INTRODUCTION: MIDDLE CLASS HOUSES
Tanzania is a nation under construction. From the steel and glass skyscrapers reconfiguring Dar es Salaam's skyline to the piles of sand and concrete balusters hoarded on domestic plots, building is everywhere. The city skyscrapers that proclaim Tanzania's recent economic growth are striking, but it is at the domestic scale that this construction is most remarkable. New suburbs stretching up to 30 km from Dar es Salaam's city centre are dotted with two-storey pastel-coloured villas interspersed with ‘not-yet houses’ (Melly Reference Melly2010), floor plans with walls and no roofs, jutting out of the ground like molars. In the countryside, new houses incorporate new architectural features and materials such as pillared concrete verandas, mirrored windows and aluminium roofing sheets made to look like red, green or blue tiles. In both urban and rural areas people are investing in building and improving their houses using new designs and materials and filling their interiors with new kinds of goods from new places. As they re-make the built landscape, these house-builders signal the growth of Tanzania's middle classes.
Class is back on the agenda in discussions of Africa, but it is the apparent emergence of the middle that is generating interest (Banerjee & Duflo Reference Banerjee and Duflo2008; Birdsall Reference Birdsall2010; Ravallion Reference Ravallion2010). Influential business and consumer market reports produced by the African Development Bank and McKinsey, among others, claim that a third of the continent's population now qualifies as middle class, using a $2–20 per capita daily expenditure definition (Leke et al. Reference Leke, Lund, Roxborough and van Wamelen2010; Ncube et al. Reference Ncube, Lufumpa and Kayizzi-Mugerwa2011), although the proportion is smaller if the ‘floating class' with a daily per capita expenditure of $2–$4 is treated separately. In 2010, 5% of the continent's population was rich (per capita daily expenditure of more than $20), 13% middle class ($4–20 a day), 21% floating class ($2–4 a day) and 61% poor (less than $2 a day) (Ncube et al. Reference Ncube, Lufumpa and Kayizzi-Mugerwa2011). Such reports assume a generalised ‘African’ middle class, even though other members of the new global middle classes are often described at the national scale (the ‘Chinese’ middle class, or the ‘Indian’ middle class, for example). The African middle class is characterised first by income, and second (if at all) by a set of assumed universal middle class traits including choices and behaviours in relation to nuclear families, education, entrepreneurship, politics and consumption patterns that include bigger houses and cars, refrigerators, computers and mobile telephones (Ncube et al. Reference Ncube, Lufumpa and Kayizzi-Mugerwa2011). Celebrated as a product of relatively high national GDP growth rates, poverty reduction efforts, the resource boom, conflict resolution and improved macroeconomic and business environments, Africa's middle class is also expected to play a role in shoring up recent macroeconomic and democratic gains. As urban consumers they are projected to spend an increasing proportion of household income on non-food items, with spending on housing alone predicted to grow by 4.5% between 2008 and 2020, ahead of healthcare, telecommunications, banking and education (Roxburgh et al. Reference Roxburgh, Dörr, Leke, Tazi-Riffi, van Wemelen, Lund, Chironga, Alatovik, Atkins, Tefous and Zeino-Mahmalat2010; The Economist 2012).
The representation of Africa's growing middle class as a homogeneous urbanised group with predictable economic, political and consumer behaviours is problematic when we know so little about them. Measuring incomes, occupations and consumption patterns is important (Visagie & Posel Reference Visagie and Posel2013), yet ultimately limited if we are to understand what middle class lifestyles are like and how they are changing social landscapes (Heiman et al. Reference Heiman, Freeman and Liechty2012). This is particularly so in places where data on income and employment is inadequate. In this paper I suggest that we can begin to gain insight into the social changes accompanying middle class growth by focusing on the house as a key site where lifestyles are visibly in flux. Drawing on a small sample of the new styles of houses being built in suburban Dar es Salaam and Machame, Kilimanjaro, I outline four middle class domestic aesthetics: respectable, local aspirational, global aspirational and minimalist. Each is characterised by distinctive uses of architecture, materials, interiors, décor and compound space. These differentiated domestic aesthetics reveal the limits of assuming an ‘African middle class’. Houses have become part of the cultural capital that can be deployed to demarcate middle class distinction in post-socialist Tanzania (Bourdieu Reference Bourdieu1984), but they also provide new ways in which the ‘old middle class’ and ‘new middle class’ seek to distinguish themselves from each other as well as from their ‘class Others’ above and below them (Liechty Reference Liechty2003). It is not my aim here to provide a dynamic model of contemporary class relations that can account for the emergence of the middle class, which would require national data on employment and asset ownership, and a close analysis of the social, economic and political interests underpinning recent struggles, particularly over land. Rather, the more modest aim is to reconsider recent claims about the ‘African middle class’ in the light of a particular practice in a specific place. It is widely recognised that as a social scientific category the middle class is notoriously slippery, defined neither by its relation to the means of production nor to capital-owning employers (Wacquant Reference Wacquant, McNall, Levine and Fantasia1991). One way of dealing with this is to focus on the social and cultural practices through which the middle classes produce and reproduce themselves (Liechty Reference Liechty2003; Fernandes and Heller Reference Fernandes and Heller2006). Building on Weber's (Reference Weber, Gerth and Wright Mills1948) insistence on the importance of lifestyle to status groups, and Bourdieu's (Reference Bourdieu1984) reformulation of social class as constituted by economic and cultural capital, houses can be understood as a pre-eminent site for the everyday reproduction of middle class distinction. This means that the house cannot be treated as a black box that indexes a standard set of middle class traits. Rather we need to go ‘over the threshold’ and into the house itself to consider the everyday ways in which middle class domestic spaces and lifestyles are made in a particular time and place (Miller Reference Miller2001).
