African Urban Quarterly

ARCHIV MOUGEO no. 116116

AFRICAN URBAN QUARTERLY

Special

Urban Agriculture in Africa

____ !I iI.1II!J!1

VOLUME 11 NUMBERS2 AND 3 MAYAND AUGUST,1996

Publishedin March 1999 AFRICAN URBANQUARTERLY INTERNATIONALADVISORY EDITORIAL BOARD

J. A. Adelamo,University of Lagos, Nigeria MichaelL. McNulty, Universityof Iowa, USA Hycith I. Ajaegbu,University of Jos, Nigeria Arcot Ramachandran, United NationsCentre for Human M. A. Al Hammad,Arab Urban Development Institute, Settlement SaudiArabia Akbar Mohamed, SUNY-Binghampton, USA Guy C. Ankerl,Inter-University Instituteof Geneva, A. C. Mosha, Universityof Botswana, Botswana Switzerland Walter 0. Okomo,SUNY-College of Staten Island, USA Elias H. 0. Ayiemba, Universityof Nairobi,Kenya AnthonyM. O'Connor,University College, London, UK Keith Baird, SUNY-Buffalo, USA Peter 0. Ondiege,University of Nairobi,Kenya SalvinoBusuttil, Universityof Malta, Malta Richard S. Odingo,University of Nairobi, Kenya Mike B. K. Darkoh, KenyattaUniversity, Kenya SimeonH, Ominde, Universityof Nairobi, Kenya AzukaA. Dike, Universityof Nigeria at Nsukka,Nigeria LeonardPlotnicov, Universityof Pittsburgh,USA M.A. Ezz Al-Din, Universityof Kuwait, Kuwait C. M. Rogerson, University of Witwatersrand, RSA Roland Fuchs, United Nations University, Japan Pius 0. Sada, Universityof Lagos, Nigeria Josef Gurgle,The Universityof Connecticut,USA Ademola T. Salau,University of Port Harcourt,Nigeria B. W. Hodder,University of London,UK EdwardW. Soja,UCLA, USA Allen Howard, RutgersUniversity, USA Elliot Skinner,Columbia University, USA B. S. Hoyle, Universityof Southampton, UK Christopher J. Smith, SUNY-Albany, USA George0. Krhoda, Universityof Nairobi, Kenya Richard E. Stren, Universityof Toronto, Canada John P. Lea, Universityof Sydney,Australia D. R. Fraser Taylor, CarletonUniversity, Canada AkinL. Mabogunje,University of Ibadan,Nigeria Peter Treuner, Universityof Stuttgart, Germany Justice N. Mlia, Universityof Malawi,Malawi Peter P. Wailer, GermanDevelopment Institute,Germany

AFRICAN URBAN QUARTERLY is an independent, Telephone: 254-02-216574; Telex: 22095 VARSITY interdisciplinaryand internationaljournal published quarterlyin NAIROBI,Cable: VARSITY NAIROBI February, May, August and November. Reprints: Minimumorders of 50 copies are available from the Editorialoffices: AfricanUrban Quarterly Limited, P.O.Box 74165, publisher. Nairobi, Kenya, East Africa. Telephone: +254-2-216574,Telex: 22095VARSITY NAIROBI,Cable: VARSITY NAIROBI. Indexing: Appropriate articlesappearing in AfricanUrban Quarterly areabstracted or indexed in: UN Selected Publisher: African Urban P.O. Box monthlyBibliography of Quarterly Limited, 74165, International East Africa. Telex: Articles; Development Abstract; Sociological Nairobi, Kenya, Telephone: +254-2-216574, Social Social and Social 22095 VARSITY NAIROBI,Cable: VARSITY NAIROBI. Abstract; Welfare, Planning/Policy Development (SOPODA); International Bibliography of Book Editor R. 0.Obudho, African UrbanQuarterly Limited, P.O. Box Reviews (IBR); Sage Urban Studies; Current Geographical 74165, Nairobi, Kenya, East Africa. Telephone: +254-2-216574, Publishers; InternationalRegional Science Review; Population Telex: 22095 VARSITY NAIROBI,Cable: VARSITY NAIROBI. Index;Anthropological Index; CIB International ConstructionData Base ICONDAof the InformationCentre for Regional Planning Subscriptionrates: Four issues and BuildingConstruction (IRB); Sociology of EducationAbstract KENYA OUTSIDE KENYA and AfricanUrban and RegionalScience Index. Individual US $ 50.00 US $ '60.00 Opinionsexpressed in signedcontributions arethose of theauthor(s) Institution US$ 100.00 US $ 120.00 and do not reflect those of the Editor or African Urban Quarterly Single copy US$ 40.00 Limited. Special rates for students in full-time education are availableon © Copyright 1996by AfricanUrban Quarterly Limited. request. ISSN 0747-6108 Airmailpostage: add US$ 40.00 Surfacemail US$ 20.00 The whole ofmaterials appearing in this periodicalis copyrighted All payments must be made in US dollars and cheques are made and may not be reproduced in whole or in part without the payableto African Urban QuarterlyLimited. permission of the.Editor and Publisher,African Urban Quarterly Limited. P.O. Box 74165, Nairobi, EastAfrica. Telephone: and Orders: Kenya, SubscriptionInquiries +254-2-216574, Telex: 22095 VARSITY NAIROBI, Cable: African Urban QuarterlyLimited, P.O. Box 74165, VARSITY NAIROBI. Nairobi,Kenya, EastAfrica. /1 (f!, IDRC B

AFRICAN URBAN QUARTERLY

Editor R. A. OBUDHO, Universityof Nairobi,Kenya

Associate Editors MICHAELCOQUERY, Universitéde Paris VII, France SALAHEL-SHAKHS, RutgersUniversity, USA ALAN FRISHMAN, Hobart and William Smith Colleges, USA J. B. OJWANG, Universityof Nairobi, Kenya HARRY W. RICHARDSON, Universityof SouthernCalifornia, USA PAUL SYAGGA, Universityof Nairobi,Kenya

AFRICAN URBAN QUARTERLYLIMITED, P.O. Box 51336, Nairobi, Kenya, East Africa. Telephone: +254-2-216574;Cable: VARSITY Nairobi; Telex:22095 VARSITYNairobi and Fax: +254-2-336885

C f 0

// C7 AFRICAN URBAN QUARTERLY Volume 11 Numbers 2 and 3, May and August, 1996

CONTENTS

ARTICLES • Introduction:An ImprovingDomestic and International Environment for African UrbanAgriculture Luc J.A. Mougeot 137

• Urban Agricultureand the African Urban Food SupplySystem Annu Ratta andJoe Nasr 154 • How Do the Urban Poor Stay Alive? Food Provisionin a Squatter Settlement of Bissau,Guinea-Bissau Ilda Lourenco-Linde!! 163 • Urban Agriculture,Food Security and Nutritionin Low Income Areas of the City of Nairobi, Kenya Alice Mboganie-Mwangi and Dick Foeken 170 • Urban Agriculturein Africa:A Comparative Analysis of Findingsfrom Zimbabwe, Kenya and Zambia Da#ie! S. Tevera 181

• Urban Food Producers' Decision-Making: ACase Study of Kibera, City of Nairobi, Kenya Pascale Dennery 189 • The Beautiful City: Gardensin Third World Cities Mary Cockramand Shelley Feldman 202

• Open Space Cultivationin Zimbabwe: ACas Study of Greater Harare,Zimbabwe DavisonJ. Gumbo and Takawira W Ndiripo 210 • AnimalFarming in African Cities Ann Waters-Bayer 218 • Urban Microfarmingin Central Southern Africa:A Case Study of Lusaka,Zambia A.WDrescher 229

• Epurationet réutilisationdes eaux usées domestiques en maraIchage pdriurbain a Dakar, Séndgal Seydou Niang 250 • Farming in the City: From Analysis to Action Julie van derBliek andAnn Waters-Bayer 259 • Farmingin the City of Kampala: Issues for UrbanManagement Gertrude Atukunda and Daniel Maxwell 264

BOOK REVIEWS • UNDP, UrbanAgriculture. Food, Jobs and SustainableCities Beacon Mbiba, Urban Agriculturein Zimbabwe. Implications for Urban Managementand Poverty Dick Foeken 277

• Alice M. Mwangi,The Role of UrbanAgriculture for Food Security in Low Income Areas of Nairobi Dan Maxwell 279

• Carlos Lopes, Enough is Enough: For anAlternative Diagnosis of the AfricanCrisis Samuel0. Akatch 281

• FAO, Supplyof LivestockProducts to Rapidly Expanding Urban Populations G. E. Otiang 'a-Owiti 283

111 ISSN 0237 - 3539 THE BANGLADESH JOURNAL OF AGRICULTURAL ECONOMICS

Volume XIX June and December 1996 Numbers 1 & 2

Article

Effectof Educationon Technology Adoption andAggregate Crop Output in Bangladesh—UttamKumar Dev and Mahahub Hossain

TechnicalChange and Elasticity ofFactor Demand in Rice Production inBangladesh—M. Kamuruzzaman, Md. Ferdous Alam and Jovnal Abedin

Changing Cropping Pattern in Bangladesh from 1971—75through 1991—93: Implications on Crop Sector Growth— Shamsul Alam andMd. JavnalAbedein

A Region-Wise Analysis of Supply Responseof Cotton Crops in Punjab—AnjuSharma and K. S. Dhindsa Research Note

Adoption of Improved Practices by Potato Farmers—M.M. Huque, M. H. Rashid andM. L. Rahman

Economics of Japanese Quail Farming in Dhaka MetropolitanCity—S. A. Siddique and M. A. S. Mandal

Yield, Price and Income Instabilityof Different Crops in Jessore Dis.trict—M. S. Alam, S. M. Elias and S. M. M. Murshed

Demand for Fertilizer in Bangladesh.A Note—P. C. Modak and B. K, Barmon

Comparative Progress of Development During Pre and Post Liberation Period in Bangladesh Tea Industry—M. S. Gazi andf. KSaba

Editorialcorrespondence should beaddressed to theExecutive Editor,The BangladeshJournal of Agricultural Economics, BangladeshAgricultural University,Mymensingh, Bangladesh. Subscription orders and business correspondence should beaddressed to the DeputyDirector, Bureau ofSocioeconomic Research and Training, Bangladesh Agricultural University, Myrnensingh- 2202, Bangladesh.Telephone: 09 1-55695to 97 Ext. 2300

The Half Yearly Journal of THE BUREAU OF SOCIOECONOMIC RESEARCH AND TRAINING BANGLADESHAGRICULTURAL UNIVERSITY MYMENSINGH African Urban Quarterly, 1996: 11(2 and 3) pp. 137—152 ©1996 by African Urban QuarterlyLtd.

INTRODUCTION: AN IMPROVING DOMESTIC AND INTERNATIONALENVIRONMENT FOR AFRICAN URBANAGRICULTURE

Luc J.A. Mougeot CitiesFeeding PeopleProgram Initiative,International Development ResearchCentre, P.O. Box 8506, Ottawa, Canada KIG 3H9

Abstract Accepted May, 1996 The introduction submits a definition of the cotzcept of urban agriculture, reviews some persisting myths and positions contemporarypractices in ahistorical perspective.The sequence ofcontributions to this issueis discussed, withafirst group documenting the roleof urban agriculture within the urbanfood supply system andwithin urbanpoor's provisioning strategies as well as the impact ofself-consumption onproducer households 'foodsecurity. A secondand more importantgroup illustrates emerging trends which shouldfacilitate the growing incorporationof urban agriculture into African urban management, over the next decades. Indeed, more and more municipalgovernments recognisethe sector as a growing urban economic function. Many are re-interpretingor revisingcurrent policies, includingplanning norms regardingurban agriculture (e.g., productive use ofopen spaces); several studies make recommendationsconcerning specificfarming systems andtheir interaction with one another More efforts are being made totap on synergies between urban agriculture andunderused urban resources (e.g., wastewaters). More public consultationsare being held to resolve aforediscussedpolicy, planning and management issues. Local authority associations and development agencies are also paying moreformal attention to urban agriculture and this bodes weilforfuture African initiatives. Résumé L'auteursoumet une definition du conceptd'agriculture urbaine, répond a quelquesmythes persistants a son sujet et situe lespratiques contemporainesdans une perspectivehistorique. On y présente les contributionsdes autres articlesde ce numéroa Ia comprehensiondu phénomèneen Afrique. (In premier groupe d'articles documentent le role de l'agriculture urbaine au sein du systéme d'approvisionnementalimentaire urbain et des strategies auxquelles ont recours les pauvres des vu/es pour assurer leur alimentation,ainsi que l'impact de l'autoconsommationsur Ia sécurité alimentaire des ménages producteurs. Unsecond grouped'articles, plus important, discutentplusieurstendances emergentesdevantaccommoderune meilleureprise en comptede l'agriculture urbaine dans Ia gestion des villes africaines, au cours des prochaines décennies. En effet, de plus en plus de gouvernmentsmunicipaux reconnaissentle secteur commefonctioncroissante de l'économie urbaine. Plusieurs en sont même a ré-interpréter ou reviser lespolitiques en vigeur y compris des normes de planification touchant l'agriculture urbaine (v.g.: l'utilisationproductive des espaces libres);plusieurs etudesfont des recommandations au niveau de systèmes deproduction specifiques et des ressources urbaines sous-utilisees (eaux usagées). De plus enplus de consultationspubliques ontlieu pour résoudre les contraintescihautprésentées au niveaudes politiques,de Ia planification et de Ia gestion. L'attention croissante que reçoit l'agriculture urbaine de Ia part des associationsd'autorites locales et des agences de développementdevrafavoriser de nouvellesinitiatives africaines dans ce sens.

A Definition in the Making acknowledged by the more conservativepublics as diverse Mindsets and policies on urbanagriculture areundergoing and very robust, if not functional, components of the a notableshift within countriesand among the international modernisation process. Progressive policy reforms have development community. This may reflect our latest wisely supported theirdevelopment, when not appropriating awakening to real-world urbanisation underway in most someof theirelements. Predictably,urban agriculture should societiessince theend of World WarII. The concept of 'urban completea similar genesisby the end of this century. agriculture' has been coming of age overthe last threedecades The conceptof urbanagriculture is now evolvingrapidly, in the wake of other formerly condemnedurban anomalies with far-reachingimplications for the future developmentof (e.g., 'spontaneoushousing' in the 1950sthrough the 1970s urbancentres in both the North and the South.Whereas initial and 'informal employment' in the 1960s throughthe 1980s). submissions were concerned with establishing the identity Originallyexpected to be crushedby the irresistibletides of of urban farming (rangesof production systems, locations, western modernisation,spontaneous housing and informal practitioners and legal status), recent interpretations have employment actuallyhave endured, expandedand evenhave stressed its relationships with the economic, social and come to prevail in many places; they ultimately would be environmental sustainabilityof urban centres(food security, Acknowledgements—I wish to express my sincerethanks to manycolleagues who have either reviewedearly drafts and/orprovided useful materials:David Drakakis-Smith,Joe Nasr, Jean-MichelCentres, Joe Phellan,Jac Smit, BowndinKing, Chris Rogerson,Beacon Mbiba, Lood Spies and Camillus Sawio. My appreciation goes to the African Urban Quarterly for their invitation and to the International DevelopmentResearch Centre for grantingresources to organisethis issue as well as writing time for this article. I alone am responsible for its contents.

137 Introduction—LucJ. A. Mougeot AfricanUrban Quarterly nutritional health, employment and income generation, pools,complementary commodities and a fewproduction entrepreneurialdevelopment, open space management and scales, systems and consumption markets. However, waste resource reuse, gender empowerment, community urban agriculture is typically much smaller-scaled and development).Still, while the phenomenon is undeniable, more spatially dispersed than rural agriculture; this most specialistswould agree that an operationaldefinition affords more intensive,integrated, flexibleand adaptive will need to be periodically revised out of practical farming. Urbanagriculture tends to be technicallymore experience,with conceptcomponents tobe further specified efficient, tapping on economies of agglomeration inurban to instructlocal interventions. centres—information, expertise, infrastructure,inputs, An attempt to merge earlier with more recent emphases services and specialty markets—which are unmatched submits the following definition to the reader: Urban in most rural areas. Urban dynamicspress upon urban agriculture is an industry located within (intra-urban) or on agriculture a measureof mobilityand organisationagain the fringe (pen-urban) of a town, an urban centre, a city or unparalleled in rural areas; this is becausein urbancentres metropolis,which grows or raises, processesand distributes farming needsto competeor combine with a variety of adiversity of food and non-food products throughintensive other land uses, to avoid or overcome more severe plant cultivation (horticultureand agroforestry) and animal ecological stress and respond swiftlyto shifts in demand husbandry(livestock, avicultureand aquaculture)which uses for perishable specialties (Smit, 1980: 499—506; May and and reusescompetitively urbanhuman and material resources Rogerson, 1994: 90). Urban dynamics constantly (e.g., household, cooperative,enterprise, corporation) mainly transforms urban agriculture, impressing on it an to the daily subsistenceneeds and market demand of local opportunism and adaptabiliiy unparalleled in rural consumers (Ganapathy, 1983; Sawio, 1993; Smitetal., l996a: agriculture. 3). (c) Urbanagriculture is nothingmore than a euphemismfor Some misconceptionsare still widespread in Africa and horticultural gardening in the urbancentre: Initial surveys elsewhere, which often stem from initial directions of the by French geographersin West Africa and most UNU literature,that the afore definition intends to correct: Food-Nexus Programme surveysof African urban centres (a) Mostagriculture labelledas locationally urbanis actually did emphasise this visually more impressing form of sited in rural settings: What exactly is an urbanarea will urbanagriculture (Vennetier, 1961: 60—84; Sanyal, 1986; vary fromcountry to country. However, delimiting urban Diallo and Coulibaly, 1987;Tricaud, 1987 and Schilter, areas according to their field of activity rather than 1991). However, UNDP's survey commissioned to The following jurisdictional boundaries, consistently has Urban Agriculture Network (TUAN) in 1991—1992, proven to be more useful when planning and managing identifiedover 40 different farming systems and their urban-based activities. Awide range of criteriahave been variants, each with its own technology, investment needs, used to determine the 'urban' location of agricultural yield rates and returns to labour and risk (Smit et al., activities in Africanist literature, from land-use zoning 1996a: 107—134). Morerecent publicationsrecognise the (Mbiba, 1994: 188—202) topopulation density thresholds diversity of African urban gardens and address a range (May and Rogerson, 1994: 69—96). A need for clarityis of non-horticultural systems (Centres, 1992; Sawio, fully justified on policy, if not on planning grounds. 1993; Egziabher, 1994: 85—104; ENDA-Zimbabwe, However, the risk is still very limited that current 1994; Greenhow, 1994; Lee-Smith and Memon, 1994: evidence on urban agriculture would actually refer to 67—84 and Drescher, this issue). agriculture in rural areas. Research in Africa, Latin (d) Within any given zoneof a city, urban agriculture is mainly Americaand Asia (e.g.: ORSTOM,IDRC, UNU, FORD, carried out in residentialbackyards: Most surveys show UNDP) and development interventions (e.g.: GTZ, that residentialbackyards actually support only a small DANIDA,DGIS, ODA, FAO, regionalbanks, WB) so share of the full activity underway in any given urban farhave mostly targettedthe larger urbancentres (above centre or within each household. Ground-level land 500,000people). The literature, in fact, has beencriticised availabilityis particularlylimited in low-income,high- for disregarding smaller centres, where the rural-urban density districtswhere the numbers of off-plotproducers divide often isquestionable. Beyond theintra/peri-urban are larger. TheUNDP surveyidentified seven large types dichotomycaptured by the definition, spatial zonations of space based on tenure and site characteristics(home (core, corridors, wedges and peripheries) have been spaces: back, side and front yards, basements, sheds, proposed, more useful both to explain the geographic rooftops, patios,balconies, walls, windowsills, doorsteps, nichesof specificfarming systems and toplan the spatial plastic tubes and sticks; community spaces; surplus or allocation and evolution of particular systems, during reserve public and private spaces; roadsides and other physical expansion and land-use development of the rights-of-way; streamsides and floodplains; wetlandsand urban area (Smit et al., 1996a: 96—101). water bodies and steep slopes). All these venues have (b) Urban agricultureis technicallynot different from rural been observed to be intensivelyused in African urban agriculture, only done at the wrong place: Although centres(Freeman, 1991; Maxwell, 1994: 47—66; Gumbo closely related, urban agriculture is distinct from rural and Ndiripo; Dennery and Drescher, this volume). agriculture and interacts with it in many ways. Both Becauseurban agriculture is muchmore than aresidential originally share common technologies and biogenetic land use, it is practisedin synergy with a wide range of

138 Volume 11 Nos. 2 and 3, May andAugust, 1996 An ImprovingDomestic and International EivironmentforAfrican Urban Agriculture

other land uses, commercial,recreational, transportation, unsustainabilityof this divorce in the face of new urban manufacturing, institutional, infrastucture or utility- challenges, particularlyin sub-Saharan Africa. related. Urbanagriculture is probably as oldas the firstgreat urban (e) Urbanagriculture is largelyan informal, whennot illegal, centres built by our ancient civilisations.What is urban is survival strategyof the urbanpoor: Although most urban historicallyor culturallydefined; what ancient civilisations agricultorsproduce for householdsubsistence, surveys considered to be urban do not fit our contemporaryreality, consistentlyshow thatall incomegroups and manypublic just as our definitionsmay not fit the urbanreality of future institutions and private companiespractice some urban civilisations. What is striking, however, is that, no matter agriculture, with varying shares of their food and non- what definition prevailed, throughout history and across food output being marketedlocally, when not nationally civilisations,food production in urban settlements has not or internationally(Mvena, Lupanga and Mlozi, 1991; been exceptional or temporary. Drakakis-Smith, 1992: 258—283; Sawio, 1993 and Investments were made into massiveand ingenious earth- Maxwell, 1994: 47—66). Better-off groups are and waterworks within and fringing ancient urban settlements. increasingly organising the production of poorer Thesewere used whollyor partly togrow shrubs,vegetables, practitioners when not employing them directly or fruits, ornamental, medicinal and other utilitarian plants for indirectlythrough outsourcing (Drakakis-Smith, Bowyer- food, feed, storage,wood and fuel, buildingmaterials, shade, Bowerand Tevera, 1995: 191). Withgrowing competition fencing and windbreaks, and to raise livestock for food, for urban agricultural inputs and services, surveys materials,traction, transport, trade, sacrificesand status. consistently showthat thepoor's access to inputs, support Ancientagriculture was not sited haphazardly in the urban and benefitsis disproportionately curtailed(Mlozi, 1993: fabric or disconnected from the rest of its economy. Urban 105—128); urban agriculture is not easily affordableby food production took a variety of forms, making efficient the poorer and more recent arrivees to cities where it is use of space,site conditions,closeness to amenities,services well-established. and resources. Urban centres combined production and There is anecdotalevidence of workers'mobility from storage areas (Knossos), associated walled gardens with other sectors or occupations into and out of urban hydraulic facilities (Thebes, Egypt) and fed livestock with agriculture, though there seems to be a tendency for brewery refuse (Akhenatan, Egypt) (Warren, 1994: 46—51; peopleto cumulate urban farming withother occupations; Jellicoe, 1989: 25 and Courtlandtand Kocybala, 1990: 126). there is also anecdotal data on mobility within urban Some urban centres built vaulted aqueducts which gravity- agriculture. Butoverall, systematicaccounts overa range conveyed water from underground cisterns to terraced of socioeconomicgroups and farming systemsduring a cropfields (Eleutherna, Crete)(Rodenbeck, 1991: 91). Roman periodof years await further investigation.More studies coastal urban centres linked their harbour by channels to refer to employment generated by urban agriculture in lagoon-based fish-farmsfor local consumption and export activities which either provide inputs or use its outputs, (Cosa), as still practised today (Orhetello, Italy) (McCann, in therest of theurban economy, (Schilter, 1991; Groupe 1994: 95—96). Across theAtlantic inpre-Columbian America, de Rechercheet d'ExchangesTechnologiques (GRET), coastal centre authorities ordered the drainage of swamps 1991; Greenhow, 1994; Drakakis-Smith, 1995; and immense waterworks supported a highly organised Lourenço-Lindell,this issue). agriculture (Edzna, Mexican Campeche) (Hammond, 1994: 132). Needless to say, a growing number of contemporary ContemporaryUrban Agriculture in Historical urban centres are reusing treated and untreated municipal Perspective wastewaters to irrigate woodlands,orchards, pastures, grain As our planet becomes predominantlyurban, the world crops, produce and fish (for an internationaloverview see cities' 'corporate memory' is acquiring new relevance. It Edwardsand Pullin, 1990; Moscoso on Lima, 1995; Thang enables us to gain perspectiveon our current crises and to on Hanoi, 1996; Niang on Dakar, this issue). inventoryhow we have applied human intelligenceto cope Urban agriculture does not appear to have been socially with the needs of an urbanising humanity, in very different demeaning or technically primitive in past civilisations. Major cultures and economies.The reasons for and ways in which political centres sited on wetlands(Teotihuacan) reclaimed much urbanagriculture was promotedin past societiesbear extensiveareas throughraised-beds anchored with fencesof strange similaritieswith its resurgenceon the eve of the 21st willows, filled in and periodically fertilised with marshy century. vegetation and topped with canalbottom; these carriedfruits, The gathering evidence, with pre-colonial African vegetables,trees and houses (Coe, Snowand Benson, 1986: urbanism still comparativelyunder-documented, shows how 104). Transplanting from reedbeds afforded multiple extravagantwas the attempteddivorce between urban centre harvestingand animal and human manure was recycled into and agriculture, advocated in 18—19th century western Europe organic gardens(Redclift, 1987: 109—110). At urbancentres and its colonies. Le Goff (1984: 331) asserts that the urban clawing on to mountain slopes (Buritaca 200, Colombia; centre-countryside contrastwas strongerin Medieval Europe Cuzco and Machu Picchu, Peru), extensive retentionwalls, than in most other societies and civilisations.Yet, he found terrace gravel beds and stone-lined drainage afforded that the Medieval urban centre contained within its walls intensive farming on steep slopes (Coe, Snow and Benson, vineyards, gardens, pastures, fields, cattle and manure. 1986: 166—167 and Butland, 1976: 162). In the late 1800s, Gathering trends at the end of this century point to the Paris still had 5,000 people working 860 hectares of highly

139 Introduction—LucI. A. Mougeot African Urban Quarterly productivemarket gardens which recycled and composted whereoften open-spacewedges betweenresidential quarters human and animal wasteand used irrigation, raisedbeds and were cultivated (a practice resumedin post-colonialcapitals glasshouses,supplying two million Parisianswith fruitsand eitherspontaneously or withexternal support in recentyears) vegetablesand exportingexcess to London (Girardet, 1992: (Winters, 1983: 14—15). Historical accounts are fewer of 52). urbanhorticulture in higher-density, more trade-boundcities Food provision was an enduring concern of urban underislamic influence, butother forms of urban agriculture, populationsand their administrations,given urbancentres' such as aquaculturein coastal ports or animal husbandry in •vulnerability to insufficiency or disruption of imported arid regions, were probablyquite important, as observed in supplies,hence to hunger or even famine. Roman authorities contemporary cities. of fast-growingurban centres commissioned complexes of Neglected by 20th century developmentagencies, urban garden houses likely for middle and lower class households agriculture historically has interacted with rural agriculture (Ostia, Rome) (Watts and Watts, 1994: 86—89). Barbarian to the mutual benefit of both. From the evidence, many invasions during the High Middle Ages would teach local ancient urban centresprobably provided the incentives and governmentsthe need to strengthenfood self-reliancewhen testing grounds for specialised, intensive farming systems. rebuilding urban centres later in the period. This happened Some advances have been borrowed by urban agriculture, even before then: when substituting themselves to urban adapted, tested and devolved back to rural agriculture. centres, monasteries developed as semi-autonomous units Technological breakthroughs in ancient urban centres have stressing food self-reliance (Guidoni, 1981:42). For instance, included sun reflectors, water collection, storage and the 9—10th century plan of Saint-Gall monastery at Fontenay, conveyance; frost protection, wetland drainage and slope Cluny, clearly shows specialised quarters for herborists, terracing. Large-scale canal networks clearly seem to have gardeners,poultry yard, cowboys, sheep, pigs, goats, horses, followed the advent of fully established urban centres cows and latrines (Le Goff, 1984: 144—145). (Adams, 1994: 15). Even today, urban centres serve as In the 8th century, Rome was growing again and faster diffusers of more efficient greenhouse, aquaculture, than it could feed itself. Pope Zaccharias' (741—752) plan to hydroponics and dairyingtechnologies to rural areas (GRET, repeopleand rebuildthe city included five 'domus cultae' or 1991 and Marulanda and Izquierdo, 1993). large farms, near main arteries several miles apart one from Thedivorce of agriculture from urbancentres and of food another, all around the city; six would be implemented by production fromurban economies is very recent in our urban Pope AdrienI by century's end, one of which entirelyto feed history. It also has not been universal.It may dateback to the the mass of workers needed to repair and restore Rome's European Renaissance, having spread throughout Western infrastructure (Guidoni, 1981: 98—99). Elsewhere,icons of Europe in the 18th century and becomingentrenched by the important cities, such as Novgorodor Cologne depictedwell- late 19th, when it also started to be introducedin European spaced housing, gardens and orchards, some fenced up to colonies. In colonial Africa, gardening and some animal keep outsmall livestock,located within inner and outer walls husbandry were often permittedon private plots but official (Guidoni, 1981: 113; Jellicoe, 1989: 34; Yanin, 1994: 123). initiatives were minimal for incorporating food and fuel Several contemporary African urban centres are making production into low-income housing plans into the efforts to set aside zones for pen-urbanfood production and management of institutionalestates and public open space some are establishingthese on wedges close to high-density, or into urbanwide zoning. In East Africa, Kironde low-incomedistricts. Planned pen-urban market gardening (1992:1277—1292), Lee-Smith and Memon (1994: 67—84) and is well-developed in Asian urban centres. Mbiba(1994: 188—202) havereviewed the sanitaryand ethnic Throughout most of mankind's history and in quite justifications for a colonial legislation largely inimical to differentcultures and climates,urban dwellers have engaged urban agriculture. to varyingextents in producing some of thefood they require, In the last 20 yearsurbanisation has been challengingthe close to or at their own residence,within or fringing urban practicalityof urban centres' exclusive reliance on distant, centre limits. Food production has not only been carried out costly and unreliable rural food production, domestic and by better-off households; those in authority often have foreign. Urbanisationhas also been challengingthe morality commissioned,built and managed massive food-producing of deprivingthe urban poor from accessing idle urban land systems,even incorporatingfood production into the design for feeding themselves and others. As the sanitation-without- of modest living quarters. agriculture approach is failing to solve the urban waste Pre-colonial African urbanismremains under-researched, management crisis, urban authorities worldwide are again to say nothing of urban agriculture. Ancient and Medieval turning theirattention to the sanitation-with-agriculture option urbanismseems to have been well-advanced in West Africa as witnessed by the planning of a 1996 World Bank and medievalreports on Africanurban centres pointed to the Symposium on the subject. The urban centre/agriculture major role of agriculture in what were low-density, shifting divorce is showing signs of weakening both in the North and urban centres with important political functions. In West the South. Africa, intensive pen-urban farming enabled urban centres to achieve near food self-sufficiencyin the face of climate Trendsin African Urban Agriculture: This Issue in constraints to long-distance transportation and long-term Context storage and littledependence on trade for daily food supplies. Surveys show that since the late 1970s, urban farming It was even more important in Eastern and CentralAfrica, has been growing inmany countries, in terms of practitioners,

140 Volume 11 Nos. 2 and 3, May andAugust, 1996 An ImprovingDomestic and International Environmentfor African Urban Agriculture spaceused, contribution tohousehold welfare and tothe urban Urban povertyand food insecurity will expand considerably economy in general (Mougeot, 1994a; Tevera, this issue).A for several factors, from demographic to fiscal. plethora of factors explain this growth: rapid urbanisation, Strategic to urbanagriculture's functionality in theAfrican crippled domestic food-distribution systems, national policies urban foodsystem are the ways in which marketing networks harmful to domestic production, reduced public spending, try to adapt to urbanisation under several constraints and subsidies and wages, retrenchmentof formal employment, opportunities. Becausedemand is ruled more by affordability soaringinflation and plummeting purchasing powerof middle than availability, it can be assumed that different income class, lax urban regulatoryenforcement. Civil strife,war and groups will resortto differentmixes of acquisition modalities, naturaldisasters also disruptrural food production and supply jointly determining the relative contributionsof rural, urban lines to urban centres. Recentstudies have further linked the and foreign production areas. Although acquisition and growth of urban agriculture to structural adjustment production are interrelated,these have usually been treated programmes and major currency devaluations in the late as different worlds in the literature. The second article of 1980s and early 1990s (Drakakis-Smith, 1995; Shapiro,Ehui this issue, by Ilda Lourenco-Lindell, attemptsto correct this and Fitzhugh, 1995: 109—119; Lourenco-Lindell; Atukunda gap. on urban and Maxwellin this issue).Conditions sufficient to dampen Lourenco-Lindell focuses the growing poor the resurgence of urban agriculture appear increasingly population and analyses household food consumption and unlikelyin most of Africa. together with household involvement in food production distribution.In order to do so she uses a broadenedconcept entitlementwhich (1) The Role of Urban Agriculture in the Urban Food of encompasses,beyond possession-based entitlements, other behaviours rooted in social and cultural Supply System and inthe Urban Poor'sFood Provisioning codes, such as charity and safety networks, institutional Strategies conditionsand Three contributions to this issue initially position this illegal practices. This enables to document the resurgence of urban agriculture within African urban food approach Lourenço-Lindell and the African urban food directcontribution of urbanagriculture toprovision strategies supply systems poor's for subsistence and sales.It indicates provisioning strategies, as well as the impact of urban through self-production on households'nutritional health. how central urbanself-production can be to such strategies, agriculture self-producing the in which be influenced The first article of this issue Joe Nasr and Annu Ratta, through many ways producers may by welfare and outlines the recent and future of urban in (mutual help among producers, community growth farming funerals and in formal and non-formal sub-SaharanAfrica's urbanfood groups) may participate supplysystems during rapid channels of in the urban food urbanisation. All as defined Drakakis- acquisition supply system. systemcomponents by Farmers and additional/seasonal Smith areas generate employment (1990: 100—120), including food-producing incomefor other basic needs link with rural and urban and networks (processedfood), up (domestic foreign),marketing the food trade, foodstuff otherwise unaffordable, and urban are affectedin and produce consumptioncentres, quantity reduce on food, enhance their own urbanisation. The with which the dependence purchased quality during ability exchange entitlement, provide food gifts and meat sharing. network to urbanisation marketing responds (growing They also, although unwillingly, contributeto food insecurity more concentratedand food spatially qualitativelychanging mitigationthrough theft of their crops, animalsand assets by affects the relative mix of methods demand) acquisition third parties, a problem far from insignificant in several and differentincome (exchange, production transfer)among surveys. and the of areas. While groups system's geography supply Urban food security has been shown to deteriorate in networks evolved from and small-scaledto petty capitalised several urban centres of the developing world, being often and large-scaledin developedcountries, the authorscontend worse than in rural areas among low-income people (von that in the world the traditional structure developing supply Braun, MeComb,Fred-Mensah and Pandya-Lorch, 1993). often with the new structure.It is this overlaps development Urbanagriculture is commonly argued to improve the urban which accommodatesthe resurgence of urban agriculture, poor's food security. However, few studies so far have big and small, in the African urban food supply system. examinedits actual impacton nutritionalhealth of producer The mix of acquisitionmethods and of supply areas is households. Thethird articleof this issue, by Mboganie Alice mediated by several policy and technology factors, both Mwangiand Dick Foeken is on Nairobi and supplements an positive and negative.According to Nasr and Ratta, several earlierstudy in Kampala(see Atukunda and Maxwell,in this macrotrendswill sustain urban agriculture's growing role issue).In both studiesnon-farming control groups were used; during the next 30 years or so, particularlyin sub-Saharan Kampala focused on anthropometricmeasurements while Africa. Thefastest urbangrowth will occur in thosecountries Nairobi added quantified nutrient intakes. In both cases, least equippedto feed their urban centres;within less than a farming householdsjustified their activity for their need of generation, Africanurban centres willcontain as many people food, regardlessof other employment.In Nairobi, while all as the whole continent holds today. This will take place groups had inadequate energy intake, the percentage of despite lagging economic growthand slow developmentof households with inadequate proteinintake was highest in the marketingnetworks; a gaj is widening between population non-farming group and differences in averageintake between growth and staple-cereal production growth, in the face of farmingand non-farminggroups worsenedwhen correcting limitedforeign exchangefor importsto make up for deficits. for householdsize. The higher energy and proteinintake of

141 Introduction—LucJ. A. Mougeot AfricanUrban Quarterly

farmers resultedfrom self-production (thesewere even better- recently have been inclining local authoritiesto realise the off materiallythough their cash incomeswere comparable to scale of the activity and revise earlier dismissalsas a short- thoseof the non-farminggroup—thanks totheir reduced food lived import from rural culture. Very few urban centres purchases). Measures of nutritional health for children otherwise possess and publish statistics on local urban (weight-for-height and weight-for-age) showed similar agriculture;existing records are dispersed among agencies, patterns favouring the farming group and this group's lead largely underestimating informal activities and excluding even increased when controlling for age distribution. some product categories,scales of production and areas of Although averages among the three groups do not vary much, the urbancentres. Still, early official initiativesdate back to these hide very different distributionsof data; malnutrition the 1970s and even before, prior to the publication of the is less widespreadwithin the farming group. first surveys by the United Nations University and the International Development ResearchCentre of Canada. (2) Incorporating Urban Agriculture into Urban A case in point are East African surveys, which have Management; Trends inAfrica resoundeddifferently among official circles depending on Several trends are underwayon the continentwhich bode which country is considered.The fourth article of this issue well for the incorporation of urban agriculture into African by Dan Tevera, introduces the reader to surveys in several urban management policies well into the 21st century. A eastern African urban centres. It notes that information second group of contributionsto this issue reflect some of availableon the considerablegrowth of urban agriculturein those trends. Zambia, Zimbabwe and Kenya shouldbe used todraw lessons as urbanagriculture spreads to other countriesof the region. (a) More African Urban Governments are Recognising Research on the spatial extent and location of urban Urban Farmingas a Permanentand Growing Function agriculture, socioeconomic characteristicsof the cultivators of the Urban Space Economy. Recognition is a requisite and theeconomic importance of theactivity, is found wanting for legitimation,legislation, promotion and regulation.As of more attention to rural-urban linkages (see Lewcock on opposed to ignorance or denial, recognition implies a Kano, 1995: 225—234) and to the relationshipbetween urban conscious association of conceptwith empirical observation. agriculture, wastemanagement and human health (Niang on Even though a local authority recognisesurban agriculture, Dakar, this issue). Despite the spatial growth of urban it may not necessarily approveor endorseall or even part of agriculture,the poor do not benefit much from it and most it; however, recognition usually has been associated with urban farming remains officially unsupported. Tevera greater support to 'legal' farming,even greater tolerance of advocates thaturban farming be encouraged under controlled 'illegal' farming.When producers agreethat 'illegal' farming conditions.This is the approach which most known local is strictly transitional where it is practised, councils often initiatives underway in the region seemto be adopting (Sawio, have been able to accommodatethese as prelude to planned 1994; Gumbo and Ndiripo; Atukunda and Maxwell, this development, without violating or having to modify issue). prevailingbylaws or zoning plans. The Harare City Council does tolerate, in council housing projects, the cultivationby (b) A Review of Policy and Planning Options Often residentsof vacant land slated for non-residential use while Follows Official Recognition of Urban Agriculture, this land awaits development. Additionally, the extensive Leadingto ConditionalLegitimation of Existing Activities, croppingof roadsides,public playgrounds, parks and police to Incorporation of New Activities into Planningand to stations witnessed throughouthigh density districts during the Regulation of Both Current and Planned Activities. 1996 (May) rainy season testifies to the absence of any Changes in policy have tended to precede changes in significantrepression by local authorities. planning. Policychanges are found inpoliticians statements, The process by which local authoritiescome to recognise councils' actualinterpretation or enforcementof bylaws and urbanagriculture are still unclearand Joe Nasr suggests more promotionof practical initiatives on social or humanitarian research is needed to clarify processes which lead to such grounds. In times of hardship, presidential and local attitudinal or practicalchanges. Whichare more determinant: governmentaldeclarations (e.g., Zambia, Tanzania, Zaire, self-reliance in the face of national crises (wars, droughts Zimbabwe)have urged people to become more self-reliant and structural adjustment), external funds and visibility in rural and urban areas. Supportfrom local technocratshas afforded by international humanitarian projects, often been lukewarm; however,electoral processes in newly powerlessness againstbeyond-point-of-no-return scale which independent countries have been making ministers and the activity may have reached urban centrewide,eagerness councillors more accountable to their constituencies (e.g., to emulate 'more advanced' urban centres, key decision- Uganda and Zimbabwe); interventions have become more makers' own first-hand experience with urban agriculture sustained and comprehensive. (many are farmers themselves) and its effectiveness to Between 1975 and 1985, governments in six African alleviatepoverty and mitigatemalnutrition or faminein his! countries supported communal gardening: Mozambique(land her family, neighbourhood,riding? provision), Ethiopia (water, seeds and fertilisers), Zaire Surveysduring the 1980s have contributed to estimate, (garbage composting, nutrition education and coops to characteriseand publicise the industry, putting numberson streamline marketing), Morocco (tax exemption), Zambia peopleinvolved, areas usedand value of output,contribution (land tenure to squattersfor home food production),Kenya to household economy. These and subsequent surveys (nutrition awareness and income generation with crops)

142 Volume 11 Nos. 2 and 3, May andAugust, 1996 An ImprovingDcmestic and International Environmentfor African Urban Agriculture

(Wade, 1987: 38-41). The new capitals of Côtd d'Ivoire, Programmes should prioritise newcomers and projectsshould Malawi and Tanzania have been designedto accommodate be sited so as to use and protect, if not reinforce, current urban agriculture and their authorities are encouraging it. local networks. Sectoral studies on urban agriculture provided grounds for Households fully dependenton food self-productionare revising master plans of major existing urbancentres which uncommon, but self-production is key to household in turn offered the frameworkfor practical initiatives as in maintenance; its availabilitygreatly affects the household's Maseru, Lesotho(Greenhow, 1994: 2) and Kampala, Uganda spendingpattern. Hence, urbanfarming programmes should (National EnvironmentInformation centre, 1994; Atukunda be promotedamong certain groups for their use value and and Maxwellin this volume. The 1975 revisionof the 1968 fungibilitybenefits at least as much as for the exchange value master .plan of Kinshasa set aside areas for horticulture in of their products. Urban agriculture may also increase the east, central and southwest sections of this city and women's financial independence and theircontrol on related provided the framework for support programmes incomemay buffer poor decision-making by other members aforementioned (Pain, 1985: 34). In Durban, Republic of of the household. However, control of production is not SouthAfrica'ssecond largest city, ametropolitan open-space automatic but rather follows from self-esteem and perception: systemdeveloped out of a 1983joint study between the City's assistance should further the developmentof women's true Engineer Departmentand the Universityof Natal, now links worth and contribution to household livelihood. nine municipal parks and incorporates urban agriculture Dennery also found that assistance for non-agricultural programmesfor food and income (Roberts, 1996: 5—6). tasks which enables the producer to devote more time to It is becoming clear to a growing number of urban farming is easier to obtain than for plot-tending;so actions governments that colonial regulations (environmental and particularly aimed at women producers, which stress sanitary) and prohibitive bylaws have remained largely household-related cooperatives (child-care, cooking or water- unenforceable, while in-field repression (harassment, slashing proision)can greatlyimprove farming labour allocation and and eviction)is becoming less dependable on humanitarian productivity. and political grounds. More governments realise that Producers docollectively engage in low-cost interventions disregard, outlawing and persecution have undercut to mitigate or avoid urgentthreats to their activities. This is proclaimed intentions: much urban agriculture has had to not uncommon on other urban issues: popularorganisations develop under conditions now revealed as ecologically are extremely plastic, coalesceand dissolvearound specific degrading,sanitarily unsafe, technicallyinefficient, socially emergencies. Attempts to introduce permanent, formal inequitable and economicallyexploitative. participation models from highly structured civil societies Crucialto enlightenedpolicies and programmesfor urban have toooften failedbecause they ignorethat informalways agriculture is an intimate knowledge of urban farmers' are the people's response to costly, ineffective and often decision-making.Very little of this has been researched so corrupted formalorganisations found locally. Dennery rightly far and the fifth article of this issue, by Pascale Dennery, warns that higher-level producer organisations will only precisely focuses on poorproducers' use of open spaces and succeed if they fully understand and build on the existing theirdecision-making in Nairobi, Kenya.Her findings carry organisationalculture: For instance, many organisational very practical implications;policy-making can be less than forms are not strictly food production-oriented, which were appropriate if it does not account for local conditions observedto actuallysupport it variously in Kibera,and vice- influencingproducers' rationality. versa. For instance,large households assign at least one member mainly to food production;it follows that self-production is (c) More Urban Centres are Reviewing and Adapting a socially valueduse of labourtime and thatrepression among Assimilated and Acquired Technical Planning Norms certain groups can severely underminetheir food security. AffectingUrban Agriculture. This is whereenabling policy Multiple-plot farming is an accepted practice because it trickles down into official supporting practice. Beyond the enableshouseholds to spread theft, eviction or bad-croprisks; inclusion of urban agriculture into master plans or sporadic it also affords a diversified or more continuousproduction. social/food emergency interventions, more urban centresare Theseare arguments for theacceptance of dispersed farming, creating permanent institutional programmes and agencies acondition whichshould increase and facilitate land provision and stressing synergiesbetween urbanagriculture and other urbanwidefor urban agriculture. activities within broader urban action plans. This responds An important observation in Kibera is that most open- to a need for indigenising African urban planning(Kironde, space producers are defacto tenants abiding by customary 1992). As synergies between urban agriculture and other usufructprinciples, not land-grabbers. And producers mainly urban sectors are valued, concepts need to better reflectand dependent on urban agriculture stand to lose the most if serve local conditions; standards need to be modified beyond evicted. These two findings suggest that producers may be urban agriculture, extending also to sanitation, water and willing toconsider usufruct withoutownership (public-private housing supply as well as urban transportation. partnerships); also some indemnification of legitimate More researchers are exploring zoning modalities for a occupantsis in order where displacementcannot be avoided, smoothtransition from rural to urban land-useon the urban particularly for thosemore dependenton theactivity. Kinship, centre's fringe.This includesflexibility to allocate,displace, ethnicity and belonging to the right social networks were upgrade, possibly combine or integrate urban agricultural found to be crucial to accessing production inputs. functions with other land uses as the urban centre grows. In

143 Introduction—LucJ. A. Mougeot African Urban Quarterly addition to the on-plot residential gardening, many urban running farmingcooperatives for sometime; several public— centres are adding communal programmes of urban private partnerships are already at work: in Lomd (port agriculture, namely, throughthe productive use of open space, facilities). Dakar (waterutility concessions), Durban (health (e.g., horticulture and animal husbandry in pen-urban clinics and factory premises), Dar es Salaam (church group greenbelts) as in Maputo (Wolfgang Stiehens, personal estate), Morogoro (university campus). These need to be communication December1995) and Durhan (Roberts, 1996: evaluated and disseminated to other urban centres, as 5—6). Secondary urban centres have often innovated, with synergies with other land uses can create win-win solutions various forms and degrees of support in Kenya (Lee-Smith for partners, with economies of scale toproducer groups likely and Memon, 1994: 83),Tanzania (Mvena, 1986), Ivory Coast to benefit consumers as well. (Della, 1991) and SouthAfrica (Rogerson, 1993: 25—26).The An important point made by Cockram and Feldman is City of Gweru, with the assistance of a national agricultural that although group forms of gardening may take place on extension agency, identified undeveloped council land communal land, this is not automatic.Several urban centres suitable for agriculturalactivities, divided it into small plots in the region insist on the need for farmers to organise and leasedthese out to for a nominal fee on a producers yearly themselves and register as groups in orderto be granted public basis (ENDA-Zimhahwc, 1996: 13). Early provisions should or council land (role for local GOs and NGOs). Consistent small enable these cities to manage further growth more with Dennery's findings, their article notes that women can effectively. use group gardening as forum to recognisestrategic gender Urban are also with governments addressing, growing interests. Also successful projectsoften stress local materials, involvement from non-governmental stakeholders, the need crops, methods and sustained technical assistance, for tenure or usufruct Creative lcaseholds improved stability. inexpensive and accessible inputs. Itconcludes with arange as tried (i.e. non-residential,purpose-specific successfully of possible official actions, from cheap to more costly, to in the North and the South) encourage knowledge and the of communalurban material investment farmers into land improve efficiency agriculture. by conservation, Related to Cockram and Feldman's review, the seventh productivetechnologies and profitablecrop selection. They articleof this issue DavisonGumbo and Takawira also farmers' them to credit and by Ndiripo legalise activity, entitling examines the recent of cultivation in technical and assistance, and growth open-space training crop protection Harare. This is a highly seasonal and mobile form of urban investment indemnification. In Tanzania, the National cultivation. Still associatedwith the nature of most will livestock shifting InsuranceCorporation soon extenda insurance urban in African urban centres to to urban areas (Camillus Sawio, personal agriculture (as opposed policy backyard gardeningin LatinAmerica and pen-urban market communication,May 1996). belts in Asia),open-space cultivation in many African urban Reflecting the growing interest for group forms of centresmay providethe canvas for a flexiblesystem of more agriculture on open spaces, the sixth article in this issue by or less temporary communaland market gardens. Mary Cockram and Shelley Feldmanemphasises communal As urban centres of post-apartheid Republic of South gardens,cooperative farms and marketgardens. This may be relevant to African urban centres which need to with Africaand other countries look into making more productive cope use of their Zimbabwe's offers sizeable social/food security challenges and can capitalise public open spaces, experience a useful reference. on traditional grarian institutions and accommodating urban Despitehistorically stringent regulations, centre open-spacecultivation has grown to become a major land- layouts. use metro Harare and few local officials would Theprivate home-plotgarden islegally endorsedin most of today countries and established and mid-income discard it as ephemeral.Between 1990 and 1994, cultivated firmly among high- doubled its 17 of groups.In Dakar, however, one NGO had to stop supporting open-space area, covering nearly percent metro Harare.Areas close to industrial districts shrankwhile backyard gardening as it was considered off-zoning and others close to districtsand and producers were threatened .with fines by public health high-density along roadways It should be wherever in new waterways have grown, even in the CBD and on parkland in inspectors. promoted possible north of the CBD housing schemes (Greenhow, 1994). Under prevailing low- upscale neighbourhoods (ENDA- income housing patterns and until confined-space Zimbabwe, t994). from the A technologies become widely available to the poor, the Someplanning implications derive findings. be located within contributionof backyard gardeningto the poor's urbanfood really extensive food-growing could security will remain limited. walking range of peripheral high-density neighbourhoods Most urban poor have little land of their own to produce where house plots lend little space for growingfood; intensive the food they need. But surveys reveal an enormous amount space-confined (soilless, vertical, rooftop) systems could were found to of public and institutional, even private,land availableat any thrive at more central locations.Homeowners one time for food production (unbuilt or unconstructible). be in bettertenure positionto establish claim on nearby off- Communal forms of land tenure or usufruct have much plot fields,closeness topermanent residence being important potential to improve the poor's access to urban agriculture, tomanagement and protection.The zoning of some areas for whereopen-space cultivation is already widespread or rapidly communal gardens or even market gardens would require growing. Community spaces and surplus or reserve public minimal usufruct guarantees to ensure that producers invest and private spaces (utility and institutional estates included) into productiveand conservationistpractices, including the are prime candidates to group forms of urban agriculture. use of compost-manure instead of or with chemical fertilisers. City council departments in Harare and Durban have been More productiveplots would also mean that these could be

144 Volume 11 Nos. 2 and 3, May andAugust, 1996 AnImproving Domesticand International Environmentfor African Urban Agriculture smaller, thus accommodatinglarger numbers of households At a micro-zoning level, bylaws must account for than currently. household-level integration of different productionsystems Intra- and periurbananimal husbandry is growing inmajor as well as for complementary on- and off-plotfarming, typical urban centres of the region (e.g., Dar es Salaam, Kampala of African urban agriculture. Closing previous system- and Bamako)and raises severalplanning concerns(GRET, specific articles, the ninth article of this issue, by Axel 1991; Mougeot, 1994b: 105—115; Centresand Rolland, 1995: Drescher, examines this complementarity on a household 137—148; Metzger, Centres,Thomas and Lambert, 1995 and level in low-income high and mid-densitysquatter quarters Shapiro,Ehui and Fitzhugh, 1995). The eighth articleof this of Lusaka. The term 'urban microfarming' refers to crop issue by Ann Water-Bayers explains the pervasivenessand production, homegardening, horticulture and livestock growth of urbananimal husbandry: high-valued multipleuses, keeping and is supplemented with wild fruit and vegetable large variety of species and systems, mobility of animals. gathering. Noteworthy is that wildfood gathering is generally Variants of systems are affected by main proiuction aim, disregarded in urban agriculture literature: 40 percent of scale, intensity, time allocation to activity, dependence on Lusaka intra-urban respondents said they practised it. This wage labour, genetic stock, husbandry methods, tenure includeswild-grass cutting,seasonally important to feeding modality. Functionsof livestockfor different incomegroups urbancattle as reported on Bamako, BoboDioulassso (GRET, vary but these functions aremore numerousand determinant 1991 and Centres, 1992) and witnessed in Dar es Salaam. on the poor's livelihood. Micro-enterprises and small- All these activities are different in terms of timing, labour livestock populations are growing despite preferential organisation and gender allocation, While in rural areas, treatment given to agribusiness operations. Staal and gardens and cropfieldsco-exist separately, in urban centres, Shapiro's (1994: 533—549) study of urban dairy husbandry dry-season homegardens may convert into wet-season in Nairobi illustrates hov the traditional and largely cropfields. Rainfedopen-space crops in Lusaka use land fully; unregulatedsupply structure,referred to by Nasr and Ratta croppingpatterns and crop choices reduce lossesdue to theft can capture markets, whenever regulatory attempts breed and drought, pointing to information flows among farmers monopolieswhich raise costs and supply-disruption risks to of different areas of the urban centre. As in Harare, urban consumers. Urban animal husbandry exchanges homegardens use animal dung and serve as nurseries for resources between urban and rural areas, with cropping species transferred to ridge-bed fields in the early wet season. systems (manure collection at compost plants and schools Termite-mound gardening in natural wetlands exemplifies for gardeners) and with agribusiness (provides inputs and site-specific opportunism in the urbanenvironment. Drescher uses processingwastes). recommends that access to resources be improvedas opposed Thereare concernswith human health, safetyand sanitary to markets (which don't seem to be a problem) and that risks of urbanlivestock, still scarcely documentedor verified. preferencebe given tomultistakeholder approaches involving Less so have these risks been assessed against other factors farmers, extensionists, decision-makers and urban planners. (includingill-enforced standards in other activities) affecting people'shealth status, traffic safetyor ecological quality; none (d) As Urban Centres Tackle Synergies Between Urban has been weighed against the benefits to producers and Agricultureand Other Urban Activities, the Former's consumersalike. Role in Urban Resource Recovery Is Drawing Growing Still a major information gap persists which could stall Attention in the Region. Conventional municipal solid and the fuller development of animal husbandry's potential in liquid waste managementremains administratively highly Africanurban centres. Granted, urbananimal husbandry has centralised, capital-intensiveand deficit-ridden, with only grown in recent decades despitehealth and safety concerns; partial and unreliabletruck collection of unsortedmaterials. these have regally failed to curtail its expansion as nutrition An even smallershare of collectedloads eventuallyends up and income benefits have spoken louder. However, where at few, often remote, sites hazardous to scavengers. the activity already exists, every effort should be made to Few urbancentres have encouragedat-source sorting and modify or introducepractices which will make urbananimal pre-collection of organics by organised groups of urbancentre husbandry safer,fairer and more profitable to bothproducers, farmers for district-level composting and/or application to consumersand the physical environment. intra-urban plots. Few urban centres have made efforts to Waters-Bayer's article concludes with very specific sell and delivertruckloads tolarger clients on the urbanfringe. initiatives throughwhich governments can improve their role. Lewcock (1995: 225—234) found that waste application to The aim is to have more disadvantagedgroups increasetheir pen-urbanagriculture has a tradition in Kano and pen-urban access to veterinary services,extension and information on farms are a developing market for large quantities of human health risks as well as their capacity to access urban minimally composted waste. This is preferred on rainfed plots centre resources(via multistakeholdercommunications and and as a complementto artificial fertilisers (crop growth): it negotiations).It is also for them to defend their interests, to enhances long-term soil structure and moisture retention. Pen- participatein the development of technologies for small-space urban farmers are willing to pay substantial amounts for production (time-saving husbandry for women) and acceptable waste and greater collaboration is required organisationalforms for more effectiveaccess to inputs, safe betweenurban wastemanagers and pen-urban users to fine- waste recycling and open-space management. The author tune safetyand delivery to their mutual benefit.On the other stresses small-scale operations using local and non- hand,WASTE's recent survey of urbancomposting in Africa conventional feeds, animal species selection accordingly, concludes that neighbourhood-level and micro- facility improvementand decentralisedfarmer markets. entrepreneurialcomposting has been workingeffectively in

145 Introduction—LucJ. A. Mougeot African Urban Quarterly

several urban centres, while most larger plants have failed. unrestricted gardening. Parasitological analyses of vegetables Multiplying small-scaleprojects in any given urban centre wateredwith untreated wastewater showedthat produceready will requirelarger local markets,more transfersites for pre- for sale contain eggs and larvae of parasites harmful to collection of solid waste in each neighbourhood, as well as humansif consumed raw. Also domesticlivestock which are labour-intensive collection systems (Hart and Pluimers, eatenup by residentsdo drink atuntreated wastewater ponds. 1996). Since no single known treatment technology exists for At-source sorting and doorstep collection are crucial to meetinghorticulture requirements of safety, productivityand increasing volumes and improving thesafety of organic waste affordability, theCambérène Stationis developing a sequence reuse in urban agriculture.This quickensat-source retrieval of treatmentcomposed ofelements from differentprocedures of organics. lowering health risks to collectors, producers (manuring under bare ground, microphyte lagooning and and consumers, includinganimals. It also reducesthe weight Pistiafloating mats) inorder to optimise agood purification, and humidityof the remainingsolid wasteload, renderingits disinfection according to WHO norms, moderate reduction disposal, incinerationor recycling cheaper, safer and more of azote and phosphorus, balance of azote and phosphorus, effective.In most African urban centres, solid organics are as well as sufficientproduction of water. so valuable to farmers that where at-source collectionis not Thepublic and environmental health risksposed by urban more extensive,this is usually due to major transportation, agriculture are a major concern among policy-makers contractual or regulatoryconstraints. Where at-source sorting throughoutAfrica. This concern is often an argument for is the exception, collectors of organics prioritise food- trying to write off open-space cultivation, even backyard processingand retailing businesses where little sorting is gardening in some urban centres. Much of those risks may needed and usable quantities are large. stem from improper handling of agrochemicals by producers, Several African urban centres operate municipal application of unsorted or insufficiently treated solid and wastewatertreatment plants coupledwith agriculturalreuse. liquid organic wastes to specific crops and crop selection! Fewer have worked with groups of horticulturists or location without due regard to site exposure to ambient air aquaculturists who reuse the more abundant untreated sewage and soil pollution. Still under-researched, this concernis likely to come up with affordableschemes for meetingintensive- to grow as more urbancentres try to reuse as much solid and production needs of those groups. liquid waste as possible to curb pollution and optimise The tenth articleof this issue by SeydouNiang discusses freshwaterresource use. In Harare where this is definitelyan the initial results of an ongoing research at the Cambérène issue(Gumbo and Ndiripo,this volume), preliminary findings Experimental Stationof theSociete nationale d'épurationdes from Bowyer-Bower and Tengbeh's research (1995) show eaux du Senegal (SNEES) in Dakar. An initial phase of this that open-space cultivation contributes more to land research was supported by IDRC. The situation in Dakar is degradation thanto pollution itself. Chemicalpollution effects found in many other sub-Saharan urban centres, where of industrial effluents and discharges, illegal dumping, untreated wastewaters are reused directly into horticulture housing sewage, and surface water are far more serious on and the produce is sold on local market fairs. In Dakar, surface water than the worst possible polluting effects of drought and soil exhaustion have caused horticultural chemicals used in urban agriculture. Agrochemicals are production to decline; given the growthin volumes of urban expensivelocally and theiruse is still very limited: potential wastewatersand their richness in fertilising elements.These pollutionfrom that source will remain low in the near future. are becoming an alternative,sometimes the sole source of On the other hand, open-space cultivation has reduced water. Public health risks are real for producersthemselves rainwater infiltration into thesoil, particularly in long-cropped and their families in the field, product handlers,consumers sandy granite soils. It has multiplied, thanks to a crusting and residents near fields. Occurrences of epidemics effect, surface runoff manyfold;rainspiash and runoff have attributableto mispractice are found around the world. In increasedenormously sedimentmovements, soil erosionand Niang'sview, simple prohibition isnot viable and legalisation siltation (steep-slope open-space cultivation). However, would allow to better manage and lower its risks. Several cultivation techniques influenceconsiderably soil erosionand actions are in order: pre-reuse treatment of wastewaterby the majority of plots have acceptablelevels of loss. Ridges extensiveand affordabletechniques, confinereuse to certain and furrows could bringerosion rates within tolerablelevels crops less sensitive to disease transmission (fruit trees), elsewhere. modify irrigation techniques(avoid submergence) and limit Initial evidence in Harare points to a need for both human exposure throughequipment and education. technical (propersiting, recourse to agroforestry) and policy The study looks at the impact of wastewaterirrigation on (land tenure/usufruct security) interventions, in order for cultural practices and sanitary quality of produce. It open-space cultivation to become more productive and characterises different types of wastewaters used in profitable, and to reduce erosion and siltation and expand horticultureand the impact of their use on producer's water the tree cover in urban centres. needs, crop selection, irrigationmodalities, quantities of water and inputs required (parkiandmanure, peanutmilling waste (e) More Urban Centres are Reso'rting to and fish manure).Physical, chemicaland biochemicaltraits MultistakeholderConsultations to Set Action Courses of wastewaters make them particularly interesting for Regarding Urban Agriculture Policy, Planning and horticulture and they are water and input saving. Management IssuesDiscussed in Items (b), (c) and (d) of Bacteriologically, faecal coliforms and streptococcus the Introduction. The City of Dar es Salaam has gone from concentrationsmake them unsuitedper WHO standards for early inclusion of urban agriculture in its 1968 master plan

146 Volume 11 Nos. 2 and 3, May andAugust, 1996 Animproving Domesticand International Environmentfor African Urban Agriculture

to running a multistakeholderexercise to enhance synergies as a legitimate land-use in the master plans of Jinja and between urban agriculture and other urban sectors for the Kampala. The recently adopted long-term plan for Jinja management of future growth (Dar es Salaam City Council promotessome land use for agriculture;since late 1994, new and Centre for Housing Studies, 1992 and Sawio, 1994). land-use regulations adopted by the KampalaCity Council Several public consultationson urbanagriculture have taken now explicitly permit "homebased agriculture,including the place or willbe held in the regionin the near future:Kampala growing of crops and keepingof livestock". (1993), Dares Salaam(1993—today), Harare (1994),Pretoria Fundamental to a favourable shift in officials'attitudes (1994), Ouagadougou (1995) and Accra (1996). A observed by Atukunda and Maxwellduring the 1988—1993 multistakeholder conference organised by the Pretoria period was the active local-level militancyof urban farmers Technikon in Februaryof 1994 and attended by 400 people and political support by electeddistrict councillors (also in had most papers assessing the potential of urban agriculture Harare in the early 1980s, Mbiba, 1994: 196—197). In this as a major productive use of public space in post-apartheid now more receptive environment, theauthors stress the need urban centres of the Republic of South Africa. As most to still convert urban planning from atechnical into abroadly- consultationsare recent and follow-ups to these even more based political process toformulate and enforce bylaws which so, little is known yet of theirpractical outcomes. No doubt can manage a real urban centre, changing bylaws from that part of the future research agenda in Africa will be to ineffective banning paper tigers into instruments for monitor, evaluate, disseminateand improvethe effectiveness correcting detrimental aspects of informal practices. Some of such exercises. policy reforms couldthreaten access to landby informalurban Symptomaticof this trend, several articles in this issue agriculture and such risk should be curbed through stress the need for research to be more participatory and appropriate zoning,compensation and reallocation of evicted produce information on key local roadblocks for issues activities. The articlepresents various methods through which prioritisedthrough multistakeholder consultations. a dialoguecan be furthered among urbanfarmers, researchers The eleventharticle of this issue by Julie van der Bliek and policy-makers. and Ann Waters-Bayer submitsa frameworkand a checklist which can be useful to urbancentre groups who needto tackle The International Context: Local Authorities and specific constraints or exploit opportunities in urban External SupportAgencies agriculture. The sequence includes (a) urban centre-level As Africa becomes urbanised, urban councils of large preliminary context analysis and assessment with the urban centres participate more actively in international producersthemselves; (b) participatory urban appraial with congresses and are more listened to. Recent fora have been neighbourhoods typical of the range of intra-urbandiversity paying much greaterattention to the roleof urbanagriculture (techniques recommended and some exemplified); (c) in urban developmentthan a decade ago. For instance, the recognition of urban agriculture functions and resources, International Union of Local Authorities had panels includingmultistakeholder prioritisation of issues to guide addressing urban farming at its 31st World Congress in specific interventions; (d) stakeholderworkshop to discuss Toronto in June 1993. One year later, the Global Forum on results of previous steps and facilitate equal-footing Cities and Sustainable Development convened 50 city negotiations and coalitions for interventions; and (e) delegations from the North and the South in Manchester, participatory monitoring and evaluation for accountability, providing them with an Advisory Workshop on Urban adjustments and dissemination. Participatory issue-definition Agriculture. and action-takingare very importantfor all partiesto benefit The Global Forum's Advisory Workshop on Urban from the state of affairs targetted by the exercise. Policies Agriculturewas attendedby membersof at least halfof the which would enhance exclusively the role of small-scale city delegations. These were surveyed at the Workshop. urban agricultors in urban food suppy strategies could be Almostall respondents said there wasagriculture both within opposed by powerful groups (May, 1994). Conversely, and at the edge of their home city. A little more than two- consultationsdominated by the latter will lack credibility thirds said it was done by households,with a third adding among the former and doing little to improve their lot. that entrepreneursand institutions were also engaged. Only Kampala is one capital city where research has a quarter said urban agriculturewas causingsome problems increasingly shifted from generating baseline data to but two-thirdssaid their city was benefittinggenerally from informing public debates and assisting policy-makingon it. A third said city farmingwas regulatedsomehow in their urbanagriculture. The twelfthand closing articleof this issue city,with only aquarter knowing of any support programmes by Gertrude Atukunda and Daniel Maxwell summarises or research underway on agriculturein their city. variousstudies conducted at the MakerereInstitute of Social In August 1994 in a Declarationon Social Development Researchwith IDRC and Fullbrightfunding. This is the first and Sustainable Human Settlements issued at the International known research to systematically and statisticallyexamine Colloquium of Mayors on Social Developmentat the UN in in a major African urban centre, the impact of urban New York,over 100 mayors from around the world invited agriculture on the nutritionalhealth of children in households various sectors of society to join them in six categories of of various income groups. An early version of this article actions for the sustainabledevelopment of their cities. The informeda public debate in Kampala in late 1993, installed top category read as follows: "Reducing urban poverty by by the Mayorhimself. Findings were used inreports produced providing productive employment for thepoor and thejobless underthe World Bank-funded First Urban Project(Uganda), in the privateand public sectors,promoting urban agriculture which ended up recommendingthe inclusionof agriculture and supportingmicro-enterprise developmentthrough credit

147 Introduction—LucI. A. Mougeot African Urban Quarterly and training, particularlythe informal sector" (International Salaam (Richter, Schnitzler and Gura, 1995: 135—140). Colloqiumof Mayors on Social Development, 1994: 10). DANIDA has beenchannelling funds throughregional banks The change of attitudes among local and national to assistcooperatives of female urbanfarmers in East Africa governments in Africais influencing acorresponding change (Mougeot,l994b: 115). FAOis leading technical assistance among the internationaldevelopment community and vice- and trainingas well as feasibility studies for a whole range versa. Even with the divorce between agriculture and the of intra and pen-urban production systems (hydroponics, urban centre still not fully resolved, urban agriculture is horticulture, animal husbandry and forestry)(Marulanda and mobilising agrowing number and very diverse setof agencies. Izquierdo, 1993; Kuchelmeister and Braatz, 1993: 3—12 and In the wake of the BrundtlandReport in 1987, the United Phelan and Henriksen, 1995: 27—34). UNCHS has been co- NationsUniversity's Food-Energy Nexus Program, the Man supporting multistakeholder consultations for urban and Biosphere Project and IDRC-sponsored surveys in management actionplans which account for urbanagriculture African urbancentres, UNDP informally conveneda Support (IDRC files). UNICEF, CARE and OXFAM have been Groupon UrbanAgriculture (SGUA) in mid 1992. Two years exploring the potential of urban agriculture within more later, institutional representation atthe SGUA's ThirdMeeting comprehensivefood security strategies(UNICEF, 1994 and in Ottawa had tripled to 18 institutions of which six are IDRC files). support agencies. When the SGUA met at IDRC in March The World Bank has been supportingthe treatmentand 1996, it agreedon creating a global support facility, defined reuse of liquid and solid waste from cities into urban and workplan priorities and took charge of specific tasks to rural agriculture (Khouri, Kalbermatten and Bartone, 1994). improve information and coordination among agencies for It recently supported projects recommending inclusion of communications, research, policy, technical assistance and agriculture as a legitimate land-use in new master plans credit and investment in urban agriculture (the reader can (Atukundaand Maxwell,this volume). correspond with the SGUA on Internet through addressing The Bank also commissioned an assessment for more his message to [email protected]). Several comprehensiveBank supportto urbanagricultural activities agenciesnow recogniseopenly past project supportto urban in sub-Saharan urban centres (Smit, Ratta and Bernstein, agriculture. Someare creatinginter-sectoral workinggroups, 1996a). The reportfinds evidenceof extensiveurban farming exploringways to incorporateurban agriculture concernsin occupying increasing amounts of land and producingbenefits currentstructures and programmes, when notformalising new in terms of food and fuel production, waste and natural programme mechanisms. More are also collaboratingone resource management in sub-Saharan urban centres. It with another on specific projects and are reviewing the identifies as common constraints inadequate institutional! industry in their annual corporatereports. legal frameworks, limited access to agricultural inputs and The followingis an incomplete landscape of international post-production services, inadequatetechnical knowledge of activity in urbanagriculture, yet it sets a relevantcontext for urban agricultural practices, organisational constraints, future local initiatives in Africa. UNDP recently published political and sociocultural biases and lackof physical security. TUAN's study on urban agriculture;it is assisting Southern The report concludes from the Bank's record that it has local governmentswith their learning more about each other overlooked urban agriculture and its potential benefits, in on how urbanagriculture can renderthe development of their both its sectorwork and lendingprogramme, even where the urban centres more sustainable.At the time of this writing, industryis alreadysubstantial. It recommendsthat the Bank several consultations were being planned to ensure that increases its involvement, through approaching urbanfarming synergies between urban agriculture and other sectors of with an interdisciplinary perspective (urban development, urban development are taken into account in proposals, urban environment, infrastructure and agriculture), agreements and follow-upto HabitatII in June 1996. IDRC accounting for urban agriculture when formulating larger is working with several multi- and bilateral agencies to environmental strategiesin urban areas or integratedcoastal support production technologyimprovements and concerted zone managementplans (especially where there is larger policy interventionsin Africa, safe community-based liquid potential for intensifying urban agricultural activities), and solid waste reuse for urban agriculture in Asian urban collaborate locally and internationally to increase funding centres and regional networking in Latin America. CIDA, efficiency and tocoordinate donor activities and investments, with the Canadian Urban Institute and GTZ have been to advise borrowers on how to choose policy instruments supporting urban agriculture as a productive use of green suited to remove local constraints and maximise urban belts around major urban centres (La Havana, Cuba and farming's contributionto food security, income generation Maputo, Mozambique). SWEDEPLANhas assisted local and natural resource and waste management(i.e. through governmentsin Lesotho and Botswana with incorporating Regional Municipal Development Program and other intra-urban food production into social housing planning developing mechanisms). (Greenhow, 1994). Dutch Development Cooperation is A Development Bank of SouthernAfrica (DBSA) paper promoting urban agriculture as part of its policy on urban on policy support to micro-entrepreneurial development poverty alleviation (Kolstee, Bijlmer and van Oosterhout, includes urban agriculture and recommendsmore compact, 1994: 72, 148); it is supportingpen-urban zoning for high- inward developmentof urban centres, temporary access to density residential use with open-space food production in land to poorest producers, reversal of trend towards ever- African urban centres (Joep Bijimer, personal smallerplots for shelterand reviewof exclusionaryland-use communication, March 1996). GTZ is assisting the zoning, to facilitate combination of housing with income- development of periurban horticultural system in Dar es generatingactivities atthe householdlevel (Rogerson, 1996:

148 Volume 11 Nos. 2 and 3, May andAugust, 1996 An ImprovingDomestic and International Environmentfor African Urban Agriculture

30, 34). The DBSA has supported several small food- An overarching message of this issue is that urban producing projects in urban centres (Rogerson,1993: 25). agriculture should not be promotedand managed as an end per se. Rather it must be integrated into cross-sectoral and Conclusion multistakeholder strategies for mutuallybeneficial urbanand This issue of the African Urban Quarterly is the more agricultural development. Greater collaboration between recent in a sequence (Hunger Notes 1992, IDRC Reports research and developmentcapacities in urban planning and 1993, Unasylva 1993, ILEIA 1994, Habitat International thosein agriculturaldevelopment is urgently neededto make 1995, EntwicklungLandlicher raum planned for late 1996 urban farmingmore effective and sustainable inAfrican urban and Environmentand Urbanisationplanned for 1997) which centres of the 21st century. On one hand urban-based has been organisingand reporting on the growingknowledge experimental research in agriculture has stressed land- and know-howon urban agriculture.This issue, however, is extensive monocultural systems under secure tenure, with thefirst of thesequence to focus exclusively onAfrica, reason high capital and input needs. On the other, urban surveys why this Introduction has prioritised Africanist literature. reveal widespread, space-confined, labour-intensive, Hopefully, equivalents for other regions of the world will integrated, waste-reusing, partly local market-oriented, food- soon be available. producing systemsoperating despite precarious tenure. Little A trend is noticeableacross the material published so far researchand development investmentcould be found in the on African urban centres. Since Vennetier's 1961 study on regionthat would bridge the void, by eitheradapting the rural Pointe-Noire and throughoutthe 1970s and mostof the 1980s, experimental and/or improving the urban empirical so as to research tended to be dispersed geographically, single- turn into constraints faced most low- baseline opportunities major by authored and academic, descriptions, loosely income urban farmers (e.g., lack of on-plot ground-level interconnected.As of the late 1980s entries reflect more space, insufficient and unsafe air, soil and water inputs). structured efforts which and build on synthesise previous Advances underway in Latin America and Asia in popular research. More address issuesand specific prescribepractical hydroponics, stall-fed animal husbandry and small-scaled interventions. integrated farming offer much potential for exchange and For this issue, no terms of reference were specific given mutual learning. to the authors as papers could not be commissioned as the authors were chosen based on consultancies.However, References their known research interests.This enabledus to order their Adams, R.M. "The of Cities", contributionsunder the trends which have been introduced (1994). Origin Scientific American (special issue). herein. A special effort was made to secure the participation of researchers from different countries, with studies young P.V. City Toronto Urban a wide of Archibald, (1993). of Support for using range methodologies (from regional Food Production.' a Garden from overviews to individual life Even the issue is Creating City. Proposal stories). so, the of Toronto Technical biased towards sub-Saharan Africa, social science City Interdepartmental Working anglophone on UrbanFood Production, revised 9 Nov 1993. and policy concerns.References to otherAfrican Group City approaches Hall. Toronto. sub-regions. otherdisciplinary perspectives and technological aspects of urbanagriculture will he found in thebibliographies of the articles. Atkinson. S.J. (1995). "Approaches and Actors in UrbanFood in Countries", HabitatInternational Vol. Collectively,the articles do not claim urban agriculture Security Developing 19 No. 2: 151—164. to be either a panacea to African urban centres' problems (food security, employment and open space and waste G. The management) or the mother of all urban epidemics.Nor do Bowyer-Bower, T.A.S. and Tengbeh (1995). Environmental Urban they pretend that urban agriculture should ever make urban Implications of(Illegal) Agriculture centres lood self-sufficient, independent from rural food inHarare, Zimbabwe. Working PaperNo.4 of ODA Research supplies. However, they argue with soundevidence that food project R5946. Presented to the "Workshop on the self-reliance is and will remain central to the urban poor's Environmental,Social and Economic Impacts of (Illegal) food provisioning strategies. Urban agriculture may more Urban Agriculture" iR Harare, Zimbabwe, University of often than not he well positioned to supply cheap, fresh and Zimbabwe, Harare. reliable sources of micronutrientswell into the next century. The articles and their referencesalso present some evidence Butland, C.A. (1976). People of the Sun: The Civilizations on spin-off effects of urban agriculture in the larger urban of Pre-ColunibianAmerica. New York: Praeger. economy. Future research systematicallydocumenting such linkagesshould further steer urbanagriculture clear from the Centres, J.M. and J.P. Rolland (1995). "Adaptation of Red 'black hole' stigma often affixed to it. The articles are Meat and Milk Commodity Systems to An Adverse particularly concerned with the significance of urban Macroscenario: TheWest African Case", in R. TrevorWilson agriculture for the urban poor, surveysalso clearly showthat (ed) Supply of Livestock Products to Rapidly Expanding higher income groups compete with the poor for resources; Urban Populations. Proceedingsof Joint FAO/WAAPIKSAS without proper and timely public intervention, urban Symposium at Seoul National University. Rome: Food and agriculture could becomea battlefieldfor the survivalof the Agriculture Organisation. urban poor.

