Introduction

As in many other parts of the world, transformation and modernity form an important part of the concept of the city that migrants carry with them when moving to Luanda, the capital of Angola, and that the already established urban dwellers also have of the place they live. Both the common citizen and other urban stakeholders, including policy makers, share the vision of a modern, ever improving city. The fascination with modernity (Mbembe and Nuttal 2004; De Boeck and Plissart 2004; Simone 2011) and the expectations it creates (Ferguson 1999) mark more evidently the recent African urban transformation. The appeal of the ‘city lights’ is likewise an Angolan phenomenon (Monteiro 1973). particularly accentuated in the recent post-war years, generating new sorts of urban fantasies (Cain 2014). new dreams of the urban life (Buire 2014) and new perceptions of the differences between the city proper, the periphery and the rural areas (Roque 2011).

While transforming the urban space and its dynamics, as agents of urban planning or as urban dwellers, the inhabitants of Luanda are also moulded and reconfigured by the city in production. The sociospatial dialectic, which understands space shaping the social as much as the social shapes the spatial, calls for an attentive look at one of the most rapidly transforming African cities today, to the agency of urban stakeholders and the implications of their urban experience. Nevertheless, within this dynamic construction of perceptions, the ‘image’ of the city and of urban life changes throughout distinctive new phases, informing perceptions and values and, consequently, policies and practices (Jenkins 2006). The rapid changes brought to African cities that were successively, within a few years, exposed to independence, civil war, structural adjustment and peacetime reconstruction call for a careful analysis of the present, taking into account the rapid transformation of the images of these cities as well. Paul Jenkins proposes an analysis of Maputo, Mozambique—with a similar recent historical past as Luanda—as a city perceived as civilization in the colonial times, as a parasite after independence and as an engine of development in free-market democracy times (Jenkins 2006). Our approach to the ‘perceptions of Luanda is aligned to this type of periodized analysis as it provides the socioeconomic background of urban realities at each very distinct context. However, as expected, the Luanda case highlights different dominant features for each period.

Based on qualitative research conducted in Angola within a multi-disciplinary research project and on the data from intensive fieldwork in Angolan cities since 2010, particularly in Luanda, the analysis aims to grasp not only the perceptions of the city from the common urban dwellers but also the stated intentions and logics behind the plans and projects for the new modern Luanda. Documental analysis provided a substantial and important set of indications regarding the future envisaged for the city. Over 50 interviews, distributed between the common urban dweller and the different sorts of policy stakeholders,Footnote 1 completed this information and added more refined meanings and perceptions of the urban reconfiguration taking place in Luanda to the analysis. Whenever relevant, some of these accounts are used to illustrate the arguments, though not all material is used given the limitations of the article.

The analysis shows very clearly in the last decades, as the city expanded, the social implications and social perspectives of urban organization and planning. While the colonial segregation of urban space was explicit, the profound and rapid changes that took place in the city after independence did not allow for planning and far less for social approaches to the urban social transformation. However, new exclusions and inclusions began to take shape in the city as the economy and society evolved. With the end of the war and under an explosive increase of investment in the city, the forces of capital, investment and social stratification became more visible, leading to new forms of exclusions and precarity, materialized in particular forms of urban residence and social configurations. The discussion of urban exclusions, central in this article, is concentrated in the final section, comprising both the broader African cities’ scholarly production and the increasing number of new debates about Angola. The article starts, however, with a fundamental background presentation of the context and transformations of urban life in the last decades.

Transformation in the Last Decades: Population, Space and (Possible) Arrangements

The growth of the city of Luanda was regular, albeit limited, since its foundation in 1576 until the late nineteenth century, when significant numbers of people came from Portugal to settle in the city, being rapidly joined by increasing numbers coming from the rural areas (Amaral 1983; Mendes 1988; Mourão 1997; Fonte 2012). However, the most extensive growth of the city, both in terms of area expansion as well as demographically, took place right after independence in 1975 and lasted until the present day (INE 2014).

Schematically, the city went through three major transformative periods in the last century that imposed sociospatial reconfigurations: (i) the colonial spatialization of socioeconomic difference through the consolidation of a dual city, Baixa and Musseque, downtown and uptown—these concepts are further discussed below; (ii) the post-independence massive expansion of the musseques and their imbrication in the colonial city and (iii) the post-war multi-centred and extensive expansion along with the requalification (and gentrification) of the centre.

The physical expansion of the city has always been determined by the arrival of population, occupying either the peripheral areas surrounding the concrete asphalted centre or moving directly into the centre, depending on the historical moment, as will be described.

Up to the twentieth century, colonial settlement was limited and slowly expanding, given the vocation of the city as trading post, initially of the slave trade and from the nineteenth century of commodities from the hinterland. From about 400 inhabitants in 1621, the city grew to only around 4500 in 1810 (Amaral 1962). Thereafter, however, a noticeable growth began to take place, which increases significantly with the replacement of the slave trade by agricultural exports. The coffee boom of the 1950s and 1960s encouraged the establishment of more settlers in the capital, and more services and trade, which was later reinforced with the development of industrial activities in the 1960s and until independence (Mourão 1997; Venâncio 1996). The number of people coming from Europe increased almost three times since the 1930s until the 1970 census (Fig. 1).