THE MISSING MIDDLE
In Tanzania the growth of a distinct class situated between a poor majority and a very small elite has gathered pace since economic and political liberalisation in the mid-1980s. National growth rates averaged 6.7% between 2008 and 2012 (National Bureau of Statistics 2012). A cursory unpicking of this headline figure reveals the polarised nature of wealth distribution in the country, in which per capita consumption between 2000 and 2007 increased for every wealth quintileFootnote 1 except the poorest, for which it dropped by 2% (Research and Analysis Working Group 2009: 144). On the other hand data on absolute poverty show that the proportion of the population living below the poverty line fell from 38% in 2000/1 to 33% in 2007 (Research and Analysis Working Group 2009: 147). While the problems of structural poverty have rightly generated considerable description and analysis, there is also something going on in the middle that merits more attention.
Sectors where the middle classes congregate have experienced growth in the post-socialist period, including the public sector, the civil society sector and the private sector (World Bank 2012; Kelsall Reference Kelsall2013). During ujamaa government employment was the most reliable route to middle classness, particularly in rural areas. But with 63% of the country's employed workforce estimated to be in the private sector by 2011 (National Bureau of Statistics 2012), this may be changing. Over the last two decades formal and informal employment opportunities for the minority have opened up in mining, telecommunications, tourism and construction, as well as in other sectors such as NGOs, new consumer services (mobile phones, shops, supermarkets), financial products (banking, insurance) and education (teaching, administration) (World Bank 2012). Small businesses have also flourished in transportation and trading within Tanzania, across East Africa and beyond. These sectors have been central to the accumulation strategies of the ruling elite, a government and business class made up of high-ranking politicians and top civil servants who control and distribute considerable state resources, and private sector capitalists or the ‘new financial elite’ (Ellis Reference Ellis2011). This elite owns the means of production such as transport, machinery and land, employs a considerable workforce and invests in mining, infrastructure and construction. But it is not only the ruling elite that has benefited from growth in these sectors. Ellis (Reference Ellis2011) notes the emergence of a middle class in Africa distinguished by the increased consumption of imported goods and new ways of making and managing money using the savings, loans and mortgage products offered by the banking sector. In Tanzania the creation of a market for financial products is the most recent in a wave of changing practices around making and spending money that were set in motion by liberalisation. In the post-ujamaa period there is now more opportunity for, and less opprobrium attached to, the visible pursuit of individual distinction through lifestyle and consumption choices centred on the house, the car, the nuclear family and the body.
Published scholarship on contemporary Africa offers surprisingly few resources with which to analyse this changing social landscape, and the middle classes have been notably absent from recent published scholarship on Africa outside of South Africa (Southall Reference Southall2004; Seekings & Nattrass Reference Seekings and Nattrass2005; Posel Reference Posel2010; cf Spronk Reference Spronk2012). The class debates which dominated African Studies in the 1960s and 1970s were preoccupied by analyses of a political economic landscape polarised between two classes, the peasantry and the bourgeoisie, who were constituted by their relations to the means of production (Bernstein Reference Bernstein1977; Sklar Reference Sklar1979). Late colonial attempts to create a rural middle class through targeted assistance to ‘progressive farmers’, particularly in the upland coffee- and tea-growing areas of Bukoba, Kilimanjaro and Rungwe (Iliffe Reference Iliffe1979) did generate a class of rich peasants, or petite bourgeoisie, but they were understood as struggling for inclusion in a local or national bourgeoisie (Raikes Reference Raikes1978; Samoff Reference Samoff1979). Indeed what exercised many of Tanzania's radical commentators after the Arusha Declaration of 1967 was the dissonance between Nyerere's (Reference Nyerere1966) claims to a classless society and the evidence of rural and urban class formation, the antecedents of which were laid down during the colonial period and exacerbated by the post-colonial government. For those radical commentators the central class dynamic was the exploitation of the peasantry by a ‘bureaucratic bourgeoisie’ (Shivji Reference Shivji1976) that shaped socialist policies on nationalisation, import controls and cooperatives to suit its own interests.
Beyond political economy the preoccupation with those at both the top and the bottom of the social hierarchy is also evident in much social historical and anthropological research on Africa. In these accounts the language of class is often eschewed in favour of ‘elites’, who commonly do not own the means of production but exercise some control through their position in the state or in traditional power structures. Elite status may also be generated through education, migration, age and wisdom (Lentz Reference Lentz1994; Werbner Reference Werbner2004). However ‘elite’ and ‘middle class’ are not interchangeable terms. Elites are usually understood as members of a ‘ruling elite’ rather than situated somewhere in the middle. Sara Berry's (Reference Berry1985) work on accumulation and class formation in colonial Yorubaland, for example, traces the trajectories of those who benefited from mission education and entered the post-colonial ruling elite rather than a middle class.
