Chapter 4 Water labour: urban metabolism, energy and rivers in nineteenth-century Rio de Janeiro, Brazil

Carrying energy and matter into the city

Despite its grandiose fluvial name, Rio de Janeiro city was not named after a river. Instead, a number of small streams, medium-sized rivers and even some large ones feed into Guanabara Bay, forming its watershed. The bay itself is a sunken estuary. So, we do not focus on any single river but on this ensemble of rivers, streams and creeks that shaped Rio de Janeiro’s waterscape. Eugene Odum claimed that delimiting ecosystems makes more sense if we consider water basins, as the flow of water adds to the cycle of solar energy in a myriad of open biological systems. Likewise, we interpret the city as an open system and consider its influence and transforming power on its surroundings. The city casts a large shadow beyond its constructed structures (buildings, walls, human-made markers). But if the city is bigger than itself, it is through the rivers that the city grows both ecosystemically and socially.⁹

We propose thus a two-tier perspective: on the one hand, we discuss the fluxes of social transformation from a historiographic viewpoint; on the other hand, we incorporate both metabolic and ecosystemic analyses for the material and energetic fluxes into that account. As we look at the rivers’ flow, we consider in this section the elements that entered the urban system through the rivers in the nineteenth century. The agency of the rivers was thus responsible for transporting energy, water, sediments and other materials into the city. At the same time, the protagonism of the rivers was tributary to their relations to other elements in their surroundings, such as the local topography, the retention of sediments by roots (trees or other plants), and even human labour. We argue that the interrelations of these agents – rivers, topography, plants and humans – were responsible for inputting materials essential to urban life.

In the environment, the geography of energy strongly connects to the geography of work. That is, rivers’ work occurs unequally, as it depends on the energetic and material influences of the local topography. How rivers established connections with their environs depended on their material reality, whether waterfalls, floodplains, water courses or estuaries. Likewise, human labour establishes relationships with the same material reality when using or changing a river. Let’s consider again our three categories: upper, middle and lower course rivers. The work by which nineteenth-century Rio de Janeiro’s residents could divert part of the Carioca upper course river to the city’s fountains was possible due to the environmental conditions resulting from the joint work of the rivers flowing downhill, the sun and the winds, as much as to the infrastructure of river pipelines. The fusion of human labour with the work of the fluvial environment was necessary to direct the waters to the fountains that supplied the city. It could hardly occur in the middle and lower course of rivers, obstructed by sediment deposition and often tainted by water contamination.¹⁰

Several water basins drained into Rio de Janeiro city. Rio’s inhabitants were well aware of their dependence on the river’s flow for fresh water. Tellingly, these inhabitants became known as Cariocas, named after one of the first rivers they exploited for drinking water in the colonial period. The Tijuca massif influenced the distribution of fresh water and the city’s expansion. Despite the existence of the Carioca River, the history of Rio de Janeiro was marked by successive water scarcity situations due mostly to inadequate water supply policies in the face of accelerated population growth. The Carioca could not supply enough fresh water. This made it necessary to invest in infrastructure and technical training of public engineers who had to learn how to manage the river for water supply. Before the 1860s, Rio de Janeiro’s water policies were improvised, often unconnected to larger planning and frequently flawed, with rare studies on water regimes. The initial public works on the rivers Carioca (1750) and Maracanã (1817) could not harness the rivers’ work, resulting in precarious water distribution and a massive presence of leaves and sediments. The 1860s marked a political watershed (pun intended) for Brazil’s imperial government regarding the infrastructure projects in the capital. The imperial administration created important institutions with their technical staff strongly dominated by engineers – first within the Ministry of the Empire and then within the Ministry of Agriculture, Commerce and Public Works (henceforth Ministry of Agriculture). As civil engineering developed and the federal administration professionalised urban services, these new technical staff formed scientific commissions to research the rivers and their characteristics to inform public policies.¹¹

Understanding the dynamics of the fluvial work was essential to carry out structural modifications and increase the city’s water use efficiency. In the 1860s and 1870s, public investment in the water commissions for dozens of rivers and streams closest to the city simplified and reduced their characteristics to four types of variables: daily water flow, elevation of the water catchment outlet, capacity of the water tanks, and the condition of the forests near the water springs. A few years before the 1860s water commissions, the imperial government was concerned mainly with river flow and elevation of the water catchment outlet. Sound knowledge about the flow of the rivers enabled greater control over their work, improving the quality and quantity of water supply.

Water was just one of the substances that rivers carried from the upper course environment into the city. Sediments were also important. Due to their upper course’s altitude and topographical gradient, the rivers could carry larger sediments and small stones. As fluvial energy dissipated along the river’s course, its capacity to carry sediments diminished. Rivers in the middle course transported smaller and smaller particles and transformed their environments by depositing sediments of different sizes. As the water descended and lost energy, the heavier elements rushed to the bottom, and the rivers’ labour could only carry the smaller ones. Thus, the river’s labour defined the kind of sediment that remained in different terrains and which kind of biotic community emerged: in the lower river course, the small deposits composed the mud of Rio’s mangrove forest, while in the upper course, only the heavier sediments remained, providing good soil for dense forests.

Humans often used these materials distributed by the rivers locally. However, they frequently added their own labour to the rivers’, and carried the sediments to the city as raw material for constructing buildings, roads and draining systems, among other purposes. Tons of stones, sand and clay taken from the fluvial systems in the upper course reached Rio de Janeiro loaded on the back of working mules. These animals – and their work – were an essential part of the metabolic transformation of the city, as we will discuss later. The mules acted both locally and across the different levels of the river systems by transporting people and materials through the mountains of Rio. Animal traction was just one of the many forms in which combined labour of the biotic and non-biotic communities contributed to the shaping of Rio de Janeiro city.

