POLITICS AND INDIGENOUS VICTIMIZATION: THE CASE OF BRAZIL

There is a dearth of criminological scholarship on how the political persuasions of governments affect Indigenous people as it relates to human rights and environmental consequences, whether positive or negative, for Indigenous peoples. To address this gap, we develop a comparative instrumental case study of the policies concerning Indigenous peoples implemented during two political periods in Brazil: the administrations of presidents Silva (2003–2010) and Rousseff (2011–2016) and the administrations of Temer (2016–2018) and Bolsonaro (2019–). We explore the consequences for Indigenous peoples of these leftist and the right-wing governments. We argue that governments of both political leanings victimize Indigenous populations, with leftist governments using structural violence and right-wing governments engaging additionally in symbolic and direct violence.


Introduction
In November 2011, jagunços (hitmen) hired by a group of land owners from the state of Mato Grosso do Sul and employees of a local private security company attacked a Guarani-Kaiowá settlement, 1 shot one of its residents, Cacique Nísio Gomes, and kidnapped his wife and two children. Events like this, confirmed by governmental (Comissão de Direitos Humanos e Minorias [CDHM] 2012; 2016) and non-governmental institutions (Conselho Indigenista Missionário [CIMI] 2014) alike, expose the all too common, yet dramatic, situation for Brazilian Indigenous peoples. As the author of a January 2012, Le Monde Diplomatique Brazil report asserted, 'the power of the large land holders of the region is so strong that neither the presence of federal government representatives nor the National Security Force are capable of intimidating them' (Zinet et al. 2012).
Throughout the colonial period and up to today, Indigenous peoples in Brazil have been systematically victimized (Coates 2004) either directly by governments or by other actors who exploit governmental inaction. Bluntly put, governments from the extreme right to the extreme left have been responsible for violating the rights of Indigenous peoples either directly or by omission and commission. In Brazil, as in the rest of Latin America, this phenomenon combines racist ideologies that sustain internal colonialism (Mondaca 2017) and the normal functioning of the capitalist treadmill of production that seeks expansion in all regions regardless of the cost in human lives (Lynch et al. 2018). Independent researchers Stuckert and Gentilli (2019), as well as public (CDHM 2017;MPF 2018) 2 and non-governmental institutions (CIMI 2019), 3 have documented widespread Indigenous victimization perpetrated or facilitated by every Brazilian government. Furthermore, Brazil is the most lethal country for environmental activists, most of whom are Indigenous: more than 150 environmental activists were murdered in Brazil between 2012 and 2015 (Global Witness 2016) and, in 2016 alone, 65 human rights activists were murdered. This is a trend that continues to grow (CDHM 2016: 33).
Nevertheless, the study of Indigenous victimization lacks a nuanced exploration of whether the violence directed at Indigenous peoples correlates with the political leanings of governments. To fill this gap, we present the findings of a comparative and instrumental case study of the four most recent Brazilian governments-those led by Lula, Dilma, Temer and Bolsonaro-covering the period January 1, 2003 to December 2019 (the time of the writing of this article). Our main argument is that, while leftist governments primarily target Indigenous peoples through structuralinstitutional violence, right-wing governments, in addition, also employ symbolic and direct violence.
Our article is divided into five sections. In the first segment, we highlight the dearth of criminological information about the impact that diverse political governmental persuasions have on Indigenous populations. In the second section, we lay out our methodology, which is designed to address the challenges of researching in the global South. In the third section, we develop the theoretical framework of our analysis, which is inspired by the notions of direct, symbolic and structural violence, first proposed by Johan Galtung (1969;1971;1990) and developed by Rosa del Olmo (1975). In this same section, we conceptualize our understanding of right-wing and leftist governments. In the fourth section, we describe the Brazilian political context of the past two decades, drawing attention to the victimization of Indigenous peoples by both leftist and rightwing governments. In the fifth section, we discuss the symbiotic nature of all types of violence and the meaning that our study has for both Indigenous criminology and the wider harm perspective. 2 The CDHM (Human Rights and Minorities Commission) of the Brazilian Federal Parliament (Chamber of Deputies) publishes annual and special annual reports, which contain specific chapters on Indigenous issues. These reports, beginning in 1999, are available at https://www2.camara.leg.br/atividade-legislativa/comissoes/comissoes-permanentes/cdhm/ documentos/relatorios-de-atividades.
