local networks can be trusted. The business of drug cultivation and trafficking then involves´affective ties. . . that increase personal and commercial security and guarantee impunity´. 34 Informal networks were crucial for protection. In the early years of the drug trade, two key state institutions in particular provided cover for well-connected people involved in the drugs business: the (state) judicial police and the army. Melesio Cuén had been municipal president of Badiraguato on several occasions during the 1940s and early 1950s. His political influence extended into the appointment of judges, police officers and civil servants. As a powerful cacique, Cuén also had substantial economic interests in mining, real estate, grocery stores, tortilla factories, a pharmacy and a funeral home. He was also involved in poppy cultivation and trafficking opium gum. He sold protection to local producers, who then acquired the right of´disímulo´(a blind eye) by the antidrug judicial police inspectors. Cuén was a narco-cacique, whose local power was based on a mixture of licit and illicit businesses, formal authority and informal political networks. The crime-state nexus was deeply integrated in governance systems in the Sinaloan highlands.
Such arrangements went beyond the local level. In Sinaloa and in neighbouring Durango, members of the De la Rocha family held influential positions in the judicial police, which was responsible for carrying out the government´s anti-drug campaigns. As head of the state judicial police of Sinaloa, Francisco de la Rocha taxed local producers (to be paid in raw opium) in exchange for protection. He was also believed to be in business with a major opium trafficker in the Culiacán area. 35 Aureliano de la Rocha held in the same position in Durango, and, in 1946, was appointed commander of an eradication campaign that covered several municipalities in the states of Sinaloa, Durango and Chihuahua, an area known as the Golden Triangle. The campaign was not very effective. 36 The involvement of the De la Rocha clan in the regional opium trade is a textbook example of the corruption of law enforcement, the entanglement of state and crime, and the colonisation of state institutions by informal clan interests.
The army was not exempt from these arrangements. Several military strongmen extended their coercive power towards positions of political and economic power, and into the gray zone of illegal transactions. Take the case of Tiburcio Garza Zamora, who built a cacicazgo in the Mexican northeast. Already the owner of Reynosa´s electricity plant during the 1940s, Garza commanded several military and militia units in Nuevo León and Tamaulipas. During the 1950s, he headed Reynosa´s defensas rurales, from which he jumped to federal deputy. He then returned to the military and became zone commander of Chihuahua and Nuevo León (the latter until 1970). He was considered as the´the patriarchal cacique´of Reynosa, where he acquired a newspaper, many nightclubs and cantinas, and was unsurprisingly involved in large-scale contraband operations. 37 The cacical practices of general Olachea Avilés, zone commander in Nayarit and Jalisco during the 1940s, included bribing troops, extorting local populations and using soldiers to eliminate competitors to his liquor-smuggling business. 38 In neighbouring Sinaloa, governor General Macías Valenzuela (1944)(1945)(1946)(1947)(1948)(1949)(1950) was heavily involved in the flourishing opium trade. 39 In a way, these men followed in the footsteps of general Abelardo Rodríguez, who as the governor of Baja California in the 1920s became rich through his involvement in Tijuana´s nightlife and gambling sector, and general Rodrigo Quevedo, the governor of Chihuahua during the 1930s, who was heavily implicated in gambling, bootlegging and other illegal activities in Ciudad Juárez. 40 In around 1950, army commanders firmly supported the ruling party, persecuted those whom they regarded a threat, used violence, and consorted with bandits and criminals: 'in the process [they] blurred the boundary between the government and the realms beyond it´. 41 Cacicazgos and political networks shrouded the informal transactions of these military strongmen and created dark spaces of illegality, crime and impunity.

