Living in a Walking World: Rural Mobility and Social

Equity Issues in Sub-Saharan Africa

GINA PORTER *University of Durham, Durham, UK

Summary. — Accessibility and mobility are embedded in the development nexus in far-reachingways. Field studies of mobility among women and men in rural settlements with poor road accessillustrate the frustrations and costs of living off-road. They are frequently marginalized andinvisible, even to local administrations. State decentralization appears to have had little positiveimpact in reducing ‘‘tarmac bias’’ and improving rural service delivery. A range of potentialinterventions, from Intermediate Means of Transport to electronic communications is reviewed,and opportunities for building social capital in off-road areas through nurturing improvements instate–civil society relations are considered. � 2002 Elsevier Science Ltd. All rights reserved.

Key words — accessibility, mobility, off-road, decentralization, Africa, services

1. INTRODUCTION

The title of this paper originates in the fore-word to a World Bank publication, ‘‘Transportand the village’’ (Barwell, 1996), where KevinCleaver reiterates a fact which, on the one handis glaringly obvious to development practitio-ners, yet at the same time is so often side-stepped in development research and planning:‘‘the African farmer largely inhabits a walkingworld.’’ My particular focus is on women andmen for whom this observation is probablymost apt, because walking is often the onlymode of travel available: those who live in ‘‘off-road’’ rural settlements.‘‘Off-road’’ has developed new connotations

in Western material culture and in Westerncultural criticism, where vocabularies of travelcurrently proliferate (for example, Wolff, 1995).In this context, the term evokes a peculiarlymasculine fantasy construction of wildernessexperience, particularly when linked to use ofthe four-wheel drive vehicle (Bishop, 1996).‘‘Off-road’’ has much starker, decidedly ‘‘un-playful,’’ connotations in a sub-Saharan Afri-can context, where the advantages whichinaccessibility may confer are rarely appreci-ated. It is defined in this paper as areas—and,more specifically, settlements—away from agood gravelled or paved road which, for at leastpart of each year, are inaccessible or accessibleonly with difficulty by motorized transport.Problems associated with poor access may be

experienced not only in very remote settlementsbut even in villages only three or four kilome-ters from a good road and close to urban areas.Although the inhabitants of such settlementsare often markedly poorer than those residentin comparable roadside locations in the sameregion, and the settlements are almost inevita-bly disadvantaged in the allocation of publicservices, they have received remarkably limitedspecific attention in the development litera-ture. 1

In a recent review of transportation studies inthe Western world, Law (1999) urges re-searchers to consider the practices and (gen-dered) meanings of mobility, while maintainingconcern for injustices in access. In a similarvein, in this paper I want to extend discussionbeyond the immediate mechanics of moving

World Development Vol. 30, No. 2, pp. 285–300, 2002� 2002 Elsevier Science Ltd. All rights reserved

Printed in Great Britain0305-750X/02/$ - see front matter

PII: S0305-750X(01)00106-1www.elsevier.com/locate/worlddev

* I am grateful for the assistance of Frank AcheampongOwusu and to the off-road villagers who spent so much

time with us on this study. The research was funded by

the Department for International Development of the

United Kingdom (DFID Crop Post Harvest Pro-

gramme, Projects R7149, R7575). However, the De-

partment for International Development can accept no

responsibility for any information provided or views

expressed. Earlier versions of this paper was presented at

the UK African Studies Association Annual Conference,

Cambridge, September 2000, and the Association of

American Geographers Annual Meeting, New York,

March 2001. Final revision accepted: 1 September 2001.

285

people and goods, which has been the focus ofmost transport research conducted in low in-come countries, to a rather broader consider-ation of the lived experiences of women andmen in off-road villages. These I consider notonly in immediate practical terms of obtaining(or trying to obtain) medical attention, marketoutlets and credit, but also in terms of thepower relationships between roadside and off-road settlements, with specific reference to ac-cess to the political process in the context ofdecentralization policies. This focus is essential,since we cannot explain people’s travel experi-ences in off-road areas without reference to theconfigurations of power which mediate rela-tionships between individuals and institutions. Iconclude with a review of the prospects forimproving off-road conditions, including thepotential role of telecommunications, and ofother interventions aimed at encouraging groupactivity and building social capital. The dis-cussion is based principally on recently com-pleted qualitative research in coastal Ghana,which included detailed studies in five off-roadsettlements (conducted over a full year in orderto assess seasonal change), but builds on myearlier work on rural transport and trade inNigeria and on a broad spectrum of relateddevelopment literature.