The repairs to the water supply system that flowed into the Carioca Aqueduct in 1861 exemplify the importance of this complex combination of human, animal and fluvial energy. The Department of Public Works, a government agency subordinated to the Ministry of Agriculture, estimated this project would require the consumption of 1000 m³ of stones, 5000 m³ of sand and lime, and 400 m³ of clay for landfilling. In the same year, reforms of the water tanks in the Tijuca mountain, part of the Maracanã River’s tributaries, demanded 2000 m³ of stone and lime. Thirteen years later, in 1874, paving with water-bound macadam a few access roads to the Tijuca massif (which rises from Gávea, Jardim Botânico and Andarahy, now Tijuca) used 2970 m³ of rocks and small stones. The estimated weight of these rocks and stones was around 5,000 tons. In the area’s steep slopes, mules and donkeys seldom carried more than 100 kg, and therefore an estimate of 50,000 trips in the year 1874 were necessary only to bring these stones to roads – or about 130 trips a day. The fact that the work site was close to the area where the materials were extracted certainly reduced the work of these animals in constructing the roadways and railroads which climbed up the mountains. Nevertheless, the imperial government depended heavily on the energy produced by the mules and counted, among its ‘staff’, hundreds of these animals who took advantage of the upper course’s fluvial environment to feed on the grass near the riverbanks and to quench their thirst in the springs.¹²

The presence of mules in the forested hills and mountains meant a drastic transformation due to the trampling of the soil. Soil compaction reduced the forest’s regenerative capacity and added to deforestation, which had been dramatic since the early nineteenth century, with the first coffee plantations established in the valleys shaped by fluvial work. Deforestation in Rio de Janeiro’s mountains had two main causes: slash-and-burn cultivation and logging. In the case of coffee cultivation, it was common practice to use fire indiscriminately to eliminate the dense forest and clear space for coffee. Ironically, coffee seedlings developed poorly when exposed directly to the sun, and therefore the areas most sought (and mostly burned) by the coffee planters were the shaded environments in the river valleys. A survey indicated that there were over 250,000 coffee trees scattered in nine locations of the Tijuca massif. The number may seem modest when we take into account the full extension of the massif (around 3,300 hectares), but we still know little about the human uses of the valleys.¹³

Logging, however, was intended to supply the metabolic demand of fuelwood and timber within the city and its immediate hinterland. Most of the wood was burned into charcoal still in the mountains and then easily transported by mule or/and by boat to feed the furnaces in the city. The data collected by Warren Dean for the year 1882 show that there were many furnaces in the city, including 173 bakeries, 33 coffee roasters, 36 sugar refineries, 60 ironworks and foundries, 66 felt-hat manufacturers, 11 potteries, 5 cardboard and paper factories, 5 glass and porcelain factories, and 22 boiler manufacturers. The 1850 census indicates that over 27,000 houses had one oven each, and the 1872 census reveals that this figure reached over 30,000 by then. In addition to the daily fuel consumption for cooking, cloth dyeing and other activities, each domestic space would have burned tons of firewood just in their construction. Even brick houses consumed huge amounts of firewood. Indeed, Dean argues that brick houses were wood houses. He estimated that a small house used around 30,000 bricks (63 m³), for which it would have been necessary to burn 18 tons of firewood.¹⁴

Despite the difficulties for historians to develop good estimates of forest loss in the nineteenth century, we know that urban society felt deforestation acutely. Forest clearing deeply damaged the symbiotic relationship between tree roots and soil microorganisms. Soil disintegration and erosion caused by the rain and the rivers hindered water circulation in the riverbeds. The consequent siltation in the plains transformed the fluvial environment in the medium and lower courses. Swamps expanded, therefore, due to deforestation in the upper courses. Physicians and other professionals trained in social medicine saw the swamps as unhealthy areas threatening the urban social body. And they were ready to fight the swamps with all the might of modern science.¹⁵ The urban elite interpreted the outpouring of mud and sand in the city centre as an unwanted environment that needed modernisation.

Probably one of the most remarkable reforestation projects worldwide in the nineteenth century, the reforestation of the Tijuca mountain officially began in 1861 under the initiative of the Ministry of Agriculture. The area was considered critical for maintaining Rio’s water supply and thus was purchased by the imperial government from private owners, mostly coffee planters. By 1889, when the imperial regime fell due to a republican coup d’état, more than 100,000 seedlings of over 100 species had been planted. In line with the rhetoric of exuberant nature pervading Brazilian culture –as can be noted, for example, in the national anthem lyrics – the reforestation project privileged native species. Between 1861 and 1889, native species (and otherwise) circulated through scientific institutions and reached the reforestation site, which became a sort of forest laboratory. The imperial government claimed several objectives for this initiative, and the most recurrent in the documents were the hopes for a healthier climate, the need to supply the city with water, a steady source of wood for urban building, and the beautification of the mountains. In the end, the mixture of scientific ideas with the classist and racist culture of the Carioca elite turned the enterprise into a project of exclusion, as we will see ahead.¹⁶

From a metabolic perspective, the change in forest dynamics severely affected rivers and their flows. The quality and quantity of water were directly related to deforestation and reforestation. Here, humans and other animals, as well as plants and rivers, worked together to transform the biophysical environment. Just like energy and matter, information entered and was metabolised by the urban ecosystem.¹⁷ Fluvial systems store information in their topography and ecological relations, but they also experience structural modifications based on information generated by humans (scientific ideas, sanitary customs, urban irregularities). The information circulating within Rio’s human society, such as ideas of forestry, race, class and private property, affected the river water regimes as much as the quantity of precipitation or the steepness of slopes.

The weight of European ideas was massive in nineteenth-century Brazil. Physicians trained at the Medical School of Rio de Janeiro, from its foundation in 1832 to the late 1860s, were strongly influenced by French social medicine. The Medical School’s first generation of professors had graduated in Paris or at least at the Academia-Médico Cirúrgica de Rio de Janeiro (1813), whose faculty had also trained in France. It was not unrelated that Paris’s urban transformation in the early nineteenth century became a blueprint to be tested in the Brazilian imperial capital, including the health rationale implicit in the new urban planning models. Very influential in Rio, these medical professionals championed the neo-Hypocratic theory of miasmas to discern which environments benefited human health and which did not. The new social medicine interpreted and imagined an ideal healthy urban environment, pointing to several places supposedly unfit for human health. This rationale legitimised state policies within and beyond the city, including ones that removed traditional or poor communities from certain areas while making others unaffordable.¹⁸