Forces from without
During decades the arrangements between drug trafficking and state actors weré housed´in the broader edifice of localised informal orders. They were embedded in networks of families, friends, and neighbours and part of local economies, cultures and governance systems. The reproduction of these arrangements was, however, conditioned by forces beyond their reach. From the late 1970s onwards, several (inter) national developments converged and reinforced each other eventually transforming the governance of crime, violence and state nexus. Even though their manifestations and effects were neither unilinear nor uniform, over time they consolidated into a qualitative change. First, US counter-narcotic efforts were increasingly effective in closing down the Caribbean routes that moved drugs from the Andean countries to the US. Pacific routes then gained importance, which made Mexico key for the transhipment of cocaine. The crackdown on the Colombian drug organisations and their eventual fragmentation opened up opportunities for existing and newly emerging Mexican criminal organisations to gradually penetrate and take over cocaine trafficking. In the 1980s, the US cocaine market exploded. It has been estimated that the amount of cocaine consumed in the US and trafficked through Mexico increased from 20% in 1984, to 30% in 1989, to 50% in 1998 and to 80% at the turn of the new century. 42 The cocaine boom massively raised the stakes of the Mexican drugs economy, as it poured unprecedented amounts of cash into the pockets of traffickers and accomplices. But the cocaine trade also affected the organisation of Mexico´s drug trafficking landscape in the sense that the drug was not locally produced. Home-grown drugs require different organisation, territorial, technical, coercive and social inputs and skills than high-value drugs produced elsewhere and´only´shipped through Mexico. In other words, cocaine trafficking in Mexico lacks the historical embeddedness characteristic of the marihuana and heroin business.
Second, the growing organisational and financial muscle of a new generation of drug trafficking organisation coincided with and was partially made possible by major transformations in Mexico's political system, some of which have precedents in the 1970s. Political liberalisation and an arduous´democratic´transition manifested themselves first at local and regional levels, but eventually led to the electoral defeat of the one-party system in the 2000 federal elections, as well as to the crumbling of key pillars of centralised political control. The pluralisation of partisan political power, especially in the north, and decentralisation policies increasingly shifted power from federal to subnational levels. Governors gained enormous political and economic weight in the face of a weakening federal state. 43 Finding refuge under state autonomy, a new Realpolitik federalism granted regional elites and regionally based drug trafficking organisations greater independence and more options to protect their interests. 44 Sociologist Rivelois even speaks of mafia regions or narco-states, where regional elites, often priísta cacicazgos, established alliances with the bosses of illegal economies and appropriated the spoils themselves. 45 In combination with increased criminal resources, the changing configurations of (sub)national power gave rise to a new generation of regional networks of corruption, complicity and impunity.
Third, global liberalisation of trade gave a huge boost to the proliferation and the effective globalisation of illicit businesses as well. 46 With Mexico´s adoption of an exportoriented development model in the mid-1980s, and especially after the launch of NAFTA (1994), cross-border trade multiplied. As trade grew spectacularly, so did traffickersć apability to camouflage illicit shipments. A 2001 government report noted that to cope with this volume, customs officials had to expedite searches and interrogations. Mexican drug traffickers exploited the decreased effectiveness of border control. 47 Neoliberal economic globalisation and criminal internationalisation became good bedfellows. When US demand boomed, Mexicans exploited the comparative advantages of the long and porous border, longstanding know-how in drug trafficking and smuggling, and endemic corruption and impunity. If drugs had been legal, Knight once observed, the achievements of the Mexican narcos would have been viewed 'a great neoliberal, NAFTA success'. 48 Finally, neoliberal economic reforms indirectly affected the political economy of drug trafficking in that they negatively shaped opportunity structures for young Mexicans in rural producing and urban trafficking areas. 49 The narcoeconomy provided these youngsters with exit options. 50 Fourth, the transformations discussed above acquired additional significance in 1985 when corrupt drug trafficking law enforcement relations led to the kidnapping and murder of DEA agent Enrique´Kiki´Camarena and his Mexican pilot Alfredo Zavala. The incident ushered in a new and prolonged phase of US pressure on Mexican authorities. The Camarena affair constituted a turning point in the recent history of state-crime governance in Mexico, as it brought to light the complicity between drug traffickers and the Dirección Federal de Seguridad (DFS), which enjoyed the support of or worked on behalf of the CIA. 51 Fierce reactions from the US, eventually led to the dismantling of the DFS. In time, 'Washington's unyielding pressures. . .increased the cost of political protection [of drug traffickers, WP] and the exposure and vulnerability of traditional mediating mechanisms'. 52 No longer able to obtain reliable state protection, coalitions of trafficking organisations and drug lords started to fracture and resort to violence.
