At the city edge
Situating peripheries research in South Africa and Ethiopia
in African cities and collaborative futures

This chapter examines the operationalising of research focused on understanding how transformation in the spatial peripheries of South African cities and an Ethiopian city is shaped, governed and experienced. We discuss both intellectual and methodological challenges and insights of undertaking the research which has at its core a desire to understand the dynamics and drivers of change and the ‘lived experiences’ of residents living on the peripheries of cities, using a mixed qualitative methods approach. We reflect upon and propose conceptualisations concerning terms such as periphery, ‘drivers of change’ and what or whose lived experiences are captured or can be known. In doing so we point to preliminary findings and consider issues of comparability and differentials in data depth and coverage. The chapter concludes by highlighting the richness of researching the peripheries.


This chapter explores how transformation in the spatial peripheries of three African city regions is shaped, governed and experienced, drawing on the findings of a three-year Economic and Social Research Council/National Research Foundation (ESRC/NRF) funded research project in South Africa and Ethiopia. We discuss both intellectual and methodological challenges, along with reflective insights of undertaking research on the dynamics and drivers of change and the ‘lived experiences’ of residents living on the peripheries of cities, using a mixed methods approach.

The chapter emphasises practices of collaborative research and reflects on our conceptualisations of terms such as ‘periphery’ and ‘drivers of change’. It then turns to question the process of knowledge production, examining what or whose lived experiences are captured or can be known. In doing so it briefly points to some initial findings of the project, but its aim is not to address these findings substantively, rather to explore the conceptual and practical challenges of undertaking research in African city peripheries. This exploration is extended through a reflection on various methodological issues, including issues of comparability and managing differentials in data depth and coverage. It concludes by highlighting the richness of researching the peripheries and the wider impacts of the work for both academic agendas and policy contexts.

Introducing mixed methods in urban peripheries research

The key research questions of the project were as follows:

  • What are the main characteristics of economic transformation and infrastructural change?
  • Who has driven these changes, and what is their significance for inclusive urban development?
  • What governance mechanisms shape these infrastructural and economic changes?
  • How are these urban changes experienced by different residents of urban peripheries?
  • What are the comparative lived experiences in urban peripheries?
  • What are the implications of such drivers and experiences of urban change for reducing urban poverty and improving urban inclusion?

The project aimed to address and answer these questions through the use of a mixed methods approach which coheres around a focus on seven case studies representing different kinds of urban peripheries across African cities. Specifically, our cases included places where older and new low- and middle-income areas are close to major new areas of infrastructure investment (Waterloo/Hammond's Farm/Verulam near to King Shaka Airport, eThekwini in Durban); a traditional authority area close to areas of formal residential, commercial property and shopping mall investment (Molweni/Crestholme close to Hillcrest, eThekwini: see Figure 2.1 for Molweni); a ‘mega-human settlement’, major social housing, lower-middle income private development and vulnerable informal settlement on the edge of Soweto, Johannesburg (Lufhereng/Protea Glen/Waterworks); places with a history of ‘displaced urbanisation’ and relocations linked to apartheid homeland policy (Winterveld, Tshwane) and apartheid-originated industrial decentralisation (Ekangala/Ekindustria/Bronkhorstspruit, Tshwane) where ‘new cities’ or mega-human settlements are planned.

Figure 2.1 Molweni, eThekwini

In the city of Addis Ababa, our two cases are Tulu Dimtu and Yeka Abado/Legetafo. In both cases there is large condominium housing investment by the state. Tulu Dimtu is a new residential area of predominantly condominium and cooperative housing located in the south-eastern edge of Addis, relatively near to an industrial park, a site of significant employment straddling the municipal boundary between Addis Ababa and Oromia State (see Figure 2.2). Yeka Abado is a relatively established centre, located in the north-east of the city, adjacent to the boundary with Oromia State and quite close to the eastern end of the city's new light railway. On the other side of the boundary but adjoining Yeka Abado is Legetafo, an area of high-end private residential villas abutted by rural farmland.

Figure 2.2 Tulu Dimtu, Addis Ababa

Though the choice of case studies was by no means straightforward, a key feature of most of the cases is their multi-nodal nature. This means they are highly differentiated both in their composition and between cases: for example, a single case may include areas of informal housing, formal middle-class housing and state-subsidised housing for the urban poor.

The project used diverse methods and activities in order to gather data. It has primarily adopted a mixed qualitative methods approach, underpinned by ideas of comparative urbanism on the one hand and a commitment to seeing the peripheries from the ‘everyday’ perspectives of those who live within them, on the other. The research activities encompassed solicited diaries, auto-photography and interviews with residents in case study sites, accompanied by surveys of a sample of residents in these sites. In addition it included key informant interviews with government officials, planners, business representatives, developers and leaders. This field research was in turn underpinned by close virtual and physical communications between team members across three countries with joint activities such as workshops ensuring a regular dialogue across the project and strong collaboration.

This chapter now moves to consider two conceptual issues that underpin the overarching aims of the project, namely how we conceptualise the periphery and what we understand ‘drivers of change’ to mean. It concludes with some reflections on methodological challenges and opportunities that have shaped the project.