Transformations within the river/urban system

Rio de Janeiro grew attached to Guanabara Bay. By the turn of the nineteenth century, the city was no longer a timid European entrepôt. Rio’s inhabitants had drained mangrove swamps or landfilled large extensions of the bay since the foundation of Rio in the sixteenth century. The rivers draining to the bay near the harbour had also been incorporated into the workings of the city. However, in the early nineteenth century, the heavy influx of new migrants gave a new impulse to the old practices and techniques. In fact, by the mid-nineteenth century, it was no longer just a matter of scale. The way Rio’s inhabitants saw the small rivers permeating the urban tissue had changed. An almost seamless tapestry of ditches, creeks, canals and streams ran from the hinterland to Guanabara Bay, a tapestry punctuated by houses, factories and buildings. Rivers and the city were joined in a complex spreading biosystem responsible for many urban services.¹⁹

The nineteenth century witnessed several changes in Rio’s river-city system. Some of them were material, and others not. While the colonial government had actively drained lagoons, dredged swamps and sought to manage water supply, the reach of the modern city was more extensive and qualitatively different. The modern city-river biosystem did require active planning, large public works and regular maintenance, and this was the first significant change. No longer a task for a small group of residents in a neighbourhood, or an impromptu project by colonial authorities, ditches had to be dug and kept clean. In many instances, rivers and creeks were channelled and forced underground, while in other cases, they were domesticated into straight lines by concrete walls, destroying the meandering trajectory that characterised their lower curse. As Rio de Janeiro was the country’s capital, the federal government invested a fair amount of its budget into local public works and hoped to tame the rivers that often overflowed during the rainy season. Thus, the modern city-river metabolism claimed more energy through public investments (for example, public funding and hired or enslaved labour). It was also a political metabolism.

The second prominent change was aesthetical in character. The rectification of rivers followed a geometrised urban design highly appreciated by Rio’s boosters. But the rivers resisted such geometrisation. Likewise, the people using these rivers persisted in their traditional practices, though maybe not in the same places. For instance, in 1822, the British visitor Maria Graham thus described the Carioca River: ‘At the entrance to the valley, a little green plain stretches itself on either hand, through which the rivulet runs over its stony bed, and affords a tempting spot to groups of washerwomen of all hues, though the greater number are black’.²⁰ The work of the river and the washerwomen merged at the local riverine topography, with a configuration of stream water, rocks and the sun that was useful for those people. Other travellers observed them ‘standing in the stream and beating their clothes upon the boulders of rock’ to clean them. And later they ‘purified in the stream and bleached [the clothes] in the sun’.²¹ This scene could no longer be seen in the modern city of the early twentieth century. The Carioca River was then surrounded by large villas forming an upscale neighbourhood, and black women with red scarves doing laundry were no longer welcome. But the city still needed the work of washerwomen, so they moved their business to other, less visible, streams and creeks. So, modernity reached certain rivers but not others. Segregated as it was, the urban fabric allowed the coexistence of transformed rivers and rivers that retained much of their traditional appearance and uses.

The third significant change concerned the cultural perception of the urban rivers. As the maintenance of modernised rivers demanded more investments, and as they changed the city’s outlook – more or less visible, more decorated or more disguised, with drained areas or harnessed in a factory – the urban rivers of Rio de Janeiro also became cultural laboratories of the hygienist scientific culture. Were they the solution for or the source of the dreaded urban epidemics? What did the transformed rivers say about the moral character of Rio de Janeiro’s inhabitants? Were there good and bad urban rivers? Were some healthy and others miasmatic?

These changes shaped the metabolism of urban rivers. The state’s intervention, the coexistence of tradition and modernity, and the new cultural perception of urban rivers forged a segregated city that was shaped by the work realised by the rivers and on the rivers. Certain activities, such as the washing described by Maria Graham, would take place in specific sites – such as slum houses near small rivers, where washerwomen took out their laundry – or in rivers far from the city centre. Other activities, such as textile manufacturing, were located on high-priced urban lots with good access to water (used for powering the mills). But these lots were not so valuable that they would compete for space with upscale houses and public buildings. Still others, such as burning fuelwood harvested at the river’s upper course, took place across the entire city as urban dwellers cooked, built houses and used river energy in their daily lives.²²

Thus, the metabolic processes within the city accelerated over the nineteenth century. By the early twentieth century, the city used more energy, consumed more raw materials and produced more effluents than a hundred years earlier. It was an order-of-magnitude change. More importantly, however, the agents that operated these changes were quite diverse. They mixed their labour in planned and unplanned operations – and the rivers were both actors and space in which these actions occurred.

Therefore, we can identify human action in cleaning ditches, dredging rivers, and general public works such as landfilling, draining and bridge building; they all used energy and transformed urban rivers. Humans also planted edible cress on the riverbanks for human consumption and grass for their mules and horses and used river water to discharge garbage and animal remains. Also, having remained a slave society until 1888, Rio’s inhabitants forced other humans to work in the rivers. These were mostly demeaning and unhealthy jobs, such as transporting and eliminating human waste, which those enslaved people known as tigres executed during the early hours of the day.

Human labour could also combine with animal labour – like the mules and the equines that transported materials from the rivers and trampled the river margins while grazing. These animals depended on the energy available in the capinzais (grass pastures) planted in the floodplains of the main rivers, especially on the city’s outskirts. Advertisements for the sale and rental of farms mentioned capinzais to raise the price. Between the 1830s and 1880s, it was common for these ads to mention how many animals could be fed, describing the size and quality of the grasses. The rivers had a wider floodplain for grass planting in the urban periphery. The largest capinzais were located in the floodplains of rivers in the northern suburbs, such as the Joana, Maracanã and others.²³

Suburban agriculture spread along the banks and designed new river paths to irrigate plants such as agrião (watercress). Agrião (Nasturtium officinale) is a plant that grows easily in small streams with gentle running water. More common in the northern suburbs, landlords near rivers had ditches dug to divert the water to where the crop would be planted. In the Catumby Valley, closer to the city centre, suburban farmers were so successful in this crop that the region became known as Zona do Agrião (Watercress Zone). After intense conflicts arising from the diversion of waters, the cultivation of watercress was prohibited in 1878.²⁴

