Risk, Vulnerability, Social Justice and the Anti-Politics Machinery in Brazil: A Brief Discussion about the Mariana Dam Disaster

by Carolina Lima

On November 5, 2015, Brazilians experienced the disaster that generated the greatest environmental impact in their history, well-known as the Mariana dam disaster. During the following 17 days, mining tailings affected riverside communities and created ghost towns along the Rio Doce basin until they finally reached the Atlantic Ocean. When I visited one of the affected districts in 2018, it was possible to see a preschool calendar where they marked the last day of classes (seen in figure 1) and how high the mud deposits got (seen in figure 2, where the objects in the lower right corner are the toilets).

From L to R – Fig. 1: Preschool Calendar in Paracatu de Baixo, Minas Gerais, Brazil (Source: Author); Fig. 2: Preschool Bathroom in Paracatu de Baixo, Minas Gerais, Brazil (Source: Author)

Despite the discussion about the people’s relationship with space, other factors like risk, vulnerability, governance and social justice gained focus in discussing what happened. This article is the result of an undergraduate course on Political Ecology in Brazil addressing environmental governance, risk and vulnerability issues in Brazil and also other peripheral countries in the Global South. The main goal here is to discuss these concepts as well as the arrangement of governance and development in the capitalist system and how it is related to risk and vulnerability conditions. Besides these topics, the article will present the lack of democratic participation of the marginalised groups in the government process due to the authoritarian nature of governments and the anti-politics machinery in the scientific field.

To illustrate these discussions, this article will focus on the case of the Mariana mining dam in Fundão in the Brazilian state of Minas Gerais, that collapsed in November 2015. The disaster caused the death of 19 people and caused significant damage, with pollutants being spread over 668 km along the river. This disaster, after five years, still reverberates in Brazil due to the irreparable damage to all populations along the river polluted by mining waste. This article will briefly discuss the conditions of risk production in these spaces and the relationship between environmental justice and vulnerability. In the end, it will present some reflections and questions related to the topic as a contribution to the research field.

Disasters punish those who are not prepared to receive them, or unable to overcome adversity (Valencio, 2010). The action of removing people from their homes is the compulsory displacement of a group, but it can be seen as an expulsion or landscape intervention. According to Valencio, in Brazil, power relations allow the naturalisation of ills, blaming “natural disasters” for the failure of an inclusive urban model. By territorialising the poorest areas as “risk areas”, subjects become “removable”. There is also the naturalisation of “technical accidents” that intensify the “reasons” for removal. By blaming the victims, who are removed and displaced, and the poor, who are criminalised for their losses and ills, the responsibility of the State and capital disappear to some extent in the debate on urban segregation.

It is quite important to understand and question the mechanisms and processes that produce risk. We need to comprehend this risk as a spatial phenomenon since places are considered under risk and also vulnerabilities (Marandola Jr. & Hogan, 2009). Is the focus of risk and vulnerability, therefore, on people or places? When dealing with places, the ecological dimension of structures and their relations with the environment must be considered. When the risk exceeds the subjects’ resistance, disaster strikes. In the case of Fundão dam, capital was not a tool to overcome the risks, since the number of towns affected was huge. But one thing is certain: people who had more accumulated capital could overcome the adversities brought by the disaster more easily.

Risk also must be understood as a component of danger: it is defined by danger to people and their properties due to the possibility of the occurrence of a geological process (Castro, Peixoto & do Rio, 2005). The risk can be natural (related to “natural” processes), technological (linked to industrial production) or social. The last category comprises the damage that society, or part of it, can cause, such as with armed conflicts and wars. One important question to consider about disasters is the existing risk: how was it being managed? While this text was being written, 4 other dams face an imminent risk of collapse in the same Brazilian state and 10 others are considered an “emergency” (Oliveira, 2020). As a Brazilian, the feeling that several lives, even mine, may no longer exist after a disaster is almost natural: a few years later, in 2019, another dam collapsed, in Minas Gerais generating the biggest work-related accident in Brazil resulting in the loss of human lives and the second biggest industrial disaster of the century (figure 3). Both disasters show how one of the main economic activities in Brazil takes lives in exchange for money: mining not only ruins nature, it has been killing people. That we live in a “global risk society” is the human condition of the 21st century (Beck, 2006).