Conceptualising the periphery

Globally, scholars have made various efforts to make sense of and conceptualise the peripheries of cities. These have informed the conceptualisations and framings that we detail below, but have also proved limited in adequately capturing the complexities evident in diverse African peripheries. Literature on peri-urban change (e.g. Mbiba and Huchzermeyer, 2002) has suffered from allegations of limited theoretical clout, though it does continue to provide important insights into particular forms of land use and tenure changes on the edges of some African cities. Nonetheless, the nature of changes increasingly documented, particularly in the African context, is evidently more complex and varied. The now extensive literature (McKee and McKee, 2004; Phelps, 2012; Stern and Marsh, 1997) on edge city development, originating in North America (Garreau, 1991) but now applied globally, is arguably a poor lens through which to explore transformations on the edges of African cities. This is because of its assumptions about the pivotal role of private vehicle ownership, private-sector investment and the peripheralisation of economic opportunities and commercial functions. Many of these are poorly evidenced in African peripheries. However, given its foundational nature, we do speak to this literature but – cognisant of the limitations of this work – we argue that theorisations from a contextualised southern perspective are thus essential (Parnell and Robinson, 2012; Watson, 2014). Our work aims to contribute to this growing scholarship through our analyses of lived experiences of the peripheries to inform grounded, contextualised understandings of complex urban change.

Beyond conceptualisations of ‘urban edges’, literature on city regions is important in pointing to the spatial complexity of these places, particularly their economic multi-nodality, multi-directional movement patterns and multiple governance arrangements, which complicate concepts of peripherality. However, this literature has not paid much attention to African cities (Beall et al., 2015). Recent research on ‘global suburbanisms’ (Ekers et al., 2012), understood as ‘the combination of non-central population and economic growth with urban spatial expansion’ (Ekers et al., 2012: 407) has documented the diverse ways in which this growth is occurring internationally. It argues that ‘suburbanisation’ is now the dominant form of urban development globally (Keil, 2018). This research has included some work on African cities. Bloch (2011) has pointed to rapid growth on the urban periphery linked to economic expansion and the rise of the middle class, while Andreason et al. (2017) argue that informal residential development on cheaper land on the periphery might be seen as a form of suburbanisation in Dar es Salaam. Both describe quite particular forms of peripheral development. Mabin et al. (2013) review African literatures and demonstrate the diversity of forms of growth in African cities, and the need for deeper exploration of these patterns. Work in this realm is limited, however (Buire, 2014; Jenkins, 2013), particularly in relation to governance mechanisms, different forms of infrastructural investment and how these explicitly shape the lives of residents. Partly tied to this agenda and focusing specifically on the urban periphery is work by Sawyer (2014). She examines the differentiated nature of change across the peripheries of Nigerian cities, noting the piecemeal ways in which urbanisation proceeds. This important work is highly relevant to our understandings and framings of the periphery but evidences a raft of processes that are less familiar or commonplace in both South Africa and Ethiopia. In these contexts, state coordination is more evident, although we can recognise that the extent of planning or state intervention is ‘loosening’ in the South African cases relative to that witnessed during the apartheid era (Harrison and Todes, 2015).

We explored the work of Caldeira (2017) and her concept of peripheral urbanisation but found it to be less useful. It collapses peripheral urbanisation into auto-construction, and uses it to describe practices of producing cities rather than examining urban peripheries per se. Auto-construction is a practice we see within peripheral spaces, but we do not necessarily ascribe it as being of such spaces. In other words, we view peripheries as more multi-faceted than an emphasis on auto-construction suggests. Essentially, we are dissatisfied with much of the literature available to us in terms of its explanatory value and relevance, and we aim to contribute to this debate.

We conceptualise the peripheries as spaces where complex socio-political, spatial and economic processes work to complicate and inform urban change, including in ways that are contradictory and variable. We view the peripheries as layered, relational spaces. Our conceptualisations encompass an understanding of peripheries which privileges the ‘peripheral’ geographical location of settlements but recognises that the idea of ‘being peripheral’ is subjectively determined. It also does not assume that the peripheries are necessarily economically, politically or culturally peripheral to the city region or spatially marginal (Pieterse, 2018); hence we include wealthier investments, spaces of power and varying ways of living in our framing of urban change, although in reality gaining insights into everyday life in these wealthier spaces has proved tricky. Some of the key features of peripheries in our understanding are that they are generally spaces located geographically some distance from a main urban core, recognising that this in itself may be fluid and relative, and that they may be close to new growing cores. They are areas of changing land use, where development may be relatively less dense and where a lack of services and infrastructure may be evident. The spaces are commonly residential but not always, and can be heterogeneous, depending on their particular histories. They often offer elements such as fresher air, cheaper land, access to affordable housing and rural features (although the meaning of ‘rural’ in this context may be complex and shifting). Such spaces often attract speculative investment and work as footholds in the city for particular residents, they may also show evidence of being incremental and unfinished, and they may be boring, feel dislocated and offer limited employment opportunities and transport facilities. Many of these spaces are in transition, and the temporal dimensions of urban change underpinning urban peripheries are key to our conceptualisations.