In fact, rivers and their human uses had been the target of municipal regulation for decades. However, efforts to tame and maintain the rivers intensified with the growth of the symbolic capital of physicians and their quest for urban health in the 1870s. In 1875, the Central Board of Public Hygiene organised the General Commission for Health. All sanitary activities carried out by the Commission focused on rivers and other bodies of water: cleaning and conservation of rivers, ditches and mangroves, backfilling swamps, and general urban cleaning. Thus, river metabolism converged more intensely with the interests of local elites. The river conservation and cleaning service began through a contract with Júlio Richard, and at the end of the nineteenth century, it became a municipal service. In the first year, the service focused on 15 km of rivers and ditches, including the Carioca, the Trapicheiros, the Joana, the Comprido, the Papa Couve, the Berquó, and a couple of ditches. As early as 1881, 464 kilometres of rivers and ditches were included in Júlio Richard’s cleaning service. The waste that had to be dealt with ranged from sediments comprising sand and mud, small plants such as bushes and banana trees, to utensils and furniture such as mattresses and beds. In addition, the Comprido, Papa Couve, Berquó and Trapicheiros rivers were widened in the 1870s.²⁵

These metabolic processes also included nonhuman agents – often in the opposite direction of what humans would plan. Thus, plants absorbed solar energy and flourished in the nutrient-rich river waters. This accelerated ecological succession, clogging riverbeds and forcing humans to regular and expensive dredging. The rivers themselves showed that they could not be tamed easily. Their ecological dynamics challenged the efficiency of the public works that sought to turn them into obedient organic machines. Frequent rainstorms, high tides and siltation helped the rivers overflow their artificial concrete beds, with flooding becoming a constant nightmare for Rio de Janeiro’s public administration.

Effluents, waste and products leave the river/urban system

As we have seen, energy and material inputs were harnessed and transformed by the work of different agents in nineteenth-century Rio de Janeiro – humans, plants, animals and rivers. From these metabolisms, the city excreted things that influenced its surrounding areas – the rural hinterland, the beaches and the oceans. Some elements left never to return. This was the case, for instance, of commercial exports such as coffee or sugarcane, which were loaded in the large ships in the port of Rio de Janeiro and entered the international capitalist market. However, some outputs immediately returned to the urban territory due to local characteristics such as topography, winds and tides. Such was the unfortunate case of urban sewage, for which the federal administration sought a solution for decades.

In 1864 the Rio de Janeiro City Improvements Company was created, mostly with British capital. It obtained a concession from the imperial government to collect and treat domestic sewage for over 12,000 houses in the city. The administration had high expectations for a solution to its sanitary crisis. Rio had more than 200,000 residents by then, and enslaved African and Afro-descendent people discharged the sewage. At night, these workers gathered human waste from homes, shops and public buildings in wood barrels and dumped them on the beaches. They also relied on the tide’s work to do the ‘sanitary service’ of cleaning the beaches. This system of sewage disposal – occasionally carried out by freedmen – survived until the end of the imperial period but began to decline with the implantation of the City Improvements Company’s sanitation system.²⁶

The City Improvements Company implanted a sewage separations system which was quite modern for the period. According to the engineers of the Ministry of Agriculture, the success of this system depended on a reliable and abundant supply of water to push the urban waste. However, in the secondary collectors, errors in the original construction and failure to consider the steep slopes caused frequent interruptions in water and sewage flows. The uneven terrain of the city provoked cracks and fissures in the thin-walled pipes that had little pressure due to the low volume of water. The consequence was ‘that puddles of stagnant shit appeared everywhere’.²⁷ After a series of complaints from the population, two imperial commissions were established to inspect, monitor the works and indicate sites needing repairs. Even after completing the repairs in 1867, the system depended heavily on the river waters. When the water was not enough, the complaints persisted. The report of the Ministry of Agriculture of 1868 said: ‘Some homeowners have [issued lawsuits] against the exhalations, that in the lateral entrances for the flushing tanks, ventilators, etc., are released, and they molest the residents of the neighbourhoods’.²⁸

Legitimised by constant public complaints, state-employed engineers exercised some control over the City Improvements Company. Thanks to the advantageous concession contract, this levelled up the unequal relationship between the company and the city. Engineers and the company agreed they had to solve the problems of obstructions and lack of water for the proper functioning of the sewage system. The obstructions always worsened after heavy rains because sand and dirt ended up in the sewage tubes. Constant work to unblock them was necessary. The Company did 177 such works in 1869, 219 in 1870, 192 in 1871 and 197 in 1872 – a total of 785 cleanups in 4 years. José Pereira Rego, a member of the Central Board for Hygiene, suggested several solutions, including the following: to increase the supply of water and the slope in specific points for suitable sewage disposal; to lay subsidiary pipes on sound, solid terrain to avoid depressions caused by irregular landfills, often poorly done; to disinfect regularly the reservoirs; and to raise the level of the streets or to bury the pipes deeper in the soil to prevent them from breaking with the weight and vibration of the vehicles.²⁹ In 1874, the exhalation issue was almost solved, mostly due to fewer obstructions, good precipitation levels (1568 mm), and the inclusion of ventilator openings. Ten years after the City Improvements Company installed the first sewer plumbings, the situation appeared stabilised and satisfactory – at least for the city’s high-end areas. Physicians and engineers emerged from the crisis as the champions for the modern city and valued members of the public administration.³⁰

The city’s sanitary metabolism became more efficient in the 1880s when an adequate daily water supply became available. In the nobler neighbourhoods of the suburbs, the system had good water intake from the reservoirs. Modernising the sanitary complex required new technologies, such as installing coal filters in the 351 ventilators and 52 lateral entrances and the replacement of the flushing tanks. The new flushing tanks enabled the release of water discharges from the rivers into the system. The City Improvements staff cleaned them, and the intermittent siphons regularly released water to avoid malodorous exhalations.³¹ Again, river energy, human labour and technology enabled the urban metabolism to eject undesirable effluents.

Between the 1860s and 1870s, effluents were released into Guanabara Bay after the filtering out of solid material and dumped by barges and ships – and later by hydraulic steam pumps. The solid material was compacted, and a small part was directed to Rio’s agricultural green belt. The larger share, however, ended up on the Island of Sapucaia, in Guanabara Bay. Sapucaia received both garbage and solid residues of human waste, functioning as a sanitary landfill from 1865 to 1949. In the 1880s, an engineer employed by the City Improvements Company developed a technology to reorient the urban metabolism’s material flows towards recycling. Using calcination techniques to treat the solid waste, the final substance would become ‘a kind of mortar, at first very fluid, but which by exposure to air and the sun hardens’.³² After experiments, he claimed to have successfully turned the waste into cement through a simple and inexpensive method. By incineration, the material lost all its water and organic matter, showing high concentrations of lime and alumina, ‘almost in the same proportions as these substances enter the Portland cement’.³³ By the end of the 1880s, the City Improvements Company used two kilns to produce cement via this technique. There is no evidence of the energy required, but we can deduce that this increased the demand for the fuelwood and charcoal obtained in the upper river courses.