Fig. 1
figure 1

Evolution of the population of Luanda (thousands of inhabitants). Source: Africapolis, http://e-geopolis.eu/; Amaral 1983; Angolan census preliminary data (INE 2014)

Full size image

Although post-independence estimates continued to provide an account of the massive growth of the population, which translated noticeably in terms of buildings and the physical extension of the city, data on the population of the capital city are only very recently definitive. The visibly accelerated growth that began in the mid-1980s and that continued with the intensifying civil war that lasted until 2002 could no longer be measured by the extrapolation of data of the last national census that had been conducted in colonial times, in 1970, or of a partial census which was completed in Luanda in 1983.

The 2014 census finally provided accurate figures for the consequences of war and the continuous rural exodus of three decades (INE 2014). With regard to the geographical distribution of the population, the weighting of Luanda regarding the total population of the country needs to be emphasized, a situation that has been even more pronounced between 1991 and 2009, either as a result of the war or as a result of rural migration motivated by the search for better living conditions (Cain 2013; Udelsmann Rodrigues 2007). The distribution of the urban population in Angola is today, a decade after the end of the civil war, remarkable: according to the latest census, 62 % of the Angolan population live in cities, 27 % of the total in the capital city. While the second most populated city, Benguela, now has two million inhabitants, Luanda concentrates 6.5 million. Moreover, the population density of the province is considerably high—347.4 inhabitants per square kilometre—as it refers to the recent extension of the province boundaries to a total of 18,834 km2, including the Quiçama National Park of 9600 km2, some 40 min drive from the centre of Luanda. Considering only the central municipalities, the density is comparable to the world’s largest cities: 23,307/km2 inhabitants in the Cazenga and 18,169/km2 in Luanda. Quickly and roughly comparing demographic figures of urban agglomerations (UN 2014), Paris had a density of 3676 inhabitants/km2 in 2013; London, of 4763/km2; New York had 10,633 inhabitants per square kilometre. In the ranking of the world’s most populated cities, these central districts mentioned could be compared to Mumbai, with 14,505/km2, or Dhaka, with 34,633/ km2.

The demographic factor has therefore been one of the most important determinants of the urban reconfiguration in Luanda. While planning and interventions aimed on several occasions to tackle the explosive growth and the precariousness it inevitably imposes on the city, they also systematically addressed the societal reconfiguration of the city (A1V2 2009; IPGUL 2014). In the colonial period, planning markedly focused on the unbalanced relations and conditions of the urban social strata, the colonizers, and the colonized. After independence, it initially envisaged the dilution of colonial segregation and the minimization of the problems caused by war. Recently, it turns again to the need to address poverty, on the one hand, but also the private investment desires of the wealthy, on the other (Udelsmann Rodrigues 2009).

Colonial Downtown and Uptown

The colonial endeavour in Angola, as in many other parts of Africa, was marked by the brutal imposition of social segregation and by the development of mechanisms for its conservation. Luanda was founded by the Portuguese in 1576 in a particularly privileged spot of the west coast of Africa, a bay morphologically adequate to the defensive demands of the urban nucleus that came to be implanted there. The geography of the location determined the design of Luanda. Since its foundation, the city developed from two distinct nodes, usually called cidade baixa (downtown) and cidade alta (uptown). The downtown, located on a line between the beach and the foot of the hill surrounding it, grew along the seafront and there developed what became its main functions, namely the port activities, commercial and residential. The uphill core became a political-administrative area par excellence, the place where the powers of the Crown and the Church settled (Amaral 1962) as well as the defensive structures (Venâncio 1996).

The slave trade, which was located on the coast, and the activities to which it was attached—navigation, conservation and maintenance of boats, as well as the logistics linked to the guarding and transaction of slaves—and the subsequent ivory trade and later trade of agricultural products made the population of Luanda very diverse from an early age. Apart from the native groups (mestizos and black), it integrated some foreign, mainly European origin traders (Portuguese, Dutch, French and English) and Brazilian traders. In 1730, 150 whites were recorded in the city, and 30 years later, 500 (Mourão 1997). Afterwards, the city initiated a systematic process of growth, as shown above. Over time, the city naturally expanded, increasing the number of civil buildings and starting to link the two city cores, reconfiguring its initial profile (Martins 2005). The city continued to change until the late nineteenth century:

Around 1850, someone wrote that the city had richly assorted shops, with all goods and large trade import and export, witnessing a lot of activity of its inhabitants. In 1824, 680 individuals exercised the most varied forms of business; in 1850 there were 2500 exploring a much diversified urban trade (Amaral 1962: 170).

