Grand corruption in Mexico: The convenient disappearance of an agrarian community

This article offers an exemplary case of grand corruption in Mexico, narrating the story of how a series of actors and institutions at all levels collaborated to obtain favours, while dispossessing an agrarian community of its property rights, and replacing its elected Commissariat with a person of their choice, paying him the equivalent of US$3 million. This manoeuvre gave right of way to President Peña Nieto’s interurban train, at the risk of ruining the community’s water supply. The study argues that we should distinguish the concept of grand from that of petty corruption. It sees neopatrimonialism as a subtype of grand corruption, describing complex networks of actors across several institutions and levels, which operate the corrupt practices, with agency at all levels. President López Obrador won the 2018 election on a promise to fight corruption. This study illustrates the difficulties he will encounter trying to realize this promise, while suggesting that combatting impunity would constitute a promising start.


Introduction
Mexico experienced a landslide election in favour of Andrés Manuel López Obrador (AMLO) for president in July 2018, largely due to popular trust in his plea to combat corruption. According to Morris (2000), such promises have constituted the hub of Mexican politics for decades, however, this time the voters confided in the candidate. Although many say that 'everybody does it', Mexicans in general apparently are exasperated with the political practice of corruption. It is seen as a practice that consumes considerable parts of public resources, often impedes the carrying out of justice and public policies, and at times blurs with organized crime. The case I discuss in this study is highly relevant in this context; it concerns a particular form of corruption conceptualized as grand or neopatrimonial corruption.
In 2014, the Agrarian Tribunal abolished the property rights of an agrarian community at the outskirts of Mexico City, and ordered its removal from the Agrarian Register. This occurred exactly at the time when the Ministry of Transport needed access to its territory in order to construct President Peña Nieto's megaproject, the interurban train between Mexico City and Toluca. The Agrarian Tribunal also designated a person -let us call him Gabriel -who certainly was a comunero, but alien to the community's elected Commissariat, to represent the community. The Ministry of Transport then negotiated the access with Gabriel, who received a personal check of 47 million pesos (US$3 million) for the possessory rights.
The train tunnel of four kilometres through a hill was built where Acopilco's main water sources are located. Most comuneros 1 did not receive any compensation, nor is there any guarantee concerning the potable water supply of the community. When the Agrarian Tribunal issued a verdict that restored the community's rights in 2016, Gabriel contested it, thus postponing its execution and perpetuating his own position. The community still lacks legal existence, property rights, elected representation, and monetary compensation. With the compensation money at his disposal, Gabriel paid an amount to the comuneros who supported him, and organized a group of thugs to subdue protests in the village. These steps produced a deep division among the comuneros.
This story demonstrates how state effects can be produced, 2 seemingly by the institutions themselves, but in reality prompted by officers in personal networks of power across the institutions where they are employed. The case demonstrates that the Agrarian Tribunal and the Federal District Court participated in this network. My study indicates that these practices are not necessarily controlled from above, thus difficult to eradicate, even with the best of intentions, though an emphasis on combating impunity, the sister of corruption, would help (Morris, 2000). Furthermore, these practices are not only carried out by particular groups, but often take the form of a generalized competition over power and resources, carried out in dynamic, hierarchical networks with scope for agency even from below.
My contribution fills a gap in the literature on grand corruption in Mexico first of all with an exemplary empirical case based on participant observation and multi-method ethnography. Rather than explaining or measuring corruption, it aims to describe and understand some of its mechanisms, piecing together fragments of a process, which took place over the years and involved a large number of actors and public agencies with no institutional connection, pointing to the dynamic workings of these networks. I argue the importance of distinguishing between petty and grand corruption, explore the utility of the concept neopatrimonial corruption, and suggest some modifications to its definition. The main contribution is to highlight the way in which the corrupt practices appear to be state effects, approximating legitimacy, calling for obedience. Thus, it is paramount for the corruption network that some participants hold positions of authority in state institutions.

Corruption, neopatrionalism and the state
Traditionally, economists and political scientists have dominated the literature in this field (Anders & Nuijten, 2007), often in unison with multilateral agencies like the World Bank, based on an understanding that corruption is linked to "underdevelopment" and should be addressed by international organizations and anticorruption agencies. These have tended to take the view that corruption is embedded in the local culture (Gledhill, 2004, p. 155), from a "developed" North perspective versus the "underdeveloped" South. The research and policy-making on corruption has often been related to geopolitical, neo-colonial interventions. This forms part of the background for the reticence of especially anthropologists, but also other social scientists using an ethnographic approach, to engage in the study of corruption. On the other hand, there is also a growing demand among local populations and in academia for exactly this type of research, contemplating that this could contribute towards a better understanding of the practice (Ruhl, 2011;Torsello & Venard, 2016), and ultimately help to fight it. However, several authors suggest there is no such thing as a remedy for this disease (Torsello, 2011, p. 4;de Vries, 2007, p. 161). These authors, nevertheless, call for micro-studies, using participant observation and other ethnographic methods; in recent decades, several ethnographically oriented scholars have contributed towards the study of this topic (see for example Nuijten & Anders, 2007;Pardo, 2004), though mostly petty corruption.
