Abstract
In this article, we discuss the repercussions of water rationing in Brasília on the theory of environmental justice based on the population's perception of the 2017/2018 water crisis. Quantitative research applied the statistical technique of variance analysis. The results show that the distance from the city center, income, and education of the population in the Brazilian capital were important aspects of environmental injustice, in addition to identifying the central role of the state in this process. The traditional and centralized model of state–private water management produced and reproduced environmental injustice through institutional violence and water alienation. Although denounced and contested in light of its partiality in interpreting and addressing the water crisis, the centralized management of water resources, even when not privatized, tends to gain strength in crises whose causes and unequal distribution of water and its costs were systematically hidden.
INTRODUCTION
Brasília currently has more than three million inhabitants, making it the third largest city in Brazil. 1 It is located in the midwest region of the cerrado biome and is close to the sources of the largest rivers in the country, Tocantins, Araguaia, and Paraná. 2 Despite having the highest per capita income, the city is one of the most unequal in Brazil. 3
In 2017, Brasília was hit by a serious water crisis, which led its governor to declare a state of emergency and enact an integrated plan to face the crisis and a decree to restrict the use of water in the capital. 4
Water restriction in Brasília extended for 513 days between January 2017 and June 2018. This restriction has reached >80% of the population of the capital, specifically those supplied by the Santa Maria and Descoberto water reservoirs. The first supplies ∼600,000 people, whereas the second reaches 1.8 million inhabitants. A total of 25 administrative regions (RAs) were affected by rationing out of the existing 33. 5
During the rationing, however, several news articles indicated that the poorest RAs in the federal district were without a water supply for >5 days. 6 At the same time, other news reports showed that rich and traditional regions were running out of water for <1 day, that is, the supply was paralyzed but returned a few hours later. 7 The context of unequal distribution of the costs of environmental problems in society is a striking feature of the conceptual framework called environmental justice. 8
In this scenario of water crisis and unequal distribution of crisis costs among RAs in the federal district, this article questions: How did the population of the federal district, hit by the water crisis, perceive, feel, and react to a given crisis? How do these actions correlate with the socioeconomic profile of the city's inhabitants? Do they demonstrate aspects of environmental injustice?
The concept of environmental justice was formed from the struggles for civil, social, environmental, and territorial rights in the United States in the 1980s. 9 , 10 The search for evidence on the unequal distribution of environmental problems and risks on a portion of the population, specifically those segregated/marginalized by race and income criteria, has been the main research object in this field of knowledge. 11
Empirical explanations for the existence of an unequal distribution of environmental problems and risks for a specific portion of the population are the tendency of governments/states and companies to follow paths of least resistance to decisions on allocation of environmental risks; dynamics of the real estate market that bring environmental risk sites closer to poor people's homes; exclusion of the local population in need of decision-making bodies; reduced importance of labor and occupational health movements; exclusion of poorer people and people of color from the environmental movement; violation of social and environmental agreements and treaties, among others. 12 , 13 , 14
DATA AND METHODS
Data collection
The survey was conducted between June and July of 2018. A total of 368 questionnaires were collected from 25 RAs in Brasília, supplied by the city's two main reservoirs, Descoberto and Santa Maria (Fig. 1). Respondents were randomly chosen when they were in the administrative region, and their data were only collected if they claimed to live in 1 of the 25 RAs affected by the water restriction plan in Brasília. Thus, data collection was carried out by field researchers in each of the 25 RAs, and the interviewees could answer the questionnaire even when approached in RAs in which they did not live.
Location of the 33 RAs in Brasília. RA, administrative region.
The questionnaire was prepared using the Google Forms platform and contained 27 questions divided into 4 groups: (1) general characteristics of respondents, (2) overview of water supply crisis in Brasília, (3) knowledge and impact of the plan to restrict the use of water resources, and (4) production of actions due to the water supply crisis (Data Access Statement section).
Data analysis
To answer the research question of how the population of the federal district, hit by a water crisis, perceives, feels, and reacts to this crisis and how are these actions correlated with the socioeconomic profile of the city's inhabitants, we selected three characteristics of the socioeconomic profile of the interviewees, which served as the independent variables of the research: distance, monthly income, and education.
