The River of Oblivion

Territorial Policies, Economic Cycles and Integrated Conservation in the Middle São Francisco

The São Francisco River Valley, in Brazil, arose relatively late as a major axis of inland settlement, particularly from the seventeenth century onwards, when successive economic cycles and administrative interests began to shape its fluvial system. Initially driven by extensive cattle ranching, this process gave rise to cultural landscapes, navigable waterways and rural estates that gradually developed into urban settlements. Over time, these settlements assumed a strategic role in Portuguese colonial expansion into the hinterland, at a time when settlement remained largely concentrated along the Atlantic coast. 

By the First Brazilian Republic (1889–1930), a relative balance between urban settlements and the natural environment had become established across this territory, following a process that had developed throughout the nineteenth century. From the second half of the twentieth century onwards, and particularly under the military dictatorship (1964–1985), the region became the target of large-scale interventions. These projects reconfigured the material and symbolic foundations of the existing territorial system, fragmenting the historic urban network, disrupting urban continuities and eroding territorial identities. Although the São Francisco River had long been portrayed as a symbol of national integration and celebrated as the “river of cattle ranches”, the twentieth century exposed the vulnerabilities underlying this development model. The resulting disruptions became increasingly apparent, causing irreversible losses to both the river’s hydrological system and its network of riverside towns. 

In this context, the river may be understood as a “river of oblivion” – a reference to the River Lethe of Greek mythology – as hydroelectric infrastructures contributed to the physical and symbolic erasure of urban settlements and traditional riverine practices. The recent expansion of energy-production frontiers associated with the energy transition lends renewed relevance to this historical question, particularly as renewable energy projects are increasingly presented as pathways towards sustainability. Against this contemporary backdrop, this paper traces the successive territorial policies that have reshaped the São Francisco River Basin over time, with particular emphasis on the Middle São Francisco, and examines their impacts on urban heritage using integrated conservation as an analytical framework. 

The study adopts a historical-interpretative approach combined with cultural landscape analysis. It draws on a critical review of the literature on the historical settlement of the São Francisco Valley and the transformations associated with the development of large-scale hydraulic infrastructure throughout the twentieth century. Historical records and photographic documentation are also used to trace continuities, disruptions and spatial reconfigurations across both fluvial and urban landscapes. The analysis is informed by integrated conservation and cultural landscape approaches, which situate urban heritage within a broader territorial system. 

A critical reading of heritage shows that it cannot be understood as a fixed or stable set of assets. Rather, heritage is a historical construct, shaped by processes of selection and the social attribution of value¹. From this perspective, considering water as part of historic urban landscapes draws attention to the connections between ecological, technical and cultural values formed over time. Heritage may therefore be understood both as a cultural practice and as a process of meaning-making², bound up with the formation of urban identities. In this sense, territorial heritage acquires particular relevance, since its values are formed through historical relationships between communities, environments and forms of territorial occupation³.

Water, in turn, has long shaped riverside landscapes. The relationship between urban space and waterfronts extends beyond the juxtaposition of historic urban form and its physical setting, since riverbanks, coastlines, port districts and linear parks carry both material and intangible dimensions across different scales. Through everyday life, subsistence practices, local economies, recreation, public health and quality of life, these spaces mediate relationships between city and nature. Such relationships also depend on the ecosystem services provided by aquatic environments⁴.

Rivers, lakes and waterfronts have therefore directly influenced urban morphology, settlement patterns and economic development. At the same time, they have acquired specific cultural meanings and forms of appropriation, although these are often obscured by strictly functional interpretations. Within this discussion, the concept of cultural landscape remains important to the understanding of water as heritage. Predating later developments in heritage theory, it brings material and environmental attributes into relation with knowledge systems, cultural practices and ways of life historically associated with riverine landscapes⁵.