149 Introduction—LucJ. A. Mougeot African Urban Quarterly

Centres, J.-M. (1992). Agricultureet urbainsetpéri- Wastewater Reclamation and Reuse for Aquaculture, urbains aBobo Dioulasso: Lesfihieres d'approvisionnement Calcutta, India, December 1988. Bangkok: UNDP/WB/ en intrants et de commercialisationdes produits agricoles. ESCAP/Government ofIndia. Asian Instituteof Technology. Paris: GRET. Département de Géographie, Université de Ouagadougou, INERA, Bobo Dioulasso. Egziabher, A.G. (1994). "Urban Farming,Cooperatives, and theUrban Poor", in A.G. Egziabher et al. (eds) Cities Feeding Coe, M. et al. (1986) An Atlas ofAncientAmerica.Oxford: People:An Examination ofUrban Agriculturein East Africa. Equinox. Ottawa:International Development Research Centre.

Cosgrove,S. (1994). Une histoire des deux villes: Canadian ENDA-Zimbabwe(1994). Urban Agriculture in Harare. Community Gardening in Montreal and Toronto. Ottawa: Draft Final Technical Report,Environment & Development Cities Feeding People Series Report 8. International Activities, ENDA-Zimbabwe. Development Research Centre. ENDA-Zimbabwe (1996).Urban Agriculturein Zimbabwe: Courtlandt, C. and A. Kocybala (1990). A Guide to the Realities andProspects. Proceedings of aworkshop organised ArchaeologicalSites ofIsrael, Egyptand North Africa. New by ETCInternational and ENDA-Zimbabwe. Harare: ENDA- York: Facts on File. Zimbabwe.

Dar es Salaam City Council/Centre for Housing Studies Freeman, D. (1991). A City of Farmers: Informal Urban (1992). EnvironmentalProfile ofthe Metropolitan Area.Dar Agriculturein the OpenSpaces ofNairobi, Kenya. Montreal: es Salaam: DSM and ARDHI. McGill-Queen'sUniversity Press.

Della, A.A. (1991). Dynamique de l'espacepéri-urbain de Ganapathy, R.S. (1983). Development of Urban Agriculture Daloa: Etude géographique.Abidjan, Ivory Coast:Doctoral in India: PublicPolicy Options. Paperpresented at theUrban thesis. Institut de geographie tropicale, Faculté des lettres, AgricultureSeminar, Singapore,July. arts et sciences humaines, Université nationale de Côte d' Ivoire. Girardet,H. (1992). The Gaia Atlasof Cities: NewDirections for Sustainable Urban Living. New York: Anchor Books Diallo, S. (1993). UrbanAgriculture Research in WestAfrica: Doubleday. Record, Capacitiesand Opportunities. Ottawa: Cities Feeding People Series Report 5. International Development Research Groupe deRecherches et d'Echanges Technologiques (1991). Centre. Améliorer l'approvisionnement de Bamako en produits maraIchers etprotéinesanimales. 1—Agriculture et Elevage DialIo, S. and Y. Coulibaly (1987). Les déchets urbains en a Bamako. Paris: GRET. milieu démuni a Bamako. Tokyo: The Food Energy Nexus Programme, United NationsUniversity. Greenhow,T. (1994). UrbanAgriculture: Can Planners Make a Difference? Ottawa: Cities Feeding People Series Report Drakakis-Smith, D. (1990). "Food for Thoughtor Thought 12. International Development Research Centre. about Food: Urban Food DistributionSystems in the Third World", in R.S. Potter and A.T. Salam (eds) Cities and Guidoni, E. (1981). La yule européenne: Formation et Development in the Third World. London: Mansell signification du quatrième au onzième siècle. Bruxelles: Publishing. Pierre Mardaga

Drakakis-Smith, D.W. (1992)."Strategies for MeetingBasic Hammond, N. (1994). "The Emergence of Maya Food Needs in Harare", in J. Baker and P.O. Pederson (eds) Civilisation",Scientific American (special issue). The Rural-Urban Interface in Africa. Uppsala: Nordiska Afrikainstitutet. Hart, D. and J. Pluimers(1996). WastedAgriculture: The Use of Compost in UrbanAgriculture. Netherlands: Urban Waste Drakakis-Smith, D. (1995). Socioeconomic Aspects of Urban Expertise Programme Working Document 1. UWEP, Goudsa. Agriculture in Harare: A Preliminary Report of Field Research in 1994/5. Paper presented at the Workshop on Helmore,K. and A. Ratta (1995). "The SurpisingYields of UrbanAgriculture, University of Zimbabwe, August 1995. UrbanAgriculture", Choices Vol. 4 No. 1.

Drakakis-Smith, D. et al. (1995). "Urban Povertyand Urban International Colloquiumof Mayorson Social Development Agriculture: An Overview ofthe Linkages inHarare", Habitat (1994). Mayors' Declaration on Social Developmentand InternationalVol. 19 No. 2. SustainableHuman Settlements. New York,August 1994.

Edwards, P. and R.S.V. Pullin (1990). Wastewater-Fed ILEIA(1994). "Farming atClose Quarters", ILEIA Newsletter Aquaculture. Proceedings of the International Seminar on Vol. 10 No. 4.

150 Volume 11 Nos. 2 and 3, May andAugust, 1996 An Improving Domesticand International EnvironmentforAfrican Urban Agriculture

Khouri, N. et al. (1994).Reuse of WastewaterinAgriculture: May, J. and CM. Rogerson(1995). "Povertyand Sustainable A Guidefor Planners. Washington, DC: Water and Sanitation Cities in South Africa: The Role of Urban Cultivation", Report 6. UNDP-World Bank Water and Sanitation Program, Habitat InternationalVol. 19 No. 2. The WorldBank. Mbiba, B. (1994). "Institutional Responseto Uncontrolled Kironde,M. L. (1992). "ReceivedConcepts and Theoriesin Urban Cultivation in Harare: Prohibitive or African Urbanisation and Management Strategies: The Accommodative?", Environment andUrbanization Vol.6No. StruggleContinues", Urban StudiesVol. 29 No. 8.

Kolste, T. et al. (eds) (1994). "Urban PovertyAlleviation". Mbiba, B. (1995). UrbanAgriculture in Zimbabwe. Avebury: The Hague: Sectoral Policy Document of Development Aldershot. Cooperation 5. Ministry of ForeignAffairs. McCann, A.M.(1994). "The Roman Port of Cosa",Scientific Kuchelmeister, 0. and S. Braatz (1993). "Urban Forestry American (special issue). Revisited", Unasylva Vol. 44 No. 173. Metzger, R. et al. (1995). L'approvisionnementdes villes Jellicoe, G. (1989). The Landscapeof Civilisation. Northiam, africaines en lait et produits laitiers. Rome: Etude FAO UK: Garden Art Press. Production et Sante Animales 124. Food and Agriculture Organisation. Lado, C. (1990). "Informal Urban Agriculture in Nairobi, Kenya", Lçinc' Use Policy Vol. 7 No. 3: 257—266. Mitlin,D. and D. Satterthwaite (1994). Cities and Sustainable Development. London: Background paper prepared for Lamba,D. (1994)."Nairobi Action Plan: City Environment Global Forum '94 . lIED Human SettlementsProgramme, and SustainableDevelopment" (unedited version of section 62 pp. on environment),in James G. Karuga(ed.) Action Towards a Better Nairobi: Report andRecommendations ofthe Nairobi Mlozi, M.R.S. (1993). "Inequitable AgriculturalExtension City Convention. Nairobi: Friedrich Naumann Foundation. Services in the Urban Context: The Case of Tanzania", in NP. Stromquist (ed.) Education in Urban Areas: Cross Lee-Smith, D. and P. A. Memon (1994). "Urban Agriculture National Dimensions. London: Praeger. in Kenya", in AG. Egziabher et al. (eds) Cities Feeding People: AnExamination of UrbanAgriculture in EastAfrica. Mvena, Z. S. K. (1986). Urban Agriculture in Tanzania: A Ottawa: InternationalDevelopment Research Centre. ResearchProposal. Ottawa: IDRC.

Le Goff, J. (1984). La civilisation de l'Occident médiéval. Moscoso, J. (1995). El uso de efluentes de lagunas de Paris: Arthaud. estabilizacionen acuicultura y agricultura. Paper delivered at Latin American Seminar on Urban Agriculture (IDRC/ Lewcock,C. (1995). "FarmerUse of Urban Wastein Kano", FAO/AIPE), La Paz, April. Habitat International Vol. 19 No. 2. Mougeot, L. (1993). "Farming on Dakar's Edge", IDRC Lourenço-Lindell,I. (1995). "The Informal Food Economy ReportsVol. 21 No. 3 (October). in a PeripheralUrban District: The Case of BandimDistrict, Bissau",Habitat International Vol. 19 No. 2. Mougeot, L. (1994a). Urban Food Production: Evolution, Official Support and Significance. Ottawa: Cities Feeding Marulanda, C. andJ. Izquierdo(1993).Technical Manualon People Series Report8. International Development Research Popular Hydroponics Gardens: Audio Visual Course. Centre. Santiago: Food and Agriculture Organisation. Mougeot, L.J.A. (1994b). "LeadingUrban Agricultureinto Maxwell, D.G. (1994). "The Household Logic of Urban the 21st Century: Renewed Institutional Interest", in A.G. Farming in Kampala", in A.G. Egziabheret al. (eds) Cities Egziabheret al. (eds) Cities FeedingPeople: An Examination Feeding People: An Examination of Urban Agriculture in of Urban Agriculturein East Africa. Ottawa: International East Africa. Ottawa: International DevelopmentResearch DevelopmentResearch Centre. Centre. Mvena,Z.S.K. et al. (1991). Urban Agriculturein Tanzania: May, J. (1994). Development Policy Options for Urban A Study ofSix Towns. Draft report. Department ofAgricultural Agriculture in South Africa. Presented to "Urban Open Education and Extension, Sokoine UniversityofAgriculture, Spaces: Potential for Productive Utilisation", Conference, Morogoro,Tanzania. Technikon Pretoria, 24—25 February. NationalEnvironmental Information Centre (1994). District May, J. and C. Rogerson (1994). "How Green is your Environment Profile: Kampala. Kampala: NEIC. Garden?"Indicator SA Vol. 11 No. 3.

151 Introduction—LucJ. A. Mougeot African Urban Quarterly

Pain, M. (1985)."Kinshasa: Symboled'uneAfrique urbaine", Shapiro,B.I. et al. (1995). "An Analysis of Constraintsand Cahiers d'Outre-Mer Vol. 38 No. 149. Opportunities in Periurban Dairy Development in Sub- SaharanAfrica", in R. Trevor Wilson (ed.) Supply ofLivestock Phelan, J.A. and J. Henriksen(1995). "GlobalIssues in the Products to Rapidly Expanding Urban Populations. Supplyof Livestock Food Products to Urban Populations", Proceedings of FAOIWAAP/KSAS Symposium at Seoul in R. Trevor Wilson (ed.) Supply of Livestock Products to National University. Rome: Food and Agriculture RapidlyExpanding Urban Populations.Proceedings of Joint Organisation FAO/WAAP/KSAS Symposium at Seoul National University. Rome: Food and AgricultureOrganisation. Smit, J. (1980). "Urban and Metropolitan Agricultural Prospects",Habitat International Vol. 5 Nos 3—4. Redclift, M. (1987). SustainableDevelopment: Exploring the Contradictions.London: Routledge. Smit, J. et al. (1996a). Urban Agriculture:An Opportunity for Environmentally Sustainable Development in Sub- Richter, J. et al. (eds) (1995). Vegetable Production in Saharan Africa. Washingtom DC: Urban. Periurban Areas in the Tropics and Subtropics—Food, Income and Quality of Life. Feldafing, Germany: Deutsche Smit, J. et a!. (l996b). Urban Agriculture: Food, Jobs and Stiftung fur internationale Entwicklung Zentraistelle fur SustainableCities. New York: United NationsDevelopment Ernahrungund Landwirtschaft. ProgrammePublication Series for Habitat IIVolume One.

Roberts, D. (1996). "A Greener Vision for the Durban Staal, S.J. and B.I. Shapiro (1994). The Effects of Recent MetropolitanArea with D'MOSS", Muniviro Vol. 13 No. 2 Price Liberalization on Kenyan Pen-Urban Dairy: A Case (May). Study Using the PolicyAnalysis Matrix, Food Policy Vol. 19 No. 6. Rodenbeck, M. (1991). Egypt from the Air. Thames and Hudson: EditionsDidier Millet. Thang, V.Q. (1996). Effects of Sewage Utilisation on Fish Farmingand Irrigation (Vietnam). Hanoi, Vietnam: Progress Rogerson,C.M. (undated).Urban Reconstruction and Urban report. Vietnam NationalUniversity. Cultivation in South Africa. Johannesburg: Department of Geography & Environmental Studies, University of Tricaud, P.-M. (1987). Urban Agriculture in Ibadan and Witwatersrand, unpublishedreport. Freetown. Tokyo: The Food Energy Nexus Programme, United Nations University. Rogerson, C.M. (1992). "FeedingAfrica's Cities: The Role and Potentialfor Urban Agriculture",Africa Insight Vol. 22 United Nations Children'sFund (1994). The Urban Poorand No.4. Household Food Security: Concepts, Evidence and Case Studies. New York: Urban Examples No. 19. UNICEF. Rogerson,C.M. (1993). "UrbaiAgriculture in SouthAfrica: Scope, Issues and Potential",Geofournal Vol. 30 No. 1. Vennetier, P. (1961). "La vie agnicole urbainea PointeNoire", Cahiers d'Outre-Mer Vol. 14 No. 53. Rogerson, C.M. (1996). Rethinking the Informal Economy of South Africa. Johannesburg: Development Bank of von Braun, J. et a!. (1993). Urban Food Insecurity and SouthernAfrica Development Paper 84. Malnutritionin Developing Countries: Trends, Policies and ResearchImplications. Washington, DC: International Food Sanyal,B. (1986). Urban Cultivationin EastAfrica. Tokyo: PolicyResearch Institute. The Food Energy Nexus Programme, United Nations University. Wade,I. (1987). Food Self-Reliance in Third World Cities. Tokyo: The Food Energy NexusProgramme, United Nations Sawio,C. (1994). UrbanAgriculture andthe SustainableDar University. es Salaam Project, Tanzania. Ottawa:Cities Feeding People Series Report 10. International Development Research Warren, P.M. (1994). "MinoanPalaces", Scientific American Centre. (special issue).

Sawio,C. (1993). "Feeding the Urban Masses:Towards an Watts, D.J. and CM. Watts (1994). "A Roman Apartment Understandingof the Dynamics of Urban Agricultureand Complex",Scientific American(special issue). Land-Use Change in Dar es Salaam, Tanzania". Worcester: unpublishedPhD Thesis, Graduate School of Geography, Winters, C. (1983). "The Classification of Traditional African Clark University. Cities",Journal of Urban HistoryVol. 10 No. 1 (November).

Schilter,C. (1991). L'agriculture urbaineàLomé:Approches Yanin, V.L. (1994). "The Archaeology of Novgorod", agronomiqueet socioéconomiques. Paris, France: Karthala. Scientific American (special issue).

152 ,. Sociology of Education Abstracts EDITOR nai4 Dr Chris Shilling, University ofPortsmouth UK ASSISTANTEDITOR Karen Woodley, Carfax Information Systems, UK Supportedby anInternational Editorial Board r Sociology ofEducation Abstracts isan essential resource for academics involved in theresearch and teachingof education throughoutthe world. Theservice draws on a wide range of international sources as a means of servingthe information needs of those concernedwith the sociological study of education, and provides comprehensive coverage of over600 journalarticles and books each year, infour quarterlyissues. Abstracts are preparedby an international team of subject specialists and are non-evaluative in form. Coverage of the service is broad to r incorporate the diverse types of materialthat are relevant to study in this area. Journals from throughout the world and a comprehensive list ofpublishers catalogues are scannedregularly for appropriate items.

Each year Sociology ofEducation Abstracts provides a complete guide to the most importanttheoretical, methodological and policy developments relevant to the sociological study of education,

SUBSCRIPTION RATES 1998 Volume 34 (4 issues). ISSN 003&0415. Institutionalrate: EU£354.00; Outside EU£354.00; North America US$634.00 Personal rate: EU £138.00; Outside EU £158.00; NorthAmerica US$248.00

ORDER FORM Please send a completedcopy of this form, with the appropriatepayment, to the address below. Name Address

th 4(0)1235401550 E.ik salesOcarx.co.uk [JCARFAX C______;j Carfax E'uNkhing I..imikd,P0 Box 25, Ahingdon.Oxfordshire OX 14 SUE. ( K African Urban Quarterly, 1996: 11(2 and 3) pp. 154—161 © 1996 by AfricanUrban Quarterly Ltd.

URBANAGRICULTURE AND THE AFRICAN URBAN FOOD SUPPLYSYSTEM Annu Ratta and Joe Nasr The Urban AgricultureNetwork, 1711 LamontStreet, NW, Washington, DC 20010 USA Phone: (1-202) 483-8130;Fax: (1-202) 986-6732 and E-mail:[email protected]

Abstract Accepted October, 1995 Thispaper examines the role that urbanfarming plays in thefood supply system ofAfrican urban centres. Aftergrounding urban agriculture within the dynamic between urbanisingtrends, a widening 'foodgap', and largely unfavourablepolicies towards it, the interaction between urbanfood demand and supply is described based on empirical evidence. Urban food production is then analysed according to its two main, interconnectedfuncti ons: farming for householdconsumption and farmingfor the market. Theauthors callforgreater recognition ofthe nexus between the urbanland-use system and thefood- supplysystem.

Résumé Get article examine le role de ('agriculture urbaine comme un tout,au sein du système d'approvisionnementalimentaire des villes africaines. Dans un premier tenps, on situe Iaproduction alimentaire urbaine par rapporta Ia dynamique existant entre I 'urbanisation de Ia production alimentaire urbaine par rapport a Ia dynamique existant entre I 'urbanisation de Ia population, ledeficit alimentairecroissant, etdes politiquessurtout defavorablesa Iaproduction urbaine. Ensuite, on décrit ('interaction entre In demcinde et l'offre d'aliments dans les villes, en s'appuyant sur des données empiriques. Suit une analyse de Ia production alinientaireurbaine, notamment ses deuxprincipaux débouchés: I 'autoconsommation et Ia mise en marché. Les auteurs réclanient une prise en compte plus explicite de l'interrelation qui existe dans les villes entre leur système d'utilisations du sol et leur système d'approvisionnementalimentaire. Introduction The aim of this paper is to introduce the place of urban Urhanisation and food shortages are both well-known farming within the nexus between the phenomenon of a phenomenain developingcountries in general and in Africa massivepopulation increase in urbancentres and the system in particular, hut they are generally perceived as distinct that suppliesthe nutritional requirementsof this population. trends. Food production may he the major urban function Urban farming in sub-Saharan Africa today is occurringin that is least noticed by urbanplanners and decision-makers. the context of rapid urbanisationcreating an economic and Simultaneously, agricultural programmesare largelytargeted foodsupply systemin which urbanfarming fits in two ways: to rural settings.It is necessary for both the urban manager, marketfarming is a competitive and significantsupply source the minister of agricultureand othersin betweento recognise in the urban food system while subsistence farming is an the intricate ties between the increase in urban populations, essential activity providing food security for swelling their growing food needs, the changing agricultural numbers of urban poor families who do not have stable production patternsin and aroundurban centres, and the areal financial resourcesto acquire food in the market as well as expansion ofland and water surfaces used for this production. for growing numbers from higher income groups who are As Kironde (1992: 1285) stated, "the receivedconcept that pauperised due to economichardships. agriculture is a rural and poor man's undertakingunsuited to Other important components of this emerging urban urban areas needs to be revised." agriculturalactivity are notdealt with here. For instance,the The call for such a new understanding of the connectionsbetween food production,wastes and resources contemporary metropolis is notbased on intellectual curiosity, have already been dealt with in Smit and Nasr (1992: 141— but on empirical observation of an evolution in human 152). Acomprehensive global overviewis presentedin Smit, settlementsin developing countries and most noticeably in Nasr and Rassa,(1996). Here,we focus on how urbanfarming sub-Saharan Africa, that is already taking place. This fits in the urbanfood supplysystem on theAfrican continent. evolution lies in a progressively blurred functional differentiation between urban areas and rural areas. Urban Urban Agriculture and Urban Policy areas in developing countries share some of the traits with Recent research in sub-Saharan Africa has established therural areasfrom whichtheir burgeoning populations came. that urbanfarming is asignificant activity in many countries These urban centres have over time established as one of in the region and contributesto the food security of African their primaryeconomic functions the core 'rural' use of land: urban families, especially the poorest ones. African urban agriculture. This use in fact had long been important in centres tend not to be solidly built, leaving many surfaces Africanurban centres. that are put into cultivation. Such an extensive farming is Acknowledgements—This paper is based partly on research containedin "UrbanAgriculture as a Strategy",a report written by TUAN's Jac Smit and Annu Ratta for the AFTES divisionof the World Bank. Thanks are due to Jac Smit for his essential comments.Jac is the author of all the photographsin this article, taken between 1991 and 1994.

154 Volume11 Nos, 2 and 3, May and August, 1996 UrbanAgriculture and theAfrican Urban Food Supply System evident when observed from the air as houses can be seen enterprise and the felt need of the farmers, rather than by scatteredin the midst of greenery (figure 1). government design. It is mostly the story of informal Surveys in differenturban centreshave found asignificant development in the face of official disapproval. Urban percentage of the urban households to he farming—over a planning laws and concepts created during colonial times quarterin Ouagadougou,Upper Volta; 35 percentin Yaounde, largelyview as inappropriate informal food production within Cameroon; over a third in Kampala, Uganda; 63 percent in the urbancentre (as opposedto formal commercial farming six Kenyan urban centres;and 68 percentin six urbancentres atits periphery).These have engenderednegative health and surveyed in Tanzania (Lee-Smith, et a!, 1987; Mvena, land-useregulations on urbanagriculture. Lupanga and Mlozi, 1991; Manshard, 1992; Maxwelland In African urban centres, animals, particularly small animalsand poultry have a long traditionas an everyday part viewof residential area Figure 1,Ibadan, Nigeria. I his aerial a mostly of life in marketcentre, centre, near the shows the of the land is family village,periodic trading airport that majority surface placed centreand Poorerfarmers follow in cultivation by the residents town, urban city. commonly thepractice of keepinga few small animals—rabbits,guinea pigs,pigeons and chickens in the house,for festival occasions or for sale. Half of all urbanfamilies in Kenyaand 30percent of the population in Maputo raise livestock(Kolstrup, 1985 and Lee-Smithet al,, 1987). Frequently, urbanhorticulturists arealso animal breeders.In Maputo'scase, 80percent ofthe urban farniers keep some livestock. However, animal husbandry is generally discouraged by government authorities. The examples of the struggles of the urban farmer are numerous. In Harare,Zimbabwe, until 1992, thecity council periodically orderedthe slashing of crops toprevent mosquito breeding and agricultural run-off into the water system (Mhiha, 1994: 188—202).In Bamako, wheremarket gardening on the riverside is supported by the government, the municipality was still cuttingdown grain ripeningon vacant lots in the l990s Zziwa, estimatedthat 10 to 25 early (Kiahou, 1991). 1992). Guyer (1987) percent when faced with a of urban Auiicansfarm (cited in Lourcnco Lindell, 1994). Recently, burgeoning agricultural combined with an to food demand Bamako, Mali, produces all of its vegetable demand and activity inability satisfy inurban theattitude of African has exports in some seasons (Diallo, 1993). Urban farming in areas, many governments evolvedfrom one of to one of tacit Karr paIn today sitisfies 70 percentof thepoultry demandof slashing crops acceptance, formal and even of urban An the city. Moreover the activity is rapidly expanding. For approval promotion farming. era of of urban was in Zambia in instance, the numbet of animals kept in Dar es Salaam acceptance farming begun 1977with urbanresidents to Tanzania grew by 40 percent in two years, from 586 to 818 apresidential speech urging grow thousand (City Council Livestock Office, cited in Kironde, their own food. This was followed by a range of actions, 1992). ranging from discontinuation of the enforcement of In many countries, urban farming provides significant prohibitivelaws, to promotional programmes,to the creation employment in the informal sector. The national census of of government-run stores (Sanyal, 1985: 15—24). Four main 1980 in Maputo,Mozambique, suggests that one-thirdof the types of agriculture can now be found in Lusaka, Zambia: urban workforce wasinvolved in agriculture (Kolstrup, 1985). gardening for family consumption (mostlyvegetables), semi- Moreover, certain disadvantaged groups find that urban commercial farming, commercial farming and rainy-season farming offers them special opportunities to provide for agriculture (primarily staples). The first is found withinthe themselves and their household. For instance, surveys have settled area while the others are more towards the periphery found that around two-thirds of urban farmers in Nairobi, (Drescher, 1994). Kenya and in Kisangani, Zaire were women (Lee-Smith et In Kenya, the Local Government Act leaves to local a!., 1987 and Streiffeler, 1991), The unique opportunities authority bylawsto decide whetherto permitor restricturban available to low-incomegroups are discussedlater, farming. The Nairobi bylaws prohibit farming along A traditionof limited urbanfarming—market gardens— government-maintained streets. Small livestock may beraised was promotedduring colonial times in someAfrican countries freely, butpermission is required for keepinglarge livestock. to supply part of the urban centres' food necessities. Some Small urban centres in Kenya are providing more support. ofthis has continuedas commercial farmingin urbancentres, Isiolo, Kenya provides support to urban farmers as well as with partial support from the authorities. The market irrigation water while Kitui, Kenya provides extension gardening includes fruits and vegetables,poultry and meat (Memon and Lee-Smith, 1993). Despite the constraints it production and high-valui crops. At the same time, small- faces,urban farming infact isthriving in Kenya and elsewhere scale farming by the poor—for subsistence and sale—has in Africa (figure 2). What is evident from this and other proliferated over the past couple of decades due to the evidenceis that urban agriculture is not a marginal activity

155 Anna Rattaand Joe Nasr African Urban Quarterly

in Africa hut rather a significantsuhsector within both the Africa will add a half billion to its urban population. In urban economy and the agriculture industry comparison, North America and Europc together will add 150 million (UNDESIPA, 1993). Figure 2, Nairobi, Kenya In an openarea adjacentto downtown, solid Urbanisationin most developed countrieswas fuelled by waste ishenig composted (left andtop ofphoto) andapplied in vegetable economicdevelopment and accompaniedby intensification production of agriculture and development of anational food distribution system. Some studies have found growth in urhanisation to he relatedto economic development(von Braun et al., 1993). However, in many developing countries, urban growth is occurringeven in the faceof low or negative economic growth and weak development of the food transportation and marketinginfrastructure. In suhSaharan Africa, this growth is caused more by rural migration (about60 percent) than by natural internal increase (UNDESIPA, 1993). All these factors are creatingpressures on the urban food supply,with foodproduction in and aroundthe city becoming increasingly competitiveand in somecases essential for survival. Out migration from rural areas has multiple impacts on the supplyof agricultural labourai d skills, It is usually the younger (and better educated) people with higher earning capacity that migrate to urban centres. In parallel, with everything else equal, this particular characteristicof out causes reduction in the of cash Urbanisationand its on Food andAgriculture migration a quantity crops Impact in rural areas and in the outflow of Urhariisationallcct the agricultural supply system in produced particularly towards the urban centres, many ways. It moves large parts of the national population agriculturalproducts Moreover, from rural sources.No more is reliance other factors such as the AIDS plague, impact the away farming supply of the rural farmer, the of food on an entirely local food pioduction possible Urhanisation productivity Thus, supplies creatcs centies of conccntrated food demand that need to hc from rural areas sufferdoubly from the labour flows and other causes, This outcome is sometimes the short supplied thiou4h an increasinglyintricate transportationand mitigated by marketing system. A ii oie formalised food marketing and distance of the migration or by the maintenance ofproductive is supplysystem cmerges. It also changes the national demand relations with the village of origin. However, migration centresthat are slructurc by affectingpcoplcs diels often to urban tot) distantand, over time, the of the urban with farmland Urban if 'as of des c loping countrics arc gr )wing three direct ties residents are reduced, timcs as List as ruial areas. ac ountine Lir an estimated84 Consequently, agricultural production must increase to percent of thc net population incrcase in those countr es iced the growing urbanpopulation, vvhetherby having higher between 1990 md 2f)20. A a iesult of this trend, the global yields per worker, by increasingthe number of w'kers, or population will go from 14 percenturban in 1960to 58 percent by tapping additional sources of production such as urban in 2020 While Noiiii Animrim a Europe and Latin America agriculture Accordingto FAOestimates, about 131 million are already highly iiihanised (ovcr 70 percent), it is in the tons of food in wheatequivalent would beneeded to feed the two regions that tie at this tune the least urhanised that we populationthat is expectedto have migrated to urban areas will see thc largest incrc ase in urban populationsover thc in dcveloping countries between 1980 and 2000, and an next quarter of a centuiy Africa (32 pLrcent), fotlowedby additional 160 million tons for the natural urbanpopulation Asia (31 percent) (UN [)ESIPA, 1993). Consequently, it is increase (UN FAO, 1984). in thesc two cont nents that the food and agriculture systems Afood gap alreadyexists in suhSaharan Africabetween will i'cel thegiearesr irnpaci of ui hanisafon Thehighest urban the higher overall rare of populationgrowth and the lower growth rates svilI he in ounti ies thit arc least equipped to rate of increase in the production of staple cereals, World feed thcir urban ccntrcs thc 49 couritrics classified by the Bank forecasts indicate that to achieve food security, food United Nations as Least t)cv eloped, which happen to betoday production will have to grow at about 4 percent per year, among the least urhaniscd nations, twice the rate of the past 30 years (Virji and Schram, 1994). While most of thesc cou aries we located in suhSaharan At the same time, the conlinent's arabIc land per capita has Africa, thc urhaaisati in lcvi Is irc not const'int across the been cut in half from 1.2 acres per person in 1965 to 0.6 continent: about 45 r ent in southcrnAfrica, 35 percent in acres today (UNFAO, 1993). If relying on an increase in West Africa 31 peicent in Sah Ii in countrics,and only 20 agricultural productivity on rural fai'msalone, many countries percent in I ist Africa (1 N DLSIPA 1993). Although these in Africamay not be able tokeep up with the food, nutrition numbersare still low they havc been i isin staggeringly fast. and other needs of the increasing urban populations. This Afterhaving incieascdIi OmIt 32 million in 1950to 206 million may be particularlycritical for countries that do not have the in 1990, Africi s urban population is cxpectcd in 2020 to foreign exchange to import food, In this context, urban exceed 700 million cqui aknt to the current total population farming is a natural economic response that must not be of the coritinc nt (Vipi and Schiam, 1994). In this period, ignored

156 Volume 11 Nos. 2 and 3, May andAugust, 1996 Urban Agricultureand the African Urban Food Supply System

The Urban Food System In more developedcountries, urbanisation was tied tothe In order to understandurban farming, it is necessary to development of a more intricate national marketing and study it as a part of the urban and national food supply and transportation infrastructure that providedthe urban centres demand system, within the context of the urbanisation with food from remoterural and foreign sources.The urban process.The urban food system, consistsof(a) the foodurban marketing structures moved gradually from the traditional residents consume, (h) the places where it is produced and petty trade structure—characteristic of periodic market (c) the often complex processes by which it gets from centres, trading centres, towns,villages and urban centres— producersto consumers. to formalisedand capitalisedmarket structures along with Urbanisation's impact on thefood demandstructure in extensive storage, refrigerationand processingfacilities that the country is multifaceted.It net only impacts the levels of increase the shelf life of food. demand for food, but also the types of food demanded. In Whatwe have found thoughis that in Africa the traditional general terms,compared to rural populations, urban residents food supply structure is being overlaid with rather than tend to eat more varied food, more animal protein and replaced by, this new structure. The remote systems could at affordable consumefewer calories. This statementshould he tempered not single-handedly nourish all urban residents by limitationsthat the poorestcitizens face in gaining access prices. Remote food production nowcomplements local ways of residents of urhanised areas withtheir nutritional to many protein-rich, less affordableproducts, and in fact furnishing needs. The of the urban food in Africa may consumemore starchy, high-carbohydrate foodstuff. On complexity system increased. the other hand, the growth in absolute numbers in middle- has, therefore, greatly income residents associated with urbanisation has resulted Drakakis-Smith (3992) has identified three main of thefood of urban centres: urban in a growth in demand for certain cereals,high-value crops, components supplysystem residents food or barter), fruits, vegetablesand meat products in many countries. acquire through exchange (purchase subsistence or transfer What is clear is that diet changes(of onetype or another) production (home production, farming) aid, donation, food and are generally associated with the urbanisation process and (food stamps feeding programmes). these have several of The sources of purchased foodmay he rural producers,urban changes repercussions. Consumption or traditional basic foods is often in urban producers import. (staples) replaced The amount of food the various sources— centres of more often non- supplied by by consumption processed—and rural and well as the such as rice and livestock urban, foreign—as crops predominantly indigenous—foods produ.cts,along each source on various with a of and convenience suppliedby vary depending factors, higher consumption pre-cooked the economic condition of the the level of foods. In due to this trend, the has including country; part change typically development of the food marketing, storage, and brought about ahigher incidenceof cancer and heart disease transportation infrastructure and system; agricultural in urban populationsin developing countries. the of the urban Urban diets have not productivity; geography region(topography, a significant impact, only locally soil conditions and climate, the of land but and on the demand etc....); availability nationally beyond, agricultural and water resources; and agricultural and urbandevelopment structure.In Tunisia,for of example,per capita consumption policies. the durum wheat residentsof small and locally grown among Whereverthe nationalfood marketing and transportation intermediateurban centres is about halfthe amount consumed system is not well developed, urban farming is particularly rural residents while for residentsof urban by large centres, competitive. For high-value, specialty, and especially it is one-fourth as much Braun et only (von al., 1993). perishable crops, urban farmers have the advantage of Drakakis-Smith on urban residents (1992: 38—58) reports closeness to market. This reduces transportation timeand cost, four times as much rice and wheat as rural ones in eating supplies crops that may otherwise not be fresh or even sub-Saharan Africa. The increased of meat consumption available, as well as providesthe meansto follow the market products by urban dwellers is supplied through intensive closelyto respond rapidly to market needsand prices. livestockfarming, especially in pen-urban areas and creates Urban farming appears in many parts of the urban food a sharp increase in demand for feed grains. system—it includes a spectrum of farming activities From where, then, do people obtain their food? Villages undertaken by a variety of farmers who farm for sale in the get theirfood supplyfrom farming within the settlementand market or for home consumption.The commercial farmers surrounding countryside.In urbanised areas, however, the range from large enterprises to middle-class family businesses immediate surroundings have to adjust their production to poor farmers who produce mostly for food security but capacity to keep up with the growing and changing food sell their product whenever they can. Farming for demand, as nearby farmland is taken over for urban uses. consumption is mostly done by the poor who lack stable Consequently, farming in the region intensifiesand adapts income to purchasefood. However,much larger sections of its crops to the new demand and additionalfood is imported the economy are found to grow some of their own food in from other parts of the country or from abroad. If the food the hard economic times faced by many countries in the supply system is viewed as a series of overlays of 'food- region. sheds'of varying diameter,shape and directionfrom theurban centre (dependingon the type of agriculturalproduct), then Urban Farmingfor Household Consumption the 'food-sheds' expand along with the urban centres they Accompanying the growthof urban centresin Africais a serve. considerable expansion in the scale of urbanpoverty. In 1988,

157 Anna Ratta and Joe Nasr African Urban Quarterly about one-third of the poor in developing countries were In the late l960s a survey found that 18 percent of the living in urban areas. By 2000, the proportionis expectedto families in Dar es Salaam Tanzania were engaged in increase to about 57 percent. In Africa, the corresponding agriculture Another survey in 1991 found the percentageto increase isfrom 23 to 42 percent(Popkin and Bisgrove, 1988: have increased to 68 percent (Mveria, Lupanga and Mlo7i, 13 23). The conditions and causesof hungerand malnutrition 1991) l'his massive shift occurred mainly in the l980s and in urban centres are very different from those in rural areas, included all economic lcvels and all portions of the city requiring very different policy interventions. (Sawio 1993) The reasons for the African urbanresidents' undertaking Reports from Kinshasa and Maputosuggest that asimilar cultivation and animal raising are manifold, By depending massivc shift to agriculture ceurred in those two urban on suppliesfrom rural areas and overseas, urbanfamilies are centres during the time of instabilityand economic decline. often finding difficulty in having access to sufficientfood to Accordingto Jamal (1988. 679 701), Kampala, Ugandanow satisfy everyone's necessities, These shortages are in turn produces twice as much )f its caloric needs than it did in linked tohigher food prices.Several surveyshave found that 1972 (40 perccnt as compared to 20 percent) In a statistical poor urban householdsspend up to 70 percent (occasionally study in Kampala Maxwell (1993) fimnd children of low- even more) of their incometo purchasefood (Khouri-Daghar, incomefarming families to hc as healthy as children of high- familics 'cud healthier th'in childrcn non 1986 and Sanyal, 1986). Amongst the very largest urban income of farming familics, centres, for instance, more than half the average household poor Urban food is in income is spent on food in several cases: 63 percent in supply partitulaily insecure countries Kinshasa,Zaire; 58 percent in Lagos, Nigeria; 55 percent in with undependahlc supplysources due topoor infrastructure Uoai da ard amY! that Algiers, Algeria and 51 in Alexandria, (e g , Zairc) ountries suffer disruption percent Egypt of networks and food due wars (Ethelston, 1992: l6l7). While these numbers are for the transportation production to or weak 'c such a total population, they tend to he even higher for the poorer governan Angola,Somalia, Mo7amblque, citizens. Some researchers have found that the urban Sudan I iherr'c and Randa During enduringcrises, urban poor ioeausul'survival 'ui the uihan and eat poorer diets, often pay higher prices and spend more of agriLuPureoffeis refugee It has bee f )und to increase in their on food, due to their low and populations Kinshasa, earnings especially and iherianu Even where incomes 1994). This is where Maputo n I fugce camps (figure 3). irregular (Lourenço-Lindell, crisis does iot cut il war or national or natural urbanagriculture enters the picture. collapse calamity thc ccurrcncc f urban farrmiiny increases in The lowerthe income group, the greater the percentage of household food that is counti ies facingpolitic ml r economicstress such as has been consumption self-produced, the case in Tan zarira over halfthe total needsin somecommunities, For providing found structural to Lusaka's food amounts to one-third of Lourenço-Lindcll (1994) adjustment poor, self-produced havc had a at've ct devaluated total food an estimated 45 large n imp (unemployment, consumption,involving percent inflc ii n urban residents in Guinea of low income families either in a home or salaries, )n) Bissau, cultivating yard Bissaic Amunu the cur es was a decline in urban in on the of the urban centre sequen gardens periphery (Sanyal, 1986). nutrition. Such a led to more ac'tr from Poor families in Addis Abaha and save lO—20 collapsc ity shifting Kampala and foi'mal to informal sector In Bissau, this meant the 37 of their income home opcning percent respectively through of new food stalls and markets in suburbanareas balancing of food 1993: In Nairobi, where production (Mougeot, 2—5), the closing of shops in thc core, E'xtensrve rice cultivation close to two-thirds of urban households of their grow part consumed entircly withinthe extendedfamily has developed own food, 77percent ofthese urbanfarmers produce entirely for their own consumption(Lee-Smith et a!., 1987). Figure 3,Abidjan, ( ôted Ivoire, I iherian refugeeshave takenover an The 1 980s were a time of economic downturnin Africa undeveloped industrial site and convert€d it to vegetable and fruft and urban populations were particularly hard hit by the production for food security and market. I he shack on the left is the recession and subsequent structural adjustment. Urban home ofa watchman populationsare also more vulnerableto breaks in the food supply, through war, civic disturbance, natural disaster or famine;Africa has had more than its share of all these since the I 970s. Consequently, urbanpoverty has grownfaster than rural poverty in many developing countries. Prices of agriculturalproducts were rising while urban incomeswere falling. Food subsidies were cut in several sub-Saharan countries.While economiesshrank, urbancentres continued to grow in size. With fewer opportunities and lower remuneration for employment and much traditional entrepreneurship, countless of thepoor along with thousands of higher-incomefamilies that were increasinglyfacing the problem of food insecurity turned to agriculture in sub- SaharanAfrica, joining those who had been practising it all along.

158 Volume 11 Nos. 2 and 3, May andAugust, 1996 Urban Agricultureand the African Urban Food Supply System on plots in Bandim, the largest unplannedsettlement in the compost as well as delivering the produce; and small urban centre of Bissau. Also found were some backyard merchants. gardens, mostly for vegetables. All this activity is labour The national 1988 censusfound that urban agriculture in intensive,with a minimum of purchasedinputs. Dar es Salaam ranks as the second employer (after petty Atkinson(1992) saw one factorthat makesthe promotion trading and day labour), with 20.2percent of all employed of urban farming superior to other food aid policies: it (or 11 percent of the populationover 10 years old) being generatesless dependenceby the recipientof the assistance. engaged init at least part-time. It was estimated that, annually, Where it thrives, there can in fact be less need for assistance about 100,000 tons of food areproduced locally for markets from outside. This was demonstratedin 1981 in Kampala, inthe city (Sustainable Cities Programme, 1992), contributing Ugandawhere, despiteintense economic problems, UNICEF 20—30 percent of its total household food supply (Sawio, concluded that supplementary feeding aid could be 1994). In 1989, around8,500 dairy cattle, 6,200goats, 15,700 discontinued due to familiesproducing part of theirown food pigs, and over 800,000 chickens were raised in the urban within the city (Mougeot, 1993: 2—5). In another instance, regionof Dar es Salaam (Mosha, 1993). Mozambiquewas facing in 1985 severe malnutrition after Often, low-income entrepreneurs play a key role in the urban market of three years of drought in the country. Import, food aid and stocking through a variety channels, street roadside stands and markets. state farmswere not sufficient to supply Maputo,but the green including food, municipal The role of and is thus belts in and around the city proved to be a significantsource cooperation organisation particularly vital for theactivities of lower-income farmers. In some of food which could be even more productivegiven proper places the men tend to the and the women do support. The Zonas Verdes programme was, therefore, (e.g., Senegal), crops the and Members of an ethnic established, covering 16 percent of the city area with processing marketing. group inDakar farm theethnic land to overwhelmingly female (Kolstrup, 1985). together producevegetables, practitioners for market. Urban the well as the middle- rice, fish and livestock the city agricultureby poor (as by from the urban middle- urbanite failure of the urban and national Market gardeners may come poor, income) implies or from small individual political and economic systems to deal with poverty and upper-income groups, running to and intensive food whether througheconomic or enterprises cooperatives large farming provide security, activity businesses. A considerable number of urban farmers are through transfer. Self-production and barter increases thefood business owners or and of millionsof urban and secondary-school graduates, supply security poor improves the result of the economic downturn their nutrition without them on food professionals,largely making dependent of the of At one urban in assistance. Local increases access and past quarter century. campus production physical Tanzania,70 of the residentfaculty were engaged in down the of food due to increased percent may bring price supply, on the side, in this from fewer middle-menand less and agriculture (especially poultry-raising) savings transportation case using governmentland. storage. All these entrepreneurs are running profitable farming enterprises in response to the demand in the market. Urban Urban Farming for the Market farmers are more in value As for within competitive perishables, higher poverty increases, farming consumption and markets such as mushrooms, flowers the household increases.But one also finds specialised shrimp, inevitably often, and 4). In these markets, closeness an increase in market low-income farmers poultry (figure physical gardening: to the market, reduced time and cost, access their income some of the transportation supplement by selling produce toregular market information, access toprocessing or export whenever There isthus thin lineof distinction possible. only a infrastructure may be someof the factors that lead to increased and betweenthe consumptionfarmer the small commercial competitiveness of the urbanfarmer. or for farmer. Many farmers grow crops raise animals both This diversification within African market gardening and sale or some consumption simultaneously, may target favoured certain farmingsystems in particular. Rich investors and animalsfor sale while others are for the to crops family and landownersare attractedto farmingsystems that require consume. Farmerssometimes begin as family gardeners and highinvestments and produce highreturns or cater tospecialty gradually begin to sell surpluses while others are market markets for export. Poultryand fish production are growing gardeners who consumetheir surplusesand less than perfect very rapidly with investments from the wealthy and from products. foreign firms, and these ventures are supported by local banks. In Africa, food sale is one of the largest economic Thus, middle-and high-income entrepreneurial farmingmay activities in urban centres, involving more than half the be viewed by the public and by the state not as agriculture enterprises (Pedersen, 1991, cited in Drakakis-Smith, 1992). but as agribusiness. This has implications for the kind of Withinthe urban food supply sector, urban food production official support the activity encounters. is an important and often invisible sub-sector. Urban Afterobserving the waythe food production and delivery agriculture goes beyond the jobs in cultivationand animal systemworks today inAfrica, how urbanfarming now plays husbandry; it generates a number of support occupations a central role in it becomes clear. This is important upstream and downstreamfrom the food production proper. information for both urbanplanners and food policy-makers. Schilter (1991: 159—168) found in Lome, Togo, ironsmiths Agricultural activity has bloomed in a primarily hostilepolicy specialisedin forging buckets and hoes; masons who build climate. Certainaspects of the farmingare frequentlyillegal: wells; potters who create clay pots; carriers for collecting growing the wrong kind of products, planting on others'

159 Anna Raitaand Joe Nar African Urban Quarterly

Figure 4,Dakar, SenegalOrnamental horticulture for thelocal market carefullycurrent prevailing practices, then sought to establish thrives on the roadside. such as and chicken Key inputs, clay pots an action topromote the (M. M. Ndubiwa, manure, are Jocally produced programme activity personal communication,25 July 1994). This illustrateshow an urban manager can act based on the knowledgethat urban food production is indeedone of the basic urban functions, and that supporting it can have positive repercussions, not only on the local food supply system, but on the nationalone as well.

References Atkinson, S.J. (1992). Foodfor the Cities: Urban Nutrition Policy in Developing Countries. Series ed. Busi Moloele. Departmentof Public Health and Policy 1 ondon: London School of Hygiene & Tropical Medicine,

Diallo,5. (1993).Urban Agriculture in WestAfrica' Research Review andPerspectives. Paper presented at the "Workshop on UrbanEnvironmental Management: Developinga Global Research Agenda", organised by IDRC, Ottawa,Canada. property, using raw sewage for irrigation, soldat an unofficial market,etc Nowadays,the urban farmerscan often expect Drakakis-Smith, D. (1992). "And the CupboardWas Bare' a tacit toleranceof their practices. However,the possibility Food Security and Food Policy for the Urban Poor", usually remains that, any day, the fruits of theii labour may GeographicalJournal of Zimbabwe Vol. 23. tie slashed helore they are harvested, This uncertainty then altects what crops are cultivated, how much care is put into Dreseher, A.W. (1994). Urban Agriculturein the Seasonal their cultivation and so on. Tropics ofCentral Southern Africa—A ('are Study ofLusaka, Where general policies, specific land-use and health Zambia. Freihurg, Germany: Institute für Physisehe the enforcement of decisions, and expectations regarding Geographic. these decisions are clear to the fatmer, the practiceof urban cultivation and animalreaiing is enhanced.The consequences Ethelston,S. (1992). "Food Costs in Cities", Hunger Notes of the actionsof policy makers are, therefore, ofdirect hearing Vol. l8No,2. on the further developmentof urban agriculture In turn, the climate on a clear African policy depends understanding by Guyer, J. (1987). FeedingAfrican Cities: Studiesin Regional urban is an essential policy—makers thai agriculture Histor, New York: InternationalAfrican Library. component of the Afiican food supply system, hut also that, when it can have serious on improperly practised, impacts Jamal, V. (1988). "Coping Under Crisis in Uganda", the urban health and environment International Labor Review Vol. 127. Such an understanding is not completelyabsent in Africa. Across the continent a number of instances could he found N. (1986). Food and Energy in Cairo: in the of decades the of a Khouri-Dagher, past couple showing beginning the Poor. Paris: The Nexus of the actual and role of urban Provisioning Food-Energy mecognition potential United Nations agriculture. There has been presidential decrees and Programme, University. declarations such as the one in Zambia in 1977 urban urging Kiabou, B. "Le mil, facteurd'insdcuritd? Les iesidents to grow their own food; organised national (1991). paysans de Bamako menacesd'exode urhain",ISSN Bulletin Vol. 358. campaigns suchas UperationIced 'moursell in Ghana, formal metropolitansupport programmcs such as those for the 'green Kironde, J. M. L (1992). "Received and Theories belt of Maputo municipal level agriculturedepartments such Concepts in African Urbanisation and The as the district agriculture office in Dar es Salaam and other Management Strategies: Urban StudiesVol. 29 No, 8. interventions in favour of urban tarming.While someof these Struggle Continues", have representedtop down approaches, manyhave originated H. Urban Food Production from the urban food producers ot thcirsupport otganisations Kolstrup, (1985). Agriculture.' in A rca hing tip and linking to the policy-makers. and Land-Use Planning Maputo.' Project Proposal. Whether responding to a crisis or seeking to seize a Aarhus,Denmark: Masters Thesis, Schoolof Architecture. perceived potential, the individual policy maker clearly can influence, within a pat ticularcountry. city or neighbourhood, Lee-Smith,D. etal. (1987). Urban Food Productionarid the the effectiveness and productivity of urban agriculture.The Cooking Fuel Situation in Urban Kenya: National Report: Town Clerk of Bulawayo (thc second largest city in Results of a 1985 National Survey. Nairobi: Mazingira Zimbabwe)is one exampleof an cnlightenedpolicy-niaker Institute. who recognised the importanceof urban farming, evaluated

160 Volume 11 Nos. 2 and3, May and August, 1996 UrbanAgriculture and theAfrican Urban Food SupplySystem

Lourenço-Lindell, I. (1994). Informal Food Production, Sawio, C.J. (1993). Feeding the Urban Masses?: Towards Distribution and Consumption in a Peripheral District of an Understanding ofthe Dynamics of UrbanAgriculture and Bissau: A Summary of a Masters Thesis. Stockholm: Land-Use Changein Dar-es-Salaam,Tanzania". Worcester, Departmentof Human Geography, University of Stockholm. MA: PhD dissertation.Clark University.

Manshard,W.(1992). "Agricultural Change: Market Gardens Sawio, C.J. (1994). Urban Agricultureand the Sustainable in West African Urban Commu.nities—The Case of Dar es Salaam Project, Tanzania. Paper at panel "Cities Ougadougou (Burkina Faso)", in Mehdi Raza (ed.) Feeding People" at "Habitat '94" conference, Edmonton, Developmentand Ecology: Essays in Honour of Professor Canada. Mohammad Shafi. Jaipur and New Delhi: RawatPublications. Schilter, C. (1991). "L'agriculture urhaine: Une activité Maxwell, D.G. (1993). The Impact of Urban Farming in crdatrice d'emplois, en dconomie de survie (Le cas de Kampalaon Household Security and NutritionalStatus. Paper Lomd)", Cahiers des SciencesHumaines Vol. 27 Nos 1—2. presented to the First Crop Science Conference for Eastern and SouthernAfrica, Symposium on Women and Agriculture, Smit, J. and J. Nasr (1992). "Urban Agriculture for Kampala, Uganda. SustainableCities: Using Wastes and Idle Land and Water Bodiesas Resources",Environment and UrbanisationVol.4 Maxwell,D. and S. Zziwa(1992). UrbanFarming inAfrica: No.2. The Caseof Kampala, Uganda.Nairobi, Kenya: ACTSPress. Smit, J. Ct a!. (1996). Urban Agriculture: Food, Jobs and Mbiba, B. (1994). "Institutional Responsesto Uncontrolled SustainableCities. New York: United NationsDevelopment Urban Cultivation in Harare: Prohibitive or Programme. Accommodative?", Environment and UrbanisationVol.6 No. Streiffeler, F. (1991). The Knowledge System and theAction Motives in Urban Agriculturejn Sub-Saharian[sic] Africa. Memon,P. A., and D. Lee-Smith(1993). Urban Agriculture Proceedingsof the International Workshop on Agricultural in Kenya. Nairobi:Mazingira Institute (Unpublished Report). Knowledge Systemsand the Role of Extension.Hohenheim, Germany: Institut für Agrarsoziologic, Landwirtschaftliche Mosha,A.C. (1993)."Urban Farming Practicesin Tanzania", Beratungund AngewandtePsychologie. Review ofRural & Urban Planning in South & EastAfrica Vol. 1. Sustainable Urban CentresProgramme (1992). Dares Salaani EnvironmentalProfile. Nairobi: United Nations Centre for Mougeot, L. (1993). "Urban Food Self-Reliance: Significance Human Settlements. and Prospects",IDRC Reports Vol. 21 No. 3. UN-DESIPA(1993). World Urbanisation Prospects: The Mvena, Z.S.K.eta!. (1991).Urban Agriculturein Tanzania: 1992 Revision. New York: Department of Economic and A Study ofSix Towns. Morogoro,Tanzania: Draft, submitted Social Informationand PolicyAnalysis, United Nations. to International Development Research Centre. Sokoine University of Agriculture, Departmentof Agriculture, UN-FAO (1984). The Stateof Food andAgriculture.Rome: Education& Extension. United Nations Food and AgricultureOrganisation.

Pedersen, P.O. (1991). The Distribution of Wholesale and UN-FAO (1993). The State of Food andAgriculture: Water Retail Trade in Zimbabwe's New District Service Centres. Policies andAgriculture. Rome: United Nations Food and Paper presented at the Nordic Association of Development AgricultureOrganisation. Geographers Conference,Kungsbacka, Sweden. Virji, H. and S. Schram (1994). Agricultureand Food Security Popkin, B.M. and E. Bisgrove (1988). "Urbanisationand in Sub-SaharanAfrica: Issues and Future Trends. Prepared Nutrition in Low-Income Countries",Food and Nutrition for the United Nations University Millennium Feasibility BulletinVol.10 No. 1. Study, Phase II. New York: United NationsUniversity.

Sanyal, B. (1985). "Urban Agriculture:Who Cultivatesand von Braun, J. et al. (1993). Urban Food Insecurity and Why?A Case-Study of Lusaka,Zambia", Food andNutrition Malnutritionin Developing Countries: Trends, Policies and Bulletin Vol. 7 No. 3. Research Implications.Washington, DC: International Food PolicyResearch Institute. Sanyal,B. (1986). Urban Cultivationin EastAfrica: People's Responseto Urban Poverty. Paris:The Food-Energy Nexus World Bank (1989). Sub-Saharati Africa: From Crisis to Programme, United Nations University. Sustainable Growth: A Long-Term Perspective Study. Washington, DC: World Bank.

161 L'ifrica Today has found a strong voice s ournl that uces3IuUy combines [nd academic analysiswith timeliness nd policy rvnce.' —Marina Ottaway

4fl I; JO ArnulUcultural journal Africa Today awsthe 1*ugi i, best scholarly—yetaccessible—work from around the world on a full range of political, economic, and social issues. The lyurnal is VOUJME 45 ('998j ssuis) committednot only to prodding information 4 SS 0001 9UE17 and insight, but also to offeringpositive solu $70 boristo the problemsfacing Africa today. $32 STU0EITS $20

Africa Today is published quarterly. Po7AoA TA'XA.A.RR S ELSf1ONS OUT- OSTHOU S oi cANAA All articlesare externally reewed. • &2000 rI ALL< M4W US OOUARB 4bS TAf A U.S. a4. Like Africa, Africa Today is reinventing CoLcwO'SOWTOTTT PLAASE ADD itself. The new initiatives taken 3 SALES TAX. by SEScLSL OTAES ThES 1O%oAHy Africa Todaypromise to make it live up LTUTPXJWPTS)HRU(ITATT TO YT* .FTScW to its name—and to make it one of the most refreshing, invigorating, and lively publications on Africa." —David Newbury

RECENTCONTRIBUTORS INCLUDE Robert H. Bates • Michael Chege R. Hunt Davis, Jr Jane I, Guyer • Ivan Karp • Ron Kassimir • Christopher C. Lowe William Martin • Julius E. Nyang'oro M. Priscilla Stone Michael Watts • Michael Wet • PaulTiyambe Zeleza

180030 S •B.. . 0080301 SUSAST (303)4440824 FAX(303) 4440824 WWW RENNLR COM African Urban Quarterly, 1996: 11(2 and 3) pp. 163—168 @1996 by African Urban Quarterly Ltd.

HOW DO THE URBAN POOR STAY ALIVE? FOOD PROVISION IN A SQUATTER SETTLEMENT OF BISSAU, GUINEA-BISSAU

Ilda Lourenço-Lindell Departmentof Human Geography, Universityof Stockholm, 106 91 Stockholm, Sweden

Abstract Accepted May, 1996 This study takes departurefrom thefact th4tsignificant parts of urban populationsin the Third World have poor access tofood. Approachesto food insecurity tend to bepartialfocusing narrowlyon food availabilityor on people' endowmentor ownership situation. Thispaper discusses how those in a verypoor condition survivein a dominantly cash economy and a rapidly changing urban environment. In particulai the paper intends to show the importanceof food gifts and poverty sharing for urbanfood security. It argues that command over food depends not only on availabilityof food and people's endowmentsituation but also on assistance networks, often determinedby cultural codes. This exercise draws on empirical data from a squatter area of Bissau, Guinea-Bissau, collected through an eclectic methodology in which qualitative and quantitative methodsare used in a complementary manner Résumé Get article se distance de l'idée courante que des secteurs impcrtants de Ia populationdans les villes du Tiers-Monde auraientdifficilement accès auxaliments dont ilson besoin. Sans Ia littérature, le traitementde I 'insécuritéalimentaire tend a être partiel, insistant defacon un peu myope sur Ia disponibilité d'aliments, Ia dotation des gens en ressources ou leur appropriation de celles-ci.L'étude montrecomment les secteursplus démunissurvivent dans une de marché et un milieu urbanen rapide evolution.Tout particulièrement, cet article souligneI 'importantecontribution des dons etdu partage de nourriture a Ia sécurité alimentaire des pauvres dans les villes. On soutient Ia these que le contrôle de Ia nourriture depend non seulementde sa disponibilitéet de I 'état de dotation des propres individus ou ménagesmais aussi de réseaux d'entraide, souvent codifiespar Ia culture locale. L'argumentest illustré par des donnés empiriques sur un bidonville de Bissau enGuinée-Bissau, recueillies grace a une approche exploitantIa complémentarité de méthodes qua litatives et quantitatives.

Introduction in that availability. Although there has long been a general Third Worldcountries have been experiencing, in the last awarenessof thedistinction between there beingenough food decades, an accelerated urban growth. Tropical Africa, and having enough food to eat, this approach continues to although one of the least urbanised areas of the world, has dominatemany national diagnoses and plansfor foodsecurity. now one of the highestrates of urbanisation. The inabilityof Among the studiesexposing the insufficiencies ofthe food urban labour markets to absorb a rapidly increasinglabour availability approach, thework of Sen (1980: 613—621; 1981; force has led to large-scaleunemployment in many African 1984 and 1986) has been a major contributionto the debate urbancentres. Rising levels of inflation,often in acontext of on food security. Besides showing empirical evidence that deepening economic crisis and Structural Adjustment starvation andfamines can develop without a decline in per Programmes (SAPs), have been followedby a fall in real capita food availability—whichis the typical situation in wagesand purchasingpower (Gilbert and Gugler, 1992 and Third World urban centres—this author developed the O'Connor, 1991) . This urbanimpoverishment has resulted entitlement approach. This approach sees starvation as inan increasingshare of urbanpopulations who fail to secure resultingfrom failure of somepeople to establishentitlement their access to purchased food, on which they depend more over a necessary amountof food. Sen introduces the notion heavilythan rural populations(Pryer and Crook, 1990). How of command over food which highlights the problem of do the urban poor survive in a dominantlycash economy unequal accessto food by different groups of peopleand the with low and irregular incomes? injustice of starvation amidst abundance. He calls this an 'acquirementproblem', which he seeks to grasp by focusing Urban Food Security and Entitlements on the relaionship of people to food and on the structureof ownership to understand starvation.According to Sen (1980: (a) Entitlements to Food 616) "ourability tocommand food dependson what we own The traditionalapproach to food security focuses on the and what we can getby exchangingwhat we own", according availabilityof food, equating it with per capita availability to a set of entitlementrelationships legitimised by the legal of food and explainingstarvation and famines with adecline systemruling a given society. Acknowledgements—I would like to thank Mats Widgren for his commentson this paper and to the Scandinavian Instituteof African Studies, Hans W-Son AhlmannsFond and Lagrelius Fond for financingfield works in Guinea-Bissau.