However, in a clear breach of its contract, the company often discharged the sewage directly into Guanabara Bay without proper treatment. The waters turned brown in several sensitive areas, and the fishermen in the bay recognised them as the ‘Aguas da City’ (the Company’s waters). Antonio Augusto Monteiro de Barros, the engineer who supervised the City Improvements Company on behalf of the imperial administration, reported 433 illegal discharges between 1883 and 1886. The fines were significant for each episode, but the discharges became less and less frequent over the years: 222 cases in 1883, 135 in 1884, 47 in 1885 and only 29 in 1886.

In 1889, at the end of the imperial era, Rio de Janeiro’s sewage extended over 253,710 metres. It had expanded more than 30,000 metres in 1885 towards Riachuelo, Vila Isabel and Andarahy Grande, areas inhabited by a population much less privileged than the ones the system first had served. But it was a thirsty system. In 1887, it consumed over 3000 m³ of water daily for the flushing tanks, urinals and latrines.³⁴

The need for more rivers

The imperial government transformed the river environments surrounding the city to ensure a specific regime of urban metabolism. The engineers interpreted the water bodies through a new technical language capable of modifying those realities. However, the city and its metabolic demand did not stop growing. To the rural population of Africans and Afro-descendants freed in 1888 – with the legislative abolition of slavery – was added a mass of European immigrants. The city’s population was over 270,000, rising to half a million in 1890 and over 800,000 in 1900. It became necessary to capture and modify new rivers for the city’s functioning.³⁵ Until the 1870s, water diverted from the rivers springing in the Tijuca massif (up to 10 kilometres away from the city) could supply 36 million litres per day, although the supply system captured just over 20 million (of which 13 were lost upon distribution). In this decade, new river commissions sought solutions to the distribution problems and to capture more distant rivers. These reports led to the collection of the Ouro, São Pedro and Santo Antônio rivers in the Tinguá mountains, just over 50 kilometres north of the city.

The works began in 1876 and ended in 1881, collecting and distributing 40 million litres daily through 53 kilometres of pipeline. The channelling of these rivers had some impasses, mainly due to local landowners who refused to cede their lots, which resulted in judicial expropriations and minor adjustments in the channel’s path. Two large reservoirs were built in Serra do Tinguá – one to collect water from the São Pedro River and the other from the Ouro and Santo Antônio Rivers – and several smaller ones across the city. Although the collection of nearby rivers was well advanced, a severe drought compromised the supply of the imperial capital, imposing an intermittent distribution and requiring a water transport service through the new railroad built to support the rivers’ channelling.³⁶ With the end of the contract, the railway began to be used to transport passengers and goods. Thus, the search for water eventually benefited the flow of people and objects.

Despite a substantial increase in the water supply to the city, its operation slowed substantially during drought events. Engineers measured the flows of rivers in the Tinguá Mountains from 1882. They collected data in rainy and dry seasons for a more realistic estimate of water resources. Environmental control required knowledge in an efficient and reductionist technical-scientific language.³⁷ Added to the then-recent collection of the d’Ouro and other rivers, the estimates of river flows in the Tinguá mountain suggested the supply of up to 60 million litres of water daily – the minimum amount for the functioning of the city, according to the engineers – in times of extreme drought. In other words, even with the new intake, the city still suffered from water shortages in years of moderate drought. The following year, the expropriation of land close to water collection sites began in the Serra Velha, Cantagalo, Brava and Macuco waterfalls and the São Pedro river headwaters in the Tinguá mountains.³⁸

At the beginning of 1889, after a drought and a severe yellow fever epidemic, several proposals for the collection of rivers emerged amid political unrest in the imperial capital. The winning bid was the most daring (and least expensive): it proposed to have three rivers diverted through temporary gutters for more than 60 kilometres in just six days. Emperor Dom Pedro II was one of the few who believed in the incredible proposal, assisting with two trains to transport materials and personnel. On 24 March 1889, after the promised six days, 14 million litres per day had been collected for the city’s supply.³⁹

Months later, the imperial regime was replaced by a republican military government. The same urban elites retained control over the fluvial metabolism based on medical theories and civil engineering practices. The modernisation of Rio de Janeiro transformed the riverscapes into supposedly healthier environments, materialising the dream of a tropical capital city comparable to Paris and its great reforms of 1902 to 1906. The electricity demand of the growing industrial park and public transport (trams) also required the modification of riverscapes increasingly far from the city. That was the case with the construction of the Fontes Hydroelectric Power Plant, 80 kilometres from the city. It was inaugurated in 1908 to supply 24 MW of electricity generated from damming 180 million cubic metres of water. Rio de Janeiro’s thirst increased exponentially throughout the twentieth century, affecting ever-distant river ecosystems.⁴⁰

Conclusion

Urban metabolism relies on rivers carrying matter and energy through the urban territory. From a relational perspective, we have shown that the work of the rivers shaped the horizon of possibilities for historical developments in Rio de Janeiro. The increase in energy and water demand led to the optimisation of resources to reduce the inefficiency of collection and distribution. This required the urban technical staff to search for larger, ever-distant rivers, which brought to the front the prestige of engineers and doctors in the eyes of the municipal authorities.

Throughout the nineteenth century, the urban appropriation of the rivers’ work occurred in myriad ways. The local topographical complexity enabled diverse riverscapes, including their use by different human social groups across rivers’ upper, middle and lower courses. As the capital of the Brazilian Empire, Rio received varied migratory inflows, which shaped a diverse urban population as to attitudes to and uses of the work of the rivers. Being profoundly marked by class, race and gender, these social relations were also expressed in the river landscapes. Examples of this were the ban on bathing in waterfalls for enslaved Africans and Afro-Brazilians, and the restriction of the washerwomen’s work.