Two studies of Zimbabwe are notable for bucking this trend. West (Reference West2002) and Burke (Reference Burke1996) analyse the ways in which a middle class emerged in colonial Zimbabwe by accumulating economic and cultural capital. They draw attention to the importance of consumption and lifestyle, highlighting the social aspirations and consumer practices in relation to particular goods that became the hallmarks of distinction among the urban middle classes. For West, the middle class were not necessarily distinguished by wealth or their relation to the means of production, but rather constituted those who ‘shared similar personal and social experiences in the areas of education, occupations, family life, consumption habits, residential patterns, and organizational affiliations’ (2002: 2). Spronk's (Reference Spronk2012) work in contemporary Kenya follows a similar approach, demonstrating how lifestyle and sexuality practices among young professionals in Nairobi tie middle classness to preferences in leisure and forms of work, particularly in the private sector. Research in southern Africa has also demonstrated that ideas about social status, and middle classness in particular, is often associated with respectable lifestyles as demonstrated through personal conduct, education, jobs and homes, much more than with income or power (Ginsburg Reference Ginsburg1996; Lee Reference Lee2005; Bolt Reference Bolt2010).
These studies indicate the significance of work, leisure, consumption and the home to what makes the middle classes across Africa. In Tanzania the meaning of class must also be understood in the context of the political ideologies of the socialist and post-socialist eras and the changing moral values through which accumulation and consumption are perceived. This can be glimpsed through the shifting vernacular of social differentiation. For example, class relations in 1960s Dar es Salaam were popularly conveyed in terms of the kabwela (poor worker) who was exploited by the naizi (an urban class of civil servants, traders and landlords; Ivaska Reference Ivaska2005; Brennan Reference Brennan2012). The moral judgements embedded in these terms indicate the significance during the ujamaa period of ideas about the legitimacy of hard work for the common good and the illegitimacy of accumulation and sloth. These came to be the moral hallmarks of ujamaa, associated with the leadership code of the 1967 Arusha Declaration that forbade holders of public office from private accumulation. In practice the undifferentiated figure of the mkulima (farmer) around whom ujamaa was built jarred with the emergence of rich farmers. Yet despite growing social differentiation, conspicuous consumption was not common during the ujamaa period. Evidence of accumulation was routinely hidden (Raikes Reference Raikes1978; Moore Reference Moore1998).
With the relaxation of import controls from the mid-1980s consumption practices across the social spectrum began to change. Conspicuous consumption of imported goods became the new mark of distinction among the elite and an aspiration for the poor and working classes, who could easily achieve ‘middle class style’ through the purchase of cheap imported clothes (Brennan & Burton Reference Brennan, Burton, Brennan, Burton and Lawi2007: 63). Moyer (Reference Moyer2003: 49) suggests that by the mid-1990s a ‘reconfiguration of class’ had taken place in Dar es Salaam in which the hegemony enjoyed by the urban petite bourgeoisie during ujamaa had been usurped by a new elite class. Popularly known as wakubwa (‘the big ones’), the recent growth of this class and its hallmark fast accumulation is widely understood by Tanzanians to be a result of the proliferation of opportunities for corruption and witchcraft since President Mwinyi came to power in 1985. Yet the petite bourgeoisie, made up of educated civil servants, professionals and intellectuals, remained unnamed in the vernacular, perhaps because they were widely considered to be the ‘unmarked carriers of the moral status quo’ (Moyer Reference Moyer2003: 61). The middle class in Dar es Salaam has then been characterised as a residual respectable class wedged between the elite and the poor that mourns the loss of ujamaa. Beyond these insights, Tanzania's emergent middle class has so far escaped academic scrutiny. Africanist scholarship in general has been better at understanding the rich and the poor. Now almost three decades into liberalisation, the ranks of the middle classes have grown and morphed, shaped by both socialist and post-socialist eras. Middle classness is no longer tethered to respectability and ujamaa.
DOMESTIC ARCHITECTURE AND AESTHETICS IN DAR ES SALAAM AND MACHAME
The discussion that follows is based on a small sample of the new style of houses that have been built in Tanzania over the last fifteen years. Eighteen houses were studied; ten built in Machame, located in the densely populated coffee–banana belt on the lower slopes of Mount Kilimanjaro, and eight built by urban migrants from Machame now living in suburban Dar es Salaam.Footnote 2 Most were built by men as family residences, several were a joint project between husband and wife, two were maintained by widows and one was being built by a single man. The diversity of the house-builders is significant and underlines the weakness of standard measures of the middle class that rely on income and occupation. Some were established members of the old middle class on the basis of their long-standing senior positions in the public or private sectors. Others had more recently joined the ranks of the middle classes through success in business or self-employment such as building and trade. These house-builders constitute Tanzania's new middle class. Some are graduates of Tanzanian or overseas institutions, others are primary school leavers. All are owners of the means of production since they own land or property through kinship relations, and many have bought more land and property through the market and employ labour on it. Indeed, what unites all of these house-builders is that they own land in a context of land scarcity. Land scarcity and rising prices are not new in Tanzania (Kombe Reference Kombe2000; Brennan & Burton Reference Brennan, Burton, Brennan, Burton and Lawi2007) although competition for land is increasing (Askew et al. Reference Askew, Maganga and Odgaard2013). As a result there is now greater differentiation between those who own land and those who do not (Moore Reference Moore1986; Owens Reference Owens2004), and in this context houses are significant because they shore up claims to land ownership. In Dar es Salaam, building helps to prevent land invasions; in Machame a house is an ‘improvement’ which could bolster a claim to land ownership in the event of a dispute (Moore Reference Moore1986). Houses have become part of middle class investment portfolios (Briggs & Mwamfupe Reference Briggs and Mwamfupe2000; Owens Reference Owens2010) that are now spread across land, housing and children's education.