Watercourses and the cultural landscapes associated with them are, however, continually transformed, as are their relationships with historic urban landscapes. Technical interventions linked to modernisation – such as dams, canals and reservoirs – alter natural conditions and produce new territorial configurations and symbolic references. In this respect, cultural landscape and Historic Urban Landscape should not be treated simply as successive paradigms. They may instead be read as complementary approaches that extend the territorial scope of heritage interpretation. Landscape can accordingly be considered as heritage, as a territorial resource and as an object of planning, bringing natural, historical and social values into the same field of interpretation⁶.

The cultural landscape approach, consolidated by UNESCO in the 1990s, established the inseparability of nature and human agency. The later Historic Urban Landscape approach addresses contemporary urban dynamics more directly, including socio-spatial change, environmental pressures and economic restructuring. In doing so, it extends the discussion of heritage conservation to questions of urban transformation, sustainability and socio-economic development⁷.

Although the two approaches overlap, they are not interchangeable. Their points of contact are particularly relevant to riverine landscapes, where natural substrates, historical continuities and social practices are closely intertwined. Urban form, in these circumstances, cannot be separated from the ways in which water is appropriated, managed and interpreted culturally. In Brazil, the National Institute of Historic and Artistic Heritage (IPHAN) recognises eighteen cultural and natural assets under the category of cultural landscape⁸. Historic urban centres, by contrast, had received institutional protection since the Vargas era, without their natural systems being clearly incorporated into their interpretation as heritage. 

The cultural landscape category was later added to existing forms of heritage designation, but was never fully incorporated into conservation practice. The recognition of Brazilian historic centres began during the Vargas administration (1930–1945) and was subsequently consolidated through a heritage model rooted in nineteenth-century traditions. This model privileged historic cores and the museification of cities, with a strong influence from European precedents such as Paris. Yet Giovanni Giovannoni had already argued that monuments should be understood in relation to the urban fabric and surrounding landscapes, anticipating questions later addressed through cultural landscape and Historic Urban Landscape approaches⁹. Within this institutional context, IPHAN developed an interpretative model that remained influential for decades. Formal and stylistic values were privileged in the treatment of urban ensembles, while social, environmental and territorial dynamics were often marginalised. Although this approach played an important part in safeguarding historic centres, it also reinforced a narrow conception of heritage. Landscape systems and everyday practices, among other dimensions, have only more recently been reconsidered through cultural landscape and Historic Urban Landscape perspectives. Questions therefore remain about the capacity of existing conservation strategies to address both built heritage and the landscapes of which it forms part. 

More recently, developments in heritage theory have brought sustainability and climate change more firmly into the discussion. Attention has also turned to urban voids along waterfronts, particularly in post-industrial cities, where interventions have in some cases sought to restore historical relationships between cities and water systems. In Liverpool, for example, the regeneration of the Albert Dock preserves warehouses, docks and basins while introducing new uses compatible with the character of the waterfront. In Hamburg, HafenCity uses canals, quays and port infrastructure as the basis for urban redevelopment. In both cases, water forms part of the heritage setting, maintaining a relationship between the industrial past and contemporary urban life. 

In Brazil, projects such as Porto Maravilha in Rio de Janeiro and Porto Digital in Recife reflect comparable processes of regeneration. Both involve the revalorisation of central areas and waterfront districts and incorporate, to varying degrees, principles of integrated conservation and the creative economy. At the same time, however, they are associated with financial restructuring that can disrupt existing social relations and land-use practices. Waterfronts, in these cases, are often treated as secondary components of the heritage context and subordinated to redevelopment strategies associated with real-estate interests. As a result, their cultural, symbolic and everyday dimensions receive only partial consideration in planning practice.

The São Francisco River Valley, however, differs from these examples. Many of its towns and communities did not experience the intensive industrialisation that elsewhere disrupted relationships between cities and water systems. More direct relationships between communities and the river therefore remain visible in everyday practices and spatial organisation. The São Francisco River continues to structure territorial identities, urban form and the life of riverine communities. This is not, however, a recovered or reconstructed heritage, but part of a living landscape in which material and intangible dimensions remain active, although increasingly vulnerable. Nor does this continuity imply an absence of conflict. Hydraulic interventions, particularly dams, continue to threaten historically established relationships between society and water in the region.