163 lIda Laurenco-Lindeil African Urban Quarterly

What Sen calls the entitlementset of a person in aprivate to the question how do the very poor manage to stay alive? ownership economy is determined by two conditions: his/ Sucha positivefocus is neededif future action is notonly to her endowmentand his/her exchange entitlement mapping. address the causes of hunger,but also to support thepopular The formerconsists of the person's bundle of ownership, that survival mechanisms alreadyat work. is essentially, what one produces with one's own resources, one's meansof production,assets and own labour. The latter (b) Urban Food Security corresponds to the set of alternative bundles that one can In spite of the dimension and gravity of the urban acquire, given his/her endowment bundle. "A person is undernutrition problem, only modest research efforts have reduced to starvation if some change either in his/her been invested in this area of study. These have clearly been endowment or exchange entitlement mapping makes it no delayed due to the myopia of the availabilityapproach. The longerpossible for him/herto acquireany commodity bundle fact that urban centres often display abundanceof food and with enoughfood" (Sen,1986:9). While this analysis has great constitutethe nodes throughwhich food aid and importsare relevance for Third World urbanareas plagued with declining channelledhas contributedto give the urban food problem a real incomes, shrinkingemployment and rising food prices, lowpriority in theresearch agenda until recently. A significant it does not exhaust all the ways through which peopleavoid part of the literature on African urbanisation has literally starvation. In fact, although Sen states that what a person is ignored food security issues, even though food is the most entitled to depends Ofl thelegal and political system, economic basic of needs(Peil, 1981 and O'Connor, 1983). Buta number forces and the social and cultural characteristics of a society, of studies have emergedwhich recognise the food insecurity this statement has little consequence in his analysis. The of the urban poor, their heavy reliance on purchased food authorchooses to focus on "the systemof entitlement relations and weak purchasing power, leaning towards Sen's operating in the economy" or possession-based entitlements entitlementtheory. However,studies on the role ofsolidarity and to dismiss the other not less importantfactors influencing and safety networks for the food security of the urban poor one's access to lood. It appears that it would he enough to are virtually non-existent. This paper isone first step towards make an exhaustive inventory of the endovment and closing this gap. exchange entitlementsituation of a person todetermine with In this research project, food provision for the poor is precision his/her level of commandover food. addressed in a comprehensive manner,looking to the various Such an approach assumes that all individuals are well ways they gain access to food—through purchasing it, informed as to the market value of their assets and labour, involvement in food production,food gifts and even stealing. have perfect knowledge of the market and always make Much of the work on urban food supply has seen the urban economically rational decisions.Types of behaviourthat do population as totally dependenton food sources exterior to not fit this homo economicus model are dismissedby Sen's the urban centre (Guyer, 1992 and Fryer and Crook, 1992), entitlementapproach as unimportant'choice failures' caused but somerecent studies document therelevance of agricultural by ignorance or apathy. Charity,for example,is considered a activitiesin and aroundurban areas (Rogerson,1993: 33—44 non-entitlementtransfer, which is notgiven any relevancein and Schilter, 1991). In particular, Rogerson (1993: 33—44) his analysis. It is the argument of this paper that such refers to the 'ruralization' of urban centres in sub-Saharan behaviours, considered irrational from the economic or Africain thecontext of economic crisis and austerityimposed western point of view can be of crucial importance for the by Structural AdjustmentPolicies (SAPs) and sees it as a food security of the urban poor. In particular, charity and resultof thestrategies of urbanhouseholds tocope with rising safety networks give many poor access to food, regardless living costs. Research on the distribution networks linking of their ownership situation.There is a need to explore more urbanproducers and consumers and the role of both urban deeplythe roleof choice toenjoy or not theusufruct of one's food production and distribution in the food securityof urban endowmentand exchange entitlementsituation and to look dwellersis very limited. These aspects of urbanfood supply more closely into the social and cultural codes that often tend to be addressed isolated from each other. The present underlie such choices, and in some contexts, might study intends to correct this by integratively analysing superimpose on capitalist behaviour. household food consumption and its involvement in food Theimportance of institutional conditions which can give production and distribution activities, understood here as or deny people the ability to exercise one's ownership and interdependent components of the 'urban food system' exchange entitlementmust be accountedfor (Woldemeskel, (Drakakis-Smith, 1993). 1990:491—495). Similarly, the importance of illegal practices such as stealing, illegal cultivation or trade should not be Research Methods ignoredsince the poor, and the urbanpoor in particular, often The findings presentedin this paper are the result of a have to resort to illegalityto survive. research strategy that uses qualitative and quantitative While some of these aspects of food security must be methodsin acomplementary manner. This is aimed atbetter delayeduntil more mature phases of this researchhave been graspingreality through combining thestrengths and reducing reached, the focus of this paper is on the role of 'irrational' the limitationsof those methods.A qualitative approach will economic behaviour and cultural obligations for avoiding reveal food insecurity as experienced by thepoor, their coping starvationand improvingpeople's accessto food.This study strategies,the cultural codes and the meaningsbehind their attempts a positive approach to the problem, turning from actions in the realm of the food economy. Quantitative the question why somepeople fail to comnwnd enoughfood, methods will allow for testing the significanceof a number

164 How Dothe Urban Poor StayAlive? Volume 11 Nos. 2 and 3, May andAugust, 1996 Food Provision in a Squatter Settlementof Bissau, Guinea-Bissau of variables identified throughqualitative work for a wider increase of percapita availability of rice, themain staple food, population and will facilitate communication of research largelydue to arapid growthin rice imports (Guinea-Bissau, results to policy makers and more positivistresearchers. 199lb). At the domestic level, the end of state control over Thisexercise takesplace in asquatter area of Bissau called tradingactivities has been expressed in a boomingexpansion Bandim. During 1992, asix-week exploratory fieldwork took of small-scaletraders as well as food markets in the urban place and from January to April 1995 a survey on the periphery, most of which are of a spontaneous nature. household foodeconomy was carriedout. The surveycovered However, although food availability seems to have increased atotal of 335 householdscorresponding to around 10 percent somewhat, it is doubtful that food security by the majorityof of the population ofone squattersettlement. Households were the urbanpopulation has been improved. selected on thebasis of randomsampling ofhquse units, since SAP's policies have resulted in further unemployment a list of householdswas not available.All householdsliving and frozen salaries in the urban-based public sector (Aguilar in eachsampled house unit were interviewed and, if that house and Zejan, 1991). This imposed austerity has hit the urban was part of a compound (a group of householdsrelated by population in a particularly strong way. Guinea Bissau is kinshipand acommon head), all households inthe compound experiencingan intenseimpoverishment process, in which a were questioned.The householdunit is consideredhere, for largeshare of its population is facingrising levels of inflation the purposes of the survey, as the group of people that had and a declining purchasing power. In this context, the been eating from the same stove (hereinafter used endowmentsituation of the urbanpoor is of great importance interchangeably with household) during at least six of the to understand the relation betweentheir loss of entitlements twelvemonths prior to the survey. This unit is merelya point and their deteriorating food security. But the entitlement of entrance into a complex reality and its limits dissolve as approach gives only an incomplete picture of the problem. one goes more deeply into the constellation of networks that One has to go beyond it to understand how the very poor, overlap in the struggle of providing food for its members. those with an endowment and exchangeentitlement situation The person of the household interviewed was the one close to nil, stay alive. supervisingor responsiblefor buying and preparingthe food for consumption or, alternatively, another person in the Food Provision in Bandim householdinvolved in those tasks and aged 16or more years. Qualitative interviews with households,different types and (a) The Bandim District scalesof foodproducers and traders as well as key informants The majority of Bissau's populationlives in unplanned both in 1992 and 1995 provided the grounds both for settlementslacking basic servicessuch as electricity, sewage elaboratingthe survey scheduleand probing the validity of or running water. Bandim is the largest of these settlements, its results. Although analysis of the survey data is still in accommodating about23,000 people (Guinea-Bissau, 1991a). process, it is already possible to present here some of the Crowdingin this district increased from 10.9 to 16.4 persons results that support the argument of this paper. per house and childhoodmortality was 230/1000 in 1987— 1990(United Nations Centre for HumanSettlements, 1994). Food Supply and the Poor in a Rapidly Changing Urban The surveyconducted earlier in 1995 showed that 68 percent Centre of the adults in the enquired householdsate only one meal In spite of its modest size of around200,000 inhabitants, per day, that is, they ate nothing either in the morning or in Bissau's populationis growingrapidly at an annualrate of S the evening. This widespread practice is locally known as percent and accounts for 20 percent of the total population the twenty-four-hour-shot. of the country (Guinea-Bissau,l99la). After independence In spite of this poor scenario, Bandimprovides significant in 1974, the socialist one-party government concentrated opportunities for food production. The district includes investments on large-scaleindustrial projects aroundBissau extensiveland beyond the built area which, by tradition,falls (Davila, 1987). These investmentsand the expansion of state under the administration of the Papels, the dominant ethnic employment in Bissau intensified the already accelerated group. Much of this area is used for rice cultivation in salt growthof capital city. The failure of both to absorba rapidly water but many vegetablegardens also exist. The Bandim growing urbanlabour force resultedin risingunemployment. area also benefits from the proximity of the mangrove, an In the context of a deteriorating economy and widening area rich in fish, crabs, birds and fuelwood as well as an fiscal and external deficits, Guinea-Bissau negotiated the artisanal port that supplies the city with fresh fish. SAPs with the World Bank in 1986. The programme was aimed at correcting internal and external distortionsof the (b) Modes of Food Provision economy, stimulating economic growth and improving the The survey conducted early in 1995 intended to yield balanceof payments.These aims were to be attainedthrough basic information on modesof food provision among Bandim exchange rate realignment, reduction of public investment dwellers for later qualitativestudy. One of the results elicited and expenditure as well as trade and price liberalisation by the survey is the great importance of self-employment as (World Bank,1987). Thus, a radical shift from a centrally a main source of cash income. This is hardly surprising, planned economy towards a liberalone has gradually taken consideringthe shrinkingwage employment. However, this place, with important consequences for the food sector. fact deserves not to he omitted due to its importance for The liberalisation of external trade has led to a more people'sexchange entitlement. One consideration to he made diversified supplyof food. It has also resulted in a significant is theirregularity of theincomes of the self-employed which,

165 Ilda Lourenco-Lindell African Urban Quarterly for those who do not manage to keep any savings,means a exchangeentitlement of its practitioners.In Bandim, among varyingcapacity to purchasefood. As asupplementary source the 114 farming households surveyed (34 percent of the of income to wagework, self-employment is no lessimportant sample), 40 percent sold more than half of the production when themonthly salary is not enoughto meet the food needs and of these 39 percent spent most of the resulting incomes of an average sized household, which is definitelythe case on food. in Bissau. Whetherurban food producers are thepoorest of the poor More enlightening is the outstanding position of food needsfurther investigation. While it can be argued that food trade (including trade of foodstuffs, drinks and directly related production demandsaccess to resources,such as land, out of products such as coal) as a source of cash income for the reachof the poor, acloser lookat the natureof such access Bandim's households. Food trade constitutes the largest is required. Many food production activities in Bandim are category among the main incomeactivities, amounting to 37 carried out with few cash expenses, with borrowed inputs percent of the surveyed stoves and accountsfor the majority and based on solidarity ties. Solidarity is the piece of the of secondary income sources. This calls attention to the puzzle missingin most analysesof food security. double importanceof urban food trade for the food security In rice farmingin Bandim,mutual help betweenfarming of the urbanpopulation. Not only does it makefood available households of the sameethnic group is important in providing for purchase but also gives a large share of the urban labour for the working seasons, in borrowing tools, population access to cash income with which to buy food. exchangingseeds and protecting the paddy from birds. Land From the 330 household enterprises engaged in food trade, for rice production is still to some extent distributedby the 56 percent spent more than half or virtually all of the earned local elders and plots for gardening are often borrowedor incomeon food alone (22 and 34 percent, respectively). Many rented from them for symbolic prices.Canoe fisheriesimply poor food retailers are caught in a vicious cycle in which higherexpenses but there is similarlymuch mutual assistance they often have to 'eat' their working capital in order to stay between fishermen, without which many could not operate. alive,which reducesthe scale of their activity and intensifies To offer a part of the catch to relatives (not belonging to the their poverty. Ironically, those food sellers operating at the fisherman's household) or to a fellow fisherman who was smallest scale and poorest conditions are of crucial not able to go out fishing is a common practice. importanceto the urban poor because they make available Thus, different components of the food economy show small amountsand/or low cost foods that thelatter can afford networks of assistance with differentconfigurations. The food on a daily basis. security of a household reflects the constellation of these In the face of decliningpurchasing power, the importance networks that go far beyond its limits. But let us now focus of alternative sources of food for the poor is probably on thosesolidarity bondsthat directlyaffect provision of food increasing. While this trend is not as yet easy to prove, the for consumption. Depending exclusively on gifts to meet survey data is quite elucidative about the involvementof one's food needs is one mode of food provision in Bandim Bandim's dwellers in food production.About 33 percentof and deserves therefore attention. Moreover, a large share of the inquired households had members involved in farming households depends to a greateror lesserdegree on food gifts. during the twelvemonths prior to the survey, 52 percenthad About 68 percent of the surveyed householdsdeclared that raisedlivestock and 39 percenthad engaged in picking fruits. they had received gifts in cash, food or meals in t.he twelve Fishing activities in Bandimare another important component monthsprior to the survey. This seems, however,a low figure of the quarter's food economy not so much because of the considering that it is usual to share meals daily with the amount of population actually involved (only 8 percent of neighbouring stoves,which many respondentsdid not regard the surveyed stoves had practisedit), but because it is asource as gifts. of incomefor alarge numberof fish retailers and an important Relatives and neighbours were the dominant source of source of fresh fish for Bissau (Guinea-Bissau, l991a). thosegifts (95 percent), belonging mainlyto otherhouseholds Food production by urban dwellers is an important living in the same house or compound(73 percent) but also componentin their food security. Such activitiesprovide the in other parts of Bandimand Bissau.It is importantto notice urban poor with foodstuffsthat they otherwisemight not be that II percent of the households had received gifts with able to afford and reducetheir dependency on purchased food originoutside of Bissau. The frequencyof these gifts varied through direct entitlement to what they produce. They also between one to several times a year and often dependedon diversifytheir diets and occasionallygenerate some cash to visits to those areas.Transfers in the oppositedirection, that pay for other basic items. The latter is clearly the case of is, fromthe surveyed households involved in foodproduction livestock raising, which often constitutes a safety valve, to their rural areas of oiigin, were also seldom mentioned— generatingcash to buy the basic staple in times of hardship. only 7 percentof the farming householdsreported to do so. In thecase of rice farmersin Bandim,who often have limited Most of the gifts of the enquiredfood producersare directed access to cash, subsistencerice farming provides them and to other stoves in the same house or compound or within their dependents with the main constituentof their meals. Bandim. But urban food production goes beyond the subsistence level, When askedwhat they did when they had no food to eat although the relative weight of subsistence and commercial or money to buy it, 43 percentof the respondentsasked for production varies from one product to another. Market- food or money from other households living in thesame house oriented food production in the urban centre improves or compoundand 32 percent turned to other relatives and availabilityof fresh foods for the urban populationand the neighbours in the urban centre. Particularlythe first figure

166 flowDo the Urban Poor StayAlive? Volume 11 Nos. 2 and 3,May andAugust, 1996 FoodProvision in a Squatter Settlementof Bissau, Guinea-Bissau seems to underestimatereality, especially considering that provision in Bandim. This is evident in the way many food 68 percentdeclared receiving gifts in the last twelvemonths. producers use their production. About 74 percent of the This is due to people's reluctanceto ask for food and to admit householdsinvolved in cultivation and fisheries gave away that they, nevertheless, rely on the moral obligation of part of their production even if it was not enough to sell. neighbours to offer them some food if they have none. Rice farmers, in particular, refuse to sell an eventualsurplus. These moral obligations concerning food gifts reaches They tend to keep it not only for the next cultivationcycle such adimension that can hardly be understood by the western hut also for gifts and for cultural ceremonies—peacewith mind.About 90 percentof the surveyedhouseholds declared the supernatural world is perceived as a requisite for well- they offered food to people in need even if the remaining being. Similar considerations can he made about how the food wouldnot be enoughto meet theneeds of the household. poor handle their assets. It is common that an impoverished This poverty sharing is of crucial importancefor the poor house owner wil not charge rent from householdsthat are because it is the guarantee that they will virtually always related to him/her by kinship or are in an equally poor find something to eat. For the donor, it represents a social situation. It is also not probable that a person who is out of capital/investmentto which he/shecan appeal for in times of cash will sell his/her house or land in order to buy food. The shortage. entitlementapproach does not accountfor choices which are not coherent with market principles and remains at a (c) Freedom of Choiceand Culturally-Based Entitlements somewhatutopic level detached from reality. The above set of data indicatesthese bonds of assistance By focusing on those means of commanding food that are stronger between stoves sharing the same roof and are legitimised by the legal system governing a given society, belonging to the same compound. Many households in Sen underestimates the importance of illegal practices that Bandim are still organisedin compounds—agroup of stoves might constitute an important mode of food provision. For interrelated through kinship and a common head—the example,13 percentof the inquiredhouseholds had reported traditionalform of residential in the country. The degreeof thefts of food in thepast twelvemonths, but the numberwould assistance between stoves sharing the same house varies be much higher if other assets had been included. Informal greatly, dependingon possible kinship ties, length of living references to such thefts were numerous. Such activities together and friendship. But particularlywithin the compound representa lossof entitlementsby the victimised households there are traces oftraditional systems of assistance, eventually and a means of gaining access to food for others. Onepossible partly eroded by the influences of urbanisation, in mutual consequence is that members of already poor households assistance. There is a division of rights and duties in the might be deprivedfrom earning an incomein case they have compound that determines what type of assistance one to stay home to protect their properties(considering the low membercan expect from another in case of need. Similarly, level of securityof the houses of the poor). The poor are also one candifferentiate the providers and the beneficiaries within those less able to defend their own legal rights, which often the stoveunit. Theseculturally determined systems are being results in losses of entitlements and reflects their political studied through qualitative methods, but due to their (not only economical) marginalisation. The political nature complexity(they are differentfor the differentethnic groups of command over food is also evident in the differentiated living in Bandim), more qualitative work has to be carried access people have to social resources and networks of out before a full account can be given. But it is clear that assistance, such as the compoundand the kin group. such cultural obligations should not be ignored because in some contexts they are of critical importanceto the survival Conclusion of the poor. The entitlementapproach dismisses such non- What determines the mode of food provision of one market oriented behaviour by focusing on endowmentand household is the ultimate challenge in this study. Which exchange entitlement. It is necessaryto extend Sen's largely modes of food provision are the most resilient to starvation possession-based entitlement concept to incorporate these is another important question that needs further research socially- and culturally-based entitlements. These beforean answer can be formulated. entitlements are notdetermined by one's endowment situation I have here deliberately focused on the positive side of butby a set of legitimateclaims that one can make todifferent the influence of culture and tradition in food provision. But entities in the community and society to provide one with one must not ignore their less positive effects on people's food. lives. Traditional relationsin somefood production activities One aspect that stands out in these findings is the inBandim give an unfavourable position toyouth and women, prominent role of the household and the community who have poor control over productiveresources and weak (understood here simply as the neighboursand the kin group decision-making powers. Anotherexample is the diet taboos in the compoundor withinreach) as basic institutionsfor the and the concentrated consumption of meat during cultural provision of food, an essential area of social reproduction. ceremonies. Theinteraction of these traditional elementswith This realityacquires new meaningin a time when the state is the modern urban environmentare the source of struggles officially withdrawing itself from former responsibilities, and conflicts which reflect the process of conservation and such as subsidisingrice topublicsector workers. dissolutionthat is typical of many pen-urban areas.Are these Anotherimportant reflection is the non-profitlogic and culturalbonds of assistancein food provisioneroding under 'unrational' economiccharacter of a significantpart of food the influenceof urbanisationor reinforcingin the face of the

167 Ilda Lourenço-Lindell African Urban Quarterly current austerity and deteriorationof living standards? The Peil, M. (1981). Cities and Suburbs: Urban Life in West answerto this question is of crucial importance to the future Africa. New York: Africana PublishingCompany. survival of the poor. Pryer, J. and N. Crook (1990).Cities of Hunger: Urban References Malnutrition in Developing Countries. Oxford: Oxfam. Aguilar, R. and M. Zejan (1991). Guinea-Bissau: A Fresh Start? Gothenhurg: Department ofEconomics. Unpublished Rogerson, C. (1993). "Urban Agriculture in South Africa: PhD thesis. Policy Issues from the International Experience", Development SouthernAfrica Vol.10 No.1. Davila,J. (1987).Shelter PovertyandAfrican Revolutionary Socialism: Human Settlements in Guinea-Bissau. London: Schilter, C. (1991). L'Agriculture Urbaine a Lomé. Paris: Karthala. InternationalInstitute for Environmentand Development.

Sen, A. World Vol. 8. Drakakis-Smith,D. (1993). "Food Securityand Food Policy (1980). "Famines", Development, for the Urban Poor", in J. DahI et al. (eds) Land, Food and Sen, A. (1981). Poverty and Famines: An Essay on BasicNeeds in Developing Countries. Goteborg: Department Entitlement and Deprivation. London: Oxford ofHuman and EconomicGeography, University ofGoteborg. University Press. Gilbert, A. and J. (1992). Cities. and Gugler Poverty A. Valuesand Oxford: Urbanisation in the Third World. Oxford: Sen, (1984). Resources, Development. Development Basil Blackwell Publisher Ltd. Oxford University Press. Sen, A. (1986). Food, Economics and Entitle,nents. Finland: J. Cities. London: The Guyer, (1992). Feeding African World Institute forDevelopment Economics Research. InternationalAfrican Institute. United Nations Centre for Human Settlements (1994). Guinea-Bissau RecenseamentoGeral da (1991 a). Populacäo Crowding andHealth in Low-Income Settlements, Case Study e Habitacão. Bissau: Instituto Nacional de Estatistica e Report, Bissau, PreliminaryDraft, 1994. Nairobi: UNCHS, Censo. Habitat.

Guinea-Bissau (199lb) Plan Cerealierde Ia Guinée-Bissau. Woldemeskel, G. (1990). "Famine and the Two Faces of Paris: CILSS et MDRA. Entitlement:A Commenton Sen", World DevelopmentVol. 18 No. 3. O'Connor,A. (1983). The AfricanCity. London: Hutchinson University Library for Africa World Bank (1987). Guinea-Bissau: A Prescription for Comprehensive Adjustment. A World Bank CountryReport. O'Connor, A. (1991). Poverty in Africa: A Geographical Washington, DC: World Bank. Approach. London: Beihaven Press.

168 Community Development Journal Volume 33 1998 4 issues Edited by K. Popple, University of Plymouth

To find out more about Community Development Journal please visit our web site at http://wwwoup.co.uk/cdj where you can register to receive the forthcomingtable of contents by e-mail, whether you subscribe or not

Published by Oxford University Press

To SEND FOR A FREE SAMPLE COPY SIMPLY PHOTOCOPY THIS FORM AND SEND OR FAX IT TO THE ADDRESS BELOW

Please send me a free sample copy of: ( Oxford Press\ University i Journal Journals X98 Community Development i Marketing I Name Great Clarendon Street I Address I Oxford 0X2 6DP i I Fax: 01865 267485 I ) City/County i' Postcode AfricanUrban Quarterly, 1996: 11(2 and 3) pp. 170—179 ©1996 by African Urban QuarterlyLtd.

URBANAGRICULTURE, FOODSECURITY AND NUTRITION IN LOW INCOMEAREAS OF THE CITY OF NAIROBI, KENYA

Alice Mboganie-Mwangi and Dick Foeken, respectively Unit ofApplied Nutrition, University of Nairobi,P.O. Box 30197, Nairobi, Kenya; and African StudiesCentre, P.O. Box 9555, 2300 RB Leiden,The Netherlands

Abstract Accepted December, 1995 This article discusses howfar farming activitiesperformed by low incomedwellers within the City ofNairobi play a role in thefood security and nutritional conditionof the householdsinvolved. A comparisonis made between three low income groups, i.e. those who dopractice urbanfarming, those who do not, anda group involved in an urbanfarming project. The results showthat as far as food security is concernedurban farming doesplay a prominentrole, but this is not translated in a betternutritional condition of the young children. Résumé Cetarticle traite de Ia mesure dans laquelleles activités agricoles pratiquéespar les citadinsde has revenus, en deçá des limites de Ia yule de Nairobi, contribuenta Ia sécuritéalimentaire et a Ia nutrition des ménagesimpliqués. L'étude tisse une comparaisonentre trois groupes a bas revenus:ceux quipratiquent l'agriculture urbaine, ceux qui ne lapratiquentpointet un troisième groupe, engagédans unprojetd'agriculture urbaine. Les résultatsmontrentque l'agriculture urbaine contrihue a Ia sécurité alimentaire des ménages,même si Ia conditionnutritionnelle desjeunes enfants dans les ménagespratiquants n 'est pas netternentsupérieure a celle des enfants dans les ménagesnon-producteurs.

Introduction al., 1992:284—294), while goats,rabbits and poultry have also Until recently, urban agriculture—which can be defined been reported in Kampala, Uganda (Maxwell and Zziwa, as any farmingtechnique in an urbanenvironment (Maxwell l992b) and Harare (Drakakis-Smith,1992: 258—283). and Zziwa, 1992b)—was believed to be an insignificant Farming within the City of Nairobi is not a new cultural practice carried over from the rural areas and was phenomenon except that its intensity increased in the late ignored by academics and planners. Recent research, I 980s. Food pricespartly explainedthe rapid increasein the however, suggests that urban agriculture is potentially a cultivationof open sites, backyards,river valleys, road and livelihood-enhancing strategy for the low income urban rail reserves in city and other urban areas in Kenya (Lee- dwellers (Sawio, 1993: 59—82). Urban agriculture is usually Smith et al., 1987). Freeman (1991) observed that farmers an activity unplanned and uncontrolledby the state. Apart growingcrops had curved out irregular boundariesfor their from farming in backyards (mainly by those with some plots depending on who camefirst. Some operatedon public unusedland space on theircompounds) and farming inrural land leased to them by 'landlords' at a rent as high as Ksh. areas which became part of the urban area due to the 1,000 per annum (Gathuru, 1988). Another study revealed expansion of the urban boundaries(Memon and Lee-Smith, that the majority of such farmers are women, very poor, 1993: 25—42, term as 'urban farmers' these traditional landless and subsistence dwellers while farming at the landownersor farmers),it involves food productionon idle backyard of residential houses is a characteristic of a few and/or reserved land as a mode of survival by many low middle and upper socioeconomic groups as these can afford incomeurban people. It is this latter type of urbanagriculture housing with unused land space (Lado, 1990:257—266).A that this article is concernedwith. survey carried out in KiambuDistrict, Kenya indicated that In East and Central Africa, urban farming has been such factors as family size, landlessness, unemploymentand practisedsince the late 1960sand early 1970s (Sawio, 1993: the need to grow food crops for domestic consumptionand 59—82). It has been officially recognised in Zambia but for sale to obtain a cash income motivated farmers (mainly attempts to bar it surface from time to time (Sanyal, 1987 females) to encroach on roadside reserves as a means of andRakodi, 1985: 53—61 and 1988:495—515).Studies carried sustaining their livelihoods (Mutisya and Lado, 1991:107— out in this region reveal that a majority of the farmers are 127). womenof low socioeconomic class. Urban agriculture tends Most studies pointed at problems of land tenure for the to be a livelihood strategy to supplement their inadequate poor urban farmers. They 'illegally' farm on land that does incomes by producing food on any available land (Sanyal, not belong to them hence they are faced with problems of 1987; Rakodi, 1988: 495—515; Drakakis-Smith, 1992: 258— crop and/or livestock insecurity. In the general survey on 283 and Maxwell and Zziwa, l992a: 91—109). Crops urban agriculture in Kenya carried in 1985 by Lee-Smith et produced include mainly vegetables and fruits plus some al. (1987), 6percent of the urban farmersmentioned to have legumes and root crops (Mlozi et al., 1992: 284—294 and experiencedeviction from their plots by the landowner or Maxwelland Zziwa, 1992b). In Tanzania,milk and poultry municipality. In the City of Nairobi, this applied to only 3 have been reported to be produced in urban areas (Mlozi et percent of the respondents.

170 Urban Agriculture,Food Security and Nutritionin Volume 11 Nos. 2 and 3, May andAugust, 1996 LowInco,ne Areas ofthe City ofNairobi, Kenya

Until now, very few studies focused on the food StudyAreas consumption and/or nutritional condition of poor urban The study was conducted in the City of Nairobi and dwellers.Alarcon and Rivera (1994:171—1 82), studyingthe covered two cluster areas that had already been identified impactof changesin purchasingpower on foodconsumption with poverty: Korogocho and the Kitui-Kanuku-Kinyago area of an urbanpopulation in Guatemala City found that the mean (figure 1). Korogocho area was intentionallyselected from a energy intake for householdswith thelowest incomes ranged list of already identified poor urban zones in the City of from 70percent to 78 percentof the requirements and protein Nairobi.Its selectionwas based on the fact that there was an intake from 92 percent to 106 percent of the recommended adequatenumber of householdspractising urban agriculture. level.Atuanya (1987:109—127)carried out a dietary survey It served as a typical low income area without any agricultural of the urban poor in Benin City, Nigeria and found that, on influence from Nongovermental Organisations(NGOs) and average,daily individual calorie intakewas about 75 percent its agricultural activitieswere purely self-initiated. The area of the requirementsand the average protein intake was also is located about eight kilometres from the urban centre, less than the recommendedsafe level. towards the northeast. its population was estimatedat 75,000 The potential contributionof urban agriculture to the food in 1990 (World Vision International,1990). Among the poor supply of the urban population, in particularthe urban poor districts of the City of Nairobi, Korogochohas been shown who practice it as a survival strategy, is revealed in various to have the lowest monthly income per household head studies. Sachs and Silk (1990) cited a numberof studies that (Kenya/UNICEF, 1990). highlight the importance of urban agriculture in several The Kitui-Kanuku-Kinyago area is situated three countries.For instance,urban agriculture has been shown to kilometresfrom the city centre. The combined villageswere improve vitaminA levels in diets in many Asian countries, estimatedto have 10,000 households with more than 40,000 suchas Itidonesia (Yeung, 1987: 14—23). In somesmall urban people living there. These were the villages involved in the gardening projects such as in Argentina, high yields were so-called Undugu Society Urban Agriculture Project obtained, while livestock could he fed with the waste from (USUAP).The USUAP is part of a wider project on slum the gardens (Wade, 1987: 29—35). developmentand organisedby the UnduguSociety of Kenya Several surveys conducted in 1990—1992 in low income (USK) for 'underprivileged'people living in the low income areas ofthe City of Nairobirevealed that malnutrition among areas. The USK startedas a small parkingboys (Street boys) young children was quitecommon. Theprevalence of wasting scheme launched in 1975/76. It has now developedinto an among children aged 6—60 monthsranged from 5 percent to extensive low income development project. The USUAP 13 percent(the national rural figure in 1987 being 2.5 percent) started in 1988 and its aim was to provide household level and stunting from 10 percent to 57 percent (19.6 percent in food security for the householdsinvolved. The initial target Kenyaas a whole; Kenya, 1991). In 1992, the level of wasting areas were the three slum villages Kitui-Pumwani, Kanuku in Kawangware,a low income area in the southwestof the and Kinyango located on the eastern side of the City of City of Nairobi was 8.3 percent (Kenya/UNICEF, 1992). Nairobi. Plots with a size of 165 square metres (3 x 55 m) Past studies on urban agriculture in Kenya have along the Nairobi River were allocated through the local concentrated eitheron the urbanagricultural activities of the governmentto 105 low incomehouseholds. The individuals urbanpopulation in general, i.e. in all segmentsof the urban were given agriculturalresult demonstrationsand assistance population(Lee Smith et al., 1987) or on urban producers on the plots for a period of two years and left to continueon only (Lado, 1990: 257—266 and Freeman, 1991). The latter their own with only technical advice from the USK. The studies focused on the characteristicsof the producers (i.e. technologies offered were mainly hiointensive including household and/or individual characteristics) and of the organic manure composting and pesticide formulation agriculturalactivities (i.e., types of crops, destinationof the (Personalcommunication, 1993). Crops grown were meant produceand land security, amongothers). A more recentstudy to be mainlyvegetables for consumption and the surplus for in Kibera (a slum settlement in the City of Nairobi) by sale. Dennery (1995) focused on factors affecting the decisions The Korogochoarea providedtwo types of households, and actions of urbanfarmers. Little is known,however, about namely, low incomeurban households practising self-initiated (a) the extent to which urban agriculture as a food source urbanfarming activities (the so-called'Korogocho farmers') contributes to household food security, consumption and and households not practising any urban farming (the nutritionalstatus among low incomeurban populations, and 'Korogocho non-farmers'). Three villages were randomly (b) how the low income urban farmers compare with their selected from a total of sevenvillages and all the households non-farmingcounterparts in this respect. The present article in these villages were listed. It was found that 30 percentof tries to bridge this gap and contributes to the existing the householdscould be classified as urban farmers. From knowledge concerning urbanagriculture in theCity ofNairobi both 'farmers' and 'non-farmers', 70 households were in at leastthree ways: (a) it focuseson low incomehouseholds randomly selected. From the Kitui-Kanuku-Kinyago only; (b) it compares householdsengaged in urban agriculture householdsinvolved in the USUAP, another 70 households with households notperforming any type of agriculture within were randomlyselected (the so-called 'USUAP farmers'). theurban boundaries, and (c)it includes agroup of households During the analysis it appeared that some householdsfrom taking part in a special urban agricultureprogramme. the Korogochofarming group had to he dropped because it

171 - . .2 9: e , \

-I Nakuru \ -I .4

Ngor research area ______1 Kitui-Kinyago-Kanuku 2 Korogocho main road — railway river EMBAKASI divisional name

north

0 5km Mombasa I I

Mapi Nairobi Source: Matrix Development Consultants, 1993 Urban Agriculture, FoodSecurity and Nutritionin Volume 11 Nos. 2 and 3, May andAugust, 1996 Low Income Areasof the City ofNairobi, Kenya was found that some of them had only a few poultry (less Kikuyu formed the largest minority (30 percent). Although than 5) and did not practice any form of urban cultivation. the ability to acquire urban land for farming seems to be Due to this and to unreliable responses, the final number of influencedby the length of stay in the city, even those who households involved were 48 Korogocho farmers, 67 may not have stayed in the city for a long time may acquire Korogochonon-farmers and 62 USUAP farmers. plots through ethnic acquaintances.On certain occasions in Data were collected with the use of an interviewer- the Korogochofields, it was said that if a Kikuyu wanted to administeredquestionnaire and thehead's wife or thefemale stop tilling a certain plot, it would be 'sold' to somebodyof head in the household as the main respondent (the male the same ethnic group as the incomingfarmer. If in any case householdhead wasrequired to respondwhenever necessary). the plot was passed on to somebodyof different ethnicity, The head's wife or female head was selected as respondent sometimes the new individual would be phased out by those because she is usually responsiblefor food production and farmingthe surrounding plots by 'digging into theplot' from preparation; hence,she is thebest respondent especially where all sides. Althoughthis isnot representative of all thefarmers, household food preparationand consumption issues are at it has some bearing as to why mostly Kikuyu are urban stake. farmers. ALuonon-farmer in Korogochocomplained to have been phased out in this manner. Demographic Characteristics Table 1 shows someselected demographic characteristics Household Resources of the three study groups. In general, the non-farmers In all groups,most householdmembers were involvedin households appear to be less far in the 'family life cycle': informal trade and food selling (table 2). This consisted of households are smaller, household heads are younger and street hawking and kiosk and market selling of raw and there aremore young children. Moreover and probably related cooked foods,new and secondhand clothes, other household to this, most of the heads in this group arrived later in the items, collecting and sellingof urbanwaste for recycling(e.g. City of Nairobi than the heads in the two farming groups, wastepaper, empty bottles, old plastics and shoe soles, among one-thirdof them after 1986. This in itselfmay be one of the others)and shoerepair. Among the Korogochogroups, casual reasons for not having access to land: they came at a time labour was the second important activity. Few household when a lot of potential farming land was already occupied members had found employment in the formal sector, while in general they are less 'settled' in the city than those especially in the non-farmers' group (and despite their who stayed longer. As far as the educational level of the comparatively high level of education). Most formally household heads is concerned, the highest level was found employed persons were from the USUAP group, although among the Korogochonon-farmers as the large majorityof no more than one-quarterof the adult population. Informal them had at least upper primary school level of education. manufacturing(carpentry, metalwork and handicraft) and The female heads among the non-farmers were also better illegal trade and practices were more common among the educatedthan the rest since 14 percent of themhad attained non-farmersas compared to the farmers. Illegal trade and post-primaryschool educationcompared to 5 percent among practicesincluded activitieslike manufactureand selling of the urban farming households.Even though the (few) heads alcoholic brews, prostitution,street begging and stealing. with post-secondaryschool education were all found in the Finally, urban farming was also mentioned as a source of USUAP group, one-third of the heads in this group had income, especially in the Korogochofarming households. received no education at all. For the USUAP households, urban farming was quite There appeared to be a distinct ethnic clustering among marginalas an income source. the groups. Most of the inhabitants of the USUAPvillages Using the monetary income generated from activities in are Kikuyu (90 percent)who have been in theCity of Nairobi which householdmembers were involved in 1993, a rough for a longer time than their Korogocho counterparts.The estimation of the averagemonthly householdincome could Korogocho farminggroup consisted for abouthalf of Kikuyu, be made. The kind of income-generating activities the one-thirdof Luo and the rest of other ethnic groups such as household members were practising were not very rewarding. Luhya, Akamba and Somali. Among the Korogochonon- The large majority (85 percent)of the individualsearning a farmers, however, the Luo dominated(60 percent)while the cash incomein 1993 had an averagemonthly incomeof less

Table 1, Demographic Characteristics by StudyGroup Korogocho Korogocho USUAP farmers non-farmers farmers (N=) (48) (67) (62) • average householdsize (no. ofpersons) 6.9 5.6 6.8 • average age ofhousehold head (years) 39 32 45 • no. of hh. members youngerthan 5 years 1.3 1.7 1.0 • % hh. heads born outsideNairobi 90 78 73 • % hh. heads at least 15 years in Nairobi* 63 39 85 •% hh. heads withat least mid-primaryschool 69 85 49 * Only thoseborn outside Nairobi Source:Supplied by the authors

173 AliceMboganie-Mwangi and Dick Foeken African Urban Quarterly than Ksh. 3,000. Only such activities as illegal alcohol findingof Lee-Smithet al. (1987)that 90 percent of the City brewing and selling, trading in used clothes and informal of Nairobi urbanfarmers consumed the crops they produced manufacturingseemed to bemore promising,but few people while only 21 percent sold any part of theirproduce. had an incomeout of that. As a result,the average household Livestock rearingdid not appear to be important. Only a income was also low. However, the groups differed few individuals had some cattle, sheep and goats, poultry substantially in this respect: with about Ksh. 2,700, the and rabbits mainly for home consumption. The major average monthly income in the USUAP group was almost hindrance was the lackof space for keepinganimals. Housing 40 percenthigher than in the Korogochogroups (aboutKsh. units were so squeezed together that there was no space for 2,000). A look at the income distribution shows that the even children to play, leave alone keeping livestock. Ten percentage of householdswith very low incomes, i.e. below years ago, the same pattern was found: 7 percent of the City Ksh. 2,500 permonth was by far the highestamong the non- of Nairobi householdsreared some livestockand the major farmers (table2). constraint was also lack of access to land space (Lee-Smith Since income is usually not reliably estimated, a very et al., 1987). simple welfare index was developed by means of the Despitethe urban agricultureproject launched by theUSK, ownership of certain items which appeared to have a the use of modern inputs among the USUAP farmers was discriminativevalue.' Table2 shows that particularlyamong less commonthan among the Korogochofarmers. There was the non-farmers the percentage of households with a low more use of chemical fertiliser, improved seeds/seedlings, welfare index was very high indeed. improvedbreeds and use of feed supplements in Korogocho. The average size of the urban plot(s) was quite modest Use of natural pesticides was more commonin the USUAP (table 2). The Korogocho farmers cultivated an area of on group. This may be because the USK advocates a blo- average 3,200 square metres (0.4 acres) while the plot size intensive kind of agriculture (i.e. where urban waste is of the USUAPfarmers was less than halfof that. In addition, recycled and used as fertiliser, pesticides and even as many householdsclaimed to have accessto land in the rural seedlings). Another reason may be, as Lee-Smithet al. (1987) areas. This applied particularly to the non-farmers. At the point out, that many urbanplots are too small tojustify use same time, however, for two-thirds of the latter, the rural of modern inputs. In their study, most of the farmers who plot(s) formed neither a food nor an income source. This is used inputs tended to have larger plots than those who did likely related to the fact that most non-farmersoriginated not. from the area around Lake Victoria, i.e. too far from Nairobi The major problem faced by urban farmers was theft to be able to exploit the land. (besides suchproblems as pests and.diseases, lackof capital, and 'plots used as toilets', particularlyin the USUAPareas). Urban FarmingActivities Generaldiscussions revealed that a substantialproportion of All urban farmers in both groups except one in the the maize and beans were harvested and consumed or sold Korogocho grouppractised urban farmingbecause they were before they were fully mature to avoid loss through theft. in need of food. About one third of the farmers indicated that Thus, theft is very importantin that it forces the cultivators they also needed income. This indicates that urban poor to harvest crops with low caloric value. This finding householdsproduced mainly for home consumption. Hence, contrastedwith that of Lee-Smith et al. (1987), where the crops grown were predominantlythe basic foodstuffs like most important problem was that the crops were destroyed maizeand beans, a finding comparable with thatof Lee-Smith by animals.The explanation for this may be that at the time et al. (1987) 10 years earlier. An important vegetable (mid-1980s) Lee-Smithand her colleaguescarried out their concerned kale (sukuma wiki) which is usually consumed study, the economicsituation in Kenya was more stable and together with ugali, a thick paste made from maizemeal. better. Therecent declining economic situationin Kenya may Potatoesand arrowrootalso appearedto be importantcrops. have escalated the problem of povertyand hungerin the urban For all the crops grown, most of the producewas consumed areas such thatthe problem of theft has outgrownthe problem at home; in other words, urban agriculturemay be said to be of animalsdestroying the crops. Eviction,like in Lee-Smith mainly for own consumption.This, again, is in line with the et al.'s (1987)study, did not appear to be a major problem.

Table 2, Household Resources by StudyGroup Korogocho Korogocho USUAP farmers non-farmers farmers (N=) (48) (67) (62) • % persons withformal employment 15 6 24 • % persons doingcasual labour 58 43 19 • % persons doing informal trade & food selling 60 45 86 • % hhs with monthlyincome below Ksh.2,500 31 60 34 • % hhs with low welfare index* 69 90 57 • average size urban plot(s) (sq. metres) 3,200 — 1,400 • % hhs with access to rural land 50 75 40 * Based on the ownershipof certain items (see footnote I and Mboganie 1995: 72) Source:Compiled by theauthors

174 Urban Agriculture, FoodSecurity and Nutrition in Volume 11 Nos. 2 and 3, May and August, 1996 Low Income Areas ofthe City ofNairobi, Kenya

Food Security ingredients differ in their nutrient composition (i.e., water To get a general impressionof the level of food security content, energy content,mineral content and proteincontent as perceived by the three categories of households, several among others), they are converted into nutrient equivalents general questions regarding food availability were asked. for assessment of the quantity of nutrients consumed. In Table3 gives a summary. It is clear that the food situation of addition, household nutrient requirements vary since the USUAPfarming group was generallybetter than that of householdsdiffer in size, sex and age distributionand other the two Korogochogroups. Asked how they coped withfood factors that influence their nutritional needs. Therefore,for shortages, almost half of the USUAP farmers claimed that analysis of survey findings, householdsize is standardised they never experienceddrastic food shortages to warrant a in terms of the numberof consumer units. Thus below, the changeof behaviour. For those USUAPhouseholds who did foods consumed are expressed in kilocalories of energy and face food problems, the most importantstrategy was buying grams of protein per consumerunit per day.2 food on credit implying that they were optimistic about On average, all three groups had inadequate energy intake obtaining some monetary income later on to pay the debts. (table4). Average kilocaloricintake was less than 75 percent The Korogochofarmers employed awide rangeof strategies. of the estimatedrequirements. Average proteinintake, on the The non-farmers, however, had to rely mainly on such other hand, seemed to be adequate for all groups.3 These poverty-indicating strategies as going for donations and results are comparable with thosefound in other urbancentres hawking. in developing countries (Alarconand Rivera, 1994: 171 —182 In all three groups,purchased food formed by far the most and Atuanya, 1987: 109—127). The finding that average important food source (table 3). Nevertheless,one-quarter energy intake is below and average protein intake is above of the Korogochofarmers indicated that their own farming the recommendedlevel should not he a surprise since other activities within the urban boundariescomprised their main studies, especially among the rural poor, have revealed the food source. For the USUAP group, this percentage was same pattern (see for example Foeken and Tellegen, 1992 lower, which can undoubtedly be related to the smallerplots and Hoorweget al., 1991). they have at their disposal. Although energy intake among all three groups was inadequate, differences were observed between the groups Food Consumption in the sense thatthe energy and proteinintakes in the USUAP The actual food consumption in each household was group were higher than in the other two groups.4 And as far recordedfor the whole weekprior to theday ofthe interview. as protein intake is concerned, the proportionof households Sincemost of the interviewingtook place inJuly and August consuming less than therecommended safe level was highest of 1994, the data give a fairly accurate picture of the actual in the non-farming group and lowest in the USUAP food consumption during these two months. Since food households.

Table 3,General Food Security Issues by StudyGroup in Percentages Korogocho Korogocho USUAP farmers non-farmers farmers (N=) (48) (67) (62) • "Always or most ofthe time enough to eat" 35 25 71 • "Do not require any improvement in foodsituation" 13 6 27 • "Go for food donationsor hawking during food shortage" 31 66 24 • "Most importantfood sourceduring past 3 years": - purchased food 67 82 68 - own urban production 25 — 16 Source:Compiled by theauthors

Table 4, EstimatedDaily Energy and Protein Intake Per Consumer Unit by StudyGroup Korogocho Korogocho USUAP farmers non-farmers farmers (N=) (48) (67) (62) • Average household size in consumerUnits (cu) 4.8 3.7 4.3 Energy intake • Average intake (kcal/cu/day) 1,904 1,804 2,151 • % ofrequirements* 64 61 73 • % hhs with intake <75% ofrequirements* 70 67 64 Protein intake • Average intake (grams/cu/day) 62 61 66 • % ofrecommended level* 123 122 132 • % hhs with intake <75% ofrecommended level5 22 30 18 Energy requirementsare estimated at 2.960 kcal/dayper consumer unit.A safe levelof protein intake is estimatedat 50 grams per consumerunit per day (for calculation, see Mboganie1995. Appendix 5) Source: Compiledby theauthors

175 Alice Mboganie-Mwangiand Dick Foeken African Urban Quarterly

For all ingredients consumed during the week under Nutritional Condition of Young Children investigation,the source was asked for, i.e. purchased,from Anthropometric measurementsof childrenaged between own urban farming or from donations by others. The results 6 and 60 months—generally considered as the most are shown in table 5. For all groups, purchased food was by vulnerable group in terms of nutritional condition—were far the most important source of energy. Nevertheless, expressedas weight-for-age (WA), height-for-age(HA) and differences betweenthe studygroups regarding the proportion weight-for-height (WH).5 TheWHO (1983) reference values of purchased energy and proteinconsumed couldbe observed. wereused todetermine the nutritionalstatus of the children. Energy intake among the Korogochofarmers was 100 kcal TheHA valuesof less than 90percent of the reference value perconsumer unit per day higherthan among the non-farmers were generally regarded as evidence of stunting while the despite lower purchases in the farmer group. Since the WH valuesof <80 percentof reference value were regarded absolutelevel of food given to them by others was the same, as evidence of wasting. The WA values of <80 percent of the conclusion seems justified that this higherenergy intake referencevalue were regarded as generallymalnourished and was the result of the Korogochofarmers' own production. those with <60 percent were regarded as severely The same applied to the intake of proteins. In addition, it malnourished. Theresults are shown in table 6. The average was seen earlier that the Korogochofarmers seemed to be WA,WH and HA in all groups were abovethe cut-off points. better off in material ownership although their monetary However, there appearedto be atrend similar to that in food income was about the same. This again could be attributed consumption. Average WA and WH among the Korogocho to the fact that ifpeople produce their own food, they spend non-farmers was lowerand average HA among the USUAP less incomeon foodand use it for other needs. In other words. farmers was higher.6 for the Korogochofarmers, urban agriculture appearsto be The differences between thegroups are more pronounced beneficial in two ways; directly because of a greaterenergy when the proportionsof malnourished, wasted and stunted and proteinintake and indirectly becauseit enables them to children are considered. Usingany of theseindicators showed spend less money on food (at least during a given period of that the nutritionalstatus of the childrenin the non-farming the year since the data collected apply to a relatively short householdswas the worst and in the USUAPhouseholds the period only). best. And althoughthe differencebetween the proportionof The higher energy (and protein) intake in the USUAP underweightchildren among Korogochofarmers and among households compared with Korogocho farmers and non- non-farmers was small, the fact that 5 percentof thechildren farmers cannot be explained in the same way.The absolute among the non-farmerswere severely malnourished gives levels of own-producedenergy and proteins in the former the farmers a better stand. The observed prevalence of group were much lower. It appears that the higher energy malnutrition was within the range found in the earlier and protein consumption among the USUAP farmers was mentionedsurveys conductedin 1990—1992 in low income mainly caused by the fact that they purchased more food, areasof the City ofNairobi. The high level of wastingamong which is obviously related to their higher welfare level. the Korogocho farmers' children was the same as that found in the low incomearea of Kawangwarein 1992. Table 5, Origin ofEnergy Intake by Study Group (kcal/cu/day) Despitethese general observations regardingthe nutritional condition of the young childrenin the three study groups, it Korogocho Korogocho USUAP farmers non-farmers farmers should be stressed that when looking only at the HA data (N=) (48) (67) (62) (i.e. at the long-term nutritional status of the children), the • From own urban production 263 — 110 difference between the Korogocho farmers and the • Given others 102 96 53 non-farmers is HA is the • by Korogocho negligible: average Purchased li2 .L2QZ L91.1 same, while in both groupsthe prevalenceof stunting isvery Total 1,904 1,804 2,151 high. Since the levels of energy intake do not show a big Source:Compiled by the authors. difference eitherand the levels of income are the same, one

Table 6, Nutritional Condition ofChildren Aged 6-60 Months Korogocho Korogocho USUAP farmers non-farmers farmers (N=) (35) (84) (30) Weight-for-age (WA) • average* 85.7 (14.5) 83.7 (14.9) 87.0 (11.2) • % malnourished (WA<80%) 37 42 27 • % severely malnourished (WA<60%) — 4.8 — Weight -for-height(WH) • average* 96.3 (10.6) 94.0 (11.0) 94.1 (7.8) • % wasted (WH<80%) 2.9 8.3 — Height-for-age (HA) •average* 92.5 (5.5) 92.6 (7.8) 95.3 (4.9) • % stunted(HA<90%) 31 38 17 * Standard deviationsin parentheses. Source:Compiled by the authors.

176 Urban Agriculture, FoodSecurity and Nutrition in Volume 11 Nos. 2 and 3, May and August, 1996 Low Income Areas ofthe City ofNairobi, Kenya is inclined to conclude that the advantage the Korogocho It could be shown,however, that theirhigher livelihood level farmershave regardingtheir food supply—in terms of having was not due to their urban farming practices. to buy less food than the Korogocho non-farmers—is not The urban planning policies and laws governing land translated into a better long-term nutritional condition,hut acquisition in Kenya do not take urban farminginto account. in a higher material welfare level. Generally, it is not legal to practice urban farmingaccording to the bylaws, although it is tolerated to some extent. Conclusion However, noting that these urban farmers rely on rainfall, In general, low income householdsin the City of Nairobi have no technical agriculturaland livestockkeeping advice are unable to adequately feed themselves on their earnings. and are faced with problemsof theft of crops, it appears that Hence, many of them try to supplementtheir food supplyby there is more potential in terms of food production through growing subsistence food. However, for many of these urban agriculture than what the results of the present study householdsthis option is out of reach becauseof the lack of may suggest.The Governmentof Kenya policies need to he access to urban land. Unlike the middle and upper income reconsidered in the light of the hardship to the low income groups (who can very easily feed themselves on their urbanhouseholds. Becausethere appearsto he great potential in urban for the of food earnings), the low income groups have no backyards.The agriculture purpose improving and the of malnutrition high densities and overcrowded conditions they live in do availability reducing problem among the urban of 'allotments' which not leavethem with spaceto cultivate and/orto keep livestock. poor populations, policies have been used and succeeded in countries and They are, therefore, more often found farmingvacant public European Malawi in urban could he reviewed or private land without any form of security. The plots are encouraging agriculture and modified for in This would assure too small, however, and the farmersmay have to move from adoption Kenya. rights like to extension services kr and one plotto anotheronce therightful owners decide todevelop agricultural high yields them. adviceo: whatto grow where. Of course there is the issue of of the for human hut this can he The study indicatesthat the length of stay in the city plays safety produce consumption controlled the Governmentof more control a role as far as purchasing power and food security of the by Kenyataking overuse ofurban waste, environmental and what low income households are concerned. The longer the management land shouldbe used for food production in collaborationwith households have been in the city, the more possibly they thecommunities and any otherorganisations. However, there establish ways and means of obtainingincome and food for is need for further research into the qualityand improvement survival. For instance, they may acquireformal employment, of urban food and livestockproduction. land to cultivate and may establish reasonable incomeactivities. It is that educationseems generating conspicuous Notes to no role inthis since thebest-educated play (anymore) group The items used were radio, sofa set and Households was also the with the lowest bicycle. (the non-farmers) group received a score of I for the ownership of any of these items percentage of persons in formal employment and with no and a score of zero for not owning the item. The scores were access to urban land. summed up so that the lowest score was zero and the highest Compared with the non-farmers, the Korogocho farming score was 3. The distributionof the householdsis shown in households weresomewhat better off interms of food security Mboganie (1995:72). Households scoring below 2 were and, be it to a lesserextent, of nutritional status. Their urban consideredto have a low welfare index and to be 'very poor'. activities contributed to this in the sense that the farming 2 self-produced food wasmainly meantfor home consumption. One consumerunit (cu) is theconsumption equivalent (here: in termsof ofa nominal adult male. The In addition,it enabledthese householdsto spend less money requiredenergy) required of all household members is derived from this and on food and more on other needs, resulting energy purchases daily on such factors as status and better conditions than the non- depends age, sex, physiological in generally living among level (see 1995: Data farmers. In in the somewhat physical activity Mboganie, Appendix 2). short, explaining higher on all foods consumed,harvested, purchased and received as livelihoodlevel of the Korogochofarmers as compared with gifts were translated into kilogramsof edible portions and into the Korogocho non-farmers, urbanagriculture certainly plays kilocaloriesand gramsof proteinper consumerunit, using Food a role. Composition Tables for Foods Commonly Eaten in East Africa. The householdspractising urban agriculture with assistance (Nairobi:Clive E. West, 1987)Technical Centre for Agriculture from USUAPhad ahigher level of food securityand abetter and Rural Developmentand Food and Nutrition Cooperation nutritional condition than those farming on their own Programmeof East, Central and SouthernAfrica (1987). initiative. They appearedto have a more diversifiedkind of - To ensurethat the results were not influenced income farming and sold more of their produce. However, due to by outliers, their small their level of food was two additional analysesof energyand protein intake, as well as very plots production quite condition low. from food these households also nutritional (see below) were done; one by using only Apart production, the households with a monthlyincome of less than Ksh. 2,500 benefitted from other activities as well income-generating and one with only the households with a low welfare index(for as the shelter that came with the improvementproject along details, see Mboganie 1995: Appendix 4). In both cases, the urban agricultureproject. Hence, theirliving conditions were resultsshowed the same trends. Hence,the original resultsare betterand they had a higher purchasingpower compared to presented here. thenon-farmers and the farmersnot involvedin any project.

177 Alice Mboganie-Mwangiand Dick Foeken AfricanUrban Quarterly

It has been shownthat there is a negativerelationship between Foeken, D. and N. Tellegen (1992). Household Resources householdsize and energy intake per consumer unit, i.e. the and Nutrition of Farm Labourers in Trans Nzoia District, larger the household the lower the energy intake (see e.g. Kenya. Nairobi / Leiden: Food & Nutrition Studies Foeken and Niemeijer, Klaver, 1991:35—36). Ifthe figures in Programme, Report No. 44. Ministry of and table4 wouldbe corrected for household the Planning size,however, major National Development /African StudiesCentre. result would be that the energy intake in the non-farmers householdswould be even lower. Freeman,D. (1991).A Cityof Farmers: InformalAgriculture Height-for-age (HA) expresses the height of a child as a in the Open Spaces of Nairobi, Kenya. Toronto: McGill percentageof the correspondingmedian height of children of UniversityPress. the same age in the reference population. The HA values of less than 90 percent are generally regarded as evidence of Gathuru,P.K. (1988).Urban Agriculture Project: Hopesand stunting,indicating that the child has failedto grow satisfactorily Challenges.Paper Presentedat Workshop on "Food for the duringlengthy periodsin the past. Therefore,HA is commonly Future" held in Eldoret, Kenya. regarded as an indicator of nutritionalhistory reflectingsocial and economic conditions. Weight-for-height (WH) expresses Hoorweg, J. et al. (1991). Economic' and Nutritional the a childas the median weightof apercentage of corresponding Conditions atSettlement Schemes in CoastProvince. Nairobi/ of of the similar in the reference weight children height Leiden: Food & Nutrition Studies No. population. The Weight-for-height(WH) values of below 80/ Programme, Report 85 can be as evidence of 36. Ministryof Planning and NationalDevelopment/African percent regarded wasting,indicating StudiesCentre. acutemalnutrition. In this report, value of 80 percent is used. Weight-for-height (WH) is an indicator of present nutritional status. The weightof a child can alsobe expressed in terms of Kenya, Republic of (1991). Fourth Rural Child Nutrition Weight-for-age(WA), often used as a 'short-cutmeasure' Survey 1987. Nairobi: Ministry of Planning and National becauseit reflectsboth previousgrowth and present nutritional Development, Central Bureau of Statistics. condition. It is used for a broad classificion of malnutrition. Children with less than 60 percent of the referenceweight for Kenya, Governmentof, and UNICEF (1990). Metropolitan their age are generallyregarded as severely malnourished while HouseholdSurvey, 1989. Nairobi: GovernmentPrinter. those with a WA between 60 percent and 80 percent as malnourished. Kenya, Governmentof, and UNICEF (1992). Childrenand A 1992. Nairobi: 6 the are influenced the distributionof Women in Kenya: Situation Analysis Usually averages by age GovernmentPrinter. the children. Children aged 1—2 years usually tend to have a poor nutritionalstatus comparedto the rest due to theeffect of weaning. For purposesof verification, the results were corrected Lado, C. (1990). "Informal Urban Agriculture in Nairobi, for age distributionin each group but the trend was even more Kenya: Problemor Resourcein Developmentand Land Use magnified. The originalresults are thereforepresented. Planning?"Land Use Policy (July).

References Lee-Smith,D. etal. (1987).Urban FoodProduction and the Alarcon,J.A. andJ. Rivera (1994). "The Impact of Changes CookingFuel Situation in Urban Kenya.Nairobi: Mazingira in Purchasing Power on Food Consumptionon an Urban Institute. Population in Guatemala City", Ecology of Food and NutritionVol. 31 Nos. 3—4. MatrixDevelopment Consultants (1993). Nairobi's Informal Settlements. An Inventory ReportPN-ABH-741 prepared for Atuanya, E.I. (1987). "Nutrition as an Indicator of Urban USAID/REDSO/ESA. Poverty,A Dietary Survey of the Urban Poor in Benin City, Nigeria",in P.K. Mwakinwaand A.O. Ozo (eds) The Urban Maxwell, D.G. and S. Zziwa (1992a). "UrbanAgriculture in Poorin Nigeria. Ibadan: Evans Brothers(Nigeria Publishers) Kampala: IndigenousAdaptive Response to the Economic Limited. Crisis", Ecologyof Food and NutritionVol. 29 No. 2. An Dennery, R.P. (1995). Inside Urban Agriculture: Maxwell, D. and S. Zziwa(1992b). UrbanFarming inAfrica: Pmducer Decision Nairobi Explorationof Food Makingin a The Case ofKampala, Uganda. Nairobi:ACTS Press. Low Income. Wageningen: UnpublishedThesis. Agricultural University. Mboganie Mwangi, A. (1995).The Roleof UrbanAgriculture FoodSecurity in Low Income Areas inNairobi. Nairobil Basic for Drakakis-Smith,D. (1992). "Strategies for Meeting Leiden: Food and Nutrition StudiesProgramme, Report No. Food Needs in inJ. Baker andP. Pedersen Harare", 0. (eds) 54. Ministryof Planning and NationalDevelopment/African The Rural-Urban Interface in Africa: Expansion and Studies Centre. Adaptation.Copenhagen: Seminar ProceedingsNo. 27. The Scandinavian Institute of African Studies and Centre for Memon, P.A. and D. Lee-Smith(1993). "Urban Agriculture Research. Development in Kenya". CanadianJournal of African StudiesVol. 27 No. 1.

178 Urban Agriculture, FoodSecurity and Nutrition in Volume 11 Nos. 2 and 3, May andAugust, 1996 Low Income Areas ofthe City ofNairobi, Kenya

MioziM.R.S. et al. (1992)."Urban Agriculture as a Survival Sanyal,B. (1987). UrbanCultivation in EastAfrica: People's Strategy in Tanzania", in J. Baker and P.O Pedersen (eds) Response to Urban Povert. Cambridge: Massachusetts The Rural-Urban Interface in Africa: Expansion Instituteof Technology. Adaptation.Copenhagen: Seminar Proceedings No. 27. The Scandinavian Institute of African Studies and Centre for Sawio,C.J. (1993). "Urban Agriculture:Eastern and Central DevelopmentResearch. Africa", in L.J.A. Mougeot and D. Masse (eds) Urban Environment Management;Developing a Global Research Mutisya, D.N. and C. Lado (1991). "Some Socioeconomic Agenda,Vol. I. Ottawa: International Development Research Factors Behind Roadside Farming in Kiambu District, Centre. Kenya", Journal of Eastern African Research and Development Vol. 21. Wade, 1. (1987). "Community Food Productionin Cities of Developing Nations", Foodand Nutrition Bulletin Vol.9 No. Niemeijer, R.D. et al. (1991). Seasonality in the Coastal 2. Lowlands of Kenya, Part 4/5: Food Consumption and Anthropometry. Nairobi/Leiden: Food & Nutrition Studies West,E.C. (1987). Food Composition Table for Energy and Programme, Report No. 38. Ministry of Planning and Eight Nutrients in Foods Commonlv Eaten in East Africa. National DevelopmentlAfricanStudies Centre. Nairobi: Technical Centre for Agriculture and Rural Development and Food and Nutrition Cooperation Rakodi, C. (1985). "Self-Reliance or Survival? Food Programmeof East, Central and Southern Africa. Production in African Cities, with Particular Reference to Zambia",African Urban StudiesVol. 21. WHO (1983). Measuring Change in Nutritional Status: Guidelines for Assessing the Nutritional Impact of Rakodi, C. (1988). "Urban Agriculture:Research Question Supplementary FeedingProgrammes for Vulnerable Groups. and the Zambian Evidence", Journal of Modern African Geneva: World Health Organisation. StudiesVol. 26. World Vision International (1 990). Field Experience from Sachs,I. and D. Silk (1990).Food and Energy: Strategiesof Korogocho Project. Nairobi: Unpublished Report. SustainableDevelopment. Tokyo: United NationsUniversity Press. Yeung, Y. (1987). "Examples of UrbanAgriculture in Asia", Food and Nutrition Bulletin Vol. 9 No. 2.

179 OXFORD'S NEWJOURNAL FOR THE SOCIAL PROFESSIONS European Journal of Social Work Welfare, Social Services and Social Action

EDITORS: Prof. (-lans-Uwe Otto, University of Bielefeld,Germany Prof.Walter Lorenz, University College,Cork, Ireland A social work journal is being launched in 199 by Oxford University Press that will provide a new forum for academic debate in the social professions. It will analyse and promote European and international developments in social policy, social service :nstitutions and strategies of social change by publishing refereed papers on contemporary key issues. Articlesin Volume 1, Issue 1 will include:- The Politicsof Care for the Elderlyin Scandinavia, Gunborg /acobsson, Finland Issues in European Child Protection Research, Rachel Hetherington, UK From Service to Social Control: Implications ofWelfare Reform for Professional Practice in the US, Neil Gilbert, Berkeley, USA The Flightfrom Universalism. The Caseof Swedish Social Policy, Staffan Blomberg, Lars Hárrisson, Jan Magnusson, Jan Peterson, Sune Sunesson, Lund, Sweden The Professional Role of Social Workin the Market, Geert van der Laan, Utrecht, TheNetherlands Between Interest in Practiceand Power of Definition. On the Development of Social Work in Spain, AndreuLópez-Blasco, Spain CALL FOR PAPERS Manuscript ii )U Prof. Hans-Uwe Li , I D-33501 SielE ?l , (

INTRODUCTORYOFFER OPEN UNTIL 1 DECEMBER1997! Volume 1 1998 (3 issues) ISSN: 1369-1457 Personal subscribers: Order before 1 December 1997and save £10/USS17! Institutions: U £80/US$150 Individuals: U £30/USS5O Students: U £20/US$33 Individuals: U £201USS33(if orderreceived by 1 December 1997) £ sterling rates applyin Europe, 115$ rates elsewhere *please send proof of studentID U I wish to subscribeto European Journalof Soda! Workas indicated above. U I would liketo ordera sample copyof Issue1 whichwill be available in early 1998 Name ______Address ______

Post code/Zip Country Please return to: loumals Marketing (EJSW3), Oxford University Press, Great Clarendon Street, Oxford0X2 6DP, UK Tel: -*44 (0) 1865267907, Fax: +44 (0)1865 267485 E-mail: [email protected] AfricanUrban Quarterly, 1996: 11(2 and 3) pp. 181—1 87 ©1996 by African Urban QuarterlyLtd.

URBAN AGRICULTURE IN AFRICA: A COMPARATIVEANALYSIS OF FINDINGS FROM ZIMBABWE, KENYAAND ZAMBIA

Daniel S. Tevera Departmentof Geography, Universityof Zimbabwe,P0 Box MP 167, Mount Pleasant, Harare,Zimbabwe Tel: 263-4-303-211; Fax: 263-4-353-407

Abstract Accepted May, 1996 With rapid urban growth rates, a diminishingability of many countries tofeed the increasing national populations, persistent droughts and escalatingfood prices, urban agriculture is increasinglybecoming afood security strategy both at the national andhousehold levels. Accordingto Freeman (1993:1)urban agricultural activities in ThirdWorld urban centres "have reached the point where their importanceas apart ofthe urban informalsector can no longer be overlooked." The extent, nature and role of urban agriculture in Africa vary considerablywithin andbetween countries as well as throughout the urban hierarchy. The main sources ofurban food are the urban house-gardens(on-plot cultivation)and the stretches of open land(off-plot cultivation) within residential and industrial areas and theperipheral areas where illegal'cultivation is quite common. Howeve, thephysical growth of most urban centres has curtailed theamount of land availablefor subsistence food production. Given the low and erratic incomes ofmany urbanfamilies in Africa, numerous householdsare confronted with severe nutritional problems that are frequently exacerbatedby distribution and marketingsystem inefficiencies. It is against this backgroundthat urban agriculture shouldbe considered if realistic solutionsare to befound. Urban agriculture refersto the cultivationof crops at both the subsistenceand commercial levels and keepingof small livestockin open spaces in urban areas. Howeve,; due to the research interestsof academicsand donor agencies the term is commonly usedto refer to off-plotcultivation which is largely considered 'illegal'. Although much activity occurs within house-gardens(where they exist) urbanagriculture has beenfostered by the availabilityof unused open spaces (Mosha, 1991 andMaxwell and Zziwa, 1992) and it utilises resources in urban ecosystems which would otherwise be unutilised (Smit and Nasi 1992: 141—152). Urban agriculture is a significanturban activity that relates to other processesin the urban centre, such as the use ofwaste water to irrigate crops. Links exist between urban agriculture activities, household consumption, petty-trade andformal recycling in theform offlows ofwaste material (e.g. the use of sewageeffluent forcattle pastures on municipalfarmsin the City ofHarare in Zimbabwe helps the city councilearn several millions ofZimbabwe Dollars annually throughcattle sales. Food is the most crucial item in the expenditure budget of the households. On the average, 50—60 percent of the total household income is spent onfood in Third World countries. Sanyal (1986 and 1987: 197—207), for instance, has indicated that poor householdsin Tanzanian urbancentres produce about one-thirdof nutritional needsin this manner while inKenya, Lamba (1993: 83—97) has observed that more than 50 percent of the urban households' expenditure is spent on food. In Zimbabwe, official surveys show that low-income urban householdsspend over half of their budgets on food items alone while the high-incomeurban householdsspend only 20percent of their budget onfood. The need to supplementinadequate urban diets compels a substantial segment of urban dwellers to engage in urban agriculture. However Drakakis-Smith (1992: 258—283)noted that there is need to differentiatebetween urbanfood productionfor self-consumption andfood producedfor sale. Thepurpose of this paper is to examine the importanceof urban agriculture in Africa by analysing the majorfindings ofthe major empirical studieson urban agriculture inAfrica, with special reference to Zimbabwe, Kenya and Zambia. The three African countries are very differentwith respect to the degree of urbanisation and the extent of urban agriculture. Howeverthese countriesshare abroadly similar settlercolonial heritage,space-economy andphysical planning traditions which have always regulated urban land use. In all the three countries,towns are planned and local authorities have a wide range ofpowers to control and shape the built environment. Résumé La culture d'espaces libres ou clairières non-construitesdans Ia trame urbaine retient de plus en plus l'attention du public enAfrique. Dans unpremier temps, 1'auteur note que la resurgence récentede I 'agricultureurbaine etses interrelations avec d'autres phénoménes urbaines restent negligee par leschercheurs africains,comparativement aI 'étuded 'autresquestions. La littérature disponiblesur lesujet a néanmoins quantfie l'ampleur et la repartition spatiales des activitésagricôles, trace le rofil socioeconomiquedes producteurs et 1'apport de leurproduction a 1'économie urbaine; elle en a examine les aspects opérationnels et a moindre titre, ses effets environnementaux sur Ia ville et son hinterland. Dans une seconde partie, une lecture comparée des principales etudes réalisées au Zimbabwe, au Kenya et en Zambie porte I'auteur aux conclusions suivantes: 1'agriculture urbaine dans ces pays n 'est definitivement pas transitoire mais croissante, laquelle donne lieu a des formes urbaines qui s'éloignent des idéaux occidentaux; lesplus pauvres béneficient peu de 1'activité; en général celle-ciest encorepeu endosséepar les autorites locales. Toutefois, I 'auteur signalle une tendancerécente vers des positions officiellesplus condescendantes,1 'incorporation de I 'activitéa I 'aménagementde nouveauxcentres urbains et le relâchementdes contrôles dans les villes existentes.

181 Daniel S. Tevera AfricanUrban Quarterly

Studies on Urban Agriculture in Africa and the diffusionof vector-borne diseases such as malaria. It is the contentionof this paper that urban agriculture is Similarly, there isneed for researchon the interfacebetween a phenomenon in Africa which has largely escaped the urbanagriculture, pollution and the degradationof the urban attention of urban geographersand academic researchers in environment. Also,Mougeot (1994: 1)has lamentedthe lack general. Clusters of research exist on urban agriculture in of "longitudinal and comparative analysesbetween farming Africaand its spatial dimensionsbut the availableliterature and nonfarming households, on nutritional statusof theurban is rather limitedand it covers only patches of an interrelated poor and their strategies for coping with food insecurity,as network of social, economic and politicalphenomena. Thus well as the currentand potentialimpact of urbanagriculture as Drakakis-Smith(1992: 258—283)has observedthere has on households and their members more exposed to been less writtenabout urbanfood supplysystems than other malnutrition risks (women, children,and the elderly)." basic needs such as shelter,despite the fact that food is the A final group of studies such as by Mazambani, 1982a: most basic of needs and that shortages are more likely to 134—138; Maxwelland Zziwa, 1992 and ENDA-Zimbabwe, triggerurban instability than inadequate provision of housing 1994 has been concernedwith the environmentaleffects of or education. urban agriculture in urban centres and their inlands. Watts Theexisting literatureon urban agriculture inAfrica falls and Bransby-Williams 1978: 101—102 made some under four basic and sometimes overlapping approaches. preliminary observations on the link between urban First, a relatively large number of studies have focused on agriculture and malaria. the identification of the spatial extent and location of Unfortunately, none of the empirical analyses on urban agricultural activities in urban areas (such as Mazambani, agriculture inAfrica, excluding Rogerson's (1992: 229—234) 1982a and 1982b: 134—138; Lado, 1990; Mbiba, 1995). These most illuminatingconceptual analyses,has tried to examine studies have been largely of baseline nature, characterised the phenomenonacross national boundaries. There is need by extensive descriptions,outlining the incidenceof urban for cross-country studies on urban agriculture to better agriculture, and informing policymakers of the positive conceptualisethe phenomenon and to make better informed features of urban agriculture. policy recommendations. Also,there is need forresearch and A secondgroup of studiesfocuses on the socioeconomic policy dialogue on the potential contribution of urban profilesof theurban cultivators and theeconomic importance agriculture to poverty alleviation and sustainable resource of urban agriculture (such as Mazingira Institute, 1985; utilisationin African urban areas. Rakodi, 1985: 53—63, 1988a: 495—515, 1988b; Sanyal, 1986, 1987; Freeman, 1991, 1993). These studies have addressed Survey of the EmpiricalStudies issues such as the spatial distribution, the activities and In this section some of the empirical surveys on urban motives of the cultivators,the financial and economic value agricultureconducted in Zimbabwe, Kenyaand Zambia will of urban agriculture and its implications on self-sufficiency bereviewed. Several factorsexplain why these three countries at the household level. Also, they have highlighted the were selected for this study. First, because these countries importanceof urbanagriculture as a socioeconomic survival have experienced amassive growth in thescale of the activity and livelihood-enhancing strategy for the urban poor. The during the past decade despite the existence of limiting study by Freeman (1993)focused on the significance of urban policiesand regulations.Second, because extensiveresearch agriculturefor the advancementof womenin Africanurban has been undertaken on urban agriculture over at least ten centres. He made the observation that female cultivators years and there is ample data for analysis for comparable encountered tremendous impedimentswhich limited their times. Third, in many ways the urban cultivationpractices advancementin the urban areas. and spatial patternsmirror those prevailingelsewhere in the A third group of studies suchas the onesby Sanyal, 1986; East and SouthernAfrica region, althoughthe scale seems to Lado, 1990: 257—266; Mosha, 1991: 83—92; Maxwelland betoo high. This suggests thatin the near futurethe prevailing Zziwa, 1992; Drakakis-Smith et al., 1995: 183—193; has urban agriculture situation in these countries is likely to be explored the operational dimensions of urban agriculture repeatedelsewhere in the region hence there is need for the within the context of urban economies characterised by othercountries to learnfrom the urban agriculture experiences several competing land uses and intertwining linkages. of Zimbabwe, Kenya and Zambia. However, the rural-urban linkages (e.g. rural-urban food flows,etc.) and how they influenceurban economic activity, Zimbabwe especially urban agriculture, have not been properly Zimbabwe has apopulation of 10.4 million and 36percent contextualised.As Wekwete (1993: 98—110) has observed, of which are accommodatedin urban areas which have been there is need to contextualise the phenomenon of urban growing at a rapid rate of about 5.4 percent per annum agriculture within the broad context of the urban economy between 1982 and 1992. Urban poverty is widely and urbanmanagement and toplan for the activity by making acknowledged and authoritiesconservatively estimate that it one of the main urban land-useactivities. This needsto be between 10—15 percent of the urban population live in seriouslyconsidered given the fact that withinAfrican urban absolute poverty in squatter settlements (on vacant public centres there are lots ofunutilised open spaces.Also, there is land) and backyard shelter (in the high-density suburbs). need to examine the linkages between urban agriculture, Urbanpoverty has been exacerbatedby lack of employment waste management and health. Comprehensive studies are generationin the formal sector during the past 10 years In required on the suggested links between urban agriculture 1992over 1.3 million peoplewere formallyconsidered to be

182 Urban Agriculture in Africa: Volume 11 Nos. 2 and 3, May andAugust, 1996 A ComparativeAnalysis of Findings from Zimbabwe,Kenya and Zambia

without jobs and the rate of unemployment officially More recently, a survey by Drakakis-Smith,Bowyer-Bower estimatedat 25 percent. and Tevera (1995) examined the linkages between urban Several factors have contributed to the rise of urban povertyand the increasingincidence of urban agriculturein agriculture inthe country. First,persistent droughts since 1982 Harare. which have reduced food flowsfrom therural to urbanareas The above group of studies produced a number of and in some cases there have been reverse flows. In Harare, interesting findings. First, that 80 percentof all house-gardens the capital, at least a quarter of the low-incomehouseholds in Harare are used to produce food mainly for subsistence receive food from their family homes in the rural areas. (Drakakis-Smith and KiveIl, 1990: 156—180) and that Second, the international recessionof the mid-1980s (coupled householdmembers who engaged in urban food production by thedrought) has been blamedfor thelow rate of economic undertook severalother roles in relation both to production and employment growth and many families have been and reproduction. Second, that in three highly differentiated compelled to boost their householdincomes by cultivating residential suburbs of Harare (i.e. Glen View high density vegetables for the market. Third, the implementation of a area, Mabelreignmedium density area and Epworth, which structural adjustment programme in 1991 has resulted in is an upgraded former squatter settlement,about 80 percent heavy social costs as, among other things, prices have risen of the interviewees grew crops in their home-gardens followingthe removal offood and fuel subsidies and as many primarily for householdconsumption (Drakakis-Smith and workers have been retrenched.The progressiveremoval of Kivell, 1990: 156—180).The study noted that inEpworth there subsidies and deregulation of prices of basic commodities was a greater tendency for families to keep small livestock like foodstuffshave been inflationary. During the 1991 and such as chickens and goats than in the other residential 1992 period food prices for lower-incomeurban households suburbs.There is ample documentationof the various foods rose by an incredible 534 percent. The price of subsidised farmers grow and the animals they rear where they live in plain maizemealrose by over 100 percent betweenJanuary the urban and pen-urban areas. 1991 and January 1994 while the price of bread rose by Third, that undernutritioncharacterises the lives of the between 100—175 percent during the same period (Tevera, low-income householdsin Harare despite the fact that most 1995: 79—80). Withinthe urban areas, however, the highfood spendover halfof their incomeon food (Drakakis-Smith and prices have forced more households to produce their own Tevera, 1993). Fourth, that extensive illegal cultivation food. regularly covers public open spaces during the rain season In Zimbabwefour studieson urbanagriculture and urban from Novemberto March and that open spaces close to the food supplies are worthnoting. Thepioneering studyon urban residential areas were extensively cultivated. Off-plot cultivationand woodfuelcollection in Harareby Mazambani cultivation is generally considered to be illegal by the (1982a and 1982b: 134—138) examined the activity in the authoritiesif it either occurs within 30 metres from water- city from the perspectivethat it was transformingan urban courses or is undertaken without written approval of the environmentinto arural landscape. Centralto this study were authorities. Fifth, that virtually all urban food production three important observations.First, that agriculture within outside house-gardens is considered illegal (with the urban and pen-urban areas is a widespreadactivity and in exception of the fields that are cultivated by cooperatives most cases is for subsistencepurposes and involveslimited with the city-fathers' blessing) and attracts a generally capital inputs. In Harare, urban agriculture is extensive in negative response from the authorities. Sixth, that urban private gardens (especially in the high-income/low-density farmers receive minimum support, if any, from central areas) and in public open spaces such as undeveloped governmentor the local authority. Generally,there are no industrialand residential areas and road or railway reserves. urban extensionservices for crop production and livestock Second, on the basis of analyses of aerial photographs rearing. Similarly, credit facilities for urban farming are Mazambani (1982b: 134—138) established that thecultivation almostnon-existent hencethe current low investment in urban of open-spaces in Harare had increased from about 267 agriculture. Seventh, that the findings reaffirmed hectares to 4,762 hectares between 1955 and 1980 and that Mazambani's(1982a) results that maize, sweetpotatoes and the cultivation was mostly occurring in vleis (wetlands). beans were the leadingcrops cultivated.Eighth, that urban Third, that cultivation was mainly for subsistencepurposes agriculture was not confined to the poor, but that there was and that maize, sweetpotatoes and vegetableswere themost incredibleheterogeneity among thecultivators. A fairly large popular crops cultivated. numberof middle-incomehouseholds are involved and these Researchon urban foodsystems and urbanagriculture in oftenhave betteraccess to land, labour and finance. The group Harare, conductedby several teams of geographersbetween included many of the urban poor but not the poorest of the 1990 and 1994, has generated a number of scholarly urbanpoor because the latter lacked securityof tenure in the publications (Drakakis-Smithand Kivell, 1990: 156—180; city and had verylimited access to land. Finally, that conflicts Drakakis-Smithand Tevera, 1993;Drakakis-Smith, Bowyer- existed between the urban poor and the authoritiesbecause Bower and Tevera, 1995: 183—193). The 1989 survey by of the latter's strictimplementation ofpolicies restricting the Drakakis-Smithand Kivell investigatedthe natureand origin cultivation of municipal land (Drakakis-Smith, Bowyer- of foodproduction activities withinHarare and itsperiphery. Bower and Tevera, 1995).However, it was noted that more A follow-up survey by Drakakis-Smith and Tevera recently officials had been more accommodativeof urban investigated the basic characteristics of the urban food agriculture as was reflected by the general moratorium on distributionsystem and its links with the urban household. maize slashingsince the 1992—1993 season.