What, then, are we talking about when we talk about corruption? The definition of corruption would seem to be straightforward, like the one quoted in Gledhill (2004, p. 159, citing Coatsworth, 1999, "the abuse of public power for private gain." Another simple, but more comprehensive definition is the one used in the international ranking list, Corruption Perception Index (CPI) by Transparency International in collaboration with the University of Göttingen. Here, data is collected from local and international business leaders and risk analysts: "the misuse of public power for private benefits; for example, the bribing of public officials, taking kickbacks in public procurement, or embezzling public funds" (Canache & Allison, 2005, p. 95). The practice and study of corruption take place all over the world, but is not understood in the same way everywhere. What would be considered corrupt, for example bribery, in one cultural context, following Gledhill (2004, pp. 161-162), could be seen as legitimate reciprocity in another. Furthermore, following Uslaner (2008, p. 58), culture is not itself an explanation of corruption; he finds that inequality underlies and explains corruption. Legitimacy does not necessarily correspond to legality, even within the same cultural context (Pardo, 2004, p. 5). Based on long-term ethnographic fieldwork in Mexico, de Vries (2007, p. 162) argues that even clear cases of bribery were considered legitimate by the peasants he worked with, provided the public officials they had to give money to did "comply with their part of the transaction". He points out that everyday corruption is so widespread that most people are entangled in webs of complicity (de Vries, 2007, p. 145).
The issue of legitimacy points us to a concern, which is of fundamental importance in my study, the distinction between petty and grand corruption. Reviewing the work of Uslaner (2008), Smith (2010Smith ( , p. 1177 underlines that people see grand corruption as illegitimate, while they often take a different view on petty corruption. Ruhl (2011, p. 33) argues that such a distinction is necessary in order to avoid distorting comparative studies; the two types of corruption are carried out at different levels and perceived by different audiences. Anders and Nuijten (2007, p. 6) are not primarily preoccupied with elaborating universal definitions, often sought by political scientists and economists in order to generate quantifiable and comparable data sets. The elaboration of a definition of political corruption that is applicable across cultures and historical times is precisely the aim, apparently unattainable, of an entire article (Philp, 1997) discussed by Gledhill (2004, pp. 160-163). Nevertheless, Anders and Nuijten underline that some basic typologies and classifications, like petty and grand corruption, provide a useful point of departure for the study of corruption.
Córdova and Ponce (2017, p. 239) offer a valuable distinction between petty and grand corruption. Petty corruption concerns the implementation of political decisions (for instance, the provision of public services) as to who will benefit from them, and when. Grand corruption, on the other hand, involves modifying political decisions, laws, and regulations themselves, in order to benefit influential politicians, high-ranking public servants or private actors. Based on my study, I would add court decisions to the list of what could be modified. Furthermore, some authors observe that grand corruption at times is linked to, or blurs with organized crime (Córdova & Ponce, 2017;Heywood, 1997, using the term political corruption). It is, however, common in the corruption literature not to distinguish between those two types, as for instance the CPI, which is used as data in corruption studies worldwide.
Political corruption is a term often used, frequently without any further definition. Heywood (1997, p. 421) defines it as "corrupt activities which take place wholly within the public sphere or at the interface between the public and private spheres." This definition does not differentiate between petty and grand corruption. Anders and Nuijten emphasize that more than conceiving of corruption as an individual act, they see it as institutionalized and embedded in the wider power relations in society, with systemic and structural dimensions , p. 2). De Vries (2007 goes to the extent of maintaining that "corruption stands for the everyday workings of power in Mexico". Although ethnographic fieldwork stands out as particularly well suited for illuminating processes of petty corruption in its many forms, several researchers have pointed out (Blundo, 2007, p. 37;Torsello & Venard, 2016, p. 37) that the core method of participant observation generally leaves grand corruption beyond its scope. Córdova and Ponce (2017, p. 239) show that it is difficult to identify the actors of grand corruption, which also applies to gaining the necessary long-term access to the processes of interest. These often include criminal activities, and thus constitute some danger for both researchers and informants (Torsello & Venard, 2016, pp. 37, 49). However, as Pardo (2004) surmises, ethnographers who have carried out fieldwork on other issues, thereby gaining the trust of their informants, may 'happen' to come across case material on corruption. It may be impossible to do outright participant observation on this practice, but in conjunction with other methods, case studies may be constructed (Pardo, 2004, p. 3). Empirical studies on grand corruption are rare; however, my study is a case in point. The lack of empirical studies is also noted concerning neopatrimonialism (Monsivais & del Río, 2013, p. 61;von Soest, 2010, p. 3); interestingly, Monsivais and del Río (2013) see it as a form of corruption.