The 6ª

Background
Most criminologists who focus on Indigenous groups carry out their research within a framework of legal definitions of crime. They thus (1) ignore harmful legal practices and (2) restrict their scholarship to a study of these communities to the extent that they come to the attention of the official criminal system, e.g. how race and ethnicity affect parole of Indigenous criminals and recidivism of Indigenous people. In this regard, criminology of Indigenous peoples has not adequately taken into account the dynamics of government-Indigenous intersections as a fertile and necessary field of criminological study. Our study engages this dynamic by focussing on how the political persuasions of Brazilian governments have played into the victimization of Indigenous communities.
In Brazil, local endogenous studies of crime and harms vis-à-vis Indigenous peoples follow a similar pattern. Brazilian scholars have mainly focussed on exploring the criminalization of Indigenous communities (de Silva 2013;Silva 2016), the processes of imprisonment and incarceration (Flores 2008), cultural conflicts between Indigenous and non-Indigenous Brazilians (Neves 2011;Wagner 2018) and the legal discussion about the procedures of attributing responsibility to an Indigenous perpetrator upon the commission of a crime (Dias 2014;Stefanini 2015;Santos 2018).
Despite this trend, criminologists around the world have begun to critique the approach to Indigenous issues that is limited to legal facets. They point to the neglect of other phenomena and histories that negatively impact the lives of Indigenous communities, irrespective of the legality of what took place (Cunneen and Rowe 2014;Martin 2014). For instance, for more than two decades now, scholars have been exploring Indigenous justice systems as an alternative form of justice for the broader society (Blagg 1997;Braithwaite 2002). More recently, criminologists in the fields of green criminology, state crime and the crimes of the powerful have begun to pay attention to the experiences of Indigenous peoples beyond their involvement in criminal systems, applying a harm perspective (see below). Goyes (2019) advocates for a southern green criminology-one that relies heavily on the knowledge and experiences of Indigenous peoples-that would approach environmental harms in the global South from a southern vantage point. This new direction takes into account equally the environment/Indigenous people and the state/corporations, the former as overrepresented victims of the actions of the latter (Boekhout van Solinge 2010; Crook et al. 2018;Lynch et al. 2018). In line with this recent trend in criminology, we move beyond a legal definition of crime in our use of successive Brazilian governments to examine the dynamic of Indigenous/environmental-state/corporation intersections. We provide the details of the research design in the next section.

Methods
The global distribution of cultural-scientific capital depends on the capacity to gather representative data and, thus, generate generalizable theses (Adams and Lawrence 2019). In the pursuit of generalizability, the academic community favours probability sampling methods-a set of techniques that require extensive economic, labour and institutional resources. Meanwhile, UNESCO has documented the 'huge disparities in research capacities across countries' (Sané 2010: iv). This disparity is marked by the ability of core countries to allocate economic capital and labour to research and the inability of peripheral countries to generate reliable, official, government-produced statistics. As a result, researchers based in economically rich, core countries monopolize the scientific capital of the world. The demand for scholarly sanctioned methods in research reinforces and widens global knowledge gaps (Connell 2007;Carrington et al. 2019).
In the development of the research behind this article, it was particularly the lack of official statistics and data on violence against Indigenous communities that represented the main challenge for us. However, considering the urgency of the topic and the snail's pace at which the transformation of the global structures that sustain global disparities in research is taking place, we applied innovative, yet reliable, procedures to gather, systematize, verify and codify the data for our project, beginning with its methodological design. We developed an instrumental, collective case study in the form of a comparative time-based research study. In other words, we used official and unofficial statistics, reports and other types of documentation of violence against Indigenous peoples in Brazil from 2003 to 2019 to evaluate the role that the political leanings of successive governments played in such violence. Stake (2005) defined a case study as the selection and exploration of a physically localized phenomenon. With this, he meant that the case was historically and materially contextualized rather than being purely theoretical. Despite being localized, the findings of case studies can be generalized without compromising the accuracy of the contextual factors. Stake (1999) classified case studies into three types: (1) intrinsic case studies where the case itself is important, (2) instrumental case studies where the case serves to broaden knowledge about a wider issue beyond the case itself and (3) collective case studies where conclusions are reached through a comparison of several cases. Our design is primarily instrumental, but we included a comparative element because our goal was to understand differences in how leftist governments and right-wing governments promulgate violence against Indigenous peoples. While we chose Brazil as the location of our study, we argue that, because of its shared history of colonization and Indigenous dispossession with other southern states (see Goyes and Aas 2019), our findings can be instrumentally useful in understanding the situation of Indigenous communities in other locations. In general, the selection of case studies does not follow logic statistical sampling (Adams and Lawrence 2019) but, rather, is based on theoretical sampling (Flick 2005). This means that selection is based on the content of the cases rather than on their representativeness. In our study, we used maximum variation theoretical sampling, as our goal was to look at the policies and practices concerning the Indigenous populations of a country of two radically diverging types of governments.