The historical formation of the São Francisco River Valley forms part of the wider process of inland settlement in Brazil and, more specifically, of Portuguese colonial expansion beyond the Atlantic seaboard (Fig. 1). From the seventeenth century onwards, the so-called “Velho Chico” became an important route between the coast and the interior, at a time when overland circulation remained limited and river navigation provided one of the principal means of movement, trade and administrative control¹⁰.

The São Francisco River thus became part of the territorial organisation of the region, whose history remains legible in the material traces of successive forms of settlement and circulation. Settlements, routes and early roads formed part of this process and, taken together, supported the development of an inland urban network through which dispersed populations were progressively connected¹¹. This territorial formation, however, cannot be attributed solely to Portuguese colonisation. Long before colonial settlement, diverse Indigenous groups inhabited the river basin and maintained their own circulation systems, land-use practices and territorial meanings. Many of the routes later appropriated by colonisers were, in fact, Indigenous in origin. The valley was therefore formed through successive and overlapping territorial layers rather than through a single founding process. 

From the eighteenth century onwards, colonial settlement intensified with the expansion of extensive cattle ranching. As sugarcane production dominated the coastal areas, livestock raising was increasingly displaced inland, where the São Francisco basin provided access to water and extensive tracts of land suited to pastoral activity¹². Cattle ranching nevertheless remained subordinate to the colonial export economy, whose principal economic base lay in sugar production¹³. The distribution of farms along the valley also reflected the interaction of different bodies of knowledge. Indigenous knowledge informed understandings of water dynamics, seasonal variation and environmental risk, while colonial practices introduced systems of land division and extensive livestock management¹⁴. Together, these logics shaped subsequent patterns of settlement.

Most farms, and later towns, were established on elevated riverbanks, locally known as barrancos. These sites combined proximity to water with a degree of protection from seasonal flooding. This spatial relationship helps to explain the heritage significance of water in the São Francisco basin, since the resulting cultural landscape bears evidence of both geomorphological adaptation and accumulated environmental knowledge. Urban settlements developed gradually in close association with navigation, administration and economic activity, while many towns served as river ports and logistical nodes linking overland routes¹⁵. The result was a distinctive pattern of urbanisation shaped by the relationship between settlement and river (Fig. 2).

Religious and administrative institutions further contributed to this process. The Portuguese Crown, together with the Catholic Church, promoted the institutionalisation of parishes and villages, encouraging population concentration while extending colonial control across the river basin¹⁶. Within the Middle São Francisco, two principal arcs of settlement can be identified. The first extends from Bom Jesus da Lapa to Xique-Xique and comprises towns historically associated with religious movement, commerce and regional circulation. Barra provides a particularly representative example. Formerly known as Vila de São Francisco, it acquired judicial and commercial importance because of its position at the confluence of the São Francisco and Grande rivers. Within the same arc, Bom Jesus da Lapa developed through a different set of relationships, largely associated with religious practice. 

Its diocesan status reinforced its regional importance, while pilgrimage to the Morro do Bom Jesus became closely bound up with the town’s urban identity. 

A decisive episode occurred in 1691, when Francisco de Mendonça Mar settled in the caves of Morro da Lapa and placed religious images there. The devotional practices that followed gradually established the site as a place of pilgrimage¹⁷, while the hill remained a lasting focus of religious devotion. Urban development subsequently followed these religious practices. Settlement initially occupied the area between the rocky formation and the river, with relatively simple and dispersed housing. Over time, however, occupation became concentrated around the sanctuary, as pilgrimage encouraged commerce, accommodation and other supporting activities¹⁸. The town consequently developed around a pilgrimage-based urban structure, with Morro do Bom Jesus as its spatial and symbolic focus (fig. 3). Its caves were transformed into sacred spaces and have remained places of continuous worship for more than three centuries. The sacralisation of the hill, in turn, intensified pilgrimage and directly influenced the expansion of the town¹⁹.