183 Daniel S. Tevera African Urban Quarterly

The research by Mbiba (1994: 188—202 and 1995: 75— bananas, flowers and the raising of poultry, pigs, goats and 86) who classified urban agriculture by typeof product (arable cattle. Second, that only 31 percent of the farmers had legal or non-arabic) by spaces used (on-plot or off-plot) and by accessto the urbanland that they cultivatedand that the low- institutional settings (legalor illegal), addressed the issueof incomehouseholds tendedto cultivate municipal land which the role of urban agriculture in Harare and the legislative was in transitional use or in peripheralareas where there was and policy issues. The study involved the use of structured little crop security (MazingiraInstitute, 1985). Third, that a questionnairesto interview 1,200 householdsand data were significantportion of thepoorest households in Nairobispend collected on the number of cultivators, typesof crops grown in excess of their total income on food through borrowing. and amountof spaceutilised. The following findings emerged Fourth, that agricultural practices are very basic and from the study. First, that the bulk of urban agriculture in dependent on hand labour with only a few simple and Harare wasarabIc and livestockrearing wasnot as extensive inexpensivetools and there is extensiveuse of organicinputs as elsewhere in Africa. Second,that themost commoncrops in Nairobi and technical advice and agricultural extension cultivated on off-plot fields were maize, beans, pumpkins, servicesare lacking. Fifth, that 62 percentof the households sweet potatoes and that fertilisers were routinely applied. surveyedwere engagedin urban cultivationmainly for their Third, that the size of the plots varied from 400 to 9,000 subsistence needs and that 56 percent ofthe urban farmers square metres and that, on the average, the cultivatorshad were women. Sixth, that as many as 1 7 percent of the cultivated the same plots for about 3.6 years. Fourth,that 95 households kept livestock, especially poultry and cattle, percentof the cultivatorslived withina radiusof 5 kilometres within the city. Seventh, that 63 percent were involved in from theiroff-plot fieldand that 63 percentof the cultivators out-of-townagricultural activitiesmainly because of lack of were women. Finally, that 88percent of the cultivators tended accessto suitable land close to theirresidential areas. Eighth, their plots while 12 percent employed contract labourersto that about 25.2 million kilogrammesof crops, worth about workon thefields. The last finding clearly suggests that some US $4 million were producedin thecountry's urbanareas in of the householdsinvolved in off-plotcultivation in the city a single season. are not low-income families and that they probably grow The major recommendation that emerged from the crops mainly for the market. MazingiraInstitute study was that there was need to support The study by ENDA-Zimbabwe (1994) involving a localauthorities, NGOsand Community-Based Organisations sample of over 1,700 cultivators, examined several aspects (CBS) in devising creative and flexible solutions for of urban agriculture in Harare. Central to this work were improvingaccess to land. several major observations. First, that the area under The MazingiraInstitute study is thefirst large-scale study cultivationhad increased by 92.6 percent between 1990 and on urbanagriculture conducted in Kenya and it did shed some 1994 and that the spatial area under cultivationin the city light on the complex phenomenon. It also paved the way had increased from 8.8 percent to 16.7 percent during the forward for several studies using similar but more rigorous same period. A related finding was that most of the increase research methodologies. had occurred in the high density areas. Second, that urban Asimilar picture of thefindings emerged from Freeman's agriculture is pursued by homeowners (61 percent of the (1991) study of 618 urban cultivators that were mainly total), lodgers (tenants) (27 percent), servants(4.5 percent) engaged in off-plot cultivationin Nairobi (Freeman, 1991). and relatives of homeowners(7.4 percent). However, about The study observed that urbanagriculture, mainly dominated 90 percent of the urban cultivatorsstayed in the high density by women who had been in the city for long periods, was areas while 7.5 percent and 2.8 percent stayed in the low widespread throughout the city. The results show that 28 density and medium density areas respectively. Third, that percent of the urban farmershad primary education while 6 62 percent of the urban cultivators were unemployedand percenthad secondary education. It is evidentfrom the study were mostly women. Fourth, that maize was cultivated on that most of the urban farmersheld urbanjobs and that their 50 percent of the area under cultivation, followedby sweet involvement in farming activities reflected a multiple potatoesand groundnutswhich took 19percent and 8 percent livelihood strategy intended to supplement the low wage of the space respectively. Fifth, that 88.4 percent of the incomes. Freeman(1991) classified the urbancultivators into cultivatorsengaged in off-plotcultivation and usedchemical four categories: backyard farmers who cultivated private fertilisers. residential land;riverside farmerswho cultivated land on river flood plains; farmerswho cultivated on roadside verges and Kenya squatterfarmers who cultivated open spacesin industrial and In the Kenyan context, two studies focusing on on-plot residential areas, land bordering the railway tracks and and off-plot cultivation, respectively, are worth noting. The parkland. study by the Mazingira Institute (1985) is perhaps the first The study observed that urban farming in Nairobi was comprehensivestudy on urban agriculture in Kenya. The constrainedby environmental factors (such as drought, land study was undertaken in 1985 and itcovered six urban areas degradation,flooding), crop loss due to pests and by zoning from which a sample of 1,576 predominantly on-plot laws which limitedaccess to land. It also stressed the need to cultivatorswas randomlydrawn. plan for urban agriculture because it had the potential to Centralto this work were atleast four major observations. contribute substantially tofood securityin thecities in Kenya. First, that urban agriculture in Kenya was quite diverse and Freeman's study on off-plot urban cultivation it included the production of vegetables, fruits, grains, complemented the earlier on-plot survey undertaken by the

184 UrbanAgriculture in Africa: Volume 11 Nos. 2 and3, May andAugust, 1996 A ComparativeAnalysis ofFindings from Zimbabwe,Kenya and Zambia

MazingiraInstitute. However,it is interestingto note that in of urban developmentaway from the colonial ideal of the both cases the studies observed that cultivation was showcasecapital, and from the notion that cities comprise predominantly for subsistencepurposes and that, despitethe the exclusivedomain of the governing and businesselites." accommodating position assumed by the various local Second, the real poor do not seem to benefit much from authorities in Kenya, urban agriculture was regarded with urban agriculture because their access to land is too lackof sympathy and this explainsthe general lack of support circumscribed and where they have access they oftencultivate to urban cultivators, for example in the form of extension tiny fields that are incapableof producing yieldshuge enough servicesand loans for inputs. to economically emancipate them. Third, this strategy is not More recently, Lamba (1993:83—97)observed that proper confined to the poor, hut includes most middle-income urban agriculture can help reduce water, waste and hazard householdswho often have better access to land, labourand problems and increase returns on water bodies in the large finance. towns of Kenya. He also noted that cultivation on home- Fourth, it is an economic activity which in the majority gardens provided a possibility to reuse domestic water and of cases does not have the support nor sympathy of local to recycle urban wastes into productioninputs. authoritiesin Africaand has either been ignoredor penalised by policymakers "partly becauseof the mistaken assumptions Zambia thatthe cultivation of maize isa health hazard"(Rakodi, 1985: In Zambiatwo independentstudies by Sanyal and Rakodi 57). Rakodi (1985: 56—63and I 988a: 495—515) also observed documented a variety of agricultural activities in the city. that urban planning, both in the central and peripheral areas Sanyal (1986and 1987: 197—207) surveyed250 low-income of several Zambianurban centres had not made an attempt households in Lusaka, the capital, and observed that 60 to accommodate urban agricultural activities. Official percent of the households cultivated food in either their responses to urban cultivation vary from country to country gardens or in open spaces in the urbanperiphery. Central to and from urbancentre to urban centre hut have tended to he this study were two important observations. First, that proscriptivesince the acti'ity is perceived as a hindranceto subsistence food amounted to one-third of all the food the developmentof the urban areas envisaged by planners consumedby thelow-income households in Lusaka.Second, and urbanadministrators. Indeed someauthorities have gone that recent migrants tended to have less access to land for so faras tolegislate against urbanfood production (Drakàkis- cultivationand subsistencefood than longer term residents. Smith, Bowyer-Bower and Tevera, 1995: 183—193). In A series of studies conducted by Rakodi in Lusaka Kenya, for example,it is illegal to keep livestock or cultivate between 1970and 1980confirm the extensive natureof urban crops within urban areas (Mazingira Institute, 1985). In agriculture in the city. Rakodi (I 988b)reports that thesurveys addition to being considered illegal those involved in off- showedseveral things. First, that over50 percent of the people plot urban agriculture are subject to sporadicharassment by interviewed practisedon-plot and off-plotcultivation. Second, urban authoritiesin Kenya and Zimbabwe(e.g., the burning that urban cultivation in the city was a key survival niche or slashingof crops and prosecutionof cultivators). particularly for low-income women with limited formal In Zambia urban agriculture appears to he tolerated by education and few marketable skills (Rakodi, 1988a: 495— the authorities(Rakodi, 1988a: 495—515 and l988h), while 515 and 1988b). Third,that the choice of crops growntended in Tanzania and Malawi it is only tolerated in new urban to be influenced by the market, the local by-laws, centres which take cognisance of this activity in urban susceptibilityto theft, labour and water needs. For instance, planning(Mosha, 1991: 83—92). Elsewhereit has never been Rakodi (I 988a: 495—515) observedthat thetendency to grow formalisedand can, therefore, not expect official support cabbages, onions and tomatoes in the home gardens was services. Forthe 'illegal' urbancultivators life is a perpetual determinedby the need to water the crops regularly and to struggle mainly becausethey have toprevail against a barrage protect them against theft. Fourth, that householdsengaged of 'complex regulations' blocking their efforts to survive. in urbanagriculture because of householdneed (i.e. need for Theprevailing repressive attitudestowards urban agriculture extra food or money)and supply factors (e.g. theavailability seem to suggest that urban administratorsand planners are of land for cultivation,labour, time, inputs). not aware of the importanceof urban agriculture. Therehas been a shift in official thinking from intolerance Discussionof Findings and Policy Responses before the 1990s to some tolerance, albeit punctuated by What lessons can we draw from the urban agriculture interludes of harassmentby the authorities.However, there experiences of Zimbabwe, Kenya and Zambia? Having has been a change in the official attitude of government identifiedthe particular features of urbanagriculture in these towards most informal sector activities including urban three countries,one could draw at least four lessons. First, agriculture. Theworsening economic situation forcedofficials there is considerableevidence that urban agriculture is not a in Lusaka to pay a blind eye to urban agriculture since the transient activity in Africa as food continuesto be produced mid-1980s (Sanyal, 1986 and 1987: 197—207). Similarly, within the urban centres. Instead, the facts show that it Mbiba (1994: 188—202) has observed that authorities in continuesto increase in spatial terms and has become quite Harare have relaxedthe controlson off-plotcultivation since extensivein scale. Consequently,urban forms have emerged the devastatingdrought of 1991—1992 and the introduction in Africa that are significantly at variance with the 'ideal' of the structural adjustment programme in 1991. western concepts of urban development. As Freeman Nevertheless, in the three countries urbancultivation is still (1993:19)has observed there is "an acceleratingdivergence looked upon as a rural activity which has no place in the

185 1)aniel S. Tevera African UrbanQuarterly urban areas.While local authoritiesmay he tolerantof urban supported under regulated conditions.Cultivation close to agriculturalactivities, legal paraphernalia inheritedfrom the streambanksshould be discouraged because that contributes colonial era need to he changed as they are still hostile to to environmental degradation and river siltation. However, urban agriculture. other open land not yet ready for development could be Severalmajor recommendations emergedfrom the studies utilised for urban agricultural purposes by individual highlightedabove. First, it seems reasonableto permit even cultivatorsor registeredcooperatives. encourage, the urban poor to grow more food on urban land There is need to establish the extent to which urban but under controlled conditions in order to avert the agriculture is a potential, albeitpartial solution to the problem degradation of the urban environment. Most African of food insecurity at the householdlevel in urban areas. Can governments would improve their ability to deal with urban agriculture be more than a coping strategy for the problems such as widespreadurban poverty by employing marginalised? Does capital-intensive horticulture or dairying the right policies and programmes, especially those that in urban areas have a role and what factors constrain the provide an enabling environmentfor thepoor. Perhaps there developmentof intensivecommercial agriculturein Africa's is need for local authoritiesin these three countries to take urban areas? the lead from Maseru in Lesotho, where the local authority In considering the spatial coexistence ofurban agriculture has taken an enabling approach to urban agriculture by not and other land uses it is important to note that urban only accommodating it but by supporting it. This could be agriculture can use land in ways that do notthreaten the urban achievedby allowingthe cultivatorsto use open spaces on a environment by avoiding use of hazard-prone or temporarybasis until it is ready for developmentinstead of environmentally-sensitive land ill-suitedfor cultivation.For letting the land stay idle or destroyingcrops. Relaxationof example,operations could be confined to specificlocations municipal regulations constraining urban agriculture will lead wherethere is minimum environmental damageresulting and to a growth of the activity and will help provide a cushion the cultivatorsshould be encouragedto cultivate crops which for the urban poor. will help achieve food security at the household level. In addition, there is need for municipal authorities to develop Conclusion formal,organised structuresto administerurban agriculture. The studies that were reviewed on urban agriculture in Zimbabwe,Kenya and Zambia focusedon the importanceof References the activity as a survival strategy of the urban poor and to a Drakakis-Smith, D.W. (1992)."Strategies for MeetingBasic lesserextent addressed the issuesof land-use planning. There Food Needs in Harare, Zimbabwe", in J. Baker and P.O. is need, however, for new studies that address issues not Pedersen (eds) Small Townsand Rural Development inAfrica covered in great depth by the earlier studies, such as the underConditions ofStress. UppsalaNordiska Afrikainstitutet. discussion of the economic benefits and costs of urban agriculture to the nationaleconomies and the contributionof Drakakis-Smith, D. and P. Kivell (1990). "Urban Food the activity to the achievement of sustainable urban Distribution and Consumption: The Case of Harare",in A.M. development, with respect to resourcesand ecology. Findlay, R. Paddison and J.A. Dawson (eds) Retailing In summary, urban agriculture is commonin most urban Environments inDeveloping Countries. London:Routledge. centres in Africa but differences exist in terms of the magnitudeand the institutionalresponses which range from Drakakis-Smith, D. and D. Tevera (1993). Informal Food being prohibitiveto being accommodative. Policiespursued Retailing in Harare. Goteborgs: Occasional Paper No. 7. in Zimbabwe, Kenya and Zambia,generally, seem to fall in Kulturgeografiska Institutionen, GoteborgsUniversitet. theprohibitive category. However,it is importantto note that at the local authority level the policy responses vary Drakakis-Smith, D. et al. (1995). "Urban Poverty and Urban considerably. For instance, Lee-Smith and Memon (1994: Agriculture: AnOverview of theLinkages in Harare", Habitat 67—84) noted that in Kenya policy responses to urban InternationalVol.19 No. 2. agriculture varied from being very prohibitive (e.g. in A.G. etal. Cities Kisumu) to being laissez-faire(in the larger municipalities Egziabher, (1994). FeedingPeople. Nairobi, of Nairobi and Mombasa)and proactive(in the smallerlocal IDRC. authorities suchas Isioloand Kitui). In Kitui urbancultivators ENDA-Zimbabwe(1994). Urban Agriculture in Harare. have access to extension servicesand enjoy crop generally Harare: UnpublishedReport. public support for their activities. Ifthe present harsh economicconditions in thesecountries Freeman,D.B. (1991). A City of Farmers: Informal Urban persist, the number of the urban poor involved in urban Agriculturein theOpen Spacesof Nairobi, Kenya.Kingston: agriculturalactivities is likely to increaseas unemployment McGill-Queen'sUniversity Press. worsens.Given the crisis created by the unprecedented drift to the urban centres there is need for African policymakers Freeman, D.B. (1993). "Survival Strategy or Business to addressthe issueof urbanagriculture in adifferent manner TrainingGround? The Significance of UrbanAgriculture for from what has been the approach so far. For instance, urban the Advancement of Women in African Cities", African agriculture should be allowed to continue, but should be StudiesReview Vol. 36 No. 3.

186 Urban Agriculture in Africa: Volume 11 Nos. 2 and 3, May andAugust, 1996 A ComparativeAnalysis of Findings from Zimbabwe,Kenya and Zambia

Lado, C. (1990) "Informal Urban Agriculture in Nairobi, Rakodi, C. (1985). "Self-Reliance or Survival: Food Kenya: Problemor Resource in Development and Land Use Production in African Cities", African Urban Studies Vol. Planning?",Land Use Policy Vol.7. 21.

Lamba, D. (1993). "Urban Agriculture: Eastern Africa", in Rakodi, C. (I 988a). "Urban Agriculture: Research Questions L. J. A. Mougeot and D. Masse (eds) Urban Environment and ZambianEvidence", Journal of ModernAfrican Studies Management. Ottawa: IDRC. Vol. 26.

Lee-Smith,D. and P.A. Memon (1994). "Urban Agriculture Rakodi, C. (I 988b)."Urban Agriculture in Lusaka,Zambia", in Kenya", in A.G. Egziabher et al. (eds) Cities Feeding in I. Dankelman and J. Davidson (cds) Women and People. Nairobi : IDRC. Environment in the Third World: Alliance for the Future. London: Earthscan. Maxwell, D. and S. Zziwa (1992). Urban Agriculture in Africa: The Case of Kampala. Nairobi: ACTS Press. Rogerson, C.M. (1992). "FeedingAfrica's Cities: The Role and Potentialfor UrbanAgriculture", Africa Insight Vol. 22 Mazambani, D. (l982a). Aspectsof Pen-Urban Cultivation No.4. and Deforestation Around Salisbury, 1955—1 980. Harare: Unpublished MA Thesis. Department of Geography, Sanyal, B. (1986). Urban Cultivationin East Africa, Food- Universityof Zimbabwe. Energy Nexus Report. Tokyo: United Nations University:

Mazambani,D. (1982b)."Peri-UrbanCultivation in Greater Sanyal,B. (1987). "Urban Cultivation AmidstModernization: Harare", The Zimbabwe ScienceNews Vol. 16 No. 6. How Should We Interpret It?", Journal ofPlanning Education and Research Vol. 6. Mazingira Institute (1985). Urban Food and Fuel Project. Nairobi: MazingiraInstitute. Smit, J. and J. Nasr (1992). "Urban Agriculture for SustainableCities: Using Wastes and Idle Land and Water Mbiba, B. (1994). "InstitutionalResponses to Uncontrolled Bodiesas Resources",Environment and UrhanisationVol.4 Urban Cultivation in Harare: Prohibitive or No.2. Accommodative?",Environment and Urbanization Vol. 6 No.1. Tevera, D.S. (1995). "The Medicine that Might Kill the Patient: Structural Adjustment and Urban Poverty in Mbiba, B. (1995). "Classificationand Descriptionof Urban Zimbabwe", in D. Simon et al. (eds) Structurally Adjusted Agriculturein Harare",Development SouthernAfrica Vol.12 Africa. London: Pluto Press. No. 1. Watts,T. and W. Bransby-Williams (1978). "Do Mosquitoes Mosha,A.C. (1991)."Urban Farming Practicesin Tanzania", Breed in Maize Plant Axils?" Medical Journal of Zambia Review ofRural and Urban Planningin Southern and Eastern Vol. 12. AfricaVol.1. Wekwete, K.H. (1993)."Urban Agriculturein Southern and Mougeot, L.J.A. (1994). "Leading Urban Agriculture Into Eastern Africa",in L.J.A.Mougeot and D. Masse (eds) Urban the 21st Century: Renewed Institutional Interest", in A.G. Environment Management. Ottawa: IDRC. Egziabheret al. (eds) Cities FeedingPeople. Nairobi: IDRC.

187 & Leading the field in policy Policy studies for 25 years Editor:RandaU Smith. Universityof Bnstol. Politics Abstracts Editor:Derek Hawes.University ofBristol.

& Politics,volume 25. 1997 Pohcy & Politics is a multi-disciplinary Policy (4 issues): journal concerned with the origins, UKiEurope Rest ofthe world impact and evaluation ofpublic policy Institutions £87 £95 across a range ofareas, including health, Individuals £39 £42 housing, governance,labour markets, Published quarterly inJanuary, April, July and education, the environment, urban October. policyand community care. Subscriptions to Policy & Politics Recent articles include: O Iwould like to subscribe toPolicy & Politics Sarah Payne,Psychiatric care in the and endose payment. (Cheques should be does fail community: it young men?; made payable to Universityof Bristol) ChristianToft, Constitutional choice, C Pleasesend me a pro formainvoice for: multi-level government and social 0 Individuals security systems in Great Britain, Germany and Denmark; Francis G. 0 Institutions Castles and A ChristopherPierson, C Pleasesend me a free inspection copyof new convergence?Recent policy Policy& Politics developments in the United Kingdom, Name Australia and New Zealand. Organisation .....__. Special issue January 1997 Address Regionalism in England: current trends and future prospects Edited jointly by Peter John and Alan Whitehead, this issue includes articles on: Returnto: The Policy Press, University of Bristol.Rodney Lodge. Grange Road. Bristol • Businesspartnerships and regional 8S8 4EA government (Coulson) TeI:+44(0) 117974 llI7, Fax: +44 (0) 1179737308 • The offices for the government E-mail: [email protected] English regions (Mawson and

Spencer) louts Policy & Politics ispublished by The Presson behalf ofthe • Regionalisation,regional institutions Policy School for Policy Studies, University and economic development (Evans of Bristol. II and Harding) I' !L!Li IS5N 0305-5736 African Urban Quarterly, 1996: 11 (2 and 3) pp. 189—200 ©1996 by African Urban QuarterlyLtd.

URBAN FOOD PRODUCERS' DECISION MAKING: A CASE STUDY OF KIBERA, CITY OF NAIROBI, KENYA

P. R. Dennery c\o Dr Luc Mougeot,International Development Research Centre, P.O. Box 8500, Ottawa, Ontario, Canada,K IC 3H9

Abstract Accepted April, 1996 This paper presentsfindings from a study of decision making by urbanfood producers'. Thefood producers resided in Kibera, City ofNairobi 'is' largest informal settlement. The study usedprimarily qualitativemethods to investigate howfood producers with limitedmaterial and monetaryresources engage in urban agriculture. The practices of urbanfood producers served as the entry pointof theexamination of decision making processes. The availabilityof cash within the household;the valueproducers place onfood for home consumption;theavailability of labour; and the risks associated with insecure land tenure, theftand landdegradation were thefactors which most influencedthe practices of urbanfood producers. Examples from the study will be used to illustrate the effectof thesefactors. An overview of the types of relationships urban food producers have with otherhousehold members andwith the community at large will be touchedon in the latter sectionsof the paper Résumé Cetarticle analyse la prisede decision des producteurs urbainsde bas revenus1 en sefondant sur des etudes.de cas dans Kibera, le lotessementinformel le plus peuplé de Ia ville de Nairobi, au Kenya. L'étude a recours surtout a des méthodes qualitatives pour documenter comment des producteurs d'aliments dotes de maigres ressourcesmatérielles ou monétaires s'engagement dans I 'agriculture urbaine. Leurs pratiques servent de point de depart pour reconstruire les processus décisionnelsdont elles résultent. Lesfacteurs les plus influentssur ces choix de pratiques sont: Ia disponibilitéd'argent comptantau sein du ménage, Ia valeur que lesproducteurs accordent a Ia nourriturepour I 'autoconsommation, Ia disponibilité de main-d'oeuvre, les risques associés a l'accésprécaire a Ia terre, au vol eta Ia degradation des sols. Quelquesexemples illustrent I 'effetde cesfacteurs sur les pratiques par les producteurs. L'article enfin brosse un des types de relations que les producteurs entretiennentavec les autres membres de leur ménage, ainsi qu 'avec leur communauté. Introduction Urban Food Production Poor urban dwellers are found in increasingproportion In the past decade, there have been two major studieson within urban centres. Husain and Lunven (1987:50—61) urban agriculture in the City of Nairobi.The first study was suggestedthat the populationslum settlementsin the city of carriedout by theMazingira Institute (Lee-Smithet al., 1987) Nairobi could be increasing by 25 to 30 percent annually. and theCity of Nairobiwas one of six Kenyan urbancentres The majorityof slum settlementdwellers have tocreate their that were studied. The majority of the householdssurveyed own meansof support since employmentin theformal sector were engaged in backyard gardening. Animal production is increasingly scarce. Thelivelihood strategies on which poor activities were includedin the study as was food production urbanites rely are often informal and sometimes illegal by urbandwellers in rural areas. The second study was done (Bibangambah, 1992: 303—313). Urban food production is by Freeman (1991: 1—22) and looked at cultivation by one such livelihood strategy, but it is often misunderstood individual producers in theopen spaces of the City of Nairobi. by urban planners and municipal governments. Both these studieswere primarilyquantitative and provided The formulation of effective policy and successful useful information regardingthe age,sex, length of residence projects which account for and promote urban food in the City of Nairobi, income and practices of urban food production activities is dependent on knowing more about producers.These two studies were also carried out at a time how food producers make decisions. So far, littleis known when unbuiltland was still relatively abundant in the city. about the factors which shape producer decisions. Many Lee-Smith et al. (1987) and Freeman (1991) presented decisions revolve around the allocationof resourcessuch as strong arguments as to why municipal and national land, labour, money and produce. Relations within the government; should move towards recognising and household and outside the household are also important supporting urban agriculture. Unfortunately,there has been considerations when examining urban producer decision little progress in this direction, especially in the City of making.The findings presented below are intended to give Nairobi. A legal or institutional framework supportive of some insight into the mind-set of the most importantactors urban agriculture and related activities (composting, waste in urban food production:the urban food producers. water and organic waste recycling, among others) is non- Acknowledgements—Special thanksto Dr Peter M. Ngau ofthe Department of Urban and RegionalPlanning, University of Nairobi for his assistance during fieldwork.The study was carried out with the aid of a grant from the InternationalDevelopment Research Centre, Ottawa,Canada.

189 P R. Dennery African Urban Quarterly existent. Oneof the few encouraging developments for urban vegetables, preparingand sellingstreet foods,making shoes agriculture in Kenya has been theestablishment of theGreen and furniture,sewing, brewing alcohol, prostitution,selling Towns project by the Ministry of Local Government. The drugsand medicinal plants,operating kiosks and construction aim of the project is to assist officialsin therapidly growing among others. It is important to note that a great variation urban centres of Kisii, Homa Bay and Busia to formulate exists between the best-offand the worst-off households in environmentally sound urban development plans Kibera (National CooperativeHousing Union Ltd., 1990). (Grootenhuis, personal communicationand Duchhart and Although the populationof the settlementis ethnically Grootenhuis, 1993: 9—10). Ironically, theGreen Towns project diverse,the area's urbanfood producers areprimarily Kamba was developed based on the results of a food production and Kikuyu. On the other side of the Motoine River and initiativeby the Undugu Society of Kenya in two informal NairobiDam, facing Kibera, there isa large open space where settlements along the Nairobi River, a few kilometresfrom most Kibera food producers have their plots (figure 1). The the Central BusinessDistrict, City of Nairobi. producers refer to this area as Langata (actually this area is only onepart of LangataDivision). The Langata open space Study Design slopes down towards Nairobi Dam and has an important Most of the studies done on urban agriculturein Kenya environmental function. Water from thedafn is usedto supply and other East African countries have been based on partof Nairobi's drinkingwater (Freeman, 1991).There are questionnaire surveys administered to large numbers of three typesof crop production:rainfed cultivation, irrigation respondents, in variouslocations across the urbancentre and, with sewage water and crop productionclose to waterways in some cases, in a few different urban centres (see for where waterlogging and.periodic floodingoccurs. example Lee-Smithet al., 1987; Maxwelland Zziwa, 1990; Typically, thosewho have been growingcrops in the area Freeman, 1991 and Sawio, 1993). As survey methods tend for several years, cultivate more than one plot (shamba in to obscure differences between producers,their households Kiswahili). There are severalreasons for this. Differentplots and differentareas of theurban centres, theresearch methods offer different ecological conditions depending on selectedfor thepresent study wereprimarily qualitative. The topography, soil propertiesand proximity to water.There are number of producers involved in the fieldwork was alarge numberof producers inthe area, therefore, it isdifficult deliberately kept low in order to take an in-depth look at to obtainone large plot. Most producers try instead toobtain decision making and resource allocation by urban food several small plots. Ifpart or all of the food growing in one producers. plot is stolen,having other plots can be an advantage. The producers who participated in the study resided in If rainfall is timelyand adequate, there are two seasons Kibera, a large informal settlement of City of Nairobi. of planting. Maize is grown mostly during the long rains Fieldworkwas carried out betweenAugust and October of (April toJune) but somefast maturingvarieties can begrown 1994. Four types of food producers took part in the study: during the short rains (Octoberto December).Beans can be case study informants (4), focus group participants(a total grown in both seasons.Aside from these two staples, sweet of 12 individuals in three focus groups), open-ended and Irish potatoes,kale (locallyknown as sukuma wiki) and questionnaire respondents (7) and special interest cowpeas are commonly grown. Smaller flood prone areas interviewees (2). The fieldwork also included seven short are commonly plantedto arrowroot (cocoyam), bananas and interviews with community residents not involved in food sugarcane. Pigeon peas, pumpkin, cassava and perennial production and conversations with representatives of four crops are occasionallyfound among the staple crops. High Non-Governmental Organisations (NGOs) and two value crops such as onion, tomato and Swiss chard (which governmentministries. Considerable emphasis was placed Kenyans call spinach) are rarely seen due to theft. on the four case studies as they providedan opportunityfor Intercropping is quite common. Crop rotationand irrigation repeated observations of individual producers within the with sewage water are practisedby someproducers. Animals context of their households. are produced withinthe settlementof Kibera. Chickens are most numerous butducks, sheep and goats are also common. Kibera Food Producers Theftand poorsanitation aresignificant constraints for animal Kibera is an informal settlementin the City of Nairobi producers. Heavy losses, particularlyof chickens,discourage which dates back to late last century when Nubian soldiers producers and cause many producers to abandon animal were allowedto settlethere. Becauseof its large areaand the production altogether. general absence of slum clearance, Kibera has grown A brief introduction of the case study informantswill be exponentially since 1974. Population estimates for Kibera given here as many of the examples below are drawn from vary from roughly 250,000 (Matrix Development the case studies. Each informanthas been given a fictitious Consultants, 1993) to 800,000 (Mulili, personal name in order to protect their identity. A summary of some communication). There are ele'en villages within the of the major characteristics of the informants is presentedin settlement. Kiberais well-known for itsvibrant small business table 1. sector. MatrixDevelopment Consultants (1993: 17) cited over Martha is 38, married, arrived in the City of Nairobi in 7,300 enterprises.Data from respondentsand observations 1975 and has seven children (1 to 15 years of age). This give an indication of the variety of occupationsproducers Kamba womanventured into food production in 1979 after and other Kibera residents engage in. The occupations her vegetableselling business failed. Martha cultivatesthree encountered include selling fish, groundnuts or fresh plotsbut she and her husband havetrouble making ends meet.

190 'C

'0 0'0 P. R. Dennery African Urban Quarterly

Table 1, SelectedCharacteristics ofthe Four Case StudyInformants in Kibera,City of Nairobi, Kenya

Martha John Joyce Robert Age 38 45 42 56 Ethnic group Kamba Kamba Luhya Kikuyu No. ofchildren 7 5 7 4 Householdmember with wife husband & son son wage employment none (factoryworker) (railways & tailor) (mason) No.ofyearsinKibera 19 14 8 12 No. ofyears cultivating 15 12 4 8 No. ofplots 3 5 3 3 Source:Dennery 1995: 59

Martha's husband has difficulty obtainingcasual work and abit ofmoney tobuy seedsand Robert startedfood production though Martha often makes eight to ten trips a day to the in 1986. Interestingly, Robert's three sons come to Nairobi neighbouringresidential areas to sell water, she can only on the weekends to grow food on plots near their father's manageto make enough cash to buy food for one day. three plots. John, 45, also Kamba, is married with 5 children (ages 3 When comparingthe characteristics of the case study to 22). His wife works at a factory. John came to the City of informants and those of the open-ended questionnaire Nairobi in 1967. He was a carpenter until 1982 when his respondents (table 2), a number of qualitative differences tools were stolen during a riot. With the help of a friend, stand out. These are the number of children, the length of John startedin food production a few monthslater and now residence in Kibera and the number of plots. The four case has five plots, of which two are irrigatedwith sewagewater. studyinformants had foodproduction as theirmain livelihood Joyce is the second wife of a Kenya Railwaysemployee, activity while five of the seven interview respondents had age42, Luhya, and has 7 children(4 of themreside in western non-agricultural remunerative activities which take up a Kenya with the first wife). She arrivedin theCity ofNairobi substantial amountof time2. The informantshad an average in 1986to join her husband. Joyce startedgrowing crops and of six children compared with two for the respondents.It raising animals in 1990 to keep from being idle. Shehas three appearsas though the more mouths to feed the more time is large plots and keeps chickens in a confinedarea besideher devoted to food production. Of the four informants,Joyce house. Compared with the other informants,Joyce and her was the only one who had resided in Kibera less than ten householdare relativelywell-off. years. Joyce had lived in Kibera for eight years, the same Robert is a 56-year-oldKikuyu man who resides alone time as the longest residing respondent. The fact that in the City of Nairobi. His wife and four adult children live informants had a larger number of plots than respondents in CentralProvince, Kenya. Robert cameto the city in search perhaps reflects both the greater food needs and the better of ajob in 1982 butcould not find one. Hisbrother gave him access to production resources of well-established urban

Table 2, Selected Characteristics ofInterview Respondents Engaged in Food Production in Kibera,City ofNairobi, Kenya

average

Respondent 1 2 3 4 5 6 7 Sex male female male female female male male

Age 28 22 50 30 34 40 32 33.7 No. ofchildren 1 1 3 2 3 5 0 2.1 Ethnic group Kamba Kamba Kamba Kikuyu Kamba Kamba Kikuyu Wageemployment runs security city kiosk/mason none guard none cleaner watchman mason Wageemployment runs / city of spouse kiosk painter none employee mechanic none n/a No.of years in Kibera 2 5 5 6 8 6 3 5.0 No.of years cultivating 1 2 I 4 3 4 2 2.4 No.of plots 2 2 - 1 1 2 1 1 1.4 Source:Dennery, 1995: 84

192 Volume IiNos. 2 and 3, May andAugust, 1996 Urban FoodProducers'Decision Making: A Case Study ofKibera, City ofNairobi, Kenya residents. An importantpoint emergesfrom the above: having the bearer of the title for the plots they used. Rather they a largehousehold serves as an impetusfor activeengagement were "de facto tenants abiding by customary usufruct of at least one householdmember in food production as a principles"(Freeman, 1991: 78), fully consciousof the fact main livelihood activity. that they did not own the land. Anothercomparison can be made betweenproducers on Insecurityof tenure has a marked effect on the practices the basis of food production practices. In both of the single of urbancultivators. Loss of access to land has a permanent sex focus groups, producerspointed out several differences and devastatingimpact. Some producers, like Robert, whose betweenestablished producers and thosewho had startedonly plots border on or are near constructionsites, feel the threat recently. New producers must purchase most or all their oflosing accessto theirplots quite acutely.Yet they continue materials and have little experience/knowledge of how to to takethe risk of losing the fruit of theirlabour. In particular, successfullyproduce food in the City of Nairobiconditions. producers whose main occupation is agriculture have few Establishedproducers have to absorb the maintenanceand other alternatives. Theseare usually among thosewho failed replacement expenses for seeds, tools and in some cases, to getjobs when they came from rural areas. The chance of agricultural chemicals. Those who have engaged in food getting one now is very remote. production for severalyears know whatpitfalls toavoid (e.g.: Cultivators try tocut theirlosses in case ofeviction. Most harvesting maize and beans as quickly as possible to avoid evictions take place without prior notificationand without theft) and the most efficient ways of carrying out tasks. New compensation. Usually neither the building contractors nor producers also claimedthat theirestablished counterparts had the legal owners are ever seen by the producers. One day, the more money todevote to foodproduction. This seemsa valid producergoes to his/herplot only to find theircrop cutdown statement as the householdsof established producers have and building materials on the site. Having more than one benefitedfrom severalyears of savings on food expenditure plot, as is common for the long establishedKibera producer, and cash from produce sales. is a form of insuranceagainst eviction. Not all the plots are likelyto be built upon at the same time and some plots may Accessto Land be spared that fate altogether. Eventually, producers may find Food production has no clear legal status in the city of themselves in aposition where they nolonger have any plots Nairobi and, in general, is not well viewed by local policy- to grow food in. Oneproducer encountered during the study makers3. This needsto be kept in mind whilelooking at issues was almost at that point. This woman said that she used to of land use for food production.In recent years, there has have many plotsbut was down totwo plots. Most of the plots been a trend towards the privatisation of Government of were simply reclaimed by the title bearer. In one case, the Kenya land, particularlyin the City ofNairobi wheremarket maize she had plantedwas simplycut down to make way for demand for land is high. The privatisation process has taken the expansion of Kibera. She was not given a chance to place withoutthe simultaneousimplementation of an overall harvest anything which had been growing there. land-useplan. Clearpolicies regarding urban food production, Perennial,long maturing and semi-permanentcrops are protectionof fragile land or of waterwaysand multiple uses avoided by most Langata area producers.Martha is the only of land are not in place. informant who had a number of different fruit trees in her Part ofthe land in theLangata Area of the City ofNairobi plot. She planted them between 1979 and 1985. When asked was formerly Government of Kenya land which the Prison why no othertrees were planted, Martha repliedthat she began Authority had been allocated. The area closest to the to realise that she would lose everythingin case of eviction. residential estates was allocated to the National Housing Plots such as Joyce's (situated on Ministry of Lands and Corporationby the Governmentof Kenya for house building Settlementproperty) or thosealonghe NairobiDam areless but construction of new estates has not yet started. The threatenedby privatisation than thoseon what was formerly remainderwas or is waterwayreserve, public or privateland. PrisonAuthority property. Thismay explainwhy Joyce, after Much of the PrisonAuthority land was privatisedstarting in four years of cultivation, has planted her first mango tree 1989. In the last two or three years, a numberof those who and why severalproducers are using vegetationto transform have obtained title deeds for the land have begun building the edge of the dam into soil. When trees are present, it is houses for rentto middleincome earners on the formerPrison usually for one of three reasons: either they were planted a Authority land, resultingin the displacementof severalfood while ago, they are located in an area which is certain or producers. unlikelyto becomea building site or their use-valueis high. An additional difficulty presents itself regarding land- Common examples are fruit trees such mango and guava, use and land tenure in Langata and metropolitan area of perennialslike banana or papayaand longer maturingcrops Nairobias a whole. Traditionally, as Freeman(1991: 74—78) such as pigeon pea. In any case, the surface areas covered carefullyexplains, the Kikuyu had in the past recognised the are relatively small compared to those devoted to growing rightsof those who occupied agricultural land as belonging fastmaturing staple crops like maize and beans. In essence, to someoneelse. They were consideredtenants of alandlord. choosingcrops which maturerapidly increases the likelihood If the landlordevicted thetenant or thetenant chose to leave, of successfullyharvesting them. "the tenant is allowedto harvest his standing crops even after There is a clear difference between longer-established termination by the landlord" (Maini, 1967 as cited by producersand thosewanting to venture intofood production. Freeman,1991: 75). None of the foodproducers encountered Accordingto one of the focus group participants,"all those during the study paidmoney to the Governmentof Kenya or with the best shambas are those who started early." In other

193 P R. Dennery African Urban Quarterly words, as would have been expected, the areas with fertile on a first come first served model. Kinship, ethnicity and soil and favourabletopography and location were occupied belonging tothe right social networks isof criticalimportance first. In a sense there are three categoriesof producers:those to those entering the arena of food production for the first who had plots in Langata prior to 1989, those who started time as well as to those already operating within it. Despite just afterthe PrisonAuthority left and those who startedlater these insights, the gap between the Langata producer and than this. The producers in the first category had the best those who have the ultimatecontrol over land-usein Langata possible situation. Having worked in the area for a number (the wealthy, NairobiCity Counciland theMinistry of Lands of seasons, those producers could identify which sites had and Settlement)remains very wide. the best soil or drainage in the event they wanted additional plots and took over the plots when the PrisonAuthority left. Food Production and Household Economic Status The second type of producers still had a relative advantage Food production has different functions for different over the last type as the former still had the possibility of households. Forlow- and very low-incomehouseholds, food choosingplots and, possibly, of adding otherplots. The trend production ranges from between being one of several ways towards landprivatisation whichexisted in the Cityof Nairobi of meetingbasic needs to being the only or principalway of beforeand during fieldworkwill probablyentail a shortened doingso. Whereon this continuuma householdlies depends engagementin food production for those who started their primarily on economic realities. In Kibera, making a living plots after 1989. Unlike Martha, Robert or John, such exclusively from agriculture is uncommon.In most if not all producersmay not benefit from a 10- to 15-year horizonfor cases, other livelihoodstrategies are used in additionto food urban food production. production, particularly in times of low agricultural By the early I 990s, the demand for vacant land in the production (in betweenseasons and justafter the rains begin). Langata area which new producers could use exceeded the However, food production plays a key role in household supply.At the sametime, moreestablished producers wishing maintenance. Each of the producers in the study indicated to expand their operations were (and are still) looking for that the economic status of their household would fall additional plots. Similarly, constructionsites have increased substantially ifthey were no longerable to produce some of in number and slum settlement expansion continues, theirown food.Freeman (1991: 110) expressed thesentiments effectively reducing the amount of land available for of producers in this way: "The vast majority of urban cultivation. At present, virtually all areas which can be cultivators,both male and female, are very poor, landless, cultivated are in use. In some cases, this includes highly subsistencedwellers for whom theirlittle shamba may mean oxidised, marginal soils and steep slopes. For established the differencebetween a precariousbut continuedexistence producers, seeking to expand, he or she can ask other in the city and a full-blownfamily catastrophe." producers whether they or someone they know is planning At the individual and household level, microeconomic to cease production (due to relocation, sickness and factors sometimes strongly influencethe decisions producers discouragement, among others). Alternativelyand probably make with regards to the resources they put into and obtain morefrequently, aproducer who is aboutto leaveapproaches from food production.For example,Martha planted sukuma someonewhom he/she thinks may be interestedin the plot. wiki and spinach,which produce leaves for sale in a matter The situationis considerably more complicated and risky of weeks in order to improve cash flow. Producer decisions for those seeking a plot for the first time. Who one knows also reflect external policy conditions like changing food becomescrucial to obtaining a plot. Individualswho do not prices and land use. In 1994, maize was planted with have the appropriate socialties are shut outof food production increasing frequency in and around the City of Nairobi, altogether. Long-establishedproducers know what gifts they despitea high risk of theft because of a sharp rise in maize must give to secure the use of an additionalplot (usually a prices. kilo of sugar and an agreed upon sum of money). Such The ability of the producer and the members of the producers also know who is the 'real'4 user of the plot and householdto generate cash and goods determinessome of do notrisk losing money or crops to temporaryoccupants. A the options availableto the producer. Employmentof either potentialproducer will probablyexperience some difficulty theproducer or a householdmember, particularly of spouses, obtainingland. He or she is unlikely to haveextensive contact raises the householdhigher up the economicladder. Among with established producers unless the produceris already a the informants, there was a clear difference betweenthe two friend, relativeor neighbour. Forexample, kinship tiesproved householdswhere a spouse was regularly employedand the useful for a producerwho startedin 1992.Upon retiring,his two where this was not the case. Other factors affecting uncle returned to his rural home and left the former the financial position were number of dependent children, shamba. Sharing a common ethnic affiliation or living or whether the children resided in the City of Nairobi or in the workingin the same place can also facilitatesecuring access rural home, access to and types of casual employment and to land. Recent migrants to the City of Nairobi usually start abilityto borrow money from friendsor relatives. off by living with relatives and then find aplace of theirown Cultivation is aseasonal enterprise.For Kiberaproducers nearby. Similarly, producerswho have plots in Langatawill it is an activity which occupies the betterpart of the year. be more inclined to pass them on to eithera relative or a co- Even with sewage water irrigation, the amount of food ethnic they know well rather than to a stranger. Thus there producedin the fields varies with the season.When the long appearsto be a form of limited controlby the producers over and short rains are good, food supplies are generally lowest land. The distributionof new plots is not arbitrary or based in October/November(during planting and weeding) and

194 Volume 11 Nos. 2 and 3, May andAugust, 1996 Urban Food Producers' Decision Making: A Case Study ofKibera, City ofNairobi, Kenya from January to April (the longer of the two dry seasons). Value of Food from a ProducerPerspective The end of the long rains and maize harvest (end of June! Food producers choose between keeping produce for beginningof July) is the time when the shamba suppliesthe home consumption and sellingor giving awaypart of it. The most produce. Regardless of the time of year, households allocation of food among various uses depends on the purchasefoods which they cannot grow. When food supplies productive capacity of the producer and on the financial are low, the household must have the cash to buy position of the household. Productioncapacity dependson supplementary food. Better-offhouseholds, such as Joyce's numerous factors. The most important of these are the area can more easily cope with the increased spending need at cultivated, available labourand soil quality. Kiberaproducers those times than households such as Martha's. In a sense, make a numberof decisionsconcerning how theirproduce is regularwage employment serves as a bufferagainst declining allocated prior to planting their fields. Essential to an food supplies. Households relying on casual work to help understanding of these decisions is the value producers make ends meet faced an additional constraint as some of attribute to the produce they obtain. From the viewpoint of thesejobs are availableonly atcertain times of the year. For the producers, market-bought food is so expensive that in example, construction generally takes place when there is Martha's words "the food we harvest has a greater value little or no rain. compared with any wageswe would getif I or another family Seasonal variations in food available from the shamba membergot ajob." Producecan be used as food (use-value) imply achange in producer and household financial priorities. or can be easily converted to cash unlike wages which have Joyce's case best exemplified this. Low production meant only an exchange value. that she spent less on more expensivefood items like meat, Kibera producers assertthat they do notproduce for sale: cookingoil and wheat flour and that more ofthe household's male and female participantsof the focus groups expressly cash was spent on food. Joyce's decision to plant cowpeas stated that they do not grow food for sale and that their plots and beans very early effectivelyshortened the period when were too small for commercial production. This is true of nothing was growing in the field. Since the rain fell at crops such as beans, sweetpotatoes, arrowroot and cowpeas appropriate times, the cowpeas she had planted were whichare primarily plantedfor home consumption. Surpluses producingleaves by mid-October. Unlike Martha,Joyce did of these crops may be sold or sent to arural home. However, not need to seek an alternatesource of incomein casual work. in the case of green maize and sukuma wiki, the producers She relied either on her husband's income or on proceeds assertion is not consistent with what actually takes place. from the saleof cowpea leavesto make foodpurchases. Thus, These two crops are in large part sold. The fear that maize production practices,the numberof mouthsto feed and wage will be stolen if left in the field until it is mature is common employment can all affectbuffering capacity. The periodsof among producers. The crop is quickly removed from the low food supply in the City of Nairobi correspondwith the fields,particularly if the plotsare notguarded at night.Maize opening of the three school terms.For low-income producer cobs whichare matureor close to maturity are dried and kept households with school-age children, this presents a for householduse. The green cobs, which usually constitute tremendous difficulty. Those households are expected to the bulk of the crop are sold. Sukuma wiki is a crop which is allocatealready scarce financialresources to school-related planted primarily for home consumptionbut for which the expenses. Poor householdshave few options. They cannot surplusis largelysold. This crop isalso useful in maintaining cut into the food budget any further and so must try to defer cash flow becausesales take placefrequently over the course fee payments or take one or more children out of school of two or three months. (permanently or temporarily).This aspectof seasonality can Sales of surplus can be considered advantageousto the potentially have very adverse consequences over the long- producer as he/she can obtain cash without cutting into the term. household's food supply. Kibera producers are particularly Intra-mensalvariations have a similar effectas seasonal reluctant to part with the portion of produce set aside for variations in production.Most wage earners are paid at the household use. However, when needs arise, a producerwith end of the month. Dependingon incomeand expenses,it is limited financial resources may feel that he/she has no choice not unusual for householdsto run shortof cash by mid-month. butto sell, compromising householdfood supply. Thosewith Certainexpenses such as rent must be paid all atonce, early comparably larger disposableincomes can choose between in the month. The results are a substantial reduction in selling produce and using non-agricultural income. Other available cash—commonin all households—notjust those options are borrowing money and doing casual work. The with regular wage earners. Decision-makingpatterns are preference for one of these options depends largely on the likely to change even within the course of a month. When producer's access to each of them and on the nature of the translated into practical implications for the urban food need. producer,.material and cash resources arelikely to bediverted Agricultural produce is not commonly traded, though away from agriculture. If seeds or other inputs need to be some Kibera shopkeepers will accept produce in exchange purchased, these expenses are put off until more cash is for othergoods. Produce may also be used as giftsat weddings available.Carrying out operations on a timely basismay be or when a producervisits his/her rural home. It may also be hindered resulting in a reduction in the amount of food given or shared with neighbours and friends. The latter obtained. depends on the will of theproducer and, in all likelihood,on the availability ofthe produce. In this respect, foodproduction

195 R. P Dennery African Urban Quarterly has both an economic and a social function.All informants to produce food for the household came out of necessity. said that they gave produce as gifts. John gave produceas a Martha and John's respective small business ventures had wedding gift, Robert contributes food at church functions, failed. Martha had been selling market-bought vegetables weddingsand funerals and Joyce gives food to neighbours, whileJohn's carpentry tools had been stolen.Robert had not relatives and friends. Martha gives small amountsof maize been able to find ajob. Robert says "I was not educatedand to friends when she is able.Arrowroot was commonly given knew I would not be able to find a job." Robert's age when away as a gift. he arrived in the City of Nairobi (48) was probably an Evidencefrom this study indicatesthat cash incomefrom additional hindranceto finding work. Joyce is the only one food sales is used specifically to meet other basic needs. In who faced no dire circumstanceswhen she first embarked the women's focus group, answeringthe question "what do on food production.Joyce began food production because they use the money they obtain from sellingproduce for?", she hated being idle. In Joyce's view, staying at the house said they use it primarily for food and basics such as paraffin with nothing to do leads to gossip and quarrels. In food and maizeor wheat flour. The participantsin themen's focus production,Joyce also saw an opportunity to diminish her group gave a similar answer. They said sales of produce foodpurchases and for having somethingon hand which she should not even be called "a sale" but rather "an exchange" can sell in times of need. The informants were not asked if as the money is used to buy something else the household they had considered alternatives to food production while needs. One man gave this example: "When I grow sukuma they were deciding whether or not to begin. It is possible wiki and my childrenwant somethingelse to eat, I sell it and that there may have been no reasonablealternatives due to buy the children whateverother food they want." Payment lack of capital to (re)starta small enterpriseand to relatively of school fees appears to be an important use for revenue low potential for wage employment. from produce sales. Produceis sometimes soldbecause school Both men and women were very close in how they fees must be paid. A single mother, with 6 adult children, perceive urban agricultural activities. Women see food said she educatedall her children through agriculture. production as part of their duty in feeding the family. Households which rely solely or partly on agricultural Obtaining income from sales was considered secondary to production for theirlivelihood have an advantage over those saving on food expenditure. Theparticipants in the women's who do not. They can substitute food for cash that would focus group added that producing food increased their otherwise be spent on food. Kibera producers considered financial independence. They did not need to ask their reduction of food expenses to be one of the main benefitsof husbandsfor cash to buy food or make small purchases.This urban agriculture. It is also an important motivation for •is an important benefit for women as their status within the continuing. The greater the quantity/acreageand the larger householdmay improve. Men appear more iiiclined to see the variety of crops produced, the more cultivatorsare able foodproduction as aform of self-employment, providingthe to satisfy a larger proportion of their food needs. Non- means for household survival/sustenance. When men talk producers in the study dependedin large part on market and about urban agriculture,they generally place less emphasis store bought food. Only one of seven stated that he on feeding the householdand more on supportingit. Sales occasionallyreceived food from his rural home.The others and profit were not a prime motivationof male producers in had to rely on theirown or their spouse'swork to put foodon Kibera.The men in the focus group repeatedlysaid that food the table. In some cases, street food sellers can use unsold production is not a business.These men specified that the wares for householduse. One respondentsaid that his wife plots are generallytoo smallto produce a substantialsurplus. was occasionallypaid with food for herlabour. Thelow food Genderdifferences were notclearly apparentwith regards supply and relatively high food prices during dry season to the proportionof food keptfor home consumptionversus probablyhad a lesserimpact on householdsengaged in food what is sold. Capacity to produce food, the number of production than on other households. Low-income food household members who depend on this food and the producers usually have the added benefits of amore diverse availabilityof incomefrom non-agricultural sources largely diet prior to the start of dry season as well as stored food determinethis proportion.Since men aremore likely to live from recent harvests. Thoughnon-producers seemmore likely in the City of Nairobi alone, it is probable that they spend to be engaged in some form of income generation,food to less on food than female producers.As Joyce and several them had neither a use-value nor an exchange-value. women in the focus group pointed out, the financial responsibilitiesof Kenyan men are inherentlydifferent than Producersin the Household those of Kenyan women. Men are expected to provide an By looking at an individual's personal history, a income by working outside the home while women are relationship between producer knowledge and current expectedto help within the home. practicesemerges. Most urbanproducers learn thebasic skills At the intra-householdlevel, conceptualisation of the they use for cultivation in rural areas. Often, one or both sexualdivision of labourtranslates intoattitudes which shape parents were food producers. The events which occurred the allocation of resources and producer practices. Women around the time the informants decided to start food in the focus group said that because they prepare food, they productionappear to have been the initial push or incentive know the needsof the householdand, therefore,decide how which led to the decision. Serious financial difficulties much produce to sell and what food to buy. The food precededthe entry of three outof fourcase study informants production activitiesof womenalso shape their expectations into food production.For John, Martha and Robert, starting as to how their spouses should allocate income from non-

196 Volume 11 Nos. 2 and 3, May andAugust, 1996 Urban FoodProducers' Decision Making: A Case Study ofKibera, City ofNairobi, Kenya

agriculturalwork. Oneof the reasons why Joyceexpects her to agriculturaltasks. At the other end, much of Martha's time husband topay for school fees is becauseher foodproduction was taken up by the care of young children and procuringa efforts largely eliminate householdfood expenditure. small incomewhich she used for food and school fees. The Male and femaleproducers had comparable viewsof who immediate needs of the family took precedence over the is involvedin decision-making.Whether decisionsaffecting longerterm need to produceher own food. Martha's decision the household were made jointly or individuallydepended to make daily survival a priority forces her to trade-off time on the importance of the decision (buying maize flour for for food production for time selling water. Martha's ability the next meal vs borrowingmoney for amedical bill) and on to control her agricultural labour time is limited by her the nature of the power relationships.Consultation with the responsibilitiesto others. spouse served as a meansof preservingmarital harmony. For For agriculturalwork, access to the labourof others can example,Joyce makesdecisions regardingthe shamba alone take the form of direct assistancein the shamba or assistance butconsults with herhusband only to maintaingood relations with other tasks which free up the producerfor such work.A and keep him up-to-date. general complaintamong Kibera producers was thereluctance Self-perception and decision-making powerare intimately of others to assist with food production.Relatives, friends related.To illustrate, Joyce's case is compared with Martha's. and neighbourswere found to be more willing to help with Joyce's social status and self-confidenceplace her in a more houseworkthan with cultivation. Women in the focus group advantageous position when making decisions. By having said they could only once ask friends to help in the shamba morematerial and social resources, Joyceeffectively has more and that these were usually people the women knew quite freedomin her choices. Joyce can save moneyfor unexpected well. However, when labour demands were the greatest, it expenses, borrow money from her neighbours or from the waspossible for most of theproducers interviewed to obtain variousgroups she is involvedin and has the optionof hiring limitedassistance either at home or in the field, particularly labour and a vehicle to transporther green maize.When her from aspouse. Yet overall,individual producers worked alone husband's salary advance and the balanceof his salary runs for the better part of the time in the field. out, Joyce can sell produce without negatively affecting Bottlenecks caused by shortage of labour were obvious household food supply. Because Joyce has access to and in John's case. For him, there were three major difficulties: controlover arelatively large amount of produceand because First, he had more land under cultivationthan it is possible she can count on a regular source of cash, the consequences for one personto manage.Although John couldkeep up with ofpoor decisions are not likelyto have an extremelynegative the work on the two plots irrigated with sewage water, the impact on her household. In Martha's case, her decision- other two plots he cultivated were weedy. The fifth plot was making power is underminedby two factors: how she views left fallow because the soil was poor and, perhaps also, herself in relation to her husband and her capacity to meet because John was not able to devote time to improving its the needs of her family alone. Martha must provide most of fertility. Second, John had only limited access to his wife's the food and cash for household use. In spite of her hard- labour. John's claim that he could ask his wife to stay away workingnature and of thefact that she almostsingle-handedly from work anytime in order to help him in the fields is ensures her household's survival, Martha still prefers to probably not completely true. The opportunitycost of lost portray herhusband as the one in chargeof makingdecisions. income is likely to exceed what would be gained from the But in actual fact, Martha makes importantdecisions which labour assistance his wife can supply. John's assertion that affect the time she devotes to food production versusother he would beat her if she refused to miss work is probably activities, the amountof produce she will eventually obtain more accurate. The third difficulty was his indebtedness. John and the availability of cash for food and other expenses had fairly sizeabledebts which could notbe paid back using (school fees for the children). In Martha's case, necessity income from sales of produce and his wife's salary. often determineschoice. Martha's options are more limited Consequently, he did casual work on occasion but was than Joyce's as she cannot sell produce without affecting reluctant to give details concerning its nature. John's householdfood supply, except when selling surplus sukuma exemplifies the problem labour shortage can pose for the wiki. urban food producer. Even though cultivationwas the main occupationof the Labour Management case study informants, considerabletime was spent on other For Langata producers, the busiest times in the fields activities. Opportunities for casual wage work and self- occur in October—November and April—May during employment differdepending on sex, age, educationand will. ploughing,planting and weeding.Since these tasks have to Several livelihood activities performedby Kibera residents beperformed on atimely basis, controlover one's own labour are physically demanding (e.g., carrying water and and accessto that of others becomemeaningful determinants constructionwork) and poorly paid. Householdmaintenance ofpotential productivity. Control over one'sown labourwas activities such as collectingwater, washingclothes are time of greater importancethan labour assistance from another consuming, particularly in dry season. Focus group data source be it spouses, relatives, friends or hired workers. indicated that what was true of Martha is also true of other Among the case study informants,there was arange of extent womenwith youngchildren: they could not go to the fieldin to which producerscontrolled their own labour. At one end, the morning until the childrenwere fed and that they had to Robert had few external demandson his time when he was cut shortthe time they spentworking in order to return home in Kibera. Robert could, therefore, devote most of his time to prepare meals. Instances of competition between food

197 P R. Dennery AfricanUrban Quarterly production and other productive or reproductiveactivities in case of crisis. In view of the costs associated with can thus occur on a daily as well as on a seasonal basis. transporting bodies tothe morgue and then (usually)to a rural Anumber of thelabour issues presented above are directly home for burial, there are a large number of funeral groups linked to the prevailing gender ideology in Kenya. Female which exist as separate entities from welfare groups. The urban producers must carry out most of the care and assistance such groups can provide to very low- and low- maintenanceof the household, regardless of the time they income households is invaluable.Alone, few producers or devote to food production or other livelihood activities. non-producers would be able to recover from the financial Women, individually and in groups, commentedthat they burden a severe illnessor the death of a householdmember tended to stay in the field longer than men (to harvest would create. The social support these groups provide is of vegetablesfor the evening meal for instance). Even if both equal or greater importance than financial assistance as men and women had worked in the shamba together, the members can encourageeach other and share their concerns. womanwas still responsiblefor fetchingwater and preparing Financialposition may serveas a hindranceor barrier to mealsonce she returnedto the house.In couples,this division group involvement. Due to her relatively low cash income, of tasksdid not vary when one or both persons had full-time Martha experienceddifficulty in participatingin one group or part-timeemployment, regardless of who was employed. and cannotjoin one group as she would like to. Martha had Even thoughJohn's wife works all day in the industrial area, to abandonmembership in an NGO-basedsavings group for she is still expected to prepare the evening meal.Part of the thepurchase of school uniformsdue to lackof funds.Martha difficulty in obtaining labour assistance stems from the shared her desire to join a group of women from her ethnic prevailing attitude towards food production, particularly group.These women performdances for which they getpaid. towards food production withinthe city. Manyurban residents Money for this group's activities was obtained from the left their rural homes to searchfor wage employment in non- registration fees of newmembers and monthlycontributions. agriculturalwork and reject the notion that food production The funds were pooled with the intention of buying land to is an appropriateurban activity. build rental houses on. Thusjoining a group or maintaining membershipin a group can be problematicwhen cash flow Producersin the Community is reduced and survival becomesthe top priority. Previous studies provide little insight into the During the fieldwork, there was surprisingly little relationships between urban food producers and the evidence of producer pooling together to purchase seed or communitiesto which they belong. Yet producers are, to a other inputs or to obtain credit. However, urban food greater or lesser extent, affected by the relationshipsthey producers, particularly those with plots in Langata, share have with those outside their household, be it with relatives, manycommon problems (e.g., theft, uncertaintywith regards neighbours,friends or acquaintances.There are also events to land and labourshortages). It appearsas thoughurban food in the village, the community or higherlevels which translate producers function relatively independently despite their into changes at various levels of society. Urban producers physical proximity. There are,however, a few groups ofurban areactive participants in a numberof differenttypes of social food producers althoughthere is no umbrellaorganisation in networks. Involvement in a networkdepends both on need which all Kibera food producers can participate. Langata and on personality. In a dense settlement such as Kibera, producersassociate together in two types of groups.The first contact with others is an integral part of daily life. Such type tends to form aroundcauses or events and are often of a contacts not only fulfil a social need but can provideaccess temporarynature. The second type usually involvesa longer to money, materialgoods and information. Urbanproducers term goal. Distinctions betweenthe two types are sometimes contribute to various groups in the hope of creatinga safety unclear. Thesepoints are illustrated below. net in case of emergency. Surprisingly, contact among Martha reported that in May 1994, the crops of several producers is usually limitedand producerorganisations tend producers were slashed down to make way for a school on to lastonly afew months.The issue ofproducer organisation land adjacent to anearby residential estate.The victimsspread is of great importance to the design and implementation of word among the producers. The groups approached the new policy. An effectiveorganisational base would facilitate Divisional Chief to demand compensationfor the victims' contactwith, participationof, and negotiationwith the urban lost crops. Compensation was later granted. Several groups foodproducers who would be affected by suchpolicies. Land form each year aroundthe time maize harvest begins.These tenure and land-use issues could potentiallybe addressed if groups raise money and manpowerfor guardingthe crops at contact between producers and policy-makers were made night. In some cases, guards are hired and in others male easier. producers contribute the necessary labour. One such group Food producers try to ensure that they have some form obtaineda letter fromthe Kibera Divisional Chief which they of assistance incase of illnessor death. Few Kiberaresidents can showif police enquire aboutthe nature of theirpresence can afford to save money as individuals.However, group inthe fields atnight. Joyce was thetreasurer of the same or a membershipprovides a good incentive for residentsto save similar group. She mentioned the organisation was also together. A common form of group involvement is involved in lobbyingthe Chief for compensationin case of membershipin a welfare or funeral group.Most welfare and land/crops loss and for permanent use of land in the area. funeral groups require a contribution of Ksh. 50 or 100 a Participation in the group appears to vary according to month.5 The assets of welfare groups are usually distributed availabilityof time and money and to the extent to which on a lottery basis. They are also loaned or given to members producers would see a need for protectionfrom theft. Robert

198 Volume 11 Nos. 2 and 3, May andAugust, 1996 Urban FoodProducers'Decision Making: A Case Study ofKibera, City ofNairobi, Kenya had also associatedwith a few producers to hire a pick-up are thus likely to affect group cohesion. Divisions already truck for the bulk purchaseof cow manure. exist along gender lines as women may not share the same Organisations motivated by longer term goals often interestsor may feel more strongly about certain issuesthan requiresizeable monetary investments. Consequently, these men. Differences in economic status may further alienate organisations are more volatile because of the temptationto producers from each other. use group funds for other purposes. John described his An important disincentive for producers to organise is experience with one of these groups. "In 1985, we formed a the powerlessness they feel with regards to their largest group and each contributed3,000 shillings ayear. When the problem: ensuring short- and long-term access to land. The money totalled 60,000 shillings, the leaders of the group problemquickly takes on political dimensionsas it involves disappearedwith the money and were never seen in Kibera Government of Kenya officials, urban developers and wealthy again. This is why I keep my involvement with other investors. Kibera producers,as individuals and a group, see producers to a minimum." John did not disclose what the the land issue as a problem with no solution or one which purpose of the group was but, due to the sum involved, it inevitably means a victory for more powerful groups in was probablyrelated topurchasing land. Mismanagement of Kenyan society. Realistically, how can low- and very low- relatively sizeablefunds engenders mistrustand may preclude income food producers, who are considered squatters and futureparticipation. Even for relatively modestventures, like reside in an informal settlement, make their concerns heard the water selling or chicken raising projects proposed by and actedupon by thosein power positions? Themost obvious participants in the third focus group, pooling and ways of tackling these issues are for producers to organise administeringfunds is an important issue. The producers in themselves in groups. One possibility for organising this group were concerned that they would not be able to producers is to form small cells of producers based on the raise all the necessary funds and were unsure where to get welfare group model. These cells could then be brought financialassistance. Availability of money may precludethe together undera larger umbrellaorganisation. The next step participation of poorer producers,reinforce unequal power would be for Community-Based Organisations or Non- relations and skew access to land and other resources. Governmental Organisations to assist producers in Problemswith the organisation of producers were brought communicating with policymakers project and thereby help up, particularlyduring thethird focus group discussion.Lack bridgethe gap throughadvocacy. Considerable care will be of informationabout the existenceof such organisations was needed in selecting advocates. Because land tenure issues frustratingfor someof those present. Martha wasparticularly easily become politicised in Kenya, advocates would have upset to hear that Robert and Joyce were leaders while she to be seen as politically neutral. had no idea these groups even existed. The requirement that Onceagain, the producers will have to know beforehand there be at least 25 participants for the group to he given how they could potentially benefit from forminggroups and official recognitionby the Chief and Ministry of Culture and having other organisationsassist them with lobbying. From Social Services was another hurdle. Individual producers the producers' perspective,the obstaclesto forming new or occasionally attributedthe fact that they did not personally enhancingexisting organisations may seem to outweigh the know the producersin the neighbouringplots to the fact that advantages of maintaining the status quo. However, an each person has his/her own concerns and should not be incentive for producers to organise could be created if the bothered.Due to theirmultiple livelihood and social activities, producers felt their voices would really be heard by the absences and schedules, organising meetings with large Governmentof Kenya. numbers of producers is difficult.Unless the producersfeel they might obtain a tangible benefitfrom group efforts,they Conclusion are unwillingto participate. Several producers complained The findings presentedabove deal primarily with three that the existing groups rarely do anything of value aside agricultural resources: land, produce and labour. These from guarding plots. Issues of prices for their produce and resources have both a social and an economic value. Land access to land were not being effectively tackled by such tenure and land-use management emerged as critical issues groups. during the fieldwork. In the Langataarea, food production is There appearedto be other difficulties as well. Seasonal in direct competition with housing. There are no clear variationsin labourdemands place a severeconstraint on the provisionsfor compensationin case of evictionnor arc there time producers can commit to group activities. During arrangements for producers to use land in Langata or planting and weeding, only the most dedicated producers elsewhere for cultivation. Although environmentally could be expected to attend group meetings. The only desirable, a land-use plan for the area which includes exception being when urgent action is needed to obtain managementof the NairobiDam watershed and preliminary compensationin case of eviction. The continuity of group treatmentof sewagefor use by food producersis unlikelyto activitiesand the planningprocess are disrupted. Someeffort be successfulunless a middleground between producers and is required to build the organisationback to where it was policymakers is created. prior to the disruption.The majority of producersare either Labour availability is akey factordetermining the amount Kamba or Kikuyu.There may be sometension between those of produce obtained. Among food producing households, twogroups and withother groups. In alarge group, the Kamba there is usually one adult who devotes most of his or her and Kikuyu will likely be more numerous than other groups time to agriculture, particularly in households with large such as the Luhya. Political preferencesand cultural values numberof dependentchildren. Genderissues were significant

199 P. R. Dennery African Urban Quarterly determinantsof access to and control over agriculturaland References non-agricultural labour. Women in Kibera and in Kenya Bibangambah, JR. (1992). "Macro-Level Constraints and overallare responsiblefor chores such as fetchingwater and the Growth of the Informal Sector in Uganda", in J. Baker preparing meals regardless ofhowmuch timethey have spent and P.O. Pederson (eds) The Rural-Urban Interface: in thefield. Male producersare more likelyto have casual or Expansion and Adaptation. Uppsala, Sweden: The regular work apart from food production than are female Scandinavian Institutefor African Studies. producers. Completing agricultural operations on a timely basis often requiresreallocation of individual and household Dennery P.R. (1995).Inside Urban Agriculture: An labour. Explorationof Food Producer Decision Making in a Nairobi Urbanfood production is animportant livelihood strategy Slum. The Netherlands: M.Sc. thesis. Department of for Kibera households.Produce is a multiple-useresource, Ecological Agriculture, Wageningen AgriculturalUniversity. providing food, generating revenues from sales, allowing savings on food expenditure and produce has an exchange Duchhart,I. and F. Grootenhuis(1993). "Urban Agriculture, value in that it is used to meet other basic needs. In most Incentive Planningand Green Towns", Urban Perspectives cases,income and employmentfrom a source other than urban Vol.4No.1. food production are needed to help make ends meet, since the amount and types of produce are not sufficient to meet Freeman,D.B. (1991). A City of Farmers: Informal Urban all household food needs. Food producers and their Agriculturein the OpenSpaces ofNairobi, Kenya. Montreal: householdsare less vulnerable to changes in market prices McGill-Queen'sUniversity Press. for food compared to their non-producing counterparts. Kibera food producers operate within a constantly Freeman, D.B. (1993). "Survival Strategy or Business changing context. Therefore decisions and their practical TrainingGround? The Significanceof Urban Agriculture for outcomesare. constantlyrevised in order to satisfy the needs the Advancement of Women in African Cities", African of the moment. Many of the practices of Kibera food StudiesReview Vol. 36. producers serve to fulfil short-termneeds, yet thesepractices are nonethelesspart of a longer term strategyfor household Husain,A.and P. Lunven(1987). "Urbanization and Hunger food self-sufficiency and for maintainingan adequatecash in the Cities", Food andNutrition BulletinVol. 9 No. 4. flow. Lee-Smith,D. etal. (1987). Urban FoodProduction and the Notes Cooking Fuel Situation in Urban Kenya—National Report: A full descriptionof the study and itsresults can be found Results of a 1985 National Survey. Nairobi: Mazingira in Dennery (1995). Institute.

2 Thepurposive sampling method used to selectinformants Maxwell,D. and S. Zziwa(1990). UrbanAgriculture: A Case and respondentspartially accounts for these differences. Study of Kampala. Makerere Institute of Social Research (Mimeograph), now publishedas Urban Farming inAfrica: Lee-Smith and Memon (1993) state that cultivation in The Caseof Kampala, Uganda. Nairobi:ACTS Press, 1992. Nairobiis allowedexcept on publicstreets while Freeman (1993) describescultivation as technicallyillegal. MatrixDevelopment Consultants (1993).Nairobi's Informal Settlements: An Inventory. Nairobi: Report PN-ABH-74l A producermay loan or rent the plot to someoneelse on preparedfor USAID/REDSO/ESA. occasion or may hire someoneto work on it for him or her. Therefore, the person currently using the land may Memon, P.A. and D. Lee-Smith(1993). "Urban Agriculture not be the initial producer and, according to Kibera in Kenya",Canadian Journal ofAfrican Studies Vol. 27 No.1. producers,the former cannot pass on the prerogativeto use that land. The initial user is the one with whom National CooperativeHousing Union Ltd. (1990).A Survey negotiationsmust be undertaken. of informal Settlements in Nairobi. Nairobi: NACHU.

At the time of the fieldwork, the Kenyan Shilling was Sawio, C.J. (1993). Feeding the Urban Masses?: Towards appreciating rapidly. 1 US dollarwas equivalent tobetween an Understanding ofthe Dynamics ofUrban Agriculture and Kshs. 35 and 45. Land-Use Change in Dar-es-Salaam,Tanzania. Worcester: Unpublished Ph.D. dissertation. Clark University.