As a consequence, by the mid-1990s, Mexico had become a key player in the international drug economy, which hugely boosted home-grown drug trafficking organisations financially, organizationally and coercively. Meanwhile, federally controlled institutions and mechanisms of political, social and criminal control were leaking power to subnational levels. This occurred in the midst of changes unleashed by neoliberal reforms and trade liberalisation, which deepened the vulnerability of many Mexicans. This is the scenario in which a new generation of powerful drug trafficking organisations exerted growing influence on local and regional societies, an example of which I will examine below. Territorial competition between and internal divisions within these organisations caused new waves of violence. After 2000, and more importantly since 2006, Mexico's outright militarization of the 'war on drugs' substantially contributed to the vortex of violence, causing a human rights crisis. 53

Disruption, cartelization and criminal sovereignty
The consequences brought about by these forces have played out differently in regions and communities across the country. While much research still needs to be done about these diverse outcomes, I argue that these broad forces deeply affected previous informal order arrangements among a range of actors and networks (partially) involved in illicit or criminal activities. Some of these arrangements were able to adapt, while others seriously were disrupted. In yet other cases, they appear to have been replaced by new arrangements in the form of criminal sovereignties. These changes led to new forms of instability, vulnerability and violence.
Anthropologist Natalia Mendoza has written a mirco-history of these larger forces during the last 10-15 years from the perspective of the town of Altar in the Sonora desert, situated along major smuggling routes towards the US border. She speaks of a process of´cartelization´, meaning the´displacement of´old school´independent traffickers, the incorporation and professionalisation of young people in structured and hierarchical regional networks, and the privatisation of the crossing border routes´. 54 A key feature has been the increase in violence and homicides. The local illegal economy, consisting of trafficking of drugs and undocumented migrants, has experienced a series of transformations by which control over illegal activities was taken away from local actors by extra-local (organized) organizations. This not only diminished local control over resources and cash flows, but also undermined social hierarchies, systems of (informal) authority, and the community´s moral order. This has little to do with the absence or presence of illegal activities per se, since for decades the people benefited from drugs and migrant trafficking. From the mid-1990s onwards, the Altar corridor had become one of the most important routes for undocumented migrants. The local population was widely involved. Around 2005, around 150 houses had been turned into lodging facilities for transient migrants, and the town had eight large hotels. Many local families earned good money servicing lodgings, food, communication and transport. Income was distributed widely among the population. Out of town drug traffickers hired local specialists to cross drugs across the border (called cruzadores), because they possessed intimate knowledge of the terrain and the different routes. Considered a sort of´commons´, it was freely used by the town´s cruzadores, who worked independently for drug producers. 55 The image that emerges is similar to that encountered by Van Dun in the cocaine producing villages of Upper Huallaga in Peru, where´the illegal industry was entrenched in everything´. Here the local cocaine patrones were not seen as socially disruptive but enjoyed popular legitimacy. They were engaged in a de facto system of co-governance with the legally constituted authorities that guaranteed safety and order for all. 56 The key transition has been that whereas before these activities were woven into social, political and cultural governance systems, and hence enjoyed legitimacy, their regulation has increasingly become disconnected from them. Regulatory powers have, as it were, been lifted to higher territorial and organisational levels. As drug trafficking gained more momentum and weight, competition between cruzadores, and especially between the drug producers and traffickers intensified. Previously independent local cross-border traffickers were either subordinated to larger criminal organisations or driven out of business. Trafficking routes to different border crossings came under the control of supra-local criminal interests. They were´privatized´. This development extended into the business of migrant trafficking. Over a period of a few years drug trafficking organisationa came to´prey´on migrant flows and tax those involved for using de facto privatised routes to border crossings. 57 In combination with the growth of the drugs business itself, the cartelization of undocumented migrant flows pushed forward processes of bureaucratization, professionalisation, and specialisation. Regional criminal organisations recruited and trained young people from the town for different roles, such as look-outs or sicarios. Mendoza found that the incorporation of these chavos in cartel-like structures disrupted their family and community ties:´They become immune . . . to local social controls of violence, and they situate themselves above or outside the moral order´. 