In the Luanda of the 1900s, teeming commercial activities with some degree of sophistication added to the business of exporting slaves (Martins 1998). The influence of foreigners and the dynamism of commercial activities provided Luanda with a certain degree of cosmopolitanism, despite the epoch and its condition of colony. Wealthy families, mostly mestizo, dominated, and their economic strength was grounded on the slave trade. Many of them also held properties in Luanda and farmed or conceded for farming ‘arimos’ (farms of different sizes) nearby. Among this elite, the figure of Dona Ana Joaquina Santos, a mulatto dealer, stands out, a character of great economic power and political weight, linked to the history of Luanda and symbolizing the power of the mestizo in the society of those days (Martins 2005). Apart from her, there were also other very wealthy mestizo families, influential and educated, and several members of these families occupied intermediate places in the administration and positions linked to economic power and, in many cases, also political positions in the city (Martins 2005). This general picture asserts the relevance of the urban culture by that time, generated by a strong entrepreneurial drive and the heterogeneity of the population.

It is therefore not surprising that the downtown, in addition to the growing number of houses, was gaining the contours of a true business centre, widening and diversifying the set of business activities developed there, contributing to the enrichment and the prestige of local merchant families. While this core expanded into a real mestizo city—because it mingled blacks, whites and mestizos busy with the development of their economic activities rather than with differences in skin colour—the upper town, spatially less extensive and demographically less representative, remained as the religious and political decision centre, hosting a population mostly made up of Portuguese (Amaral 1983; Mendes 1988; Mourão 1997).

The abolition of slavery, decreed by Portugal in 1856, fundamentally changed the economics and consequently the social organization of Luanda. Portugal then tried to strengthen the presence of the Portuguese population in Luanda by sending contingents mostly made up of military, given the fears that Angola could be attached to Brazil, a country from which it depended very strongly and with whom it had very old solid ties (Mourão 1997). The hitherto very influential and economically wealthy slave-trading Luanda bourgeoisie could no longer continue to pursue their traditional business, and this changed dramatically its power and socioeconomic opportunities. More than a new economic cycle, a new period in the history of Angola was inaugurated, most particularly in the history of Luanda (Amaral 1983; Martins 1998). While maintaining its status as the administrative centre of the colony, the city lost the condition of being the largest wealth generator. African traders in Luanda, tightly attached to the slave trade, had some difficulty in adapting to the new economy based on commodities and agricultural exploitation, as well as to the instability of their production cycles and fluctuations in prices (Martins 1993).

At the beginning of the twentieth century, Luanda was a city with a relatively established socioeconomic development. Its population had grown moderately, and in the mid-40s, with the end of World War II, the production of major crops, including coffee (and later sisal), had become vital for the Angolan economy (Oliveira 1963). All this activity had substantial impacts on the economy of Luanda and on the continued attraction to the city of both the Portuguese population and populations from the Angolan provinces.

The increased presence of Portuguese, especially from the 1950s, will change the city’s design and, to a large extent, the relationship between the white, black and mestizo and the possibilities of access to the urban idealized life. Segregation became more than racial difference to be especially marked by the economic capacities of different groups of people, among which blacks and mestizos constituted most of the less fortunate (Amaral 1962). However, again, afterwards, the racial, social and political discriminations became more evident (Martins 1993; Udelsmann Rodrigues 2009). The economic impetus and the increase of the population of Luanda have reflected in the urban area of the city. The low purchasing power of the manual unskilled rural labourers, as well as the cost of city living, socially and residentially sidelined much of this population. The musseques, poor neighbourhoods of precarious homes that have existed in Luanda for centuries (though a precise origin in time is not very well defined, Amaral 1983). continuously accommodated urban growth. These periphery areas integrated both the precarious type of settlement and afterwards the middle classes, when the lands—particularly those owned by the wealthier black and mestizo families—were expropriated to provide space for the needs of urban modernization and management and for the physical expansion of the city. This tension has, among other social effects, nurtured the self-determination emergent ideas and the formation of liberation groups (Martins 2005). while at the same time confirmed the delusions of urban life to some.

Post-independence Expansion of the Musseques In and Out of Town

In the first three decades after independence, the causes of rapid urbanization were fundamentally linked to war and the rural exodus associated with it. In this long period, the population grew exponentially and the city integrated people originating literally from all 18 provinces of Angola. This mass of displaced poor rural people thickened the already predominant precarity of the city, introducing it to central areas as there were spaces left vacant by the Portuguese fleeing to Portugal (Mendes 1988). In 1980, the province of Luanda was divided into Luanda and Icolo and Bengo for administrative reasons, but this did not stop the musseques and bairros continuously expanding to the Icolo and Bengo, Viana or even to the Quiçama Park. The city and its surrounding musseque area began to occupy an area several times larger than the colonial city. This growth in some neighbourhoods turned them gradually into municipalities, which accounted for nine until very recently: Kilamba Kiaxi, Rangel, Maianga, Sambizanga, Viana, Ingombota, Cacuaco, Samba and Cazenga (A1V2 2009). The latest spatial and administrative reorganization of 2011 resulting from Law No. 29/11 (1 September) extended the administrative level of the province, municipalities and commune boundaries to accommodate the increasing expansion. The province now incorporates the natural 9000-km2 park of Quiçama and is organized in seven municipalities: Cacuaco, Belas, Cazenga, Icolo e Bengo, Luanda, Viana and Quiçama. The municipality of Luanda coincides with the former limits of the city of Luanda, integrating the old municipalities of Ingombota, Kilamba Kiaxi, Viana, Rangel, Samba and Sambizanga, which became urban districts (IPGUL 2014). The new political-administrative division, compared with the previous, not only shows the great expansion in terms of area that the city recorded but the demographic growth as well (Fig. 2).