The first step in our research was to use both official and unofficial statistics to identify prevalent forms of violence used against Indigenous peoples, and their fluctuations, during our sample period. The statistics of the NGO CIMI (Missionary Indigenous Council) were the basis of our study. 4 These statistics were the initial indication that defined the path of our posterior qualitative analysis. Second, we complemented these statistics with a broad systematic qualitative analysis of court rulings, official and unofficial reports and media accounts. At this stage, we connected the statistical trends in violence against Indigenous communities with more in-depth and detailed qualitative accounts of the violent events. Third, we connected the violent events with developments in national Brazilian politics, in the form of policy development and political statements, the latter because the symbolic plane is fundamental in this study.
For the development of these three steps, we relied heavily on secondary sources, particularly the reports issued by local NGOs and media. We were forced to rely on unofficial reports to map violence against Indigenous peoples because the Brazilian government has not systematized its data, produced many official reports or made available the few that exist. While official government reports published in the last decade are of higher quality-exemplary of this are the reports published by the Fundação Nacional do Índio (National Indigenous Foundation [FUNAI])-the information they contain is still insufficient. Furthermore, since the Temer government, myriad government agencies have suspended the collection and publication of data. All this to say that finding reliable, official, government-produced statistics for the production of knowledge in the global South is challenging.
Using unofficial, second-hand material is far from ideal, so in our study we implemented a set of techniques to check and augment the reliability of the unofficial material we used, increase its depth and constitute it as a homogenous, systematic database rather than a patchwork of information. These strategies allowed us to secure reliable data despite the challenges existent in the field. The main strategies we used to check the reliability of unofficial statistics were the following.

Identifying the overlap between official and unofficial sources
We sought to evaluate the reliability of our data based on the variations in the figures offered by governmental and independent institutions. Because official reports from the FUNAI lack data on violence, when seeking official information, we relied on the research developed by the CDHM (Human Rights and Minorities Commission) of the Federal Chamber, which is part of the Brazilian legislative power. Issues concerning Indigenous peoples have been extensively explored in their annual reports. Besides providing information for our analysis, reports from the CDHM exhibit a significant overlap with the data produced by the CIMI. Furthermore, the CDHM and the CIMI produced a joint study of the massacre of the Guarani-Kaiowá (CDHM 2015-16). Another relevant official source was the 6ª Câmara de Coordenação e Revisão (Sixth Chamber of Coordination and Review) of the Federal Public Ministry. The 6ª Câmara promotes the inspection, control and defence of the Indigenous population. This chamber validated CIMI's 'Map of Violence against Indigenous Peoples' report as an official document. The endorsement of CIMI's outputs by state entities and the significant overlap between official and CIMI information is a first indicator of our data's reliability and validity.

Analysis of CIMI's methodological criteria
We evaluated the soundness of the methods employed by CIMI to produce statistics. CIMI creates its database based on the general number of deaths of Indigenous peoples reported by two official sources: the Secretaria Especial de Saúde Indígena (Special Secretary of Indigenous Health) and the Secretarias Estaduais de Saúde (State Health Secretariat). CIMI uses this information only as an estimate because it is not restricted to violent deaths. Subsequently, CIMI proceeds to compare the data with information regarding violent situations provided by the press and reports collected by its on-thefield technical teams. CIMI deploys more than 170 members in 11 regional agencies throughout Brazil for this task. The soundness of this methodology is a second indication of the reliability of CIMI's data.

Mapping the criminal investigation of homicides reported by CIMI
We double validated the homicide cases reported by CIMI. First, we verified violent events through a document analysis of the criminal cases. Second, we compared the events with media reports. Of the many cases reported by CIMI, 91 per cent were confirmed by sources other than CIMI and 63 per cent were confirmed by the criminal justice system. These high figures further attest to the reliability of CIMI's documentation.
In addition to the statistics from CIMI, CDHM and the 6ª Câmara, we built a database using court rulings, police inquests, media reports and relevant policy developments in our sample period. For the category of news report, we always secured additional sources to ensure the reliability of the report. Using this database, we developed a critical qualitative analysis to identify the motives behind and dynamics of violence committed against Indigenous peoples, in connection with the timeline of Brazil's governance.