The affected people, in addition to their physical displacements, suffer a social and cultural displacement. This is because they go through a process of forced sociability, forged in political processes that are alien (Zhouri et al., 2017). Disasters are not limited to the critical situation that caused them, but they unfold in lasting processes of crisis and social suffering. However, they are generally considered only objective damages. For this reason, there are a series of strict conditions to be fulfilled for people to be recognised as affected and compensated. This shows the socio-political mechanisms of an arrangement that assigns roles to companies that neglect the realities affected by the disaster.

The emotional dynamics strongly depend on the dimension of the place, because it is impregnated with fundamental memories, which reverberate with the subjects during the removal processes. Another important thing is that more than 80% of the residents of the places affected by the Fundão dam burst are non-white (Acselrad, 2017). This shows that the production of risk can be related to aspects related to race, further exacerbated by hierarchical power relations, which exclude victims from the processes of space production.

If risks are something produced (since it is always considered as something to be lost by societies), like vulnerability, would it be consistent to see them as natural phenomena? It is essential to understand the capitalist production of space to analyse the dynamics, as there are mechanisms for risk assessment and prevention, which are not democratised. How can social justice and environmental governance be mobilised to reinvent the right to the city and territorial management in favour of the masses and memory and not capital? Would capital have the capacity to overcome risks and remove vulnerability from the site to avoid disasters? Or would risks not even be produced in those places inhabited by capital holders, because they have access to risk prevention and assessment mechanisms?

To further discuss governance and participation, I present some authors and their ideas. There has been growing participation of civil society in Brazil, since the 1988 Constitution (the most recent one, also known as the “Citizen Constitution” because it presents a lot around citizenship), which foresees participation as a constituent of citizenship (Dagnino, 2004). Along with this movement, there is a transfer of the State’s social responsibilities to civil society — this appears as a synonym for the third sector. Thus, there is a distinction between what it means to be a citizen (based on actions carried out by members of civil society as “exercise of citizenship”) and a “needy human”. Additionally, NGOs often represent scattered interests in society, reducing effective participation. There is an insoluble link between culture and politics, so political projects produce cultural meanings. The state, then, operates as a minimal neoliberal state, with social policies left in the hands of capital.

According to the World Development Report, governance could be the medium through which state and non-state actors interact to design and implement public policies (World Bank Group, 2017). Governance must be based on commitment, which allows actors to trust policy coordination, which ensures the organisation of the market and investments, and cooperation, for the execution of projects. However, there is an asymmetry in the distribution of power, engendering a capture of politics by some groups that interfere with effectiveness. The report, from the World Bank, presents the concept of clientelism, which results in a set of laws that reflect the interests of groups with greater bargaining power.

Participation and governance are intimately linked with sustainable development, considering the preservation of land for future generations and usage by root actors (people who live on the land and do not play roles such as state, market, NGOs or social movements). However, sustainable development would be incompatible with the logic of capitalist accumulation (Carneiro, 2005). The field of environmental policymaking proves to be consumed by the general economic and political imperatives that determine the use of natural conditions as conditions for the accumulation of capital. To compose participatory policy councils and, indeed, to influence disputes, it is necessary to have cultural, representative and economic capital, which are essential for bargaining. The holders of these specific capitals make up the countryside elite.

Bourdieu (1976) points out that the scientific (and technical) field is a space for struggle, in which scientific authority, which is a kind of cultural capital, is at stake. He points out that epistemological conflicts are inherently political and from these, it would be impossible to conceive of a neutral science. Because they are a form of capital, scientific capital (i.e., the number of academic titles and knowledge that people have access to) can be accumulated, thus generating an unequal struggle between agents (state, market and root actors) in the field of disputes. Another author who may be interesting to include in this discussion is Ferguson (1994), who presents the concept of the “anti-political machine” as the net result of development policies in the third world. According to him, it is what is left of the discourse, which is neither positive nor critical.