Drawing on the multiple forms of peripheral experiences and contexts we encountered in our research, we have through collaborative working initially identified five different categories of urban periphery which we aim to reassess, amend if needed, and expand upon in future publications and theorisations. We do not view these categories as hierarchical, exclusive, all-encompassing or finite: rather, they can operate in overlapping and hybrid ways, recognising that multiple categories can usefully be applied to each of our seven case study areas and that categories bleed into each other in important ways.

The five categories are as follows:

  • The speculative periphery. As the labelling suggests, this refers to urban spaces targeted (usually but not always or not exclusively) by private capital investment for the purposes of profit generation. This may be in the form of housing estates, commerce, industry or agri-business (or even major new multi-use developments) or may refer to spaces where particular investments generate and extend power bases of individuals or institutions. This category evidences the presence of relatively cheaper land, with sometimes easier (or less controlled) mechanisms of access, ownership and financing – although this is not a straightforward relationship.
  • The vanguard periphery. This refers to peripheral spaces within cities, which are at the forefront of urban change or are spaces of urban experimentation, state ambition, innovation and development. These may include new forms of housing, including mixed and sustainable housing forms, areas of integrated development, or areas with experimental forms of urban governance, which seek to manage potentially conflictual histories of rule and power broking.
  • The auto-constructed periphery. This is akin to Caldeira's (2017) conceptualisation, which privileges an understanding of the role of informality in parts of the urban periphery referring to both the efforts of poor households but also forms of investment which may overstep or ignore planning legislation, building codes and environmental concerns. Hence, auto-construction can cut across wealth barriers and describe forms of urban change usually directed by individuals, but most commonly evidences poor informal housing developments on parcels of land on the edges of urban centres.
  • The transitioning periphery. This captures many features of change often evidenced within peri-urban descriptors of spaces including changing land use from rural uses including agricultural to more urban uses such as residential, institutional or retail. Processes may include the densification of spaces through the reduction of plot size, growth in housing and other built forms as well as the building of infrastructure which transforms spaces, such as roads, electricity, water provision, bus shelters and shopping facilities. These may be areas with long settlement histories, but where change is very evident. This category also accounts for transitions in forms of governance across urban peripheries.
  • The inherited periphery. This describes spaces on the urban edge which were often produced by the state and commonly evidence decline or a failure to progress in various ways, such as employment opportunities, investment (particularly in business), infrastructure and basic services. Such spaces may continue to exist as spaces of obligation for the state, created for particular political and historical reasons and which continue to exert expectations and pressures on a weak authority. Economic opportunities may be narrow and vulnerable to change. The areas may be spatially fragmented, and for some residents living there, they can be poorly connected or have unaffordable transport provision. These are areas where residents’ narratives of change or the lack thereof evidence hopelessness, marginality, feelings of being trapped and neglected. At the same time, they may also contain social networks or investment in housing valued by residents, or be places where more diverse activities can occur, so they are not necessarily perceived in uni-dimensional ways. They may also include more middle-class residents with historical investment in housing or work in the area, and thus may be more differentiated than is immediately apparent. Finally, inherited peripheries may contain aspects of governance typifying that of the ‘informal strong-man’ who is powerful and effective but can also be dangerous, operating at times beyond the reach of the state or alongside relatively democratic structures.

As devices these labels are useful to us in several ways: they broadly categorise and summarise the types of urban peripheries we have encountered; they provide a shorthand to reflect differences and distinctions between them; and they suggest something of their ‘status’, role or trajectory within larger urban conglomerates. But, as noted, the categories are not mutually exclusive and may exist alongside one another in nearby localities, or form overlapping layers in an area undergoing change. We invoke these categories of urban peripheries in the three city regions studied in our project. This focus on the scale of the city region is a conscious spatial and analytical practice. We argue that the geographical peripheries of city regions are relatively absent in research terms, particularly in relation to African cities (aside from a few key analyses outlined above), and that there is a lack of understanding of the state and non-state developments they are attracting – and of their potential for dynamic change, intervention and decline.

Our struggles over our conceptualisations of the urban peripheries were keenly tested through the process of case study selection, where we debated various tensions around the multiple meanings of ‘peripheral’, including the geographic question of being peripheral relative to different legal and political boundaries around cities, regions and states. A case in point is that of Lufhereng, Protea Glen and Waterworks, which operates as a case peripheral to the city of Johannesburg but is less peripheral when considered within the wider city region of Gauteng. This case, as well as that of Yeka Abado in Addis, forced us to question the geographical and spatial assumptions inherent in the idea of the periphery and what centres or cores we were privileging in our analyses. We foreground the notion, identified above, that living in the urban peripheries is a relational practice, and we take as central residents’ own interpretations of where they live and how they define their home spaces. Thus we view the periphery as geographic, relational and lived as these quotes from participant diaries indicate:

We are still a rural area of which they say we are urban.

(PK, Diary, Ekangala)

Kangala is a township situated in a semi-rural area, although it being a township with semi urban lifestyle habits … you still experience the beauty of African cultures around here; people still practise tradition and isiNdebele.

(Q, Diary, Ekangala)

Living in this area has a distinct feeling. I mean, I've never lived in such an area before. It has very nice features. In any other part of Addis Ababa, you don't see farmers farming, collecting, preparing and storing their harvest and that creates a certain form of joy.