2. BACKGROUND: RURAL TRANSPORTCONDITIONS

Rural transport conditions in sub-SaharanAfrica are remarkably poor by comparisonwith Asia and Latin America. There is a sub-stantial literature which highlights the lowquality of the rural road network in Africa, thefact that rural roads (paved and unpaved) aredefective not just due to poor design and con-struction but, above all, due to lack of propermaintenance and that this greatly increasesvehicle operating costs (Platteau, 1996). Ruraltransport charges are higher than in any otherregion in the world. Ellis and Hine (1998)compare villages in Zimbabwe, for instance,which have an average of one motorized vehicleper 300 people, with Sri Lanka, where the levelis five times as great. They report transportcharges for journeys of up to 30 km as up totwo and a half times more expensive in Africathan Asia.Investment in transport infrastructure in

Africa tended to come late, generally in thecolonial period, when most countries experi-

enced their first major phase of rail and roadconstruction. This produced a dendritic system,firmly directed to the evacuation of primaryproduce to coastal ports for onward shipmentto Europe (Taffe, Morrill, & Gould, 1963). Infavored export production zones, such asGhana’s cocoa areas, motorized transport ex-panded rapidly from the mid-1920s, eventuallyproducing nodes of high accessibility in someareas by the late 1950s, which assisted move-ment of goods from rural areas to major bul-king points (Gould, 1960, p. 110). Across ruralsub-Saharan Africa, however, head loading—mostly by women—has continued as the dom-inant mode of transporting goods from and torural households (Barwell, Edmonds, Howe, &De Veen, 1985; Dawson & Barwell, 1993;Doran, 1990; McCall, 1985; Moore, 1979).Economic decline and subsequent imposition

of structural adjustment programs tended toexacerbate strains on existing inadequate roadtransport systems in the 1980s and early 1990s.There was a deterioration in road constructionstandards and road maintenance, in purchasinglevels of new vehicles and in vehicle mainte-nance (Levy & Malone, 1988; Riverson et al.,1991; Riverson & Carapetis, 1991). In Nigeria,where there had been an enormous expansionin road construction and vehicle ownership inthe 1970s, fuelled by the oil boom, the impactof recession and the imposition of a structuraladjustment program on road transport condi-tions was particularly devastating (Filani, 1993;Porter, 1997). Guyer (1997, pp. 86–91) providesa fascinating, closely observed view of boomand decline and their transport implications inthe Ibadan area, where the Datsun pickup be-came a ‘‘social institution’’ in the halcyon yearsof the oil boom. The impact of adjustmentpolicies in Nigeria was generally to reinforcefurther the transport focus on roads still ingood condition as a result of investment in thepreceding oil boom and to reduce serviceselsewhere.In Ghana, where there had been severe re-

cession for some years prior to the impositionof structural adjustment policies in the early1980s, road conditions were very poor bycomparison with Nigeria. Clark (1994, p. 66),suggests that in the Kumasi region ‘‘many vil-lages fell off the transport map entirely’’: a vividmetaphor for the loss of access which occurred.In 1988 only 28% of Ghana’s paved roads werein good condition, compared to 67% in Nigeria,where the impact of the oil-financed road con-struction program was still evident (World

WORLD DEVELOPMENT286

Bank, 1994). In some other countries such asUganda and Mozambique conditions weresubstantially worse again, with only 10% and12%, respectively, of paved roads in goodcondition.Subsequently, there has been more emphasis

on road maintenance in sub-Saharan Africa(encouraged by the World Bank SSATP RoadMaintenance Initiative). Labor-based roadconstruction and maintenance programs arenow strongly favored by donors (Plumbe,Bascou, Etounga-Manguelle, & Maiga, 1995;Stock & de Veen, 1996). While some improve-ment on major routes has thus occurred,funding has tended to remain inadequate toattend to the majority of the network. Off thepaved road, access is usually even more diffi-cult. Bryceson (1999) concludes that, whilemarket liberalization eventually led to majordonor-financed road rehabilitation programs insub-Saharan Africa, ‘‘liberalisation has tendedto increase the transport service gap betweenon-road and off-road villages.’’Some of the most detailed research con-