200 New at SAGE in 1997! JOURNAL OF EUROPEAN SOCIAL POLICY Edited by Graham Room and AnneJamieson both at Universityof Bath, UK "extremelyimpressive.., the journal can be thoroughly recommended." - EuropeanAccess TheJournal of EuropeanSocial Policy features leading specialistsand academicsacross Europe providing expert analysisof key socialpolicy issues. Bothempirical research findings and theoretical developmentsare reported. Up-to-date European Briefing "An extremelygood Digestsection follows developments in social Europe." - Notabene The Digest sectionoffers an invaluable guide to the latest European social trends, new legislation, analysis of new and existingEuropean Commissionprogrammes and recent research. Publishedquarterly. ISSN: 0958-9287

OrderForm for New Subscribers- Subscribe atthe Introductory Rate SAGE Publications Ltd. 6 BonhillStreet, London EC2A4PU, UK Call our Subscription Hotline on +44 (0)171 330 1266 USAorders to besent to: ______P0 Box 5096, Thousand Oaks,CA 91359 1ctI,odsof Pnij,izcnt Name ______0 1enclose a cheque (madepayable to SAGEPublications Ltd)for: ______Address______Please invoicemy credit card U MastercardU Visa Amount: 6 91 ______U Iwanttosubscribe to Journal ofEuropean Card No:Tfl! I I I I I I I I I I I I I SocialPolicy starting with Volume 7(1997) ______U IntroductoryRate for Individuals Expiry Date: £31/US$49 (Usual Rate £39/US$62) U Institutional Rate £115/US$184 Date: / / LI Please send me a brochureon thejournal Signature: African Urban Quarterly, 1996: 11 (2 and 3) pp. 202—208 ©l996 by African Urban QuarterlyLtd.

THE BEAUTIFUL CITY: GARDENS IN THIRD WORLD CITIES

Mary Cockramand Shelley Feldman, respectively 10 Frederick Street, Apt. 301, HertfordConnecticut 06105, USA, Email: [email protected] [email protected]; and Departmentof Rural Sociology,Warren Hall, CornellUniversity, Ithaca, New York 14853

Abstract Accepted May, 1996 Internationalcomparisons of research on theagronomic, social,and policy aspects ofurban agricultureshare theconclusion that urban food production is an important component of household survival strategies. Most urbanfarmers are poorly educatedwomen withfamilies who possessonly afew ofthe skills that are valuedin the marketplace. Such women are among thosemost likely to investtheir labour butfew other inputs, inurban agricultural production. These urban farmers creatively gain access to rights-of-wayand use vacant land which they usuallydo notown to growstaple foods and raisesmall livestock near their dwellings. Hostile governmentpolicy andaccess to waterare key constraints to theirproductive capacity. These conditionshelp explain why successfulgardens are often limitedto the rainy season. Yet, urban agriculture, whichflourishes during difficulteconomic times, provides between 10percent and 30percent of the householdbudget. Thepaper concludes that urbanfarmers, like their rural counterparts,engage in work that is gender stereotyped. Since subsistenceproduction is perceived to be part of women's responsibilityfor feeding thefamily, it is hardly surprising that women are the dominant participants in this art of urban survival.

Résumé L'article trace un bilandu statut deI 'agricultureurbaine contemporainequi permet desituer I 'experienceafricaine dans une perspective globale, notammentau niveau des cultures potageres. Les lois et normes culturelIes interagissent avec les conditionsagronomiques et techniquespour encourager ou contraindrelocalement le développementde cette agriculture. Enpremier lieu, I 'article caractérise lesproducteurs eux-mêmes, puis exposeles contraintes agronomiques et techniques les plus souvent affronteespar ceux-ci, ainsi que les strategies auxquellesus ont recourspour contourner ces contraintes ou en diminuer l'impact sur kurs activités. Les auteurs discutent ensuite certains obstacleset difficultés d'ordre socio-politique (politiquegouvernmentale, contrôle des ressources,division du travail et distributiondes tdches au sein des ménages).

Introduction opportunities for employment in theformal sector are limited It is nota new observationthat the world's populationis and incomefrom the informal sectoris low. Urbanagriculture growingat a tremendousrate; it has more than doubledsince isnot anew idea, but canbe aneffective developmentstrategy 1950. Mega-cities and large urban centres, especially in the in a supportivepolicy environment. This paper reviews the Third World, are experiencing unprecedentedpopulation status of urban agriculture through an analysis of current increases—Mexico City, for example,grew by 9 million,an grassroots urban agriculture,projects and literature on this increaseequal to the populationof New York City injust the issue which is scatteredthroughout a range of disciplinary last 15 years. In 1950, only five large urban centres in the journals.The firstsection examines thepresent status of urban world had a populationof over five million,but by 1990, 30 agriculture around the world. Although knowledge is large urbancentres had passed the fivemillion mark (United relatively limitedand few comparable studieshave been done, Nations, 1991). Poor people make up the vast majority of some common characteristics of urban agriculturists do this urban increase and the lack of adequate shelter, emerge. The second section addresses common technical employment and foodcauses serious problems inurban areas. constraints to farming in urban centres, focusing on the Urban centres are unable to provide services for new obvious elements of land and water. The third section inhabitantsand poor people often have little accessto health addresses social barriers and contradictions to sustaining care and education, few or inconvenient water and sewer urban agriculture in the current policy environment, services, poor roads and inadequatetransportation. examining as well issues related to household resource This paper examinesone way that urban residentsmake control, gender divisions of labour and household labour urbancentres more livable and more productiveby practising allocation. urban agriculture. Widely practised in the past and in the present, food productionin and aroundurban areas provides Urban Agriculture much-needed food for urban residentsand recycles organic Urbanagriculture isbroadly defined to includecultivating wastes. Many urban dwellers,predominantly poor women, plants, raising animalsand fish, and growing fungi within a find that growingfood is not only possible in the urbancentre greatermetropolitan areaor an urbancentre. Plants,including but is necessary for family maintenance. Gardens are herbaceousannuals and perennials,shrubs and trees may be commonlyfound in Third World urban centres and are used consumeddirectly for food or may be used for fodder, fuel, by the poor as a means to guarantee subsistence where buildingmaterials, medicine, ornamentation or industry. Food

202 Volume 11 Nos. 2 and 3, May andAugust, 1996 TheBeautiful City: Gardens in ThirdWorld Cities crop production will be the focus of this paper because it is in the household food supply and nutritional level. Not the most common type of urban agriculture in low-income surprisingly, malnutrition is common in urbancentres around neighbourhoods.Contrary to most expectations from more the world. developed countries (MDCs), most urban gardens do not Food production saves thehousehold money by replacing compriseprimarily herbs and vegetablesbut includestarchy foodthat would havebeen purchased or, invery poorfamilies, staples, fruit trees and livestock. Urban agriculture as a supplementing purchased food. In Africa, there is a "large researchconcern has been largelyoverlooked by the academic proportionof very low-incomehouseholds for whom urban and developmentcommunities. Though urban agriculture is farming becomes an important activity to supplement relevantto urban-focuseddisciplines such as urbanplanning householdbudgets" such that "urban farming has becomea sociology, geography, rural, environmental or naturalresource critical variable in sustainability" (Wekwete, 1992: 131). departments, it does not fit neatly into any of these. Most Squattersettlement families in Nairobi, Kenyasaid that food published works on urban agriculture are anecdotalreports was their first concern (Hake, 1977) and 65 percent of all with few references or descriptive studies based on small householdsin the capital grew food on urban or periurban sampling populations. Thefocus of mostof thesestudies were land (Freeman, 1991). Brownrigg (1985: lii) reported that gardenersor low-incomeurbanites, although researchin Asia an advantage of periurbansquatter settlements is the "chance is often based on census data.The limitedresearch attention to plant a few fruit trees and vegetablesand raise some small tothe questionof urbanagriculture may also bepartially due livestock". to the Westernconcept of an urban centre as aplace for non- Agriculture's contribution to the household can be agricutural activities. In ThirdWorld countries, urbanofficials significant, providingup toalmost half of ahousehold's food often believe thatgardens are a remnantof rural life that will needs. Gardens provide 10 percent of the householdbudget disappearwith acclimatisation to the urbanenvironment, even in Lima,Peru (Ninez, 1985: 6—14) and 10—30percent of food thoughresearch does not support this. Officials are, therefore, costs in Buenos Aires, Argentina, the equivalent of 5—20 reluctant to support policies or projects which reflect the percent of the total household budget (Gutman, 1987: 37— legacy of rural life and its unmodern, rural and 42). Agricultural productionin low-incomeareas of Lusaka, unsophisticated tie with the land. Despitethe ideathat gardens Zambia was 10—15 percent of the food budget in the mid- are tied to rural life, the urban poor continue to cultivate in 1970s (Ledogar, 1978: 57—62). Lucenet(1988) found a range urban centres as a way to address economicconstraints and of 50—75 percent participationin agriculture among seven malnutritionand as a meansto diversify household resources. studies of Lusaka's low-income residents. Low-income They not only serve as a backstop against hunger but also residentsin Kampala, Ugandaproduced nearly45 percentof decrease the cash requirementsof thehousehold and provide their own food in 1989 and calculations from gardeners' better quality food than poor urbanites can purchase.As in estimates indicate that 36 percent of the city's households rural areas, home gardens are viewed as insuranceagainst a wereagricultural producers,of whom three-fourths werelow- break in food supply. income(Maxwell and Zziwa, 1990). Large-scale employment The primary reason people grow food in urbancentres is of garden workers in food production has also been to feed their families.Again and again, case studiespoint to documentedand is most common in urban centreswith large economic necessity as the driving force behind urban green belts such as Maputo, Mozambique, where a third of agriculture and to the important role gardens play in the urban workforce was involved in agriculture in 1980 household survival. In Kampala, Uganda, "urban farming (Kolstrup, 1985). Clearlyurban centres do encompass a wide tends to be an aspect of life for poor urban dwellers,who range of peopleengaged in production.Urban agriculture is supplementtheir meager income by producing food on any a remarkablywidespread and importantactivity among the availableland" (Maxwell and Zziwa, 1990: 21). Gardenscan low-incomeurbanites. In Asia, food gardens are part of the provide an element of stabilityduring inflationor economic urban landscape.As von Fleckenstein (1978) noted, food crisis, when they increasein frequency. A surge of people in gardensare a part of Melanesiancity life and are found in all Kampala, Uganda and Lusaka, Zambia began gardening yards "except perhaps those occupied by expatriates". during the recent economic declines of the 1970s and early Panjwani (1986:29) estimated that urban agriculture and 1980s. Likewise, underconditions ofjob uncertainty or under- associatedwaste recycling employsover a million people in and unemployment, farmingis likely to increaseas Thaman Calcutta, India.And in China,20 percent of the urbanlabour (1977) found in Papua New Guinea and Fiji. force is engagedin agriculture (Bjorklund, 1987: 2—15). In theurban cash economy, urbanresidents are expected Urbangardeners are mostoften low-income, middle-aged to be able to meet all their needs through the market, yet women with families to support, little education and few wagesare often belowthat requiredto meetdaily needs. Food marketable skills. In Kampala, gardeners weremostly married is the most important requirementfor survival and the cost women with a householdof seven and a similar profile of of food alone can consume half or more of the household the urban farmer is found in Lusaka (Sanyal, 1984; Rakodi, budget of low-incomepeople (Sanyal, 1984 and Freeman, 1988:495—5 15 and Maxwelland Zziwa, 1990).The average 1991). Studies of 10 Latin American urban centres indicate Nairobi farmerwas a38-year-old woman withsix dependents. that the lowest income quartile householdsspend 50 to 69 Almost half had previously gardened before starting their percentof their incomeon food and beverage(Austin, 1980). own plot(Freeman, 1991). Two-thirds ofPort Moresby, Papua When a household spends such a large proportion of its New Guineafarmers werelow-income women with families incomeon food, fluctuationsin incomeoften lead tochanges to support and a few years of primary education (Vasey,

203 Mary Cockra,n and Shelley Feldman African Urban Quarterly

1985:37—43). In Bogota, Colombia, BuenosAires, Argentina system based on communes, although there is evidence of and Santiago. Chile, gardeners are overwhelminglylow- friction between plannersand farmers (Skinner, 1981: 215— income (Page, 1986: 38—44; Gutman, 1987: 37—42 and 280). Some of the most economically viable of Russia's Robson, 1989: 28—29). In China, unlike many other places, remaining state farms are those in or near urban centres urban farmers are full-time market producers who usually specialising in high value horticultural and ornamental specialise in vegetable production, organised in collective production; the communal structure is less cohesive with teams by the state (Skinner, 1981: 215—280). lowervalue staples in the rural areas.In Russianurban centres, Thoughpoor, most gardenersare neither recent migrants institutions such as hospitals and prisons areturning to rooftop to the urbancentre nor refugees; plantinga garden seems to gardensto provide food in adifficult economic environment. lllow once a family has a place to live and has decided to In Maputo,cooperative farms, market gardens and individual stay in theurban centre.This is because gardening requiresa subsistence farms co-exist on land set aside for urban degree of certainty of continued urban residence and a agriculture. commitmentof time in order to benefit from the investment Market gardens are also found in urban areas, typically of' planting and caring for a garden. For example,gardeners at the urban centre's edge. In Singapore and Hong Kong, had lived in Kampala for more than 10years before beginning market gardens are very common and provide the islands a garden (Maxwell and Zziwa, 1990) and in Nairobifor more with 20—45 percent of their vegetable needs and 70—100 than five. Over 70 percent of Nairobi farmersindicated that percent of poultry (Cheng, 1976; Tempelman and they planned to remain in Nairobi and not move back to the Suykerbouk, 1983: 62—72 and Wong, 1983). These market countryside(Freeman, 1991). gardenstend tobe largerand moreintensively cultivated than Thesesubsistence gardeners grow staples and vegetables subsistencegardens. Market gardeners usually have secure for theirhouseholds and might marketor exchangesome food land tenure, and thus they invest more in the garden than with friends and family. Staples such as maize,sweet potatoes subsistence gardeners, growing high value vegetables and and cassavapredominate in most gardens. Vegetables arealso sometimes livestock (Niiiez, 1984). Market gardens are grown, particularly in Asia, where vegetablescomprise up usually the primary employmentfor at least one household to half the daily diet. Many gardensbegin as a few bananaor memberand often outsidelabour ishired. Men in agriculture maize plants and increase in size and species variety as the are much more likely to be found in market gardens than gardenergains skill and confidence. The resourcesused for subsistence production. Gardens serve as a source of cultivation, includingtools, inputs and planting materials are employment, including paid day labour in subsistence usually obtained from family or friendsand thegardener and gardens, long-term jobs in larger market or cooperative her family usually provide all the labour. Although most gardens and self-employmentfor millions of individuals. subsistencegardeners prefer to garden on their houseplot, Urban gardens provide employment for women and others many also plant gardensat a considerabledistance from their who are disadvantagedin the labour market, both through home on land they do not own nor have securetenure. self-employment and waged work. In Kampala,for example, Although household subsistencegardens are by far the even small subsistencegardeners occasionally hire labour to most common type, communitygardens, cooperative farms help with harvest or planting, although workers regard and market gardens also exist in urban areas. Community agriculture as undesirableemployment because of theheavy gardens—which usually comprise groups of individual manuallabour involved(Maxwell and Zziwa, 1990). gardenersbanding together to use larger tracts of land or to The prevalence of food production in urban areas share information,tools or other resources—rarelyemerge indicatesthat manylow-income people perceive it as atheans spontaneously. In the United States of America community to meet basic daily needs in a difficult environment.That gardeninggroups in urban centresaid members in acquiring urbandwellers cultivatein climatesas diverseas dry Maputo, access to land, water and a range of other services,but these tropical Fiji and temperate Shanghai indicates that they formal organisations are much less common in the Third consider agriculture a worthwhileendeavour to meet their World. Urban gardening projects are usually designed as food and incomeneeds. community gardens because they help to facilitate the sharing of information and resourcesand also because gardens help Technical and Agronomic Issues to advance the social agenda of some urban development Thephysical surroundingsof an urban centre, including initiatives which are generally aimed at poverty alleviation. the climateand thequality and availabilityof landand water, As some note, self-reliance through urban agriculture can play a major role in the appearance of urban cultivation. have a powerful psychologicalimpact and be empowering, Access toland and waterare the mostimportant and generally especially when groups of gardenerscommunicate with each the most difficult garden resources for the poor to acquire. other (Page, 1986: 38—44; Wade, 1987: 29—36). Silk (1982: Agronomic issues such as the agricultural methods used, 16) argued, for example, that through gardens "inner city inputsavailable, and thecrops growninteract with gardeners' residentshelp to develop neighborhoodsolidarity and thus motivation and needs to produceunique cultivationsystems grow not only vegetables,but hope". in each city. Communal gardens and cooperative farms where Land is, obviously, a critical component for urban gardeners share the labour and divide the harvest and its gardens. Both the high cost of urban land and agriculture's proceeds are found primarily in socialist and communist low profitability relative toother industriesinhibit agriculture societies. China boasts a well-developed urban agriculture in urban areas and contribute to the common though false

204 Volume 11 Nos. 2 and 3, May andAugust, 1996 The Beautiful City: Gardens in ThirdWorld Cities beliefthat agriculture is seldom found in urbancentres. Large such as for cooking and cleaning compete with agriculture. amounts of urban land are idle, however, due to Gardenersare willing to pay for water they use for gardens. inaccessibility, speculation or because it is unsuited for Except for family labour, few inputs are used in urban building. Smit (1980: 499—506)estimated that there are 22 centre plots. This is not surprising since most subsistence square miles of arable land in New York City and four or gardenershave little capital to invest.Planting materials are five in both Cleveland and Detroit, USA. Only half of the obtained from family and friends, kitchen wastes or the land in Bogota, Colombia is developed (Lowe, 1992: 18— marketat relatively lowcost. Secure tenureof land correlates 25). In planned urban centres, significantopen space may be directlywith higher investmentand inputs; Lusakan farmers designatedto remain open and these planned areas may be with secure land tenure invested much more than their themost hospitable to urbanagriculture. Nairobi, for example, counterparts without tenure (Sanyal, 1984) and market was designedby theBritish with about one-fourthof theurban producers tend touse more inputs than subsistence gardeners. area deliberately left as open space (Freeman, 1991). In African urban centres where gardens are illegal, farmers Althoughthere is alarge amountof unused land in urban refrainfrom using chemicalfertilisers and pesticidesfearing areas, much of it is inaccessibleby legal means to the poor, that these inputs would draw attention to the garden. for whom buying land for agriculture is seldom a feasible Moreover, chemical fertilisers and pesticides often are not proposal. Over half of non-gardening Lusakan, Zambia availablein the urbancentre in the small quantities an urban survey respondents did not cultivate because they did not gardenermight need.Although the urban system producesa have enough land to start agarden (Sanyal, 1984). The urban large range of organic wastes which could he used in poor usually controlonly the lot on whichtheir home is built, agriculture, few are actually recycled. In Kinshasa, Zaire, which may be no larger than the house itself. Still, roughly gardeners pay to have market wastes transported to their one-thirdto one-half of urban gardens worldwideare found garders,where they sort the wastesand add the organic matter on the gardener's land usuallyon the houseplot. Many people to the soil. The Chinesehave a complex nightsoil collection prefer to garden close to home because these gardens are system for agricultural utilisation, hut cultural barriers and easier to integrate into the routine of other work, because health concernsconstrain the use of nightsoil elsewhere. produce can be harvested as needed and immediately Agricultural technology from nearby rural regions may consumed, kitchen wastes can be added to the garden and or may not be appropriate to the urbanenvironment as climate these gardensare less vulnerableto theftthan are moredistant conditions can differ over short distances. Plants face gardens. additional stress in the urban environment, including wind, For many people, the householdplot is inappropriate or compacted soils and damage from people and animals. The insufficient for cultivation. Some plant unauthorised most productive urban agriculture systems use intensive 'guerrilla' gardens on vacant spaces along canals, roads and technology, includingthe addition of organic wastes that are industrial installations in urban centres worldwide, despite plentiful inurban areas. Gardeners in Asia and Latin America uncertainty about harvesting what they have planted tend to use more intensive agricultural techniques such as (Brownrigg, 1985). Themost commonspaces includevacant raised beds and soil amendments than do most African lots and land set aside for parks or future development, cultivators. Intensified horticulture outside Shanghai has periurbanland, riverbanksand rights-of-wayfor highways, resultedin production levels equal to the best Japaneseland railroads and utilities. For landowners,allowing gardens on and the harvest of up to nine crops per field annually (Ash, vacant land may be a means to prevent other unwanteduse 1981: 188—221). The techniques urban gardeners use to of the land, such as trash dumping and shantytown produce food depend in large part on the materials at hand construction. In Rio de Janeiro,Brazil, the electriccompany and the specificskills and knowledgeof the gardener. leased land underpower lines to gardeners preciselyfor these Most commercial hydroponic techniquesare ill-adapted reasons and the 172 hectares in cultivationproduced food to poor urban areas because they have high start-up costs, worth US$ 10 million in 1983(La Rovere, 1985). In China, require preciseapplications of imported chemicals, rely on where controlover both land and agriculturalproduction and dependable electricity and require a significant investment management iscentralised, land surrounding the largestbuilt- to learn the method. Hydroponic projects have had a varied up parts of the urban centre is allocatedto agricultureand in success rate and the most successful have generally particular to intensive vegetable production.As the urban emphasised low technology, sustainedtechnical support and! centre grows, the state allocatesmore land atthe urbanfringe or a working model of the method (Wade, 1986: 30—34 and to agriculture andmakes an effort tokeep improvedvegetable 1987: 29—36; Kendall, 1990:18 and ILEIA, 1994). In arooftop fields in production. gardening project in St Petersburg, Russia, which taught a Like land, wateris required for urbancultivation. In urban container gardening method and two hydroponic methods centreswith adry season, gardens areoften rainfed and appear adapted to local materials and conditions,gardeners preferred only in the rainy season. In other urbancentres where a water the container technique because it was easier to understand source for irrigation is convenient,gardens occuryear-round. and to maintain. Port Moresby residents who have functional water taps While the availability of land is critical to urban irrigate daily, for example, whereas residents without agriculture,people have found ways to cultivate tiny plots, convenient access to water do not irrigate their gardens often integrating small livestocksuch as poultry, rabbits and (Thaman, 1977). This is because few gardeners are willing swine with the plants. Where in-ground gardens are not tocarry water togardens and becauseother demands for water possible becauseof lackof space, poor soil or other problems,

205 Mary Cockram and Shelley Feldman African Urban Quarterly gardeners prove to be creative adapters. They plant in in neighbourhoods where local authoritieswere favourable containers that range from stacked tyres and tins to walls to urban agriculture (Rakodi, 1985: 53—63). Maputo, with built-inplanters, although container gardening generally Mozambiquehas large green belts which were established requires more frequentcare and fertilisation(Urban Resource by the socialist government at independenceand are now Systems, 1986 and ILEIA, 1994). As we will show in the loosely regulatedby city government. next section, land tenure affects the amount of investment Asian governmentshave a longertradition of supporting and labourhouseholds make in urbangardens. urban agriculture than in most of Africa. The Chinese government has long promotedfood self-sufficiency in urban Social Determinants areas partly because the urban food supply is the state's While the availabilityof land, water and other resources responsibility. Chinese urban centresproduce 85 percent of determine whether urban cultivationis possible and affects their own food and the 17 largest urbancentres produced 60 agricultural productivity, social factors influence the percentof theirneeds in the early 1980s(Skinner, 1981: 215— accessibility of these resources and establishes a social 280 and Bjorklund, 1987: 2—15).Many aspects of agricultural climate for gardening. For example,the government policy, production are centrally planned, including major production gender division of labour, allocation of householdlabour and decisionssuch as the quantity and type of crop. This leads to otheremployment demandsshape interest in urban agriculture disagreement between planners who strive for uniform as does the availability of other social infrastructureand production of all crops and peasant producers who want to support. This means that laws and cultural norms interact grow more profitable vegetables for 'self-advancement' with agronomic and climaticconditions to encourageor limit (Bjorklund, 1987: 2—15). the developmentof urban gardens. Government support of urban agriculture usually Governmentpolicy plays an important rolein legitimising increases the incidenceof urban agriculture, but does not or restricting urban cultivation. Land-use regulations affect necessarily improve conditions for poor urban farmers. whetheragriculture is a temporary or permanentphenomenon Housingprojects in Butare, Rwandawere designedwith lots through urban land-use policy, zoning laws, tax structures, big enough for gardens and with agronomists to provide and urban statutes, all of which can be modified to promote technical advice tohomeowners. Despite the fact that banana or discourage urban agriculture. Official attitudes toward and sorghum are traditional backstops against famine and urban agriculture range from mandates in China to benign providesignificant organic matterfor the soil, thegovernment neglect in Argentina to open hostility in Kenya. prohibitedtheir cultivationwithin urban areas in favour of Government policy toward agriculture is most 'modern' methods (Pottier, 1989: 461—477). The constraining in Africa, where cultivationof food plants is agronomists' recommendationswere very different from banned outright in many urban centres and discouragedas thoseof traditional cultivators and were not alwayssuccessful backwardand rural in others. Since no action is takenagainst in the relatively poor soil and low-input conditions of the the non-food plants and trees that are common in urban project. centres, it isclear that crops are singled out precisely because Urban agriculture policies most affect low-income, they are food crops. Interviews with urban officials in middle-agedwomen with familiesto support,little education Kampalapointed to a cultural difference between gardeners and few marketable skills: those who depend most on and authorities.Whereas local officials tended to be aware agriculture.The preponderanceof women engaged in urban of the importantrole of agriculturein householdfood supply, food productionis due in part to gender roles. Women often higher officials were less tolerant of farming which they have or assume responsibility for food and water believed to bea temporary holdoverof rural culture(Maxwell provisioning, and cooking fuel. Many women ir Lusaka and Zziwa, 1990). As noted earlier, however, most urban considergardening as part of their domestic work much like gardeners have lived in the urban centre for several years cooking.Urban gardening projectsmay help meet women's beforeplanting agarden and engagein agriculture as an urban immediatepractical gender needsfor food, butfrequently do solution to an urban situation. Indeed, many of Nairobi's not address deeper issues of gender equity and employment gardeners have lived in the city their entire lives. opportunities unless women use the gardening project as a Where governmentshave recognisedthe importantrole forum to recognisestrategic gender interests. that agriculture plays in low-income households, policies Gardeningis also a response to women's disadvantaged favouring urban cultivation are found. These are most positionin the labour force. Women are frequentlypaid less common in urban centres where food supply is very tight than men for the samejob. And in mosturban centres, female- and government officials want to limit the funds spent on headed households represent a disproportionately high imported foodstuffs. In Addis Ababa, Ethiopia, the city percentageof thepoor because they have a high dependency governmentallocated vacant land for gardensand employed ratio and limited access to services and employment. poor men to produce vegetableswhich were sold to the poor Although urban agriculture is a source of employment for (Nelson, 1978: 25—46). In the early 1980s, the Government women and otherpeople disadvantaged in the labourmarket, of Zambia promoted urban gardens in the national these people may already be fully occupied meeting their developmentplan in order to reduce food imports,increase basic needsthrough other forms of non-wage labour. Gardens urban food self-sufficiency and enable the poorto make their are relatively flexible in the amount and timing of labour house payments(Sanyal, 1984). Locallevel officials, though, neededand can productivelyabsorb extra labour. Labour is had abigger impact on gardens,which were twice ascommon the single largestinput in urban cultivationgreater than any

206 Volume 11 Nos. 2 and 3,May and August, 1996 The Beautiful City: Gardens in ThirdWorld Cities

other input, averagingone hour of labourper week for every or settingaside undeveloped areas of urbanor periurbanland 4—12 m2 across urban centres worldwide, excluding travel for agriculture.Other supportiveactions include subsidising time to and from the garden (Cockram, 1993). In the US, inputsin thesmall quantities urbanfarmers need, encouraging irrigation was the biggest labour investmentin dry season or establishing composting systems for urban wastes and gardens. Information on returns on labour time was not providing technical support for gardeners. Even without availablefor manyThird Worldurban centres, but returns in government support, urban food production contributes USAgardens range from less than $ 1/hour to more than $13! substantially to the well-being of urban centres' poorest hour, depending on the value of the crop grown (Cleveland residents. Increased government approval ofagriculture could et al., 1985: 694—696). create a viable solution to the growing problem of urban poverty. Conclusion Urbanagriculture in theThird World isa means to combat References poverty and increase food security on an individuallevel. It Ash, R. (1981). "The Quest for Food Self-Sufficiency",in is a strategy used by poor urbanites to survive in harsh Christopher Howe (ed.) Shanghai: Revolution and economicconditions as it provides significantquantities of Development inan Asian Metropolis. Cambridge: Cambridge fresh food and incomefor householdmembers. For women UniversityPress. with limitedaccess to servicesor employment,gardens serve as a relatively reliable source of food and income that they Austin, J.E. (1980). Confronting Urban Malnutrition: The themselvescontrol. Urban gardens includea range of staple Designof Nutrition Programs. Washington, DC: WorldBank crops and vegetables where low-technology, low-input Occasional PaperNo. 28. gardens reflect limited access to land, capital and other resources. Government policies can either facilitate or Bjorklund, E.M. (1987). "Olericulture and Urban constrain urban food production. Development in China", Tijdschrift voor Economische en Given thewidespread interest in urbangardens by people Sociale GeografieVol. 78 No.1. aroundthe world,it ishoped that this review ofurban gardens will stimulate further research of and action on this 'rural' Brownrigg, L. (1985). Home Gardening in International activity in the urban setting. More research is needed on Development: Whatthe Literature'Shows. Washington, DC: several fronts since available information remains quite LIFE. limited. In particular, comparablestudies from different urban centresand countriesare neededto determine who cultivates, Centro de Educacióny Tecnologfa (CET) (1989). El Huerto why, and under what constraints production occurs. What Familiar Urbano. Valparaiso, Chile: Cetal Ediciones. farmer information and input needs are most limiting?How can agriculture use organic wastes produced in the urban Cheng Siok-Hwa (1976). "The State of Agriculture in environmentand avoid contamination of crops?What kinds Singapore", Journalof Economic Development and Social of policy proscription would best and least expensivelylead Change in Asia Vol.1 No. 1. to improvedsupport for the urbanpoor's gardens?What are the effects of land-use policies which encourage urban Cleveland, D.A. et al. (1985). "Economic Value of Home gardens? How can extension and other support for urban Vegetable Gardens in an Urban Desert Environment", gardens be structured so they are useful and an efficient HortScience Vol. 20 No. 4. expenditurefor the producer?What role can NGOsplay and how can governments work with NGOs to support urban Cockram,M. (1993). Agriculturein Third World Cities: The cultivation? How successful are garden projectsin community Invisible Food Supply. Ithaca, New York: Master's thesis, development? CornellUniversity. Urbanagriculture projectshave beenproposed to improve healthand nutrition,to increaseincome and topromote social Freeman,D.B. (1991). A City of Farmers: Informal Urban cohesion, but unless the community perceives the same Agriculturein the Open Spaces ofNairobi, Kenya. London: problems, they seldom achieve their goals. Successful McGill-Queen'sUniversity Press. projectsoften emphasisedlocal materials,crops and methods and provided sustainedtechnical assistanceto overcomethe Gutman, P. (1987). "Urban Agriculture: The Potential and inevitableproblems that arise. Projects requiringexpensive Limitationsof an Urban Self-Reliance Strategy", Food and or inaccessibleinputs seldom lasted after the donor left. Nutrition BulletinVol. 9 No. 2. Both national and urban governments can choose to increasethe efficiency of small agriculturists througha range Hake, A. (1977). AfricanMetropolis. New York: St Martin's of actions. Inexpensive and immediate options include Press. endorsements by prominentfigures and therepeal ofpunitive laws whichinhibit agriculture.Relaxing land-use and zoning ILEIA (1994). "Farmingat CloseQuarters", Special issue of legislation, especially in areas unsuited for construction, ILEIA Newsletterfor Low External Input and Sustainable would also encourage agriculture. More costly incentives AgricultureVol. 10 No. 4. include increasingthe size of houseplotsin housing projects

207 Mary Cockram and Shelley Feldman African Urban Quarterly

Kendall, S. (1990). "Rooftop Hydroponics in Bogota" Sanyal,B. (1984). UrbanAgriculture: A Strategyof Survival Development Forum Vol. 18 No. 3. in Zambia. Los Angeles: Ph.D. dissertation, University of California. Kolstrup, H. (1985). Urban Agriculture: Food Production and Land Use Planning in Maputo—A Project Proposal. Silk, D. (1982). "ThePotential of Urban Agriculture: Growing Aarhus: Master's thesis, School ofArchitecture. Vegetables—and Hope", Alternative/Appropriate Technologies inAgriculture Vol. 3 No. 2. La Rovere, E.L. (1985).Food and Energy inRio defaneiro: Provisioning the Poor. Paris: FoodEnergy Nexus Programme, Skinner, G.W. (1981). "Vegetable Supply and Marketing in United Nations University. Chinese Cities", in Donald L. Plucknett and Halsey L. Beemer, Jr. (eds) Vegetable Farming Systems in China. Ledogar,R.J. (1978). "Food and Survival in Lusaka's Self- Boulder:Westview Press. Help Townships",Carnets de l'enfance Vol. 43. Smit, J. (1980). "Urban and Metropolitan Agricultural Lowe, M.D. (1992). "City Limits", World Watch Vol. 5 No.1. Prospects",Habitat InternationalVol. 5 Nos. 3 and 4.

Lucenet, FP. (1988). Land for Housing the Poor Through Tempelman, G.J. and F.J.J. Suykerbouk (1983)."Agriculture UrbanAgriculture: The Case ofLusaka, Zambia. Cambridge in Singapore: Problems of Space and Productivity", Mass: Master's thesis, Massachusetts Institute ofTechnology. SingaporeJournal of Tropical Geography Vol. 4 No. 1.

Maxwell, D. and S. Zziwa(1990). UrbanAgriculture: A Case Thaman, R.R. (1977). "Urban Gardening in Papua New Study of Kampala. Kampala: Makerere Institute of Social Guinea and Fiji", pp. 146—168, in The Melanesian Research. Environment. Canberra: Australian NationalUniversity Press.

Nelson, J. (1978). "Pen-Urban Malnutrition: A Neglected United Nations(1991).World UrbanizationProspects 1990: Problem", Carnets de l'enfance Vol. 43. Estimates and Projectionsof Urban andRural Populations andof Urban Agglomerations. New York: United Nations. Niflez, V. (1984). Household Gardens: Theoretical Considerations on An Old Survival Strategy. Lima: Urban Resource Systems (1986). City Food: Crop Selection International PotatoCenter. in Third World Cities. Oakland: Platypus Press.

Niflez, V. (1985). "Working at Half-Potential: Constructive Vasey, D.E. (1985). "HouseholdGardens and Their Niche in Analysis of Home Garden Programmesin the Lima Slums Port Moresby, Papua New Guinea", Food and Nutrition With Suggestionsfor An AlternativeApproach", Food and BulletinVol. 7 No. 3. NutritionBulletin Vol. 7 No. 3. von Fleckenstein, E (1978). DooryardFood Gardens in Port Page,D. (1986). "Growing Hopein Santiago'sUrban Organic Moresby: An Original Study of Morata Together With a Gardens", GrassrootsDevelopment Vol. 10 No. 2. Comparisonof Other StudiesPast and Present. Occasional Paper, Economics Department, University of Papua New Panjwani, N. (1986). "Calcutta's Backyard:Food and Jobs Guinea.Photocopy. From GarbageFarms", Development:Seeds of Change Vol. 4. Wade, I. (1986). "Food, Transport, and Zoning", Development: Seeds of Change Vol. 4. Pottier,J. (1989). "Three's a Crowd: Knowledge, Ignorance, and Power in the Context of Urban Agriculturein Rwanda", Wade,I. (1987). "CommunityFood Productionin Cities of AfricaVol. 59 No. 4. the Developing Nations",Food andNutrition Bulletin Vol.9 No.2. Rakodi, C. (1985). "Self-Relianceor Survival? Food Production in African Cities, With Particular Reference to Wekwete,K. (1992). "Africa", in R. Stren, R. White and J. Zambia",African Urban Studies Vol. 21 No.1. Whitney (eds) Sustainable Cities: Urbanization and the Environment inInternational Perspective. Boulder:Westview Rakodi,C. (1988). "Urban Agriculture:Research Questions Press. and ZambianEvidence", Journal of ModernAfrican Studies Vol. 26 No. 3. Wong, C.T. (1983). "Land-Use in Agriculture",in T.N. Chiu and C.L. So (eds) A Geographyof Hong Kong.Hong Kong: Robson, E. (1989). "Growing Crops WithoutLand", World Oxford UniversityPress. Development Vol. 2 No. 6.

208 ( A Gurrnf Bibligrnphy n AFRIGAN AFFAIRS

Editor: Paula Boesch • Book Review Editor: Kwame Nantambu

EDITORIAL BOARD New Feature Section: new is a first-of-its-kind Bayo Adebowale Africa-On-Line, our section, guide Daniel Compagnon to helpthe African affairs community explore and interpret Frederick I. B. Kayanja the ofinformation options available on C. Tsehloane Keto exploding diversity intendsto Guy Martin the WorldWide Web.This new service provide Daniel G. Matthews essays, articles,bibliographical tools, directories and profiles Sibyl E. Moses otherrelevant telecommunication background information Serah Mwanycky to stimulate access to key resources on African affairs. Roger Pfister Peterj. Schraeder This timely reference source, although primarily aimed at both limited and unlimited Internet and WorldWide Web users, also will help those panning to go on line with the swarm of available information options. Africa-On-Line,in The-SteadyDrumbeat short,reflects Baywood'sserious and sustained commitment In a world where image Current on African Affairs on the the to keep A Bibliography is everything, steady information drumbeat of A Current "cutting edge" by providing access to vehicles Bibliography on African and eschewing the questionable distinction between Affairs has done quite academic and popular offerings. well in a new attracting Aims and Scopes: generation of users.The pages of this journal To present a thoughtful selection of references to a (begun before many of considerable diversity of materials, not only in subject and nations in Africa today's geographical selection,but also to pointthe user toward an became independent) increasing range ofresources focusing on a sociotechnical offera compelling view of Africana from the mid to multiethnic approachto African affairs. late 20th century. For all those interested in Africa,A Current Bibliographyon Journal African Affairs provides a current, comprehensive and features: annotated bibliography. Our source materials are both Commentary African and international. The Bibliographical Section is AfricaOn-Line divided intotwo parts—Generaland Regional. Headings in these areas will to accommodate current interests. Features adjust The scope of the Journal will continue to grow reflecting Book Reviews the main viewpoints that exist for given topics. BibliographicalSection CumulativeAuthor Subscription Information: Index Price pervolume—4 issues yearly ISSN: 0011-3255 Institutional Rate: Regional Index $159.00 Postage & handling: $6.50 U.S. & Canada, $11 .75 elsewhere Complimentary sample issue available upon request

Baywood Publishing Company, Inc. 26 Austin Avenue,Amityville, NY 11701 Call (5 16) 691-1270 Fax (516) 691-1770 Call ToIl-free (800) 638-7819 e-mail: [email protected] . web site: http://baywood.com AfricanUrban Quarterly, 1996: 11(2 and 3) pp. 210—216 ©1996 by African Urban Quarterly Ltd.

OPEN SPACE CULTIVATIONIN ZIMBABWE: A CASE STUDY OF GREATER HARARE, ZIMBABWE

Davison J. Gumbo and Takawira W. Ndiripo ENDA-Zimbabwe, p. 0. Box 3492, Harare, Zimbabwe;Fax: 263-4-301-156

Abstract Accepted September, 1995 This article characterises the recent evolution ofopen-space cultivation in the City of Harare, Zimbabwe. Analysesare based on air-photographic surveysin 1990and 1994and afield survey including questionnairesapplied in 1994to a large sampleoffarming householdsin sectors where open-spacecultivation is concentrated.The cultivated acreage nearly doubled over theperiod, more of thefarmed areas in 1994being associatedwith the more denselypopulated neighbourhoods, most of which are low-income areas. The study discusses the probable causes of this growth; its household survey indicates that economic factors do prevail. Also, homeowners are more likely than tenantsor others to secure access to open spaces where to grow crops. Women predominate among producers and harvests mostly go to household subsistence. The study warns about degradation risks associated with the inappropriatefarming of vulnerable land. Insecure land tenure deters most open-spaceproducers from adopting conservationistpractices. Changesare recommended to currentlegislation and planning process which should allow urbandevelopment to incorporateopen-space cultivation as a transitional or mixed land-use, while also providing producers with sufficienttenure stability and encouraginga labour intensive type ofmarket-oriented production.

Résumé Cet article caractérise I 'evolution récente de Ia culture des espaces libres dans Ia yule de Harare au Zimbabwe. Les analysessefondent sur I 'interpretationde couvertures aéro-photographiquesde 1990et 1994 etle traitementde questionnaires administrés en 1994 a un vaste de ménagesproducteurs dans les principaux secteurs cultivés de Ia ville. Les croissante entre ces aires et les quartiers résidentiels plusdensément occupés, la majorité revenus. L'étude disserte auteurs notent un quasirdédoublement des aires activéesdurant Iapériode, uneassociation sur les causesprobables de cette croissance, l'enquête indiquant que les mobilesd'ordre dominent. Ce sont lespropriétaires de residences qui ont le plus aisément accés aux espaces libres pour les cultiver On note la participation prédominante de Ia femme et l'autoconsomtnationde Ia majeurepartie des récoltes. L'étude alerte sur les risques d'erosion que pose Ia mise en culture inadequatede terres susceptibles.Les auteurs soulignent le statut precaire d'occupation qui decourage 1'adoption par les producteursde mesuresconservationnistes. Ony recommande des changements au niveau législatifet de Iaplanification qui permettraientsu développementde Ia ville d'y incorporer et reproduire Ia culture des espaces libres comme utilisationdu sol transitionnelleou mixte, touren assurant un minimum de stabilité auxproducteurs ety stimulant une production de marché qui soit travail-intensive.

Introduction properties (on-plot activities) to cultivation of a variety of The City of Harare considers itself to be the 'Sunshine foodcrops inopen spaces around Harare (off-plot cultivation). City', a city in which the cultivation of open spaces was Urban agriculture occurringin open spaces (off-plot)is either considered to be repugnant.This positionhas been bolstered illegal or semi-legal. For themajority of thecultivators, urban over the years by legislative framework that forbade agriculture is illegal since they merely cultivate any open cultivationwithin the Greater Harare, Zimbabwe. However, space without any permission from any authority. For the contrary to the 'tradition' and the stringentenforcement, there few urban agriculture groups or cooperatives, urban has been a gradual increase in the cultivationof open spaces agriculture is semi-legal because these practitioners are in Harare over the recent years.Whereas there was only4,822 allocated land by the local authorities.This form of urban ha cultivatedin 1990, this figurerose to 9,289 ha in 1994,an agriculture is semi-legal because although permission to increase of 92.6percent (ENDA-Zimbabwe, 1994). That such cultivatethe landis given by thelocal authorities, nostatutory an increase has occurred against the backdropof toughlaw amendmentshave been made to the land-use regulations. enforcementmechanisms signals the existence of somepush Theextent, natureand intensityof urbancrop cultivation factors on the householdsthat are involved. has some importantramifications on the economiesof both Urban agriculturemay be describedas the use of urban householdsand urbanauthorities. Most importantly, however, land for agricultural purposes. This occurs most often on it has some implicationson the natural environment in the undeveloped land that may be designatedor zoned for uses given urbanareas. Thus, the subjectof open spacecultivation otherthan agriculture.Urban agriculture practicesrange from in Harare shouldbe given greaterattention, especially in the vegetable and small stock production around residential fields of policy and physical planning.

210 Volume 11 Nos. 2 and 3, May andAugust, 1996 Open Space Cultivation:A Case Study ofGreater Harare

Although the land-use laws and regulations governing household survival strategy, the potential harm it has on the urban land-use have changed a little since 1980, there has environment needsto be explored.Thus, it is imperativewe been a growing culture of tolerance by the local authorities critically analyseurban agriculture in Harare inorder tocome allowing the practice to go on unhindered. There are up with comprehensiveactions and policies that enhancethe humanitarian, political and economic reasons for this. Of health and safety of the urban environmentand lead to the much significance are the woes of the late eighties and establishment of sustainable urban centres in which all nineties. The effects of persistent droughts and the harsh inhabitantsstand to benefit. realities of the EconomicStructural Adjustment Programme (ESAP) have exacerbatedthe plight of the urban household Urban Agriculture in Harare in Harare and in particular the poor Harare has a history of sustainable urban agriculture which predates independence. Work carried out by SomeAspects of Urban Agriculture Mazambani showed this trend. In 1955 some 267 ha were Urban agriculture within Third World urban centres is cultivatedand this went up to 4,762 ha in 1980 (Mazambani, now a reality. It includes all forms of agricultural activities I 982a and 1982b: 134—138). A surveycarried out by ENDA- and is carried out in basements, roof-tops,backyards, way- Zimbabwe(1994) showed a dramatic increase of 92.6percent leaves and open spaces, among others (Smit and Nasr, 1992: of the open spacesarea undercultivation in Harare from 1990 14 1—152and van der Bliek, 1992). While urban agriculture to 1994. The table 1 shows the changes in land under could be said to involve all income categories,this activity a survival for the urban Factors provides key strategy poor. Table 1, Changes inArea under Urban Cultivation inHarare between promotingthe urbanagriculture growth include deteriorating 1990and 1994 national policies, such as (removalof subsidies,low wages, rising cost of living) unemployment, among others Actual (Mosha,1991).The same applies to Greater Harare. Area cultiv ated increase Percentage Urban agriculturein Harare dates to as far back as 1985 Land use zone 1990 1994 (ha) increase (Mazambani, l982a and 1982h:134—138).However, prior to Lowand medium 1.588 3,376 1,788 112.6 Zimbabwe'sindependence in 1980, local authorities inHarare High density 2,391 5,323 2.932 122.6 tendedto bestrict in the urban Industrial areas 843 590 (253) (30.0) enforcing planningregulations Total 4,822 9,289 4,467 92.6 that forbade urban agriculture.There were strict controlsto Source: Compiled b y the authors curb urbancultivation, particularly of open spaces.The main driving force for the increase in the practice of urban in Harare is economic. Most of economic cultivationbetween 1990 and 1994(Figures 1 and 2). agriculture Since 1990 the land undercultivation in residential areas to be rooted in the ESAP, one hardships appear squarely has more than doubled. The locationsin Harare whosemajor objective was to improvethe quality of life of high density the low income and in Zimbabwe. demonstratethe highest increase of 123 percentor 66 percent disadvantaged groups of the overall increasein Harare. It is to note that ESAP has had some For the has important impacts. instance, country for the suburbs showed a witnessedsome fundamental in the Cost high density sampled,they density changes economy. of 2,391 ha for 1990 while Mazambani's (l982a) data for of has risen, with a effect on the living sharply devastating 1980 was 4,762 ha. The differences could be due to the fact urbanlow incomehouseholds and the due to theremoval poor that 1990 data were derived mainly from photographstaken of subsidiesfrom basic cost measures commodities; recovery in the dry season. Thus, some vital information may have instituted in health and education; real wages have fallen been lost. largelydue to inflationand worsenedthe position of the low While there has been a notable increase of area under incomegroups in Zimbabwe;formal sector employment has cultivationin the residential areas, the industrial areas have also fallensince the beginningof ESAP and as aconsequence experienced a significantreduction. A possible reason for of this, low incomehouseholds have startedto engage in urban this could bethat industriesare located away from residential agriculture.Urban agriculture should be considered as one areas so that cultivatorshave to travellong distancesto such of the strategies for combating food shortages because it areas. Harare is estimatedto cover a land area of 55,558 ha requireslittle investmentcapital and provides an alternative and thearea underurban agriculture has risen from 8.8 percent sourceof cheap food (in the face ofmounting prices of basic in 1990 to 16.7 percent in 1994. commoditiesand sourceof directhousehold income through It can be seen that areas with high population sale of surplusproducts or indirect throughsavings made by concentrationssuch as the high density suburbs tend to have consumingown produce). thelargest area under crop cultivation. Most of the low income From the above, it will be realised that urban agriculture earners are also found in these areas. However,just a small has the potentialof playing a crucial role in theeconomies of areaclose toindustrial sites is cultivated.This may be due to householdsengaging in it. Harare has notbeen an exception. lack of surroundingopen space or arestrictive long distance However,the envisaged benefitsof urban agriculture should from the residentialareas. On theother hand, there has been be balancedagainst the impact this activity has on theoverall an upward trend in the increaseof area under cultivationin urban environment. Much as it can be argued that this is a the residential areas.As more peopleget into theactivity and

211 HARARE Open Space in Cultivation

I Source: IDRC project 93-0024 O51 2 3 km. Digital CartographyDH. Douglas HARARE Open Space

i Source: DRC project 93-0024 Digitat CartographyD.H. Douglas Davison I. Gumboand Takawira W. Ndiripo African Urban Quarterly thebuilt environment continues to expand, landfor cultivation due to slope cultivation, the health standards of Harare will be scarce very soon. There is, therefore, need to residents are alsolikely to be affected as siltationand pollution rationalise the use of urbanspace. aremost likely to affect the majorrivers such as theMukuvisi, which supply water to Lake Chivero, the city's main source Natureand Characterof UrbanAgriculture Practitioners of water. The majorreasons why peopleengage in urban agriculture In all cases, open spaces indicated as being under is mainly economic. In Harare, this activity tends to centre cultivationabove have theirown potential negativeimpacts. on low income earners and the majority of these reside in For instance cultivating near: road/railway causes high density suburbs as indicated in table 2. obstructions to such thoroughfares;vleis/streams leads to erosion, siltation and eutrophication of water bodies; and and other causes erosion.All forms of Table 2, ofHouseholds with HouseholdAccess to Land hilislopes steep slopes Proportion erosion from urban loss trees and for UrbanAgriculture by HomeOwnership and ResidentialDensity agriculture lead to the of As % Sample this can be harmful to the environment. Low Medium High Home ownership density density density Total Crops Grownby Households Landlords 2.6 1.8 56.6 61.0 As already indicated, households that are involved in Lodgers 1.4 0.5 25.2 27.1 urban agriculture come from all categories.Crops grown by Servants 2.9 0.3 1.3 4.5 these households and as shown in table 4, the main Relative 0.6 0.2 6.6 7.4 vary crop (of landlords) is maize. TOTAL 7.5 2.8 89.7 100.0 N= 1,784 Source:Compiled by the authors Table 4,Distribution ofHouseholds Growing SpecificCrops in Off- PlotFields by Residential Density The high density areas have 89.7 percent of all urban Proportion grow ing by density agriculture activities and in these areas urban agriculture is Low Medium High Total carried out by landlords (homeowners),accounting for 56.6 Crop (%) (%) (%) (%) percent of allurban agriculture activities followed by lodgers Maize 3.84 1.47 44.67 49.98 (tenants) (25.2 percent) and relatives of homeowners (6.6 Groundnuts 0.81 0.03 6.76 7.60 Homeownersdominate urban activities Rice 0.03 0.69 1.11 1.83 percent). agriculture 18.74 because the theirresidence in area. Sweet potatoes 2.25 0.66 15.83 of permanenceof a given Sweet sorghum 0.12 0.06 4.93 5.11 They,therefore, are better able to establish 'permanentland Beans 0.33 0.00 4.27 4.60 claims' in open spaces for urban agriculture. Servants and Water melons 0.00 0.00 0.48 0.48 lodgers who tend to change residences often reduced their Pumpkins 0.57 0.00 9.25 9.82 accessto land for urban In the low areas Roundnuts 0.39 0.00 1.29 1.68 agriculture. density 0.00 0.00 0.15 0.15 servants tend to cultivate more than their Cucumbers open spaces TOTAL 8.35 2.91 88.74 100.00 employers (homeowners) because the homeowners have Source:ENDA-Zimba 1994:16 relatively higherincomes and are, therefore, not into as much bwe, open space cultivationas their counterpartsin high density areas. Maize is the main crop grown by at least 49.98 percent of the participants.This is mainly grown by high density Location of Open-Space Fieldsand Extent of Cultivation residents.Sweet potato cultivationcomes second. These two Table 3 demonstratesthe type of open spaces that were crops form the staple food crops of most urban low-income under cultivationin 1994 and their specifiedlocations. households. Most of the cultivation occurs near roads and on hill The survey also revealed that access to open spaces for slopes.It has been found that streambank,swamp and vleis crop cultivationis directlyrelated to homeownershipand or cultivation isrampant in such high density suburbsas Mbare, access to ahouse. Those who own houses have greater access Sunningdale, Highfield, Budiriro,Mabvuku, Dzivarasekwa tosurrounding open spaces. Outof 1,784 sampledhouseholds and Kuwadzana.This practice,if uncontrolled, imposes great engagedin off-plotcultivation 61 percent are landlords and risks of siltationand eutrophication. Besidesland degradation most of them (56.6 percent)are in the high density suburbs. Lodgers also demonstratea relatively higher access to off- Table 3, LandTypes and Extentof Cultivation plots (27.1 percent), whereas servantsand relatives have the least access to open spaces. Location offield Uses ofspaces s % Table 5 shows the participation of family members in Road sides 37 cultivation. Women most dominant Streambanks 9 open space are the Railway line way leave 4 participantsin open spacecultivation. This is consistent with Onhills 31 other work that depicts the dominanceof women in urban On swamps/vleis 19 agriculture. TOTAL 100 A number of reasons explain why women dominate in Source;ENDA-Zimbabwe, 1994: 13 urban agriculture. Most men are employed elsewhere and

214 Volume 11 Nos. 2and 3, May and August, 1996 Open Space Cultivation:A Case Studyof Greater Harare

statistical information to the need to and Table 5,Participation ofHou sehoId Members in Off-Plot Activities points accept by Residential Density (As a % of Households Engaged in the promotefood production as acritical urban function. Official Activity) support ofurban food production would ensurecheap supplies of sufficient and nutritiousfood for the of the urban Suburbdensity majority Type of poor. participants Low Medium High Total Mother 3.4 1.5 39.7 44.6 The Practice of Open Space Cultivation and Its Impact Daughter 0.5 0.3 12.9 13.7 on the Urban Environment Father 1.5 0.7 13.8 16.0 Son 0.2 0.2 20.4 20.8 and Hired labour 0.8 04 3.7 4.9 (a) Technology Inputs Open space cultivationin Harare is largely done by the TOTAL 6.4 3.1 90.5 100.0 use of hand tools (hoes). ENDA-Zimbabwe(1994) survey N = 2,754 results show that 20 percent of the sampled households use ENDA-Zimbabwe, 1994 fires for land preparation.The use of fire has adverse effects on the urban environment and in particularit destroys organic cannot find time for urbanagriculture. Furthermore, women matter and this reduces soil fertility. Besides the rislcof have traditionally tended to be engaged in food production spreading to other land-use zones and air pollution, fire for householdconsumption. Results of theENDA-Zimbabwe destroys the vegetation and exposes the soil to erosion. (1994) study showed that 62 percent of urban agricultural However, destruction of the vegetation results in potash practitioners were unemployed and the bulk of them are (fertiliser) needed to improve soil fertility. women. This is because unemployed womenhave more time Of the sampledpractitioners 88.4 percent use chemical at their disposal and so can be engagedin urban agricultural fertilisers,2.1 percentorganic fertilisersand 9.5 percent use activities. A relationship was also observed between size of both. Most Harare open space cultivatorsprefer the use of householdand its participation in urban agriculture.Large chemicalfertilisers over compostto boost production levels. households with 5—9 members participated in urban Furthermore, 2.04 percent of the sample use pesticides. agriculture most. These same householdsalso tended to fall Dimethoateand somepowders (Malathion dust) arethe main thelow middle Given this into to incomegroups. background, pesticides applied. These fertilisers and pesticides have it can be that involvementin urban is concluded agriculture significantimpact on the urbanenvironment, especially the to householdincomes. basically supplement water bodies. It is important that institutions and policy- and as if to assert the fact that urban Interesting enough makers take due regardof these inputs in creatingan enabling is for low income these data agriculture mainly households, environment that supports the livelihood of the urban show little use of hired labour across all areas. very householdsand the safety and health of the city's residents at Production Levels in 1994 large. Crop On the other it has been shown that Table 6 demonstratesthat urban households hand, open space many rely cultivation occurs on hill road and rail reserves, more on maize cultivation for family consumption slopes, results show that 15 of swampsand stream banks.Most of these cultivated areas do requirements.Survey only percent not have mechanical conservation measures to control maize harvested was sold in 1993. Maize sales are expected any tohave increasedin the atleast 5 soil erosion. As most of the cultivated crops take very long subsequentyears by percent is per annum. As participants realise the accrued income to providean adequatesoil cover,erosion inevitable. benefits, it is most likely that they intensify their activity in order to gain more. (b) Accessto Land for Cultivation Theestimated total harvestedcrops emphasisethe greater All forms of open space cultivation in Harare may be reliance of urban householdson such crops as maize, sweet viewed as illegal as there are no clear (if any) statutory potatoes, beans and groundnutsfor food security, incomeand provisionsthat legalise the activity. Consequently,there are cash saving, especially in the high density areas. All the no formal means by which land is allocated to selected

Table 6, Estimated TotalHarvested Crops Across the Three Residential Sectors for the 1994 Harvest Season High density Mediumdensity Low density Major crops Produced/Consumed Produced/Consumed Produced/Consumed Maize 3,994.7 2,557.8 46 46 383.1 128.3 Roundnuts 291.8 291.8 — — 0.2 0.2 Groundnuts 103.5 80.1 — — 16.4 7 — Rice 17.3 16.3 — 1 1 S/potatoes 533.9 408.6 2 2 43.7 16.7 (buckets) Beans 777.9 777.9 — — 7.4 7.4 Source: Compiledby the authors

215 DavisonJ. Gumboand Takawira W. Ndiripo African Urban Quarterly individuals.Land acquisition is mainly by survival of the accommodative and supportive.Planning authorities should fittest. Under these circumstances, cultivators are not incorporate urbanagriculture as aland-use activity that needs guaranteedaccess to the same piece of land in successive tobe planned for and managed like any other land-useactivity. seasons. Planners must consider the drafting of a cultivator's lease Even though some local authorities have 'informally' withthe minimum conditionsset to ensure sustainabilityand allocated cultivation land (which may be due to intense environmental protection.The grantingof the lease willgive pressure by urban households to engage in the activity), urban agriculture practitioners a sense of security of tenure securityof tenureis notguaranteed to the cultivatorsas there and this will ensurethe futureof urbanagriculture within the are no legal contracts specifyingtheir tenure with the local urban economy in terms of productivity. Legalising urban authorities. An insecureland tenure system does notprovide agriculture could pave the way for the provision of support sufficient incentives for the cultivators to invest in services, including extensionand trainingservices and credit conservation and other land protectionstructures. facilitiesthrough the relevantinstitutions. To revitaliseurban Facedwith sucha situation, what then can be done to the economies, urban agriculture should be commercialisedto practice of off-plot cultivationin Harare in order to sustain ensure that it becomes labour dynamic. This could help to the economy of the city at large? lowerthe increasing levels of unemployment that has become part and parcel of the Zimbabweaneconomy. Once urban Conclusion agriculture takes shape within the urban economy,there is Policy-makers and urban planners should formulate need to set up a multidisciplinary environmental impact strategies that are geared towards sustainable economic monitoring committee which will be responsible for developmentof urban centres. Aholistic approach has to be monitoring and evaluating urban agricultural activities. As adopted, which looks at the locational and functional the City of Harare continues to grow and the country's dimension of open space cultivation both as a land use and economic problemscontinue, it is highly unlikelythat urban within the context of the urban economy. This should be agriculture can be halted by mere policy pronouncements.A substantiated with necessary legislationchanges, particularly deliberate policy of supporting it will be essential so that it the Regional,Town and Country PlanningAct of 1976 and becomesenvironmentally friendly. alliedActs which have colonialundertones and are generally hostile to urban informal sector activities(particularly land- References use activitiesdeviating from rigid plans). This is mainly due ENDA-Zimbabwe (1994). Urban Agriculture in Zimbabwe to the fact that urban agriculture is a reality. Although and Its Impact on the Environment: A Surveyof the City of forbidden, urbanagriculture will continue. It is interestingto Harare. Harare: ENDA-Zimbabwe. notethat 76 percentof urbanagriculturalists interviewed were knowledgeableand aware of Council bylaws. These bylaws Mazambani D. (1982a)."Aspects of Pen-Urban Cultivation are not likely to abide withinthe present circumstances. and DeforestationAround Salisbury 1955—1981". Harare: Urban agriculturein Harare is here tostay becauseof the MA Thesis, Geography Department. University of important rolethat it continuesto play in the urbanhousehold Zimbabwe. economyin the face of worsening economichardships. Yet in Harare, it takes many years between land acquisition, Mazambani,D. (1982b). "Pen-UrbanCultivation in Greater servicing and ultimate construction. With appropriate Harare", The Zimbabawe ScienceNews Vol. 16 No. 6. planning, such land could legallybe madeavailable for urban agriculture during the period between acquisition and MoshaA.C. (1991). "Urban FarmingPractices inTanzania". construction. After all, it will still be cultivated illegally, Review ofRural and UrbanPlanning in Southern andEastern exposing it to possible degradation which could make final Africa. construction expensive. Thesignificance and prospectsof urbanagriculture have Smit J. and J. Nasr(1992). "UrbanAgriculture for Sustainable been narratedand a number of recommendationshave been Cities: Using Wastes and Idle Land and Water Bodies as drawn. What remains is the role politicians and planners Resources",Environment and Urbanisation Vol. 4 No. 2. should play to ensure the provision and promotion of sustainable agriculturalpractices into urbanland-use plans. van derBliek J. (1992). UrbanAgriculture —Possibilities for There is need for central and local governmentauthorities to EcologicalAgriculture in UrbanEnvironments as aStrategy createan enabling legislation that supportsurban agriculture. for Sustainable Cities. Leusden, Netherlands: ETC This will ensure urban planning policies that are Foundation.

216 g ri cultural

ubscripior S2UU for dvduaIs S43 00 for institutions: $15 00 for S udents I d $ 00 postage for foreign orders) University of CaUfornia Press 2120 Berkeky Way Brkeley CA 94720 Edtor: Morton Rothstein, Ag cuItura hs a y Canter Unversity of California, Davis CA 95616 African Urban Quarterly, 1996: 11 (2 and 3) pp. 218—226 ©1996 by African Urban QuarterlyLtd.

ANIMAL FARMING IN AFRICAN CITIES

Ann Waters-Bayer ETC Foundation, p. 0. Box 64, NL-3830AB Leusden,Netherlands Address for correspondence:Rohnsweg 56, D-37085 Gottingen,Germany; Tel: +49-551-485751, Fax: +49-551-47948 Email: [email protected]

Abstract Accepted September, 1995 in this article, attention is drawn to the importanceand recent growthof livestock rearing insideAfrican urban centres. Ways of classifying the existing production systems are suggested. The functions of animal farming for different income groups and the opportunitiesthat livestockrearing offers forurban economies are outlined. Problemscaused by the keeping of livestock in the urban centre and major constraintsfaced by the livestock rearers are discussed. Finally, proposals are made for research and developmentaction to address these constraints and to grasp the opportunities offered by animal farming in urban centres, particularly with respect to recyclingof urban wastes. Résumè Cetarticle confirmeI 'importanceet Ia croissancerécente de 1'elevage de detail dans les villes africaines. Onpropose des alternatives de catégorisation pour les systémes de production courants. L'auteur note lesfonctions que 1'activité remplit auprés des produteurs, lesquellesvarient se/on leurniveau de revenus, ainsi que les opportunitésqui s'offrent aux villespour accroItre I 'apport de I 'élevage urbain a leur On y propose enfin des lignes de recherche et d'action pour le développement, nécessairesa Ia resolutionde contraintes eta / 'optimisationd 'opportunités identzfiées, notammentau niveau du recyclagedes déchets urbains.

Background 'modern' production ofmeat, milk and eggs by transnational Keeping animals in urban centres is not a recent or domestically-owned companies located on the outskirts phenomenon. Throughout history, urban dwellers in both of large urban centres. Governmentincentives included tax northern and southern countries have depended on their exemption,low-interest credit, subsidies on inputs and/or livestockfor security and survival, especially during periods outputs, liberal policy concerning imports of equipment, of political unrest. Duringwars inprevious centuries, people machinery, exotic breeding stock and feeds. Technical sought safetyfor themselvesand their animalsbehind urban assistance of donor countries and sales efforts of centre walls, suchas in Kano in northernNigeria. But also in agroindustries were geared to these high-input systems the decades since independence,as immigrantshave set up (Krostitz, 1984: 17—23). satellitesettlements beyondthe old wallsof theurban centres Recent economic changes have put these systems under and as urban growth has engulfed the surrounding villages, considerable pressure. Large-scale intensive livestock the open spaces between the many quarters which make up production in periurbanareas is declining in many countries most Africanurban centresare commonlyused for livestock whichare struggling withStructural AdjustmentProgrammes, rearing (figure 1). especiallyafter devaluation in the francophonecountries of In less than 20 years, the urban populationin developing Africa. The costs of importedinputs have skyrocketedand countries will have doubled,at current growthrates (Carty, the luxury urban marketsserved by the 'modern' enterprises 1991: 2—4).Worldwide, urban dwellers will outnumber rural have dwindledas aresult of recession. Consumersare buying dwellersby the year 2000 (Sachs, 1986: 2—11). Many urban products from livestock reared on low-cost local resources planners—andalso many livestockdevelopment planners— and sold through unofficial channels, often directly from regard urban livestock keeping as a transitoryphenomenon producerto consumer rather than buyingthe more expensive in developing countries. However, there is no sign that products from dairy plants or certified butchers. livestock and other forms of farming decrease when urban In the meantime, despite government bans, livestock centres grow larger and people reside longer in the urban rearing in the urban centres is being developed by micro- centres. Indeed, studies in Africa suggest quite the contrary entrepreneursthrough processes of indigenousdevelopment, (e.g. Sawio, 1993). without external support. With rapid urbanisation and In the colonial period, livestock were banned from most deterioration of economic conditions,a growing number of urban centres in Africa. After independence,the bans were urban families are raising small animals suchas poultry, goats, usually retained, while government support was given to sheep, pigs and rabbits. Some also keep a few dairy cattle.

Acknowledgements—The authorthanks Jean MichelCentres, Groupe de Recherche et d'EchangeTechnologiques (GRET), France;Davison Gumbo, Environmentand DevelopmentActivities Zimbabwe (ENDA-Zimbabwe); Luc Mougeot, International DevelopmentResearch Centre (IDRC),Canada; and Jac Smit,The Urban AgricultureNetwork, USA, for generouslymaking reportsand informationavailable for this review; and WolfgangBayer and Gerrit Zemmelinkfor their constructivecomments on thetext.

218 Ce

Fig. 1 SITES OF PERI—URBAN, INTER—URBAN AND INTRA—URBAN LIVESTOCK REARING IN A TYPICAL AFRICAN CITY. KEY City Wall

Built up urban area I ______Rood 4 4 Railway River Mkt. Market place

Cd ('1

Ce C

CCd

'0 C 0 .0 C — Ce C C 0 Ce

Cd 0 C

Ce Ann Waters-Bayer African Urban Quarterly

Development agencies and government officials are only demands on their services: the pen-urban commercial beginning to recognisewhat is happening,even though they enterprises of large to medium size situated on the edge of themselves may live in the same urban centre. urban centres, operated by the government, parastatal organisations or privatefirms and specialisedin poultry,egg, Growthin Livestock Numbers beef, pork or dairy production. Only in recent years has urbanagriculture received much The formal livestock services generally have very little attention from research and the focus has been on croprather knowledge of the small-scale animal fanning activitiesinside than animal farming.Animals in Africanurban centreshave urban centre quarters (intra-urban or core-area livestock been largely overlooked not only because they are not rearing)and between quartersof the same urbancentre (inter- supposedto be there but also because they are mobile, they urban or corridor livestock rearing). People who raise use land only temporarilyand, especiallyin thecase of small animalsin the urbancentre can be roughlydivided into 'on- animals kept in the home or backyard,they are not readily plot' livestock rearers who use private residential space in visible. Therefore, only few research data are available. backyards, insidethe home,or on balconies or rooftopsinside However, these data alreadyindicate considerablegrowth in the residential areas of the urban centre; and 'off-plot' livestocknumbers in African urban centres in recent years. livestock rearers who use unoccupied private or state land In Dar es Salaam, Tanzania for example, the chicken suchas unbuiltplots, hospital or school grounds, public parks, population increased from about 500,000 birds in 1985 to wetlandsand streamsides unsuitablefor building and strips 800,000 in 1989 while the number of pigs, goats and dairy along transport and utilities (telephone, electricity) lines, cattle doubled(Mougeot, 1994). Yet the fresh milkprocessed often in the corridorsbetween the residential areas. by the parastatalTanzania Dairy Limitedplunged from more than 14 million litres in 1979to less than 5 million litres in (a) On-Plot 1989 (Ngigwana, 1992: 115—122). Likewise, small-scale Livestockkept on-plot are often enclosedby afence, wall dairy productionis increasingin and aroundNigerian urban or cage or are tethered and feed and water are brought to centres, hut less than 10 percent of the milk enters official them. Larger animals may be allowed to graze part of the market channels (Walshe et al., 1991). Also inAddis Ababa, day, at least seasonally. Besideshomeowners, also employees it was found that most dairy producers are sellingdirectly to living on institutional compounds such as hospitals and individual consumers (Debrah, 1992: 257—268). The informal schoolskeep animals at theirplace of residence. Forexample, ade in unprocessed milk in Nairobi, Kenya has also 81 percent of staff living on the university campus in Zaria increased in the last few years(Staal and Shapiro, 1994: 533— in northern Nigeria keep livestockthere, mainly poultry and 549). The informal sector of animal farmingand marketing small ruminants (Gefu, 1992: 295—302). A particularly of animal produce in major African urban centres is widespread system of on-plot animal farming is practised in flourishing. Otherquantitative studies made inAfrican cities West and North African urban centres, where stall-kept or hac revealedthat more than 20,000 households in Bamako tetheredsheep are fattened for Muslim festivals. These are rear animals in the city, and thousands more derive income very labour-intensive production systems using high from supplying them with production inputs or marketing quantities of local, usually purchased, feed. They are, services (Centres, 1991); over one-third of households nevertheless, profitable on account of the extremely high surveyed in Harare rear animals in the city, mainlychickens prices for fattenedrams (Centres, 1991). hut also rabbits, pigeons, ducks and turkeys (ENDA- Zimbabwe, 1994); 51 percent of households in six major (b) Off-Plot Kenyan urban centres own livestock and 17 percent keepthem Livestock kept off-plot tend to be grazing animals such in urban centres (Lee-Smith and Memon, 1994:67—84); as sheep, goats, pigs, cattle, buffalo and donkeys. They are official 1990 records for Nairobi show about 25,000cattle, herded, tethered, hobbled or allowed to roam freely on land 30,000 small ruminants, 30,000 pigs, 8,500 rabbits and which isused with or withoutthe landowner'sconsent. Some 350,000 poultry (van der Bliek, 1992); in Dar es Salaam, herds are brought in by pastoralistsfrom outside the urban urban farming is the second largest source of employment centre in order to graze the urban lowlands during the dry after petty trade and labour, and 74 percentof urban farmers season (Mbiba, 1994: 22—25). Many of the off-plot grazing keep animals (Mougeot, 1994); and even in very densely animals belong to landless families: eitherlong-term squatters populated Cairo, Egypt 5 percent of the households rear in densely-populated slum settlements,or pastoralists who animals, especially chickens and pigeons; about 15 percent settle temporarily on vacant land on the outskirts of urban of eggs consumed in the city are home-produced (Khouri- centresor beside traffic routes in the urban centre. Dagher, 1987). These data are eitherofficial statistics or the In somecases, people with only afew animals eachjointly information that researchers couldobtain from urbandwellers hire a herder to care for their animals. This is done, for who may not havedared tobe fully open abouttheir (officially example, by some dairy-cattle owners in Bamako. The illegal) activities. The number of animals actually kept in animalsare kept togetheralso at night: the corrals are within these African urban centres is probablymuch higher. the urban centre limits on sites which are designatedby the authoritiesand can be shifted as demand for housing grows. Classification of Livestock Systems In additionto daily grazing,the animalsare fed very limited The livestock production units best known to people supplements bought by the owners (Achuonjeiand Debrah working in governmentservices are those which make most 1992: 223—238).