58 It increased the heightened sense and experience of insecurity and fear. Cartelization meant centralisation of decision making and´settlementsẃ ith the regional authorities (state police), concentration of illegal profits, and the extraction of resources from local communities. In other words, it resulted in the disruption of locally managed illicit economies, which were embedded in socially, politically and culturally legitimate institutions and informal arrangements. Despite the private and coercive appropriation of public space by drug trafficking organisations in association with state police, Mendoza has signalled some (subordinated) accommodation by local players. Elsewhere in Mexico, however, developments went further and eventually triggered an armed uprising by autodefensas or community police forces. While the hot and dry Altar region has long been a trafficking region, the hot and fertile tierra caliente in Michoacán, and the adjoining coastal mountain region have been both drug producing and trafficking areas. These differences underline my argument about the regionally dependent and multidirectional transformations of the crime-governance nexus. What happened in Michoacán is, in a way, a´radicalizedv ersion of Altar. Here, during the 2000s, the main features of cartelization evolved in an´amplified´manner and covered a larger territory, thereby giving rise to a system of criminal sovereignty. It subsequently set into motion a counter-dynamic involving state and non-state violent actors that eventually brought down the criminal governance system of the Knights Templar (Templarios). Michoacán had become engulfed by drug-related violence and armed actors. It was here that president Calderón, himself from Michoacán, started his militarization campaign at the end of 2006. At the time, the recently appointed Minister of the Interior boasted about´. . .the recovery of public spaces grabbed by organized crime. . .´and en end to impunity of criminals. 59 As it turned out, the worst was still to come. A few months before, an organisation called La Familia Michoacana proclaimed its existence by tossing five severed human heads onto a nightclub dance floor. A few years later, two grenades were hurled into a crowd during Independence celebrations in the state capital, killing eight and injuring hundreds. 60 In May 2009, federal law enforcement arrested dozens of municipal authorities and imprisoned them in Mexico City on suspicion of collaborating with La Familia, then the most powerful criminal organisation in the state. This so-called michoacanazo seriously undermined the relationships between levels of government. 61 In March 2011, Los Caballeros Templarios de Michoacán emerged from the ranks of La Familia. The Templarios combined an enigmatic regionalist and messianic discourse with brutal extortion and political domination. According to official statistics, between December 2006 and May 2013, 4635 people were killed in Michoacán. 62 What were the main features of these regional criminal organisations? Due to its geographical isolation and its reputation as a´peripheral, indomitable, indolent place´, southern Michoacán has historically been ruled by mestizo caciques, rancheros and hacendados. 63 Since the late 18 th century, they had migrated towards the Pacific coast conquering´new territories and pushing out indigenous communities. 64 In the process, they developed a strong sense of practical autonomy and social identity, and kept external state institutions at bay. A strong ranchero society and culture valued family, hard work, and popular Catholicism. 65 Rancheros exercised local control over economic resources and the means of violence.
Drug cultivation and trafficking have deep roots in the area. After the Second World War, state-led development projects reached southern Michoacán, pouring massive resources into the area in the form of highways, credit, irrigation, hydroelectricity and, finally, the seaport and mining-metallurgic complex in Lázaro Cárdenas. A prosperous agrarian regional economy emerged, especially in the tierra caliente, which produced cotton, and then limes, mangoes and melons, largely for the US market. However, economic change and substantial population growth also came with an increasing concentration of wealth and agrarian conflicts. In addition, the region had a fair share of delinquency and pistolerismo. The region was rife with conflict and violence. 66 The transformation of the regional economy and the emergence of transnational commercial networks also benefited local drug traffickers. In the 1960s, poverty pushed many into marihuana cultivation, a development that generalised in the 1970s. In 1974, the Attorney General's office claimed that Michoacán was where most marihuana was produced and most drug-related detentions occurred. 67 The southern Sierra Madre witnessed a spectacular rise in marihuana cultivation during the 1980s. With the neoliberal abandonment of development projects an illicit drug economy started to fill the void. 68 With the arrival of cocaine, Michoacán evolved from a major producing area to one that combined production and transportation. The port of Lázaro Cárdenas became an attractive reception point for South American cocaine and chemical precursors from China, turning the tierra caliente into´the largest amphetamine factory in the world´. 69 Cocaine and synthetical drugs raised the stakes of organised crime and transformed Michoacán´s illegal economy. 70 The region´s first home-grown drug-trafficking organisation was run by the Valencia brothers. Rooted in the drug-producing areas of southern Michoacán and with networks in the US, they later went into business with Colombian cocaine producers. While the Valencias (or the Milenio cartel) shipped tons of cocaine through Michoacán to the US, they also became involved in the production of synthetic drugs. They did so in close cooperation with the Sinaloa drug trafficking organisation, while Michoacán became a much-coveted territory for rivals. The Valencia cartel became part of the violent conflicts between the Sinaloa and Gulf organisations. It was in this context that the Gulf cartel sent the paramilitary force of Los Zetas to the region. After 2000, confrontations between Los Zetas and the Valencias turned southern Michoacán into a battlefield. The take-over by Los Zetas was widely viewed as an´external intervention', which directly affected local communities. 71 Ranchero sovereignty and autonomy came under threat.