Fig. 2
figure 2

New administrative division of Luanda. Source: Development Workshop, 2013

Full size image

As mentioned above, following independence, there were few or practically no investments in the construction of the city on the part of the government that channelled all resources into the civil war. However, the city was erected by the urban dwellers who spontaneously and informally built their houses (Development Workshop 2011). The result was a vast surrounding periphery with precarious housing, poor or inexistent infrastructural conditions and a mixed population of wealthier and poorer residents who settled on the available land on the outskirts of the safety of Luanda (Cain 2013). In the centre, the city became a confused mixture of people and housing of all types, hosting the poor and the wealthy in the same spaces, somehow trying to match the post-independence socialist ideologies. Throughout these years, co-residence made the inhabitants of Luanda subject to similar infrastructural conditions, precarious to almost all of them (Mendes 1988).

‘In the early 1980s, we lived in the newly built Cuban buildings but many of our Angolan colleagues, engineers or teachers, lived in the musseques’ (interview in Luanda, male Portuguese aid-worker [cooperante], June 2014).

While the first years of this period were characterized by socialist politics and economics, in 1991, the country initiated the passage to a market economy and preparation for the first elections of 1992. While the elections failed, returning the country to civil conflict until 2002, the economy evolved to the free market and to a booming informal economy. This created the conditions for the emergence of new wealthy entrepreneurs, especially in the informal economy. An important part of this urban informal economy became focused precisely on real estate: with the war, little new construction was taking place while at the same time major international businesses—particularly the oil industry—sought housing in Luanda and were willing to pay high rents.

‘With time, the city started to run down. The big firms, particularly the big construction firms—Portuguese, Italians, French, Americans or the Dutch—built the first villages for their workers’ (interview in Luanda, June 2014, Brazilian female cooperante in the 1980s).

These residential reconfigurations initiated again the transformation of the city and its spatial composition, with the centre slowly being re-occupied by the wealthy and the poor slowly moving again to the peripheries, succumbing to the appeal of the real estate market (Udelsmann Rodrigues 2009; Buire 2014; Cain 2013; Croese 2012; Gastrow 2013/14). Reconfigurations within the new socioeconomic context represented then, to an increasing number of urban dwellers, new delusions with the urban life and the expectations created, particularly among the massive number of migrants seeking the city safety and opportunities. In the meantime, the political and economic transformation was accompanied by an opening of the country to other international partners, which brought along new influences, both culturally and socially, as often stated in the interviews, and particularly new conditions for urban economic transformation.

Post-war Multi-centred and Extensive Expansion

Since the peace in 2002, urban growth has been fuelled by the continuing exodus of the rural population in search of better conditions and opportunities of life plus a significant natural growth. The city continuously outspread to the peripheries, and additionally, from 2002 on, it became even more attractive to private investment.

The new administrative divisions of the province of Luanda of 2011, with the integration of Icolo e Bengo (headquarters in Catete) and Quiçama and the concentration of the former nine municipalities into seven, ended up conceding a much wider extension to the limits and area of the province. The vast special economic zone (ZEE, Zona Económica Especial), which crosses the municipalities of Icolo e Bengo, Viana and Luanda, is now part of the broader city limits. This extension also made it clear that the expansion will continue exponentially beyond what people would think the city could grow to. Even in the low-density Quiçama natural park, which occupies 72 % of the territory of the new provincial area, new urban features are likely to be planned and implemented, although it is still considered a reservation area dedicated to sustainable rural development, tourism and leisure until 2030 (Decree 59/11, page 1576).

In 2008, the number of new constructions in Luanda had boomed: over 576,000 houses were licensed (54 % of the total) while nearly 440,000 infrastructure projects for trade and services were registered (41 %), covering an area of 1,063,678 m2. In the following year, this number had decreased but was still maintained at very high levels: over 4000 housing projects representing 72 % of the total new constructions, covering 569,585 m2 (IPGUL 2014; A1V2 2009). This new construction, registered by the provincial authorities, is radically different from the spontaneous informal ventures of the post-independence period, although much self-construction is still being done, as a strategy to access the improved expected urban life. Occupations of land with improvised tin-can lodging, particularly in areas where construction projects are being set, are a very relevant dynamic within the recent expansion of Luanda, revealing the lasting tensions between the formal and informal projects, the new city being built and the increased limitations to have access to it. Beyond the (less represented) reasons for (re-)settlement due to natural causes, the majority of new settlements in non-built areas are made by occupants who perceive an opportunity in the compensations from the state or from private builders (Croese 2013; Gastrow 2014).