Neopatrimonialism
The concept was coined by Eisenstadt (1973), referring to regimes, which had inherited institutional models of modernization from the colonial periods, but developed "many characteristics which differed greatly from the 'nation-state' or revolutionary models of modernity" (1973, p. 13). The term often has referred to a particular form of domination, or regime type, relying on the observation of a series of political practices, that is, traditional or patrimonial practices within formally modern states (hence the neo-) with rational-legal rule and professional bureaucracies (Bach, 2011;von Soest, Bechle & Korte, 2011). Von Soest et al. (2011, p. 1309) advice that the concept should be used analytically, not normatively, and that at any rate "the systemic character of neopatrimonialism is a matter of degree and not principle." Bach (2011) contributed an extensive overview of the literature on neopatrimonialism on a global basis, discussing the variations in its use on several continents; it has been used as a prototype for the "antidevelopment state" in Africa, but not to the same extent in Latin America. Oszalk (1986, p. 229) has, however, suggested neopatrimo-nialism as an adequate description of several states of the region, maintaining that "personalist government turns states into the private government of those possessing the necessary power for the exercise of political domination." It is often implied, though not necessarily explained, that this concept has an explicatory power (Monsivais & del Río, 2013, p. 58). In this respect, concerning Latin America, Bach (2011, p. 282) refers to the legacy of three centuries of Spanish and Portuguese colonial presence.
There is a debate whether neopatrimonialism should be considered a third, hybrid type of regime, inserted between the authoritarian and the democratic type (Durazo, 2010;Erdmann & Engel, 2007;Oszalk, 1986). While some scholars underline that neopatrimonialism does not constitute a transition from authoritarian to democratic rule (Durazo, 2010), Pitcher et al. (2009) characterize this term used about states as a misunderstanding of Weber, who they claim used the term 'patrimonialism' to refer to a form of legitimate authority, not a type of regime. In their comparative study of eighteen Latin American countries, Monsivais and del Río (2013) find that neopatrimonialism does coexist with both authoritarian and democratically elected regimes. These authors argue for using the term, not about regime types, but about a set of political practices, which develop around public institutions, as a specific form of corruption, compatible with clientelism and patronage (Monsivais & del Río, 2013, p. 49). In the corruption terminology I have set out before, it would represent a subcategory of grand corruption.
Monsivais and del Río (2013, pp. 48-49) define neopatrimonialism as a form of particularistic capture and manipulation of public norms, agencies and resources. Particularistic because they are used to benefit a delimited and concrete group of actors. Capture and manipulation, since these actors have the capacity to use government rules and resources. They further underline a point, which is paramount in my study; the actors can do this discretionally and strategically. Under certain conditions, these rules and resources fill their formal function, but under other circumstances, they are evaded or used arbitrarily. Concerning the issue of benefits for delimited and concrete groups, my case points to a more dynamic process. In order to account for the ways in which the corruption process evolves, I make use of Nuijten's (2003) concept of the force field referring to what emerges around substantial resources that become available.
Following Monsivais and del Río, neopatrimonialism is a specific form of corruption among other forms. What differentiates it, are three characteristics: 1) some actors occupy formal positions of authority; 2) these individuals are connected with concrete groups hierarchically structured in networks; and 3) these individuals are responsible for actions that are in tension or directly in contradiction with the legal order (Monsivais & del Río, 2013, pp. 48-49). These authors, referring to the last point, indicate the need for a profound immersion in the context in order to follow the decisions, resources and recipients of these transactions.

The state
The state plays an important part in this study, since certain practices are labelled corrupt by reference to a set of legal or other norms (Anders & Nuijten, 2008, p. 13); furthermore, the resources used in the corrupt transactions are public resources. In addition, the actors in the process need to occupy important positions within state institutions in order to carry out these transactions and make them appear to be state effects. The state, in a Weberian context, is the territorial monopoly of legitimate violence, and this monopoly is the key enabler for state enforcement of law and order and exercise of justice (Weber, 1978, in Ong, 2018. Public institutions make up the state-system, which already Weber did not consider an object (Neumann, 2005, p. 195).
The state-system, following Abrams (1988, p. 82) is "a nexus of practice and institutional structure centred in government." He further surmises that a state-idea provides unity to the state-system, and that the claimed reality of the state is what legitimizes the institutionalization of power, conducive to establish political authority and legitimacy, as Krohn-Hansen and Nustad (2005, p. 5) point out. The type of legitimacy on which a system of domination relies, shapes the form of obedience that is called for (Durazo, 2010, p. 87). This point helps us understand some of the background why the actors in this study would go to such lengths in order to make the corruption process appear to be a state effect, seemingly legitimate, demanding obedience.