The fourth step was the qualitative analysis of the data, which we conducted by employing grounded theory, a methodology where codes emerge from data rather than codes being derived from a predefined theoretical framework (Strauss and Corbin 2012). The codes that emerged during the open coding phase regarding the dynamics, forms and motives of violence against Indigenous peoples closely paralleled Galtung's typology of violence. Thus, we chose that typology for the step of axial coding (linking categories and subcategories) and to structure the article. Fifth, in the process of writing, we chose to describe only emblematic cases, primarily due to space constraints. They are, nonetheless, representative of the broader set of data of this project. We now explain the contents of the theoretical framework that guided the axial coding.
Theory Galtung (1971) proposed three types of violence: direct, cultural and structural. His intention in proposing this pyramid of violences was to move away from seeing violence as individualized and point towards systemic and structural social arrangements as the main drivers of violence (Goyes 2018). For Galtung (1969: 170), most social sciences and state policies were overfocussed on direct violence 'where there is an actor that commits the violence as personal or direct'. According to Galtung, this focus ignores the two types of violence that are at the root of direct violence, cultural and structural violence. Cultural violence stems from 'those aspects of culture, the symbolic sphere of our existence . . . that can be used to justify or legitimize direct or structural violence' (Galtung 1990: 291). This is not unlike Bourdieu's (2001: 5) symbolic violence, which is 'exerted for the most part through the purely symbolic channels of communication and cognition (more precisely, misrecognition), recognition, or even feeling'. Galtung uses Carmichael's (1968: 151) definition of structure as 'the overall operation of established and respected forces in the society' to form his definition of structural violence: the harms that are subtle and not clearly the responsibility of any specific institution but are the result of the abstract and broad combination of official rules that govern social life.
Galtung's typology helped shape the work of the Latin American Critical Criminology Group in the 1970s. Del Olmo (1975), a prominent member, noted that Latin American violence differed from violence in developed 5 countries where most criminological theories were and are still produced. Drawing inspiration from Galtung, del Olmo categorized violence into four types: individual (or direct), institutional (materialized in legal repression, mainly, but not exclusively, by omission), structural and revolutionary (guerrilla violence). For del Olmo, the difference between southern and northern countries was that violence in the south was predominantly structural, from which all other forms of violence arose. Despite her observations about the distinctiveness of southern violence, del Olmo noted that only people who participated in individual or guerrilla violence were criminalized, whereas perpetrators of structural and institutional violence were not penalized (for more, see Goyes 2019).
In this article, we combine Galtung's and del Olmo's typologies to analyse the differences in governmental violence directed at Indigenous peoples, depending on the political leaning of the regime. We use Galtung's three types of violence as a matrix of analysis: direct, symbolic and structural (of which institutional is a subtype).
Including direct, symbolic, institutional and structural-institutional violences as foundational concepts in our analysis also means embracing the harm perspective or zemiology. We distance ourselves from understanding a crime only as those social conducts defined by penal law to also include legal acts that are harmful. We draw our inspiration from Schwendinger and Schwendinger (1975) who questioned the criteria used to define what was criminal and simultaneously highlighted that most actions of the powerful were permissible by law but were, nevertheless, harmful to most of society. Pioneering scholars in the field of zemiology consequently advocate for the replacement of the notion of 'crime' for the broader one of 'social harm'. The latter encompasses all social events and conditions that affect people adversely throughout their life trajectory (Hillyard and Tombs 2004).
Finally, we broadly characterize Lula's and Dilma's governments as leftist and Temer's and Bolsonaro's as right wing. Here, we follow the seminal work of Maddox and Lilie (1984), who proposed a political spectrum that ranges from modern liberalism, to libertarianism, to conservativism, to populism. In their spectrum, modern liberalism corresponds to leftist governments, which advocate for governmental regulation of the economy to promote the welfare of society and for the free development of individuals without governmental restrictions. In contrast, conservativism corresponds to right wing. For the conservative, 'left to his own devices, man is dangerous' and, consequently, 'human action needs to be restrained, guided in accordance with correct principles', such as those provided by 'religion, natural law, right reason, and tradition'.
In this position, 'the Government's function is to interpret and enforce these correct principles through law, thus ensuring justice and civility through the use of authority' (Maddox and Lilie 1984: 16). In short, we refer to leftist governments as those that are liberal in terms of culture and individual development and that seek to regulate the economy to ensure the well-being of a society. Right-wing governments are those that seek to regulate culture and behaviour based on religion or traditional values and advocate for no governmental intervention in the market. In the next section, we describe some important actions of the four Brazilian governments that support our characterization of leftist and right-wing governments.