(010, Diary, Tulu Dimtu)

There is nothing unique about this area when you compare it to other areas. Considering that it's a rural area, I assume the change is that there is now water and power provision and a school is constructed.

(039, Diary, Yeka Abado)

We recognise that distance, accessibility, visibility and mobility are all critical, but we ask from and to what or where, rather than assuming the city centre as the obvious focus. The siting of interventions and the relative governance practices and engagement all shape the lived experiences of the periphery.

Debating ‘drivers of change’

Our project makes use of the idea of ‘drivers of change’ in order to capture the multi-scalar processes shaping urban change in the peripheries. This term allows us to explore a wide range of processes, including economic, governance, environmental, political, individual, etc., at multiple scales. We argue that the diversity of such drivers of change is significant and it suggests simplistic accounts of the peripheries are highly problematic. Our research reveals that in some areas, large-scale formal investment is evident (such as in Tulu Dimtu in Addis), while other areas are predominantly characterised by informal development or a complex mix of formal and informal processes, as in Molweni in eThekwini and Winterveld in Tshwane. We argue that theoretical framings, which focus only on growth, are misleading (although growth is evident, such as in northern eThekwini through planning and other forces) (Todes, 2017). Edges can also be places of economic and population decline, with Ekangala an example of decline of industrial opportunity over time. Governance can be weaker on the edge than in the core, or divided between adjacent authorities or between different forms of governance, as in Molweni or Yeka Abado/Legetafo. We use these ideas to move beyond work on African cities which either overlooked peripheral areas or focused on a donor-driven conception of the ‘peri-urban’ concerned primarily with changes to land use and farming (Mbiba and Huchzermeyer, 2002). However, we note that interactions with the land vary in such spaces, with some offering opportunities for subsistence farming or as places where traditional farming still occurs (as the quotes from residents above illustrate).

Rather than a conceptual focus on the interface between urban and rural, this research explores urban peripheries as distinct sites that can be subject to major investments, new urban visions, contingent governance practices and processes of growth and decline. It aims to understand how people live in these critical spaces of twenty-first-century urbanism, as well as the potential of these sites for economic development and poverty reduction. Given very high rates of urban growth and complex forms of urban spatial expansion across most of the African continent (Doan and Oduro, 2012; Fox, 2012; Parnell and Pieterse, 2014), the challenge of peripheral urban governance and poverty reduction is affecting African states and residents in complex and diverse ways. In South Africa, for example, city governments result from local government consolidation and often cover large areas, with some city regions crossing other administrative jurisdictions, e.g. Gauteng. Other countries on the continent are dominated by a capital or ‘core’ city in a highly unbalanced urban system, placing particular strain on the land surrounding one specific city (Thuo, 2013) and associated governance institutions. In Ethiopia and Uganda, this process of core city expansion is further complicated by ethnic dimensions of territorial governance, as the city spills over borders into the surrounding region (Goodfellow, 2010 and 2017; Gore and Muwanga, 2014) and as our Addis cases reveal, associated conflicts over land result.

Fringe locations, where local economic activities such as mining or manufacturing can either integrate into the city region or alternatively decline – through restructuring, for example – can be transformed through state housing or speculative land improvements. Development in these areas can result in mobility and access challenges which impact employment opportunities. However, our findings show that this is often very varied locally, with adjacent neighbourhoods experiencing quite significant differences. Developments in such areas can strain viable infrastructure and service delivery at scale, requiring private-sector (including transnational) investments in the face of energy, telecommunications and water shortages (Simone, 2014; Todes, 2014). Major infrastructure projects financed by foreign aid or international assistance, particularly from China as in the case of Addis (e.g. the light railway), are rapidly transforming urban edges in contradictory ways, fostering inclusion for some, but exclusion for others (Liu and Lefèvre, 2012).

Peripheries may also be characterised by environmental challenges (Aguilar, 2008), social exclusion and low levels of cohesion. In other instances, the benefits of smaller, well-serviced peripheral communities may prove highly desirable. Similarly, cross-border ethnic or political differences can paralyse negotiations between actors, but simultaneously privilege those who are able to exploit different governance systems and policy approaches between bordering municipalities (Todes, 2014). For some residents, then, urban peripheries are localities of choice, but for others they represent spaces of curbed choice. Nonetheless, people's presence on the edge has significant implications in terms of services, welfare, employment, labour force and markets for current and future development by both state and private market actors. The relationships between particular drivers, including the private sector, and their associated ‘markets’ can be intense and complex in terms of the specifics of what is included in particular engagements: the importance and multi-functional role of the Spar supermarket in our northern eThekwini case is an example of the significance of such spatially particular relationships.

The ways in which infrastructure and economic changes are conceptualised, realised and distributed (given their unevenness) and how these relate to everyday urban practices are key (Simone, 2014). Much of the recent research on and interest in infrastructure, which dominates urban theory currently, focuses on larger-scale infrastructural investments (Nugent, 2018). Our project reveals how in fact it is the smaller-scale, localised infrastructure interventions which are often the most significant drivers of change for residents. These may include investments in toilet blocks, road surfacing and local transport. These ‘micro drivers’ of change only emerge through a focus on the lived experiences of urban change and would be easily missed if such an engagement had not occurred.