The relationship between river and hill remains visible today. Boat processions and religious practices continue to connect water with the sacred landscape and, through this continuity, sustain aspects of regional cultural identity. Yet the cultural landscape of the Middle São Francisco is not defined by ritual practices alone. Its material dimension can also be read through different types of vessels, which record successive technological changes: Indigenous rafts (jangadas), wooden boats carrying symbolic figureheads (fig. 4), and, later, shallow-draught steamships such as the Vapor Benjamin Guimarães (fig. 5)²⁰. These technologies did not sever the relationship between river and town; rather, they became part of it. Intangible practices circulated through the same riverine network. The work of washerwomen and dances such as São Gonçalo followed routes connecting towns and ports, giving these places cultural functions alongside their logistical ones.

The territorialisation of the São Francisco River Valley began in the seventeenth century, intensified during the eighteenth and reached a period of relative stability in the early twentieth century. This process involved more than territorial control. It also organised the fluvial system as a network of social, economic and spatial relationships. Over time, this network formed the basis of a cultural landscape in which the river connected settlements and structured patterns of circulation. Towns did not function as isolated units, but as parts of an interconnected urban network in which water provided both a means of movement and a basis for spatial organisation. In the Middle São Francisco – as in the Upper and Lower sections, although under different geomorphological conditions – settlement patterns therefore attest to a long-standing relationship between water and urban form. This continuity can be attributed partly to the region’s distance from the principal centres of political authority and partly to the persistence of local economies dependent on ecosystem services. 

Even where industrial activities were introduced, they did not entirely disrupt this pattern. Relations between society and the river retained a degree of continuity over time, allowing forms of territorial organisation, cultural practice and everyday life associated with water to persist. Mnemeia, in this sense, refers not simply to the recollection of a past landscape, but to the persistence of relationships through which the river remained part of lived experience. 

The second fluvial arc, extending from Xique-Xique to the neighbouring cities of Juazeiro and Petrolina on opposite banks of the São Francisco River, initially followed a pattern of cultural landscape formation comparable to that of the first arc. Here too, settlement developed through a long-standing relationship with the river and its seasonal rhythms. The diversity of the São Francisco landscape shaped the use of ecosystem services and, in turn, many aspects of everyday life. Floodplain agriculture, the seasonal use of sandbanks for recreation and relationships with the riparian vegetation of the Caatinga all formed part of this landscape, while paleochannels, seasonal streams and other hydrological features influenced patterns of settlement and land use. 

This relationship was profoundly altered in several municipalities, particularly Pilão Arcado (fig. 5), Sento Sé (fig. 6), Remanso and Casa Nova. In these places, the riverine landscape was reshaped by an energy policy devised and implemented through a top-down model of territorial intervention, with little local participation and a strong emphasis on large-scale hydroelectric development. One of the principal agents of this transformation was CHESF (Companhia Hidroelétrica do São Francisco), Brazil’s first state-owned electricity company. 

From the mid-twentieth century onwards, state intervention in the valley became increasingly institutionalised. The São Francisco Valley Commission (CVSF) was established in 1948, followed by SUVALE in 1967 and CODEVASF in 1974. The succession of these institutions marked a shift towards development-oriented territorial planning based on policies of regional integration²¹. Against this institutional background, large-scale infrastructure reshaped the valley, as dams, hydroelectric plants and irrigation systems altered both the landscape and its spatial organisation. The scale of intervention increased considerably in the 1970s with the construction of the Sobradinho Dam, whose reservoir inundated extensive areas and displaced approximately 72,000 people. Entire urban settlements were dismantled and rebuilt elsewhere²².

The environmental consequences were extensive and long-lasting. Changes to the hydrosedimentological regime altered river flows and increased siltation both upstream and downstream of the reservoir. The effects, however, extended well beyond physical changes to the river and its landscape. These interventions also produced a profound process of symbolic and cultural deterritorialisation, as riverside populations lost places through which belonging, memory and everyday life had been sustained. Former towns, productive areas and familiar landscapes disappeared beneath the reservoir. What was lost, therefore, was not only the built environment, but also a historically constituted relationship between society and the river. 