220 Volume 11 Nos. 2 and 3, May andAugust, 1996 AnimalFarming in African Cities

In urban centres in northernand central Nigeria.off-plot market:expatriates, supermarkets and restaurants. Likewise, dairy production is practised by family units operating individuals raising large ruminants in urban residential independently from each other: either settled Fulani compoundsare middle- and high-incomeearners (Centres, pastoralistswith small herds or more mobileFulani who keep 1991; Mbiba, 1993; Egziabher, 1994: 85—104; ENDA- only their milking cows in the urban centre. The cows are Zimbabwe, 1994; Lee-Smithand Memon, 1994: 67—84 and grazed on unoccupiedland in and near an urbancentre and Sawio, 1994: 25—46). Two-thirds of all animals kept in Dar are fed somepurchased supplements such as groundnut cakes es Salaamare found in the largegardens in formerly colonial, and crop residues. The Fulani women process the milk into now elitist Africanresidential areas (Mougeot, 1994). fermented products and butter and selldirectly toconsumers. Particularly businessmen and civil servants, active or The manure is sold as fertiliser to nearby gardeners and retired,have diversifiedinto livestockrearing. They depend farmers. In the case of the Fulani who keep only the milking whollyor partly on hired labour. Many are inexperiencedin cows in the urban centre, the men take the rest of the herd to animal husbandry as may be their employees as well. The graze further afield, where they set up a second iemporary animals are fed mainly with purchased roughage and camp. A similar system,involving small dairy herds of about concentrates: crop residues and agroindustrialby-products five cows each is reported from Mogadishu, Somalia such as brewery waste. These are the private producers who (Giorgettiet al., 1983: 372—380). Thus, the relatively scarce take advantage of extension and veterinary services. Their feed and space inside the urban centre are used for the most social links also give them access to exotic breeds, either productiveanimals. imported or obtained from large state farms (Laurent and The above classificationof urban livestock systems is Centres, 1990; Centres, 1991 and 1992; Achuonjei and madeprimarily according tolocation: periurban versus urban, Debrah, 1992: 223—238 and Nell, 1992: 31—44). on-plot versus off-plot.There are variousother possibilities Much more experienced are the on-plot livestock rearers of classificationsuch as according to mainproduction aim: in urban centres with a long tradition of urban farming. For commercial, semicommercial, subsistence; scale of example,large and fairly well-offHausa families in northern production: large, medium, small, micro; intensity of Nigerian urbancentres keep cattle in the inner courtyardsof production: high, medium or low level of external inputs; theirhomes. Theanimals are fed with grassand crop residues, relative importanceas economic activity: full-time vs part- which are eitherpurchased or fetched by workers. The milk time; labour resources:individual or family vs hired; animal is consumed mainlyby the largefamily. Themanure is packed resources: indigenous, crossbred, exotic; husbandry methods: in sacks and transportedby donkeysto the family farmplots. free-roaming, herding, tethering, stalifeeding or a This was describedalready by the German travellers Barth combination of these; knowledge base of producer: and Staudinger in the 19th century and is even more experiencedvs inexperienced; origin of producer: indigene widespread today (Fricke, 1979). vs immigrant,ethnic and/orreligious group and land tenure: In some cases, these animal-farming activities of the private, usufruct rent or lease, informal agreement, better-offurban dwellers are intendedas a source of highly- unsanctioned.In any given setting, the classificationmost valued food and as a source of supplementary income useful for understandingthe existing systems and guiding (although economists' calculations often cast doubt on the development planning will depend on the historical profitability).In other cases, they are more of a hobby than a development, the settlementpatterns and the major resource business.In urbanas in rural areas,livestock also play a social constraints.For example, if space is a major constraintfor role for family and religious ceremonies.However, especially livestock rearing in an urban centre, then classificationof in recent years as the gap widens betweenreal wages and the systems accordingto scale and husbandrymethods will help costs of urban life, an important reason why employees in in designing appropriateaction. the formal urban sector raise animals is to reduce food expenditures (Maxwell and Zziwa, 1992 and Mlozi et al., Functionsfor Different Income Groups 1992: 284—294). Urbanlivestock systemscould alsobe classified according to the incomelevel of the entire household, i.e. not only the (b) For thePoor incomefrom livestock. Studieshave revealed that small-scale Off-plot livestockrearing such as alongroadsides appears farming including animal farming is practised by urban to be mainly an activity of lower-incomegroups. On-plot householdsin all incomegroups (e.g. Sawio, 1993;Maxwell animal rearingby poorer urban dwellersis largely restricted and Zziwa,1992and Mougeot, 1994). However,some types to micro livestock such as poultry, rabbits and guinea pigs, ofanimal farmingare more widely practisedby therelatively and a few small ruminants. A large number of low-income rich and others by the relatively poor. It is important to urban farmers many of whom rear smallstock are women distinguishbetween these if developmentis supposedto be (Maxwell and Zziwa, 1992; ENDA-Zimbabwe, 1994; Lee- aimed at alleviating poverty, as the functions of livestock Smith and Memon, 1994: 67—84 and Sawio, 1994: 24—46). and the strategiesfor improvingthe production systemswill Theseproducers have little accessto veterinary care and can differ. afford only very limited amounts of purchased feeds. Very poor urban dwellers sift through garbage to find food for (a) For the Rich their smallstock; they can also be observed spreading out Periurban 'modern'poultry, pig and dairy unitsare owned garbage piled on urban streets to let their goats select from by the relatively rich and generally cater to a high-income it.

221 Ann Waters-Bayer African Urban Quarterly

Many of the poorest households, especially recent One of the greatest strengths of small-scale urban migrants to squatter quarters, would like to have livestock, livestockrearing is its great mobilityand flexibility.Animals but cannot because they do not have enough money,space, can be moved fromone day to thenext; field crops and trees security and knowledge of local possibilities to start. cannot.Livestock rearinggives value tomunicipal and private Especially for the lower-incomeurban families, livestock land momentarily not being used for other purposes, thus rearing plays a vital role as a source of food, income and making 'opportunistic'use of land in a positive sense. security. For them, the main functions of livestock are to A key opportunityoffered by urban livestock rearing is provideanimal proteinfor the family which itcould notafford waste recycling. One of the biggest problems in urban to buy; generate some incomethrough sales ofanimals, milk, centres—garbage—can be a resource for animals: organic manure and/or transport services; serve as a buffer against wastes from households, streets, marketplaces and high rates of inflation; save capital in a form that provides agroindustries can provide valuable feed for animals. higher interestrates (offspring) than money in the bank and Removal of these wastes for use in animal farming can at that can he used as a reserve to cover unexpectedexpenses; least partly replace municipal garbage-disposal services. The provide additional employment opportunities for family value of recycled wastes for livestock has long been members, especially those without formal training; make recognised in Asian urban centres, but also in many large productive use of 'free' resourcessuch as kitchenwastewater, Latin American urban centres, where a lucrative source of food-processing residues, and garbage in city streets; and income for smaliholders is backyard pig keeping based on transport, water, feed, food and trading goods in the case of household scraps and food wastes from restaurants. In donkeys,for example. Montevideo,people with horse carts collect garbage useful Urban animal farming provides a source of employment for pig feeding before the garbage trucks come (Serd and not only for the livestock rearers themselves but also for Neidhardt, 1994: 10—14). people operating in the informal supply systems: herders, Alsourban wastewater can bea resource for urbananimals watchmen, collectorsand sellersof milk, eggs and birds, and and crops. For example, Mexico City, Mexico pumps over sellers of leaves and grasses (Centres, 1991). Some poorer halfits sewageto irrigatedalfalfa fields.The feed is trucked women also go from house to house to buy cerealbran which to the urban centre and sold in shops to backyard livestock they then resell to livestock rearers in the urban centre. On producerswho, in turn, sell the manure to urban growers of urbanmarkets and roadsides, bundlesof cut grass or lopped vegetablesand flowers (Smit and Nasr, 1992: 141—152). On browse, groundnut hay and other crop residues are offered the outskirts of Harare, the City Council grazes cattle on for sale. This is done not only by farmers but also by poor pasturesirrigated with treated wastewaterand sells the meat urban dwellers, who make daily forays several kilometres to urban market outlets (Mougeot, 1994). Some herders in out of the urban centre in order to collect feed to sell in the Zimbabwe have takenwastewater management intotheir own urbancentre (Centres, 1991). Somefarmers without animals hands: they are blocking sewer mains passing through low- even grow forage for sale, such as those who grow Napier lying land between residential quarters of the city so as to grass aroundNairobi to sell tourban livestockrearers (Lado, flood the area and provide green pasture for their cattle 1990: 257—266). (Mbiba, 1994: 22—23). Anotheruse of wastewateris aquaculture, a fast-growing Opportunitiesfor the Urban Centre form of urban livestock rearing throughoutthe world. Fish Animal farming serves useful functions not only for the can be raised in wastewater purified less completely than producers;it also offers the urban centre as a whole several needed for direct human consumption.,The water can be opportunities to make better use of its resources. Public purified by biological means such as duckweed, which can benefits offered by keeping livestock in cities include: also be used as feed for fish and other livestock (Smit and bringing a return from land which is unsuitable for Nasr, 1992: 141—152). construction (e.g. seasonally floodedareas), still undeveloped Just as urban wastes can be a resource for livestock, the or not being used for other purposes (e.g. land foreseen for wastesfrom urban livestock can be aresource for cultivation airport expansion),or suitable for multiple use (e.g. urban and also a source of income for livestock owners. Thus, a woodland); providing employment, also upstream (input potential problem of animal farming—manure supply) and downstream (processing and marketing); accumulation—isturned into a valuable input for urban reducing transportand energy costs(and associated pollution) production of staple foods, vegetables and fruits. Indeed, it for input supplyand producemarketing, as distances between is reported from Indonesia that manure mixed with rejected producers and consumers are short; reducing public feed and sold as fertiliser can make up a large part of the expenditures for municipalservices or land maintenance, e.g. cash generated from small ruminantskept in stalls (Orskov, for garbage collection, or for cutting vegetation along 1994:24). Therapid recycling of animal wasteshelps reduce roadsides and utilitylines and on otherpublic land;improving the health risks caused by keeping livestock in the urban the supply of fresh, perishablefoods such as milk and eggs centres. Informal systems are being developed to transfer to urbanconsumers, especially those who lack refrigeration manureand animal-processing wastesto nearbycropland and facilities and providing urban dwellers with food more fishponds. In Bamako, for example, market gardeners buy cheaplythan imported food, as producerscan sell directly to animal manure which is carried to school by the children of consumers, and the costs of processing, packaging and livestock rearers; the gardeners also fetch slaughterhouse handling are reduced. wastes to fertilise their plots (Centres, 1991). Women, who

222 Volume 11 Nos. 2 and 3, May andAugust, 1996 Animal Farming in African Cities keep the majority of small ruminantsin northern Nigerian An importanttask of researcherswill be to determine the urban centres, also benefit from selling the manure to local real impact of urban animal farming and informal sale of gardeners (Hill, 1972). animal products on human health and to compare this with Efficient recycling of sewage water and organic wastes the benefits. For example, with regard to milk, what is the for and from livestock will be a main focus of research and occurrence of brucellosis (and in turn,Malta or undulantfever extensionwork in urbananimal farming.Appropriate forms in man) and tuberculosis among consumers of the milk of local organisation are needed to strengthen the links products sold through informal channels as compared with between livestock, cropping, food processing and waste that among consumersof industrially-processed milk?What collection. Household separation of garbageand decentralised would be the impact on nutrition and, thus, health if banning collection of the organic wastesfor urbanlivestock and crops the sales of unpasteurised milk and introducing expensive withindifferent urban centres quartersshould he encouraged processing meant that poorer segments of the population by emphasising the economicbenefits (Lardinois and van de could no longerafford milk?To what extent do local traditions Klundert, 1994: 6—8). There are already some promising of preparingfoods derived from livestock (e.g. fermentation to examples,particularly in large Asian urban centres, of how of milk, lengthy cooking of meat) reduce the risks such local recyclingof nutrients can makeurban centres more consumers? Ratherthan repeatingthe suspicionsof potential animal should self-sustainablein terms of food security (Newcombe, 1977 health riskscaused by urban rearing, scientists out 179—208; Smit and Nasr, 1992: 141—152).A notable initiative hecarrying interdisciplinary, includingepidemiological, to of the in health and in Africa is that of the Zabbaleen community, which studies provide evidence problems other sectors affected livestock Then, attention specialises in collecting, sorting and recycling solid wastes by rearing. can be to the most serious real of in Cairo, Egypt. The livestock keepers graze their pigs and given solving problems for human welfare. goats on organic wastesand, in turn, supply the community's high priority The of is that millions of in African with livestock wastes (Mougeot,personal reality today people compostingplant urban centres are with either their own or communication,1995). living livestock, that of their neighbours. The people have good reasons for animals and faceconsiderable in Problems and Constraints keeping they problems doing so. They, their neighbours and their customers could benefit Raisinganimals in urbancentres brings not only benefits; greatly from support in dealingwith these problems. it can also give rise to very real problems. When the density Thefew detailed studiesmade thus farsuggest that a major ofanimal populationis high, diseasescan spread more easily problemexperienced by animal farmers in the urban centres and quickly. The proximity of animals to humans increases is the high rate of livestockmortality. Forexample, the value therisk of transmitting diseases also to humans.Manure, dirty of animalsthat died annuallywas found to be higher than the materials, feed rests and the wastes of animal bedding value of animals consumed or sold by urbancattle rearers in if not can attract flies and rats processing, properly handled, Nairobi and Memon, 1994: Also a and can water. As of urban livestock are (Lee-Smith 67—84). study pollute products of urban sheep production in Côte d'Ivoire came to the sold informal control of mainly through channels, hygienic conclusion that productivity could be increased most conditionsand food becomes It is quality impossible. already efficientlyby reducingmortality (Armbruster, 1987). difficult in the formal in view of the lack extremely sector, The challengefor livestockservices is to find economic of laboratoriesand staff in appropriatelyequipped qualified ways of doing this. Most small-scaleanimal farmers cannot African countries. Traffic accidents are caused many by afford large cash expenditures. The solutionsare more likely animals. often aboutthe noise roaming Neighbours complain tobe found in therealms of cost-effectiveprophylaxis rather and odours from animals in the urbancentres. kept than curative measuresand in hygiene educationrather than animal and sales of non-controlled Banning farming regulations than cannotbe enforced. animalproducts is not the answer, as this woulddeprive many But animal farmers' problemsare by no meanslimited to urban families of a vital source of livelihood and security. thoseof animal healthand hygiene. Studies of urbanlivestock Besides, in many urban centres, the bans are already in the rearing in Mali and Burkina Faso revealed that the greatest books,but cannotbe imposed. Most importantly, bans prevent problem of small-scale producers isobtaining sufficient feed, government services from intervening to improve animal particularlyin cramped inner sectors of urbancentres where health and productivityand to minimiserisks tohuman health. there is little or no room for grazing. This problem is likely It is in the interest of the large-scaleperiurban livestock to be closely linked with the high animal mortality rates. operations,which have high overheads and high production Another problemin theinner urban centresis lack of adequate costs, toeliminate the competition from the small-scale urban space for the animals. When it rains, small ruminants and livestock rearers who can offer cheaper products with less poultry are kept in the entrance,the kitchen or a room of the processingand packaging. Before governmentssuccumb to house; otherwise,they wanderthe streets. Addressingthese the pressures ofthe influential 'modern' producers for stricter problems of animal feeding and housing will help in controlson productionand marketing,a closer look needsto improving animal welfare and reducing losses. It may also be taken at who will benefit from such controls. The health be possible to overcome space constraints by promoting a arguments were put forward in colonialtimes tojustify bans shift to smaller species of animals. and possibly to eliminate indigenouscompetition; the same Where producers are so close to coisumers, marketing arguments are still widely used today. does not appearto be amajor problem.In the studiesdone in

223 Ann Waters-Bayer African Urban Quarterly urbancentres such as Bamakoand Bobo Dioulasso, the need if they can derive benefits from it. For example, they could for better organisation of input supply was deemed to be far be granted rightsto grow and trim trees which line the roads more importantthan improving the organisation ofmarketing or to plant and cut grass on unbuiltslopes or public land, in (Centres, 1991 and 1992). order to obtain animal feed.

Room forAction (e) Policy Favouring Small-Scale Low-External-Input As the urban centres expand rapidly and as increasing Systems numbersof urban dwellersseek security and profit in raising Governments need to set up a policy framework which crops and animals, governments have to re-examine their favours small-scale animal farmingbased on inputs available position with respect to urban agriculture. In view of the in and near the urbancentres. This would involve encouraging potential dangers of livestock being kept in close proximity the use of local and so-called 'non-conventional' feeds and to humans, but also in view of the potentialbenefits offered taking advantage of the wastesfrom urban agroindustries such by keeping animals on the urban centre doorsteps, as breweries. This would also involve focusing on animal developmentplanners are faced with many challenges. species and breeds which can use the available roughages and depend less on concentratesand which are suitable for (a) Recognition urbanareas becauseof their small size and docile nature.As The first big step for public authorities is to recognise a rule, it can be said that the smaller the animal, the more that animal farming in African urban centres does exist. affordableit is to a wide range of income groups and the Official recognition makes it easier for veterinary and more easily itcan make use of limitedurban spaces (Mougeot, extension services to deal with the potential dangers of 1994). This suggests that activities focused on smalistock livestockto public health and environmental quality. such as chickens,ducks, pigeons, guineapigs, rabbits, goats and sheep are more likely to benefit also the urban poor in (b) Information crowded inner sectorsof urban centres. Needs to be disseminated to both producers and consumersabout livestock-related dangersfor human health (f) ImprovedMarketing Facilities and how to avoid or reduce them. Ratherthan tryingto reorganise marketing to replace the existinginformal systems, assistancecould be given simply (c) Action-Oriented Research by offering improved marketing facilities such as To obtain relevant information and to ensure proper decentralised 'urban farmer markets' which allow direct assessmentof the potentials and problems of urban animal producer-consumercontacts under hygienic conditions, for farming,more researchis neededinto the existing livestock- example, with access to clean water. This is particularly rearing systemsin African urban centres. The research must important for the women selling dairy products and other be designed to yiel the kind of information that urban home-processed foods. farmers, consumers and planners requirefor making decisions and taking action and the results of such researchneed to be (g) Women as Partners made widely known among all stakeholders in clear and Mostreports on urban livestockrearing make no reference concise language. Aboveall, action-oriented research in urban to the gender of livestockowners. However,it is frequently animal farming shouldbe aimed at buildingthe capacities of mentionedthat the work of feeding,watering and healthcare the poor and landless urban dwellers to assess and expand of animals in the urban centres is done mainly by women their possibilities to use urban centre resources to rear just as it is in many rural areas and that women ar doing livestock and to argue their interests in the politicalarena of muchof the informalmarketing of small animalsin the urban the urban centre. centres. This suggests that particular attention needs to be given to the role of women in urban livestock rearing and (d) AppropriateTechnology and Inputs that womenwill bekey partners indevelopment efforts. The Development agents should work together with low- work done by women in caring for animals particularly income urban dwellers to develop appropriatetechnologies providing them with feed and water is time-consumingand for rearinganimals in small spaces as well as organisational even more so in densely-settled areas. Many women have forms to give poor urban dwellers access to the inputs chosento investthis time, possibly to the detrimentof other necessary for healthy and productiveforms of animal farming. household chores.This is afact whichreflects the importance A relatively simple start would be to make inputs such as that the women accord to livestock. It is likely that these veterinary products and feed supplements available in small womenwill bemore interested in finding ways to save labour units suitable for backyard farming.As emphasised above, in livestock husbandry rather than to increase livestock there is particular need for innovationsn technology and production (Laurent and Centres, 1990), and this must be local-levelorganisation to permitefficient and safe recycling taken into account in technology developmentactivities. of organic wastes for and from livestockin the urbancentres. Innovativeways of combininganimal farmingand 'greening' (h) Communication and Negotiation of the urban centres should also be sought.Urban livestock Development agencies should create opportunities for keepers mayeven invest in urbanenvironmental improvement livestock rearers and other urban dwellers to communicate

224 Volume 11 Nos. 2 and 3, May andAugust, 1996 AnimalFarming in African Cities

with each other so as to understandeach other's actionsand (eds) Cities Feeding People: An Examination of Urban problems and to encourage changes in behaviour. One Agriculturein EastAfrica. Ottawa:JDRC. possibilitywould beto promote resource-user organisations: common-interest groups which can negotiate withother local ENDA-Zimbabwe(1994). Urban Agriculture in Harare: stakeholders injointly definingand implementing regulations Report on an IDRC-SupportedProject. Harare: ENDA- for using urban resources such as grazing areas or garbage. Zimbabwe. Particularlyurban crop farmersand livestockrearers need to be brought togetherto work out forms of interactionwhich Fricke, W (1979). Cattle Husbandryin Nigeria: A Study of can bring them mutual benefits. Legal arrangements for Its Ecological Conditions and Socio-Geographical temporary or multiple use of urban resources must be Differentiations. Heidelberg: Heidelberger Geographische facilitated in such a way that the great strength of urban Arbeiten52. Geographisches Institut. livestockrearing—its flexibility—is maintained. Municipal authoritiesare in a position to provide temporary securityof Gefu, J.O. (1992). "Part-TimeFarming as AnUrban Survival usufruct rights on public and private land and could also Strategy: A Nigerian Case Study", in J. Baker and P.O. provide assistance in conflict resolution. Pedersen (eds) The Rural-Urban Interface in Africa: Expansionand Adaptation. Uppsala:Scandinavian Institute Conclusion of AfricanStudies. The challenge for government services faced with the of urban animal is to interact in a Giorgetti, A. et al. (1983). "Caratteristichedegli Allevamenti reality farming positive Bovini neIl'area lJrbana di Rivista di and enabling manner with theproducers rather than making Mogadiscio", e Vol. 77 No. 3. vain attempts at banning the animals. Both producers and AgricolturaSubtropicale Tropicale consumersare in need of well-founded and locally relevant information and low-cost which Hill, P. (1972). Rural Hausa: A Village and a Setting. appropriate technologies Press. permitmore healthy and productiveanimal farmingin African Cambridge: CambridgeUniversity urbancentres. All stakeholders the urbanauthorities including N. in Cairo: will then be in better tomake wise decisionswhen Khouri-Dagher, (1987). Food and Energy a position Nexus the use of urban resources for livestock Provisioning the Poor. Paris: Food-Energy jointly planning United Nations rearing and other sources of livelihoodfor urban dwellers. Programme, University. W. in the References Krostitz, (1984)."Poultry Development Developing Countries", WorldAnimal Review Vol. 52. Achuonjei,P.N. and S. Debrah (1992). "Efficiencyof Fresh Milk in the Bamako Area of Mali: Some Marketing C. Informal Urban in in R.F. Brokken and S. Lado, (1990). Agriculture Nairobi, PreliminaryResults", Seyoum(eds) Problemor Resourcein and Land Use inSub-Saha ran AddisAbaba: ILCA. Kenya: Development DairyMarketing Africa. Planning?Land Use PolicyVol. 7 No. 3. Armbruster,T. (1987). Die Produktivitätder SchaJhaltung I. and A. van de Klundert of in der der Lardinois, (1994). "Recovery Regenwaldzone Elfenbeinkaste. Gottingen: Organic Waste in Cities", ILEJA Newsletter Vol.10 No. 3. GottingenUniversity, unpublished PhD thesis. Laurent, C. and J.M. Centres (1990). Dairy Husbandry in W.P. "Towardsan Urban Earthwatch Carty, (1991). World", Tanzania: A Development Programmefor Smaliholdersin Vol. 43. Kilimanjaro and Arusha Regions. Versailles-Dijon- Mirecourt, (Documentde travail, URSAD). Centres, J.M. (1991). Améliorer l'approvisionnement de Bamakoen produits maraichers et enproteines animales. 1. Lee-Smith, D. and P.A. Memon (1994). "UrbanAgriculture Agricultureet elevagea Bamako. Paris: GRET. in Kenya", in A.G. Egziabher et al. (eds) Cities Feeding People:An Examination of UrbanAgriculture inEast Africa, Centres, J.M. (1992).Agriculture et elevage urbains etperi Ottawa: IDRC. urbains aBobo Dioulasso: Lesfilieres d'app rovisionnement en intrants et de commercialisationdes produits agricoles. Maxwell, D. and S. Zziwa (1992). Urban Agriculture in Paris: GRET. Africa:The Case ofKampala, Uganda. Nairobi: ACTSPress.

Debrah, S. (1992). "Dairy Marketing by Intra-Urban, Pen- Mbiba, B. (1993). Socio-Economic and Gender Aspectsof Urban and Rural Dairy Producers Near Addis Ababa, Urban Agriculture in Harare, Zimbabwe. Regional Ethiopia", in R.F. Brokken and Seyoum S. (eds) Dairy Development Planning Occasional Seminar, 1 October, Marketingin Sub-Saharan Africa. AddisAbaba: ILCA. Institute of Social Studies, The Hague.

Egziabher,A.G. (1994). "Urban Farming,Cooperatives, and Mbiba,B. (1994)."The Cattle ofChutungwiza: Conflicts on the Urban Poor in Addis Ababa", in A. G. Egziabheret al. the Rural-Urban Fringe.ILEIA Newsletter Vol.10 No. 4.

225 Ann Waters-Bayer African Urban Quarterly

Mlozi, M.R.S.et al. (1992). "UrbanAgriculture as aSurvival Landuse Change in Dar-es-Saiaam, Tanzania. Worcester: Strategy in Tanzania", in J. Baker and P.O. Pedersen (eds) PhD thesis, Graduate School of Geography, ClarkUniversity. The Rural-Urban Interface in Africa: Expansion and Adaptation. Uppsala: Scandinavian Institute of African Sawio,C.J. (1994). "WhoAre theFarmers of DaresSalaam?, Studies. in A.G. Egziabher et al. (eds) Cities Feeding People: An Examination of Urban Agriculturein East Africa. Ottawa: Mougeot, L.J.A. (1994). Urban FoodProduction: A Survey IDRC. of Evolution, Official Support and Significance. Paper Presented to "Habitat '94", Edmonton,Canada. Serd, C. and R. Neidhardt (1994). "Stadt-Land-Integration im Rahmen PeriurbanerTierproduktion", Entwicklung und Nell,A.J. (1992). "An Overviewof Dairying in Sub-Saharan ländlicher Raum Vol. 2 No. 94. Africa", in R.F. Brokken and S. Seyoum (eds) Dairy Marketing in Sub-Saharan Africa. Addis Ababa: ILCA. Smit, J. and J. Nasr (1992). "Urban Agriculture for SustainableCities: Using Wastes and Idle Land and Water Newcombe, K. (1977). "Nutrient Flow in a Major Urban Bodies as Resources",Environment and Urbani7ation Vol. Settlement: Hong Kong", HumanEcology Vol. 5 No. 3 4 No. 2.

Ngigwana, L.L.M. (1992). "Dairy Imports and Their Staal, S.J. and B.I. Shapiro (1994). "The Effects of Recent Influence on Domestic Dairy Marketing, With Particular Price Liberalizationon Kenyan Pen-Urban Dairy: A Case Reference to Tanzania", in R.F. Brokken and S. Seyoum Study Using the Policy Analysis Matrix Approach", Food (eds) DairyMarketing in Sub-Saharan Africa. AddisAbaba: PolicyVol. 19 No. 6. ILCA. Walshe, M.J. (1991). Dairy Development in Sub-Saharan Orskov, R. (1994). "Landless Livestock Keepers", ILEJA Africa: A Study of Issues and Options. Washington DC: Newsletter Vol. 10 No.4. Technical Paper 135. World Bank.

Sachs, I. (1986). "Work, Food and Energy in Urban van derBliek, J. (1992). Urban Agriculture:Possibilitiesfor Ecodevelopment",Development: Seeds of Change Vol. 4. EcologicalAgriculture in Urban Environments as a Strategy for SustainableCities. Leusden: ETC. Sawio,C.J. (1993).Feeding the Urban Masses?Towards an Understanding of the Dynamics of Urban Agricultureand

226 Volume 11 Nos 2 and , Mayand Auguat 1996 AnimalFarming in African Cities

Cairo,Egypt. Chickens, clucks and poultryraised in backyards in urban and periurbanareas are regularlysold in city streets

Photo: WolfgangBayer

Kaduna,Nigeria Unattended flotks of shrep wanderthrough the informal srttlements and engulfed villages whichnow make up large parts of northern Nigerian cities

Photo:Wolfgang Bayer

227 Ann Wa1ers.Bayer African UrbanQuarterly

Smallstock scavenging for food,a typical scene in towns throughoutBurkina Faso

\VoU ang ltaycr

URBANMICROFARMING IN CENTRAL SOUTHERNAFRICA: ACASE STUDYOF LUSAKA, ZAMBIA

A.W. Drescher Departmentof Applied Physiogeographyof theTropics and Subtropics(APT), University of Freiburgi. Br., Germany

Abstract Accepted January, 1996 The relationshipbetween urbanfood production, food securityand the urbanenvironment has been largely neglectedin thepast. In Lusaka,Zambia as in many othertropical urban centres, gardening andcropping receive very littlesupport from local authorities. Indeed, city councils often prOhibit these activities. Productionof staplefoods prevails in the wet season, and vegetableproduction in the dry. Both activities largely depend on access to resources like water and land. Within the high-and medium-density squatter quarters, vulnerability in termsoffood securitydiffers. In the Zambiancase, it wasfound that dry-season cultivationis not practised by the most vulnerable householdsbut rather by those which have access to essential resourcesfor this activity. In Lusaka,garden size decreaseswith increasingpopulation density. The walking distance to sources ofwater is muchfurther in the high-density areas, making microfarming more difficultthere. Accessto both land and water is lowest in the high-density, low-income compounds in Lusaka. Microfarmingobviously contributesto householdfood security in the city both directly by providingfood and indirectly by generating income. There are significantdifferences betweenthe role ofwomen andmen in urban householdfood security. Womenare the major actors in urban microfarmingbut face many obstacleswith respectto incomegeneration andaccess to resourcesand markets. There is urgentneed for appropriate support of urbanagriculture in thefuture. The paper tries to give some ideas.

Résumé Par le passé on a largement negligé Ia relation qui existe entre Ia production alimentaire, Ia sécurité alimentaire et 1 'environnement urbain. A Lusaka en Zambia, le maraIchage et autres cultures restent trés peu appuyéspar les autorites locales. En effet, les administrations urbaines interdisentsouvent ces activités. Typiquement, a Lusaka Ia production de denrées de base tire profit de Ia saison des pluies tandisque les cultures maraIchèresperdurent en saison sèche. L'accès des producteursaux ressourcesdetermine largement Ia localisationetl'ampleurde leurs cultures.Au sein des bidonvilles aforte et moyenne densité d'occupation, Ia vulnérabilitédes gens a I 'insécuritéalimentaire vane. Dans le caszambien, les cultures de saison sèche ne sontpas pratiquéespar les ménagesles plus vulnérables, mais bienpar ceuxqui ont accés aux ressources requises. A Lusaka les parcelles cultivéessont plus réduites dans les secteurs plus densément peuplés. Les distance de marcheaux points depuisage sont plus grandes dans ces secteurs, ce qui y rend la microculture plus difficile. L'accèsau sol et a l'eau estfortement limité dans les secteurs denses a bas revenus de Lusaka. En général, Ia microculturecontribue a la sécunité alimentairedes ménagesen ville, directement en lui procurant des alimentset indirectement en générant des revenus de vent. Le role des femmes se distingue clairementde celui des hommes dans les strategies du ménagepour réduire son insécuritéalimentaire. Malgré lefait qu 'ellesdominent Ia pratique de Ia microculture, lesfemmessont désavantagées en ce qui a trait a Ia génération de revenus et l'accès aux ressources et aux marches. II y a besoin urgent d'appui adéquat a I 'agnicultreurbaine et I 'article soumet quelquesrecommendations en ce sens. Urban Microfarming: A Neglected Issue staple food productionor on vegetableproduction. There is Food availability and stable access are both critical to an urgentneed to realise that this issue shouldbe looked atin household food security. The means for procuring food amore holistic way. As emphasised during arecent workshop (produced, purchased and gathered) and the social (DSE/ATSAF, 1994), research and planning in urban mechanisms that buffer households from periodic shocks agriculture requires interdisciplinarity. The term urban determine the households' access to food (Frankenberger, microfarming is used here to reflect this need for a 1992).Thus thepossibility to garden isthe basic requirement comprehensive understanding of agricultural land-use in toself-sufficient or additionalfood production of ahousehold, urban centres. It encompasses urban crop production, Food might be available on the markets but not attainableto homegardening, horticulture(both vegetables and fruits)and the householdsbecause prices are too high. livestock keeping. Also the gathering of wild fruits and When agricultural activities in urban centres are vegetablesis a strategy of urban people to achieve greater examined, a one-sided view is often taken, for example,on food security. Thesurvival of many urbanfamilies in southern Acknowledgements—Very special thanksto FriedaBos (Agricultural Universityof Wageningen, The Netherlands),for contributingin field work and theoreticalframework and especially for the support of the socioeconomic componentof the research project. Paul Muwowo (University of Zambia)strongly supported the fieldwork, helped a lot to understand the people and was aperfect student.Thanks also to the FAOProject (TCP/ZAM /2553), LusakaiZambia for the support of the research project in 1993. Dr Ann Waters-Bayer (ETC, The Nether- lands)did a great deal in revisingthe article and contributingwith good ideas how to make itbetter.

229 A. W Drescher African Urban Quarterly

Africa depends on these activities, which help to produce becameone ofthe most important sources of foodand income additional food and income, thus increasing food security in urban areas (Schilter, 1994 and TUAN, 1994). and the standardof nutrition. As will be shown with respect to the timing and organisation of labour inputs, there are The Potential of Urban Microfarming in Africa: Some significant differences between these activities including Examples gender-specific differences. Globally, about 200 million urban farmers provide food In Zambia, field crops are grown in the wet season to and income for about 700 million people (Smit et al., 1996). produce the staple foods. In rural and pen-urban areas, In 1993, in high-density areas of Harare, Zimbabwe, more gardens and field crops are distinctly separated,but in urban than 300 t of maize were harvested (ENDA-Zimbabwe, areas a plot near the house is often used as a homegarden in 1994). In Dar es Salaam, Tanzania, urban agriculture was the dry season and as a crop field in the wet. Here, as in the secondlargest employer in 1988. In Kenya, 25 percentof many other countries, the homegardenis not exclusivelya the urbanfamilies are absolutely dependenton self-produced vegetable garden. It may includecrops thatare mainly sources food. Accordingto Maxwell and Zziwa (1992) in Lusaka, of food carbohydratesrather than what Western horticulture Zambia,one-third of the urbanpoor produce all of theirown classifies as vegetables as well as fruits, fuelwood, herbs, food. Low-income groups are highly involved in vegetable medicinal plants and plants grown for their ornamentalor gardeningin Maseru, Lesotho: 69 percent of these families magical value (Brownrigg, 1985). Under the climate produce vegetables while only 39 percent in high income conditionsin Zambia, staple food productionis restricted to areas do so (Greenhow, 1994). the wet season. Both men and women areinvolved. The staple In high-density areas surveyedof Harare, 86 percent of crops in urban areas include maize, sorghum, millets and the households keep small livestockand 49 percent of these cassava as well as substitutes for staples such as sweet households keep chickens (ENDA-Zimbabwe, 1994). Poultry potatoes and groundnuts. In addition, vegetables such as keeping in the backyard is very common in many African pumpkins,beans, okra, tomatoes and cucumbers are grown urbancentres. ForKampala, Uganda,70 percent of thepoultry as intercrops. This type of urbancultivation has the nutritional needs are produced inside the city (Maxwell, 1994: 47—65). role of bringing a combinationof carbohydrates,vitamins From many countries, it is reported that activities in staple and micro-elements into the human diet. Urban food production, vegetable growing and small livestock homegardeningis done mainly by women, with the help of keeping are greater in high-density than low-density their children in wateringand weeding.It may bedone year- residential areas. In Maseru, Lesotho,low-income areas on round or only in the dry season,when it is highlydependent rocky soils have relatively more livestock production (pigs on the availabilityof water. The main nutritional roles of and poultry) than other areas with better soils (Greenhow, urban homegardeningis to diversifythe diet and to provide 1994a). This type of urban agriculture is particularly vitamins (mainly through vegetables). In Zambia, urban importantfor generatingcash. In urban areas,not only small- livestock keeping, which is even more ignored than other scale subsistence livestock-keeping but also commercial componentsof urban microfarming,is also mostly done by production is found. Milk, pig and poultry production are women, with some important support of children. It brings commonin many African cities. benefitsin terms of both cash and food for the families. This activity, as well as the hunting of small animals has the Urban Microfarming: Case Studyof Lusaka, Zambia nutritional role of providing animal protein. Microfarming Lusaka is one of the fastestgrowing urban centres in the is partof the bioticcomponent of the urbanenvironment and developing world and had a population of 1,192 million competes with the abiotic componentssuch as buildingsand inhabitantsin 1991 (Zambia,1990 and 1992, figure 1). Since roads. Therefore, urban microfarming needsto be considered 1980, its populationhas nearly doubled,growing at a rate of in urbanplanning activities. Development ofmore sustainable 6.1 percent between 1980 and 1990, when the population cities will require an improved balance between the biotic density was 2,728 personsper km2 (Zambia,1990). Looking and the abiotic environment(van der Bliek, 1992). at these figuresand takinginto account the overall economic Urban microfarming has been largely ignored by situation of the country, it becomes obvious that food researchers,extension services and policymakers (Mougeot, production in urban areas is a necessity for the survival of 1993: 2—5; Lamba, 1993: 17—18; Streiffeler,1993 and 1994; the poor. These days, to eat a single daily meal is becoming Ratta and Smit, 1993; Greenhow, 1994a and Drescher, 1994). commonplace inAfrican urbancentres (Vennetier, 1988: 209— It has been regarded as economically marginal. In Africa, 226, cited by Mougeot, 1994). Consequently malnutrition, womenase traditionally responsible for feeding their families. microelementdeficiencies and hunger are growing worse. In urban areas, this role is not planned for as it is assumed In Zambia, politiciansdislike dealing with issuesof urban that subsistenceactivities do not take place in urban centres microfarming,in the past as in thepresent. It was, therefore, (Lee-Smith and Lamba, 1991: 37—40). Another reason for quite exceptional when former President Kenneth Kaunda the neglect of urban microfarming has been the focus told the National Assembly in 1977 that the inhabitants of especially by development agencies (Drescher, 1995), on Lusaka must grow their own vegetables and cereals. His staple foodproduction for the urbancentres, whichis thought messagedid not reach the LusakaCity Council (LCC), even to be possibleonly in rural areas.As the economies of many in the drought year of 1992, with its extreme shortage of African countries started to deteriorate in the 1970s and food, the LCC suppressedurban agriculture by threatening consequently the living conditions worsened, agriculture to slashthe maize.One of the official reasons given was the

230 Volume 11 Nos. 2 and 3,May andAugust, 1996 Urban Mkrofarming in Central Southern Africa: A Case Studyof Lusaka, Zambia

Figure1, Population growth in Lusaka, Zambia(1950—1995)

Pç.thiii Qmsthin Lusaka(Zanña) (1950-1995) Penin Thsmds

14OO 1192 1o .----. -

9N2 -.. - 8O0 . - - -.-

- .- .. ..

mirce.SO 1990, 0 19J 19S '63 '69 '74 '89 '90'92 '95 Years

231 A. W. Drescher African Urban Quarterly possiblehealth hazard:maize crops supposedlyfacilitate the were recorded and counted by individuals to determine breedingof malaria-bearingmosquitoes (Sanyal, 1985: 15— species diversity and yields. 24), an argument that is repeated season after season.In the case of Lusaka, it is obvious that the lack of environmental (b) Wet Season Cropping Survey hygiene (non-existentor non-functioning sewage systems) In the 1992/1993 rainy season the survey on wet season is much more likely to favour malaria than are urban crops croppingactivities covered some of the open spaced areas or other vegetation. In the meantime,it has becomeobvious iosideLusaka. The fields were measuredand cultivated plants that urban agriculture is not, as assumedby most authorities, were recorded as well as intercropping systems, planting "a remnantof bush life" (Sanyal, 1985: 15—24) hut rather a distances and special cultivationpractices. survival strategy of urbanpeople. However, it is still necessary to convince urban authoritiesof its importance for the well- (c) Compound Survey on AgriculturalActivities being of the populationand the sustainabledevelopment of The compoundsurvey was carried out in 1992/1993 in urban centres. According to Jaeger and Huckabay (1986), eleven compoundsof Lusaka by using a brief standardised three kinds of cultivationcan be distinguishedin urban and questionnaireincluding four questions such as (i) Do you pen-urban areas: small-scale commercial agriculture on practise gardening? (ii) Do you cultivate in rainy season? smaliholdings; houseplot gardening (homegardening) and (iii) Do you water in dry season?, and (iv) Do you own or scattered cultivation on vacant land in and aroundtownships. use more than one garden?The purpose of this survey was to This article focuses on vegetable production in small find out how many peopleare involvedin gardening and in urban houseplots and staple food production in the fields, rainy wet season cultivation. In each compounda minimum describing the main results of the survey and highlighting of 50persons (25 womenand 25men) were askedrandomly. thedifficulties faced by thepoorer familiesin their gardening In total 648 people were interviewed. activities. Part of this article is based on a Food and AgricultureOrganisation (FAO) reportprepared by Erescher (d) Market Survey and Bos (1994). The marketsurvey was carried out on four retail markets in different areas of Lusaka (low- and high-income)and on Methodology the mainmarket in Lusaka for 12 monthsin 1992/1993. Aims In 1992/1993 a survey on urban microfarming (with were the surveyof price developmentfor selectedproducts, specialemphasis on homegardens) wascarried Out in Zambia. the standardisation of informal weights and the estimation It covered the urban and pen-urban areas of Lusaka and two of possible income generation by selling homegarden rural areas in the Southern and NorthwesternProvinces of products. the country (figure 2). Components of the research programme were the homegarden survey, the wet season Wet-SeasonAgriculture cropping survey, the compound survey on agricultural Van der Berg (1982) and Jaeger and Huckabay (1986) activitiesand the market survey. This article concentrateson identified the geographic localities of different types of the situation of the urban populationof Lusaka. The main microfarming in concentriccircles around the urban centre. objectives of the survey were to determine the role of Thecentral zoneis dominated by homegardenswhile the semi- homegardens for the households;determine the contribution commercial and commercial typesare situated in the periphery. of garden products to the diet of householdmembers and to The latter forms aremainly confinedto therainy season(end householdbudget; make an inventoryof the mainproblems of October to mid-May).Gardening is practised year-round faced by households in their homegardening; discover why only near permanentsources of water (streams,dams, rivers certain householdspractise gardening whereas others do not. and natural wetlands). In addition, inside and on the edge of For the data analysis a PC based database was created. the city, there are wet-season plots of staple foods and vegetables.These plots had anaverage size of300 m2 in 1986 (a) Homegarden Survey (Jaeger and Huckabay, 1986). Schultz(1976) described them People living in low-income, high- and mediumdensity as "subsidiarygardens", while Schlyter and Schlyter (1980) squatter quarters are mostly affected by lack of food and calledthem "distantgardens". Thesefields aremainly on fallow incomein Lusaka. Therefore,the study concentrated mainly areasinside urban centres which are being usedillegally. Van on these groups. In Lusaka, 34 homegardens were studied in derBerg (1982)described this land-usetype as "semi-vacant detail, in some cases in both dry and wet season, and the landuse".The limited land resourcesare fully used in thewet owners or users were interviewed with the help of a season:also public property like theuniversity campus, areas standardised questionnaire to obtain data on the socio- aroundhospitals and roadsidestrips are farmedillegally. economic background, objectives of gardening, access to In the 1992/1993 wet season, 37 of such fields were resources like water, land, credits and markets, input and measuredin different quarters ofLusaka (tables I through garden practices, labour, seeds, and the gardeners own 5). Since 1986, the averagesize has increased significantly perception of the gardening activity. Complementary to 423 m2, possibly because of the deterioration in the fieldworkwas carriedout to obtain objective information on economic situation. Similar observations were made in the presenceof pests, diseases and predators, the quality of Harare, Zimbabwe where the cultivatedarea of staples has compost (if used), the size of the gardens, distance to the greatly increased in the last decade (ENDA-Zimbabwe, watersource and homestead,among others. Theplants grown 1994). Very rarely are the cultivated areas larger: only two

232 Volume 11 Nos. 2 and 3, May and August, 1996 Urban Microfarmingin Central Southern Africa: A Case Study ofLusaka, Zambia

Figure2, Survey area in Lusaka, Zambia

ZAMBIA

- SurveyAreas -

c. / 1% / I F / 1 • / — cI,

Eastern Provinc*

— _. Province Boundary

Districts included in the survey

Province Administration

() Kilometer 200 — _ artograp IC 0 Miles 125 C. Stockmar

233 A. W Dreseher African Urban Quarterly

Table 1, Land-Use at Kalingalinga 'Friday Corner' Field Size (m2) Maize Beans Pumpkin Sweet Potato Banana Okra Tomato Observations 252 • 2 270 • • Sweetpotato on ridgebed 3 456 • • • Bananas on centralfurrows 4 594 . . . 5 600 • • 6 700 • • 7 1,125 • • • • Averagesize: 571 m2 Source:Compiled by the author large plots were observed (one more than 3,500 m2 and practised gathering (figure 3). All these families also have another more than 2,000 m2). In 96 percentof the observed gardens, which are complemented by their gathering fields, maize was grown, followed by beans (46 percent), activities. pumpkins (36 percent) and sweet potatoes (20 percent).As In the wet season, any open space in the urban the leavesare eaten from the three latter plants, they help to environment is used for cultivating crops. The following cover part of the vegetable needs in the wet season. In tables give details on the size of some of the fields surveyed addition, many wild plants (e.g. Amaranthusspp.) are used inside Lusaka in the 1992/1993 wet season and on the as vegetables.Nearly 40 percent of the respondentsliving in intercropping systems found in these fields. Comparisonof Lusaka town still gather wild vegetables; in pen-urban and thecropping patterns in the differentareas suggeststhat there rural areas, 80 percent of the households in the survey is some informalexchange of information between the city farmers. Friday Corner,Chainama Hills andIbex Hill show Table 2, Land-Use at Lusaka Great East Road'Chai nama Hills' very similar patterns while patterns at City Airport and the Field Size (m2) Maize Beans Pumpkin University Dambo differ (tables 1, 2, 3 and 4). The high 8 544 • • exposure to theft in these two areas may explain the lower 9 564 • diversityof crops, especially of vegetables.Another reason 10 690 • may be lack of access to resources (e.g. seeds) of the very II 864 low-incomegroups. Most likely, the basic combinationof Averagesize: 666.5 m2 crops of all areas is related to the supply of seed. Seeds are Source: by theauthor Compiled normally obtained throughthe distributorsof Zamseed, the

Table 3,Land-Use at Lusaka 'Ibex Hill' Field No. Size (m2) Maize Beans Sweet Potato Observation 12 102 13 140 • 14 253 • 15 301 • • Ridgebed 16 360 • Average size: 231.2 m2 Source:Compiled by the author

Table 4, Land-Use at Kalingalinga 'CityAirport' Field No. Size (m2) Maize Beans Pumpkin Sweet Potato Okra Groundnut Cucumber 17 60 • 18 80 19 112 20 189 • • • 21 270 • • 22 336 23 392 24 400 • . 25 496 26 525 27 549 • . . 28 816 29 880 • • 30 966 31 1,050 • . . Averagesize: 474.7 m2 Source: Compiledby the author

234 Volume 11 Nos. 2 and 3, May andAugust, i99 Urban Microfar,ningin Central Southern Africa: A Case Study ofLusaka, Zambia

Figure 3,Gathering still playsan importantrole in nutrition, even in the urban centreof Lusaka

Gathering [%J n =85 100 80 60 40 20 0 Uba Peti - utan Rixal

•WIdFnits OWid Vegetables]

(Source:Field Survey 1992/1993)

235 A. W Drescher AfricanUrban Quarterly national seed company. Maize seed is often kept from People living in Kalingalinga, anold squatter settlement, previous year's harvest. use theareas around the City Airport, ChainamaHills Hospital The pure maize stands are spaced at 90—100 cm between and the University for wet-season cropping. Bowa et al. rows and 60—90 cm between plants (figure 4). This is much (1979)reported that 73 percentof Kalingalingaresidents had wider than the official recommendationfor spacing: 30 cm a distant garden. This was confirmedby the recent survey, betweenplants (Purseglove,1988). The wide spacing at the and has obviously not changed in the last 15 years. These beginning of the wet season allows flexibility for mixed people belong to the lowest income group in Lusaka. This cropping: vegetablescan be plantedlater in theempty spaces. confirmsSanyal's above-mentioned statementthat the poorest A field may contain up to four different crops at one time. of the poor are highly involved in wet-season cropping. The different combinationsare indicatedin the tables.Most Nevertheless,their involvementdepends very much on the common is an intercroppingsystem with maize, beans and availabilityof arable land. Nearly 50 percent of the women pumpkin. The 10 most commonlyrecorded crops in mixed butonly 35 percentof the men practiceagriculture in Lusaka. systems are maize, bean, pumpkin, sweet potato, banana, Also therole of childrenshould not beneglected. From quite okra, tomato,cucumber, groundnut and cassava. Bananas are a young age, they provide a valuable labour force in food perennialsand have the ecologicalfunction in intercropping production for thehousehold and contribufe also inother ways systems of providing shade and organic material(mulching) to household income. In all townships, women are more and protectingcrops from wind. involved than men in both dry-season gardening and wet- Accordingto Sanyal (1985),wet-season croppingis done season cropping. The productionof staple foods prevails in mainly by the urban dwellers with the lowest per capita the wet season, and vegetableproduction in the dry season. income(figure 5). This group doesvery littlehomegardening, Similarobservations weremade in rural areas ofZambia and whereas the somewhatless poor, low-income families do both Zimbabwe. wet-seasoncropping and homegardening. In the case of the South of the Sahara, the labour of women is more poorest, their lack of access to resources is thus partly important than the labour of men in all aspects of food compensatedby (usually illegal) staple food production in production (Fresco, 1985). Women providethe labourneeded the wet season. in every phase of the food cycle to guarantee the family's Of 648 persons asked about their involvement in nutrition. At the same time, they cannot neglect their other agriculturalactivities in Lusaka,42.6 percentanswered "yes, tasks of food preparation, childcare, fetching water and we practice gardening". As figure 6 shows, there are fuelwood, washing, housecleaning and lookingafter the small significant differences between the different townships of animals(Purseglove, 1985). Besidesthese tasks, womenalso Lusaka with respectto involvementin agriculture (referring generate income, which is often more than half of the total here to theproduction of staple foods). Thereare also gender- householdincome (Due, 1985 andFresco, 1985).Thus, in a specific differences,women being much more involvedthan multitude of ways, women play an important role in men in, e.g. Kanyama township, whereas almost as many household food supply: through their productive labour; men as women practise agriculture in George. The lowest through theirdecision-making aboutproduction, consumption level of involvement in urban cultivation was recorded in and divisionof food; and through the income they generate, this township.Originally a squatter settlement, this is a very which is often used to buy food. Variousstudies have shown densely-settled area with nospace for growingstaples. Some that women's income has a greater positive impact on the people rent a piece of land from farm owners near the health and nutritional statusof the children than does men's townshipfor wet-season plots, and pay in cash or kind (e.g. income(Maxwell and Smith, 1992). bags of maize or groundnuts). Gardening on termite mounds in natural wetlands Kanyama is a squatter settlementin southwest Lusaka. (dambos) is a specialtype of wet-season microfarming.One Thereis afair amountof open space aroundit, butagricultural example is the so-called 'UNZA Dambo' south of the use is limitedby the many rock outcrops.Here, people shift Universityof Zambia(UNZA) and adjacent to Kalingalinga to roadstrip cultivationalong old MumbwaRoad. Olympia township. Termitemounds provide ideal sites for gardening Park and Materoaremedium-density areas,where cultivation because they are not influenced by groundwater and are is possible both between the houses and in nearby open thereforenot waterlogged in the wet season. The activity of spaces.In both townships,involvement ofinhabitants in wet- the termitesprovides nutrients to the crops. Table 5 gives an season cropping is very high. overviewof somewet-season plots at UNZADambo. Sweet

Table 5, Land-Use on WetSeason Plotsat UNZA Dambo Field No. Size (m2) Maize Pumpkin Sweet Potato Okra Observation 32 24 • Ridge bed 33 140 34 189 35 196 36 520 37 560 Averagesize: 321 m2 Source: Compiledby the author

236 !.i,Iu,ne II Nos. 2 and 3. Mayand August. 1996 Urban Microfar,ninin Central Southern .4friea:.•% Cccv Stud ofLusaka. Zambia

Figure 4. Planting distances in maize on different sites in Lusaka (averages of 10 single measurements. l93/II94.wet seuson

98un 84n M M

aizefield at the University Danbo X MM 67 n 100 an 4______Maizefield at ' Kanyaula —f..--.--_.- = i maize plant Maizefield at thainana Hospital

237 A. W. Drescher AfricanUrban Quarterly

Figure5, Natureof cultivation by percapita income

50 40. no 30 0 Plot garden only • . .. 20. . B 10.

0 ______I I — I I 012 12-25 25-40 >45 Monthly per capita income (in Zambian Kwacha) (1 Kwacha = US$ 0.75 in 1980)

(Source:Adapted from Sanyal, 1985)

238 Volume 11 Nos. 2 and 3, May andAugust, 1996 Urban Microfarmingin Central Southern Africa: ACase Study ofLusaka, Zambia

Figure 6,Percentage ofrespondents practisingurban staplefood production

Urban Agriculture

Average ______I I I Matero ______— I I ______O'ympia Park I I ______Kalingalinga ______U, I ______Chilenje ______-J E Mtendere ______[ifernale (%) Ornate(%) Garden ______-j U ] Kanyama ______-J I ___ I Jack ______-J Chawarna ______—J • I Mandevu ______George ______

o 10 20 30 40 50 60 70 80 90

239 A. W Drescher AfricanUrban Quarterly potato is often planted on ridge beds. This keeps the roots season homegardening become obvious when figures 7 and above the waterloggedarea and enlarges the space for tuber 8 are compared. Nearly 30 percentof therespondents practise development. Sweet potatogrowing appears to haveincreased irrigatior?or watering in thedry season,i.e. they keep gardens. in recent years. There is a direct connectionbetween wet- In total, 32 percent of the womenand 24 percent of the men season cropping and humegardeningwith respect to sweet practise gardening. In Chilenje and Matero, nearly half of potato: the garden serves as a nursery for the plant, which is the women practisegardening, comparedwith only about 25 transferred to the fields at the beginningof the wet season. percent of women in other townships like Matero, George As reportedby Hardon-Baars(1990), sweet potato is used in and Chawama. Kanyama is a special case with very little The Philippinesand in Indonesia as a buffer crop for food gardening activity. Here, on account of the rock outcrops, security because they can remain in the soil for a longperiod the high populationdensity and the poor water supply, only and are harvested directly for consumption. The termite 10 percent of the women and 7 percent of the men do any mounds have been cultivatedby Kalingalingaresidents since gardening. the 1950s. They are used only in the wet season. Of the Homegardening comprises various activities with unskilled workers (cleaners and securityguards), 76 percent different demandson labour and time. These activities are: sell part of the produce, while only 9 percent of the skilled clearing/preparing the garden bed; sowing in the nursery; workers do so (Jaeger and Huckabay, 1986). These figures transplanting ofseedlings; weedingand watering; harvesting; clearly showthe differences betweenlow-income and higher- and processingand marketing.The tasks of men and women income groups and the importance of wet-season urban do not differgreatly, with the exception of fencing, which is cropping for income generation. There is no established amale task. Fences must berenewed as often as once a year, method of obtaining a termite mound. In the case of depending on thematerials used. Preparing the garden bedis undevelopedland, it is "first come, first acquired" (Jaeger heavy work, especiallyif the plot was not cultivatedduring and Huckabay 1986). Near the University buildings, the the wet seasonand must be cleared again. Clearing is mainly arable area is often occupied by the employees of lecturers. amale task. Younger children provide valuable labour. They In this case, the fields are considered as belonging to the start to help with wateringthe gardenat theage of 9—10 years accommodations, and the owner is the authority who (sometimes even younger), and also help in weeding. On informallyallows their use. average, the respondentsspend about 15 hours per week on gardening. Here, there is a significant difference between Urban Vegetable Production women and men: men active in gardening do this for an The broadest definition of the homegarderi "as a average of almost 20 hours per week, whilewomen gardeners distinctive system of agriculturalproduction carried out by spend only 12 hours per week in the garden. households" is given by Brownrigg(1985). The homegarden A relationship was also found between the hours spent is classified as a separateand unique system of agricultural on gardening and the aim of gardening: the more production. Observationsmade duringthe fieldwork revealed commercialisedthe activity, the more hours are spent on it. that Zambian homegardens differ considerably from No correlation could be found between size of garden and homegardensas described in mst literature. Seasonalityof hours spent gardening, possibly because people with larger climaticconditions is one reason for these differences.Most gardensfound ways to make the work less labour-intensive. reports on homegardens are based on data collected in the Ifwater or labouris lackingat crucial periods, the gardening humidtropics, where seasonality is less prevalent. Most, but activity may be abandoned. The main gardening activities not all, definitionsof homegardenslimit their location to a take place in the dry season. In the wet season most of the place near the residence. They are sometimesalso called gardensare transformed into crop fields, mainly with maize. "house gardens" (Thomson, 1954) or "kitchen gardens" The year-round labour calendar is as fjllows: April/May: (Brownrigg, 1985). Such definitions exclude agricultural preparing the garden and planting; June: planting, production located at somedistance from the home.There is transplanting, start of watering; July: planting, watering, little information in the literature about the location of weeding, harvesting of the first crops; August/September/ homegardensin tropical unimodalclimates, i.e. withone dry October: watering, weeding, harvesting and November/ season and one wet season. Establishmentof a homegarden December: harvesting, preparing the fields and planting in the dry season requires the existence of a water source staples (transformationof gardens to crop fields). near the homestead. In urban areas, some householdshave The workingcalendar shown in figure 8 is based on that accessto water near or even insidethe house. Depending on developedby Jaeger and Huckabay (1986) for wet-season the economic status of the area (varying from high-cost cropping of termitaria fields in Lusaka. In principle, wet- housingto squatter areas), thewater supply comes from piped season cropping in other parts in town follows the same water in thehomes or from communalwells. Ifspace allows, calendar; even the crops are the same. New crops of a householdcan—in principle—usethe area for cultivation. importance were added to the calendar,as well as data on To avoid further discussionsabout distanceto the house and dry-seasongardening derived from field observations.As it size of the garden, the homegarden is defined here as an was not possibleto mention all the plants cultivated in the essential activity of the household, no matter how far the dry season, only themost important ones areindicated. Figure garden lies from the home. 8 shows that theagricultural cycle continues year-round. Wet- The differences in involvementof the urban population seasonfarming starts with field preparation in mid-November, in wet-season cropping (staple food production) and dry- just beforethe first rains are expected.At this time, watering

240 Volume 11 Nos. 2 and 3, May andAugust, 1996 Urban Microfarmingin CentralSouthern Africa: A Case Study ofLusaka, Zambia

Figure7, Percentage ofrespondents practising urban (vegetable) gardening

Urban Gardening

Average Matero Chilenje

e Kalingalinga Dfemale(%)t Olympia Park a •male(% E Jack Mtendere Chawama I I Garden Mandevu

George Kanyarna —II 0 10 20 30 40 50 60 70 80 90

241 A. W Drescher African Urban Quarterly

Figure8, Working calendarfor urban micro farmingin Lusaka,Zambia

of :::: A Wateringand fli Preparrng fields / gardens Planting Weeding grasses Harvesting Ill forcuation seeds and loosening ofsoil weeding

NB: Design anddata on wetseason adapted fromJaeger and Huckabay(1986) and compiled withgardening data from field observations

242 Volume 11 Nos. 2 and 3, May andAugust, 1996 Urban Microfarmingin Central Southern Africa: A Case Study ofLusaka, Zambia and weeding of the gardens stops and the last crops are south ofLusaka. The other halfcomes from boreholesin and harvested. After the rains have stopped in April/May, the near the town. In principle,the supply system is designedto preparation ofthe gardens starts. Somegardeners start earlier, servea maximum of 300,000people. The company has about some later, depending on labour availabilityand the yield of 35,000registered clients,or 3 percentof the townpopulation. the wet-season crops. If the yield was good, people start Nearly halfof the water islost on its way to theclients, either gardening later; some explained that they were tired of by leakage or theft. More than 80 percent of the population fieldworkafter the rainy season. Decemberand January are uses water without payment, also for wateringgardens. The difficult months with respect to food security. Garden water supply is irregular and unreliable and, during drought products and staple food reserves came to an end, and the periods, it is liable to breakdown. In 1992a large partof the fieldcrops arenot yet ready to beharvested. Leafy vegetables populationof Lusaka was on a mass movementin search of are alwaysthe first to appearin thecycle: leavesof pumpkin, water. The average size of the gardens in the survey area of bean and sweetpotato in the wet season(available beginning Lusaka is 184 m2 acrossall samples, with alarge range from of February) and leaves of rape and pumpkin in the dry 17 to 865 m2. Nearly 90 percent of the gardensare less than season. 300 m2 (55 percent less than 150 m2) and all of them are generally very near (< 20 m) the house. One main group of Land Availability and Dry Season Water Supply non-gardeners could be identified: urban inhabitants of and water are the basic for Space supply requirements townships with high populationdensity. activities. Inside the urban the are garden centre, gardens Thecase of Kanyamais a strikingexample for the limited on the house but sometimes a mostly plot off-plotalong path accessto basic resourceslike land and water.This township or dust road. The area on the size of garden mainly depends is one of the most densely settled areas of Lusaka, without the and varies from 70 to several hundred houseplot, square outdoorspace and water for gardeningand with very limited metres. The ill-defined land and the tenureship illegal land resources for wet-season farming because of the rock characterof someof the urban the compoundspartly explain outcrops. This example also shows how the 'poorest of the insufficient and/or of resourcessuch as availability security arecut off from basic needs like gardeningbecause of land and water. These a poor' legal aspects represent major limitedresources. Household as a strategy limitation to activities. Pen-urban is gardening coping agricultural agriculture for survival cannot be practised here. Even if tools, money also limited the of official and unofficial being by sprawl and seeds were available, these families would remain residentialareas and commercialactivities. Schultz Already, vulnerable because of limited land and water resources. (1976) out that the of cultivation"around the pointed "ring Another of non-gardeners are Female Headed central urban zone is likely to be pushed outwards by this group Households (FHH). This group is in our process of urban expansion. Schlyter's very detailed study underrepresented survey. This indicates that FHH are not much involved in of George Compound, Lusaka showed how this process land for inside gardening in Lusaka. Literature shows, that there is a destroys gardening townships (Schlyter, 1991). of FHH in the of The only areas where farming has increased is where disproportionate representation group livestock householdsin rural areas in Zambia (Carr, roadstripswere created by road construction(although also keeping that, it seems that due to serious labour this type of farming is illegal). At the same time, population 1991). Beyond FHH concentrate more on activities pressure on resources is increasingand will probablycause constraints, generally which can be carried out near the house but do not serious social problems in the future.According to Schlyter require too much work. (1991), in 1969 61 percent of the residents of Old George Compound practised gardening on plots of about 40 m2 Other Constraints: Skill average size. This figure has significantly decreased to 42 Input Resources, Development, Theft percentgardeners in 1989,cultivating plots of 12 m2 average size only. Results of our survey showed that only about 20 Seeds for the garden are obtained from different sources: percentof the respondents in this township now have gardens. last season's harvest; purchased (from commercial Many gardens disappeared because of the increasing institutions, NGOs, street vendors or other households); gifts. population,the building of infrastructureand changesin the Exotic vegetables dominate in gardens, and most of the constructionpattern (figure 9). households buy these seeds. Urban gardeners depend on Availability of water is one of the main limits to purchased seeds only, or on a combinationof last season's gardening. Waterfrom varioussources is used. In urbanareas, seedsand purchasedseeds. Very rarely do they use only seeds piped home water, (communal)taps and (communal) wells collected from previous harvests. More than 75 percent of are most common.In Lusaka, a strong relation can be found the respondentsuse animal dung to fertilise the vegetable between the economic standardof the townshipand the type garden. Less than 45 percent use mineral fertilisers. Use of of water supply: inpoorer townships, there is eithercommunai compost is uncommon. A few people try to prepare some watersupply or none atall, and peopletry tohelp themselves sort of compost,but generally the quality is very low, because by digging wells, where possible. all kinds of garbageare mixed with the organic matter. Inside Thewater supply of Lusaka is incumbentupon theLusaka town,poultry is kept by about 55 percentof therespondents. Water and SewerageCompany, whichhas theduty todeliver Animal husbandrydoes not appear to play an importantrole waterthroughout the town and toorganise the sewage system. withinthe groupof homegardeners, but this does notexclude Nearly halfof the water comesfrom the KafueRiver, 60 km animal husbandry as an importanturban activity. The survey

243 A. W Drescher AfricanUrban Quarterly

Figure 9,Decrease of gardens in George Compound from 1969to 1989

George - Compound I Lusaka

1968 1977

1980 1989 Houses Homegardens

(Source: Adaptedfrom Schlyter,1991)

244 Volume 11 Nos. 2 and 3, May and August, 1996 Urban Microfarmingin Central Southern Africa: ACase Studyof Lusaka, Zambia

was designedto cover gardenersrather than peoplepractising Income is generated by marketingsome garden products. other forms of urbanmicrofarming, such as livestock keeping. The strategies vary: Direct sales to neighbours,barter, and Gardeningrequires specific skills to bring knowledge into sales on wholesaleor retail markets. Only 17 percent of the practice. Since manyplants found in the gardensdiffer from gardeners sell on markets.The followingreasons were given those foundin the wet-season fields and includeexotic crops, for notdoing this. "Donot produce enough". The households the development and/or acquisitionof relevant knowledge have small gardens and produce just enough to feed the and skills is important to urban gardeners. For all the family. Any surplus is notbig enough to be sold at themarket. householdsin the survey, parents and relativesare extremely The surplus is soldor givento neighbours, relativesor friends. important sources of experience. Schools or courses are "Problemswith marketing" was mentionedmainly by female second in importance.Own experience (experimenting) is gardeners, who referredto problemsof lack oftransport, lack another importantfeature. Extensionworkers do not play an of time, distance to market or lack of market. It is quite important role in conveying knowledgeabout grdening. It difficultto obtain precise information on incomeearned from seems that this activity is barely noticed by the extension the garden,for variousreasons, namely, the gardenersdo not services. sell regularlyand thereforedo nothave aregular income over More than one-third (38 percent)of the urbanrespondents the year; the cash earned is often spent immediately at a had problems with theft of their produce.Urban women are market or nearby shop; people do not like to tell how much more affected by theft. Forty-sevenpercent complain about they earn, often because they are afraid of other people's it while only 29 percent of men do. The frequency of theft jealousy; not everythingis sold for cash, part of the produce increased with greater distance of the garden from the is also bartered, e.g. for clothes; and because harvesting is homestead. Similar observations were made by Jaeger and continuous,it is difficultto estimateyield. Huckabay (1986). Some households have found ways of In 'most cases, incomefrom gardening is not spread over preventingtheft: fencing to make the garden less accessible the entire year. It is restricted to the dry season. However, if and even sleeping in the garden to guard it during the the information given by respondentsis expressed in terms harvestingperiod. In general, the respondentsfind it difficult of income per month, people selling on markets have an to take measuresagainst thievesbecause of the relationships average incomeof about ZK 11,700/month and non-market and commitmentsbetween peoplein the quarterand because sellers about ZK 5,700/month. At the time of the survey, the of the fear of witchcraft. exchangerate was ZK 560 to US$ I. Thus, at least US$ 20 per month could be gained by gardeners who sold on the HomegardenOutput and Income Generation market. There are considerable differences in the income All the households stated that garden produce is used gainedby women and by men: the averageincome of women mainly for home consumption. Sixty-seven percent of the is nearly ZK 9,100/monthwhile the averageincome of men urban households sell part of the products to get additional amountsto ZK 15,500/month. The householdsselling on the income, and nearly 100 percent give some of the produce market earned more money than those doing informal away. Gifts to other households are a common way of marketing.This can be explained by the fact that people start maintaininggood relations and help to strengthenthe 'social goingto the marketonly when they produce enoughto make security' networkof the household. Comparing the above- this worthwhile. The figures presented here are rough and mentioned farmers' information with our field observation there are some doubts about the reliability of the data, as and measurement of individual plants in the gardens it is they are basedonly on gardeners' responses,but they give at assumed that indeed more than 50 percent of the urban least an indication of the contribution of gardening to gardeners produce is partly market oriented. The main food household income. The market survey carried out in 1992 plants found in the gardens are vegetablesand fruits (table and 1993 proved that prices on the retail market in high 6). Tomatoes were found in 74 percent of the gardens, density squatter areas are much higher for some products followedby rape (67 percent),chinese cabbage(59 percent), (e.g. tomatoes and onions) compared to other markets. onions (50 percent), pumpkins (41 percent), maize (35 Vennetier (1988, cited by Mougeot, 1994) considers the percent), sweet potato (27 percent), cabbage (24 percent), microretailing of food as a strong price-hiking factor in sugarcane and spinach (23.5 percent). Less important are African cities. This is caused by the complex marketing carrots, beans and lettuce. Most important fruit trees are system and the low purchasing power of the poor retail papaya and mango (32 percent), followedby peaches (23.5 marketers. Consequently, the access to food of the poor percent) and lemon (14.7 percent). householdis reduced.The main problemfor these people is

Table 6, Homegarden Produce Choumolier/Kale Cleome gynandra Watermelon Shallots Rape Amaranthusspp. Pumpkins Onions Cabbage Green beans Baby marrow Lettuce Chinese cabbage Green pepper Spinach Papaya Cauliflower Irish potatoes Tomatoes Peaches Sugarcane Sweet potatoes Cassava Mango Maize Chilli Carrots Lemon Source: Compiledby the author

245 A. W Drescher African Urban Quarterly access to land and water.Population pressure is very high in extensionservices for urban smallholders, improvementof these townships, leaving almost no space for gardening. water supplyin townships; regionalor even local differences Accessto land in the environs of the townships is restricted in urbanmicrofarming: case studiesof differenttowns tolook by physical conditions, such as rock outcrops, which make at the possibilities of microfarming, inputs and outputs, soil cultivation impossible. Land availability is further problems and potentials; ecological impact of urban restrictedby uncertainland tenureship.Those households are microfarming: benefitsand disadvantages, utilisation of waste not able to create neither the buffer of homegardens nor the and sewage for microfarming purposes;and the gender issue: staple food productionfor householdfood security. the role of women in microfarming and household food security in the urban environment, coping strategies of Recommendations for Further Research and Action female-headed households, intrahousehold allocation of Although urban microfarming appears to bring resources, intrahousehold distribution of food and income, advantages to thehousehold as a whole, little is known about motivations, decision-making processes, relationship between the actual distribution of the food and income among illness and householdeconomics. householdmembers. Further research into this is neededin One most crucial point is theperception of unimportance order to determine the nutritional impact of microfarming, of urban agriculture,which is a psychological problem of preferences of urban consumers and impacts of possible the decision makers in policy and councils. Therefore changes in diet on nutrition. More insight is needed into the awareness-creating of decisionmakers and impowermentof actual decision-makingamong household members about city farmers is the first step to improve the situation. Even microfarming. It will be important to gain a better city farmersare not awareof the importanceof theiractivity understandingof controlof resourcesused inthese activities for urbanfood security. This seems to be one of the reasons and control of incomefrom them.In-depth interviews or case for the relatively small output of this system. Promotionof studies could be appropriatemethods for investigating such urban agriculture presumes action on different levels: the issues (Bos, 1994; Drescher and Bos, 1994; Dennery, this farmers themselves (empowerment and training), the issue). Through the fieldwork, it was noticed that illness extensionlevel (training)and thedecision makers and urban seriously affects agriculturalproductivity and the capability planners (changing perception and creating awareness). of poor familiesto cope with problems.Many illnessessuch Transferof researchfindings to implementingauthorities and as malaria are still prevalent, and AIDS is on the increase. decisionmakers stipulates beforehand closecontacts between This can seriously affect the functioningof the households. the twoparties. Therefore decision makers areone important Moreresearch is neededinto the relationship betweenillness targetgroup for special workshops on urbanagriculture. There and household economy (Bos, 1994). It is vital to any is urgentneed of a different approach to support femaleurban understanding of householdcoping strategies and, ultimately, farmers. This group needs special attention through well to the effective design of food-security strategies that the trainedextension servicesand external support structureslike relative importance ofdifferent income sources, their seasonal e.g. credit facilities. Future action should therefore include fluctuations,and sustainability, as well as the responses of workshops on urbanagriculture and foodsecurity for decision individuals and households, be well understood (Maxwell makers and planners, farmers and extension services. 'On and Smith, 1992).Policy decisionsshould no longerrevolve garden' researchis one important step towardsempowerment around external modernisation (and commercialisation) of of urbansmallholders. Thisfact gotobvious during fieldwork small-scaleagriculture, but rather face the questionsof how in Lusaka. Therefore more research with a participative to integrate appropriatetechnologies and new management extensionapproach is needed in this field. One step further strategies.Schulthes (1992) defined integration as the key to can be the support of small-scaleenterprises for vegetable successfulmanagement and clearly points out how difficult production on commercial basis and the establishment of this appears to national bureaucracies.Legalising squatter appropriate training institutions for vegetablegardening. townships and the establishment of site-and-service-and The basis for urban microfarmingis the availabilityof upgrading-projects are possible ways of integration, as resources. The potential for developingthe urbanagricultural proposedby Gaebe (1994: 570—576). Thequestions and, still sector is high. In most cases, the formal economic sector in more important,the answers of themicrofarmers themselves urban centres of the tropics is underdeveloped and, as a totheir problems can significantly contribute tounderstanding consequence, does notprovide adequate income for theurban and improvingtheir practices and systems (Drescher, 1994). dwellers.Urban microfarmingoffers an alternative activity Many authors agree upon the need for more researchon and a buffer for household food security. It is based on urbanmicrofarming. Someessential questions are: the quality traditional activities of the indigenous people, and is only and quantityof urban resourcesfor microfarming: access to now being discovered by development agencies and resources (land tenureship, who cultivates where), quality policymakers as a possibletool for development.The intra- and quantityof water resources,availability of land, access urban land-use systems and, more important, the toarable land, qualityof intra-urban soils;ways of promoting socioeconomic and sociocultural frameworkof microfarming a 'self-help' approach to resource management and as partof theurban food securitysystem arehighly sensitive cooperation between urban microfarmers; how to change issues.Therefore, any actions initiatedfrom outside have to policy attitude towardsurban microfarming: appropriate town be discussedcarefully with all stakeholders. planning, support programmes for gardening, provision of

246 Volume ii Nos. 2 and 3, May and August, 1996 Urban Microfarmingin Central Southern Africa: A Case Studyof Lusaica, Zambia

References Bos, F. (1994). The Role ofHousehold Gardens in Household Greenhow, 1. (1994a). Urban Agriculture in Lesotho, Food Security. Reeks publikaties van de vakgroep Botswana andSweden: Some Brief Observations. Stockholm: Huishoudstudies,No. 14. Universityof Wageningen. Swedeplan.