Leftist and Right-Wing Governments in Brazil Background
Brazil ceased being a fully democratic country in 1964 with its first military coup of the 20th century, which removed the democratically elected João Goulart from office. 6 The coup lasted until 1985, but the first democratic election only took place in 1989. The dictatorship of the coup implemented an Indigenist policy characterized by assimilationist and tutelary policies with a goal to progressively 'civilize' the Indigenous population (Friede et al. 1975; for a discussion of assimilationist policies beyond Brazil, see Cunneen and Tauri 2017). The 1988 Brazilian constitution marked a shift in this trend by recognizing Indigenous peoples as actors entitled to full rights and capable of defining their own interests (art. 232). That recognition included an acknowledgement of their right to own traditional lands as a fundamental aspect of their cultural and physical survival (art. 231, §1º). The constitution states that 'the lands traditionally inhabited by Indigenous people will be reserved for their permanent possession, including its exclusive usufruct of the richness of the land, rivers and lakes' (art. 231, §2º).

Leftist governments: Lula and Dilma
Luiz Inácio Lula da Silva (2003Silva ( -2010 was the first leftist, popularly elected president of Brazil after re-democratization. In 2011, Dilma Rousseff, former Minister in Chief under Lula, became president. She held the position until 2016, when she was impeached following corruption charges. During Lula's and Rousseff's presidencies, almost 30,000,000 Brazilians were lifted out of poverty following an economic and political program aimed at reducing the social divide within the country and strengthening Latin American bonds. In addition, beginning with the presidency of Lula, annual deforestation rates declined by more than 80 per cent between 2004 and 2012 (see Nevertheless, during their governments, environmentally risky investments continued to play a major role in the distribution of public resources and public policy. The agribusiness model was advanced under pressure from landowners, the so-called 'rural lobby'. Especially threatening was the construction of hydroelectric dams, mining projects and the installation of transmission lines on Indigenous land, initiated without prior consultation with the Indigenous groups affected (United Nations 2016). For instance, Norte Energía, the corporation responsible for the construction of a hydroelectric dam in Belo Monte (initiated by Rousseff), did not consult with affected Indigenous people as required nor invest in the mandatory safety measures. For none of this was the company penalized (Aroeira 2019). Furthermore, during Rousseff's government, only 3 per cent of the 24,000 dams were monitored by the relevant government agencies (Justicia Global 2019).
Rousseff's government had a very low rate of land allocation for Indigenous communities (Justicia Global 2019). Two examples illustrate this. The Xukuru from Oruba initiated their land claim in 1989 and, even though they were granted land in 2001, subsequent administrations (including Lula's and Rousseff's) have not honoured the guarantee of non-intrusion in the area (Justicia Global 2017). In 2019, the Inter-American Commission on Human Rights (IACHR) recognized the international responsibility of the Brazilian state to ensure the rights of the Xukuru to collective property ownership and judicial protection. Likewise, the quilombolas (Afro-Brazilians) have demanded that they be allocated land and their communities protected but have received no response from the government. To further their case, they have filed a petition with the IACHR (Justicia Global 2018). Both cases dealt with by the IACHR challenged leftist Brazilian governments' actions towards Indigenous communities.
In short, during the leftist governments, victimization of Indigenous peoples mainly took the form of institutional-structural violence, i.e. governmental omission. This omission can be seen in two areas: (1) governmental agencies responsible for establishing the boundaries of Indigenous territories and their protection from intrusions lagged in their duties and (2) militarized structures inherited from previous dictatorships were not transformed into human rights institutions guaranteeing the fundamental rights of Indigenous communities, as promised. These omissions of Brazil's leftist governments total 13 years of neglecting to act on the promises that the Labour Party made to Indigenous peoples.

Right-wing governments: Temer and Bolsonaro
Michel Temer, who became president upon the impeachment of Rousseff, carried out labour reform, pushing for job cuts within the civil service and the privatization of public companies. The administration of Temer, which, in a strict sense, was transitionary, implemented a federal law by which lumber companies were granted timber rights to the land (Human Rights Watch [HRW] 2019). Temer, however, did not allocate any land to Indigenous peoples. Up until Bolsonaro's inauguration in 2019, he had been the most detrimental president in terms of Indigenous policies since the return of democracy (CIMI 2018a;2018c). Between August 2017 and July 2018, the destruction of the Amazon forest hit an unprecedented level-1,185,000,000 trees were destroyed. An area equal to two soccer fields was deforested every minute (Dantas 2018;Exame 2018;Greenpeace 2019). These reports are confirmed by official sources that rely on the live updates of the DETER monitoring tool (INPE 2020).