‘Most tin-can houses are a way to mark the ground in areas to be urbanized just waiting to get compensation for an illegal occupation’ (Interview in Luanda, male 45, April 2013).

The recent development of Luanda and the major works are state-led, including the reconstructions of the waterfront, The Marginal, and of the central business district, the re-qualifications and the new centralities, targeting both the low-income populations and/or resettled populations (Croese 2013; Cain 2014). The re-entry of the state into the provision of housing reveals the shift from relatively unplanned and so-called chaotic growth (of the 1990s and early 2000s) to large-scale urban regeneration and development planning driven by the state. In the private sphere, peace, coupled with strong domestic and foreign investment in the capital and the emergence of a more distinct middle class, gave a strong impetus to the construction of new buildings. Luanda is now a city of multiplying numbers of ‘gated’ mid/upper class condominiums—which had already begun to appear in the 1990s/2000s—and of new ‘centralities’, i.e., brand new entire towns of tall buildings, for the middle and lower classes, and new neighbourhoods of the ‘million houses’ (Croese 2012) project targeting the poor and the resettled population.

The Rationale of Urban Policies: The Modern Bright City

Successively, urban planning has been manifestly behind time regarding the growth of the population. Although the early colonial settlement, primarily oriented towards trading, did not attract a large population for centuries, the settlement waves of the 1950s and 1960s of colonists coming from Portugal and migrants arriving from the hinterlands, rapidly and unpredictably—i.e., not foreseen either in colonial policies or by their administrators—changed the aspect and functioning of the city calling for informal precarious urban solutions that materialized in the musseques, in the vast slums.

Still during the colonial period, the first indications that a planned expansion of a ‘modern’ city was needed were the preparation of several urban plans with the objective of the organization of the city and the regulation of the expansive ‘informal’ market-driven occupation of the peripheries by the population of the central districts. The government of Portugal had for this reason established in 1944 the Office of Colonial Urbanization (Gabinete de Urbanização Colonial)—later called Overseas Office of Urbanization—that took charge of the urban planning of the colonial cities.

Regarding urban planning properly, the first plans for Luanda were made in 1942 by Étienne de Groer and David Moreira da Silva. Already by this time, the proposal for the creation of five new satellite cities, new dormitory centralities and the transfer of the population to them was a solution envisaged to control the growth and congestion that the city was already experiencing and address urban development (Fonte 2012). As this was not approved, a whole set of new proposals for the Luanda urban plan were made and successively not approved: in 1947, 1949, 1952 and 1956. All of them followed the 1942 rationale of the satellite cities promoting better urban life and relief of urban congestion. Despite the modernist approach of the plans, these still maintained the distinction between first and second class residential areas, racially and socially distinct. Even the early 1960s plans like the Bairro Prenda, foreseeing racial integration (Venâncio 2013) and an international demonstration of integrational policies by Portugal (Mendes et al. 2005). was planned with buildings for the white population and for the black population.

Despite some partial studies and actual interventions for some locations in the city, such as the project for the bay, the Kinaxixi market, the Finance and the Public Works’ buildings, the Port of Luanda and the bank of Angola, the main characteristic of urban planning of that time was that the proposed plans were systematically not approved. Instead, the implementation of projects has always followed the perception of the diverse stakeholders at each occasion of what was best. Only in 1961 was a plan finally approved, also based on the core ideas of the Groer-Silva proposal, despite not being fully implemented. New plans were again prepared between 1970 and 1973, and the new plan of 1974, elaborated by the firm Omnium Technique d’Aménagement, included a more coherent and developed proposal for a satellite city, an evaluation of the possibilities for legalizing most of the musseques and the renovation of the downtown.

But independence in the next year and the new social ideologies prevented this plan from being pursued, not only because it envisaged the separation of the musseques from the rest of the city but also because realistically the country was at war and the massive exodus had begun with masses coming from the warring countryside flocking to the safety of the capital city. Self-construction became the norm and practice at the beginning of the 1980s, despite some scattered government initiatives like the creation of the Musseque Rehabilitation Office (Gabinete para Reabilitação dos Musseques, GARM)Footnote 2 in 1987 or the much later 1999 LUPP—Luanda Urban Poverty Programme. Despite the evident increasing precarity, the city maintained, however, its capacity to attract the population, fundamentally representing a safe haven but also being a place where the few opportunities in wartime existed.