ducted on transport issues in Africa has takenplace in Ghana (see Bryceson & Howe, 1993;Doran, 1990; Grieco, Apt, & Turner, 1996;Hine, 1993). Despite the considerable effortthat has gone into road rehabilitation programsin Ghana in recent years, road condition stud-ies indicate that there has been limited overallimprovement, with only 29% of Ghana’s pavedroads and 10% of gravel roads in good condi-tion in 1997 (Wilbur Smith Associates, 1998).So far as feeder roads are concerned, govern-ment policy in Ghana now emphasizes main-tenance of the so-called maintainable network:some of the ‘‘nonmaintainable roads’’—whichrequire reconstruction—are consequently fall-ing into disuse. Feeder roads are maintained bycontractors using labor-based techniques whichreduces foreign exchange costs (Ashong, 1996).In off-road settlements in areas such as

Ghana’s Central Region, the roads and pathswithin and close to the village are generallymaintained by residents: women carry sand,gravel and stones to the potholes, men fill andlevel them, using their own shovels and pickaxes. When the road needs attention—whichmay be as often as once per week during heavyrains—the gong-gong is beaten and every able-bodied inhabitant, including the Chief andolder children, is expected to participate. Finesare often imposed on those who do not assist.The more important roads between settlementsare usually the responsibility of the Ministry of

Roads and Highways’ Feeder Roads Depart-ment: a common complaint is that ‘‘they justweed along the roadside—not much else.’’Many of these roads are shown on maps in usein government offices in Accra as ‘‘gravel;’’ostensibly a class of road quality below tarmacbut higher than earth roads. This is entirelymisleading since many of the so-called gravelroads are indistinguishable from dirt tracks. Inthe rolling topography of Central Region, forexample, the 4–6 in. gravel surface is rapidlylost during the rains and gulleys appear. Afterfour years the surface may have totally disap-peared. Graders are used to improve conditionson these roads occasionally but, in 1999, of thefive graders operated by Feeder Roads De-partment in Central Region, four were broken.The administrative decentralization programcurrently in progress in Ghana (discussed be-low) includes the Department of Feeder Roads,which has been piloting decentralization ofroad maintenance responsibilities to local gov-ernments (Malmberg-Calvo, 1998). But short-ages of staff including skilled engineers andsurveyors make for logistical difficulties.Ghana, like most sub-Saharan countries,

imports large numbers of second- hand vehiclesfrom Europe. These are often old models forwhich it is difficult to find spare parts (whichare taxed): the popular slogan ‘‘no condition ispermanent,’’ gaudily displayed on so manyWest African vehicles, has very ironic conno-tations for travellers whose journeys are regu-larly disrupted by vehicle delays andbreakdowns. The particularly rickety yellowtaxis commonly used to ply off-road routes arejokingly referred to as ‘‘tetanus cars.’’ Evenobtaining parts for new models can take con-siderable time, since agents cannot afford tokeep large numbers of spare parts in stock. Thevast majority of motorized vehicles in ruralareas are based in settlements along the pavedroads and there is widespread reluctanceamong vehicle owners to take their vehicles onunpaved roads unless the rewards are high.Relatively low densities of demand for trans-port off-road—in part arguably the result ofpoverty induced by poor access—tend to mili-tate against this.Transport charges are high in Ghana, as

elsewhere in sub-Saharan Africa, but off-roadcharges are particularly exorbitant. Interviewswith transporters in Ghana’s Central Regionsuggest it is common to charge approximatelydouble for journeys over bad roads, sincein these conditions vehicles deteriorate more

LIVING IN A WALKING WORLD 287

rapidly. In 1998–99 the standard charge by tro-tro (minibus) was around 100 cedis per pas-senger per unpaved road mile, compared toaround 50 cedis per paved road mile, over adistance of about 10 miles. (Ellis & Hine, 1998,present figures for paved roads & poor qualityearth roads in Zambia which suggest differencesof even greater magnitude). Not surprisingly,Anyinam (1994) reported that agriculturalprices in some remote rural areas of Ghanahave dropped because of increased transportcosts associated with road deterioration. Ac-cording to Beenhacker (1987) tyre life is re-duced 25–50% of normal life under poor roadconditions: when six new truck tyres costaround £300 (Ghana, May 1998) such consid-erations are important. Finding spare parts is adifficult and time- consuming business, too.They are generally available, but only in themajor urban centers, in Ghana. Many arecounterfeit copies and soon break.