Traditional practices were affected accordingly. Floodplain agriculture, artisanal fishing and river navigation declined or had to adapt to new environmental and territorial conditions, despite having long supported both subsistence and cultural identity along the river. In Sento Sé (fig. 7), Pilão Arcado, Remanso and Casa Nova, the losses extended across several dimensions of everyday life. Houses, churches, public spaces and informal meeting places disappeared and, with them, many of the spatial references through which collective memory had been sustained.

The construction of the “new towns” added another dimension to this process. Although relocation addressed the immediate requirement to rehouse displaced populations, it was not matched by effective strategies for socio-economic reintegration. Adaptation to new productive systems was consequently difficult and uneven. The transition from floodplain agriculture to irrigated farming was often exclusionary, while artisanal fishing declined as ecological conditions and river dynamics changed. Oral testimonies from displaced residents repeatedly describe experiences of loss, fear and social rupture. Moreover, these effects were not confined to the period of displacement; they continued across generations and altered community life over the longer term. The hydroelectric programme therefore did more than reorganise territory. It contributed to the erasure of pre-existing territorialities and to the progressive silencing of lived urban experience among the affected populations²³.

The contrast between the former and new settlements is particularly evident in their relationship with the river. Pilão Arcado, Remanso, Sento Sé and Casa Nova had historically developed with their urban form oriented towards the São Francisco. Their replacements, by contrast, were conceived through a markedly technocratic approach to planning, in which standardised layouts and administrative requirements took precedence over historically established relationships between settlement, landscape and river. This top-down approach reduced the complexity of existing territorial arrangements to simplified spatial solutions, thereby breaking with a river-oriented pattern of settlement that had developed over centuries. Street layouts and circulation were increasingly aligned with roads and geometrical planning principles, while the river lost much of its former place in urban life and, in many cases, became visually and functionally peripheral. 

The spatial consequences can be seen particularly clearly in Pilão Arcado (fig. 8) and Sento Sé, where the new settlements no longer open towards the river but are oriented away from it, breaking with earlier patterns of spatial dependence on the São Francisco. Remanso and Casa Nova, although situated closer to the reservoir margins, were likewise reoriented towards highways and regional road networks. A comparable planning rationale was later applied to other towns in the Lower-Middle São Francisco, including Rodelas and Petrolândia (fig. 9), particularly following the construction of the Luiz Gonzaga hydroelectric plant. 

Large-scale energy infrastructure thus altered not only individual settlements but also the relationships that had sustained the wider fluvial network. One of the most significant changes concerned navigation. River transport, long dependent on seasonal and hydrological conditions, gradually lost ground to road transport, while roads became the predominant means of moving people and goods. This shift, in turn, changed the position of individual towns within the regional urban network. Places that had previously performed intermediary and connective functions within the fluvial system lost much of this role as investment and mobility were redirected towards road-based networks. Riverside towns were consequently relegated to more peripheral positions and, no longer serving as transit points within the regional system, many became increasingly isolated within new patterns of circulation. 

These changes weakened the position historically occupied by riverside towns within the territory, while reducing economic activity and disrupting long-established social networks. More importantly, the replacement of a river-based territorial order by increasingly standardised and road-oriented forms of planning weakened the relationships through which these communities had historically inhabited and understood the São Francisco. The consequences therefore extended beyond displacement or changes in urban form: they affected the continuity of a cultural landscape shaped over generations by navigation, work, social life and everyday contact with the river.

Given the scale and complexity of the São Francisco River cultural landscape, the hydroelectric schemes undertaken by CHESF during the 1970s and 1980s set a pattern for later large-scale interventions, including the São Francisco River transposition project. Rooted in a developmentalist agenda, the transposition project – located in the third geographical sector, between Juazeiro-Petrolina and Paulo Afonso – sought to overcome perceived geographical constraints by diverting water to drought-affected regions. Yet, rather than departing from the interventionist practices of previous decades, it prolonged a history whose social and territorial consequences bear comparison with the lasting “traumas” caused by the sacrifice of parts of the São Francisco Valley in the late twentieth century. 