The new house styles built by the middle classes are the latest in a series of changes in domestic architecture in Tanzania since at least the start of the twentieth century (Kyhn Reference Kyhn1984; Lewinson Reference Lewinson2006). On Kilimanjaro for example, architecture has changed from compounds of conical huts made of banana palms, each with a specialised function (for men, women and cattle), to simple mud-brick two-room bungalows containing a bedroom and living room, to larger extended variations of single-storey bungalows built in permanent materials. Houses have grown in size and rooms with specialised uses have proliferated, including the use of internal rooms for activities that used to be done outside in the compound. New houses are more likely to have internal kitchens and bathrooms, and a sitting room furnished for leisure use (large comfy chairs, TV, stereo). Windows are generally larger in new house styles because they are now used to give light to rooms, not just ventilation. The use of architecture to achieve distinction is also not new. In the 1950s the urban elite commissioned architect-designed homes (Leslie Reference Leslie1963) while in Kilimanjaro they built permanent houses with wooden ceiling boards, decorative verandas and wooden windows with glass panes. Today's ideas and materials come from an even more globalised market. Locally made windows and doors using Tanzanian hardwoods are being replaced by imported tinted windows and PVC mouldings, and roofing materials are imported from Dubai, China, Malaysia, New Zealand and South Africa. Bricks and blocks are still made locally with sand, mud and cement, although cement is also imported to meet demand. Most house-builders do not employ architects but prefer to use local builders who become important agents in the spread of certain designs and techniques. Despite builders' attempts to distinguish their houses from those of their neighbours by incorporating different styles and materials such as classical columns, Arabic doors, European chimneys, curved walls and arches, and ‘complicated roofs’, the fact that these are mediated through local builders and craftsmen means that the design and decoration of new houses is strikingly similar.
Below I discuss some of these houses and their builders according to four domestic aesthetics. I use them heuristically and recognise that not all houses fit one type. While the relative size and quality of all the houses indicate the middle classness of their builders, the emphasis is on differences in taste, values and budget as evidenced in the architecture, interiors and compound spaces of the houses. The different domestic aesthetics provide one way in which the differences between Tanzania's old middle class and new middle class, with their different sources of income, consumption patterns and lifestyles, are being made.
The respectable house
The respectable house is a bungalow, that global architectural form introduced by British colonial officials to West Africa in the 1880s (King Reference King1984). Built from the 1950s by Tanzania's elite, it became the preferred style of the old middle class during the ujamaa period and has endured, despite inevitable changes, into the present. Today's respectable bungalows still reference the domestic aesthetic of ujamaa. These are modest houses, sparsely furnished with locally made furniture, and functionally decorated. The materiality of the house and its contents are of less value to inhabitants than the zero-grazed cattle, poultry, coffee and vegetables kept outside.
JosephFootnote 3 started to build his house in Machame in 1954 when he was a teacher at the local Lutheran church. He inherited the land from his father, who had acquired it when he moved down the mountain from a village higher up to establish a Lutheran church. He started out with three rooms for himself and his wife (living, dining and sleeping), gradually extending the house to the current nine rooms as children arrived. In contrast Godwin's house in Kimara Temboni, a relatively prosperous suburb approximately 25 km west of the centre of Dar es Salaam, was begun in the early 2000s. An accountant at the local office of a large international car manufacturer, Godwin had moved to Dar es Salaam from Machame in 1982 after finishing secondary school in Old Moshi, living first near the airport before moving to his more desirable current plot. The house accommodated Godwin, his wife and their four children. Godwin was keen to point out that the house was ‘not yet finished’ since they had planned to lay floor tiles for some time but had not got round to it – in fact they had just finished paving the large compound. For both Joseph and Godwin, these houses represented one in a series of investments in land and property. Joseph had only built one house but had invested in several plots of land in the lowlands below Machame and nearer to Moshi town for his sons to inherit, while Godwin owned farmland in Kibaha (in the neighbouring Coast Region), a small house that he had built on inherited patrilineal land in Machame, and the suburban compound in Kimara Temboni. He and his wife planned to retire to Kibaha and to build anew there. The current house, and the house in Machame, would both be rented out.
Both houses were single storey bungalows built with locally made sand and cement blocks, a pitched roof of local timber and corrugated iron and windows made of timber frames with glass panes and iron grilles. Both had a covered veranda at the entrance to the house with concrete flooring and decorative bricks on which a few chairs or a bench were placed to receive guests. The houses sat in large compounds demarcated by bougainvillea and masaleFootnote 4 or a concrete and corrugated iron fence. Both Joseph's and Godwin's compounds provided a driveway with parking space for a car and a small pick-up truck while a path led to the house through a well-tended garden consisting of a lawn, flowers and trees. Outbuildings in the compound included structures for livestock (zero-grazed cattle, chickens) animal feed, a store, an external kitchen for cooking with charcoal, and small sleeping quarters for domestic employees. Godwin had provided rooms for two youths, one from a neighbour's family in Machame and the other from Pare, who took care of the cows and chickens in the compound. Since the area was unsurveyed with no government services he had also installed his own tanks for potable water, sewage and livestock. Joseph's compound in Machame was connected to the local potable water supply and also made use of furrow water. His house was flanked by two modern houses built by sons to whom he had allocated small parcels of patrilineal land. Joseph's compound (kihamba) led to a large and productive banana–coffee grove (shamba) the size of which he declined to estimate but which was nevertheless one of the larger plots in the village (where average holdings range from one quarter to four acres). Family graves are also usually located in the shamba.