At government level, prominent drafting of urban action began especially after the 2002 peace agreements. The city, seen as a reborn phoenix, a renewed location for development, became ‘a construction site’ in the words of the government, and reconstruction is foreseen to continue for many years. The first large-scale post-independence housing programme of the government in Luanda was the Nova Vida project, which started 1 year before the end of the war and was completed in 2005. In 2008, the government announced a new more ambitious project, the National Program on Housing and Urban Development (PNUH), which included the construction of One Million Houses by 2012, projects for ‘new centralities’ or urban sprawl, urban renewal projects and infrastructure in some areas. Some of these projects are aimed at state employees in particular, but also the poor. The most important international support to these ends is the China International Fund (CIF) targeting the construction and reconstruction of the country, which was launched in 2006 and foresees the construction of 215,500 houses across the country, with a higher percentage proportionally in Luanda.

In organizational terms, the Office of National Reconstruction (GRN) was created with direct connection to the presidency and to the Institute for Urban Planning and Management of Luanda (IPGUL) and also to other localized structures of urban planning and intervention, such as the Office for the Reconversion of Cazenga and Sambizanga (Gabinete Técnico de Reconversão Urbana do Cazenga e Sambizanga), right in the heart of the city. In terms of overall planning, the city was the subject of an Urban Growth Management Plan in 2000, which later in 2011 materialized in a law (Presidential Decree 59/11 of 1 April) laying the foundation of the Integrated Plans of Urban Expansion and Infra-structure of Luanda and Bengo. However, the production of a true Master Plan of Luanda only began in 2013 (Metropolitan General Director Plan of the City of Luanda, PDGMCL), with plans to have the first phase and main guidelines completed in 2015, extending its conclusion to 2030, in an evolutionary manner adapted to the conditions and urban context. The general orientation is for a multi-centred expansive city, able to accommodate the population of over six million (Fig. 3).

Fig. 3
figure 3

Recent urban orientations for Luanda. Source: Google maps, based on A1V2 (2009)

Full size image

As shown, today, Luanda is following a decentralized multi-polarized plan of urban expansion in such a way that the multiplicity of typologies, simultaneously social, historical and infrastructural, are increasingly more difficult to define and appear somehow intertwined in the urban tissue. One urban study that was recently conducted in Luanda showed precisely the multiple factors at stake for classifying the multiple cumulative urban processes of the city (Development Workshop 2011). Both the pre-independence quarters and areas of the city—i.e., with over 35 years—and the post-war new typologies translate the socioeconomic differentiation of the city. The former include the Old Urban Centre and the surrounding bairros: the Musseques Ordenados, the Old Musseques and Bairro Popular. As mentioned, while in the colonial times the distinction of the Centre from the musseques clearly identified the socioeconomic differentiation, the intertwining of new urban expansion with sociospatial reconfigurations in the last decades have made many of the old musseques gentrified areas. The single mid-way typology identified by the study was the post-independence Transitional Musseques, which in general followed the ‘gentrification’ path of the colonial musseques as the city expanded. The post-war typologies, on the other hand, arise with new construction booming and continue to indicate the socioeconomic composition of their inhabitants: New Suburbs (Condominiums), Social Housing Zones and Owner-built on Planned Sites (see also Cain 2014; Buire 2014).

Luanda is therefore today a patchwork of urban typologies, including vast areas of rural settlements within the new administrative arrangements. The most notorious types of new residential concepts are the private upper-class condominiums, the centralities—brand new cities for the middle classes, of which Kilamba stands out in Luanda—and the resettlement bairros for the lower classes, all of them located on the outskirts of the city. The perception of social and spatial differences and the way they inform urban planning since colonial times reproduces the conditions for exclusions, despite the envisaged development and improvement of urban life. Consequently, the urban life projected and imagined by the urban resident is increasingly conformed in designated areas through the urban renewal. Despite the variety of residential situations today, the trend towards urban improvement, gentrification and modernization is often repeated by the national level stakeholders and by the common urban citizen of different types: ‘there [in the centre] there is more development; but it will come to this area soon’ (male farmer, 46 years old, inhabitant of a rural peripheral bairro).

The Urban Change Seen as Improvement and Experienced as Exclusion

Despite the optimism crossing both the perspectives of policy makers and of the common urban resident regarding the ‘new’ city being (re)built, the dissonances between the ‘expectations of modernity’ and real life in the city persist (Ferguson 1999; Simone 2011). Both in colonial times and during the war, disappointment with the city and the harsh conditions, especially for the poor, has been a constant realization (Monteiro 1973; Mendes 1988). and so it remains today. Poverty affects a large number of people in the city, although national poverty is higher in the countryside (INE 2011). Buire, in a 2014 article, provides examples of the contraposition between the ‘dream’ and the ‘ordinary’ in two types of neighbourhoods in Luanda—the Panguila relocation settlement and the new centrality of Kilamba—showing how, despite perceiving the order and organization in these new locations as ideal imaginaries of the urban life, the inhabitants engage in important efforts to re-initiate life in these new locations, with implications to their daily lives (Buire 2014). Modernity and ‘the urban fantasies’ (Cain 2014) are right at hand but still difficult to access for the majority.