In order to understand modern states we need to explore how power works, including "the everyday rules and normalizing technologies that govern conduct and render populations governable in the first place" (Shore, 2005, p. 239). Following Trouillot (2001, p. 131), the state is a "set of practices and processes and their effects." He specifies four state effects, out of which the legibility effect, which concerns ways to classify and regulate populations, is of special interest with regard to the agrarian reform and the topic of this study. The agrarian community in this study had its territory recognized by the state in 1992, and only the state could remove it, but needed legal justification to do so.
The process of state formation became a material reality for rural communities all over Mexico with the implementation of the agrarian reform after the revolution 1910-1920 (Baitenmann, 2005), producing a remarkable stability (Mallon, 1994, p. 72). Institutions were set up to distribute and entitle land and transform the rural population into controlled subjects and potential beneficiaries. This process took place in the articulation between rural communities, lawmakers, agrarian institutions, bureaucratic apparatuses and technical staff all over the territory, by means of state effects like tracing land histories, measuring land, producing censuses, documents and maps. The rural communities consequently needed to produce letters and applications, localize contacts, lawyers, and other specialized personnel.
By 1992, half the country's arable land had been distributed to some 3.5 million persons organized in 30,000 ejidos 3 and agrarian communities (Secretaría de la Reforma Agraria 1998, p. 313, in Baitenmann 2005. This massive interaction between the central government and the local peasant communities contributed towards the formation of the modern Mexican state (Baitenmann, 2005). It was characterized by an ambiguity, which became typical of the power relation between the state and the rural communities (Nuijten, 2003). The communities became potential beneficiaries, while at the same time controlled subjects. This constant and ambiguous interaction between the community and the state, particularly the agrarian authorities, play a significant part in the process under study. In my case, the power network in operation had actors positioned within all branches that make up the state system, except the armed forces and the legislative power. 4

Methodology
There is a striking absence of empirical case studies concerning grand corruption and neopatrimonialism in the research literature, mainly due to the difficulty and danger involved in using qualitative methods, especially participant observation, in the relevant fields. My contribution is an exemplary case, offering concrete, contextualized knowledge, which more than proving something, opens up for learning something (Flyvbjerg, 2006, p. 224). The case is the outcome of knowledge I happened to come by in the course of ethnographic fieldwork comprising participant observation on other topics (Pardo, 2004) in Acopilco, the community under study, which gained me confidence among many local actors. I did fieldwork while living there for a period of more than 24 months in the timespan 2003-2018 (see for example Hagene, 2007;2010a;2010b;2010c;2015a;2015b;2015c;Hagene &González-Fuente, 2016). Once the topic of grand corruption emerged, I pursued it by doing interviews in the community, the administrative centre, and in the relevant tribunals and ministries downtown, 5 as well as searching for written information in the media, letters, receipts, laws and verdicts, and in my previous field notes. The mere piecing together of the case story was a complex and contradictory process, as Flyvbjerg (2006, pp. 237-238) warns. He argues, however, that while these narratives are difficult to summarize, they possess an irreducible quality, and are in themselves findings.
To have access to the field is one important outcome of such long-term relationship with a community, but in addition, it is vital to come across some actual fragments of the chain of events that constitute part of the process of grand corruption. The central document of my study (Magistrado Titular, 2016) is public and available for everyone, but in order to scrutinize this uninviting text of 47 pages, first one needed to suspect that something important was hidden there. Due to my immersion in the community, I became aware of incongruences and paradoxes, a method known as abductive reasoning (van Hulst et al., 2015, p. 1). I shall refer to this document as the Verdict. Within the Verdict another central document emerged, authored by a Federal District Court (Juzgado Séptimo de Distrito en Materia Administrativa, 2014), the Oficio 3974.
The centrality of these two documents in this study calls for an explicit source criticism. Kjeldstadli (2007, pp. 169-190) proposes three steps. First, the origin and purpose of the documents. The origin is already stated, while the purpose is ambiguous, both taking and giving rights to Acopilco, and replacing the Commissariat with Gabriel as its representative. I will expand on this later. The second step, following Kjeldstadli, concerns the interpretation of the source. The two vital events, to take the rights away, and replace the representation, occupy a minimum of space, fourteen lines, to be precise, within the forty-seven pages of the Verdict. I argue that these issues have been hidden in the bulky text of the Verdict. Third, the credibility of the sources; this is undergirded by emerging from courts, which clearly adds to their capacity to imbue the operation with authority and legitimacy. The decisions stand out as state effects, which demand obedience, despite the complete lack of logic in these decisions. As for the terminology applied in my study, I shall use two Spanish terms, insubsistente / insubsistencia (unfounded, not valid), and comunero (a person with registered rights in an agrarian community). An agrarian community owns collective property recognized by the agrarian authorities, and is governed by the Commissariat elected by the comuneros. Finally, in order to protect the identity of informants, I use fictitious names, or terms like Comunero A, B, C, or officer A, B, C. To these I always refer as "he" or "him", regardless of their gender.

