Diepsloot - Housing & the Informal City

HOUSING & THE INFORMAL CITY RESEARCH PROJECT

Diepsloot Reception Area Johannesburg, 2008

Project partner: Goethe-Institut, Johannesburg
Collaborator: Prof. Lone Poulsen
26’10 Team: Thorsten Deckler & Anne Graupner (principals), Tahira Toffah, Guy Trangoš, Shameemah Davids, Claire Lubelll

In this research inquiry, Diepsloot’s Reception Area has been used as a basis against which to ‘test’ existing formal housing typologies and to develop alternative strategies which respond to and learn from the dynamics of the Informal City. Diepsloot is a predominantly informal settlement located 40km north of Johannesburg’s city centre. As a post-apartheid township it acutely represents the challenges facing the South African state on matters of housing, service delivery and effective local governance. Based on detailed mappings of Reception Area produced by 26’10 south Architects, a master class involving various architects and housing specialists was held exploring the complex relationship between housing density, occupation density, built form and customisations such as rental rooms, small businesses and the delineation of communal and public spaces.

In South Africa, density is generally accepted to be low at less than 40 units/ha, medium at between 40 and 100 units/ha and high at more than 100 units/ha. Reception Area’s existing density at 300 units/ha (predominantly one storey), its quality of spaces and diversity of functions point to the potential for low rise, high-density solutions which may be freehold and offer opportunities for sub-letting and small scale enterprise. A one hectare piece of land situated along a busy street of Reception Area was used to compare different formal housing approaches measured in terms of the displacement or additional accommodation each settlement type achieves.

When considering growth over time (in the form of self-constructed rental rooms or backyard shacks), the RDP type in fact performs as well as the much touted row-house and delivers densities approximating those of Reception Area. However, the siting of the RDP house in the middle of its plot results in poor in-between spaces when rental rooms are added over time.

Simultaneously, the existing settlement infrastructure and social amenities are often overburdened by the increased occupation. Interestingly, Chitungwiza on the outskirts of Harare, Zimbabwe presents an effective alternative in which two freestanding houses are joined along a shared boundary leaving enough space for rental rooms to define a quality semi-private communal space. The row-house type, promoted as the alternative to the free-standing RDP, starts out with higher initial densities, but has limited growth potential over time due to the inconvenience of tenants passing through the main house to access their accommodation. Units with passages on one side or access lanes at the back of stands may alleviate this to a degree. Whilst BNG (‘Breaking New Ground’) principles have been developed to render houses ‘safe’ as collateral for bank loans, current densities are not particularly efficient in the use of land and infrastructure. The banks furthermore frown upon self-constructed additions.