‘I left my job in Luanda and decided to become an independent business woman. I applied for a house in Kilamba and got it. I like it here but it was 2 h to get to the center in the morning and 2 h to get back. I had to make an option between living here or in a house with no conditions in the center’ (female resident in Kilamba Kiaxi, 35 years old, interview in January 2015).

Nevertheless, while the dissonance within these processes of adaption to the new city may be seen as conflicting logics—the techno-managerial and modernizing systems of state planning versus the marginalized poor trying to survive—the reality is that they do not invalidate each other. The idea of modernity and improvement is shared by the poor, the rich and by the urban policy leaders and technicians. On the other hand, informality and getting by within complex and precarious social and economic environments is a well-known part of the continuous reconfiguration of the city: ‘people in the city have to struggle more for survival’ (male street vendor, 28 years old, inhabitant of peripheral bairro). The current reality of daily life today, beyond the glamour of the shiny urban plans and the brand new constructions, is still one of precariousness for the majority.

While the creation of new residential solutions is perceived both by the government and by the private sector as a way of improving living in the city, life in the new neighbourhoods is confronted with many new challenges and the inhabitants recognize these difficulties. These are mainly related to the lack of infrastructure or transportation (see Power 2012). but in general the new developments are considered as ‘a good solution for the shortage of housing in Luanda and also because it is possible to access credit’ (F., civil society key-informant, 55 years old, resident in Kilamba). On other occasions, the modernity envisaged by the urban planners and promoters does not correspond entirely to the urban dwellers’ perspectives regarding the urban way of life in the new urban modalities. While technical studies have, for instance, shown that the new residents of the Nova Vida quarter are generally satisfied with their new homes (Lara and Bekker 2012). details concerning house space and accessibilities in the new neighbourhoods are not always adapted to the needs and ways of living in the city.

‘Tables, chairs, sofa sets do not go through the door [of new houses]. It is just showing-off that the government gave a person a house (I., male, 43 years old, resident in new peripheral bairro of Belas, July 2014).

Also, tensions between the perceptions of housing as a social right and housing as marketable asset are quite evident in Luanda, as would be expected in the present-day scenario of rapid changes. These affect and erode relations between the state and, for instance, the evicted citizens, and the grievances are often voiced (Gastrow 2013/2014). In sum, in the process of reconstruction of the city, the areas where tension and conflict are actually taking place or potentially can occur are varied. The responses to these dissonances are therefore constantly reformulated.

Global cities of the south such as Luanda do not necessarily follow a similar trajectory as those of the West, and the local responses here are too complex to fit segregation or polarization (Shatkin 2007). The reconfiguration of the city comprises multiple types of relocations, forced and voluntary, self-exclusions in upper-class bairros or resettlements in lower-class locations, demanding adapted solutions on the part of the urban dweller. While efforts of social (upward) self-segregation demand major investments in housing in the condominiums and keeping up with a certain lifestyle, displacement caused forcefully by evictions from the centre or areas where major projects are being implemented call for new survival strategies: ‘they have confined us here; we have to find other solutions’ (resident of the peripheral bairro Zango III, university teacher, 60 years old); ‘One day they came and told us we had to move from Chicala to Zango because our house was in a place that had problems. When we got there, the house had no window-glasses and there was a snake inside’ (woman, 29 years old, domestic cleaner).

Recent urban reconfigurations then call for multiple responses, most of these addressing the new exclusions being constructed. Selling property located in the centre or taking compensation to move into areas further away from the centre had become a common urban strategy even before the end of the war, in the late 1990s, with consequences to livelihoods as resources, and markets were shown to be less favourable in the peripheries, at least in the beginning of resettlement. The scheme has long been detected, and ways of narrowing the possibilities for fraud are increasingly more creative.

‘The sobas [local traditional authority] know better than anyone else who the people of their land are and who came to take advantage of a situation’ (A. C., administrator of the municipality of Icolo e Bengo, IPGUL 2014: 5).

Other strategies closely related to urban expansion and reconfiguration include investments in real estate, at a modest scale or involving higher resources from private middle-class investors. Resiliently, new economic opportunities are being created in the new neighbourhoods, including commerce, domestic services and transportation, at smaller or larger scales, depending on the location and capacity of the new urban entrepreneurs. More specifically, the new urban strategies are adapted to the new urban realities, in the wealthier, the poorer and in the middle-class areas.