The entry of water into the urban ecosystem occurred both through unmodified channels and a technosphere comprising aqueducts, gutters and other devices for the collection and distribution of water. Here, the work of the rivers facilitated, through gravitational energy, the supply of water through fountains spread through the city. The labour of people and other animals constructed and maintained the water technosphere, including forest restoration, a pioneering initiative in the nineteenth century. Therefore, the rivers worked in situ, shaping the complex relationship between forests, springs and humans, but also from afar, providing water and energy for the city.

Upon entering the river-city system, river work allowed for different activities in the suburbs of Rio de Janeiro, and water bodies were domesticated through channels, ditches and reservoirs. The transformation of riverscapes conformed to diverse interests, such as those of suburban farmers, with their use of floodplains, banks and ditches; textile manufacturers, who designed water systems for running mills and cooling them, as well as expelling the effluents; and doctors and engineers. According to the sanitary mentality of the latter, nothing could stand still, especially water. In this way, the rivers and their works were intensely modified by draining flooded areas with stagnant water, dredging silted areas and constricting riverbanks by straightening channels in the more densely built-up areas. But the rivers resisted these domestication attempts, refusing to become fully compliant organic machines. At the exit of the rivers-city system, the rivers performed their most sordid work: pushing out organic waste and industrial efflux. The local topography did not facilitate this work, resulting in serious problems. The system for collecting and directing domestic sewage was redesigned, requiring greater consumption of river water.

The reliance on rivers’ work for the city’s ever-expanding functioning resulted in the search for larger and more distant rivers. The demand for water and energy led to systematic efforts by municipal engineers, who mapped the rivers of greatest interest for urban metabolism. In 1870, the rivers captured by the city were up to 11 kilometres away. In the first decade of the twentieth century, this distance increased to 80 kilometres. At this moment, a paradigmatic transformation in technology enabled the conversion of river work into electrical energy, which started to be directed to the urban territory to perform new works. By transforming local riverscapes and appropriating the rivers’ work for electricity generation, the capitalist system connected Rio de Janeiro much more intensely to the river network of the hinterland. This process foreshadowed later developments covering the entire country, as hydroelectricity would become predominant in the Brazilian energy matrix throughout the twentieth century.

Notes

1. 1. Friedrich Engels, The part played by labour in the transition from ape to man. Paris, Foreign Languages Press, 1975.

2. 2. Karl Marx, Capital: a critique of political economy, vol. I. London, Penguin Random House, 1976, 284. Since then, other scholars, notedly Hannah Arendt in her The human condition (Chicago, University of Chicago Press, 1998) have explored the distinction between labour and work. It is not the purpose of this text to engage in this debate. We propose, however, that as rivers do have historical, non-intentional agency, they realise both labour and work, as they on the one hand move energy and on the other hand transform the world; Justin Williams, ‘Theorizing the non-human through spatial and environmental thought’. In: Teena Gabrielson, Cheryl Hall, John M. Meyer & David Schlosberg (eds.) The Oxford handbook of environmental political theory. Oxford, Oxford University Press, Oxford Handbooks online, 2016.

3. 3. River (Rio) was the Portuguese word used for large bodies of water in the sixteenth century, when the Europeans first reached Guanabara Bay. Often considered a misnomer for the bay, it was nevertheless the appropriate word.

4. 4. The sun is the object of the plant – an indispensable object to it, confirming its life – just as the plant is an object of the sun, being an expression of the life-awakening power of the sun, of the sun’s objective essential power; Karl Marx, ‘Critique of Hegel’s philosophy in general’. In: Economic and Philosophic Manuscripts of 1844. Moscow, Progress Publishers, 1959, available at https://marxists.org.

5. 5. Karl Marx, ‘Estranged labour’. In: Economic and philosophic manuscripts of 1844. Moscow, Progress publishers, 1959, available at https://marxists.org; John Bellamy Foster, ‘Marx’s theory of metabolic rift: classical foundations for environmental sociology’, American Journal of Sociology 105, no. 2, (1999), 366–405.

6. 6. Erik Swyngedouw, ‘Circulations and metabolisms: (hybrid) natures and (cyborg) cities’, Science as culture 15, no. 2 (2006), 105–21.

7. 7. Bruno Latour, Reassembling the social: an introduction to actor-network-theory. Oxford, Oxford University Press, 2005; André Vasques Vital, ‘Water spells: new materialist theoretical insights from animated fantasy and science fiction’, Historia Ambiental Latinoamericana y Caribeña (HALAC) 12, no. 1 (2022), 246–69.

8. 8. Christopher Pearson, ‘Beyond “resistance”: rethinking nonhuman agency for a “more-than-human” world’, European Review of History 22, no. 5 (2015), 709–25.

9. 9. Eugine P. Odum, Ecology: a bridge between science and society. Sunderland, MA, Sinauer Associates Incorporated, 1997.

10. 10. Richard White, The organic machine: the remaking of the Columbia River. New York, Hill and Wang, 1995.

11. 11. Jorun Poettering, ‘Paradise for whom? Conservatism and progress in the perception of Rio de Janeiro’s drinking-water supply, sixteenth to nineteenth centuries’, Journal of Latin American Studies 50, no. 3 (2018), 703–27; Alida C. Metcalf, Sean Morey Smith & S. Wright Kennedy, ‘ “A mere gutter!”: The Carioca Aqueduct and water delivery in mid-nineteenth-century Rio de Janeiro’, Urban History 49, no. 1 (2022), 61–78; Diogo de Carvalho Cabral, ‘Águas passadas: sociedade e natureza no Rio de Janeiro oitocentista’, Raega-O Espaço Geográfico em Análise 23 (2011), 159–90.

12. 12. Bruno Capilé, ‘Os muitos rios do Rio de Janeiro: transformações e interações entre dinâmicas sociais e sistemas fluviais na cidade do Rio de Janeiro (1850–1889)’, Doctoral dissertation, Graduate Program on Social History (PPGHIS), Rio de Janeiro, Universidad Federal do Rio de Janeiro, 2018.

13. 13. Mauricio de Almeida Abreu (ed.), Natureza e sociedade no Rio de Janeiro. Rio de Janeiro, Secretaria Municipal de Cultura, Turismo e Esportes, Departamento Geral de Documentação e Informação Cultural, Divisão de Editoração, 1992.