This appreciation of the lived experiences of places and these ‘drivers’ of change’ processes has an important intellectual trajectory as well as methodology within urban studies, geography and planning. Often it draws on deep insights from urban anthropology (Bank, 2011; De Boeck and Plissart, 2004; Ross, 2010) and sociology (Mosoetsa, 2011) which are concerned with how people live, work, eat, move, consume, sleep, parent (Meth, 2013), love and die in place. Much urban research, particularly that informing meta-scale urban intervention, relies on limited survey instruments often assessing quantitative outcomes. Charlton (2013) and Meth (2015) have argued that there is a lack of understanding of the social outcomes and the lived experiences of major interventions, as well as of more micro-scaled material changes. These are arguably significant drivers of local change, including the provision of state-provided housing, or the upgrading of communal facilities in poor informal settlements. Meth and Charlton (2016) and Charlton and Meth (2017) reveal how such housing shapes livelihood challenges and how it has mobility implications, positive impacts on identity and security, but with gendered distinctions around power, violence and sexuality. These insights inform an analysis of housing concerned with welfare, social change and poverty, i.e. a lived experience interpretation of key drivers of change. Our project thus uses a methodology closely attuned to lived experiences to shed light on how larger as well as more micro drivers of change are experienced, and how they shape the urban peripheries in complex ways.

Capturing the everyday: emphases and omissions

We have argued here that conceptualising the everyday and researching the lived experiences of urban change are valid intellectual exercises. In practice, however, as we have progressed through our data collection we have debated and faced challenges with the question of who or what characterises the everyday. As outlined above, we commenced with the intention of giving voice to varying everyday lives occupying the peripheries, no matter what social class or housing form they occupied – our aim was to ‘sample’ those who lived there in order to capture multiple experiences of the peripheries. Our starting position was to avoid producing a summary of poverty on urban fringes and also to avoid overstating ‘niche’ experiences. We have used multi-nodal cases (see Figure 2.3) to reach diverse ‘everydays’ in most of our case study areas (i.e. a mix of very different housing types which largely but not entirely maps onto different classes of residents).

Figure 2.3 Examples of multi-nodal cases: (a) Northern eThekwini and (b) Yeka Abado

However, a variety of factors worked to contort and subvert these aims while other events and decisions simultaneously worked to satisfy and extend our aims, in complex and varied ways. Our choice of case study locations and their multi-nodal components is an obvious starting point for recognising who or what gets included or excluded. Initial decisions about what to include were overruled and reconsidered as data collection proceeded, as we recognised that our findings were exclusionary or lacking, or where we sought parity between data sources. For example, our inclusion of Waterworks informal settlement in the Lufhereng case in Gauteng occurred relatively late as we recognised its significance within the area, particularly in relation to imminent economic investment plans.

Gatekeepers were important too in affecting who we connected with and sampled, usually played out along party political lines in the South African cases. For example, Waterloo and Hammond's Farm exhibited elements of gatekeeping by the African National Congress (ANC), perhaps in attempts to reclaim their power, given that the formerly ANC-controlled ward was now led by the opposition Democratic Alliance party. But gatekeeping was also tied to ‘strong-man’ politics in the Ekangala case, where implicit ‘prevention’ of access occurred or where engagement with participants was highly managed and circumscribed. Language played an important role in ex/inclusion and was significant in all case study areas. In the Gauteng cases, the variety of languages at times stretched beyond the capacity of the immediate research team, requiring the ‘farming out’ of transcripts and recordings to translators and reducing the ease of communication in face-to-face engagements. In contrast, different language skills among the research team meant that particular foreign nationals were engaged and included, picking up on quite distinct experiences in the peripheries.

The weekday working-hour timings of field research in South Africa inevitably shaped patterns of inclusion and exclusion, privileging those working at home, unemployed, elderly or young. Efforts were made during interviews and surveys to question beyond the individual present, but it is important to recognise the ways in which such temporally particular engagements tended to yield very specific insights into people, places and processes. In contrast, data gathering for the Addis cases commenced later than the South African cases and all qualitative data were gathered during weekends and work holidays, thereby extending the range of residents included in the research. Some of our cases were very peripheral and highly spatially fragmented, with Winterveld in Gauteng our most extreme example. The dispersed and hard-to-access nature of this and other cases really brought home the grind of peripheral living, but also directly affected the frequency, duration and timings of researcher engagement.

Additionally, some of our cases contained neighbourhoods and areas that felt and presented as open and accessible, including Lufhereng in Gauteng, Molweni and northern eThekwini – and in the two Addis cases, especially the areas of condominium housing. This openness was primarily a function of housing type, but also related to social structures, relative safety, political calm, street pattern and the daytime presence of residents in the area. These nodes within our case study areas were in distinct contrast to others which were highly secured, closed and impenetrable, including Crestholme and Crestview in eThekwini, and to a lesser extent Verulam and Protea Glen as well as Country Club Development in Addis (in Yeka Abado). As sites of relative and extreme wealth, urban housing security in the form of gated access, walls, fencing, intercoms and guards, and more importantly a generic sense of mistrust, apprehension and ‘busyness’, worked to limit researcher access in sometimes small ways. These included the complete barring of access and non-engagement: e.g. the survey company we employed in South Africa struggled with accessing participants in ‘wealthier white/Indian areas’. Similarly we faced challenges when trying to recruit for our diary-writing task, as well as in recruiting those to undertake interviews in these areas. We obviously took measures to overcome these moments of exclusion, in particular through using alternate methodologies to reach individuals, including WhatsApp message groups, or use of gatekeepers to facilitate access. Nonetheless, overall our data is somewhat biased towards less wealthy, more accessible residents who through their presence and willingness shared their everyday lives with our team.