3. LIVING OFF-ROAD

Getting to market to sell produce, getting toschool, obtaining medical attention, findingemployment, buying spare parts, farm inputsand consumer items not available locally, try-ing to arrange a loan at the bank—these can bedifficult tasks for the rural poor in general, butfor the residents of off-road settlements thehurdles to be crossed are additionally complex.Recounting a study of 36 communities acrossNigeria, Francis et al. (1996, p. 8) note thatrural populations describe poverty in terms ofaccess to physical infrastructure (notably roadsand heath centres) rather than services per se,and similar perceptions were encountered inGhana, but the implication of infrastructure isperceived to be service provision. Across Af-rica, banks, clinics, hospitals, secondaryschools, extension agents, produce markets andthe like all tend to be located in larger ruralservice centres, usually situated on paved roads.For women, the financial and time con-straints—and, in some cases, the cultural con-straints—on mobility (and thus on access tothese facilities), can be particularly restrictive.A review of three aspects of access: to medicalcare, markets and credit facilities, illustrates thedifficulties, frustrations and costs of off-roadresidence. All have far-reaching developmentalimplications, not least for agricultural produc-tivity (Porter, 1997): all contribute to a vicious

circle which impacts most severely on off-roadinhabitants.

(a) Access to health care

I commence this review of access problemswith health care because this is usually the firstissue to which both men and women draw at-tention when asked about their access problemsand priorities in off-road settlements. Duringthe course of my research in Ghana, I heardmany sad stories of villagers, particularly chil-dren, who failed to reach hospital alive, despitethe best efforts of the community to get themthere. Like most travellers driving in remoterareas of rural Africa, I am particularly aware ofthis problem, since I have often been waiveddown by strangers needing to take sick peopleto hospital.Villagers emphasize the need for access to

health care in emergencies, but health facilitiesof any kind are rare in off-road locations.Meagher (1999), for example, describes a sub-stantial settlement of over 12,000 people inMoslem northern Nigeria, about five kilome-ters from a main road, where there is a clinicwith no medicines and not even one trainedfemale health worker. In Morocco, places withbetter rural roads reportedly have twice the useof health care facilities (World Bank, 2000,p. 78). Ironically, however, off-road inhabitantsare often the most in need of medical assis-tance, since water supplies are frequently poorand poverty levels above regional averages. Myresearch in Ghana and a recent Uganda Par-ticipatory Poverty Assessment (cited in Booth,Hanmer, & Lovell, 2000) both indicate thatvaccination programs tend to miss off-roadsettlements, exposing these populations to fur-ther risk. To compound these problems, ill-health is often at its peak in rural populationsduring the wet season (related to high workloads and low food availability) at precisely thetime when travel conditions, even for pedestri-ans, are at their worst.At Sampa, in coastal Ghana, most water is

drawn from the River Odi, half a kilometerdistant: there is a (padlocked) standpipe but, at50 cedis per bucket of water, most peoplecontinue to depend on river water. Sampa ex-periences sporadic outbreaks of guinea worm,but the only health personnel in the settlementare two traditional birth attendants. Drugsellers occasionally visit the village by bicyclewith a very limited range of medicines. Thenearest clinic is seven miles away. If anyone is

WORLD DEVELOPMENT288

seriously ill the usual procedure is to walk tothe village at the road junction five miles awayand hire a taxi to return to the village to collectthe patient. It will cost 20,000 cedis to take thetaxi to hospital and, if the taxi is kept for thereturn journey, 30,000 cedis (over £10), thisbeing the cost of daily hire of a taxi.Such arrangements are the norm across

coastal Ghana and elsewhere in Africa, thoughin less densely populated areas the distances tohospital are often much greater and it may beimpossible to make the journey at all. Researchsuggests that visits to clinics for more routinereasons fall off rapidly after four to five miles(Howe & Richards, 1984, p. 15). As Airey(1992) concludes from a study of hospital uti-lization in Meru district, Kenya, high costs ofmedical treatment—exacerbated by the intro-duction of hospital user fees in many countriesas part of structural adjustment—are likely tobe an even greater barrier to treatment thandistance. But where high medical costs arecompounded by high travel costs the barriers totreatment must inevitably be even more pro-nounced.