Security during Temer's tenure was equally problematic. The police did not intervene when hitmen, hired by landowners, attacked the community of Akroá-Gamella in Maranhão. Two residents were murdered and two had their hands cut off (Felipe 2019; Greenpeace 2019). In total, 68 people, most of whom were Indigenous, were killed in land conflicts in 2017, which is the highest number since 2013 (CPT 2018; HRW 2019). Exacerbating this was Temer's removal of the ministerial status of the Secretaria de Direitos Humanos (Human Rights Secretariat), weakening human rights protection as a matter of governmental policy (Global Witness 2016).
CIMI's report 'Violence against Indigenous Peoples in Brazil', maps the rights violations perpetuated against Brazil's Indigenous populations during Temer's presidency. CIMI categorizes the violence as either institutional or interpersonal. The report draws attention to institutional violence (omissions of the government) in education and health, resulting in malnutrition, alcoholism, child mortality and general mortality caused by preventable diseases like malaria. According to 2017 data, there were 42 cases of negligence in health (causing eight deaths) and 41 in education. Similarly, the CDHM (2017) highlighted the significant cuts in social policies. In total, CIMI reported 142 cases of violence as the result of governmental omission. That same year, CIMI reported 702 deaths of children under five years and 128 cases of suicide (CIMI 2018a: 163). CIMI also highlighted interpersonal and collective violence in 2017: (1) 96 crimes against the environment, mostly invasions into Indigenous lands, illegal exploitation and destruction of natural resources, (2) 20 conflicts over land rights, (3) 847 cases of omission in the allocation of land to Indigenous communities, (4) 68 murders, 8 27 attempted murders, 12 injuries, 32 threats, 16 sexual violations, 18 cases of racism and 8 cases of abuse of power (CIMI 2018a: 162). Of these 95 attempted and de facto homicides, and following the procedure described above, it was clear that these crimes were mainly motivated by conflicts over land in combination with hate dynamics. 9 Jair Messias Bolsonaro became president in 2019. During his campaign and once in office, Bolsonaro also raised security as one of the most important government concerns and made it clear that he would use authoritarian measures to ensure it. He also strengthened government links with the financial/agribusiness/mining, religious and military segments of society by establishing Frente Parlamentar da Agropecuária (Agricultural Parliamentary Front), Frente Parlamentar Evangélica (Evangelical Parliamentary Front) and Bancada da Bala (Bullet Stand), respectively (Benites 2018;Elbaum 2018a;; Folha de São Paulo 2019; Isto 2019). Teresa Cristina, Bolsonaro's Minister of Agriculture, Livestock and Food Supply, clearly expressed the importance of the government alliance with the Frente Parlamentar da Agropecuária: 'You can be sure that President Bolsonaro's government appreciates and has a special bond with landowners who initially supported and believed in him' (Publica 2019). In addition, Elbaum (2018a) points out that no other Brazilian government after the dictatorial period has had as many ministers from the military as Bolsonaro. This highlights the Bolsonaro government's commitment to using force to regulate social life, a characteristic of extreme right-wing governments.
Bolsonaro's presidency has meant the beginning of a critical phase in the struggle for the fundamental rights of Indigenous Brazilians. The desperateness of the situation is evident in the explicit cultural/symbolic violence targeting Indigenous people present in Bolsonaro's rhetoric. Bolsonaro already expressed such violence as a congressman but it increased during his presidential campaign. This symbolic violence is fundamentally an act of denial: he does not recognize Indigenous people as a collective or Indigenous individuals as subjects with rights. When referring to Indigenous people, he told the members of the Rio de Janeiro Trade Association, 'They are outcasts and we must treat them like terrorists. . . . Private property is sacred. We must criminalize the actions of those marginal people as terrorism. If they intrude onto private lands, they'll get "lead" [bullets]' (Pennaforte 2018). In an interview with the TV broadcaster Bandeirantes immediately after his election in November 2018, Bolsonaro promised 'as far as I'm concerned, there will no more Indigenous land'. Political guidelines and discourses broadcast in the media and on social networks have exacerbated violence against Indigenous populations (Oliveira 2019). However, there is no consolidated data available for an academic analysis.