Finally, the challenges of urban and political unrest and the reality and threat of crime both worked to stifle access to some of our cases and reduce time spent in the field. In Addis, the ability of our team of researchers to visit the farmers’ settlement/informal areas of Yeka Abado, located right on the Oromia border, during mid-2018 was significantly undermined by ongoing political tensions. Access to various parts of Winterveld in Gauteng was reduced by concerns over researcher safety due to significant crime levels in the area.

This discussion has inevitably focused on the more restrictive realities that undermined data access, but the project has multiple ‘good stories’, too, of key individuals in each of the case study areas who extended themselves beyond what was expected to assist us in accessing participants and learning about their neighbourhoods. As examples, in northern eThekwini, two migrants from the Eastern Cape were pivotal in introducing us to the informal settlements of Canelands and Coniston, and in Lufhereng a local community development organisation worked with researchers to engage other participants in the area. These various factors have all worked together inevitably to emphasise some voices and omit others.

Comparability: cases, methods, stories, contexts …

We avoided strictly uniform criteria for case selection, although ultimately areas that were within geographic peripheries and where some form of investment had occurred were key. We aimed to include areas of decline, and we deliberately kept our notion of investment or intervention broad in order to capture diversity. Our cases are not necessarily comparable in size terms, population numbers, etc. As explained above, most are multi-nodal as they seek to capture a variety or diversity of lives in the periphery, but the characteristics of this multi-nodality varied from case to case. These variations shape processes of urban comparison, and we utilised multiple practices of comparison to structure analysis, conceptualisation and argument building. To support this we are able to use comparison to identify ‘base’ analyses, including comparisons of the history, rationale and length of settlement, along with the population figures and changes in these over the years. Our work compares the varied political affiliations and governance structures present within case study areas, identifying patterns across cases. These include identification of the generic and relative strength of the national governments of South Africa and Ethiopia in shaping the urban (in contrast to the kinds of governance identified in Sawyers’ (2014) work, for example, in Nigeria) alongside the actions of local committees and civic groups; the relative weakness of certain municipalities in servicing and supporting particular areas within cases; and the presence of problematic strong-men in some cases where formal governance appears weak and stretched or even absent. Thus our comparisons reveal that governance patterns and structures vary dramatically, with evidence of strong party politics in some, alongside problematic clientelistic relationships, power-wielding individuals, effectively functioning local democracies and committee structures, and competing traditional leadership in others.

Comparisons also reveal the varied significance of city region, municipal authority and traditional leadership boundaries in shaping investment forms and the financial clout of investments, as well as the relationship between state policy (at different scales) and actual practices of implementation on the ground. Housing types, and investment in state housing in particular, prove a significant area of comparison across all cases, throwing into sharp relief the viability of different state-subsidised schemes and the successes and failings of different architectural forms on everyday lives and urban quality. Employment, and most significantly unemployment, presents as the single biggest point of comparison across all South African cases, providing insights into how joblessness and job seeking are explicitly shaped through peripherality or manifest in diverse forms of peripheral settlement. Comparisons of unemployment reveal how difficulties in access, transport and affordability all work to undermine livelihood opportunities in seemingly very different spatial contexts. In Addis, having employment is a more dominant feature (and a prerequisite for access to condominium housing), but comparisons with the South African cases over declining affordability of new housing and the costs of living generate clear insights into the viability of life for many residents, supporting the findings of Yntiso (2008).

The role of capital in shaping the peripheries is an important comparative theme, but one which has no clear or singular narrative and is differentiated across cases depending on the types of investment (e.g. retail versus manufacturing), the relationship to residents (e.g. as prospective employee, or as customer) and the forms of capital (e.g. Chinese investors, global firms, local traders). Our findings do not point to a singular story about neoliberalism and the dominance of private capital to the detriment of state investment, although this may be important in some cases. Instead our material reveals that certain investments by the private sector – supermarkets and shopping malls in particular – can play a remarkably important role for residents locally, encouraging us to critique and counter more singular accounts of the dangers of capital penetration. Nonetheless, across our cases the volatility and fickleness of capital and its varying relationships (e.g. cosy, dominating, compromised) with different scales of the state produces patterns of investment, decline and change which directly shape access to work and services, usually revealing substantial inequalities in benefits to residents. Finally, we were struck by the significance of quality of life indicators across all cases and how influential these are to residents’ well-being. Variations on a lack of access to water, electricity, services, health care, schools, local shops, the police, etc. are common points of comparison (obviously varying in their details) across the cases and illustrate the significant needs felt by residents for local infrastructural investments in their areas.