By 2006, Los Zetas were driven out of Michoacán by La Familia, which became the major criminal organisation in the region until 2011, when it was replaced by the secessionist branch of Los Templarios. 72 Despite their localist appeals in opposition to Los Zetas, the modus operandi of both organisations was modelled on that of Los Zetas, as they combined drug trafficking with taxation of social and economic actors. At the same time, they introduced new criminal styles and strategies. 73 They aspired to construct a comprehensive project of rule and sovereignty that went beyond criminal force and violence. It claimed to be grounded in regional ranchero history. It was built on three pillars: political influence and protection, a particular ideological-cultural project, and a complex system of illegal activities and resource extraction.
Ever since the michoacanazo (2009), political influence of criminal organisations in Michoacán was widely debated. 74 A drug trafficker associated with the Caballeros Templarios openly spoke of how they worked with all political parties, supported some election campaigns, pressured uncooperative candidates to withdraw and acquired influence over political appointments and government contracts. 75 The return of the PRI to the government of Michoacán in 2012 (after 10 years) was said to be due tó a political electoral strategy and mutual agreements´between the Caballeros Templarios and senior advisors of the later PRI governor. 76 At the local level, so-called jefes de plaza controlled everyday affairs.
La Familia and Los Caballeros Templarios presented themselves as the legitimate defenders of Michoacán against outside forces (Los Zetas or federal law enforcement). 77 They developed a regionalist discourse that emphasised their local embeddedness and their wish to return order to the state. In 2006, La Familia promised to end kidnapping, extortion, thievery, and the´humiliation´of the people of Michoacán. 78 To achieve this they employed´divine justice´. The Templarios cartel shrouded itself in quasi-religious and spiritual symbols and narratives ostensibly designed to provide their criminal governance system a moral legitimacy. The discourses of regionalism, autonomy, Catholicism, sacrifice, work and violence resonated with ranchero communities. In their initial phases, La Familia and Los Caballeros Templarios had some legitimacy. Several accounts have shown that the ideological and political project gained content through conflict resolution, the establishment of order, social development and economic opportunities:´Ironically, they presented themselves as guarantors of order and security in the communities and the roads. It was a precarious and illegal order, but one that no one else could offer. . .´. 79 The Caballeros Templarios functioned as an´alternative governance´system or as 'a state, a local and regional government, a legislator and efficient organizer of daily life'. 80 But these criminal systems were fraught with contradictions, internal divisions, excesses of violence and the escalation of criminal activities themselves.
When the Templarios took control of the tierra caliente, they promised cattle ranchers, farmers and businessmen to limit themselves to drug trafficking. That was acceptable to most inhabitants, who had long lived alongside drug producers and traffickers. 81 But after a while, and despite claims to protect and defend the people of Michoacán, they started to prey on the local population and economy with cuotas on businessesa type of property tax. The Templarios eventually engaged in kidnapping, extortion and racketeering, and took over key parts of business chains. 82 The fight about lime production and distribution in the western part of the tierra caliente, one of the most fertile and productive regions of the entire country, would play a key role in the emergence of the self-defence forces in early 2013. While the extortion of legal businesses generated millions, the Templarios also preyed directly on families. People had to pay for their property, cars and number of children in school. 83 The local bosses of the Templarios managed a racketeering system that took 10% of municipal budgets. In this way alone, the organisation monthly cashed in around twenty million dollars. 84 The depth of this extortion system fuelled feelings of anger and frustration and clashed with a historical sense of economic and social sovereignty of ranchero communities. 85 Autodefensas: family, locality and drug trafficking On the 24 th of February 2013, inspired by the experiences of indigenous communities elsewhere, whilst tapping into ranchero traditions of autonomy and sovereignty (the defensas sociales from the mid-20 th century come to mind), cattle ranchers, farmers and professionals from the towns of Tepalcatepec, La Ruana and Coalcomán launched the first autodefensas. 86 It marked the beginning of an armed uprising that wanted to take decision-making powers about work and income, and life and death back from the Templarios. It created a triangular conflict between organised crime, the state and armed citizens. It was an uprising against an oppressive system of economic control, dendritic extortion networks, the subordination and corruption of local political and administrative institutions, shored up by idiosyncratic quasi-religious and regionalist narratives, all of which formed the criminal sovereign order of the Templarios. The autodefensa movement spread quickly spread across the tierra caliente and towards the coastal mountains. 87 Shortly before Christmas 2013, La Huacana was declared the 30 th municipality´liberated´from organised crime. Each municipality under its control founded its own autodefensa unit. In hindsight it is remarkable how quickly the Templarios cartel collapsed. In approximately eighteen months the autodefensas wiped out one of Mexico´s most powerful criminal organisations. By then tense relations between autodefensas and the federal state had taken centre stage, as the latter was determined to disarm the former and incorporate them into government-controlled police forces. How should we interpret what happened in Michoacán between the beginning of 2013 and the end of 2014? I will elaborate briefly on three closely related themes: family and gender, locality and identity, and the ambiguous relations between autodefensas and drug trafficking.