This persistent attacking of Indigenous peoples legitimizes specific forms of direct violence. As we explained above, symbolic violence is at the root of the other types of violence. Castilho (2018: 701) contends that 'racist discourses are not disconnected from an effective action against the rural populations. They keep happening, in the voices of diverse parliamentarians, throughout the recent political history of the country'. Bolsonaro's statements have already been concretely materialized. The media outlet Brasil de Fato (2019) has documented how Bolsonaro's discourses prompted rural farmers in the north of the country to start a movement intent on burning the Amazon forest on August 10, 2019. That movement contributed to the record high destruction of 2,500 km 2 of Amazonian jungle in a single month (BBC News 2019). As Watson (2019) reports, 'all fears were well grounded, and this racist administration is openly attacking, as never before, the Indigenous populations of Brazil, with the purpose of destroying them as communities, forcefully assimilating them, and plundering their lands'.
In addition to the symbolic violence in his rhetoric, Bolsonaro engages in direct violence against Indigenous peoples. In April 2019, he authorized the use of police force against those parading to commemorate International Indigenous Day in the Federal District (Uribe 2019). Due to numerous conflicts between the executive branch and environmentalists in 2019, media referenced an 'institutional war' against Indigenous peoples (Watson 2019;Fellet 2020). Bolsonaro also reinforced Brazil's structural violence through a range of public policy decisions. During his first week in office, he transferred the responsibility of establishing Indigenous zones from the FUNAI to the Ministry of Agriculture, thereby curtailing the expansion of Indigenous territory (G1 2019). In May 2019, he legalized the ownership of heavy weaponry, enabling rural inhabitants to defend their land from Indigenous 'threats' (Sorano 2019).
Bolsonaro's ministerial selection points to the same disregard for Indigenous rights. He appointed Ricardo Salles, an anti-environmentalist, as the Minister of Environment (Gortázar 2019) and chose the former head of the agribusiness lobby as his agricultural minister. Bolsonaro has made it clear from the beginning of his tenure that he plans to open up Indigenous land to agriculture (Bellstrom 2019). Today, the worst news is that of the destruction of Amazonia. The statistics published by the INPE (2019) show that the deforestation of the Amazon increased by 90 per cent in June 2019 and 278 per cent in July 2019 over the same months in 2018 (see also Moreno 2019). In July alone, 2,253 km 2 of vegetation was cut down, the greatest surge in deforestation rates since the INPE adopted its current tracking methodology in 2014. DETER reported that more than 4,200 km 2 of forest were chopped down in the Brazilian Amazon between January 1, when Bolsonaro took office, and July 24. That is 50 per cent more than in the first seven months of 2018 and more than double the area cleared in the same period in 2017 (Escobar 2019). Not only do the updated figures for 2020 attest to this trend (INPE 2020) but also President Bolsonaro himself confessed his opposition to the environmental agenda. In a statement made in Riyadh, at the end of 2019, in the Future Investment Initiative panel, he declared that he 'potentiated' the fires in the Amazon to the detriment of the environmental policies that had been adopted by previous governments (Amaral 2019;O Globo 2019).
In sum, Brazil went from unfulfilled promises under leftist governments to a landscape of fulfilled promises of violence under right-wing governments. The right-wing period has seen an increase in structural violence, along with notable instances of symbolic and direct violence. Based on that analysis, we argue that, while leftist governments mainly victimize Indigenous peoples through structural violence, right-wing governments also engage in symbolic and direct violence. With the backdrop of this finding, in the next section, we discuss how the diverse types of violence directed at Indigenous communities by different governments is a symbiosis of violence. Governmental violence does not disappear when a new administration takes power. On the contrary, in Brazil, it has merged with the new forms of violence brought about by subsequent governments in a spiral of violence that exacerbates the harsh conditions under which Indigenous peoples live.

Violence, Political Leanings and the Renewal of Indigenous Criminology
Drawing our inspiration from Galtung, we differentiate four types of violence that victimize Brazil's Indigenous peoples: (1) structural violence, the product of instrumental thinking that organizes and commoditizes the environment and people in a capitalist logic; (2) institutional violence, a subtype of structural violence, materialized in the actions and omissions of the criminal justice system, including its law-making, executive and judiciary branches; (3) direct violence executed by rural squads and funded by large landowners and (4) symbolic violence transmitted in the racist speeches negating the otherness of Indigenous communities.