In what regards the new sociospatial reconfigurations, in the centre of the city, and particularly in the above-mentioned central districts of Cazenga, Luanda and Sambizanga, the state tone is to reconvert, which is accompanied by private investment leading to gentrification, in the Angolan style. While gentrification was primarily postulated as a process of ‘invasion’ of lower-class quarters—in the original British study, the ‘working classes’—by the middle-classes (Glass 1964). the process is no longer restricted to certain parts of the developed world but has rather become global, interconnected to the circuits of global capital (Smith 2002). The spreading out to countries such as Angola calls for a consideration of which classes are being replaced in the centre as the dynamics of the urban growth were, as previously described, due to massive displacements related to conflict and not due to industrialization. In fact, the non-industrial character of African urban growth is not Angolan-specific (Potts and Bryceson 2006a, b). But what marks the originality of the process in Luanda is that it not only mobilizes significant financial resources from the state but also very significant private investment. The result is an increasing number of poor succumbing to the temptation of central land prices and to the imagined improvement in living standards in the new resettlement peripheries. To some, the main consequences are not only the renewal of the inner city but also the fact that ‘the poor [are] increasingly alienated from their assets’ (Cain 2013).

‘The elite thought that Luanda as it was did not suit their ambitions of an “international” place. In fact, Luanda has many “first world” spaces but right next to them miserable places’ (F., male architect, 42 years old, interview in Luanda in June 2014).

However, new adaptations are, as mentioned, taking place, leading not only to new ways of living in the city but also to new social reconfigurations: ‘My parents had a hard time adapting when they moved to Zango. But now, as they have a small plot, my mother already started to grow vegetables and flowers. They no longer want to come back to the city; one day I will move to Zango too’ (woman, 29 years old, domestic cleaner).

In the areas surrounding the centre, the realities are, as mentioned, mixed. The co-existence of the wealthy segregated spaces of middle-class gated communities and ghettos of poverty is not, however, unanimously accepted. By ‘investing in core areas of the city that are already affluent in order to demonstrate global strength for both the outside observer and internal elite (to encourage further investment), existing segregation is deepened’ (Lemanski 2007: 450; see also Beall et al. 2002). In Luanda, despite the multi-centred character of the new city being built, the tensions between socially differentiated residential spaces are acquiring new contours. The reconfigurations of spaces and society involve different degrees of sociospatial repositioning: ‘leaving Vila Alice [in the centre] and moving to Kilamba means worsening your location; some middle-class people that lived in the centre have also ended up being placed in Zango’ (F., civil society key-informant, 55 years old). As mentioned in interviews (inhabitant of the centre, student, 28 years old), the generalized perception that ‘before, rich and poor lived together but not anymore, with the new condominiums being built’ becomes increasingly more well-defined to the vast majority of urban inhabitants. ‘It is not the common low-class citizen that lives in Talatona or Luanda Sul’ (D., inhabitant of periphery, driver, 35 years old, April 2014). The average US$600,000 four-bedroom villas of the condominiums of Luanda—which require a 30 % deposit and the remaining balance within 8 to 10 months afterwards—are accessible only to the upper strata of Angolan society and foreign companies with investments in Luanda. ‘The population that tries to escape the monthly rental in the city centre comes further to the peripheries, to so far rural bairros like Tanque or Quenguela’ (group discussion, Tanque bairro, July 2014). Objectively, social difference materializes through the differentiated building of the city and, simultaneously, is reconstructed in the minds and perceptions of urban life.

Considerations on the Attraction of City Lights and the Dangers of Exclusion

Like other urban contexts of the world, Luanda has through the years attracted a significant number of inhabitants and activities, motivated by the actual objective living standards and by the subjective perspectives of urban modern life. While the individual life courses and perspectives vary, a few commonalities persist throughout the years, and this article has devoted particular attention to the expectations of the improved urban life as they do not fit the realities of everyday life lived by a significant portion of the urban population. The way the city evolved, transformed and reconfigured along the years is both a result of state logics and of those of urban dwellers, both based on notions of improved lives and modernity, depending on the changing socioeconomic conditions and integrated in a process of sociospatial interactions. In this article, the broad description of the contextual conditions that gave rise to the development of a modernizing urban planning of Luanda is combined with accounts of the perceptions not only of this idea of the city but also of the delusions of exclusions.

From the small coastal nucleus to the twenty-first century Luanda, there is an organizational continuity, based on the sociospatial differences. Since independence and with a massive growth over the 30 years of civil war, the city has been targeted by multiple planning and reconstruction initiatives, also continuously based on perceptions of modernization and development. This is leading again today the city to important spatial reconfigurations, articulated as modernity and improvement but often experienced as continued precarity. While this dissonance has been a constant along the years and constitutes an important feature of the evolution of the city, the forms it assumes today are noteworthy because investment in the city has never been so high. The significant wealth of the country is being channelled into urban investment in an unprecedented way, accompanied by foreign and national private investments and even by the small enterprises in this area created by the average urban dweller (de Oliveira 2015). Poverty and precarity, however, continue to be the evident reality for the majority of the urban population and residence in the city alone does not eliminate the excluding conditions, as shown in this paper. While the situations are varied, the discrepancy between the imagined bright lights of the city and the precariousness of daily living calls for multiple responses. The ongoing construction and reconfiguration of the city is therefore still difficult to define entirely, although signs of new exclusions—including self-exclusions—become more evident in some particular cases, as in the upper-class condominiums, in the low-class resettlement bairros or in the middle-class new centralities.