When doctor José Manuel Mireles, the charismatic autodefensa leader from Tepalcatepec, was asked about the autodefensas he argued that what really detonated the movement went beyond economic extortion and political domination and had to do with the Templarios starting´to mess (meterse) with the family´. When the Templarios raped girls and demanded the daughters and wives of rancheros for sexual abuse, when they undermined the moral, sexual and gender basis of ranchero communities, they revolted. Community members told French researcher Romain Le Cour Grandmaison that what antagonised the Templarios was´extortion, disappearances, public displays of violence, and above all violence against women in the form of rape and abductions´. It showed that they no longer respected moral rules and informal arrangements. 88 Closely connected to the threats of the intimate sphere of family and gender relations is the dispute about local belonging and protection at the level of the´pueblo´. In essence, the autodefensas fought to re-localise control over economic activities, political power, territory, justice, religion, and security, as well as family and gender values. Recent research by Le Cour Grandmaison has greatly increased our insights into how the struggle for re-localization involved processes of identification, differentiation and antagonism. 89 The Templarios built a network of jefes de plaza responsible for the administration of illegal businesses and the cartel´s law, financial management, and political control, including over elected mayors. Since in most cases these jefes and their henchmen did not come from the locality they controlled on behalf of the Templarios organisation, they were widely perceived as representatives of an external oppressive entity. They behaved as gente de fuera (outsiders). 90 And they were the first people the autodefensas went after. Many fled with their closest allies, which retrospectively confirmed their status as outsiders. Interestingly, a member of the Buenavista autodefensa told Le Cour Grandmaison:´At that moment, we realised that the jefes were finally fairly alone. They weren't from the area and they didn't have any family or ties in the municipality. So they left in a hurry. And once you've left, it's impossible to come back if you're not from around here.´9 1 In contrast, the members of the autodefensas all came from the municipalities they represented. Although these groups would move across the tierra caliente during the first phase of the uprising, they always returned to their own towns and villages. T-shirts with the name of whatever town they came from expressed their geographical roots, and enabled them to be recognised as such by the population at large. Constructing popular legitimacy, the autodefensas asserted nativist discourses and localist solidarities. They did so in opposition to the Templarios, who had betrayed their initial claim´to protect the inhabitants and the sacred territory of the free, sovereign and secular state of Michoacán´. 92 During the violent encounters with the Templarios, the autodefensas labeled them as´outside actors´. In fact, the (armed and political) struggle itself was successfully waged on the basis of intimate local social and territorial knowledge and skills. 93 Autodefensas leaders portrayed themselves as authentic defenders of the´peoplé , and prided themselves of having successfully reclaimed popular sovereignty from an alienated criminal organisation. A few months after the outbreak of the uprising, Mireles said:´The village ['pueblo' in Spanish, WP] has been taken back by its own people [also 'pueblo' in Spanish, WP], by forcing the criminals out of all of our communities. . . . It wasn't taken back because of the army or the federal police. It was thanks to the locals' courage that we could battle.' 94 The´people´/autodefensas vs. 'outsiders'/Templarios antagonism placed the latter beyond the local social, political, and moral order. Though rooted in Michoacán, the Templarios cartel had morphed into something that became disconnected from the local moral economy. In Michoacán, a particular brand of criminal sovereignty clashed with and alienated locally embedded informal community arrangements. The latter brings me to my last point about the ambiguous relations between drug traffickers and the autodefensas. An interesting starting point is that members of autodefensas differentiate between´good´and´bad´narcos. 95 A closer look at this contributes to our understanding of how ordinary people evaluate shifts in the relations between crime, drug trafficking, and violence on the one hand and particular social and political orders and governance systems on the other. It will become clear that, not unlike the case of Altar, the key issue revolves around the question if illegal economies are embedded in locally regulated (informal) orders or if they have become disembedded and integrated into larger organisational units perceived as foreign and inimical to local communities.