The structural and institutional violence of Brazilian governments is expressed, first, in their perception of land as a site for the expansion of market logic. Governments have reneged on their obligation to allocate and protect Indigenous land, as outlined by the constitution, which mandates governments to protect 'the original rights over the lands that they [Indigenous groups] have traditionally occupied' (art. 231). In other words, the executive arm has failed with its administrative procedures to establish and protect Indigenous land and guarantee Indigenous access to and benefit from it. Direct violence is materialized in legal developments that enable attacks against Indigenous peoples by private actors, e.g. the creation of rural vigilante squads funded by large landowners and encouraged by the state in the legalizing of heavy weaponry. Symbolic violence is evident in the racist speeches of President Bolsonaro that negate the otherness of Indigenous communities. Brum (2013: 399) makes this clear in her denouncement of the government that understands Indigenous lands as spaces 'where there is no one, only Indians'. Denying the personhood of Indigenous people is on par with justifying the violence directed at them. It is in that sense that we understand that the diverse dimensions of the violence directed at Indigenous communities is reinforced by the negation of otherness of Indigenous people, the possibility of them being and living differently.
These diverse forms of violence directed at Indigenous people are not insular but symbiotic, and at their centre is state racism. As we explained above, one of the main sources of structural violence against Indigenous peoples under both leftist and right-wing governments has been the non-fulfilment of constitutional mandates regarding the allocation of Indigenous land. Such structural violence precipitates an intense conflict over land, as the most recent Amnesty International (2018: 92) report documents: 'The conflict over the land and the invasion of gold miners and timber traffickers in Indigenous territories will continue, resulting in episodes of violence against the latter'. The report goes on to outline the state's responsibility in this violence: 'The government and the Brazilian tribunals undermined the existent of institutional structures and national policies, thereby further slowing down the already slow process of land allocation and intensifying the conflict over land possession in Indigenous territories'.
Governmental sanction of the illegal occupation of Indigenous lands mainly by agribusiness and mining companies drives the violence that disrupts the lives, freedom and dignity of Indigenous communities. In light of this, we argue that the structural, symbolic and direct violence of Brazilian governments are interlinked. The logic of exploitation that objectifies the environment and Indigenous people guides Brazilian macroeconomic policy and, consequently, legitimizes ongoing crime, whether isolated or systematic.
By documenting the symbiosis of direct, structural and symbolic violence against Indigenous peoples, and the important role that political leanings play in the definition of the dynamics and forms of violence, we point out the fact that any study of violence involves elements of 'seeing and not seeing, and can draw attention to absence as much as presence, mediating reality and posing important questions over what lies beyond the frame' (Carrabine 2018: 632). In other words, in our study, we sought to bring new elements into the representations of violence in Indigenous criminology. Of the weaknesses of criminological approaches to Indigenous issues that we identified at the outset of the article, we focussed on the role that governmental political persuasions play in the type of violence exerted towards Indigenous populations. Throughout the article, we also used Galtung's pyramid of violence as a conceptual framework to consider the nuances and varieties of violence against Indigenous populations. We have not only contributed to overcoming the overdependence, and consequential blindness, of the subfield of Indigenous criminology on legal definitions, institutions and processes but also introduced the multiple theoretical developments in the discipline of criminology.
Our reliance on the harm perspective was pivotal to the identification of violence that escapes the radar of orthodox criminologies. Nevertheless, our study also showed that applying the principles of zemiology is not without its issues. While the harm perspective indicates that 'more adequate, effective, comprehensive, and practical responses to the sources of injustice can be derived from studying both harms and crime instead of dealing exclusively with the latter', our study shows that a key challenge in this perspective is the power relations, most clearly represented in governmental political leanings, that determine 'who has the power to define what is understood as harmful' (Goyes 2018: 62). Whereas the goal of zemiology is to 'incorporate a wider spectrum of voices to participate in this definition [of what is understood as harmful]', our study demonstrated that dynamics that have been ongoing for over five centuries have sedimented institutions and structures that drown and silence that 'wider spectrum of voices' (Goyes 2018: 62). Illustrative of that phenomenon is the manifesto of the leaders of the Guarani-Kaiowá: To whom will we denounce the violence committed against us? To which justice system in Brazil, when it is our own federal justice system that is generating the violence against us. . . . . We don't have, and never will have the possibility of a dignified and fair life, not here on the riverside nor somewhere far from here. . . . Acknowledging that historical fact, we want to die now and be buried together with our ancestors where we are. That is why we are asking the government and the federal judiciary system to, instead of decreeing our expulsion [from our land], decree our collective death and bury all of us here. We ask, once and for all, that you declare our total annihilation. . . . Declare the collective death of the Guarani and Kaiowá of Pyelito Kue/Mbarakay and bury us here. Consider that we have decided that we will not leave this place-alive or dead (Conselho Guarani-Kaiowá 2012).
A further development of an Indigenous criminology conscious of its shortcomings, beyond contributing to a better and more nuanced understanding of violence within and around Indigenous communities, has the potential of providing new depth to the harm perspective and, consequently, the entire discipline of criminology.