Inside, these respectable houses were decorated according to a functional aesthetic, as investment in children's education or income-generating projects were prioritised over interior decor and possessions. The paint on the walls and ceilings was tired and items rarely replaced or updated. The largest room was the sitting/dining area from which the other rooms (kitchen, bathroom, up to four bedrooms) extended down a corridor. Joseph's and Godwin's interiors were furnished with locally made items in dark, heavy wood. The centrepiece of their sitting rooms was a large sofa and chair set covered in brown or black patterned heavy fabric and swaddled in foam padding to maximise comfort. The sofa and chairs were ranged against the walls around a large low wooden table flanked by ‘stools’ (small moveable tables for serving drinks to guests at their seat), and oriented inwards towards a television, housed in a wooden unit at one end of the room that included space for a satellite receiving box, loose files and papers, and framed photographs of nuclear family members. Heavy wooden bureaus displaying crockery and glassware indicated hospitality. Decorations included clocks, calendars produced by local businesses or parastatals (usually an employer and often several years out of date). Joseph's wife had bought several mass-produced prints of Jesus Christ and of Asian children accompanied by English sayings at a popular market in Moshi town. Soft furnishings included net and heavy curtains in plain colours hung at windows, plastic flowers in small vases on tables, and vitambaa, the crocheted, knitted and appliqued squares placed on sofa arms and backs that extended the longevity of the furniture as well as signalling a home managed by a respectable woman active in the local church women's group. Family photographs were an important feature, demonstrating respectability and success by showing members of the nuclear family at key life stages such as baptism, graduation and marriage. On display in Godwin's sitting room were photographs of his eldest son and daughter graduating from the universities of Dar es Salaam and Dodoma respectively, and a framed certificate given to his wife on completion of an income-generation course in animal husbandry led by a local NGO with international funding. They had just constructed a modern chicken coop for 3,000 chicks in the compound. ‘It is all for school fees’, explained Godwin, whose two younger children were still at boarding school.
Respectable houses like Joseph's and Godwin's are the most common housing style among the urban and rural middle classes in Dar es Salaam and Machame. They dominate because houses take a long time to build, so any house started between the 1980s and the late 1990s is likely to conform to this model, and even older houses such as Joseph's share the same interior aesthetic. However, houses that have been built over the last ten years by new members of the urban and rural middle classes look quite different. There is a shift away from the respectable and functional aesthetic preferred by the old middle class towards house styles that evidence a distinct new consciousness about architecture, interiors and compounds. These houses demonstrate distinction from the homes of the poor and of the old middle class through the command of globally produced consumer goods. They have become sites of consumption and leisure rather than production.
The local aspirational house
The local aspirational house is very similar to the respectable house except that it is in the process of being upgraded in terms of style and function, and is often smaller. Emmanuel, a builder in Machame, had begun constructing his house in his parents' compound in 2003 when he was just 17. He raised the money by growing tomatoes to sell locally and in Dar es Salaam. He started with just two rooms (sitting and bedroom) and a veranda, while slowly collecting the materials necessary to extend the house. First he built the kitchen, currently used as a store, and a second bedroom, later adding a third bedroom and an inside toilet. The sitting room was the only completed room in terms of furnishing and decoration, and it was here that Emmanuel's aspirational vision was most evident. Walls were freshly painted, the ceiling decorated with moulded gypsum, and furnishings included soft chairs, a low table covered with vitambaa and plastic flowers, a small TV and hospitality cabinet. All other rooms were in process, not yet furnished or even in use. Additional rooms for anticipated children were used as stores for farm produce or equipment. The bathroom and toilet were yet to be plastered, fitted out and plumbed in, because there was no water supply to the house. In the kitchen unplastered spaces waited for power sockets and kitchen cabinets, while upturned empty oil drums served as kitchen worktops. The majority of the cooking, washing and cleaning went on outside the house in Emmanuel's parents' compound.
The aspirational character of the house was also evident in new architectural features not commonly found in respectable houses. The style and shape of Emmanuel's roof was made of the new roofing materials and more ‘complicated’ than the plainer pitched roof style, with a ridge vent that gave it a distinctive shape, the gate slightly more elaborate, and the concrete balusters decorating the veranda replaced by sleeker decorative arches. The windows were bigger than those found in older style houses, with tinted windows rather than clear glass.
Overall the local aspirational aesthetic indicates a shift in taste and values that turned (parts of) the house into a site of display. There was an effort to match interior design and furniture and enthusiastic use of the newest imported materials such as tinted windows and new paint colours. Aspiration was evident in the attempt to distinguish the house from neighbours by choosing a different kind of flooring (tiles, lino or concrete) or a piece of furniture such as a large dining table and matching chairs.
The global aspirational house
Global aspirational houses were immediately striking in architectural terms. Often built as a two-storey bungalow – a real mark of distinction in a context where single storey housing is the norm – global aspirational houses were patched together from a range of architectural styles associated with different times and places. Architectural plans were sourced from abroad (Canada, Switzerland, Dubai), new features were incorporated that nodded to other places such as rural England (bricks and chimneys), the Middle East (Arabic window and door frames) or Zanzibar (Zanzibari doors), or to other eras such as classical Rome. Small verandas decorated with painted concrete balusters were replaced by grander spaces flanked by small staircases and classical style columns. Roofs mixed pitched and hipped shapes that rose to a peak in the middle like a small mountain range, and were made of imported sheets of aluminium, zinc and galvanised iron coated with granulated PVC in bright colours (red, green, blue) to make them look like tiles and to distinguish them from the usual plain (and inevitably rusting) iron and aluminium corrugated sheets. Chimneys were decorative. External walls were painted in one of the newly available colours (pink, apricot, green). Windows were tinted, large and covered with custom-designed iron grilles.