14. 14. Warren Dean, With broadax and firebrand: the destruction of the Brazilian Atlantic Forest. Berkeley, University of California Press, 1995.

15. 15. Dean, 1995; Abreu, 1992.

16. 16. Claudia Heynemann, Floresta da Tijuca: natureza e civilização no Rio de Janeiro, século XIX (vol. 38). Prefeitura Da Cidade Do Rio de Janeiro Secretaria, 1995; Alexandro Solórzano, Gabriel Paes da Silva Sales & Rafael da Silva Nunes, ‘O legado humano na paisagem do Parque Nacional da Tijuca: uso, ocupação e introdução de espécies exóticas’, Fronteiras: Journal of Social, Technological and Environmental Science 7, no. 2 (2018), 43–57; de Carvalho Cabral, 2011.

17. 17. Ramón Margalef, Perspectivas de la teoria ecológica. Barcelona, Blume, 1978.

18. 18. Jaime Larry Benchimol, Pereira Passos: um Haussmann tropical. Rio de Janeiro, Prefeitura da Cidade do Rio de Janeiro, 1990; Richard Sennet, Flesh and stone: the body and the city in western civilization. 1st edn. New York, W.W. Norton, 1994; Sidney Chalhoub, Cidade febril: cortiços e epidemias na Corte imperial. São Paulo, Cia. das Letras, 1996; Inês Andrade, ‘Notes on the therapeutic values of historic gardens in neoclassical hospitals in Rio de Janeiro (1830–1900)’, Gardens and Landscapes of Portugal 5, no. 1 (2018), 4–21.

19. 19. Maurício de A. Abreu, Evolução urbana do Rio de Janeiro. 4th edn. Rio de Janeiro, Instituto Municipal de Urbanismo Pereira Passos, 2006.

20. 20. Maria Graham, Journal of a voyage to Brazil and residence there, during part of the years 1821, 1822, 1823. London, Printed for Longman, Hurst, Rees, Orme, Brown, and Green, and J. Murray, 1824, 161.

21. 21. Daniel Parish Kidder & James Cooler Fletcher, Brazil and the Brazilians: portrayed in historical perspective and different sketches. Philadelphia, Deacon & Peterson, 1857, 120.

22. 22. Abreu, 1992.

23. 23. Capilé, 2018.

24. 24. Bruno Capilé, ‘Os idealizadores da socionatureza urbana e a transformação da paisagem fluvial carioca’. In: Alexander Costa e Luisa Schneider (eds.) Rios urbanos: diferentes abordagens sobre as águas nas cidades. Curitiba, CRV, 2022, 19–37.

25. 25. J. A. C. Oliveira, Livro de registros dos trabalhos executados pela Comissão Geral de Salubridade, para o combate à febre amarela. Rio de Janeiro, 18/02/1875–22/01/1876. Biblioteca Nacional, Manuscritos, 14,04,001.

26. 26. Teresa Meade, ‘ “Civilizing” Rio: reform and resistance in a Brazilian city, 1889–1930’. University Park, PA, Penn State University Press, 1997.

27. 27. Diario do Rio de Janeiro. Noticiário: companhia de esgotos. Front page, issue 197, 18 July 1864.

28. 28. J. A. F. Leão, Relatório do Ministério da Agricultura, do Comércio e das Obras Públicas do ano de 1868. Ministerial Report: Agricultura, 102, 1869.

29. 29. Jose Pereira Rego, Relatório apresentado à Academia Imperial de Medicina. Ministerial Report: Agricultura, 1873.

30. 30. Meade, 1997; Chalhoub, 1996.

31. 31. J. A. Saraiva, Relatório do Ministério da Agricultura, do Comércio e das Obras Públicas do ano de 1881. Ministerial Report: Agricultura, 1882.

32. 32. A. J. Oliveira, Relatório do Engenheiro fiscal da Rio de Janeiro City Improvements Company Limited. Ministerial Report: Agricultura, 1882, 5

33. 33. Oliveira, 1882, 6.

34. 34. R. A. Silva, Relatório do Ministério da Agricultura, Comércio e das Obras Públicas para o ano de 1887. Ministerial Report: Agricultura, 1888, 82.

35. 35. Benchimol, 1990.

36. 36. A. A. M. Penna, Relatório do Ministério da Agricultura, Comércio, Obras Públicas para o ano de 1883. Rio de Janeiro: Typographia nacional, 1884.

37. 37. Upon reflecting on the role of technical-scientific knowledge in urban societies, Milton Santos emphasised the ‘technicisation’ of landscapes. For Santos, information is infused into things, being the primary vector of territorial transformation. After the Second World War, this profound environmental interference, limited initially to large cities, expanded to rural and protected areas; see Milton Santos, A natureza do espaço. Técnica e tempo. Razão e emoção. São Paulo, HUCITEC, 1996; see also James Scott, Seeing like a state: how certain schemes to improve the human condition have failed. New Haven-London, Yale University Press, 1998.

38. 38. José de Santa Ritta, A água do Rio: do Carioca ao Guandu: a história do abastecimento de água da cidade do Rio de Janeiro. Rio de Janeiro, Synergia/ LIGHT/ Centro Cultural da SEAERJ, 2009.

39. 39. Lise Sedrez & Bruno Capilé, ‘Os modernos rios cariocas’. In: Lorelai Kury, Bruno Capilé, Lise Sedrez & Marcelo Motta, Os rios do Rio. Rio de Janeiro, Andrea Jakobsson Studio, 2021, 72–129.

40. 40. Elisa Müller, ‘O padrão tecnológico da Light’. In: Eulália Maria Lahmeyer Lobo & Maria Bárbara Levy (eds.) Estudos sobre a Rio Light: relatório de pesquisa. Rio de Janeiro, Instituto Light / Centro de Memória da Eletricidade no Brasil, 2008, 533–64.