The context
Acopilco is located on the western outskirts of Mexico City, in Cuajimalpa Delegation (now Municipality), bordering on the State of Mexico, at 3,000 m above sea level, with a population of 24,000 (INEGI, 2010), the majority of which earns less than two minimum wages (INEGI, 2000). 6 The villagers see themselves as divided into two main groups: the natives and the avecindados. Each group has roughly the same size. The natives are further subdivided into those registered with agrarian rights (2,344 comuneros) and those who are not. The Mesoamerican village of Acopilco had its land rights confirmed by the Spanish King in 1559. However, after the Mexican revolution 1910-1920, the village needed formal recognition by the Mexican President in order to be legally constituted as an agrarian community with collective ownership to its territory. Acopilco's process of recognition and entitlement started in 1964. 7 The struggle lasted 28 years, until the presidential resolution finally materialized in 1992, for collective ownership to 1,608 hectares of land, mostly forest, with many water springs. The community also experienced other agrarian trials concerning borders with neighbouring communities; they all went on for dec-ades. According to a former Commissioner, there has been at least four cases like this. He calculated that in his 3-year period, he might have dedicated about 80 per cent of his work time to winning one such trial. 8 The constant preoccupation with these court cases created what I would term a lifeworld of agrarian trials. In her article about an ejido in Jalisco, Nuijten (2004) describes this kind of endless processes as one aspect of state power wielding; the process never ends, the community never wins, but is kept busy fighting.
In the case under study, one such labyrinthine, endless process turned out to be a useful ingredient in another case of what appeared to be state power wielding, that of obtaining the right of way for President Peña Nieto's project, the interurban train Mexico-Toluca. What had been a normal state of affairs in Acopilco, keeping an agrarian trial going for decades in the tribunals, suddenly proved to have disastrous results. Peña Nieto took office as president of Mexico in December 2012. In July 2013, he launched three megaprojects of rapid trains in the Investment program of the Ministry of Transport (SCT, 2013). The interurban train covering 57 kilometres from Mexico City to the capital of his home state, Toluca, was among those. The plan was to build it in his six-year presidential period (sexenio), and it turned out to be the only one to be executed, though not completed. The construction initiated in July 2014 (MXQ Noticias, 2014). Apparently, the president's prestige was at stake; firstly, to be able to execute the project, secondly, to reach completion within his term. The hurry was spectacular; for instance, there was no study of the environmental impact (MIA) of the train. The Ministry of Transport posted a superficial document concerning the first stage of the project, located in the neighbouring State of Mexico (SCT, 2014). However, for Acopilco, which is located in Mexico City, there was nothing. The community's main water sources, the manantiales, were at risk, which caused serious preoccupation among the villagers.
Furthermore, the enormous amount of resources mobilized in such a megaproject appears to have worked as a huge honeypot, creating something similar to Nuijten's concept of the force field (2003, pp. 12-15), which forms around huge resources. In this case, public officials apparently offered to iron out obstacles for the train project, in return for other benefits. Some agrarian institutions and juridical specificities play an important part in my study, which is why I detail them here. The Tribunal Agrario (Agrarian Tribunal) and the Procuraduría Agraria (Attorney General's Office for Agrarian Affairs) were created in 1992. They are not organized under the judiciary power, but under the Ministry of Agriculture and Territorial Development (SEDATU), according to the Legal Agrarian Framework (Presidente Salinas de Gortari, 1992). From this document and from the Verdict we surmise that there are cases of complaints, where the federal courts under the judiciary function as superior instances for the Agrarian Tribunal. These cases go to one of the Federal District Courts in Administrative Matters. 9 The Verdict refers to such procedures several times in the Acopilco case, for instance in 2014, when the Federal District Court (Juzgado 7° de distrito) instructed the Agrarian Tribunal to have Acopil-co's entry in the Agrarian Register deleted. This structure adds to the labyrinthine quality of agrarian court cases, combined with the different types of legal resources at the disposal for contestation of a sentence or decision: amparo (protection), inconformidad (unconformity), revision (review), queja (complaint), and impugnación (impugnment).
The Agrarian Law regulates the practices of agrarian communities: the Commissariat should be elected every three years, assemblies must be held at least twice a year. There are exact regulations of who can convene an assembly, and requirements for assemblies to be valid. According to these regulations, not one of the assemblies that Gabriel organized to attain legitimacy was valid.