The interiors of global aspirational houses had complicated floor plans with open-plan spaces, many rooms, staircases and corridors. The hub of the house was a large open-plan sitting area connected to other rooms by archways. Kitchens and bathrooms were plumbed, powered and used. Kitchens used a variety of cooking methods depending on the food being cooked: gas and electricity stoves for foods requiring short cooking times while staples such as maize, bananas and beans were cooked on a charcoal stove inside or outside the house. While only the sitting room was decorated in respectable and local aspirational houses, in the global aspirational house every room received attention. Rooms had individual colour schemes, extensive use was made of local hardwoods for door frames, built-in wardrobes and ceiling boards, and floors were laid with shiny tiles imported from Kenya, India, Dubai or China.
External space was oriented more towards consumption and leisure than towards production. In both Machame and Dar es Salaam value and wealth in the Chagga homestead for much of the twentieth century was associated with land and the amount of people, bananas, coffee (in Machame), and zero-grazed livestock that could be accommodated and made productive on it. Global aspirational houses supplemented this with new values attached to the material culture of the house and compound, introducing new ideas about how to use compound space such as felling coffee and banana trees to make way for large houses and lawns, decorative gardens and paved driveways. This was an aesthetic choice, but it was rooted in changing investment strategies among the middle classes as coffee production on the mountain increasingly became the preserve of the land-rich. Simon, a tailor and farmer in Machame explained this shift in his own family, ‘I have four children who have all been schooled and got higher education. I advise them – I have no shamba to give them. Go to Moshi or Arusha or Morogoro. There is no space here! No space for livestock keeping!’ The shift away from coffee and banana production on the patrilineal shamba towards new investment strategies centred on education, land and property away from the mountain had become common among the middle classes.
The houses built and furnished by the globally aspirant new middle classes are thus quite different from the respectable houses of the old middle class. Their choices about furniture and decor index both their membership in a global consumer class and their social and economic distance from the poor. Their interiors were quite literally full of stuff, producing a domestic aesthetic at odds with the principles and values of the ujamaa period. But even within this group of house-builders there were important differences in terms of the values, tastes and lifestyles that their houses accommodated.
Young entrepreneurs like Mr and Mrs Masawe in Machame had a taste for keenly priced items of imported furniture, electronics and decoration. They embraced the global circulation of consumer goods in their own home since these were the very items that generated their own recent wealth. Mr Masawe owned several businesses including a building materials and hardware shop in Arusha and an agricultural shop in Mbeya, sourcing much of the stock himself on biannual trips to Asia during which he and a colleague filled a shipping container between them. He preferred to ‘buy cars and spare parts in Hong Kong, and laptops in China’ rather than trading with locally based Asian merchants (for example in Kariakoo), because ‘what they sell there is poor quality, they only want to make a big profit’. He also sourced furniture, lighting, building materials, clothes and veterinary goods in China, Dubai and Thailand. Explaining his market, he said, ‘people want sofas and tables. They want curtains. They want lights like these [from Dubai]…a sofa from China is not cheap, but people want to keep up with the times.’ Mrs Masawe's businesses included a shop in the local market that sold clothes imported by her husband, cake-making and event decorating (for weddings and baptisms), vitambaa tailoring and small-scale animal husbandry (chickens and two dairy cows). Her newest plan was to acquire cake moulds from Dar es Salaam to make tiered wedding cakes. Mr Masawe started building their house in 2008 when he inherited his late father's kihamba. It was almost finished by 2012. The house was a single-storey bungalow distinguished by imported building materials such as ‘tiles from China [that are] better quality – they are first class', and popular furnishings including a sofa set and table from China and lights from Dubai. The ceiling and cornicing were decorated with elaborate patterns in gypsum. Electronic goods (television, DVD player, computer) on display in the sitting room were imported from Dubai. There was not a single piece of locally made furniture in the Masawe's sitting room. Imported items such as the engraved glass-topped table nested in between the large sofas were not only cheaper to import than to buy locally due to the expense of wood, but as Mr Masawe explained, ‘you can't get it here in Tanzania’ because local craftsmen cannot (yet) make etched glass tables.
The Masawes valued the imported over the domestically produced. In other global aspirational homes these values were reversed. These tended to be the homes of older, publicly employed and more educated men and women who preferred what they considered to be quality domestic goods. Mama Thomas, a widow retired from a senior civil service post, lived with her son Thomas in Mbezi Beach, a prosperous planned suburb 20 km north of Dar es Salaam city centre. She and her late husband (also a civil servant) had bought the land for farming at a very low price in the 1970s, but were compelled to build on it in the 1980s when it became a planned residential area. The banana trees behind the house had recently been felled to make room for 2,000 egg-laying chickens and for Thomas to build his own house. Mama Thomas and her husband built the large two-storey house modelled on an architectural design acquired by her husband from Switzerland, and eventually moved into it from their government-owned flat in Upanga close to the city centre, which Mama Thomas had found noisy and lacking in privacy.
Mama Thomas and her son eschewed the imported furniture and consumer goods favoured by other globally aspirant middle classes as cheap and of poor quality, preferring items they considered to be of high quality such as bespoke furniture made by known craftspeople using local hardwoods. Imported, mass-produced items were dismissed as ‘cheap stuff from China that doesn't last’. Thomas demonstrated the corrosive effect of the humid coastal air on a TV stand and a lamp they had brought back from a trip to visit his sister in the UK, before dismissing his mother's most recent purchase, an imported golden Arabian-style lampshade hung over the breakfast table. ‘Let's see how long this lasts!’ he said.