Having sketched out some of the initial points of comparison, we recognise the need to develop analytical tools, which enable us to conduct comparison alongside an appreciation of contextual specificity and beyond the intricacies within each case study. There is also a need to further compare cases within and across city regions, and across two countries. We note that Gauteng, eThekwini and Addis Ababa have different contexts but important similarities too and that thorough contextual awareness in relation to each case is critical to in-depth understanding, especially in relation to drivers of change.

Methodologically we have worked to ensure that the data collection instruments are the same in each case and follow the same analytical processes where appropriate, and that we rely on the overall research questions to drive comparative analysis. Our capacity for comparison will be affected by differentials in data depth and coverage, which relate to the variety, volume and depth of multiple data types across and even within case study areas. This is partly a function of a significant difference in access to information and to existing resources. For instance, the presence of the Gauteng City Region Observatory and Gauteng expertise means much stronger baseline statistical data and overall level of contextual information in some cases. Yet positive surprises have shaped ‘better’ data collection than expected, including the positive responses to completing solicited diaries in Addis Ababa, a method initially assumed to be unfeasible in this city region. Our abilities to ensure consistency in depth and range of data collection are also a function of differences in access and different modes of access to residents in case study areas (detailed above in relation to gatekeeping) or because of the sampling of surveys or the benefits of social media. Finally, variability may be shaped by the ordering of data collection as it rolled out in a different sequence (via surveys first or diaries and interviews first) for reasons beyond our control.

Conclusions: the diversity of the peripheries

This chapter has set out various intellectual and methodological realities and challenges as we reflect on how we have ‘operationalised peripheries research’. We argue that there is no such thing as an urban periphery, rather we see multiple peripheries and peripheries within the periphery. We have produced an early categorisation of peripheries to illustrate this multiplicity, recognising its overlapping nature. The urban peripheries are significant and are experienced at multiple scales shaped through interventions, investments and other drivers of change, including many that are state directed. Importantly, we note how micro interventions, such as the building of a school or shop, are critical to everyday life and resident well-being. In contrast, we argue that big changes sometimes pass some residents by because of their lack of connections (political and/or physical), skills or social capital to benefit.

Our chapter has considered how a range of very real methodological choices and experiences worked to structure the production of knowledge and the breadth and depth of coverage of lives and voices across the urban peripheries. While this chapter has focused in more detail on those that ‘deviated’ from initial plans and intentions, our collaborative working experiences have served to illustrate just how productive and insightful our research endeavours have been, and our abilities to generate arguments and publications which draw on the points of comparison briefly sketched out above illustrate this achievement. To conclude, we are more convinced than ever that the urban peripheries are a critical site of urban transformation which offers insights into a true diversity of urban change. We are confident that our methodological choices to examine these dynamic peripheries through the lens of the everyday, alongside an appreciation of wider structural change, have yielded important insights and understandings which will help pave the way for a better understanding of urban peripheries.


We are grateful for funding from the British ESRC (Project Grant Ref: ES/N014111/1) and South African NRF (Project number 101579), which made this project possible. We also wish to thank all residents and interviewees across our seven case study areas for their participation and engagement with this project.


Aguilar, A. (2008). ‘Peri-urbanization, illegal settlements and environmental impact in Mexico City’. Cities, 25(3): 133–45.

Andreasen, M.H. , Agergaard, J. and Møller-Jensen, J. (2017). ‘Suburbanisation, homeownership aspirations and urban housing: Exploring urban expansion in Dar es Salaam’. Urban Studies, 54(10): 2342–59.

Bank, L. (2011) Home spaces, street styles: Contesting power and identity in a South African city. London: Pluto Press and Johannesburg: Wits University Press.

Beall, J. , Parnell, S. and Albertyn, C. (2015). ‘Elite compacts in Africa: The role of area-based management in the new governmentality of the Durban city-region’. International Journal of Urban and Regional Research, 39(2): 390–406.

Bloch, R. (2011). ‘Africa's new suburbs: Growth, expansion and governance’. Draft paper for the Global Suburbanisms Workshop on Suburban Governance, Leipzig, 1–2 July.

Buire, C. (2014). ‘The dream and the ordinary: An ethnographic investigation of suburbanisation in Luanda’. African Studies, 73(2): 290–312.

Caldeira, T. (2017). ‘Peripheral urbanization: Autoconstruction, transversal logics, and politics in cities of the global south’. Environment and Planning D: Society and Space, 35(1): 3–20.

Charlton, S. (2013). ‘State ambitions and peoples’ practices: An exploration of RDP housing’. PhD thesis, University of Sheffield.

Charlton, S. and Meth, P. (2017). ‘Lived experiences of state housing in South Africa's cities: Johannesburg and Durban’. Transformation, 93: 91–115.

De Boeck, F. and Plissart, M. (2004). Kinshasa: Tales of the invisible city. Ghent: Ludion.

Doan, P. and Oduro, C. (2012). ‘Patterns of population growth in peri-urban Accra, Ghana’. International Journal of Urban and Regional Research, 36(6): 1306–25.

Ekers, M. , Hamel, P. and Keil, R. (2012). ‘Governing suburbia: Modalities and mechanisms of suburban governance’. Regional Studies, 46(3): 405–22.

Fox, S. (2012). ‘Urbanization as a global historical process: Theory and evidence from sub-Saharan Africa’. Population and Development Review, 38(2): 285–310.