The case story
During fieldwork in Acopilco in 2015, I attended an agrarian assembly (1 February 2015), which bore directly on the present study. A man who was considered an impostor in the community had convened the assembly. This was an important reason why only 179 comuneros (out of 2,344) attended it. However, this man, Gabriel, 10 claimed to be their representative. In the village, and in the assembly, I now heard people talking about their legal document, the presidential resolution of 1992, using a new term: insubsistente (unsubstantiated). The meaning of this was that Acopilco had lost its legal recognition as an agrarian community. In addition, rumour had it that the Commissariat, which the comuneros had elected, no longer had the legal personality to represent them. Gabriel, who now claimed to do so, explained: "the law says, when a presidential resolution becomes insubsistente, the last person to represent the community before the presidential resolution should represent it." 11 He never said to which law he was referring. He was the convener, moderator, and main speaker at this meeting, and the main topic concerned the legality of the assembly and his own status, and the pending construction of the interurban train under their aquifers. He also asserted that the rumour about him supporting the train project was not true. When I was back in the community in October 2016, I heard about the Verdict that the Agrarian Tribunal had issued, recognizing Acopilco's agrarian rights. However, Gabriel impugned the Verdict, thus it was not executable, which is the case even today (August 2019).
The story of Acopilco's dispossession starts innocently enough as an agrarian trial, but at one point turns into a process of grand corruption, with the participation of a series of public institutions. Since the sequence of these events is of the utmost importance for grasping the link between the agrarian trial and the project of the interurban train, I present the story strictly according to chronology. The Verdict concerned an agrarian dispute with a neighbouring community, Atlapulco, which claimed to be the owner of some land (543 hectares) that had been included in the territory recognized in the presidential resolution in favour of Acopilco in 1992. The dispute started in 2000 with Atlapulco's complaint to the Agrarian Tribunal, which then meandered through various types and levels of tribunals without any consequence for any of the parties. In February 2009, the Agrarian Tribunal acted on instructions from a Federal District Court, 12 declaring insubsistente Acopilco's resolution from 1992, with the explicit purpose to grant the complaining party, the community of Atlapulco, the right to be heard concerning the borders between the two communities. 13 The Tribunal initiated the process to emit a new recognition and entitlement of communal land to Acopilco. Both Atlapulco and Acopilco had their teams of experts, one after the other presenting measurements and maps indicating the border between the two communities, but without reaching an agreement.
In January 2013, a person from Atlapulco handed in un escrito (an unspecified document) to the Common Office of Correspondence of the Federal District Court, and a not specified conversation took place with the archive manager (Oficio 3974). 14 In July 2013, the president launched the interurban train (SCT, 2013). Then, in October 2013, the Federal District Court registered the document that Atlapulco had handed in in January as a complaint on behalf of that community. Finally, on 25 March 2014, the Federal District Court decided that the Agrarian Tribunal must declare Acopilco's register insubsistente, and order the Agrarian Register to have it deleted within 24 hours (Oficio 3974). The Verdict referred to this instruction from the Federal District Court, confirming that the Tribunal had carried out this order. Then, in May 2014, the Tribunal sent the case over to the secretary, who would formulate Acopilco's new entitlement.
In the first months of 2014 engineers from the train project carried out technical tests on Acopilco's territory, where the tunnel was going to be constructed, perforating the ground and hitting a vein of water that flowed in great quantities from the ground (Comunero C, 2015). In this way, the tunnel became a concrete issue in the community, as well as its connection to its water supply. The construction of the interurban train started in July 2014, from both ends, thus the tunnel on Acopilco's territory was at some distance from both sides, the whole stretch being 57 kilometres. Nevertheless, every tract of land where the construction would pass had an owner, and according to Officer B (2018), the Ministry of Transport contracted companies to investigate who the owners were, and contact them. When I inquired which company had been in charge of clearing the rights for the tunnel in Acopilco, neither he nor Officer C would provide this information.
The tunnel, which concerns us here, was located in Cuajimalpa Delegation, governed by a Delegation Chief, who was elected in 2012. He had employed Gabriel in the area of Natural Resources and Protected Areas in 2012, and then promoted him to executive director of this area on 1 January 2015. Documents from the Delegation confirm this. 15 Gabriel's promotion occurred just one month before he organized the 1 February 2015 agrarian assembly I observed, and on 5 February, the Agrarian Tribunal designated him as the representative of Acopilco community. The Verdict stated, "besides, the representation of the de facto community was recognized in favour of Gabriel", without giving any reason for this choice. Officer A (2018) in the Agrarian Tribunal likewise would not provide any explanation.
The following month, March 2015, the Ministry of Transport started talks with Gabriel concerning the interurban train. I asked Officer B (2018) in the ministry how they found out with whom to negotiate access to this tract of land. He told me they would normally check with the Agrarian Register, but in this case, they had found it in the Diario Oficial de la Federación, where the Verdict was published. However, this document was not published until August 2016, and even the Agrarian Register in its institutional capacity could not have had him on file, since the community had been deleted. Thus, the information must somehow have flown through the power network.