From the 2000s onwards, attempts were made to address some of these effects through limited environmental restoration measures, including riparian reforestation and erosion-control programmes presented as environmental compensation. These measures, however, remained piecemeal and were not sustained at a scale commensurate with the interventions they sought to mitigate. At the same time, the transposition project renewed the discourse of “national integration”, while exposing the limitations of the development model on which that discourse rested. Territorial integration remained partial and uneven, perpetuating imbalances in social relations as well as in the relationship between communities and the natural environment. 

Alongside these federally funded interventions, particularly those delivered through the PAC (Growth Acceleration Programme), large private landholdings have continued to expand across the Middle São Francisco Valley. As a consequence, irrigated fruit production and large-scale grain cultivation have further altered patterns of land use and territorial organisation. Along the Juazeiro-Petrolina axis, for instance, export-oriented agriculture has grown substantially, supported by hydraulic infrastructure and closely connected to global commodity chains. Regional production has consequently become increasingly specialised and directed towards external markets. 

Further west, in Bahia, mechanised agribusiness follows a different pattern. Extensive and land-intensive, it has been closely associated with land concentration and the homogenisation of the landscape. Municipalities such as Barra, Muquém de São Francisco and Xique-Xique have experienced increasing pressure from agribusiness, particularly through schemes such as “Baixios de Irecê”. In turn, large-scale farming has altered established patterns of land use and intensified rural displacement and socio-spatial change. The result is an increasingly fragmented territory, where highly modernised agricultural enclaves coexist with marginalised areas that derive few benefits from such investment. Economic growth, therefore, has not necessarily brought greater territorial cohesion. At the same time, the river has gradually lost some of the functions it once performed in regional economic and spatial relations. 

More recently, other economic activities have expanded across the Middle São Francisco. Although less directly associated with the main river course, they remain connected to the wider catchment and its ecological systems. This is the case with wind-energy installations and extractive activities in the Serra do Assuruá (fig. 10), near Xique-Xique, as well as mineral prospecting around Santo Inácio, a district of Gentio do Ouro. 

Within this more recent phase of territorial change, wind-energy development has had a growing influence on the contemporary cultural landscape. It cannot, however, be understood solely in terms of technological innovation or energy transition, since its development is also tied to changes in global patterns of investment, particularly following the financial crisis of 2007–2008. Between 1998 and 2009, wind-energy development in Brazil was largely driven by domestic circumstances, including the electricity supply crisis and the need to increase generating capacity. From 2010 onwards, by contrast, international investors acquired a much greater presence in the sector, while wind-energy projects increasingly formed part of financial diversification strategies and global investment portfolios²⁴.

This change was accompanied by a geographical shift. Unlike earlier projects concentrated along the coast, wind-energy development after 2010 moved rapidly inland, with the semi-arid Northeast, including the Middle São Francisco, receiving a substantial share of this investment. During this period, approximately 98.4% of wind projects were located inland, marking a pronounced change in the distribution of new developments²⁵.

As a result, the São Francisco Valley has become part of a new geography of energy production. Riverine landscapes now accommodate wind, hydroelectric and agricultural systems within a territory already shaped by successive layers of public and private intervention. These projects, however, do not occupy empty or previously unused land; rather, they encounter established forms of settlement, production, tenure and cultural practice. Different histories of occupation and land use therefore coexist within the same territory, often giving rise to disputes over access, production and control. Thus, although wind-energy development is commonly presented in terms of energy transition and climate-change mitigation, its local effects include land appropriation, changes to land-use arrangements and restrictions on access to territory. 

Rural and traditional communities are directly affected by these changes. There is, therefore, a tension between the scales at which these projects are conceived and experienced. At the global level, wind energy is presented as “clean” and sustainable; at the local level, however, its implementation can intensify inequality, displacement and socio-spatial vulnerability. Moreover, these effects are shaped by the involvement of the state, transnational corporations and investment funds. Decisions over location and operation are frequently driven by economic interests defined elsewhere, leaving affected communities with limited influence over how their territories are used. 