(b) Access to traders and markets

Produce traders are often unwilling to moveinto rural areas with poor access, to purchasegoods, unless supplies from accessible areas areinsufficient (Porter, 1995, 1997). Moreover,when urban-based traders visit areas whereaccess is difficult, the competition from othertraders is likely to be limited and pricesachieved by local farmers will thus be poor,especially for perishable produce. The bar-gaining power of visiting traders will also bestrengthened if, as appears to be the case insome regions, off-road farmers have lessknowledge of supply and price conditions inmajor bulking markets (Lyon, 2000).Not surprisingly, many off-road farmers

prefer to send produce for sale in major mar-kets usually located on paved roads. This hasparticular impact on women in sub-SaharanAfrica since, in much of the continent, they arethe principal marketers and porters of agricul-tural produce. When it is necessary to takeproduce to market, women who live off-roadfrequently face arduous treks. This is a problemwhich has certainly grown in recent years inGhana (and probably elsewhere in Africa,though I have only anecdotal evidence), as aresult of the decline of off-road rural periodicmarkets. My research in northern Nigeria,

central Nigeria and coastal Ghana, and inci-dental information in other studies (for exam-ple, Filani & Richards, 1976), suggests thatthere has been a major reorganization and ra-tionalization of rural market systems associatedwith the expansion of motorized transport sinceindependence. This has led to the decline andclosure of many small markets with poor access(Porter, 1988, 1995). Indeed, market declineand closure in off-road locations appears fre-quently to be one of the unforeseen effects ofroad construction programmes. In some partsof sub-Saharan Africa, such as Moslem Borno(northeast Nigeria), where there are culturalconstraints which restrict women’s movements,in addition to physical and economic con-straints, off-road market closures followingroad construction have had a particularly se-vere impact on women living off-road.Off-road journeys to market by motor vehi-

cle—when vehicles are available—tend to bemore expensive, because of the increased costsof maintenance. In coastal Ghana journeys offthe paved road are roughly double those alongpaved routes. Women are less likely than mento be able to afford transport, even if vehiclespass near their village. Thus, while it is possi-ble, for example, to get from Sampa to itsnearest market center by vehicle in under onehour, many women trek for three hours withmaize, cassava and tomatoes and then returnhome carrying fish and groceries to sell in thevillage. Men are supposed to give their wivestransport money but do not necessarily do soand, if they do not, the wives must spend theirown money, or walk. About four or five tro-tros and taxis come to Sampa to pick people upon market days in the main agricultural season(there is no transport based in the village), butsometimes only one or two vehicles turn up. Inthe minor season there was no transport fromSampa to the main road in 1999. AlthoughSampa is a relatively wealthy off-road settle-ment (by comparison with some of the othersurvey villages) only two inhabitants—bothmen—own bicycles.At another study village, Adabra, women in

the first few months of the research similarlytalked about the difficulties of getting produceto the main market at Kasoa, 15 miles away.Adabra had once had its own market, but themarket collapsed about 25 years ago, report-edly due to deterioration in local road condi-tions. No transport was based in the village andfrequently no vehicle whatsoever arrived in thesettlement on market day. Sometimes, after

LIVING IN A WALKING WORLD 289

heavy rains, the road was completely washedaway and it was too dangerous for anyone totravel. If vehicles arrived, they were often al-ready full. The women interviewed here em-phasized that men are richer and ‘‘can affordmore transport.’’ In the dry season womengenerally cannot afford the fare to Kasoa andhave to walk. (In this settlement, later in theseason, the accessibility problem diminishedremarkably, albeit temporarily, as a result ofthe grading of the main road and a consequentincrease in passing traffic.)Interviews with women traders in Ghana