Gabriel himself must have considered his position lacking in legality, since he convened an assembly in Acopilco again, on 23 August 2015, to have his representation ratified, although he was not entitled to convene an assembly unless he already had the legal personality to do so. During this assembly, he also obtained acceptance for the location of the rails, and an authorization of himself to sign an agreement to transmit the possessory rights of the communal land needed for the interurban train. To sell the land would have been illegal, according to the Agrarian Law, 16 but possessory rights do not imply selling. Later on, some comuneros from Acopilco sent a letter to the Ministry of Transport inquiring about the credentials of Gabriel, with whom they had negotiated. 17 The legal department of the Ministry referred to an agreement signed 13 February 2015 by the magistrate and the administrative director (secretario de acuerdos) in the Agrarian Tribunal, conferring the legal personality on Gabriel. 18 The comuneros also inquired what investigations the Ministry had carried out about Gabriel's legal personality. The reply was that no investigation was necessary. He had presented a document dated 17 February 2015, stating that he was "proprietor of the representation of the community of communal goods of San Lorenzo Acopilco" (author's translation), 19 endorsed by the Agrarian Tribunal and the Attorney General's Office for Agrarian Affairs. 20 The Ministry of Transport paid Gabriel 47,1 million pesos for possessory rights to 26 hectares of land destined to the construction and operation of the interurban train on 28 August 2015. 21 At this point, Gabriel was still a public servant under the Delegation Chief, an employment that ended on 31 October 2015. 22 The legal department of the Ministry of Transport emphasized that the payment to Gabriel took place after the 23 August assembly. Considering that the payment was made 5 days after the assembly, it is likely that the amount had been negotiated before the assembly. This would make the 103 comuneros present inclined to vote in favour of any proposition at the prospect of personally getting their share of the millions he received. 23 I do not have a first-hand account of how much each participant at the assembly obtained; rumour sug-gests 20,000 pesos. This would be about the amount if all 2,344 comuneros had had their share.
Interestingly, the Ministry of Transport paid Gabriel by a personal check, despite instructions in the Agrarian Law to place the money in the community's account in the corresponding trust fund, FIFONAFE. 24 For Gabriel, the millions of pesos to his name gave more options than if he had been a legally elected Commissioner, with the money in the trust fund. Part of this money probably was invested in a group of thugs, which Gabriel led against protests and even information in the village. The first public appearance of this group occurred on 24 April 2016, assaulting the comuneros who addressed a public outdoor assembly of some 600 villagers, concerning the interurban train (Comunero B, 2016;Comunero D, 2016). The attack was also reported in several news media (Samaniego, 2016). Gabriel also received an additional 14,7 million pesos from the Ministry in December 2016, amounting to a total of 61,8 million pesos. 25 Local gossip calculated that Gabriel had kept 33,7 million pesos to himself. 26 When the Verdict came in August 2016, Gabriel impugned it, allegedly due to the community's loss of 31 hectares of land, which was the result of the trial in favour of Atlapulco. Normally, the execution of the Verdict would have implied elections of a new Commissariat, organized by the superior agrarian authority. However, with the impugnment of the Verdict, Gabriel kept his position. The tunnel at any rate was under construction, and the money paid. Exactly who played what part, in return for what reciprocation, is not obvious, but it is clear that the following central actors were involved: the President through the Ministry of Transport (various departments), the Federal District Court in Administrative Matters, the Agrarian Tribunal, the Agrarian Register, the Attorney General's Office for Agrarian Affairs, the Delegation Chief, and Gabriel. This basically comprises every government branch except the armed forces and the legislative power.
According to the Agrarian Law, 27 the Ministry of Transport could have chosen to expropriate the necessary land for transportation purposes from Acopilco. Officer B (2018) explained that they often used this procedure in agrarian communities, since the Agrarian Law prohibits the sale of communal land. Instead, in the Acopilco case, two central actions took place: the community's rights were deleted, and Gabriel was designated as the community's representative. The first action was arranged by the Federal District Court in Administrative Matters in March 2014 by transforming an unspecified document from Atlapulco in January 2013 into a complaint, on which it made the decision in March 2014, to have Acopilco's rights deleted (Oficio 3974). Exactly who approached the Federal District Court to request this outcome is by no means clear. Considering that on this occasion the task was only to delete rights, it would have been sufficient for someone in, or related to, the Ministry of Transport to inquire about information from the Agrarian Tribunal or the Federal District Court if there was any useful case pending. Alternatively, the other way around, from either court instance someone might have informed contacts in the Ministry of Transport about such possibilities. What is clear is that some actors had the required knowledge about the needs of the Ministry of Transport, and the possibility that the pending agrarian case constituted, combined with the necessary network contacts to inform the interested parties at higher levels about this possibility.