Garreau, J. (1991). Edge city: Life on the frontier. New York: Anchor Books.

Goodfellow, T. (2010). ‘“The bastard child of nobody”? Anti-planning and the institutional crisis in contemporary Kampala’. Crisis States Research Centre Working Paper, LSE.

Goodfellow, T. (2017). ‘Taxing property in a neo-developmental state: The politics of urban land value capture in Rwanda and Ethiopia’. African Affairs, 116(465): 549–72.

Gore, C.D. and Muwanga, N.K. (2014). ‘Decentralization is dead, long live decentralization! Capital city reform and political rights in Kampala, Uganda’. International Journal of Urban and Regional Research, 38(6): 2201–16.

Harrison, P. and Todes, A. (2015). ‘Spatial transformation in a “loosening state”: South Africa in a comparative perspective’. Geoforum, 61: 148–62.

Jenkins, P. (2013). Urbanization, urbanism and urbanity in an African city: Home spaces and house cultures. London: Palgrave Macmillan.

Keil, R. (2018). Suburban planet: Making the world urban from the outside in. Cambridge: Polity Press.

Liu, X. and Lefèvre, B. (2012). ‘Chinese influence on urban Africa’. Studies No. 06/12, Institut du Développement Durable et des Relations Internationales, Paris.

Mabin, A. , Butcher, S. and Bloch, R. (2013). ‘Peripheries, suburbanisms and change in sub-Saharan African cities’. Social Dynamics, 39(2): 167–90.

Mbiba, B. and Huchzermeyer, M. (2002). ‘Contentious development: Peri-urban studies in sub-Saharan Africa’. Progress in Development Studies, 2(2): 113–31.

McKee, D.L. and McKee, Y.A. (2004). ‘Edge cities, urban corridors and beyond’. International Journal of Social Economics, 31(5/6): 536–43.

Meth, P. (2013) ‘Parenting in informal settlements: an analysis of place, social relations and emotions’. International Journal of Urban and Regional Research, 37(2), 537–55.

Meth, P. (2015). ‘The gendered contradictions in South Africa's state housing: Accumulation alongside erosion of assets through housing’. In C. Moser (ed.), Gender, asset accumulation and just cities, 100–16. London: Routledge.

Meth, P. and Charlton, C. (2016). ‘Men's experiences of state sponsored housing in South Africa: Emerging issues and key questions’. Housing Studies, 32(4): 470–90.

Mosoetsa, S. (2011). Eating from one pot: The dynamics of survival in poor South African households. Johannesburg: Wits University Press.

Nugent, P. (2018). ‘Africa's re-enchantment with big infrastructure: White elephants dancing in virtuous circles?’. In J. Schubert, U. Engel and E. Macamo (eds), Extractive industries and changing state dynamics in Africa: Beyond the resource curse, 22–40. Abingdon: Routledge.

Parnell, S. and Pieterse, E.A. (2014). Africa's urban revolution. Cape Town: UCT Press.

Parnell, S. and Robinson, J. (2012). ‘(Re)theorising cities from the global south: Looking beyond neoliberalism’. Urban Geography, 33(4): 593–617.

Phelps, N.A. (2012). ‘The growth machine stops? Urban politics and the making and remaking of an edge city’. Urban Affairs Review, 48(5): 670–700.

Pieterse, M. (2018). ‘Where is the periphery even? Capturing urban marginality in South African human rights law’. Urban Studies, 56(6): 1182–97.

Ross, F. (2010). Raw life, new hope: Decency, housing and everyday life in a post-apartheid community. Cape Town: UCT Press.

Sawyer, L. (2014). ‘Piecemeal urbanisation at the peripheries of Lagos’. African Studies, 73(2): 271–89.

Simone, A. (2014). ‘Infrastructure, real economies and social transformation: Assembling the components for regional urban development in Africa’. In S. Parnell and E. Pieterse (eds), Africa's urban revolution, 221–36. Cape Town: UCT Press.

Stern, M.A. and Marsh, W.M. (1997). ‘Editors’ introduction: The decentered city: Edge cities and the expanding metropolis’. Landscape and Urban Planning, 36(4): 243–6.

Thuo, D. (2013). ‘Exploring land development dynamics in rural–urban fringes: A reflection on why agriculture is being squeezed out by urban land uses in the Nairobi rural–urban fringe?’. International Journal of Rural Management, 9(2): 1045–134.

Todes, A. (2014). ‘New African suburbanisation? Exploring the growth of the northern corridor of eThekwini/KwaDakuza’. African Studies, 73(2): 245–70.

Todes, A. (2017). ‘Shaping peripheral growth? Strategic spatial planning in a South African city-region’. Habitat International, 67: 129–36.

Watson, V. (2014). ‘The case for a southern perspective in planning theory’. International Journal of E-Planning Research, 3(1): 23–37.

Yntiso, G. (2008). ‘Urban development and displacement in Addis Ababa: The impact of resettlement projects on low-income households’. Eastern Africa Social Science Research Review, 24(2): 53–77.

African cities and collaborative futures

Urban platforms and metropolitan logistics


All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 220 220 55
PDF Downloads 69 69 35