To conclude this article, I would like to share some questions for my fellow researchers and colleagues. These are simple questions that can sustain similar future research in the Global South —

  1. How would civil society be able to handle the responsibilities of the State if, at times, they do not even have access to the scientific capital of the areas they seek to defend while dealing with actual social policies?
  2. If we consider the dimensions of democracy, would it be consistent to think about environmental governance practices based on diffuse authoritarianism and clientelism?
  3. If epistemological conflicts are also political conflicts, to what extent do capital pressures act within the production of knowledge on alternatives for South-South cooperation and favouring grassroots agents?


Acselrad, H. (2017). Mariana, Novembro de 2015: A genealogia política de um desastre. In A. Zhouri (Ed.), Mineração, violências e resistências: um campo aberto à produção de conhecimento no Brasil (pp. 155–174). Editorial iGuana.

Beck, U. (2006). Living in the world risk society: A Hobhouse Memorial Public Lecture given on Wednesday 15 February 2006 at the London School of Economics. Economy and Society, 35(3), 329–345. https://doi.org/10.1080/03085140600844902

Bourdieu, P. (1976). Le champ scientifique. Actes De La Recherche En Sciences Sociales, 2(2-3), 88–104. https://doi.org/10.3406/arss.1976.3454

Carneiro, E. J. (2005). A oligarquização da “política ambiental” mineira. In Zhouri Andréa, K. Laschefski, & D. B. Pereira (Eds.), A Insustentável leveza da política ambiental: desenvolvimento e conflitos socioambientais (pp. 65–88). Autêntica.

Dagnino, E. (2004). Sociedade civil, participação e cidadania: de que estamos falando. In D. Mato (Ed.), Políticas de ciudadanía y sociedad civil en tiempos de globalización (pp. 95–110). essay, Universidad Central de Venezuela. Viceretoría Académica.

de Castro, C. M., de Oliveira Peixoto, M. N., & do Rio, G. A. P. (2005). Riscos Ambientais e Geografia: Conceituações, Abordagens e Escalas. Anuário Do Instituto De Geociências, 28(2), 11–30.

Ferguson, J. (1994). The Anti-Politics Machine: “Development,” Depoliticization, and Bureaucratic Power in Lesotho. University of Minnesota Press.

Losekann, C. (2017). “Não foi acidente!”: O lugar das emoções na mobilizaçāo dos afetados pela rupture da barrage de rejeitos da mineradora samarco no Brasil. In A. Zhouri (Ed.), Mineração, violências e resistências: um campo aberto à produção de conhecimento no Brasil (pp. 65–110). Editorial iGuana.

Marandola, E., & Hogan, D. J. (2009). Vulnerabilidade do lugar vs. vulnerabilidade sociodemográfica: implicações metodológicas de uma velha questão. Revista Brasileira De Estudos De População, 26(2), 161–181. https://doi.org/10.1590/S0102-30982009000200002

Oliveira, N. (2020, October 2). Minas tem 4 barragens com risco iminente de ruptura e 10 em nível de emergência. Cidades. https://www.otempo.com.br/cidades/minas-tem-4-barragens-com-risco-iminente-de-ruptura-e-10-em-nivel-de-emergencia-1.2393363.   

Valencio, N. (2010). Quem tem medo da remoção? A violência institucional contra moradores de ‘áreas de risco’ . In 34° Encontro Anual da ANPOCS. Caxambu; ANPOCS.

World Bank Group. (2017). World Development Report 2017: Governance and the Law. The World Bank.

Zhouri, A., Oliveira, R., Zucarelli, M., & Vasconcelos, M. (2017). O desastre no rio doce: entre as políticas de reparação e a gestão das afetações. In A. Zhouri (Ed.), Mineração, violências e resistências: um campo aberto à produção de conhecimento no Brasil (pp. 28–64). Editorial iGuana.

Carolina Maria Soares Lima is a Geographer and a Geography student enrolled in the Master of Science program at the Federal University of Minas Gerais, Brazil. Her research interests include the notion of representation and public spaces in Latin America. She also collaborates in a post-secular architecture research group and a cultural diversity observatory.

Copyright © 2022 Co:Lab | Audioman Child by