From the start, the autodefensas movement in Michoacán was suspected of being partly driven (and financed) by drug traffickers who saw the uprising as an opportunity to advance their own interests. 96 As it turned out, rancheros that made up the bulk of the autodefensas did indeed build coalitions with local traffickers. While this may seem inconsistent and reprehensible from a´formal order´perspective, it does not from the lived experience of informal arrangements and the practical morality of ranchero communities. For decades, these rancheros had been familiar with drug cultivation and trafficking or had lived alongside traffickers. 97 Local traffickers, for example, had been forced into taxation arrangements with the Templarios. But over time, the jefes de plaza disrespected the agreements and demanded increasingly high taxes from previously independent traffickers or producers. A member of the Buenavista autodefensas council told Le Cour Grandmaison:´In the end, with the Templarios, the narcos from around here only kept about 30 percent of profits. The rest went to the cartel and the leaders. Even crystal meth production stopped being profitable!´(my emphasis). 98 A leader of the selfdefence group of Aguililla, a notorious drug trafficking region, acknowledged that´[T] here has always been trafficking here. But before, the narcos didn't pick on the population . . . If you didn't bother them, they didn't bother you´. 99 Moreover, when autodefensas occupied territory and chased out (or killed) Templarios bosses and their henchmen, it did not mean that all narco-activity and traffickers would disappear. Community members who had been involved at the lower end of the Templarios organisation could be pardoned and return to local society by joining the local autodefensa organisation or by undergoing a symbolic sentence. 100 Justice-making from below was yet another expression of re-localising decision-making power.
In the context of dismantling the Templarios criminal organisation people distinguished between belonging (or not) to that particular organisation and the involvement in drug trafficking and cultivation per se. Those who had occupied senior positions in the cartel and engaged in certain practices, such as extortion of ordinary citizens and business, were severely punished. The issue was less with crime in itself. Local populations did not so much oppose drug trafficking as a livelihood strategy, but rather a particular way of governing drug trafficking. 101 The latter connects to Mendoza's concept of 'cartelization', i.e. the consolidation of structured, professionalised and regional criminal organisations or sovereigns, which disrupted previously existing informal arrangements of the local drug economy. This organisationl transformation, Le Cour Grandmaison argues, is accompanied by a shift in the social representation of traffickers, who now became exclusively identified as beholden to the cartel. In contrast, reference to a local '. . . moral economy in relation to drug trafficking. . . was related to a desire to restore a pre-cartel era when this activity was not directly controlled by structured criminal organizations and when the 'employees' were above all acquaintances, friends and family members who were not defined as narcos. For populations that had been left exasperated by the cartel's predatory practices and abuses, the self-defence movement indicated a return to this 'order', indeed an informal order (my emphasis). 102 Social and cultural belongings shape notions of crime, violence and illegality. 103

Conclusions
When the bishops from Oaxaca and Chiapas wrote in early 1984 that money and violence enabled powerful drug mafias to take 'practical control of people's lives' they were on to something. As I have shown, they were witnessing how illegal (drug) economies once embedded in local social, political and moral arrangements and supported by a modicum of popular and elite legitimacy, became disrupted as they were lifted to other societal and organisational levels. These processes involved the cartelization, professionalisation and the para-militarization of Mexican drug trafficking organisations. All this came with an increase of violence. In my approach to drug trafficking, I distinguished between distinct historical manifestations of the crime, violence, governance nexus.
This article has drawn on a combination of theorising on the informal order, personalistic political culture and caciquismo, and the findings of recent historical and anthropological research on the regulation of drug trafficking in Mexico during different historical periods. The analysis enriches conceptual debates and shifts the conversation beyond normatively laden´democratization´or´security´perspectives on drugs, crime, violence and insecurity in contemporary Latin America. Most importantly, this article aspired to demonstrate that theories of the informal order and caciquismo, as well as careful historiographic and ethnographic research enable us to leave behind the interpretive logic of opposition, dichotomy and antagonism when it comes to grasp the relationships between crime, violence, governance and state-making, and, instead, always understand and visualise them in terms of connections, entanglements and hybrid arrangements between state and non-state actors, formal and informal institutions and legal and illegal practices. 104 Notes