Bowa, M. et al. (1979). Gardening in the City. Lusaka: Greenhow, T. (l994b). Urban Agriculture: Can Planners ZambiaInstitute ofAfrican Studies, Universityof Lusaka. Make a Difference? Ottawa: Cities Feeding People Series Report 12. International Development ResearchCentre. Brownrigg, L. (1985). Home Gardening in International Development:What the Literature Shows. Washington DC: Hardon-Baars, A.J. (1990). "Household SystemsPerspective Leaguefor InternationalFood Education. toAgricultural Research and Development: A MissingLink". In User's Perspective With AgriculturalResearch and Can, M. (1991). Womenand Food Security—The Experience Development (UPWARD). Baguio City, Philippines. of the SADCC Countries. IT-Publication. Jaeger, D. and J.D. Huckabay (1986). "The Garden City of Deutsche Stiftung für International Entwicklung/ Lusaka: UrbanAgriculture", in Williams, G.J. (ed.) Lusaka Arbeitsgemeinschaftfür Tropische und Subtropische and Its Environs. Lusaka: Geographical Society of Zambia. Agrarforschung(1994). Vegetable Productionin Pen- Urban Areas in the Tropics and Subtropics: Food, Income and Lamba, D. (1993)."The Neglected Industry of Kenyan Quality of Life. Summary Report on Workshop held in Cities", IDRCReports Vol. 21 No. 3. Zschortau, Germany, from November 14 to 17. Lee-Smith, D. and D. Lamba(1991). "The Potentialof Urban Drescher,A.W. (1994). Urban Agriculture in the Seasonal Farming in Africa", ECODECISION, December. Tropics ofCentral SouthernAfrica: A Case Study ofLusakal Zambia. Contributionto the "InternationalPolicy Workshop Maxwell, D.G. (1994). "The Household Logic of Urban on Urban Agriculture: A Growing Development Tool", Agriculture in East Africa", in A.G. Egziabheret al. (eds) UniversityCollege London (NRI/CPU), 29 June. Cities FeedingPeople: AnExamination of UrbanAgriculture in EastAfrica. Ottawa: International DevelopmentResearch Drescher, A.W. (1995). Afrikanische Hausgarten als Centre. Gegenstand angewandter physisch: Geographischer Entwicklungsforschung. APTReports No. 1. Dept. ofApplied Maxwell, D.G. and S. Zziwa (1992). Urban Agriculture in Physiogeography of the Tropics and Subtropics (APT), Africa: The Case of Kampala, Uganda. Nairobi: African Institutefor Physical Geography, Universityof Freiburg. Centre for Technology Studies.

Drescher, A.W. and F. Bos (1994). Report on the Zambian Maxwell,S. (1990). "FoodSecurity in Developing Countries: HomegardenProject andthe Rural HouseholdFood Security Issues and Options for the l990s", IDS Bulletin Vol. 21 No. Survey. Publication of the FAO Project FAHNIS "Food, 3. Healthand NutritionInformation System", Lusaka,Zambia. Maxwell,S. andM. Smith (1992). Household FoodSecurity: Due,J.M. (1985). Womenmade Visible: Their Contributions A ConceptualReview. Sussex: IDS. to Farming Systemsand Household incomes in Zambia and Tanzania. Illinois Agricultural Economics Staff Paper, Mougeot, L.J.A. (1993). "Urban Food Self-Reliance: Urbana. Significance and Prospects",IDRC Reports Vol. 21 No. 3.

ENDA-Zimbabwe(1994). Urban Agriculture in Harare. Mougeot,L.J.A. (1994). Urban Food Production:A Survey Harare: Report on an IDRC-SupportedProject. ofEvolution, Official Support andSignificance (With Special Reference to Africa). Paper Presented at the "Habitat '94", Frankenberger, T.R. (1992).Indicators andData Collection Edmonton,September 20. Methodsfor Assessing Household Food Security. Tamnal College of Agriculture,University of Arizona, unpublished Purseglove,C. (1985). "A Household Economic Approach report. to Malnutrition and Poverty in Rural Africa", in Badir D.R. (ed.). Proceedingsof the FirstAll Africa Home Economics Fresco, L. (1985). "Vrouwen en Voedselvoorziening. Meeting. Accra,Ghana. Hoofdstuk 6", in Honger op ons Bord: Over Politiek en Voedselveiligheid. Amsterdam: Evert Vermeer Stichting. Pursglove, J.W. (1988). Tropical Crops: Monocotyledons. Essex: Longman Gaebe, W. (1994)."Urbanisierung inAfrika, Geographische Rds Vol. 46 No. H.1O.

247 A. W. Drescher African Urban Quarterly

Ratta, A. and J. Smit (1993). "Urban Agriculture: It's Much Streiffeler, F. (1994). Socialand Political Problems ofUrban More than Food", Why Magazine. No. 11. Agriculture. Contribution to the International Expert Consultationon "Vegetable Productionin Pen-Urban Areas Sanyal,B. (1985). "Urban Agriculture:Who Cultivates and in the Tropicsand Subtropics: Food, Income and Quality of Why?A Case-Study of Lusaka",Food andNutrition Bulletin Life", held from November 14 to 17, 1994, in Zschortau, Vol.7 No. 3. Germany.

Schilter, C. (1994). Possibilities and Limits of Urban Thomson,B.P. (1954). Two Studiesin AfricanNutrition; An Agriculture (the Case of Lomé). Paper Presented at the Urban and a Rural Community in Northern Rhodesia. "International Policy Workshop on Urban Agriculture: A Manchester: The Rhodes-Livingstone Papers, No. 24. Growing DevelopmentTool". University College London ManchesterUniversity Press. (NRI/CPU), 29 June. The Urban AgricultureNetwork (1994). Urban Agriculture DC: The Urban Schlyter, A. (1991).Twenty Years ofDevelopment in George, InformationBrief Washington Agricultural Network. Zambia. Stockholm: SwedishCouncil for BuildingResearch. van der L. In the Shadow Lusaka: Land and Schlyter, A. and 1. (1980). George: The Berg, (1982). of Schlyter Development under Pressure Urban Growth.Lusaka: Studies in a Settlement in Lusaka, Zambia. Stockholm: People of of Squatter Zambian 6. of Zambia. Swedish Council for Building Research. Society, University van der Bliek, J.A. Urban Possibilities Schulthes, J. (1992). "Neuorientierung in der (1992). Agriculture: for Ecological Agriculture in Urban Environments as a Ernahrungssicherung",Entwicklung + LandlicherRaum 4. Strategyfor Sustainable Cities. Netherlands, Leusden: ETC. J. "Land-Usein Zambia: PartI: The Schultz, (1976). Basically Vennetier, P. "Urbanisation, et Traditional Land Use and Their Part II: (1988), production agricole Systems Regions. autosuffisance alimentaire: Réflexions sur le Cas Africaine", Landuse Map". Afrika Studien 95, Weltforum Verlag, Cahiers d'Outre-Mer Vol. 41 No. 63. MUnchen. Zambia, Governmentof (1990).Zambia in Figures. Lusaka: Smit, J. et al. (1996). Urban Agriculture: Food, Jobs and Central Statistical Office. SustainableCities. New York: UNDP. Zambia,Government of(1992). Zambia in Figures. Lusaka: Streiffeler, F. (1993).General PrinciplesandApp roaches for Central Statistical Office. Sustainable Urban Greenbelts with Special Reference to Africa. FAOreport.

248 estudios de ASIA y AFRICA VOL. XXXII, SEFTIEMBRE-DICIEMBRE,1997, NUMERO 3 104 ARTICULOS

ElisabettaCorsi Términoschinos utilizados come sinónimo de "perspectiva"ylos ecos del naturahsmo en algunos textossobre pintura del siJoXVIII Hilda Varela 'Nochedegran lunay destinoinorado!' La historiapolitica deIa Repüblica deSão ToméyFrlncipe (1975-1996) Romer CornejoBustaniante Confucianismoy desarrollo económico Simone A. Nguyen Dac Economia rural vietnamita:Quinto plan quinquenal (1991-1995)

ESTUDIOS DE ASIA Y AiRICA es una publicación cuatrimestral de El Colegio de Mexico, A.C. Suscripción anual en Mexico: 100 pesos. Estados Unidos y Canada: individuos, 20 dólares; instituciones, 30 dOlares. En Centro y Sudamérica: individuos, 20 dolares; instituciones, 30 dOlares. En otros paises: individuos,25 dólares; instituciones, 35 dólares. Si desea suscribirse, favor de enviar este cupón a El Colegio de Mexico, A.C., Departamento de Publicaciones, Camino al Ajusco 20, Pedregalde Santa Teresa, 10740 Mexico, D. F. Adjuntocheque o giro bancario nám.: ______For Ia cantidad de: A nombre de El CoIeio de Mexico, A. C., conio importe de mi suscripción por un año a ESTUDIOS DE ASIA V AFRICA. Nombre: Dirección:______Código postal: Pals: ______Estado: ______Ciudad:______EL COLEGIO DE MEXICO African Urban Quarterly, 1996: 11(2 and 3) pp. 250—257 C1996 by African Urban QuarterlyLtd.

EPURATION ET REUTILISATION DES EAUX USEES DOMESTIQUES EN MARAICHAGE PERIURBAIN A DAKAR, SENEGAL (Purification and Re-Use of Domestic Wastewater in the Suburbs of Dakar, Sénégal)

Seydou Niang Chercheura l'lnstitutFondamental d'Afrique Noire (IFAN) Ch.A.DIOPDakar, Sénégal

Résumé Accepted January, 1996 Depuis son apparution au Sénégal en 1937, le maraIchagea globalementconnu une evolution, tant sur le plan des surfaces utilisées que sur Ic plan des rendements. Malgré tout, les besoins ne sont pas satisfaits et le Sénégal continue a importer des legumes. En effet, avec les années successives de sécheresse, Ia principalesource d'approvisionnement en eau que constituait les Niayes a montré ses limites. Des lors, les eaux usées se sont présentéespour les maraIchers, comme une alternativeassez intéressantepouvant comblerIc deficiten eau. Les caractéristiquesphysicochimiques et biochimiquesdes eaux usées, compares a ceux des eaux de céanes (puitspeu profonds) eta ceux du réseau d'eau potable démontrentque les eaux usées sont trés intéressantespour le maraIchage. L'analyse de I 'impactde I 'utilisationd 'eaux usées dans lemaraIchage inontreune economicd'eau et d'intrants. Cependantsur Icplan bacteriologique, les concentrationsnotées en coliformeset streptocoquesfécauxfontque ces eaux ne sontpas aptes, selonles normes OMS, a l'irrigation en maraIchagesans restriction. Ainsi, lanaiyse parasitologique des legumes arrosés avec des eaux usées brutes montre que ces legumes, prêts a Ia vente, sont charges d'ieufc etde larves deparasites qui sont capables de transmettredes maladieschez les consommateurs, d'oà Ia nécessitC d'épurer les eaux avant réutilisation. En matière de traitementdes eaux usécs dans lespays en développement, les techniquesextensives d'épuration des eaux usées constituentlesprocédés les plus a laportéedes budgetsnationaux. Mais du fair de / 'utilisation de végétauxdans ces processusd'épuration, les eauxproduites sont, suivant Ia technique utiiisée, soitde faible quantité, soit dCséquilibrées (rapport azote-phosphore). De plus, comme Ic montre les résultats obtenus a Ia station expérimentale de C'a,nbérène, aucundes systèmes ne respecte les normes bactériologiques de I 'OMSpour Ia réutilisation sans restrictionen maraIchage.Les recherchesdoivent donc s 'orienter vers une optimisation de ces techniquesd 'épu ration extensive en vue dune réutilisationen maraIchage.

Mots Clés—épuration,caux usées, maraIchage, intrants,parasites, coliformes fécaux,lagunage, macrophytes,microphytes, marais

Abstract This article reports on an experimentin progress. Since first noticedin Sénégal in 1937, horticulturein the country has grownboth in terms ofcultivated acreage and cropyields. Nevertheless, demandcontinues to exceeddomestic supply and the country still imports vegetables. With successive years of drought, the main sourcefor water supplyfor local crops, the Niayes, has reached its limits. Wastewaters are now perceivedand used by vegetable growers as a valuable alternative to make upfor the deficitfrom traditional supply sources. Physiochemicaland biochemicalqualities of wastewaters do put them at a clear advantage for use in horticulture, when compared to deep-well water and that which is supplied by the drinking-watersystem. An analysis of wastewaters'impacton vegetable farming does point to savings by users in both water and other inputs. Howevet observed concentrationsof colforms andfaecal streptococcimake these waters unsuitedfor unrestrictedirrigation ofvegetables. Parasitological analysesof crops watered with untreated wastewatersshow that ready- for-saleproduce is loaded with eggs and larvae ofpotentially disease-transmitting parasites. Thesewaters therefore should be treated beforethey are reused in crop irrigation. Land extensive processesare thosewhich remainwithin the reach ofmost national budgets, but these make use of vegetableagents and the treated waters, dependingon the technique, are produced onlyin limitedquantities or exhibitimbalanced nitrogen-phosphorus ratios. Moreover as tests at the Cambérèneexperimental station show, none of the systems studied meets WHO bacteriological norms. Research underway aims at optimisinga combination ofavailable techniques to meet horticultural irrigation requirements.

Key Words—purification, wastewater, marketgardening, input,faecal coliforms, lagooning, macrophytes,microphytes, marsh

Introduction I'installationd'un paysanat(Kane, 1973). Depuiscette date, C'est en 1937 que le maraIchagea dté initié au SénCgal. lemaraIchage a globalement connu une evolution,tant sur le Cette premiere operation, portant sur 2.400 hectares, plan des surfacescultivées qui sont passées de 2.400 ha en destinée alamise en valeurdes Niayes (series dedepressions 1937 a 8.876ha en 1990, que sur le plan des rendements.En interdunaires le long de Ia côte Nord du Senegal) et a 1960, ils estimés a 10,8 tlha alors qu'en 1990 us

250 Epuration et Réutilisationdes Eaux Usées Doinestiques Volume 11Nos. 2 and 3, May and August, 1996 en MaraIchage Périurbain aDakar, Senegal atteignaient 17,9 t/ha (Ngingue, 1992). Malgré tout, les Malgré tout, cette pratique pose un grave problème de besoins nesont pas satisfaitset le Sénégalcontinu d'importer sante publique. De manière genérale, ii existe quatre des legumes. categories depersonnes sur qui l'utilisationagricole des eaux Au niveaudu pays, la regionde Dakar occupeIa premiere résiduairesfait peser un risque potentiel: placedans cette activité,par Ia quantitéde legumes produite, - les exploitantsagricoles travaillant dans les champs par leur diversité et par le nombredes exploitants.De 1.000 et les membres de leur famille; hectares effectivement mis en culture maraIchèreen 1980 - les manutentionnairesdes produits de Ia récolte; (Delvaque,1980), la surface exploitée est passée a 2.506 ha - les consommateurs; en 1990 pour une production de 27.279 tonnes (Ngingue, - les personnesvivant aproximité des champs(Bartone 1992), soit 18 percent de la productionnationale. et Arlosoroffin Mara et Cairncross, 1991). Deux secteursse partagent cette production: Des cas concretsde foyers imputabies - Ic secteurtraditionnel (objet de notre c'est Ic plus a cette pratique peuvent être cites a travers le monde. Pour important(90 percentde Ia productionlégumiere du pays) exemple, en 1987, une de fièvre typhoIde et (Ngingue, 1992); paratyphoIdeAct B a dans Iaregion de Dakar. 400 cas - le secteurmoderne. répertoriés. Les enquetesepidemiologiques ont alors Depuis l'origine, Ia zone des Niayesa toujours constitué, révélé que les responsables de Ia contamination des a partir des mares qui disséminéesun peu partout, Ia maraIchers qui utilisaient des eaux insuffisamment (ou pas principale source d'eau utilisée pour le maraIchage. Après du tout) traitées pour arroser leurs legumes(Fall, 1992). les années successives de sécheresse que Ic Sénégala connues interdire cette ne serait pourtantpas seules les céanes Simplement pratique récemment, (puits peu profonds) une démarche facile a envisager car, outre Ic fait que maintenaient cette culture. cette eau douce Cependant, l'interdictionne ferait que rendre plus difficileIa surveillance sur une d'eau salée ne reposant 'loupe' permet qu'une et Ia reglementation sans résoudre Ic problème, die limitée des céanes, sous de exploitation peine pollution occasionnerait, dans Ia conjoncture economique actuelle, irreversible. d'énormes difficultés Cconomiques pour cette frange de Ia sous l'influence combinée de la Actuellement, pression et pour les pays concernés. de Ia baisse des ressources en eau Ia population urbaine, (suite a La solution seraitalors de légaIiserIc système pour mieux de celle de Ia fertilité des sols sécheresse)et (trop exploités Ic contrôler et minimiser les Pour cela, un certain assez esten nette diminution risques. et pas amendés), Iaproduction nombred' actions sont nécessaires: par rapport aux autres regions (Ngingue, 1992). - Ic traitementdes eaux rCsiduaires avant réutilisation A deces dernièresannées, les eaux usées urbaines, partir (dans Ic cas oü un réseau d'égout existe) par des parleur volume de plus en plus important(en rapport avec Ia dites extensives croissancedemographique de Ia yule) et par leur richesseen techniques fertilisants, se sont comme une ahordabies; présentées - Ia limitation de Ia réutilisation des eaux usées a alternative assez intéressante pouvant combler le deficit en certaines cultures sont les moms sensibles a Ia eau des exploitations. Danscertains endroits, dies constituent qui 1' source d' transmissiondes maladies(arhres fruitiers); unique approvisionnement. - Ia modification méthode en Ii faut reconnaltre l'utilisation des eaux résiduaires de Ia d'irrigation; effet, que on Ia méthode submersion en agriculture est une pratique qui, depuis une vingtaine peut d'irrigation par et faire des méthodes d'années, s'estconsidérablement développée, aussi bien dans (arrosoirs, asperseurs...) adopter les pays industrialisésque dans les pays en développement. d'épandageplus hygiéniques(epandage souterrain); - Ia une Cependant, ce sont les zones andes et les regions a longue limitation de l'exposition humainc par saison sèche qui ont souvent les endroits les plus education en matière d'hygieneet par Ic port de bottes et de concernés.Les statistiquesmondiales sur l'emploi des eaux gants. résiduaires pour l'irrigation (Bartone et Arlosoroffin Mara Ainsi, non seulement les risques sanitaires seraient et Cairncross, 1991) donnent les résultats suivants: atténués, mais les productions pourraient aussi être - ressources Johannesburg (Afrique du Sud), 1.800 ha; augmentees et les en eau potable - Brunswick (Allemagne),3.000 ha; A Dakar, une partie des eaux usées brutes rejetées est - ensemble des villes de la Chine, 1.330.000 ha; récupéréeet directement réutiliséepour Ic maraIchage. Les - Lubhok, Texas (USA), 3.000 ha; produits issus de cette pratique culturale sont revendus - Calcutta (Inde), 12.500ha; indistinctement dans tous les marches de Dakar, si bien que - Tunis (Tunisie),4.450 ha; toutela population de cette villeest concernéepar les mêmes - Mexico(Mexique), 90.000 ha. risques. L'utilisationdes eaux résiduaires sembleêtre la résultante Notre s'intéresseal'impact de I'utilisationdes eaux de plusieurs facteurs dont les principauxsont: usées urbaines sur Ia pratique culturale et sur Ia qualité - Ia raretécroissante des autres sources d'eau; sanitaire des legumesproduits. - le coüt des engrais artificiels; Dans la premierepartie de l'étude, nous présentonsles - l'acceptation socioculturellede cette pratique; caractéristiquesdes différents types d'eaux utilisées pour - Ia difficulté relativea convaincreles exploitantsdes l'irrigationde maraichage et l'impact de I'utilisationdes eaux risques encourus. usées urbaines sur:

251 Seydou Niang African UrbanQuarterly

- les besoinsen eau des maraIchers; Après réalisation de l'enquête, le a - questionnaire le choix du type de culture; codifiéet les données traitées al'ordinateur grace au logiciel - le mode d'irrigation; de traitement d'enquete socio-économique 'MAC SS'. - Ia quantitéd'eau utilisee; - et les besoins en intrants (engrais et produits (c)Prélèvement et analyse des eaux. Les résultatsprdsentes phytosanitaires). sur les eaux de céanes proviennent d'analyses faites sur La seconde partie de l'étude montre l'impact de quelquesprdlevements instantanCs aCambérène, complétés 'utilisation des eaux usées surIa qualité sanitaire des legumes par des analyses réalisées par Ic CDH (Centre de produits. Développementde l'Horticulture) au niveau du même site. La troisièmepartie montre, apartir des premiersrésultats Concernant les eaux usées, l'analyse a réaliséea partir obten'is alastation expérimentale d'épuration des eaux usées de plusieurs instantanéset rñoyensprélevés au par voie naturellede Cambérène,les limites de réutilisation niveau du site de Ouakam. des eaux traitées en maraIchage. Nous analysons ace niveau, Le pH, la conductivité, le résidu sec, les chiorures, les les perspectivesd'avenir pour une réutilisationefficace des MES(matières ensuspension) la DBO5(Demande Biologique eaux traitCes. en Oxygene), Ia DCO (Demande Chimique en Oxygene), l'ammonium(NH4), Ies phosphates(P04- ), Ie NKT (azote Materiels et Methodes Kjeldahl ou total) et Ic Pt (phosphore total) constituent I'ensemble des parametres Utilisation directe des eaux usées brutes dans le (1) de Ia maraIchage (2) Analyse qualité sanitaire des legumes L'étude est localisCe au niveau du site de dans (a) Choix des sites de maraIchage. Trois points ont Ouakam, l'enceinte de Sur des fraIchement ciblés, regroupant Ia majorité des maraIchers de Dakar. Ce I'aeroport'. legumes sont: récoltés, prêts a la vente, des sont prélevés. Les sont rincés dans un contenant de I'eau - le site de Cambérène-Thiaroye oü les maraIchers recipent travaillentexciusivement avec l'eau de 'céane' stérilisée. Puis 100 ml des eaux de rincage sont prélevés et (puits au laboratoire. Les ont sur la de I a 3 m de et de 4 a 5 m de analyses analyses porte profondeur largeur); contamination - le sitede Pikine-Niayes les exploitants utilisent parasitologique. des eaux usdes et I'eau de 'céane'; (3) Epurationdes eaux usées par voie extensive - Ic site de Ouakam-Aéroport oi les exploitants (a) Principe. Le principe de l'expérience est base sur la travaillent avec des eaux usées exciusivement,mais de I'efficacité réelle de oii ii des maraIchers branchesau réseau comparaison plusieurs techniques y aegalement naturelles de traitement des eaux usées a en eau de Ia SONEES (lagunages d'approvisionnement potable a souterrain, Nationale des Eaux du microphytes, lagunages macrophytes, epandage (Société d'Exploitation enplantations ligneuses,...). Ces différentsprocédés Sénégal). d'épuration sont places dans Ies mêmes conditions d'alimentationen eaux usées et sous Ic même climat2. Procedure s'est déroulée (b) d'échantillonnage. L'enquête La Station Expérimentale de Cambérène (banlieu de en trois phases: Dakar), construite suivant cc principe par Ic promoteur du - une d' un elle avait but enquêtepréliminaire jour; pour procéde, est suivie regulierementdepuis trois ans par une de discuter avec les exploitantssur l'objet du travail, de chercheurs senégalais en collaboration avec les Iaméthode defamiliariser les enquêteurs avec sites; l'equipe de ViVille (cf. notes 2). utilisée l'entretien informel; elle nous a permis defaire une premiereevaluation denotre questionnaire (b) Protocole experimental. La station experimentaleest et d'estimer la taille de notre installée sur le site de la station d'épuration classique des une deuxième enquête, dite pré-enquête,a servi au eaux usées de Cambérène.EIIe reçoit ainsi des eaux usées Ia cours d'une journée a tester le questionnaire; representativesdes rejets urbains de Dakar. Elle comporte méthode utilisée cette fois l'entretien semi- 18 bassins en tôle de I m2 répartis en six cascades de trois structure avec un premier questionnaire sur niveaux. La stationse proposede tester six types de procédés Ia base de questionsouvertes; d'épuration des eaux usées urbainespar voie extensive. Les - enfin, une fois Ia taille de l'échantillon fixée (a 100 cascades sont regulierement alimentées en eaux usées enquêtes par site), le questionnaire réajustd et prétraitées (dégrillage, dessablage et deshuilage) par une définitivement fixé a 80 questions, une liste des pompe commandéepar une minuterie. personnes quitravaillent dans chaque sitea dressée, Le premier niveau des cascades est représenté par six puis a l'aide d'une table statistique des nombres au bassins de stabilisation.Les deuxième et troisième niveaux hasard, l'échantillon a choisi;l'enquête se déroula sont organisés comme suit: alors pendant cinq jours au rythme de 20 - unecascade plantée, a Typha australis L.,représentant questionnairespar site et parjour. un marais a Typha; 300 questionnairesont administrésdans les sites de - une cascadeplantée, a Phragmitesvulgaris Schum& Cambérène,Pikine et Ouakam. Thonn.,représentant un marais a Phragtnites;

252 Epuration et Réutilisationdes Eaux Usées Domestiques Volume 11 Nos. 2 and 3, May andAugust, 1996 en MaraIchage Périurbain a Dakar, Senegal

- une cascadeplantée, a Pistia stratiotesL., representant Tableau 1, Caracterstiquesdes eaux utilisées la prairie flottante a Pistia; Table I, Characteristicsof different waters used - une cascade non plantée, représentantle lagunagea microphytes; Types d'eau d'rrigation - une cascade plantde, a Eucalyptus camaldulensis, Eaux Eau représentant1' dpandage sous plantation ligneuse; Paramètres Céane usées potable - une cascade non sur plantee, représentantl'épandage pH 3 8 7,7 sol nu. Conducticitéen ts 400 a 2800 1900 1100 Les cascades une alimentation quatre premieres reçoivent Résidussec en mg/i 300 a 2100 900 800 en eau translation au dessus du sol, tandis les deux par que Chioruresen mg/i 100a 1200 400 200 dernières sont alimentdespar percolationa travers le sot. MESen mg/I 20 - 1200 0 en 50 500 2 Résultats DBO5 mg 02/I NH4enmgN/1 22 127 I en P/I 0 16 2 (1) Qualitédes eaux P043 mg d'irrigation Coli.fécauxenN/IOOmI 0 2,8.10 0 Dans la zone periphdrique de Dakar, on observe une Strepfécauxen N/lOOm! 0 i,8.I0 0 activitéintense de maraIchage. Troistypes d'eau sont souvent utilisés suivant le pouvoir d'achat du maraIcher. Ceux qui ontle plus demoyens sont branchesau réseau d'eau potable, suivant les normes OMS, que ses eaux sont inaptes a ceux qui en ont moms utilisent soit des eaux de céanes, soit l'irrigation de maraIchage sans restriction. des eaux usées, soit lemélange des deux,suivant la proximité ou non d'un réseau d'dgout. Nous avons analyse les (2) Utilisation des eaux usées dans maraIchage caractéristiques physico-chimiques etbactdriologiques de ces (a) Impactsur les besoins en eau. D'unemanière genérale eaux. Les résultats sont notes dans le tableau 1. (figure 1), plus de 80 percent des exploitants,au niveaudes Une eau de bonne qualité destinée au maraIchage doit trois sites disent ne pas disposer de l'eau nécessaire avoir unpH variantentre 5,5 et 7,5 selon les legumes cultivés. pour l'irrigation des surfaces disponibles.Malgré tout, les Elle nedolt pas être trés salée:le résidu sec etla concentration taux de satisfactionles plus importants sont enrégistréschez enchlorures pour une eaudouce sont dvaluées respectivement ceux qui utilisent les eaux usées brutes. a moms de 900 mg/l pour le premier et moms de 600 mg/l Seuls les exploitants qui utilisent l'eau de Ia SONEES pourle second.La conductivitéoptimale est situde autour de (Société Nationale d'Exploitation des Eaux du Sénégal) 2,500 Is. arriventa satisfaireleur besoin en eau: cela se comprendpar Pour ce concernela des eaux qui qualité microbiologique le fait que l'acquisition d'un robinet dans Ia parcelle assure destinées au selon les normes ces eaux maraIchage, OMS, de I'eau en permanence.L'inconvénient lie a cette situation ne doivent contenir une concentration en coliformes pas est que cette eau doitêtre payee assez chère. fécaux supérieure a 1.000/100ml et une concentration en cuufsde nématodesintestiriaux 1/1.000 supérieure a ml, pour (b) sur fe choix du de culture. La culture des une sans restriction. Impact type irrigation legumes,celle des arbres fruitierset l'associationde cesdeux Surle du de Iaconductivité et de laconcentration plan pH, deculture sont les nous avons observées. en les trois d'eau considérés sont a types pratiques que chlorures, types aptes Suivantle mode eneau l'irrigation en maraIchage. d'approvisionnement d'irrigation, les exploitantsprivildgient l'une ou l'autre pratique (figure Les eaux usées brutes sont fortement chargéesen MES. La la directe estle de Iacouche 2). consequence plus colmatage Ceux en eaux de 'céane' cultivent superficielle du sol pouvant aboutir a long terme a une qui s'approvisionnent surtout des Ii en est de même les imperméabilisation puis une asphyxie. legumes. pour exploitants utilisentles eaux usées brutes etles eaux de 'céane'.Cette La DBO5 indique la teneur de l'eau en matière organique qui rentable a court attire les biodegradable. Comme la degradationde celle-ci produit a culture,plus terme, beaucoupplus d'un minimum d'eau de terme des fertilisantstels que l'ammoniaque, les nitrates et exploitants qui peuvent disposer les phosphates, les eaux usées prdsensente sur ce point un 'céane'. reel avantagepar rapport aux autres typesd'eau d'irrigation. Les maraIchers, qui s'approvisionnentexclusivement en L'ion ammonium(NH4) est la fonnela plus réduite de eaux uséesbrutes, optentpour Ia culture associéede legumes l'azote mindrale. Sous cette forme,il est assimilablepar les et arbres fruitiers. Conscientsde l'interdiction d'utiliser les bactéries et les vdgetaux supdrieurs. La forte teneur notée eaux usées brutes pour arroser des legumeset contraints par dans les eaux usées peut être considereeegalement comme leur situation economique de le faire, par prudence, ils un avantagepour l'irrigation en maraIchage. associentles deux cultures pour de se retrouver sans Les phosphates (P043-) est assimilépar les bactdries et ressource lors d'une dventuelle intervention des Services les plantes supdrieures. Sa presencedans les eaux usées est d' Hygiene. aussi un avantageen maraIchage. Notonsque lesexploitants qui necultivent que des arbres Les concentrations notées en coliformes fécaux et en fruitiers sont peu nombreux (moms de 10 percent). streptocoquesfécaux au niveau des eaux usées montrent,

253 SeydouNiang African Urban Quarterly

(c) Impact sur le mode d'irrigation. L'irrigation par Fig. 1, Degré de satisfaction des besoins en eau par mode aspersion a l'arrosoir, par tuyau souple et par système de d'approvisionnement eneau canalisation sont lesprincipaux modes d'irrigationrencontrés Fig. 1, Degreeof satisfaction of water needs accordingto the way watered dans Ia zone. Le système dominant est l'irrigation par 100 aspersion a l'arrosoir. Cependant, on observe aussi une 90 • Eau suffisante Eau N. combinaison de ces systèmes (figure 3). 80 Les en eau de 'céane' exploitantsqui s'approvisionnent 70 utilisentexciusivement l'aspersion a l'arrosoir. Les 'cdanes' 60 situés en contrebaspar rapport aux parcellescu1tives, le transportde I'eau exige un apportd'energie qui est ici la 50 force musculairedes maraIchers. 40 Les exploitants qui s'approvisionnenten eau usde brute 30 utilisenten majoritdl'aspersion a l'arrosoir, mais 20 percent 20 utilisent les canaux associés a parmi eux, d'irrigation 10 l'arrosoir.Pourtant, les eaux uséespourraient être acheminées 0 au niveau des parcelles par de simples systèmes de SON. Cda. EU /E.U canalisation. En souci de conservationde cette rdalité, par Source: Compiled by theauthor eaule plus longtemps possibledans la parcelle,les exploitants creusentdes trous assez profondsoü us ddvientet stockent Fig. 2, deculture mode en eau leseaux usées. Celales obligent par la suite autiliser I'arrosoir Type par d'approvisionnement to the watered pour l'irrigation des parcelles. Fig. 2, Type offarmingaccording way Les a la fois en eau de exploitantsqui s'approvisionnent • Legumes • Leg. + Fruits E Fruits 'cdane' et en eau usée brute, utilisentl'aspersion al'arrosoir 80 et Ia combinaison arrosoir-tuyau souple dans les mêmes 70 proportions: 60 La majoritédes exploitantsqui s'approvisionnenta l'eau 50 de Ia SONEES utilisent a l'arrosoir. Nous l'aspersion 40 supposons que ce système est choisi uniquementpar souci d'économiede l'eau.En effet, l'eau qui arrive par lerobinet 30 a suffisamment de pression pour être transportée par des 20 canauxd' irrigation (tranchéesrustiques), ou par tuyau souple. 10 D'ailleurs, lacombinaison de l'arrosoir avec le tuyau souple 0 11 et l'utilisation de ce dernier aniquement, occupent des proportionsassez importantes chez ces exploitants. Fig. 3, Mode d'irrigationen fonction du moded'approvisionnement en eau (d) Impact sur Iaquantité d'eauutilisée. La quantitéd'eau Method to the watered utilisée semble lide au mode d'approvisionnementen eau. Fig. 3, ofirrigation according way 70 percent de ceux qui s'approvisionnenten eau de 'céane' 100 disent qu'ils utilisent entre 4 et 20 m3 d'eau parjour, tandis que 86 percent de ceux qui s'approvisionnenten eaux usdes 90 disent ne pas dépasser 4 m3 parjour. 80 Les surfacesdes parcellescultivdes dans lemaraIchage a 70 Dakar sont del'ordre de500 a600 m2 parparcelle (Ngingue, 60 Les céanes etles chemins nécessairesau de 1992). transport 50 l'eau représentent 40 a 50 percent de la surface occupée (Ngingue, 1992).En fin de compte, les surfacesrdellement 40 cultivéespar parcelle se rdsument a 200 ou 300 m2. 30 La quantitdd'eau utiliséepar ceux qui s'approvisionnent 20 en eaux de céanes se situe alors entre 20 et 66 litres par m2 10 en parjour, tandis que les maraIchersqui s'approvisionnent 0 eaux usdes ne dépassent pas 14 litres par m2 par jour. Ii sembleraitdonc que l'utilisation des eaux usées diminue la consommationen eau. Selon Beniest et al. un Légende: (1987), pour SON SONEES bon maraIchage,ii faut 6 a 10 litres par m2 parjour jusqu'à Céa Céanes la rdcolte. Ceci signifie que le système d'arrosage utilisé ici EU. Eaux UsCes conduit a un gaspillage d'eau. CIE.U. CCanes + Eaux Usées EauN. Suffis. Eau Non Suffisante Leg + Fruits Legumes + Fruits

254 Epuration et Réutilisationdes Eaux Usées Domestiques Volume 11 Nos. 2 and 3, May andAugust, 1996 en Maraichage PIriurbaina Dakar, Senegal

l'utilisation des De manière (e) Impact sur intrants. Tableau 2, Concentration en parasitesdes legumes par100 ml d'eau les utilisentma! générale, maraIchers, dépourvusde conseils, derincage ou insuffisament II n'existe de engrais et pesticides. pas Table 2,Parasites concentration in vegetablesper 100 mlof rinsing out rotation culturale systématique etle traitement phyto-sanitaire water toujourscuratif, jamais prdventif,est souventeffectué trop tard (Ngingue, 1992). Ces conditions conduisent a un Legumes rendement au metre cube d'eau utilisée trés inférieur a ce Parasites Laitue Persil Carotte que le climat et le travaildépensé seraienten droit de laisser Amibe minuta 13 kystes 6 kystes 0 espérer. Amibe hystolitica 9 kystes 14 kystes 0 Les engrais chimiques, les fumures organiques(fumier Anguillules 0 20larves 0 de parc, terre ou poudre d'arachide et fumier de poisson)et Ankylostomes 2larves 3 ceufs 0 lesproduits phytosanitaires constituentles principaux intrants Ascaris 40 ceufs 0 0 utilisés. Levures bourgeonnantes beaucoup 17 0 II apparaIt nettement dans la figure 4 que le taux Némathelminthes 0 7 eufs 11 ceufs d'utilisation des intrantsest fortement influence par lemode Plathelminthes 0 4 ceufs 4ceufs d'approvisionnementen eau. Trichomonas 13 ceufs 12 ceufs 3 ceufs Tnchocéphale 4 ceufs 0 0 Fig. 4, Utilisation des intrants parmode d'approvisionnement en eau watered Fig. 4, Input using by way of Quant aux ndmathelminthes et plathelminthes,qui sont des parasitesnécessitant la presence d'un hôte intermediaire, 90 leur presencen'est pas pour autant sans danger. En effet, les animaux constituent leshôtes 80 domestiques qui intermédiaires, dans les maisons et s'abreuvent souventavec les 70 divaguent eaux de rinçage des legumes et mangent leurs 60 Comme ces animaux sont consommés par Ia population, le 50 cycle de contamination est vite 40 30 (4) Aptitudes des eaux traitées pour le maraIchage 20 La StationExpérimentale de Cambérènese propose avant tout de tester sur différents de traitements 10 place procédés extensifs des eaux usées en comparantles rendementsobtenus 0 SON. Céa. E.U. C/EU. par chaque système dans les mêmes conditions experimentales(même climat, mêmes eaux usées, mêmes Légende: debits). L'objectiffinal est cependant,de mettresur pied une PP.San. ProduitsPhyto-Sanitaires d ' TAr. : Terred'Arachide (voire plusieures) chaIne(s) composée(s) F.Pois. : Farine de Poisson d'éléments provenant des différents procédés en experimentation. Les résultats prdsentds concernent la La richessedes eaux usdes en nutritifsconstitut disponibilité en eau après traitement et les rendements un apport assez important. Cela se manifeste chez les dpuratoires. maraIchers utilisant ces eaux par une nette sur les dCpenses en intrants. (a) Disponibilité en eau après traitement. Les résultats concernentIa périodedu 15 février au IOjuin 1994(pendant (3) Qualité sanitaire des legumesarrosés a l'eau usée brute Ia saison seche); soit 116 jours. Ils indiquent les pertes en Les analyses portentsur Ia contamination parasitologique eau de chaque systèmelors de I'épuration. et concerne des legumes qui sont susceptihies d'être Sur le plan strict de Ia disponibilité en eauaprès consommds crüs (laitue, persil, carotte). (figure 5), le système a sol nu offre le plus de possibilité L'analysedes rdsultats prdsentés dans le tableau 2 montre pour le maraIchage. II est suivi par Ic lagunagea microphyte que certains legumes sont contaminés par des parasites puis par Ia prairie flottante a Pistia. Le système a Typha comme les amibes sans temps de latence et a faible dose semble le moms intCressant surIc plan de la disponibilitéde infectante. Cela veut dire que si ces legumes sont consommés l'eau après iraitement. crüs et ma! lavés, us sont capables de provoqueraussitôt Ia maladie chez le consommateur. (b) Rendement des différentes cascades. L'ankylostome, l'ascaris et le trichocéphale sont des L'analyse du tableau 3 montre qu'aucune des cascades nématodes intestinaux qui ont une période de latence expCrimentdesne prdsente un rendementdpuratoire supérieur importanteet n'ont pas besoin d'hôte intermCdiaire pour se aux autres pour tous les paramètresde pollution considérés. transmettrea l'homme. La presence de leurs ufs et larves Dc plus, certaines cascades produisent une eau traitde montre Ic degre de contaminationde ces legumes. ddsdquilibrée du point de vue du rapport phosphate/azote,

255 Seydou Niang AfricanUrban Quarterly

Fig. 5, Evaporationet moyennes des sixcascades - une production d'eau en quantitdsuffisante. and sir tanks Fig.5, Averageevaporation evapotranspirationof groups of Si Iapremiere condition est réalisdepour l'essentiel des cascades, on constate ndanmoins un ddcalage entre les Volume en I/m2.jour retentionsd'azote et de phosphore(lagunage a microphytes, prairie a Pistia et dpandagesous sol nu). De plus, lorsque l'dquilibre est atteint,le volumed'eau recueilli ala sortie du système est tellement faible au point de compromettre une rdutilisation (dpandagesous Eucalyptus,marais a Typha). Ainsi, chacun deces systèmesd'dpuration comporte des points fortset des points faiblesmanifestes dans letraitement des eaux usdes pour une rdutilisation en maraIchage. Ce qui IIIIIIIIIIIIIIIIIl montre que seule une combinaison d'écosystèmes Typha Phragmites Lagunage Pistia Eucalyptus Sol flu différents permettrade modulerl'eau a la sortie en fonction de final au traitement. Source:Niang, Diop, Mbéguére,Radoux, 1994 l'objectif assignd Discussions Tableau 3, Rendement des cascades en % La disponibilité de l'eau constituele plus grand handicap Table 3, Purification yieldof different organising tanksin % du maraIchage: 81 percent des exploitantsestiment ne pas d'assez d'eau leurs cultures. Le branchement MES DCO NKe P.Total Coli.fécaux disposer pour au réseau d'eau potable de Ia SONEES est coUteux et ne Cascade I 90 80 84 71 9,6.10 saurait donc constituer une solution d'avenir quand la Cascade 2 31 -6 65 26 3.i0 demandeen eaud'alimentation humaine n'est pas satisfaite. Cascade 3 66 17 68 9 3,8.10 Il sembleque pour un bon maraIchage, il faut environ 8 Cascade 4 94 71 27 83 13.10 litres d'eau par jour et par m2 pendant toute la pdriode Cascade5 84 81 68 50 5.10 culturale (Delvaque, 1980), or, selon nos enquêtes, les Cascade6 62 86 55 15 10,3.10 maraIchers utilisent au moms 14 litrespar m2 parjour. Ii y a Eau brute 464 584 460 35,3 3,2.1011 donc l une question d'utilisation efficiente de l'eau disponible. Légende: Malgré tout, une des solutionspour le développementdu MES Matièresen suspension en mg/I urbain a être la réutilisationdes Demande en en 0/1 maraIchage Dakarpourrait DBO5 Biologique oxygène mg eaux d'autant le volume de ces eaux DCO DemandeChimique en Oxygene usées, Ce, plus que NH4 ammoniumen mg d'azote par litre augmentera proportionnellement a Ia croissance de la P043- Phosphatesen mg de phosphore par litre populationde Ia yule. P.Total PhosphoreTotal Deplus, l'utilisationdes eaux usdes semblejouerun role NKe. Azote Kjeldahl (total) sur la d'eau utilisde : 86 des Coli.fécaux Coliformesfécaux en nombre par 100 ml important quantite percent Strep.fécaux Streptocoques en nombrepar 100 ml exploitantss'approvisionnant en eaux usdes disent utiliser moms de 14 litres par m2 parjour, alors que 70 percent de ceux eau du fait Ia rétension de l'un dans le est 7 qui s'approvisionnent en de 'céane' disent en que système parfois consommerentre 20 66 litres m2 fois a celle de l'autre Ceci ne et par parjour. supérieure (cascades2, 3,4, 6). des eaux usées réduit le nombre d'intrants être conseillé une eau destinée au a L'utilisation peut pour maraIchage, 76 des disent utiliser moms de moms de a des corrections. employés: percent exploitants procéder trois 87 de ceux qui même si Ofl types d'intrants, pendant que percent Sur le plan de la ddsinfection, observe un en eau de disent abattementdes Coliformesfécaux de l'ordre de 7 s'approvisionnent 'cdane', qu'ils emploient puissances au moms trois d'intrants. de aucune cascade ne aux normes OMS de types 10, rdpond la réutilisationdes eaux usées brutes réutilisation en sans restriction Coli Cependant, présente maraIchage (1.000 par l'heure actuelle Dakar un de sante 100 a a grave problème ml). Les analyses parasitologiques réalisées sur des Ainsi, on note des differences de rendement entre les publique. legumes susceptiblesd'être consommés crâs le conjirment. diffdrents typesde systèmes d'épurationextensive testes mais Ce risque est accentud par le mode d'irrigation, car au niveau des volumes d'eaux recueillis qui considérablement les être très faibles certains l'arrosage par aspersion augmente peuvent pour systèmes (dpandage de contamination. Or, 73 percent des exploitants sous marais a , marais a Phragmites). risques Eucalyptus, Typha employant des eaux usées irriguent leurs cultures Dans une derdutilisation des eaux traitdesen par perspective aspersion a I'arrosoir. maraIchage, l'iddal serait d'avoir: - et Nouspensons que l'avenirdumaraIchage se trouvedans une bonne dpurationprimaire secondaire, la réutilisation des eaux résiduaires un de - aux normes moyennant effort une ddsinfectionconforme OMS, la des notammentdans ledomaine du - un abattementmodérd en azote eten avec part pouvoirspublics, phosphore traitement des eaux usées par des techniques extensives maintien d'un azote/phosphore efficace, techniquement etfinancièrementaccessibles.

256 Epuration et Réutilisationdes Eaux Usées Domestiques Volume 11 Nos. 2 and 3, May andAugust, 1996 en MaraIchage Périurbain a Dakar, Senegal

Les rdsultats enrégistrés a Ia Station Expdrimentale de Delvaque, J. (1980). Etudepour une Planificationdes cultures Cambdrènesemblent montrer que les six procédésextensives maraIchèresau Senegal. Tome3. Production, Couverturedes d'dpuration des eaux usdes testes a Dakar ne produisentpas besoins, Previsions regionales, Organisation et Mesures une eau traitée satisfaisante, pour une réutilisation en Strategiques. Conclusions et Recommandations. Dakar: maralehage,même si par ailleursles rendementsdpuratoires Centrepour leDdveloppement de l'Horticulture du Sdndgal. respectentles normes de rejets. Les recherches doivent donc s'orienter vers une Fall, C. (1992). Communication orale, Semaine de L'IFAN optimisation des rendements par des combinaisons du4au9mai 1992. judicieuses d'dcosystèmes artificiels au sein d'une méme chaIne Kane, A. (1973). L'exploitationmaraIchère et fruitièredans Iaregion de Sangalkam, mémoirede maItrise de Gdographie, Notes Ddpartement de Geographie, Faculté des Lettres et Sciences Les résultats présentésici proviennentd'une réalisée par Humaines,Universitd de Dakar. Fall C. en 1995 sur l'impact sanitaire de l'utilisation des eaux usées brutes en maraichage. Mara, D. et Cairncross, S. (1991). Guide pour l'utilisation sans risquesdes eaux résiduaireset des excreta enagriculture 2 Cette a mise au en 1978 M.Radoux méthodologie point par et aquaculture.Genève: OMS/PNUE. a la station expérimentalede Viville en Belgique. au Références Ngningue,M. (1992). Les cultures maraIchères Séndgal: en vue d'une raisonnéedes itindraires Beniest, J. et al. (1987). Guide du au Diagnostic approche pratique maraIchage Dossier documentaire, mdmoire de fin d'dtude, Sdnégal, ISRA/Centre pour le Ddveloppement de techniques, Dakar, EcoleNormaleSuperieured'EnseignementTechnique I'Horticulture,Dakar, Coll. Cah. d'Inf. n° 1. et Professionnel, UCAD. Chastel,J.M. (1982).Etude des Systèmes deProduction dans Radoux,M., S. Niang, B. S. Dirdand M. B. (1994). Iazone deculture maraIchère des Niayesde Lompoul. Grande Mbégudnd Epuration des eaux usées urbaines par voie naturelle et côte du Sdndgal. Mémoirede fin d'dtudes Dijon, ENSSAA. rdcupdration des eaux traitdes pour Ia productionde bois de Chaire des Sciences Economiques. feu et l'irrigation de maraIchage: Suivi scientifique de Ia Station Expdrimentale deCambdrène. Rapportannuel, Institut des Sciences de l'EnvironnementUCAD.

257 REVUE TIERS MONDE n° 152 Tome XXXVIII Octobre - Décembre 1997 Revuetrimestrielle publiée par l'Institut d'Etude du DéveloppementEconomique et Social Université de Paris I Panthéon-Sorbonne

n° 152

Regis Chavigny Economies en transitionet economiesen développement une comparaison Pierre Huyette Mobilite et informalité des nouvelles formes aux rèulations modernes de 1 emploi en Colombie Jean-Pierre Lachaud Exclusion du marché du travail, tnalités et genre dans les capitales africaines une methode nouveflede mesure Laurent Parrot Les modèles d equilibre generalcakulable un nouveau cadre d'analyse pour comprendreles consequences des politiques macroéconomiquessur La yule ou le village Taladidia Thiombiano La controverse empirique et thorique posee par le comportement des producteurs-consontmateurs StéphaneTizio Yves-AntoineFlory Linitiative de BamaKo sante pour tous ou maiadie pour cnacun7 Patrick Plane La privatisation de 1 en Côte d Ivoire evaluation et interpretation des premiers résultats Karine Bennafla Entre Afriquenoire et monde arabe nouvelles tendancesdes iriformelstchadiens

Documentation

Michel Lelart Un exempled integration institutionnelle 1 evolutionde la zone franc Debat

FredricDurand L exportationdu bois brut peut-elle aider a preserver les foréts tropicales " Le cas de I Indonésieet des Philippines

Bibhographi - Analysesbibliographzques

Tabledes matieres du tome XXXVIII, 1997

Si vous désirez en rendre cornpte, un exemplaire vous sera adressé sur demande

Rdaction etadministration : lnstitut d'Etude du Dveloppenient Economique et Social 58, BdArago750l3Paris-tel:01 44087307-fax 0147078175 Abonnements et ventes : Presses Universitaires de France - Departement dts Revues 14, avenuedu Bois deI'Epine B.P. 90, 91003Evry Cedex - tel : 01 60778205 - fax 0160792045 Prixdu nuinéro: 145 F - Abonne-"ents pour l'anriée 1997: France 480 F- Etranger530 F Publié avec le concoursdu Centre national des Lefties AfricanUrban Quarterly, 1996: 11 (2 and 3) pp. 259—262 ©1996by African Urban QuarterlyLtd.

FARMING IN THE CITY: FROM ANALYSIS TO ACTION

Julie van der Bliek and Ann Waters-Bayer ETC Foundation,P.O. Box 64, NL-3830AB Leusden,The Netherlands Address for correspondence: Rohnsweg 56, D-37085 Gottingen, Germany; Tel: +49-551-485751; Fax: +49-551-47948; Email: [email protected]

Abstract Accepted September, 1995 it is time for research into urban agriculture to shift from description to action orientation. This article outlines an approach which involves urban dwellers, local organisations, municipal authorities and interdisciplinaryresearchers in analysing the current contextof urbanfarming, identifying problems andpotentials, andplanning joint action.

Résumé Les auteurs notentqu 'unplus grande effort de recherchesur 1'agriculture urbaine doit maintenant être consacré a informer des interventions pratiques dans des situations spécfiques. S'inspirant de I'experience en evaluation rapide participative, dans le cadre de strategies communautaires dedéveloppement en milieurural, lesauteurs soumettent une approche pour la recherche-actionen agriculture urbaine. Cette approcheoffre un cadre de reference, adaptable pour chaque yule donnée, permettant, avec le concours des citadins, organisationslocales, autorities municipales et cherceursde diversesdisciplines, d'analyser enpremier lieu le contexte localpoury reconnoItrefonctions et ressources disponibles, d'identiflerensuiteproblCmes et opportunites d'interventionpar diagnosticparticipatif tout ceci menant a l'adoption d'interventionsconjointes par les acteurs concernés, tantau niveau des unitesde voisinagequ 'a Ia ville-même. Background husbandryin the urban centre; In recent years, numerous examples of urban farming - extensionservices for urban farmers, and the capacity throughoutthe world have been documented, analysed and and specificallyurban-related technical knowledgeof discussed. However, this research has not yet led to these services; widespreadaction topromote and supporturban farming and - research institutions with interest and capacity to to reduce potential dangers. Studies have often been limited support urbanfarming; to describing what is happeningwhere and who the farmers - Community-BasedOrganisations (CBOS) and NGOs are. To make a step forward,a more action-orientedapproach with interestand capacityto becomeinvolved in ur is needed. This approach starts with a preliminaryanalysis farming; of the context of agriculturalactivities in a particularurban - urbancouncil and other governmentbodies relevant for centre and leads into an appraisal carried out together with promoting urbanfarming, and theircapacity to become current and potential farmers among the urban population. involved;and The process may be initiated by universities or local - incentives or disincentives for urbanfarming: subsidies, governmental or Non-GovernmentalOrganisations (NGOs) provisionof allotments,rules and regulations,taxes which are interested in promoting urban agriculture or by (c) socioeconomic situation: external donor agencies that have recognised the farming - knowledge: basis of indigenous knowledge (also of initiatives of the urban poor and seek to support them. immigrants) for existing urban farming, potential sources of knowledge(ethnic background and place of Analysing the Context origin of urbanresident farmers,length of residence in The initial contextual analysis is meant to provide an and familiaritywith the urbancentre); overview of the situation. This can be conducted by - sources of livelihood,and relative importanceof urban researchersand/or local governmentplanners or the staff of cultivation, tree growing and animal husbandry; NGOs, depending on who has taken the initiativeand who - access to resourcessuch as land, water,capital, labour, has thecapacity to carry outsuch an analysis. Attentionneeds productioninputs; to be paid to the following aspects: - rural-urbanlinkages: sources of income, inputs (forage, (a) physical constraintsand potential: manure, seed etc.), knowledgeand labour, flowingin - land: availability,slope, drainage, alternativeuses; either direction; - water:rainfall, evapotranspiration, sources, availability - nutritional status of families,and contributionof urban and accessibility, quality; and farmingto family nutrition; - soil: quality (physical and chemical composition), - pattern of food availability from urban farming: availability of external nutrients (e.g. ingredients for seasonality and variationsfrom year to year; compost),pollution - economic position: identification of groups under (b) institutionaland legal framework: economic stress, dynamicsof this stress in recent - legal statusof crop cultivation, tree growingand animal years; and

259 vet' aer Bliek Julie and Anne Waters-Bayer AfricanUroan Quarterly

(Box 1:Parl/cipa/ory Planning by P00, Women/nDaka,, Senegal This example from Dakar, Senegal, which concernsthe planningof micro-enterprises in food processing, illustratesan approach that could also be taken in planningincome generationfrom urban farming. One of the options to improve income opportunities for poor women in Dakar is selling traditional,non-alcoholic drinks. GRAF, an NGO in Senegal which was already working with women on income-generatingactivities, initiated a participatorystudy to analyse the situation, problems and opportunities. This includedthree steps: Actoranalysis Different actorsin the productioncycle (producers, vendors and consumers) were identifiedand interviewed. A checklist of guidingquestions helped to create a pictureof each group: * Who are they? * Whatare their resources? * Whatis their workingcalender? * Dn work as * they individuals or are there collectiveactivities? What are the advantages I reasons for choosing this activity? * How and where do or sell? * they produce Whatore the consumers preferences? /?e,tct!on The reuit of the nterviews were presentedto the interviewees in meetings with each ofthe different groupsof actors. Dunj these meetings, the groupsreflected on their present activities and strategies. Lookr?g ahecd During further dc issions, the grcups focusedon the constraints they had identifiedand the resourcesavailable tothem. Possible solujons, sugnested by group members aswell as by GRAF staff, were considered. Source vcn 'eku'd v.n Veldhu/zen, /993 /9

- organisationalcapacity and initiatives taken by local with to urban Box 2: Using Bioresource F/ow Diagrams /0 Mon//or groups respect farming. Change This contextual analysis at the 'urban centre level' a overview of the and Bioresourceflow diagramsdepict flows of nutrients provides general physical and other into a unit and between socioeconomic situation with urban and inputs production respect to farming, differentcomponents of a unit. They can also be used of ihe governmental and non-governmental institutions which to depict flows of nutrients (e.g. food and fertiliser are involved or concerned. purchases,consumption and salesof nutrients in foods and by-products, wasteremoval or recycling)within a city neighbourhood.This is a useful tool for stimulating ParticipatoryUrban Appraisal a processof planning, experimentation, monitoringand An appraisal is then carried out by urban dwellers, improvement in resource use. from ICLARM researchers from variousdisciplires, NGOs and possibly local Assistedby researchers (International Centre for Living Aquatic Resources Management), governmentofficials sympathetic to urbanfarming. As there farmers in Malawi made ajoint inventoryand analysis can he considerabledifferences (socioeconomic, ethnic and of local resources by drawing resource flows into, out and within their farms. The farmers then discussed topographical among others) between different sections of of an urban this should be done at the and sought (e.g.through visits to research stations and centre, appraisal other farmers) alternativeways of using the available neighbourhoodor quarter level. Not only current but also resources,and commenced their own experiments in potential farmers should be involved, particularly poorer integratedfish farming. keen farm cannot By comparing a series of their diagrams of people who are to in the urban centre but bioresource flows overtime,the farmers could monitor because of lack of access to the required resources. The the transformation oftheir farming systems. The amount participatory appraisal focuses on the advantages and of material, the frequency of flow and its monetary of urban current value were recorded. Quantities were expressed in disadvantages farming, problems, local terms. This procedureenabled both the farmers opportunities and observed trends which may point to and the researchers to monitorprogress and change. potentials, other local actors in farming itself as well as in and also provided farmers witha toolfor improving their in resource 'upstream' (input supply) and 'downstream' (processing, decision making and skills management. handling and marketing)activities, existingknowledge, gaps Source:L,'htfoot and Noble, 1993 1134 in skills and techniques, motivations and cultural perspectives. ParticipatoryRural Appraisal (PRA) tools are becoming 1994). Particularly such PRA tools as Venn (institutional) increasingly popularin analysing rural situations and planning diagrams,social mapping,livelihood analysis, resource flow action by numerous NGOs and more recently also by diagrams (see Box 2), servicesand opportunitymaps, various governmentinstitutions (RRA Notes, recentlyrenamed PLA calendars,ranking and focus groupdiscussions can stimulate Notes, issued by the InternationalInstitute for Environment analysis with wide participation. In urban settings, games and Development,3 Endsleigh Street,London WCIH ODD, and jointly-constructed models have also proved to be UK). There are now several examples of the application of effective in stimulatingresidents to consider together what these tools also in urban settings (Mitlin and Thompson, can and should be done (Gibson, 1994: 41-47).

260 Volume 11 Nos. 2 and 3, May and August, 1996 Farmingin the City: From Analysis to Action

Recognising Functionsand Resources situations, there will be a shortage of one or more of these Urban farming performs several functions in urban resources. These relative shortages will likewise differ dwellers' livelihoods and in their environment. The betweenurban quarters. participatoryurban appraisalshould help in recognisingand It is, therefore, importantto identify clearly thelimitations prioritisingthese functions accordingto thecriteria of urban within which each neighbourhoodoperates and how these dwellersas well as municipalauthorities. This will providea can be overcome. The (potential) farmersmust be involved basis for designing site-specific activities to improveurban in making arealistic assessment of thepossibilities. Creative farming. These functions may include the following: solutions will be neededto deal with these shortages,and it nutritional: providing more balanced nutrition and thus is often the enterprising people already practising urban economic: and income improved health; production farmingwho can show the way. generation through selling products or increasingavailable income on food environmental: throughsaving expenditures; Stakeholder Workshops improvedcity ecology, through betterwater drainage, creation The resultsof the contextual at the urban centre and analysis of favourablemicroclimate, waste recycling reduction level and the in the various in and aesthetic/leisure: in the participatory appraisals dust, among others, especially are and discussed at one richer the of more areas neighbourhoods brought together cities, aspects gardens, parks, green actionsto be taken at and allotments for residents without or more stakeholderworkshops. Here, garden housegardens level are identified be raised. theurban centre and neighbourhood jointly may in the realmsof: The relative of these and possibly also other importance a environmentwhich functions will depend on a number of variables such as the (a) policydevelopment: creating policy economic and social situation of the urban family or enables urbanfarming to flourish and ensures that poorer urbandwellers have fair accessto the resources; community,the availabilityand location of land (e.g. in the required researchand in both 'hard' and backyardor furtherfrom the house), thepreferred production (b) development (production) close collaboration sectors (horticulture, crop husbandry, animal husbandry, 'soft' (organisational)technologies: in and aquaculture)as well as the prioritiesof the local authorities. with the farmers themselves identifying, testing The relative importance of these functions will also differ adapting notonly production technology optionsfor urban between quarters withinone and the same urban centre. farming but also organisational options such as the Besides the knowledge and labour energy of the urban allocation of temporaryresource-use rights for specific dwellers,the main urban resources used for agriculture are purposes, e.g. unbuiltland for cultivation, or roadsidetrees water, nutrients (primarily from waste) and land. In many for harvest (also for animal feed); and

Framework for (Box3. Analysis CONTEXTUALANALYSIS: analysisat 'city level', focus on generaloverview of physicaland social situation in the city related to urban farming, governmentaland non-governmental institutions involved GENERAL ANALYSISOF: - physical constraints & potential - institutional & legal framework - socioeconomicsituation - functionsand resources PARTICIPATORY APPRAISAL: diagnosis with farmers and potential farmers in (selected) neighbourhoods, (dis)advantages ofurban farming, problems, opportunities, changestaking place, who does whatand why, traditional knowledge, gaps in knowledgeand techniques, cultural perspectives

ACTION AT 'CIlY LEVEL': creating an enablingenvironment through improvedplanning capacity, removal of constraints, facilitatingpolicies STAKEHOLDERWORKSHOP(S) TO DESIGN ACTIONFOR: - policy development - institutional development - research & technology development - immediateactivities ACTION AT NEIGHBOURHOOD LEVEL: - participatoryappraisal ofproblems and opportunities - collaborationin developingtechnological solutions - organisational development 9urce.' Compiledby the authors

261 Julie i'ander Bhiekand Anne Waters-Bayer African Urban Quarterly

(c) institutional development: supportingthe development or involvingrepresentatives of stakeholdersat all levels, would strengtheningof neighbourhood-level institutions which also provide a platform for re-negotiation of rights and can interact with higher levels of organisation in responsibilities. This will benecessary as the face of theurban promotingand implementingurban farming. centre and thus, of urban farmingchanges over time. All actors concerned must be represented when a programmefor action in urban agriculturaldevelopment is References designed. The needs and responsibilities of the city Alders, C. etal. (1994)."Farming at Close Quarters".Special government must be confronted with the needs and issue of ILEIA Newsletter, Volume 10 No. 4. responsibilitiesof the urban dwellers,and agreements have to be negotiated. Gibson, T. (1994). "Showing What You Mean (Not Just Stakeholder a for workshops provide good opportunity Talking About It)", RRA Notes Vol. 21. negotiations on an equal footing and for the formation of in the of activities: coalitions, particularly following types Leitmann, J. (1994). RapidUrban EnvironmentalAssessment: aimed at urban and (a) lobbying improved planning Lessons Cities in the Developing World. of which from Washington, imaginative development governmentpolicies DC: WorldBank Urban ManagementProgramme. favour the opportunities, remove the constraints and minimise the which are faced and created problems by Lightfoot, C. and R. Noble (1993). "A Participatory urban farmers; and Experimentin Sustainable Agriculture", Journal of Farming (b) participatory action research, involving farmers, scientists, Systems Research-Extension Vol. 4 No. 1. extensionists,artisans and other input suppliers, in order to develop appropriate technologies for the specific conditions of in that or Mitlin, D. and J. Thompson (1994). "Special Issue on farming city neighbourhood Tools and Methods".London: lIED Sustainable et al., 1992; van der Bliek and van Veldhuizen Participatory (Reijntjes & Human Settlements 1993 and van Veldhuizenand de Zeeuw, 1992). AgricultureProgramme Programme, PRA Notes No. 21. Participatory Monitoring It is that be to immediate Reijntjes, C. et al. (1992). Farming for the Future: An important support given introduction to and Sustainable implementation of theactivities identifiedduring the process Low-External-Input London: Macmillan. of contextual analysis and participatory appraisal and Agriculture. planning.Initially, it may be possible to support action only in selected neighbourhoods. This needs to be accompanied van der Bliek, J. and L. van Veldhuizen (1993).Developing by participatorymonitoring and evaluationof the activities. Tools Together: Reportof a Study on theRole ofParticipation Only in this way can lessons be learned for improving and in the Development ofTools, Equipmentand Techniques in expanding the programmeof urbanagricultural development. Appropriate Technology Programmes. Leusden: ETC It will notbe sufficient to establisha programme for urban Foundation. agriculturaldevelopment based on onlyone roundof analysis and planning. Someinstitution mustbe created which permits van Veldhuizen, L. and H. de Zeeuw (1992). Learningfor continuation of a process: monitoringchanges and impacts, Participatory Technology Development:A Training Guide. and redesigningthe programmeaccordingly. This institution, Leusden: ETC. which could take the form of a 'city farming committee'

262 The PAKISTAN DEVELOPMENT REVIEW V0309729 An International Journalof Development Economics

Editor SARFRAZ KHAN QURESHI LiteraryEditor The ofSocio-economic Factors on Fertility AURANGZEB A. HASHMI Impact Behaviour: A Cross-country Analysis Book Editor Review Rehana 107 MIR ANNICE MAHMOOD Siddiqui Co-editors M. GHAFFAR CHAUDHRY (;overnment Interventions, Market Integration, MOHAMMAD IRFAN and Price Risk in Pakistan's Punjab SHAHNAZ KAZI Takashi Kurosaki /29 ASHFAQUE H. KHAN SHAMIM A. SAHIBZADA I Husband-wife Roles asa Correlate of Editorial Board Contraceptive RASHID AMJAD and Fertility Behaviour PERPINSTRUP ANDERSEN MuhammadIqhal Zafar /45 ROBERT E. BALDWIN PRANABBARDHAN FREDRIX BARTH Determinants of Rurall'ovcrtv - in Pakistan: JERER. BEHRMAN A Micro Study J. C. CALDWELL Shahnawai Malik /7/ ANSLEV .J. COALE S. I.COHEN P. A. CORNELISSE BOOK REVIEWS ANGUSS. DEATON E. V. K. FITZGERALD Overseas DevelopmenlAdminislration, A Guideto Social MAURICE GODELIER for in Countries NADEEM ULHAGUE Analysis l'rojects Developing til,rAnisire Itln/in,ood I /9 M. AYNULHASAN S. HIRASHIMA NURUL ISLAM Bina Agarwal A Field of One's Own: Gender and A. R. KHAN Land in South Asia M. ALl KHAN Rights MAHMOOD H. KHAN Yacineen Mo/nudthn /9/ MOHSIN S. KHAN KAROLJ. KROTKI Women DAVIDP. LAIDLER Loes Schenk-Sandbergen led). and Seasonal ROBERTO S. MARIANO Labour Migration JOHNW. MELLOR Rehana Siddiqui /97 GUSTAV F.PAPANEK SAMUEL H. PRESTON GERRY RODGERS SHORTER NOTICES 200 VERNON W. RUTTAN ANWAR SHAH ISMAIL SIRAGELDIN T. N. SRINIVASAN WILLIAM C. THIESENHUSEN PAN A. VOTOPOULOS

Volume 35 Summer 1996 Number 2

Annual SubscriptionU.S. $ 100.00. Pak Rs 300.00 by surface mail and U.S. $ 125.00, Pak Rs 400.00 by air mail. Perissue U.S. $ 30.00, Pak Rs 75.00 by surface mail and U.S. $ 35.00, Pak Rs 100.00 by air mail for 1997 only. Subscriptions are entered on a calendar year basis. Payment should be made directly through a demand draft, issuedin the name of the Pakistan Institute of Development Economics, Post Box 091, lslamabad-44000(Pakistan) African Urban Quarterly, 1996: 11 (2 and 3) pp. 264—275 ©1996 by African Urban Quarterly Ltd.

FARMING IN THE CITY OF KAMPALA: ISSUES FOR URBAN MANAGEMENT

Gertrude Atukundaand Daniel Maxwell, respectively Makerere Institute of Social Research,Makerere University, P.O. Box 16022, Kampala,Uganda; and LandTenure Center, Universityof Wisconsin, 1357 UniversityAvenue, Madison, WI 53715 USA

Abstract Accepted September, 1995 In Kampala, Ugandaas in many otherAfrican urban centres, lowerreal wagesandfewerformal employment opportunities have led to increased reliance on subsistenceincome, much of itfrom urbanagriculture. Despite being a pervasive activity, urban agriculture is technically illegal in Kampala, and isfurther constrained by a normative bias on the part of urban authorities.This paper provides adetailed descriptionof urban farming inKampala and presents comparative data on urban householdsthat do anddo notengage in urbanfarming. The paper then discussesconstraints on urbanfarming, highlighting particularly the issue ofmunicipal bylaws and policies. Methods ofpromoting a dialogue among urbanfarmers, researchers and urbanpolicy makers are presented andpotential solutions to policyproblems are discussed. Résumé Dans Ia yule de Kampalacomme aillers enAfrique, leflechissementdes niveauxsalariaux etIa contractiondes débouchés sur le marchefonnel de I'emploi, au cours des récentesdécennies, ontpoussé bon nombrede gens a dépendrede revenusde subsistence.Une bonne partde ces revenusest assurée par / 'agricultureurbaine. Bien qu 'ellesoit une activitéfort répandue, l'agriculture urbaine reste le plus souvent techniquement illégale Ia yule de Kampala, contrainte de surcroIt par une normatisationdefavorable. Cet article presente une description détaillée de I 'activité agricole a Kampala et dresse une analyse comparée de données recueilliesauprés de ménages urbains qui pratiquent l'agriculture urbaine et d'autres qui ne Iapratiquent point. Conclucet des lieux, I 'article discuteles obstacles auxquelsfaitface I 'agricultureurbaine a Kampala, particulièrementen ce qui a trait aux politiques et réglements municipaux. Les auteurs proposent desfacons depromouvoir un meillier dialogue entre producteurs, chercheurs et décideurs urbains, ainsi que de possibles avenues de solution aux obstacles politiques courants etprobables.

Prologue "I am ready to cut down all the beans, maize and cassava On a dry-seasonafternoon in Kampalain June, 1993, an within the perimetersof [Kampala] even if it means life or elderly gentleman was weeding a small plot of maize and death." (Filliam, 1993:4) beansthat he was growingin avacant lot downtownwhere a This briefincident captures severalcritical aspects of the buildinghad been bombedduring the 1979 war against the story of farmingin the urbancentre in contemporaryAfrica. regimeof IdiAmin. In a few weeks,the crops wouldmature, In this case, the farmer was an elderly man; more typicallyit and heplanned toharvest themto use for homeconsumption. would have been a woman with children. But in either case, In the mid-afternoon, a groupof uniformedlaw enforcement the individual involved was poor and relied on informal officers from Kampala City Council (KCC) appeared and accessto otherwiseunused land to grow some food for home without warning or notice, slashed down his small garden. consumption. While clearly considereda sociallylegitimate This act so enraged passers-by that they immediately began activity by the generalpublic, farming is technicallyillegal shoutingat the KCC workers and attackedthem with stones. inmany African urbancentres. Kampala's urbanfarmers are A passingjournalist wondered what the fighting was about not subjected to this kind of official harassment every day, and inquired, first from passers-byand later from the KCC but the economic activities and land-use practices in which official who had sent the law enforcementofficers into the they engage fall into a category in which social legitimacy field. To the passers-by,the issue was simple: the old man and legal statusare atleast somewhat at odds with eachother. ws minding his own business, growing food for his own And of course,the official quoted was beingoverly dramatic: consumptionusing land that was otherwise going to waste To him, urban agriculture is not a matter of life and death. and growing up in elephant grass, which KCC was not However, to the thousands of Kampala residents who farm botheringto keep slashed anyway. He was being needlessly in the city, it is certainlya matter of quality of life, if not life harassed by KCC law enforcement officers who, therefore, and death. And to the city as a whole, which is emerging deservedverbal abuseand stoning. The KCC official,on the from a two-decadeera of economic crisis, coming to grips other hand, stated that he was merely carrying out his with informaleconomic practicesfalling in a gray areawhere prescribedduties under the UrbanAuthorities Act, and added, legitimacyand legality are not synonymous,urban farming

Acknowledgements—We would like to express our thanks to MakerereInstitute of Social Research for institutionalsupport during field work,and to theUS Departmentof Education (Fuibright-Hays Program)and theUS NationalScience Foundation (Sociology Progam) for funding support for this research. We would also like to thank the International DevelopmentResearch Centre and Luc Mougeot in particular, for fundingsupport for the seminar/workshop at the conclusionof the study,which is describedin the paper.