emphasized the costs of late arrival at market.Delays in getting to market occasioned bytransport unreliability can have serious impli-cations for off-road women, since dealers in themarkets may have already purchased sufficientproduce from better located local women whoarrived earlier. Delays resulting in failure to finda buyer at market, or total failure to get tomarket, can also result in major losses throughspoilage. It is not only fruit and vegetableswhich are affected. Fresh cassava, for example,changes color around four days after harvestingand, if it has not been possible to take sell itimmediately at market, it then has to be driedand sold at a lower price. Hine (1984), reportingwork conducted in the Ashanti region ofGhana, observes that ‘‘little evidence was foundto suggest that produce was lost because ofimpassable roads,’’ and this may be the case inthe Central Region study villages (though somelocal farmers indicate otherwise), but it is im-portant to also consider the reduction in marketprices received due to deterioration in producequality and to late arrival at market. 2 There areother benefits from market visits too, for it isusually possible to purchase items unavailablein the village and transport them home for re-sale. In settlements such as Adabra, which oncehad a market, the difficulties of market accessseem to be most keenly felt. Old men andwomen here reminisce about the days when theyhad a big market and ‘‘everyone came. . . fromAccra, Mankessim, Fettah. . .’’ The costs ofhead loading heavy baskets of maize and otherproduce to market are born by women not justin terms of time unavailable for other tasks orleisure, but also in terms of physical injury (seeDoran, 1990, p. 58; Steele, 1993).

(c) Credit and rural banks

Access to capital and credit is a vexed prob-lem for rural communities across Africa. The

significance of credit availability and costs ofinformal credit for farmers and others is atheme taken up in much recent literature. InGhana, Sarris and Shams (1991, p. 131), andGrieco et al. (1996, p. 120) citing Steel andAryeetey, suggest interest rates for informalcredit as high as 100% over 9–12 months. Dif-ficulties for women in obtaining formal creditare particularly great, partly because of theirlow social status, partly because they lack col-lateral. Grieco et al. (1996, p. 33), make theimportant point that lack of access to capitalamong women traders has consequences for thesize of load generally transported and the fre-quency of trips made.In remote and off-road areas, however, the

problems associated with obtaining credit seemoften even more unsurmountable than in moreaccessible locations. This has implications forproduction, marketing and, of course, ability topurchase vehicles. Some studies have observedthat farmers in remote areas of Ghana haveparticular difficulties in obtaining formal bankcredit for setting up storage and processingactivities (Creightney, 1993; Hine & Riverson,1982). As Richards (1985, p. 127) pointed outin the context of rural Sierra Leone, this ishardly surprising since loan recovery is moredifficult from distant creditors and the costs intransport and staff time may exceed the cost ofthe loan. This is a particularly pertinent pointwith reference to creditors in off-road villages.In Ghana, the rural banks established (in the

mid-1970s) in response to the need and demandto make institutional credit and banking ser-vices accessible to small farmers and othersmall-scale rural entrepreneurs, might seem apotential source of credit for off-road dwellers.They are community institutions, locally-owned, controlled and managed by the peopleof the locality, mobilizing resources from thecatchment area (c.20 miles radius) and on-lending to customers in the same area. Thebanks have been required to give a minimum of50% of their loans and advances to agricultureand 30% to cottage industries. But, since thedecision in 1990 to pay salaried workersthrough the banking system, the banks’ ad-vances go mainly to workers whose salariespass through the bank, because agreed fixedmonthly deductions can be made from theirsalaries (Nikoi, 1996).Few permanent off-road residents have jobs in

the formal sector and thus have very little op-portunity to access credit from the banks. Forthe majority of rural dwellers in the informal

WORLD DEVELOPMENT290

sector, loans and gifts from friends and relativesare a crucial means of surviving hard times. Inoff-road villages in Ghana’s Central Region,such sources tend to be few, and the sumsavailable small, since fellow residents are oftenequally impoverished, particularly as incomesources tend to be very similar in off road areas.There are often fewer opportunties for partici-pation in off-farm activities in such locations(Madulu, 1998) and consequently most villagersare likely to be dependent on (the same kind of)agricultural produce. Meagher (1999), writingabout an off-road village in northern Nigeria,suggests that while rising transport costs understructural adjustment have reduced contactwith urban relatives, demands for assistance arenow concentrated on rural relatives, often littlebetter off than those seeking financial help. Forthe most part, in the five villages I studied incoastal Ghana, the credit granted and receivedamounted to minuscule sums. Elderly cooked-food sellers talked of having to wait a wholemonth to receive payment from other villagewomen for tiny quantities of food. They them-selves purchased corn on credit and even rentedbowls to make food, because they had insuffi-cient funds to purchase their own.Even better-off off-road residents can en-