In their preference for domestically produced goods the aesthetic choices of Mama Thomas and her son echoed the values of ujamaa. There was pride in patronising skilled local craftspeople and in local materials that could withstand the climate. These values distanced them from the Masawes and their taste for imported consumer goods, but their taste for quality and quantity also distinguished them from the respectable and the local aspirational middle classes.
The minimalist house
The minimalist house was the rarest of the four types, and was championed by younger, educated people recently returned from postgraduate study abroad. It was typically found in the city, where international migrants were more likely to set up home on return. This style was also distinguished by size, but this time the surprise of a modest house in a neighbourhood where large houses predominate.
Sarah and Joshua had built such a house in Mbezi Beach. Sarah had studied for her degree in Nairobi while Joshua had lived and studied for his postgraduate degree in the UK. Since his return to Dar es Salaam in 2006 they had developed a building project management business. On the large one acre plot they had purchased they had built a practical house using the latest high-quality and cost-effective building materials such as locally made interlocking bricks. Their house was a single-storey bungalow structure with a simple pitched roof of tiles, PVC guttering, large grilled PVC windows and doors. It sat within a large plot but had a much smaller footprint, containing a large central sitting and dining area leading to three bedrooms and an open-plan kitchen.
The aesthetic of Sarah and Joshua's house was modest, durable and appropriate for the local climate. The design and materials were chosen for their cost-effectiveness and ease of maintenance: for example, refusing to put in en suite bathrooms minimised housework. The interior decoration echoed this approach, with walls painted in complementary shades of the new colour palette, and a minimum of decorative objects in the sitting room – no vitambaa, soft toys, plastic flowers or mass-produced Biblical prints. Instead Sarah and Joshua had a taste for mass-produced modern art bought while travelling abroad. Furniture was minimalist with plain, low-slung corner-unit sofas in neutral fabric facing low shelving units that housed a TV and small framed family photographs.
The house had some of the trappings of respectable and global aspirational houses such as an entrance veranda, satellite dish, a large paved external area, and a small garden that Sarah had cultivated with a few banana palms from her mother's rural homestead, some vegetables and fruit trees. Commenting on the importance of her garden, Sarah said ‘we [Tanzanians] are going to have to get used to the idea of living in apartments – the change will come’ with increasing land scarcity. But for the moment, she said, ‘people want to build a house – and to have a garden’. In other words, building a house is bound up with owning and cultivating land.
DOMESTIC ANXIETIES
Tanzania's middle classes may have built houses characterised by different lifestyles and values that suggest intra-class differences, but they were all haunted by the prospect that their middle classness was transitory (Bourdieu Reference Bourdieu1984; Liechty Reference Liechty2003; Heiman et al. Reference Heiman, Freeman and Liechty2012). This anxiety was captured in the very materiality of the house. Building was a complicated process requiring the careful orchestration of resources, labour, money and materials over long time frames. Houses thus illuminated the precarious nature of claims to middle classness in contemporary Tanzania. They had as much to do with anxiety and failure as with the accomplishment of middle class membership. A house could serve as a very effective reminder of what one was not, of what one had failed to do, as much as what one had achieved (Miller Reference Miller2010).
Constructing a house was stressful and time-consuming. It involved acquiring land and the associated documents, accumulating enough cash to start building the foundation, and supervising labourers. Some learned the hard way, as did Juma, a young regional sales manager for a large telecommunications company in Dar es Salaam. He paid local labourers to manage and build his two-storey house in Kunduchi Salasala, a new planned suburb 23 km from the city centre, before going overseas to study for his Masters degree. While abroad he was called by his caretaker cousin and informed that the house was literally cracking up. The labourers had bought fewer bags of cement than required, mixing in more sand and low-quality aggregrate and pocketing the difference. On his return Juma had to demolish an entire section of the house and start again from the foundations, this time buying the materials and supervising the labourers himself. Most house-builders thus acted as their own architect and project-manager, supervising every stage of the building and finishing process. The stress could go on for years, even decades, since construction relied on cash flow rather than credit. Many houses stood half-built for long periods of time while their builders accumulated enough money to proceed. Inflation rates, which reached 20% at the start of 2012, also delayed building. Large, ambitious projects were likely to run into problems once the foundation was laid or the walls constructed, since large foundations could cost tens of thousands of dollars. From the suburbs of Dar es Salaam to those of Machame, the built landscape was littered with ‘not-yet-houses’ (Melly Reference Melly2010), reminding the builder and his or her neighbours of their failure to accomplish the key indicator of middle class status. Houses could be too big to complete.
Even when a house was completed the inhabitants did not necessarily live the middle class lifestyle anticipated by the business and policy press, or indeed by the builders themselves. The Masawes in Machame, for example, had spent a lot of money and effort to create a sitting room that reflected their membership in a global consumer class. Interiors such as theirs might suggest increased leisure time spent in the private space of the home. In practice however, the Masawes rarely used the sitting room. As Mr Masawe explained, ‘in the evenings when we are tired we just come in, put on the TV and sit here [a large unfinished corridor between the sitting room and kitchen] before going to bed. The sitting room is more of a space for guests.’ Yet domestic spaces were rarely entered by those outside the nuclear family. Interiors were therefore more about self-projection than they were about demonstrating wealth and status to others, even if only one room projected the desired subjectivity (Clarke Reference Clarke and Miller2001). Not-yet-kitchens and not-yet-bathrooms, as in Emmanuel's house, also indicated aspirations for middle class membership. Because these were usually the last rooms to be completed, families lived inside for a long time while using outside kitchens, toilets and washing areas. Since the act of building took so long anticipation was central to the building experience: waiting for sufficient money to complete a building, for family growth, and for goods and services to become available at affordable prices. Not-yet-kitchens and bathrooms signal the optimistic future-orientation among the aspiring middle classes.