The second part is not so straightforward either. Gabriel must have known about the project of the interurban train, due to his position in Natural Resources in the Delegation. As a comunero, he would also know about the insubsistencia of the rights of Acopilco. He could have taken the initiative to offer his services to give access to the necessary Acopilco land for the interurban train project. The Delegation Chief, who was Gabriel's boss, likewise would possess this knowledge. He had obtained his position by bringing an extensive clientelist network of voters from the PRD (Partido de la Revolución Democrática) to the PRI (Partido Revolucionario Institucional) months before the election in 2012 (Hagene, 2015). Thus he would be interested both in demonstrating his abilities and loyalty to the PRI, while also possessing the necessary network capacity and connections to pull off this stunt. Again, persons within the of Ministry of Transport, or in one of the companies hired to investigate the issues of right of way, could have used their networks to localize a suitable person to play the part as representative of the community. What is clear, is that the Verdict offers no reason whatsoever why he was chosen for this position, despite Gabriel's claims in the 1 February 2015 assembly. The Agrarian Tribunal designated him 5 February 2015.

Conclusions
The exemplary case of grand corruption explored in this text dispossessed an agrarian community of its property rights and gave right of way to President Peña Nieto's interurban train. The case narrative demonstrates the importance of differentiating between grand and petty corruption. The long-term, complex process comprising a network across multiple institutions and actors is strikingly different from a simple bribe to a public official, and should fall in a different category, even if they both could be described as "the abuse of public power for private gain", as quoted by Gledhill (2004, p. 159 ). While the processes of the two forms of corruption are different, so certainly are the effects; grand corruption in this case involves public officers at all levels and in many institutions, and produces what appears as state effects of an extremely serious nature. Issues of legitimacy are embedded in this debate, since people tend to see grand corruption as illegitimate. Yet, as my case shows, if practices of grand corruption succeed in standing out as state effects, this imbues them with legitimacy.
I follow Monsivais and del Río (2013) concerning the concept of neopatrimonialism in considering this as a subtype of corruption, and not a regime type, which was the denotation originally given by Eisenstadt (1973). Even if actors in the network capture state institutions and resources for their own benefit in some cases, on other occasions the officials fulfil their formal functions. Following Monsivais and del Río (2013, pp. 45-47), if the state is too weak to function, it does not qualify as neopatrimonial, but as failed. Furthermore, since I differentiate between grand and petty corruption, I specify that neopatrimonialism is a subtype of grand corruption. My findings also indicate a need for some further modifications to the definition these authors give of the concept. They underline the hierarchical structure of the networks, to which I add that there may also be a scope for agency from below. Monsivais and del Río categorize groups that engage in neopatrimonial corruption as "delimited and concrete groups." In my case, the actors and beneficiaries are not delimited and concrete groups, but dynamic and ad hoc, testifying to a lifeworld of networking and a remarkable capacity to accumulate social capital.
Concerning the issue whether grand corruption is controlled from above, my case indicates that the network participants must rather have experienced acceptance, confidence and protection. The numerous network participants apparently trusted impunity to protect them. The initiation of the process took place as a forming of force fields (Nuijten, 2003) emerging around resources that became available.
I thus propose the following descriptive definition: neopatrimonialism denotes a form of grand corruption, describing a political practice, which appropriates public funds and institutions for the benefit of particular actors. It operates across institutions, in personal networks, hierarchical, but with a scope for agency also from below, and it should include actors with formal positions of authority who can undertake activities in tension or conflict with the legal order. This dynamic definition is potentially useful in other corruption studies, as one of several possible forms of grand corruption.
In his campaign in 2018, the now ruling president, AMLO (López Obrador), made a plea to combat corruption. On the basis of my case study, what are his chances of success? Following my descriptive definition of neopatrimonial corruption, several aspects point in the same direction: corruption networks form spontaneously around public resources, benefitting actors at all levels, across institutions, without being initiated or controlled from above, allowing agency even from below. Large parts of the population share a lifeworld of networking, which may qualify for participation in these processes. To the extent that my case is a common one, my findings indicate that the scope for a president to combat corruption may be limited, especially when it comes to neopatrimonial corruption. Impunity, according to Morris (2000, pp. 230-231), is the sister of corruption. Thus, fighting impunity would constitute a promising start. * * * Turid Hagene is professor emerita at Oslo Metropolitan University, Department of international studies and interpretation. She has carried out ethnographic research in a pueblo originario on the outskirts of Mexico City between 2003 and 2018, researching a series of different topics like democracy and participation, electoral practices and political clientelism, rumour and gossip, democracy and the environment, lenguaje ritual.