References

• Abreu, Maurício de A., Evolução urbana do Rio de Janeiro. 4th edn. Rio de Janeiro, Instituto Municipal de Urbanismo Pereira Passos, 2006.
• Almeida Abreu, Mauricio de (ed.), Natureza e sociedade no Rio de Janeiro. Rio de Janeiro, Secretaria Municipal de Cultura, Turismo e Esportes, Departamento Geral de Documentação e Informação Cultural, Divisão de Editoração, 1992.
• Andrade, Inês, ‘Notes on the therapeutic values of historic gardens in neoclassical hospitals in Rio de Janeiro (1830–1900)’, Gardens and Landscapes of Portugal 5, no. 1 (2018), 4–21.
• Arendt, Hannah. The human condition. Chicago: University of Chicago Press, 1998.
• Benchimol, Jaime Larry, Pereira Passos: um Haussmann tropical. Rio de Janeiro, Prefeitura da Cidade do Rio de Janeiro, 1990.
• Capilé, Bruno, ‘Os idealizadores da socionatureza urbana e a transformação da paisagem fluvial carioca’. In: Alexander Costa e Luisa Schneider (eds.) Rios urbanos: diferentes abordagens sobre as águas nas cidades. Curitiba, CRV, 2022, 19–37.
• Capilé, Bruno, ‘Os muitos rios do Rio de Janeiro: transformações e interações entre dinâmicas sociais e sistemas fluviais na cidade do Rio de Janeiro (1850–1889)’, Doctoral dissertation, Graduate Program on Social History (PPGHIS), Rio de Janeiro, Universidad Federal do Rio de Janeiro, 2018.
• Chalhoub, Sidney, Cidade febril: cortiços e epidemias na Corte imperial. São Paulo, Cia. das Letras, 1996.
• Dean, Warren, With broadax and firebrand: the destruction of the Brazilian Atlantic Forest. Berkeley, University of California Press, 1995.
• de Carvalho Cabral, Diogo, ‘Águas passadas: sociedade e natureza no Rio de Janeiro oitocentista’, Raega-O Espaço Geográfico em Análise 23 (2011), 159–90.
• Engels, Friedrich, The part played by labour in the transition from ape to man. Paris, Foreign Languages Press, 1975.
• Foster, John Bellamy, ‘Marx’s theory of metabolic rift: classical foundations for environmental sociology’, American Journal of Sociology 105, no. 2 (1999), 366–405.
• Heynemann, Claudia, Floresta da Tijuca: natureza e civilização no Rio de Janeiro, século XIX (vol. 38). Prefeitura Da Cidade Do Rio de Janeiro Secretaria, 1995.
• Latour, Bruno, Reassembling the social: an introduction to actor-network-theory. Oxford, Oxford University Press, 2005.
• Margalef, Ramón, Perspectivas de la teoria ecológica. Barcelona, Blume, 1978.
• Marx, Karl, Capital: a critique of political economy, vol. I. London, Penguin Random House, 1976.
• Marx, Karl, ‘Critique of Hegel’s philosophy in general’. In: Economic and philosophic manuscripts of 1844. Moscow, Progress Publishers, 1959.
• Marx, Karl, ‘Estranged labour’. In: Economic and philosophic manuscripts of 1844. Moscow, Progress publishers, 1959.
• Meade, Teresa, ‘ “Civilizing” Rio: reform and resistance in a Brazilian city, 1889–1930’. University Park, PA, Penn State University Press, 1997.
• Metcalf, Alida C., Sean Morey Smith & S. Wright Kennedy, ‘ “A mere gutter!”: The Carioca Aqueduct and water delivery in mid-nineteenth-century Rio de Janeiro’, Urban History 49, no. 1 (2022), 61–78.
• Müller, Elisa. ‘O padrão tecnológico da Light’. In: Eulália Maria Lahmeyer Lobo & Maria Bárbara Levy (eds.) Estudos sobre a Rio Light: relatório de pesquisa. Rio de Janeiro, Instituto Light / Centro de Memória da Eletricidade no Brasil, 2008, 533–64.
• Odum, Eugine P., Ecology: a bridge between science and society. Sunderland, MA, Sinauer Associates Incorporated, 1997.
• Pearson, Christopher, ‘Beyond “resistance”: rethinking nonhuman agency for a “more-than-human” world’, European Review of History 22, no. 5 (2015), 709–25.
• Poettering, Jorun, ‘Paradise for whom? Conservatism and progress in the perception of Rio de Janeiro’s drinking-water supply, sixteenth to nineteenth centuries’, Journal of Latin American Studies 50, no. 3 (2018), 703–27.
• Santa Ritta, José de, A água do Rio: do Carioca ao Guandu: a história do abastecimento de água da cidade do Rio de Janeiro. Rio de Janeiro, Synergia/ LIGHT/ Centro Cultural da SEAERJ, 2009.
• Santos, Milton, A natureza do espaço. Técnica e tempo. Razão e emoção. São Paulo, HUCITEC, 1996.
• Scott, James, Seeing like a state: how certain schemes to improve the human condition have failed. New Haven-London, Yale University Press, 1998.
• Sedrez, Lise & Bruno Capilé, ‘Os modernos rios cariocas’. In: Lorelai Kury, Bruno Capilé, Lise Sedrez & Marcelo Motta, Os rios do Rio. Rio de Janeiro, Andrea Jakobsson Studio, 2021, 72–129.
• Sennet, Richard, Flesh and stone: the body and the city in western civilization. 1st edn. New York, W.W. Norton, 1994.
• Solórzano, Alexandro Gabriel Paes da Silva Sales & Rafael da Silva Nunes, ‘O legado humano na paisagem do Parque Nacional da Tijuca: uso, ocupação e introdução de espécies exóticas’, Fronteiras: Journal of Social, Technological and Environmental Science 7, no. 2 (2018), 43–57.
• Swyngedouw, Erik, ‘Circulations and metabolisms:(hybrid) natures and (cyborg) cities’, Science as Culture 15, no. 2 (2006), 105–21.
• Vasques Vital, André, ‘Water spells: new materialist theoretical insights from animated fantasy and science fiction’, Historia Ambiental Latinoamericana y Caribeña (HALAC) 12, no. 1 (2022), 246–69.
• White, Richard, The organic machine: the remaking of the Columbia River. New York, Hill and Wang, 1995.
• Williams, Justin, ‘Theorizing the non-human through spatial and environmental thought’. In: Teena Gabrielson, Cheryl Hall, John M. Meyer & David Schlosberg (eds.) The Oxford handbook of environmental political theory. Oxford, Oxford University Press, Oxford Handbooks online, 2016.