This geography of energy production extends beyond generation itself. The Serra Dourada Project, for example, consists of a high-voltage transmission corridor of more than 1,000 km, connecting wind and solar generation areas in Bahia and Piauí to the national electricity grid. Although presented primarily in terms of technical efficiency and grid integration, its territorial consequences extend well beyond the transmission of electricity. Its principal function is to carry electricity over long distances from the areas in which it is generated. In doing so, the corridor crosses rural properties, highways and waterways, including the São Francisco River itself, imposing land-use restrictions along its route and adding to existing patterns of spatial fragmentation. 

The relationship between infrastructure and local development is equally important. Unlike generation sites, transmission infrastructure provides relatively little continuing employment once construction has been completed. Claims of sustained local economic benefit therefore require qualification. More broadly, an emphasis on operational efficiency can divert attention from the longer-term consequences for land use, settlement and local spatial relations. 

Seen together, these processes constitute a third period in the transformation of the São Francisco Valley considered in this study. When read alongside previous cycles of intervention, they point to persistent shortcomings in the application of integrated conservation. Current debates continue to treat agriculture, energy and water infrastructure largely as separate fields, usually concentrating on their immediate socio-spatial effects. Yet each acts upon the same territorial and hydrological setting. Agricultural expansion, hydroelectric development, river transposition and wind-energy deployment are all part of the continuing transformation of the riverine landscape. Treating them separately, therefore, makes it difficult to account for the historical relationships between these processes and perpetuates sectoral approaches to landscape analysis. 

This does not mean, however, that these forms of production are necessarily incompatible with riverine communities. They are already part of the economic life of the region and may, under appropriate public policies, contribute to the continuity of riverine livelihoods. Such a relationship depends, in turn, on safeguarding ecosystem services and on forms of territorial planning capable of negotiating competing demands without disregarding existing ways of life. 

Under present conditions, these activities interact in different ways with cultural practices, local economies and environmental systems that have given historical depth to the valley. Municipal government, however, remains largely reactive. Large-scale projects are frequently accepted without adequate consideration of their wider territorial consequences, while short-term employment is often treated as sufficient evidence of development. In this respect, the contemporary situation retains a clear connection with the earlier cycles discussed throughout this article: fragmented public policies persist, reproducing long-standing asymmetries in the governance of the São Francisco Valley. 

The “river of oblivion” describes the gradual loss of cities, practices and landscapes along the São Francisco River as part of a longer historical process in which relationships between society, nature and memory have become increasingly fragmented. These transformations have been shaped by technical, economic and political interventions largely guided by priorities defined outside the territories concerned. As a result, patterns of settlement, land use and circulation have changed, while historically established relationships between riverine communities and the fluvial system have progressively weakened. 

As the São Francisco has increasingly been treated as a resource for energy production, irrigation and large-scale infrastructure, its place in collective life has also changed. This has fragmented the landscape and weakened historical continuities through which it acquired cultural meaning. The experience of the Middle São Francisco, however, extends beyond a regional case. Similar tensions can be found in other riverine landscapes where modernisation has altered relationships between environmental systems, urban form and accumulated cultural practices. 

The São Francisco therefore demonstrates the need to understand rivers not solely as natural resources or technical systems, but as historical landscapes shaped through long-standing relationships between environmental processes, settlements and social practices. From this perspective, integrated conservation provides a means of considering the river, its settlements and associated practices as parts of the same historical landscape, rather than as separate objects of protection. Such an approach also contributes to wider debates on the conservation of riverine cultural landscapes, where contemporary demands for energy, water and production must be considered alongside the historical and cultural meanings of the territories they transform. 

The author acknowledges Valter Caldana and Rita Patron (Mackenzie University), Romeo Farinella (Università Degli Studi di Ferrara), Joelmir Marques (Federal University of Pernambuco), Francesca Ubiali (Politecnico di Milano), Hálisson Barbosa, and Regina Xavier (Federal Institute of Bahia) for their intellectual contributions, exchanges, and support throughout the development of this research.