Others chose not to live in the house they had built.Footnote 5 This was a particular problem for house-builders with plots on the very edges of Dar es Salaam's expanding suburbs, who spent so much time commuting to the city centre that they were quite literally never at home. Public sector workers got around this problem by staying in their government-owned flats in the city centre. Other examples of the apparent contradiction between the built form of the house and the lifestyle of its inhabitants included houses with insufficient electrical wiring or no water supply, and houses decorated with shiny floor tiles that caused residents to slip and injure themselves in the rainy season. Tension over land and property either between neighbours or within the family was also common. In Dar es Salaam's suburbs many house-builders maximised their space by building right up to the plot boundary wall, contravening planning laws that provide for a metre gap between plots. In Machame, the most common problems dealt with by the Village Land Committee concerned disputes over boundaries and inheritance. The politics of land in Tanzania is not confined to the middle classes and has received much attention (Brockington Reference Brockington2002; Benjaminsen et al. Reference Benjaminsen, Maganga and Moshi2009; Gardner Reference Gardner2013; Askew et al. Reference Askew, Maganga and Odgaard2013), but land is just as politicised among and between families (Moore Reference Moore1986).
CONCLUSIONS
Tanzania's house-builders are changing the country's built and social landscapes. The new styles of houses utilise new materials that are globally sourced and incorporate new design ideas about how to configure, decorate and use inside and outside spaces. They are significant because they constitute new means of social differentiation between the middle classes and those above and below them. Furthermore, the four domestic aesthetics outlined here reveal differences between the lifestyles, values and consumption habits of the old and new middle classes in the post-socialist era. Given these differences, attempts to characterise an ‘African middle class’ on the basis of income, aspirations, values and consumption patterns, such as those outlined at the outset, seem unconvincing. Even if the middle classes can be identified in terms of relative income, it does not necessarily follow that they will share values and lifestyles. Contrary to the expectations heaped on Africa's urban middle class consumers, Tanzania's middle class house-builders emerge as a disparate, and not exclusively urban, group. The similarities in house-building between suburban Machame and suburban Dar es Salaam remind us that class formation takes place in the countryside as well as in the city – indeed it is often a process that straddles both (Berry Reference Berry1985).
The new house styles described here can be understood as a response to the social, political and economic changes that have taken place in Tanzania since at least the first half of the twentieth century and which have accelerated in the post-socialist period. House-building is an attempt to fix value and values and to protect wealth and status in uncertain times. But houses often fail to stabilise social position. The complex materiality and scale of the house project renders any attempt to fix wealth and status through building as inherently precarious. Middle class lifestyles are often imagined by those who aspire to them as relatively comfortable and secure, yet the experience for many of Tanzania's middle classes is one of anxiety, thus amplifying the need to demonstrate distinction from one's ‘class Others' (Liechty Reference Liechty2003).
The process of house-building is characterised by striking continuities with the socialist period. During ujamaa there was a state and popular narrative about the morality of accumulation and consumption. New house styles index some quite stable referents from this time concerning self-reliance, morality and respectability, and the importance of family, nation and religion. This is not to say that the pursuit of distinction through housing is new in the post-ujamaa period. It has certainly been the practice on Kilimanjaro since people began to build permanent domestic structures in the mid-twentieth century. By the end of the 1980s the houses of the local middle class in Machame were noteworthy for their size, exterior decoration, well-furnished parlours and agricultural innovation. Neighbours still judge each other today on the quality of their livestock pens as much as on the quality of their veranda.
What has changed is the frequency of the new house styles in both the city and the countryside. There are now more rooms and objects inside houses, suggesting a retreat into the private and specialised spaces of the house. There is a new consciousness about interiors as spaces that reflect personal judgement and taste, achievements and aspirations. Most of the domestic material culture in Tanzania was locally made until the 1990s, but the new buildings and interiors are now patched together from a globalised shopping mall, a trend that is reflected in Tanzania's trade deficit with countries such as China. The contradiction is that, despite this apparent globality, and with the exception of the minimalist house, most interiors are filled with the same kinds of things. The distinguishing factor is simply how much of it can be afforded.
House-building takes place within a broader construction boom in Tanzania that has attracted plenty of debate about buildings and the morality of accumulation. Building is seen as an efficient way to spend cash quickly. Both public and private buildings are consistently linked to high-profile people in business and politics and the corruption scandals they are rumoured to be connected to. But, given most people's everyday experience of not having enough money, the construction boom seems particularly vexing. In the suburbs, new houses make visible the outcomes of increasing inequality. In everyday interactions people avoid drawing attention to what they have for fear that it will be coveted, perhaps malevolently, by others, particularly friends and family members. Houses make public claims about status, wealth and achievement, all of which are known to be unequally distributed in Tanzania, but which are nevertheless underplayed or unspoken in public. Hidden behind large walls, houses materialise this very ambivalence about Tanzania's post-socialist trajectory.
ACKNOWLEDGEMENTS
The author would like to thank Peter Bofin, Hazel Gray, Emma Hunter, Aikande Kwayu, Ben Page and two anonymous referees for their comments on an earlier draft of this paper, and the London School of Economics which provided research funding.