264 Volume 11 Nos. 2 and3, May and August, 1996 Farming in the Cityof Kampala: Issuesfor Urban Management is symptomaticof other contentious issuesin contemporary mainstreamresearch on urbanmanagement inAfrica. Recent urbanmanagement in Africa. reviewsof African urbanisation (Coquery-Vidrovitch, 1991: While farminghas long been an urbanactivity inAfrican 1—98), urban employment (International Labour Organisation, urban centres (Southall and Gutkind, 1957), 'urban 1989) and urban planning (Mabogunje, 1990:121—203) do agriculture' remains at best an oxymoronic conceptto urban not mention the topic at all. A major review of the urban planners and administrators and, sometimes, as this story food security problem in Africa mentions urban agriculture depicts, an embarrassmentto be stamped out—a "remnant tangentially (Guyer, 1987). of bush life" in the words of novelist Naipaul (1981). This The definition of 'urban agriculture' is somewhat paperwill present anoverview of urbanfarming in Kampala, amorphous. Most empirical literature on Africa defines it Ugandaand relate empirical evidencefrom Kampalato the simply as farming within the boundaries of the municipal broader literature on urban agriculture in Africa in order to area. In many countries in eastern and southern Africa, draw out relevant issues for urban managementand policy; farming is technically illegal in urbancentres. Nevertheless, make specificrecommendations for policy makers and urban farming is widely if uneasilytolerated in most urbancentres, planners and highlightmethods ofaddressing the urgentneed in part because at least some officials recognise informally for dialogue among farmers, researchers and urban policy that urbanfarming is an importantcomponent of urbanfood makers about urban farming and related practices. sources and in part because of the sheer impossibilityof the enforcementof bylaws. Most of the research that has been Urban Farming and Urban Management: Some Salient carriedout portrays urbanagriculture as ahousehold survival Concerns strategy,2one ofthe many ways inwhich urbanfamilies have Structural Adjustment Programmes (SAPs) have now redeployedtheir labourand other resourcesin the struggleto been implemented across Africa, with varying levels of survive in an increasinglyhostile environment. success. However,even where these programmeshave the Fairly common biases on the part of urban managers intended effect on overall economic growth, the sectoral toward urbanagriculture are reported in these studies.First, impactin the short to medium termon theurban wage earning it is almostuniversally perceived as an activity of the urban class and the poor have been well documented, both in poor that is of only marginaleconomic significance (Sanyal, Kampala and elsewhere:Austerity measures have cut the 1985:15—24 and Maxwelland Zziwa, 1992). Because it is a social programmes (Loxley, 1989) and have also cut civil second or third form of employmentor economic activity, service wages (Chew, 1990:1003—1114)and led to major census figures underestimate,often by large margins, the cutbacks incivil serviceemployment; economic liberalisation number of people or households engaged in urban farming has led to the failure of somedomestic industries (Mamdani, (Uganda, l992b). Household income surveys contribute to 1990: 427—467)and the steep increase in the price of food the perceptionby enumerating only cash income, which is (Cornia, Jolly and Stewart, 1987 and World Bank, 1993). often quite minimal compared to the value of subsistence Feeding the urban centres remains one of the salient policy consumption. Second, urban agriculture is widely perceived questions at the root ofvarious agricultural marketing reforms to be a public health hazard in terms of providing a habitat throughoutthe era of the SAP(WorldBank, 1989 and Guyer, for variousdisease vectors, both insects and rodents(Sanyal, 1987). 1985:15—24; Freeman, 1991 and Maxwelland Zziwa, 1992). Donor agencies,research organisations and governments Someauthorities also believe that urbanagriculture is a threat that are concerned about the potential food security risks in to security by providing ready hiding places for thieves in the short and mediumterm to vulnerable groups inthe context densely populated areas. Third, agriculture is perceived by of the general decline of urban formal economiesand the urbanauthorities to bean activity that simply does notbelong process of economic restructuring, have recently become in the urban centre. Farming is viewed as 'backward', and aware of the r otential of urban food production to address its presence in the urban centre reflects negatively on the someof these risks (World Commission on Environmentand standard of urban development. Complaints about the Development, 1987; Cornia, Jolly and Stewart, 1987; (;ontempOrary 'ruralization' of urban centres are frequent Atkinson, 1992; von Braun et al., 1993 andBart, 1994). The (Bibangambah,1992 and Mlozi,Lupanga and Mvena, 1992) conventionalargument is that SAPis designedto make rural and it is widely believed that it is only the most recently- farminga more economically attractivelivelihood, which will arrived migrants from rural areas who practice farming in solve the urban food problem through increased incentives the urban centre (Sanyal, 1985:15—24; Maxwelland Zziwa, for rural production.However, ia the short to mediumterm, 1992). the burden on low and middle income urban householdsis Accordngly, most studies set about to disprove some of increased. This has resulted in a rapid expansion in the these notions. The first one is reasonably simple to argue: practiceof urbanagricultur and intereston thepart of donor most studies showthat it is not onlythe urbanpoor who farm agencies and researchers and to some extent, national in the urban centre. Sawio (1993) in particular showed governments.' evidencethat all socioeconomicgroups in Dares Salaamare Only fairly recently has urbanfarming inAfrica been the engaged in farming and that, in fact, the urban poor may be subject of seriousresearch (Sanyal, 1985:15—24; Lee-Smith getting squeezed out of urban agricultural production by et al., 1987; Rakodi, 1988:495—575; Mvena et al., 1991; middle and upper class urban residentswho can gain access Freeman,1991; Maxwelland Zziwa, 1992and Sawio,1993). to land more readily.Sanyal (1985:15—24),Freeman (1991), Many of these studies have not been incorporated into and Maxwell and Zziwa (1992) showed clearly that urban

265 (ertrude Atukunda and Daniel Maxwell African Urban Quarterly farmersare primarily not the most recently-arrived migrants, concluding section of this paper suggests some means of that in fact it takes some time for a migrantto get settled in addressingthis need. the urbancentre before being able to gain accessto land for cultivation. Overview of Urban Farmingin Kampala, Uganda The secondbias is less easily challenged becauseit is not The following discussion draws primarily on research an agricultural or social question, but rather regards public carriedout by theauthors in Kampalafrom December, 1992 health. The public health implications of free-roaming to October, 1993, building on a preliminarystudy in 1988— livestock are self-evident, but in fact, much of proclaimed 1989 (Maxwell and Zziwa, 1992). The study consisted of a health danger from urban farming regards cultivatedcrops. seriesof 44 comparative householdcase studies,a two-round Sanyal (1986)noted municipal ordinances banningplants that survey of 360 householdsin three areas of the city selected may harbour mosquitoes—singlingout maize in particular. through a multi-stage sampling procedureand interviews with But healso cites a study inwhich ateam of Zambianscientists municipal authoritiesand politicalleaders. Focusgroups and showed that mosquitoes do not breed in maize. A similar key informant interviews wereused tosupplement these three study in Kampalashowed numerousother urbanhabitats to sources of primary data. The survey enumeration areas of be the primary sources of mosquito-bornedisease (Wayira et the city are depictedin figure 1. To present a briefoverview al., 1992). Because agronomic crops are singled out for of urbanfarming, three points will be discussedbelow. These regulation but ornamentalplants are not, many researchers include the origins of farming in Kampala, the extent of dismissthis issue as simply being amanifestation ofthe third farming in the city and a comparison of farming and non- bias. Only rarely, however, has research provided urban farming households in terms of general demographic managers with hard evidence that urban farming may characteristics, labour allocationand nutritional status. contribute positvely to public health, specificallyin terms of improvedfood security and nutritionalstatus. (a) The Origins of Farming The third bias ismost difficultto countersince it is purely Urban farming and the use of urban land for agricultural one of definitionand normativeviews. Perhaps as aresult of purposes has long been a common practice in Kampala iesearching a topic against which such strong normative (Southall and Gutkind, 1957). The crisis of institutional biases are held, many researchersof urban agriculture have breakdown, economicmismanagement and civil conflictthat tended to become advocates of the practice. For example, characterisedthe social life of Uganda from the onset of the Freeman (1991:121) characterisedurban farmers in Nairobi Aminregime until the mid to late 1980s created particularly as "vigorous, energetic and determined. Their plots of difficultcircumstances for Kampalaresidents (Ochieng, 1991 cropland are gardens of hope, not wastelandsof despair". and Obbo, 1991). The political consequencesof the Amin Sanyal (1985:14—24) characterised urban farming as an regime are well known, but economically,the major impact "innovative response" to theproblems that confronted urban was the precipitous decline in real wages paid to the urban dwellers in the l980s; Mvena et al. (1991) noted that urban labouring class (Banugire, 1985: 52—66; Jamal, 1987 and farming in Tanzania is a "normal way of life". Dettweiler Jamal and Weeks, 1993). (1985) described the urban farths of Senoufo migrants in The responses at the household level to this massive Bamako as constituting an "unqualified success". Much of decline in real income from wages have been characterised this advocacy revolvesaround employment and theprovision in various ways, but in general can be described as the of a non-market source of food, particularly for the urban diversificationof incomesources and the informalisationof poor. work (Bigsten and Kayizzi-Mugerwa, 1992: 1423—1441; Almostwithout exception, theconclusions of thesestudies Jamal and Weeks, 1987 and 1993: 271—291; Jamal, 1988: reflect the original conclusion of Sanyal (1985:19) that in 679—701; Banugire, 1985: 52—66and Obbo, 1991).However, Lusaka"urban agriculture is predominantly a coping strategy this generaltrend has taken many forms. The firstof these is adopted by households whose monetary incomes are the dramaticallyincreased participation of urban women in insufficient for purchasing adequate amounts of food." income-generating activities(Obbo, 1991;Basirika, 1992 and Sanyal's study preceded recent research on individual Manyire, 1993). Much of this activity has been in the form strategies and thedifference between 'coping' and 'adapting'. of small-scale, informal trading and in the preparation and Streifeller (1987: 8—13) specificallyemphasised the need to sale of food. The second,partly overlappingwith the first, is have a food source that avoids middlemen. Memonand Lee- vastly increased proportionsof the population engaged in Smith (1993:25—42especially p. 38) underline the point that small-scale manufacturing, informal trade or informal "subsistenceproduction is a common but poorly documented delivery of services, given the extent to which municipal strategy adopted by the urban poor to feed themselves." infrastructure had broken down. Much of this activity However, at least one recent study claims that it is not constituted a second or third source of income at the primarily the urbanpoor who benefit from urban agriculture household or even individual level. A third category of (Mbiba, 1995). responseswas linked largely to the high cost of food in the The strong differencesin normative views toward urban city. These included making changes in the basic diet of the agriculture between researchers who have investigatedthe urban population, mostly substituting cassava or grains for practiceand urban managersresponsible for policiestoward more expensive plantains (Jamal, 1987) and strengthening the practice suggest an urgent need' for dialogue. The linkages with rural kin, tradingservices for food (Jamal and

266 Volume 11 Nos. 2 and 3, May andAugust, 1996 Farming in the Cityof Kampala: Issues for Urban Management

Figure 1, Administrative divisions ofKampala, Uganda

NAKAWA

CENTRAL RUBAGA

1. Kiswa Parish

2. Najjanankumbi Parih

3. Luwafii Parish

(Etimeration areas highlighted)

267 Gertrude Atukunda and Daniel Maxwell African Urban Quarterly

Weeks, 1987: 271—291; and Obbo, 1991). And, of course, market.The vast majority offarmers areproducing primarily also related to the high cost of foodin thecity, urbanresidents for home-consumption, thoughthey may sellvarying amounts and, in particular, urban women increasingly used small of their producedepending on the need for cash, alternative parcelsof land in the city for agriculturalproduction. While sources of income, and intra-household distribution of farming has always been a minor factor in Kampala's income. With the exception of the small, commercially economy, during the past two decades it has become much oriented group,urban agriculture in Kampala representsthe more prevalent even in the most densely populated parts of phenomenon referred to by Mingione (1991) as semi- the city. proletarianisation: reliance on either wage labour or petty Tn 1968 Kampala was enlarged and this territorial trading (or both), as well as on home-productionfor expansion brought into the city pen-urban areas in which consumption. both population density and land-use practices were no The vast majorityof labourin urban farming in Kampala differentfrom the surroundingrural areas.Hence, for thepast is that of women. Survey data show roughly the same 20 years, there have been two different forms of agriculture proportionsof engagement in urbanfarming across all income in the city—increaseduse of vacant land within the city's groups, ethnic groups, ages and across differing levels of centre for farming on a small scale, and continued farming education among both men and women (table 2). Thus, of land withinthe expandedboundaries of thecity. Bothhave been subject to the same set of municipal rules that ban Table 1, Extent ofF in , Uganda farming in the city. Nevertheless, the extent of urban arming Kampala agriculture in the city has meant that, municipal bylaws Proportionof households Area of notwithstanding, the practice has largelybeen toleratedeven farming land farmed while remainingsubject to occasionalcrackdowns. Enumerationarea N % Median Mean SD Data fromthe case studiesand survey showthree different Luwafua 64 54.7 820 time when numbers of farming 1,204 1,040 periods large people began Kiswab 33 28.7 125 206 269 in the The first of these has been city. groups always engaged NajjanankumbiC 24 20.7 260 416 416 in and the areas where live have been farming, they TOTAL 121 34.8 380 776 913 into the Asecond in incorporated city. group began farming aA low to middle-income area; around time that residents rapidly growing pen-urban population the mid-1970s, the city began density less than 30people/ha in theenumeration area facing severe economic problems in the wake of the Amin bA low-income,densely populated inner city slum; populationdensity regime's 'economic war'. A third group began in the late around 190 people/ha in the enumeration area I 980s. Whilethis timeperiod corresponds with thebeginning cAn older,mixed-income residentialsuburb; population density around 70 in theenumeration area of the policy impacts of SAPs, it also corresponds with people/ha Source:Authors' 1993 dramaticallyimproved security in the city, making possible survey, economic activities that were too risky to carry out prior to 1986. Evidence from thehousehold case studiessupports both farming is an activity largely carried out by women, but it the impact of structuraladjustment and improvedsecurity as cuts across most other social groupings in the city. The only reasons why more people began farming in the late 1980s. group that is significantly less involvedare people who have only recently arrived in the city, who have not been able to (b) The Extent of Urban Agriculture acquireany land for farming. Kampala is comprisedof 213 square kilometres(21,300 The most commonly cited immediate reason given for hectares). The most recent estimate for the city puts farming is the need to provide a source of food for the agricultural land-use at a total of 11,942 hectares, or 56.1 household. Income is somewhat more commonas the major percent of the total land area of the city (Uganda, 1992a). motivefor livestock-keeping. However,there are three main This figure includes some areas which, while within the reasons whypeople do not engagein farming.First and most municipalboundaries, cannot reallybe describedas 'urban', obvious was the issue of access to land: A number of case with populationdensities as low as six people per hectare. study respondentsnoted that they had previouslyfarmed in Nevertheless, all the area withinthe municipal boundaries is Kampalabut had lost access to their land. The majority were subject to the same urban bylaws, which technicallydo not simply unable to acquire land at all, and so had never been permit the use of land for agriculturalpurposes. Survey data able to farm in the city. The second category of constraint in table 1 show an estimated 34.8 percent of Kampala revolvedaround the issueof time, including the large amount householdsare engaged in some form of agriculture within of time it may take to locate land for farming. For well-off thecity. Using 34.8 percentas theestimator for theproportion respondents, the issue of time is related to opportunitiesfor of households engaged in farming within the city, urban earning alternativesources of 'moonlighting'income. A third agriculture directly affects the livelihood and/or diet of and relatively minor category simply has no desire to farm roughlyhalf the residentsof Kampala.3 in the city or to producefood.

(c)Who Farms and Why? A Comparison of Farmingand (d) StatisticalComparison ofFarming and Non-Farming Non-Farming Households Households There is arelatively small numberof commercial farmers Table 2 comparesfarming and non-farminghouseholds within the city producing primarily for sale to the urban in the present sample for a number of variables. Several

268 Volume 11 Nos. 2 and 3, May and August, 1996 Farmingin the City ofKampala: Issuesfor Urban Management

Table Characteristics ofHouseholds: Total Number of 2, Farming Table 3, Mean Characteristics of Households: Fanning and Non- and in Uganda Non-Farming Kampala, Farmingin Kampala, Uganda Total Farming Non-farming Non- Variable N % N %b N %b Total Farming farming 1-test Total Sample 348 100.0 121 34.8 227 65.2 Variable Mean SD Mean SD Mean SD Diff p-value FemaleHousehold Household Head 94 27.0 34 36.2 60 63.8 Size 4.9 3.1 6.3 3.1 4.2 2.8 2.1 <0.01 MaleHousehold Number Head 254 73.0 87 34.3 167 65.7 Employed 1.4 0.7 1.4 0.8 1.3 0.6 0.1 NS Very Low Income 55 15.8 20 36.4 35 65.6 Children Low Income 219 62.9 71 32.4 148 67.6 <5 Years 1.0 1.0 1.3 0.9 0.9 1.0 0.4 <0.01 LowerMiddle Education Income 53 15.2 22 41.5 31 58.5 of Man 10.1 3.9 9.0 4.8 9.5 4.0 0.5 NS UpperMiddle/ Education High Income 21 6.1 8 38.1 13 61.9 ofWoman 7.6 4.2 7.8 4.3 7.9 4.1 0.1 NS Farm Outside Ageof Kampala 86 24.7 26 30.2 60 69.8 Man 34.1 10.1 37.2 11.8 32.1 8.7 4.8 <0.01 Permanent in Age of Kampala 219 62.9 92 42.0 127 58.0 Woman 31.1 11.8 34.4 13.3 29.1 10.3 5.3 <0.01 apropoion oftotal Length of bpropoi.tion ofcategory Residence in Kampala 15.5 13.5 20.2 15.2 12.9 11.6 7.2 <0.01 Source:Authors' survey, 1993 Source:Authors' survey, 1993 important facts are brought out by these data. First, there is scarcely any difference between farming and non-farming Table 4, Demands on Women's Time: Farming and Non-Farming householdsin many of socioeconomic categories.Just over Houseboidsa one-thirdof the households(34.8 percent)in the samplehad Farming Non-farming 1-test someone engaged in urban agriculture. This proportion Activity Mean SD Mean SD Diff p-value remains relativelyconstant if analysed separately in terms of Wage Labour many other variables depicted: all income groups, intended and/ or permanence of residence in the city, even sex of head of Trading 2.7 3.8 4.5 4.0 1.8 <0.01 household(table 2). House Work' 4.5 2.4 4.4 2.5 0.1 NS The only variables that differ significantly between ChildCare 5.2 4.3 3.2 3.7 2.0 <0.01 farming and non-farminggroups arehousehold size, number Farming 2.0 1.7 — — — — of children,age of both male and female adults and length of Personal residence in Kampala(table 3). Farmingrespondents reported Timec 1.2 2.5 1.7 3.1 0.5 0.02 being in the city an average of 7.2 years longer than non- Total Time 15.5 3.9 13.9 4.0 1.6 <0.01 On had farming respondents. average, a farming respondent aDatafor round one ofsurvey, April, 1993 lived in Kampala for 9.5 years before beginning to farm, bHouse work is combinedtotals of cooking, cleaning, fetching water. althoughthe answers were quite variable. Thesedata indicate and buyingneeded householditems a lack of any strong relationshipbetween urban agriculture cpersonal time is combined totals of leisure time, time devoted to and income formal or levels of formal volunteer activities,and aresidual category that mostfrequently referred level, employment, to time in formal education education. But do that householdsare spent they suggest farming Source:Authors' likely to be larger, include more children and have survey, 1993 significantlyolder adults. They are also likely to have been and have less free time. the most salient is that resident in Kampala longer. Perhaps point farmingpermits mothers greater time to devoteto directcare (e) Allocation of Women's Labour of their children than do other economic activities. These data are fromround one of the carried out the A smallgroup of respondents listed lackof time as amajor survey, during season. Data from round two were constraintto farming; all but one were involved ineither wage rainy similar, though differences tended to become smaller in amounts of time labour or trading away from their householdsfor a major period of time every day. Table4 comparesthe demandson devoted to wage labour and trade, as well as to direct child women'stime infarming and non-farminghouseholds during care. the first round of the survey. These data bring out several points. One is that farming is an activity in additionto, not (f) Nutritional Statusof Pre-School Children instead of, other activities both within the household and Table 5 presents data comparing nutritional status of outside of it. Farming women devote somewhatless time to children under the age of five years in farming and non- outside economicactivities such as wagelabour or trade, but farminghouseholds in differentincome groups. The measure devote the same amount of time to other householdchores used is height for age expressed as Z-scores, or standard

269 Gertrude Atukunda and Daniel Maxwell African Urban Quarterly

Table 5, Height for Age by Income Group and Farming (Round I: UnadjustedHAZ, confirmed birth dates only, N=293) Farming Non-farming Income T-test Group groupa N HAZb N HAZb N Diff p-value means VL 29 -0.561 26 -1.918 55 1.357 0.001 -1.203 L 71 -0.468 100 -0.900 171 0.432 0.010 -0,721 LM 21 -0.161 23 -0.727 44 0.565 0.079 -0.457 UM/H 7 0.549 16 0.962 23 -0.413 0.548 0.836 Group Means 128 -0.383 165 -0.856 293 0.473 0.002 -0.650 ANOVA F= 1.980 F=20.697 F= 15.692 p=0.119 p

270 Volume 11 Nos. 2 and 3, May andAugust, 1996 Farmingin the Cityof Kampala: Issuesfor Urban Management

Despiterepeated attempts both in 198 8—89 and 1993 toreview (a) Facilitating the Research/Policy Dialogue municipal bylaws in Kampala, we were never able to get The strongly-heldviews of urban authoritieson the one accessto them. In 1993, we concludedthat while suchbylaws hand, and the hypothesison the other that urban farming probably existed at some point, documentation of their represents an importantstrategy of urban women to protect existence may have been lost during the chaotic years of the food security of their children and families, led us to public administrationin Kampalaand the bylawshave never believe there was an urgent need not only to collect and been revived. analyse data measuring the impact of urban farming, but to Various respondents,including a number of localpolitical undertake the research in such away as to facilitate adialogue leadersthroughout the city, notedthat there had been aseries between researchers and policy makers, and ultimately to of 'decrees' during theAmin period, someof themfrom Amin facilitatea dialogue between urban farmers themselves and himself, alternatively encouragingpeople to cultivate so as policy makers4.We devised a methodological process to clear areas of the city that had grownup in tall grass, and wherebythe views of a broad range of policy makers were exhorting people to stop farming because it made the city sought early in the research, after which we attempted to look 'bushy'. These were not formal decrees, they were maintain contact with these policy makers, asked them to mostly announcementson Radio Uganda. At some point in commenton early findingsor interpretations of findings,and, the mid to late 1970s, one of these decrees made it illegal to in manycases, askedthem to accompany us in the field.This cultivatecrops over afew feet high for 'security' reasons.In process helped us to further understand official's worries other words, maize,cassava, plantains, sugarcane, etc. would about urban farming;it also helpedus to identifythose whose not be tolerated, but beans, vegetablesand sweet potatoes ideason thesubject were different from the abovedescription, would.This wasfollowed by amajor crop-slashing operation or who were open to change. Werecognised early on that the in a number of areas of the city. While the outcry over the difference in views was not just between the municipal crop-slashing was such that no subsequentcity authorityhas bureaucracy and ourselves, or the bureaucracy and urban ever tried a massive operation again, the 'no-crops-higher- farmers: local political leaders in the city often differed than-two-feet'rule is still commonlycited. sharply with municipal authorities over not only urban Thoughofficials we interviewed rarely expressed the view agriculture, but a whole range of informal economic openly,the 'legal status' of urban farming may be due more activities.5 to a normative view held by city authorities than to either Toward the end of our field work, with supplementary municipalbylaws or national legislation. This normative view funding from the International Development Research Centre, holds that the city shouldbe amodern, well-organised place. we organised a day-long workshop/seminarto which we The 'bushy' appearanceof acity in which peopleare farming invited members of the research community and the every piece of open land may imply that the city is notbeing municipaland ministerialpolicy makers and authoritieswith professionally managed. The views of city officials are whom we had maintainedcontact. We also invited many of evolving—indeed there were fewer complaintsin 1993 than the local political leaders from areas of the city where we in 1988—1989 from farmers about harassmentby municipal had carried out field work, and representatives of the authorities, although, as the vignettein theprologue indicates, individuals and households whom we had interviewed, harassmentand crop-slashingdo still occur.However, direct includingurban farmersand some respondentswho were not harassmentof farmers and slashingof crops isonly one issue farmers. A portion of the workshop was devoted to a related to the legal status of urban farming. Officialsof the presentationof our research findings, but much of the day Ministry of Agriculture and various non-governmental was spent in debate.The format of the workshopmeant that organisations noted that they would have an interest in skepticalmunicipal authoritieswere often not engaged in a offeringvarious forms of projectsupport to small-scaleurban debate with the researchers(ourselves), but either with their farmers, but they currently shy away from the idea because own colleagues who were also present, with local political of the legal status, or at least the perceived legal status, of leaders, or, in severalheated exchanges, with thevery women urban agriculture. who do the bulk of the farming in the city. This setof legal circumstanceshas three impactson urban Whereas many urban officialshad little time to discuss agriculture.First, it hinders the more commonlythought of urban agriculture in 1988—1989 and many were openly forms of assistanceto small farmersin terms of inputs, credit skepticalabout our researchin early 19936there was anotable and training (some of which occurs anyway, in spite of the thawingof officialattitudes by the end of 1993. Our research lack of policy support for it). Second, it prevents specific was only one of severalpossible explanations for this change programmesfrom being formulatedin areas such as soil and in view. A long-range planning group within the Kampala water conservation, wetland degradation and control of City Council funded by the WorldBank First Urban Project livestockwastes, all of which do present real environmental duringroughly the same time periodas our field work sought and health problems in the densely populated,built-up area to legitimise and incorporate into urban planning efforts that constitutes the city. Third, it prevents some of urban small-scaleand informal sector activities of all types in the agriculture'spotential torecycle waste(Smit and Nasr, 1992), city (van Nostrand, 1993). The late 1980s and early 1990s and leaves land-use planning in the somewhat anomalous have seenconsiderable mobilisation ofinformal-sector actors situation that over half of the land in the city is being used themselves in protection of their livelihoods, particularly 'illegally' (Uganda, 1992a). throughlocal politics.While urbanfarmers are notorganised in the same manner as are, for example,the hawkers who

271 Gertrude Atukunda and Daniel Maxwell AfricanUrban Quarterly carry out much of the informal retailing in the city, or the on this principle,but politicalleaders are often at odds with taxi and mini-busowners who constitutethe city's informal thetechnical staff of theKCC over preciselythis issue. Urban transportation system, urban farmers have been active in agriculture is onlyone of manyexamples that couldbe raised. politics at the most local level, andthe impact of thishas not The second is that, to some extent, the formulation and gone totally unnoticed at higher levels. preservation of bylaws relates more to the definingof an ideal In 1988—1989 it was quite common simply to be told by city than to the managementof a real city. So long as this is officials that there was nothing to discuss about urban the case, and so long as farmingis viewed as somethingthat agriculture. In 1993, however, many interviews with officials simply does not belong in the city, the chances of legalising ended in virtually reversed roles—with officials asking us farming in the city arediminished, even though planners are for information about urban farming rather than vice versa. aware that over half the land in the city is used for farming. At the seminar/workshop, urban planners and city political Manybylaws simply abolish certainunwanted practices— leaders generally agreed that farming in the city should urban agriculture isjust one of many examples.This clearly perhapsbe regulated,but that itscurrent legal statuswas both does nothelp in addressing whateverit is that is undesirable unfair and unproductiveand does not address the real areas about the unwanted practice. A primary example is the of concern about health and environmental hazards. environmental impact of urban agriculture,particularly in a A recently-adopted long-term physical plan for Jinja, physical location such as Kampalawhere much of the land Uganda's second largest city, actually promotes some land availablefor farminghas been deemedunattractive for other use for agricultural purposes, both for environmental forms of land use. Several other important issues arise protection (preservation of green-belt areas), and in including soil and water conservation; the protection of recognitionof its prevalence in the city (Jinja City Council, wetlandsand the role of wetlandsin the preservation of the 1993). Similar recommendationshave been formally made urban ecology; the potential for urban waste recycling and in the case of Kampala's long-termplanning (van Nostrand, the possibilityof thepollution of foodstuffs grown in cities; 1993). and the health impact of livestock production in densely populated areas.All thesecertainly urgently require attention. (b) SpecificPolicy Implications in Kampala Yet, if the general practiceof urbanagriculture remainsillegal, Provided that some space has been created for dialogue littlecan bedone to regulate someof its mostenvironmentally between urban farmers, political leaders and municipal damagingaspects, or suggest incremental improvements. A authorities about the merits of urban agriculture, some similar observationcould be made with regard to offering commentis required regardingspecific initiatives that could support to women's groups that are involved in farming—a be undertaken to support the practice. Several categories of numberof NGOs in the city expressed interest, butmost were suggestionscould be offered and include the legal status of somewhat afraid of the legal implications. All these issues farming, environmental concernsand urbanland policy. suggest that the legal status of urban agriculture should be While the precise legal statutes regarding urban changed. agriculture were never locatedand in fact probablyno longer Resolution of the legal issue would also permit the exist in writtenform, virtually all respondents in this research formulation of aland-use policy for urbanagriculture. Access agreedthat farming in Kampalais, in some way, technically to land for farming in Kampala has to some extent taken illegal.The consensus arising out of theworkshop conducted advantage of nichescreated by thecomplexity of land tenure with policy makers, RC leaders, researchers and urban inthe city and by administrative turmoil.Institutional reforms farmersat the conclusion of field work in 1993 was that the in urban land are currently being implementedor proposed legal status of urban agriculture was unfair and unhelpful that would increase the tax base and streamline urban land and should therefore be changed (Maxwell, 1993b). administration (Nsamba-Gayiiya, 1993 and Mabogunje, Unfortunately,there is no one bylaw that can simply be 1992), and which would represent a substantial threat to deleted from the books.There is, rather, athree-decade long continuedinformal access to urban land for farming.8The series of decrees, written and unwritten rules and their extent of agriculture withinthe city and theproportion of the interpretations,and biases against urban agriculture.These populationthat relies on urban agriculture as an important can be changed, but they cannot be wiped out by safeguardto food securityare both importantconsiderations administrative fiat or a political vote. Underlying these in a land-use policy. Such a policy would, at a minimum, interpretations and biases are two fundamental issues address the issues of mixed use zoning, permanent regardingurban planning and urban management. designationof certain areas for farming as 'productivegreen The first is the question of whether urban planning, or belts', and the questionof compensationfor eviction or loss developmentplanning more broadly, isprimarily a technical of land for farming.9 process to be engaged in only by technically trained In her wide-rangingreview of urbanisation in Africa, individuals, or a political process in which everyone is Coquery-Vidrovitch (1991) suggested that urban policy involved. Clearly,the preparingof detailedphysical plans is shouldbe encouraged to improve upon, notsuppress, informal atechnical process, butthe determination ofpriorities in what economic activities. Altering formal urban planning activitiesare planned for what locationsis, or ought to be, a proceduresor zoning and land-use ordinances to enable the broadly-basedpolitical process. The mode of local politics growth of informal economies is a notoriously difficult inKampala and elsewherein Uganda7is fundamentally based process. Little research has been carried out on the ways in

272 Volume 11 Nos. 2 and 3, May and August, 1996 Farming in the City ofKampala: Issuesfor Urban Management which formalpolitical processesat the local level incorporate The Resistance Council System, see note 5 supra. or accommodateinformal economic activities. There is need a 8 for such not with to urban This paperhas not analysed in detail the question of informal research, just regard agriculture, access to urban land for The of but also to informal economic activities of in agriculturalpurposes. majority many types urban farmers do not own or hold customary rights to the land many places, if urban authorities are to be able to play a theyfarm; abouttwo-thirds of all parcels being farmedhad been genuinelyfacilitative rather than prohibitiverole. This paper accessed through some informal means—illegal sub-division, has outlined, both procedurally and substantively, one borrowing and squattingbeing the most common means. New approachto such research. tenure rules that crack down on these practices, or take away the incentives for land lending, would have an obvious impact Notes on urban farming. See Maxwell(1995) and (forthcoming). In addition to the works just mentioned, this includes global initiativesthat have resulted (or will result inthe nearfuture) in In late 1994, new land-use regulations were adopted by the publication of books on the subject by the International KampalaCity Council that, amongother things, permit "home DevelopmentResearch Centre of Canada (Egziabher,Memon, basedagriculture, including thegrowing of crops and thekeeping Mougeot,Lee-Smith, Maxwell and Sawio, 1994), and the UN of livestock."(Kampala City Council, 1994: 3) DevelopmentProgramme (Smit, Nasr and Ratta, forthcoming). Entire issues of severaljournals have recently been devotedto References the topic (seeHunger Notes, Vol. 18(2);IDRC Reports October, Atkinson, S. (1992). Food for the Cities: Urban Nutrition 1993; International AgSieve, Vol.6(2); and ILEIA Newsletter, in Countries. Urban Health Program, as this of Policy Developing Fall, 1994, as well edition African Urban Quarterly). Health Unit. of Public Healthand Policy, In when this research first in a Policy Department 1987, began Kampala, London Schoolof and Medicine. computerisedliterature search turnedup a total ofone reference Hygiene Tropical to urban agriculture in Africa; a comprehensivesearch today might turn up 75—80 references. Banugire,F. (1985). "Class Struggle,Clan Politics and the Economy",Mawazo Vol. 6 No. 2. 2 'Magendo' Findingsfrom ourstudy question this interpretation,suggesting rather that urban farming is largely a women's strategy. C. Urban Food A Guide to Critical the of food is of the Bart, (1994). Security: Responsibility for provision one most Issues. Atlanta: CAREFood Unit. contentiousissues in urban intra-householdrelations (Maxwell Security 1994a and 1994b).This point is importantin the consideration of policy alternatives. Basirika,E. (1992). StructuralAdjustmentand Womenin the InformalSector: A Study ofMarket Women. FriedrichEbert The precise figure is 473,800 out of a population of roughly Foundation/UgandaEconomics Association. 900,000 in 1993. This figure is conservativelycalculated by the 1991 and Census taking Housing Population figures J. "Macro-LevelConstraints and the (Ministry of Planning and Economic Development,1992) for Bibangambah, (1992). the number of householdsin Kampala, the lowest of the three Growthof the Informal Sectorin Uganda",in J. Baker and P. projectedpopulation growth rates for the city (van Nostrand, 0. Pedersen (eds) The Rural-Urban Interface in Africa. 1993),andsurvey estimates for the proportion of households Uppsala: Scandinavian Instituteof African Studies. engaging in agriculture in the city and for average size of household in With the of the engaging agriculture. exception Bigsten, A. and S. Kayizzi-Mugerwa (1992). "Adaptionand figure for the population of the city, all of these are probably underestimated.And this calculationdoes not accountat all for Distress in the Urban Economy: A Study of Kampala World Vol. No.10. commercialproduction within the city. Households", Development 20

' Theseincluded officials from KampalaCity Council, the District Chew, D. (1990). "Internal Adjustments to Falling Civil Administrator's Office, the Ministries of Agriculture, Lands, Service Salaries: Insights from Uganda", International Housing and Urban Development, Health, and Economic Labour Review Vol. 18. Planning. C. "TheProcess Urbanisation In Kampala,as elsewhere in Uganda, local politicsis organised Coquery-Vidrovitch, (1991). of under theResistance Council (RC) System, a pyramidalsystem in Africa",African Studies Review Vol. 34 No.1. where at the most localised level, there is direct democratic participation,but wherethe electedleaders at one levelbecome Cornia, G. et al. (1987). AdjustmentWith a Human Face. the 'constituency'at thenext higher level in the system—in other Oxford: Clarendon Press. words, ward,division and city-widepolitical leaders are elected indirectly. We interviewedpolitical leadersat variouslevels and locationswithin the city throughout the RC system and found Dettweiler, S. (1985). Senoufo Migrants in Bamako: markedly different views towards urban farming than among ChangingAgricultural Production Strategies and Household bureaucraticallyappointed municipalauthorities. Organization in an Urban Environment. Bloominstein: Unpublished Ph.D. dissertation,Indiana University. 6 In 1989, an interview was terminatedrapidly when the officer concernedsaid that urban was and hedid not agriculture illegal et al. Cities Ottawa: grant interviews on 'criminal activities' (Maxwell and Zziwa, Egziabher, (1994). Feeding People. IDRC Books. 1992:53).

273 GertrudeAtukunda and Daniel Maxwell African Urban Quarterly

Filliam, E. (1993). "City Farmers Face KCC's Wrath", Mamdani, M. (1990). "Uganda: Contradictionsof the IMF Monitor; June 11:4. Programme and Perspectives", Development andChange Vol. 21. Freeman, D. (1991). A City of Farmers: Informal Urban Agriculturein the Open Spaces of Nairobi, Kenya. Toronto: Manyire, H. (1993). Macro Trends, Micro Processes and McGill UniversityPress. Emerging Social PolicyIssues: SomeEvidence from Urban Women'sMarket Tradingin Uganda. Paperpresented to the Gibson, R. (1990). Principlesof Nutritional Assessment. New Annual Meeting, CanadianAssociation of African Studies, York: Oxford UniversityPress. Toronto, May 12—15.

Guyer, J. (1987). Feeding African Cities. Bloomington: Maxwell, D. (1993a). Land Access and Household Logic: Indiana UniversityPress. Urban Farming in Kampala. Kampala: MISR Research Paper, MakerereInstitute of Social Research. International Labour Organisation (1989). "African EmploymentReport, 1988", AddisAbaba: ILO/JASPA. Maxwell, D. (1993b). SeminarReport: Farming in the City. Kampala: MakerereInstitute of Social Research. Jamal, V. (1987). "Ugandan Economic Crisis: Dimensions and Cure", in Dodge and Wiebe (eds) Beyond Crisis: Maxwell,D. (1994a). UnplannedResponses to the Economic Development Issues in Uganda. Kampala: MakerereInstitute Crisis? Urban Agriculturein Kampala. Paper presented to of Social Research. the Workshop on "DevelopingUganda", Centre of African Studies, University of Copenhagen, held at Lyngby Jamal, V. (1988). "Coping Under Crisis in Uganda", Landbrugsskole, Roskilde, Denmark,June 2—5. Forthcoming International Labour Review Vol. 128, No. 2. in N. Hansen and P. Twaddle (eds), Developing Uganda. London:James Currey. Jamal, V. and J. Weeks (1987). "The Vanishing Rural-Urban in Sub-Saharan International Labour Review Gap Africa", Maxwell, D. (1994b). Internal Struggles over Resources, Vol. 127 No.3. External Struggles for Survival: Urban Women and Subsistence Household Production. Paper presented to the V. and J. Jamal, Weeks (1993). Africa Misunderstood. 37th Annual Meeting of the African Studies Association, London:MacMillan. Toronto, November3—6. Council Recommended Structure Jinja City (1993). Jinja Maxwell, D. (1995). Labour; Land, Food, and Farming: A Plan, 1994. Council. Jinja: Jinja City Household Analysis of Urban Agriculture in Kampala, Uganda. Ph.D. Dissertation, University of Wisconsin- Kaijuka,E. et al. (1989). Demographicand Health Survey. Madison. Entebbe:Ministry of Health. D. "Alternative Food Kampala City Council (1994). Kampala Structure Plan. Maxwell, (forthcoming). Security Annex1: Land Development Regulations. Kampala: Kampala Strategy: A Household Analysis of Urban Agriculture in City Council. Kampala", World Development. Lee-Smith, D. etal. (1987). Urban Food Productionand the Maxwell,D. and S. Zziwa(1992). Urban Farming inAfrica: Cooking Fuel Situation in Urban Kenya. Nairobi:Mazingira The Case ofKampala, Uganda. Nairobi: ACTS Press. Institute. Maxwell, D. and S. Zziwa (1993). "Urban Agriculture in Loxley, J. (1989). "The IMF, the World Bank and Kampala: Indigenous Adaptive Response to the Economic Reconstructionin Uganda", in P. Campbell and M. Loxley Crisis", Ecologyof Food and NutritionVol. 29. (eds) StructuralAdjustment in Africa. New York: St Martin's Press. Mbiba, B. (1995). Urban Agriculture in Zimbabwe: Urban Urban The A. "Urban and the Post- Implicationsfor Management, Economy, Mahogunje, (1990). Planning and Gender Harare: of Colonial State in Africa: A Research Overview", African Environment,Poverty University Zimbabwe. StudiesReview Vol. 33 No. 2. A. on Urban Land and Memon, P.A. and D. Lee-Smith(1993). "Urban Agriculture Mabogunje, (1992). Perspectives in Canadian Journal Studies Vol. 27 No. Urban ManagementPolicies in Sub-SaharanAfrica. World Kenya", ofAfrican Bank, Washington, DC: WorldBank Technical PaperNumber 196.

274 Volume 11 Nos. 2 and 3, May andAugust. 1996 Farmingin theCity ofKampala: Issuesfor Urban Management

Mingione, E. (1991).Fragmented Societies:A Sociology of Smit, J. and J. Nasr (1992). "Urban Agriculture for Economic Life Beyondthe Market Paradigm. Oxford: Basil Sustainable Cities: Using Waste and Idle Land and Water Blackwell. Bodiesas Resources",Environment and UrbanisationVol.4 No.2. Miozi, M.R.S., I. J. Lupanga and Z. S. K. Mvena (1992). "Urban Agriculture as a Survival Strategy in Tanzania, in Smit, J. et al. (forthcoming). Urban Agriculture:A Neglected Baker and Pedersen (eds) The Rural- Urban Interface in Resourcefor Food, Jobs andSustainable Cities. Washington: Africa. Uppsala: The Scandinavian Institute of African KumarianPress. Studies. Southall, A. and P.Gutkind (1957). Townsmenin theMaking: Mvena,Z.S.K. et al. (1991). UrbanAgriculture in Tanzania: Kampala and its Suburbs. Kampala: East African Institute A Study of Six Towns. Morogoro, Tanzania: Sokoine of Social Research. Universityof Agriculture. Streifeller, F. (1987). "Improving UrbanAgriculture in Africa: Naipaul,V.S. (1981). "A New King for the Congo: Mobutu A Social Perspective", Food and Nutrition Bulletin Vol. 9 and the Nihilism of Africa", in The Return of Eva Peron. No. 2. New York: VantageBooks. Uganda,Government of (1992a). City ofKampala: Revision Nsamba-Gayiiya, E.(1993). Reportof the National Workshop ofStructure Plan. Kampala: Ministryof Land, Housingand on Urban Land Policy andManagement. Kampala: Ministry Urban Development, Department of Physical Planning. of Lands, Housingand Urban Development. Uganda, Governmentof (1992b). The 1991 Population and Obbo,C. (1991). "Women,Children and a 'Living Wage' " HousingCensus. Entebbe Ministry of Planningand Economic in N. Hansen and P. Twaddle (eds) Changing Uganda. Development: Departmentof Statistics. London:James Currey. van Nostrand, J. (1993). Kampala Urban Study Phase / Ochieng,E. (1991). "EconomicAdjustment Programmes in Report. Kampala: John van Nostrand Associates. Uganda, 1985—8", in N. Hansen and P. Twaddle (eds) Changing Uganda. London: James Currey. von Braun, J. et al. (1993). Urban Food Insecurity and Malnutritionin Developing Countries: Trends, Policies,and Rakodi, C. (1988)."Urban Agriculture:Research Questions Research Implications. Washington, DC: International Food and the Zambian Evidence", Journal of Modern African PolicyResearch Institute. StudiesVol. 26 No. 3. Wayira, J.R.M. et a!. (1992).Report on Ecological Malaria Riley, I.E. (1987).Nutrition and Health Survey of Kawempe Control Pilot Project. Kampala: Kampala City Council. Sub-District. Kampala: Save the ChildrenFund. World Bank (1989). Sub Saharan Africa: From Crisis to Sanyal, B. (1985). "Urban Agriculture:Who Cultivates and Sustainable Growth. Washington, DC: WorldBank. Why?", Food andNutrition Bulletin Vol. 7 No. 3. World Bank (1993). Uganda:Agriculture. Washington DC: Sanyal, B. (1986). Urban Cultivation in East Africa. Paris: The World Bank. Food andEnergy Nexus Project, United NationsUniversity. World Commission on Environment and Development Sawio, C. (1993). Feeding the Urban Masses? Towards an (1987). Our Common Future. Oxford: Oxford University Understandingof the Dynamics of UrbanAgriculture in Dar Press. es Salaam, Tanzania. UnpublishedPh.D. Dissertation,Clark University.

275 I

An InternationalJournal on Human Geography and Environmental Sciences

Editors: H. van der Wusten University ofAmsterdam, Dept. ofHuman.Geography, the Netherlands. 0. Gritsa Russian Academyof Sciences, Moscow, Russia Editorial Board: Ewan Anderson, UK; M. Barlow, Canada; Georges Benko, France; Stanley D. Brunn, USA; Cesar N. Caviedes, USA; S. Conti, Italy; Claudio Egler, Brazil; AntonGosar, Slov..nk;Yehuda Gradus, Israel; G. Heinritz, Germany; Arild Holt-Jensen, No,way; C. Gregory Knight, USA; David B. Knight, Canada; LeszekA, Kosinski,France; Vladimir Kotlyakov, Russia; Yasuo Miyakawa, Japan; B. Messerli, Switzerland; A. Paasi, Finland; D. Rumley, Australia; Sam Ock Park, South Korea; Ulf Strohmayer, UK; T. Stryjakiewicz, Poland; David Turnock, UK

GeoJournal is an international periodical focusedon the links between the transformationof modern society, technological development and environmental change, as interpreted by humangeography and related sciences, The sphereof interestof thejournalencompasses aH relevant processes reshaping human activity patterns in different parts of the world, the methods of their analysis and the forms of application of geographical knowledgein planning and forecasting.The editors are committed to make GeoJournal a forum forscientific discussions going onwithin the International Geographical Union as well as avehicle forthe dissemination of IGU information. It is their intention to encourage high quality contributionsfrom a wide variety of nationalbackgrounds and substantiveviewpoints for presentation to themembers ofthe international scientificcommunity, geographersand othersinterested in this field.

SubscriptionInformation: ISSN0343-2521 1998, Volumes 4446(12 issues) Subscription Rate: NLG 2385OO/USD 1224OO, including postageand handling. Urban Agriculture. Food, Jobs and Sustainable Citiesi African Urban Quarterly, 1996: 11 (2 and 3):pp. 277—278 Urban Agriculture in Zimbabwe. Implicationsfor Urban Management and Poverty

URBANAGRICULTURE. FOOD, JOBS AND SUSTAINABLECITIES by United Nations Development Programme (NewYork: UnitedNations DevelopmentProgramme, 1996, 302 pp. soft cover, price not available) URBANAGRICULTURE IN ZIMBABWE. IMPLICATIONS FOR URBAN MANAGEMENTAND POVERTY by Beacon Mbiba (Aldershot:Avebury, 1995, ISBN 1 85628 857 9, 220 pp., Hardboard,price not available)

Withthe rapidly growingurbanisation in the world—and That cannot be said of Urban Agriculturein Zimbabwe. particularly in the developing countries—theliterature on Implications for Urban Managementand Poverty, written urban agriculture is also growing. Many publications by BeaconMbiba, although this bookis certainlyof interest. regardingthe variousaspects of urban agriculture have seen The book starts with a discussion on urban agriculture in the lightduring recent years. Hence, there is a need for an general and on theissues and dimensions of urbanagriculture overview of 'where we stand' as far as urban agriculture is in Harare (Chapter 2). The next two chapters provide the concerned. Urban Agriculture. Food, Jobs and Sustainable results of a study among 'off-plot' (on public or otherwise Cities, published by the United Nations Development illegally cultivated land) cultivatorsand among 'on-plot' (in Programme(UNDP), provides this overview. The book was own gardens)cultivators, respectively. Both chapterscontain writtenfor the secondUnited NationsConference on Human some interestinginformation, although it is dealt with in a Settlements (Habitat II) in June 1996. It has four main rather superficialway. Moreover,the reader would like to purposes(p. xvii): (1) to present a comprehensivepicture of know more on such things as the actual use:of the plots, urban agriculturein Asia, Africa and Latin America; (2) to farmingtechniques, inputs, and the destination of theproduce. define a distinct industry that needs to be recognised and Chapter5 is abit of asideline and deals with the urbanmaize treated as such; (3) to persuadeleaders in government, non- markets.It is the only chapter which is not written by Mbiba governmentalorganisations, research institutions and other (and is also theonly chapter withouttyping errors).Chapter public and privateentities to conductresearch, support action 6 deals with an issue that is very important and at the same projects and eliminateunnecessary constraints to the growth time still very much underexposed in the urban agriculture of the urban agriculture industry;and (4) to foster a climate literature, notably the role of the different government that empowerspractitioners and the agenciesthat back them institutions. It providesthe readerwith anhistorical overview to fulfil the industry's potential for improving public well- of the attitudes and policies of the different authorities in being and the quality of urban life. Because of these broad Harare, which can be described as prohibitive.In the next and ambitious objectives thebook is "neither atechnical how- chapter the author lets some women cultivators speak for to manual nor an academic document" (p. xxiv)but something themselves regardingthe opportunities and constraintsin the in-betweenand is meant for a wide array of readers. Harare environment. It is apity that what ispresented is very The book consists of four parts. Part I consists of a fragmentary and concise, otherwise it could be a very conceptual and an historical chapter. Part II deals with the revealing chapter. Chapter 8 consists of 26 photographs farmers themselves, with the locations where farming is showing various aspects of urban agriculture in Harare. If practised, with the type of farming and produce, and with photographsare meant to be "tools of social research", as the organisationsinfluencing urban agriculture. In Part HI the author states, a good quality of the pictures is a the benefits, problems and constraints concerning urban prerequisite. Unfortunatelythis is not the case, but it is not agriculture areoutlined. Part IV discussesthe futureof urban clear whetherthe photographsor the printer of the book is to agriculture. The book is not just based on earlier studies, blame. The next chapter is "an experiment" and tries to covering the period from the early 1970s up to 1995, but analyse whetherany lessons can bedrawn fromthe European also on more than 300 interviews and more than 100 site (i.e. English) garden allotments for subsistence urban visits in 18 countries in Asia, Africa and Latin America. cultivationin Zimbabwe. This kind of comparisonis always The book is very well-written, making it accessible to very tricky and in fact the authordoes not come any further non-academics as well (which was a major purpose). For than the conclusion that in times of crisis (economic, war, those less acquainted with urban agriculture,the book is an etc.) urban farmingbecomes more importantin terms of the excellent source to start with. However,also for those who nationalfood supplyand becomes also more acceptedby the (think they) do know something about the topic, the book authorities. Chapter 10 is very useful in that it deals with contains a lot of infonnation the reader will not (yet) know, urban agriculture as a counter-productive activity.However, because it provides a very complete picture of the many thechapter is very brief, althoughit poses somevery essential aspects of the industry. In this context, the 50 'cases' are questions.The following chapter concernsa more positive worth mentioningas they offer a very revealing picture of example,notably the use of sewage water for farming in the the potentialof themany differenttypes of urbanfarming as pen-urban zone of Harare. Chapter 12 deals with an aspect well as the problems related to it. In short, the book is an of urban agriculture which is very controversial: urban absolute must for everybodyinterested in urban agriculture. livestockrearing. It should be added,however, that the case

277 Book Review—Dick Foeken African Urban Quarterly describedhere is different from what is usually considered huge single study but on relatively small subject specific as urban livestock rearing in that it concernsherds of cattle studies conducted over the years" (p. 6). Secondly, the from outside the town (Chitungwiza)that are herded on the presentationof the results of these specific studies remains open fields that exist within the city boundaries. Therefore rather superficial. To state that the book "goes a long way in the conclusion that "laws that prohibit cattle in urban filling the gapsexisting in the literatureon UrbanAgriculture settlementsare not as far removed from public opinion as in Africa"(p. 211) is somewhat exaggerated.Still, the author others may want us to believe" (p. 204) is not very well raises fundamentalquestions and some of them have so far fbunded. hardlybeen dealt with. Therefore,the book is certainlyworth On the whole, the book is a bit disappointing in two reading for those interestedin the topic of urbanagriculture. senses. First, the structureof the book is rather fragmentary. The title of the book suggests that the reader will get a thorough picture of urban farming in Zimbabwe. Instead, Dick Foeken the book discusses a number of rather diversifiedaspects of African Studies Centre urban agriculture in this country.A better title would have P.O. Box 9555 been "Aspectsof Urban Agriculturein Zimbabwe".This is 2300 RB Leiden probablydue to the fact that "the book is not based on one The Netherlands.

278 Volume 11 Nos. 2 and 3, May and August, 1996:pp. 279—280 The Roleof Urban Agriculturefor Food Security in LowIncome Areasof Nairobi

THE ROLE OF URBAN AGRICULTURE FOR FOOD SECURITY IN LOWINCOME AREAS OF NAIROBI by Alice M. Mwangi (African StudiesCentre, Universityof Leiden: Food and Nutrition StudiesProgramme, ReportNumber 54/1995, 1995, Soft cover, 77 pp., price not available)

Increasing evidence of the growth of urban poverty and consumed; but the erratic nature of urban incomesources is food insecurity inAfrica over the past decade has paralleled often mentionedas a constraint to food security, especially the increasedresearch attentiongiven tothe practice of semi- in the short-term. subsistence agriculture in African cities. Yet, despite the The conclusions of the study are partially limited by the frequently repeated claim that urban agriculture (UA) size of the sample. The statistical significance of the contributesto both improvedfood security at the household difference in food and nutritionoutcomes among these groups level and increasedoverall urbanfood supplies, such claims is not stated,but the differences arefairly small. It is also not arerarely empiricallytested. This study by Alice Mwangiof entirelyclear whetherthe differences found are due to urban two low-incomeareas in Nairobi, therefore, is a welcome agriculture, or to someother, pre-existing factor. The USUAP addition to the literature on UA, because it addresses the group and the other urban farming group rank higher on a household element of this claim in a cross-sectional, 'welfare index' and in terms of stated income. It is not clear comparativemanner. from the study whether these households are better off Nairobiis, in fact, becomingone of the best-studied cities because they engage in urban farming, or they have better inAfrica where UAis concernedand the literatureon UA in food security because they are wealthier anyway: such a East Africa is beginning to be characterised by certain question is verydifficult to address without longitudinal data. 'stylised' facts: The primary motivation for agricultural Other studiesin easternand southernAfrica have noted that production among low-income city dwellers is to have a UA is often not a viable prospect for the poorest urban source of food, not a source of income; crops produced are, residents.Mwangi does note, however, that in terms of their therefore, primarily relatively low (economic)value staple own perceptions,USUAP participants are better off having food crops; UA is primarily a women's activity; and the joined the project. participation of low-income groups in UA is primarily This raises a thorny methodological problem for constrained by lack of access to land. Mwangi confirmsall measuringthe impact of UA on food security and nutrition, of this with regard to both farming and non-farming impliedbut notexplicitly stated by Mwangi. In order toassess respondentsin her study. impact, comparison across farming and non-farming The study tests two basic hypotheses: First, households householdsmust somehow control for income, hut this is engagedin UA have betterfood securityand betternutritional extremely difficult under circumstances where income statas in children under age five (in comparisonto non-UA sources are numerous and varied (Mwangi notes multiple households);and second,technical assistance in UA froman sources of income/household in her study), and where NGO or other organisationcan furtherimprove food security subsistence production is morehighly valued by theproducer and nutritional status. The study confirms both of these than simply its 'farm-gate'or even 'retail' price. This remains hypotheses by comparing three low-incomegroups in two a challengeto future researchon the food security impact of Nairobi neighbourhoods: householdsnot engagingin urban UA. agriculture; householdsengaging in urbanagriculture on their A baffling finding from Mwangi's study is the linkage own; and households engaging in urban agriculture with (or lack thereof) between caloric intake, long-term food assistance from an NGO—the Undugu Society Urban insecurity and nutritional status. The non-UAgroup had the Agriculture Project, or USUAP. In general, households lowestaverage caloric intake for the 7-dayfood consumption participationin the USUAPprogramme had higherlevels of recall undertaken by the study and reported the highest percapita caloric intake, and better nutritionalstatus among number of months of food shortage (10—12 months of the pre-schoolchildren than other urban farmers not receiving year for nearlya quarter of this group). Yet there was virtually technical assistance. This second group had higher caloric no difference in either short-term (weight-for-height) or intake and better nutritional status than the third group who longer-term (height-for-age)measures of under-nutrition. do not engage in any kind of UA. Does this imply that urban groups unable to gain access to Particularly important to note is that the proportion of land for farming have developed other, equally viable food consumed by these households originatingfrom their strategies to protect the nutritional status of their children? own farming is not that high: Among the USUAP farmers, Neither the caloric intakedata presented, nor the information for instance,it is only 5% of caloric intake. This impliesthat on strategies employed to cope with food shortages,offer a theimpact of UA is not so much interms of improving overall ready answer to this question. This remains an important householdfood availabilityas it isin termsof having asource question for future research. However, Mwangi's valuable of food available when cash is short for market purchase. study shows that UA can have apositive impact on household Market purchaseremains, by far, the major source of food foodsecurity and nutritional status. Nairobi'sUndugu Society

279 Rook Review—Dan Maxwell African Urban Quarterly deserves particularcredit for its willingnessto embracethe from one angle, it is important to improve production practiceof UA despiteits legal limbo (otherNGOs have been practiceson this land,and Mwangishows that the investment reluctant to support an activity that is in contravention of in productivity pays off in terms of food security. From municipal bylaws). But, like otherpreceding studies, Mwangi another angle, however, itis important toexplore and improve noted that UA is not an option for all low-incomeurban upon all the food security strategiesof the urban poor. households. The rapid growth of low-income urban populationsin Africa tends to imply that more people will be Dan Maxwell competing for the use of less land available for UA. Thus, Nosudri Memorial Institute for Medical Research P.O. Box 25 Legon, Ghana

280 Volume 11 Nos. 2 and 3, May and August, 1996: Pp. 281—282 Enough is Enough: Foran Alternative Diagnosis ofthe African Crisis

ENOUGH IS ENOUGH: FOR AN ALTERNATIVEDIAGNOSIS OF THE AFRICAN CRISIS by Carlos Lopes (Uppsala:Scandinavian Institute ofAfrican Studies, DiscussionPaper No. 5, 1994, paper, price GBf 5.95)

Enough is Enough is a discussionpaper which offers an succeed inAfrica, the continent's economicbehaviour must alternative interpretation to the African crisis. The author, be understood and local cultures and traditions respected. Carlos Lopes, is a sociologistfrom Guinea-Bissau.He has a He goes on to list some of the economicvalues and attitudes PhD from Universityof Paris 1 Pantheon-Sorborne, and was such as primacy of consumptionover savings,among others. the founder and first director of Guinea-Bissau'sNational Lopes criticises the assumption by international financial Institute for Studies and Research (INEP). He later joined institutions that macro-economicequilibrium in Africa will UNDP where he is thecurrent deputy resident representative be attained througha rise in exportscoming from traditonal in Zimbabwe. products. He argues that comparative advantage no longer TheAfrican crisis is characterisedby heavydependency works on the same principles as the 1970s; there are other on borrowing (financial aid); deteriorating terms of trade competitors in the same riches with better environmentsand which are cutting the value of Africa's traditional exports; access to capital, as well as superior productivity;and that crumblinginfrastructure, deteriorating or vanishingservices; market protectionism from the North denies developing escalating unemployment; poverty; high child mortality to countries access; and many other assumptionsnot based on name but a few. Several views have been given as to the indigenous elements. The author also analyses the cause of the crisis and in most cases, Africa is blamed. The performance of thoseAfrican countries that implemented the World Bank, for example, is of the view that Africa's structural adjustmentprogrammes advocated by the Breton economicproblems are not primarilydue to externalfactors Woods Institutions; and discovers a generally poor but rather to internal incompetencies. The incompetencies performance. The World Bank attributes the lower level of include lack of political leadership, ignorance,poor policydevelopment inthe continentto inadequate humanresources, making, and so on. Indeed, a WorldBank official described poor institution development, crumblinginfrastructure and the first three decades of African independence as having poor governance. No acknowledgement is made of the been "an economic,political and social disaster". Africa is budgetary constraintof externaldebt and exogenousshocks. viewed as a "continentwithout direction" and considereda The Bank however acknowledges that the economies of sick partner in the planet. It is with these perceptions and variousAfrican countries are not as uniform as it had been opinions about Africa that Carlos Lopes asks: Is it true that assumed. Africahas nothingto show?Are we accepting productgrowth Chapter three of the discussion paper concentrates on rate, high population growth rate and an ever increasing discussing the major trends and challenges. One of them is external debt burden, to name but a few. He then asks: Who the need tobuild social and politicalmodels basedon African is to blame for this predicament?Is there a need for sharing cultureso as tobe able to preserveexisting capacitiesas well responsibility for the failures? (p. 12). To answer the as generate new ones. However, this is impeded by the questions,the authoranalyses the state of the global economy politicalconditionalities imposedby thedonors beforegiving between 1960 and the early 1990s. He is of the view that any developmentassistance. The other challengeis to invest both donors and many western-educatedAfrican leadersare inhuman capitalprogressively and not regressively to ensure to blame for Africa's current predicament,because of their future growth. The third is the need to feed the population belief that development could be achieved by systematic and hence greater investment in agriculture. The fourth applicationof economic models developedin the West and challenge is the need to mobilise resources within the which in most cases were based on principles incompatible continent withoutrelying heavily on external funding. with the beliefs and practicesof African society. He finally The fourth chapter defines Africa's role in the world calls for the cathasis movement. economy as well as the democratic debate and ends with a In chaptertwo, theauthor discusses the lessons tobe learnt call for Africa to define its priorities.The author advocates byAfricans from theeconomic experience ofthe past decades. an integrated approach to development, that is, an approach He asserts that Africans must understandthat development takinginto consideration the linkagebetween the economic, is an internal effort and that application of development political and cultural factors, all accounting for a social models, as logical and appropriate as they may sound, is equilibrium. Theauthor is ofthe viewthat for Africato attain bound to fail if they are not based on solid historical and the desired economic growthrate, she must take control of cultural grounds. He cites several developmental models, the pace of implementationof the reform programme and from the basic needs theory to the structural adjustment also refuse conditionality. programmes,which have failed to solve Africa's problems In conclusion, theauthor concludes that theAfricans must and blames this on their theoretical development outside realizethat they arethe ones to developtheir country, not the Africa, with very little involvementof the latter.The author donors. They also have to recognise that neo-classical is of the view that for market-orientedeconomic policies to monetarypolicies have not proven adequate to the African

281 Book Review—Samuel0. Akatch African Urban Quarterly values. This can only be done when the feeling of defeat is for both economists and policy makers in Africa. In fact, defeated. anybodyconcerned withthe future development of theAfrican This reportis an African input into the debate onproblems continent should read it. and prospects in Africa. It is an important reading material Samuel0. Akatch Departmentof Urban and Regional Planning University of Nairobi P.O. Box 30197 Nairobi, Kenya

282 Volume 11 Nos. 2 and 3, May andAugust, 1996: P. 283 Supply ofLivestock Products to RapidlyExpanding Urban Populations

SUPPLY OFLIVESTOCK PRODUCTS TO RAPIDLYEXPANDING URBAN POPULATIONS editedby R. TrevorWilson (Rome: Food and AgricultureOrganisation, FAO/WAAD/KSAS, 1995, 233 pp, soft bound,price not given)

The bookis divided into four sections, partone consistingof 20 May, 1995. The pre-conference version of these four keynote papers that form the theme of the book, part proceedings was edited by W. Y. Kim and J. K. Ha and made two deals with the four conceptual areas that constitute the availableto participants for the conference. basisof livestock developmentsystems within any set-up in All in all, the chapters are simple, well-written and theworld, part three comprisesnine case studiesthat address relevantto the theme of the book. In addition, they give the selectedagricultural/livestock production systemsin certain readers,researchers and policymakers a very clear overview urban populations in Africa and Asia and part four covers of the livestockdevelopment systems and the challengesthat the participants' deliberations and recommendationsthat aresignificant to itspromotion in therapidly expanding pen- centredon three crucial areas of livestockproduction, namely, urban and urbanpopulations worldwide. policiesand strategies,marketing legislation and regulations, and constraintsand opportunities.Finally, there is an annex comprisingof 17 abstracts fromposters presented during the G. E. Otiang'a-Owiti Symposium entitledSupply of LivestockProducts to Rapidly Department of Veterinary Anatomy Expanding Urban Populations which took place at Hoam Universityof Nairobi Faculty Club, Seoul National University, Seoul, Korea, 16— P.O. Box 30197 Nairobi, Kenya

283 IDRC / CRDI

IIII 111111111111111111111111111111111 305809

Call for Papers Evolving Environmental Ideals Changing Ways of Life, Values and Design Practices lAPS 14 Conference Stockholm, July 30 — August 3 1996 Evolving Environmental Ideals, Changing Ways of Life, Values and Design Practices is a theme selected to promote a discussion between social, behavioural and architectural scientists, ecologists, designers, and planners. The conference should articulate the role of the built environment in ecologically conscious sustainable development. Other contemporary areas ofinterest in design researchcan be usefullyreexamined in terms of their ecologicalimplications for the built environment. Among such issues are: the growing diversity of ways of life in contemporarysocieties; the impact of migration within and between countries on the quality of housing, living and working conditions; the role of new informationtechnology in everyday life, and differences in ways of life, values and design ideals between "underdeveloped" and "overdeveloped" countries. Cultural and social responses to environmentalchanges constitute important issues for discussion. In some countries, anticonsumerist ways of life, in which sharing of resourcesis more importantthan individual levelsof consumption, have been fostered in various ways, including housing and other aspects of design. It may be asked whether these "postmaterialist"ways of life are gaining momentum, or waging a losing battle against other ideals, such as economic growth. Papers, in English or French, are welcome on any subject within the broad field of person—environment studies. Instructionsto authors will be sent to those who apply for more information. Prospectiveparticipants are asked to send 200—250 word abstractsby 31st October, 1995. People without papers are also welcome. The conference is beingjointly organised by the International Association for People— Environment Studies (lAPS), the Departmentof Architectureand Townplanning at the Royal Institute of Technology (KTH), Stockholm and the Nordic Association for Architectural Research (NA) as follows: A limited number of key note speakers will be invited to talk on the main subthemes of the conference. The formal part of the conference will end with a panel discussion. Participants will be assigned to groups for intensive sessions accordingto the theme of their papers. Papers will be sent in advanceto all participants in the same group. Groups of authors may submit papers for inclusion as a symposium on a specific subject in the frameworkof the conference. One author should act as coordinator.lAPS Networks (Housing, Landscape, Education, Spatial Analysis, Cultural Aspects of Design,Migration and Built Form) will meet. Excursions include field trips and sightseeing. Exhibitions, books, etc. related to the theme are welcome. The Secretariat should be informed in advance. The conference will be held at the Hasseludden Conference Centre, beautifully situated in the Baltic archipelagoof Stockholm.It can be reached by bus or by boat. In addition to full conference facilities, the centre offers recreation in attractive surroundings. Hasseludden offers full board. Accompanying people may enjoy the recreationalfacilites and the excursions. Furtherinformation is available from lAPS 14 Secretariat, Dept. of Architecture and Townplanning The Royal Institute of Technology, S-i 00 44 Stockholm, SWEDEN Tel +46 8 790 85 22, fax +46 8 790 85 80 (Dick Urban Vestbro) E-mail:[email protected]. Tel and fax +46 8 643 11 73 (Madi Gray) GUIDELINES FOR AUTHORSAND EDITORIAL POLICY

Authors writing for African Urban Quarterly (AUQ)must Editor's written permission. There is no strict limitationon aim tocontribute knowledge and discussionof urbanisation length of manuscript. and planning issues of developingcountries of the world. Authors should give special attention to the problem of 6. Four copies of manuscripts shouldbe submitted, typed with communicationwith a wider audience; the use of jargon lines double-spacedand with wide left-handmargins. shouldbe avoidedat all costs. 7. Whenever references are to be given for additional 2. The Editor welcomes the submissionof papers on topics information,insert the last name of theauthor and theyear as relatingto all aspects ofurban, regional and rural planningin shownbelow. All references mustinclude, whenever relevant, developingcountries, especially in Africa, but papers on the the name of the author(s), year, title of work,publishers and followingkey topics will be particularlywelcome: housing; placeof publication as indicated in the examplegiven below: theuse and developmentof energyresources; transport and book—Obudho, R.A. (1983) Urbanization in Kenya.London: communications; employment and labourmarkets; marketing University Pressof America; Article—Peil, M. (1976)"Urban and service provision in urban and rural development; SquatterSettlement: A ComparativeStudy", Urban Studies demographic change;surveying and planningtechniques; the Vol. 13 No. 2: pp. 155—166; Chapter in books—Obudho, R. growth of large metropolitanareas; communityhealth and A. and G. 0. Aduwo, "Small Urban Centres and Spatial planning; the administrationof planning and development; Planningin Kenya" in Jonathan Baker (ed) Small Town in technical cooperation and planning education and the Africa Uppsala: ScandinavianInstitute of African Studies, relationships betweenphysical and socioeconomic planning. 1990, pp. 51—68; and Conferenceand Seminar Papers— Publicationswill be inEnglish. Obudho, R. A. (1991) The Role of Megacities and Spatial Planning paper presented at the Annual Meetingof Kenya 3. The study of urbanisation and planning, in relation to the Geographical Association, Nairobi,Kenya, February 14—20, quality of human life, deals with aspects of medicine, 1990. agriculture, demography, transportation, political science, geography, history, sociology, economics, statistics, 8. Appendices should follow immediately after the main text; mathematics, urbanisation, anthropology, archaeology, any notes and references should be placed next. education,law and environmentalstudies; and they must be Acknowledgements should followthe notes, references,tables broughtto bear on the developmentof human life on rural as and charts. well asurban areas.AUQ welcomes manuscripts which focus onany of theproblem areas just mentionedprovided that the 9. Tables shouldbe sequentiall numberedin the orderin which link betweenthe subject treated and the central problemof they arereferred to in thetext and should be includedin the the comparativeurbanisation and planning in Africa within text.Each table should havea brief descriptive title and source, the rest of the world is reasonablyclear. In particular, we both of whichare writtenin lower case. shall be happy to have manuscriptsthat are: (1) reports of empirical investigations; (2) studies of methodological 10. Illustrations should accompanythe manuscript. All maps, problems;(3) studiesof realworld situationsin Africa alone diagrams,charts and photographs should be referred to as. orcompared tothe rest ofthe world; (4)philosophical analyses 'Figures' and should be numbered sequentially in the order ofbasic urbanisationand planningtheories; (5) prescriptions inwhich theyare referredto in thearticle. All figures should of policy research and practice, and (6) reviews of current have brief descriptive captions and sources also should be research,findings, information and interpretations of general cited. All illustrationsmust be camera ready. Any extra work interdisciplinaryinterest to Africanstudies. Ingeneral, AUQ on illustrations will be chargedto the author. encourages and publishes quality papers to reformulate traditionalideas within the contxt of quantative and 11. Photographs shouldbe goodquality glossyprints, preferably behavioural research. AUQ IS A HEAVILY REFEREED whole plate (8.5 x 6.5 inches). Each should have lightly JOURNAL. written in pencilon theback, the authors name, figure number and an indicationof the top of the picture. Line drawings 4. Editorialcorrespondence should be addressedto: The Editor, should be in black and should be in a form suitable for African Urban Quarterly,P.O. Box 74165, Nairobi, Kenya, reproduction. The AUQ reserves the right to redraw East Africa. Telephone: +254-2-216574or +254-2-334244 illustrations. extension2086; Telex: 22095 VARSITY NAIROBI. Cable: VARSITY NAIROBI. 12. If requested,AUQ will return all illustrationsto authors, but cannot be responsible for any damage incurred in plate- 5. The acceptance of an article for the review implies an making. undertaking by the author(s) that the article will not be submitted to other journals and that, once published,it will 13. Authorswill receive two free copies of the issues in which not be printed in whole or part, in any form, without the theirarticles appearas well asoffprints copies free ofcharge. Published by UQL AFRICA N URBAN QUARTERLY LTD. PUBLISHEROF ACADEMIC BOOKS ANDJOURNALS WORLDWIDE C

Withfinancial supportfrom

Technical Centre for Agricultural and Rural Co-operation ACP-EU (CTA), Wageningen, The Netherlands

The French Institutefor Research in Africa (IFRA), Nairobi, Kenya

InternationalDevelopment Research Centre (IDRC), Ottawa, Canada

Distributed by SUBA Books and Periodicals Distributors Ltd. P.O. Box 51336,Nairobi, Kenya, East Africa Telephone: +254-2-449231; Fax: +254-2.444110

Printed by: ICIPE SciencePress P.O. Box 72913, Nairobi, Kenya S Tel: +254-2-861680-4 S Fax: +254-2-803360/860110 Email: [email protected]