Distance refers to the distance of the interviewees' homes in relation to the city center of Brasília. The distance was measured in kilometers from the headquarters of the regional administration of the administrative region (RA) where the respondent lives and from the Plano Piloto bus station, regarded as the city center. For the average monthly income of respondents, secondary data calculated by the District Household Sample Survey (PDAD) 15 for each of the considered RAs were used. For the level of education, the self-declaration of the interviewees at the time of the interview was considered.
We separated the 27 questions into 10 categories, which will act as dependent variables. These categories were interviewees' position in relation to the city's water rationing, respondents' perception of water pressure reduction in their homes, respondents' perception of the existence of a water supply outage in their residence, perception in relation to the days when the water supply in their homes was cut off, perception of changes in the quality of the water supply, perception of health problems arising from water rationing, perception of government advertising to encourage conscious consumption by the population, knowledge regarding the existence of customer service channels provided by the government, perception of equality in the treatment of the federal district government in relation to the application of the water use restriction plan, and perception of changes in family routines during the period of water restriction (Table 1).
Variables Analyzed
To assess the effect of each of the three variables considered (distance, monthly income, and education) on the items surveyed, an analysis of variance (ANOVA) was performed using the three variables considered as the independent variable and the 10 categories as dependent variables.
RESULTS
The results were divided into three categories: distance from interviewees' homes to the center of Brasília, average monthly income, and education.
Distance
The ANOVA did not show a statistically significant effect of interaction between the variable distance from the respondent's home and the variables: position in relation to rationing, water quality, advertising on conscious consumption, use of customer service, and equality. It can be assumed that distance from home has no effect on these variables.
Regarding the other variables, the numbers related to interviewees' distance of residence in relation to the center of Brasília are statistically significant and can be analyzed under specific situations and reveal important discriminatory situations. The first discriminatory situation perceived in the management of the water crisis in the Brazilian capital is present in the relationship between distance and the perception of reduced pressure in the water supply [F(1.366) = 47.09, p < 0.001].
The results showed that the greater the distance that the citizens of Brasilia live from the city center, the greater their perception of the reduction in the pressure of the water that is supplied to them daily. The indication of the field data, therefore, is that residents of the city's periphery, recurrently the poorest, perceived a reduction in water pressure with greater intensity than those who lived closer to the city center.
A second important point is that the respondents' perception of the cut in water supply varied as a function of the distance category [F(1.366) = 19.52, p < 0.001]. This means that there are different perceptions about the cut in water supply to homes, and that this difference is based on the territorial distance from the center of Brasília to the point where survey respondents live. According to the results, the greater the distance from the residence to the city center, the greater the number of respondents who said that they had noticed the supply cut.
A third important aspect of the distance category is related to the fact that this category has a significant influence on the perception of the degree of disturbance of the Brasília government's water intervention plan in family routines [F(1.366) = 16.29, p < 0.001]. Respondents who lived in neighborhoods farther from the center were the same ones who noticed greater interference in their family routines during the intervention period of the government plan. In general, respondents who lived in more peripheral regions had some kind of change in their routines with regard to topics such as “bathing and personal hygiene,” “kitchen and clothes,” “yard cleaning,” “gardening,” and “car washing.”
This population was also those who most actively sought ways to adapt to a new routine by reducing bathing time, reducing the frequency of household activities, and implementing strategies for reusing water, purchasing water tanks, and storage containers.
The fourth important aspect is related to the distance and water quality. Although there was no statistical correlation between these two proxies, when we asked about the possibility that respondents had health problems due to changes in water quality during the rationing period, positive responses about health problems were statistically significant and associated with the most central regions of the city [F(1.366) = 17.49, p < 0.001]. In other words, the closer to the city center, the greater the perception of health problems arising from changes in water quality.
Income
The ANOVA did not show a statistically significant effect of the interaction between the income variable and the following variables: position regarding rationing, water quality, advertising on conscious consumption, use of customer services, and equality. It can then be assumed that income does not affect these variables.
Following the trend of the distance category, the income category brings about some important issues for reflection based on the concept of environmental injustice. The first was related to income and water pressure in taps, which was statistically significant [F(1.366) = 28.08, p < 0.001]. The data showed that the higher the income of the interviewees, the lower their perception of the reduction of water pressure in their taps. One of the hypotheses for this result may lie in the fact that apartment buildings in the city's most upscaled areas have their own water storage systems (cisterns), which would make the water pressure have no variation even at a time of crisis.
A second important point was the relationship between income and perception of water cuts, which presented a statistically significant result [F(1.366) = 46.16, p < 0.001]. The results showed that the higher the income of the interviewees, the greater the number of people who did not notice the reduction in the water supply in their homes. In addition, the RA, known as the structural city, which has the lowest income in the federal capital, reported the longest cutoff time for water supply, with an average of 48 hours of lack of tap water.
At the same time, RAs with high income, such as Park Way, Jardim Botânico, and Cruzeiro, responded that the cuts in water supply lasted <24 hours, that is, less than established in the intervention plan proposed by the government of Brasilia.
A third important point is the relationship between income and interference in routine, in which it presents a negative and significant relationship. The higher the income, the lower the perception of the influence of the water supply crisis on their routine. This result is similar to that presented previously on distance and interference in routine, demonstrating that people who live farther away from the city center are also those with the lowest income. Therefore, the disturbance of citizens' routines is related to both low income and distance from the city center [F(1.364) = 6.951, p = 0.008].
Education
The variance analysis did not show a statistically significant effect of the interaction between the variables education and water quality, advertising on conscious consumption, and customer service use. Thus, it can be assumed that the respondent's education does not have an effect on these variables.
However, the level of education of respondents proved to be an important proxy for the other independent variables surveyed. The first result was the association between education and position in relation to rationing [F(1.366) = 3.65, p = 0.05]. The higher the educational level of the respondents, the greater the number of people in favor of rationing, denoting that there is a strong effect of education on one being against or in favor of rationing.
The second aspect was related to education and perception of water pressure [F(1.366) = 11.52, p = 0.0007]. The results showed that there is a negative relationship between the level of education and the perception of a reduction in water pressure in households. That is, the higher the respondent's level of education, the lower their perception of pressure reduction.
A third aspect points in the same direction, but in the relationship between education and the perception of cuts in the water supply [F(1.366) = 18.5, p < 0.001]. Respondents with less education responded, with greater intensity, that they perceived a cut in the water supply. In a negative correlation, respondents with incomplete higher education or with less education than this were those who responded positively to the perception of cuts, whereas those with an education level equal to or greater than higher education responded negatively to the perception of cuts in water supply, that is, they did not realize that there was a restriction of supply.
Similarly, the number of days when the water supply was cut off in Brasília's homes was negatively correlated with the population's level of education. The lower the education level, the greater the perception of water outage days [F(1.366) = 21.26, p < 0.001].
The fourth aspect concerned education and its relationship with routine in the midst of a water crisis [F(1.366) = 35.15, p < 0.001]. With a negative correlation, the higher the respondent's education, the less interference in his/her routine during the water crisis in the capital city. However, the opposite was true.
The fifth and last aspects were associated with the relationship between education and the perception of equality in the government's treatment of its citizens [F(1.366) = 6.5, p = 0.01]. In a positive correlation, it was shown that the higher the level of education of the interviewees, the greater their perception of inequality in treatment in the midst of a water crisis. Although those interviewed with less education received a greater burden of the water crisis, this inequality in the distribution of the negative effects of the crisis is clearly shown in the collection instrument but not in the direct perception of these citizens.
In this sense, there is greater insight among more educated populations in relation to the inequalities and injustice promoted by the government's water policy, possibly because of greater availability, time, resources, and capacity to access and interpret information.
DISCUSSION
Two aspects are brought for debate.
The first is institutional violence. Institutional violence is understood as the disproportionate production of environmental and social ills among population groups through an active and conscious action of the state, which is then seen as the central entity in the construction of injustices. 16
The results show that the state is a discriminating agent and producer of violence. Since the beginning of the water crisis in the federal district, the actions of public agents have been discriminatory. The start of rationing by the Descoberto Reservoir, which supplies the neediest population in the region, was extremely criticized by specialists and even by the population. The argument was that in a water management system with characteristics of integration between reservoirs, such as that of the federal district, it makes no sense for rationing to start with a specific reservoir and reach a particular population group.
In addition, the results show that there is a correlation between the distance of the home from the city center, income, and education of citizens and their perceptions of water pressure reduction, supply cuts, and changes in routine. In other words, the state, with the management of its water use regulatory agency and chief executive, provided for the unequal distribution of rationing and its costs, and the poorest population in the federal district received the heaviest burden of the challenges posed by the water crisis. The state acted as an institution that builds and distributes violence, trouble, and aggression to a specific portion of the population of the federal district, whereas another portion is spared. This posture has also been verified in studies conducted in different countries and policies. 17
Considering the great and historical Brazilian and regional social inequality in the federal district, 18 it seems that the state's role in managing the water crisis is guided by a disregard for existing and latent inequalities, leading to its actions deepening injustices, in a logic very similar to that presented by Carrillo and Pellow. 19 In the management of this crisis, the state was selective and active in the interests of specific groups. It favored the richest population and those with relations in the political and economic intricacies of the federal district.
These were the most valued people, who spent 513 days on water rationing with no or few restrictions and interference in their routines. Relief for this group, however, meant a burden for the poorest and most peripheral population groups. In this case, the state reveals its contradictory facet, affirming and violating the Brazilian constitution by treating each population differently and disproportionately distributing suffering and violence. 20
Here, institutional violence is closely related to what Nixon 21 called slow violence. For him, violence is not spectacular and instantaneous, but incremental, cumulative, and relatively invisible. Slow violence draws attention to how various manifestations of environmental injustice, in terms of time, costs, suffering, and other forms of restrictions, generate dire repercussions for certain segments of the population at different time scales.
The second point of discussion is related to the concept of water alienation. Despite evidence of inequality and injustice in the distribution of costs and suffering from the water crisis in Brasília, the fact that they are perceived only by a share of respondents with a higher level of education suggests the relative success of what we could call water alienation. Water alienation can still be seen in research, as the population is against rationing, especially its most marginalized part.
We understand water alienation as the process of dissociation between the daily lives of water consumers and socioecological water production relations, mediated by production structures and reproduction of (lack of) knowledge about water management processes. These structures, although claiming greater confidence in centralized technical hydrology systems, generate greater dependency and disarticulation among users. Furthermore, they hide and normalize power asymmetries and the primary causes of the water crisis, denying the socionatural and sociotechnical characteristics of water. 22
They generate a false sense of water security when they do not have control over several factors that exacerbate the crisis and are strongly associated with a development model that presupposes an infinite supply capacity. As water alienation tends to simplify the socionatural processes of water production and ratify the undisputed power of the centralized management of water resources, it fulfills an important function of depoliticizing, neutralizing, and reducing the emergence of water conflicts, hiding the unsustainable nature of water production, and the unequal distribution of damage and suffering among the population.
CONCLUSION
The water crisis in Brasília illustrates and anticipates the perverse effects of the unfair distribution of the worsening of climate change.
Thus, studies related to engagement in the production and dissemination of information on the crisis and water inequality and the expansion and deepening of mechanisms of social control over water policy seem to present themselves as an important agenda for future research in environmental justice. At the same time, the defense of water as a universal right and an instrument for promoting social justice, providing mechanisms to compensate for conditions of structural inequality, such as public funding for water collection and storage equipment, exemption from charging and differentiated time for water supply during rationing, need to be better reflected and treated as possible ways of mitigating water crises in underdeveloped countries through future research.
Footnotes
AUTHORs' CONTRIBUTIONS
M.G.M.C., M.T.d.S.M., B.B. and B.C.S.C.M.d.O. contributed to the design and implementation of the research, to the analysis of the results, and to the writing of the article. M.G.M.C. conceived the original and supervised the project.
ETHICAL COMPLIANCE
All procedures performed in studies involving human participants were in accordance with the ethical standards of the University of Brasilia and Brazilian Government Research Ethic.
DATA ACCESS STATEMENT
AUTHOR DISCLOSURE STATEMENT
No competing financial interests exist.
FUNDING INFORMATION
The study was supported by grant from the CAPES—Coordenação de Aperfeiçoamento de Pessoal de Nível Superior.