counter difficulties in their business transac-tions due to their location. In Assin Aworabo,the richest of the five villages I studied, andhome to a few relatively large cocoa farmers,the two cocoa buying agents in the village payfarmers by check. The farmers must take theirchecks to the bank at the district headquarters(on the paved road) 25 km away to cash them.Sometimes, having found and paid for trans-port, farmers say they arrive to discover thatthe bank is without cash, or the cashiers maytell them that the cash they are holding is ‘‘for[formal sector] workers, not cocoa farmers.’’ Inthat case, the farmers have to rely on credit inthe village—usually no one has ready cash—until the bank has replenished its reserves.

(d) Gendered patterns of mobility amongoff-road villagers

The foregoing discussion has suggested, atvarious points, the particular disadvantagesfaced by women resident in off-road settlements.There is growing evidence of the enormity ofwomen’s transport burdens in sub- SaharanAfrica (Bryceson & Howe, 1993; Doran, 1990;Malmberg-Calvo, 1994a; Urasa, 1990). Theneed to learn more about gender issues recently

prompted the commissioning of a series of 31case studies by the International Forum forRural Transport and Development across Asiaand Africa (Fernando & Porter, forthcoming).While it is true that women often cannot af-

ford transport, even when they live in settle-ments with good road access, the burden facedby women resident off-road is additionallyheavy because of the roles they are commonlyexpected to play in sub-Saharan Africa as por-ters and (particularly inWest Africa) as producetraders. In addition to their many householdduties, women are widely required to transporttheir husbands’ produce (as well as their owngoods) to the nearest good road, or directly tobulking markets (which are now mostly situatedon paved roads). In coastal Ghana (and else-where) they may also have to travel to richerroadside settlements for other household taskssuch as taking maize to the grinders for foodpreparation, since poorer off-road settlementsoften do not have their own grinding machines.Kaur (2000, cited in Booth et al., 2000, p. 70),provides a rare on-road/off-road comparison ofwomen’s income which illustrates this point.Women in a village on a main road in Camer-oon were found to be able to make an averageincome over double that earned by women in anisolated village 90 minutes travel from the road,because of the greater time they had available toproduce food to sell.In common with women resident in roadside

locations, off-road women have less funds thanmen to pay fares when transport is availablebut, additionally, as discussed above, the off-road fares tend to be substantially higher thanthose for journeys on paved roads. Given theirlow purchasing power, sometimes coupled withcultural constraints, off-road women are alsofar less likely to own a bicycle, animal-drawncart or hand cart than their menfolk (Bryceson& Howe, 1993), an issue discussed in greaterdetail below. Thus, while this paper points tooff-road disadvantage in general, the genderednature of disadvantage requires specific ac-knowledgement.

4. DECENTRALIZATION AND THEPOLITICS OF OFF-ROAD RESIDENCE

To live off-road is to be invisible: this is thestrongest message which has emerged from myrecent work in Ghana and earlier work inNigeria. As villagers in one small lagoon-sidesettlement put it, ‘‘we are part of Keta district

LIVING IN A WALKING WORLD 291

but are forgotten.’’ Lack of a good road andother facilities is not infrequently blamed byinhabitants of off-road settlements on theirpowerlessness to influence decision-making bypolicy makers. In countries across sub-SaharanAfrica, personal contact is often an essentialcomponent in the decision-making process. Off-road villagers argue that it is particularly diffi-cult for them to maintain the intensive lobbyingat government offices required to obtain facili-ties such as clinics and schools and, indeed, toget their road improved (Porter, 1997). Theremay be other factors at work too: off-roadresidents are sometimes viewed condescend-ingly as ‘‘bush people’’ by local professionals.This is well illustrated by the comments of avillage headmaster resident in one Ghanaianoff-road settlement who, only half in jest, be-moaned his position: ‘‘It’s like a doctor livingwith the lepers. . .. I’ve been here six years. . ..I’m becoming like them.’’The link between political influence and road

construction is an interesting one. Political in-terests have had a strong influence on roadalignments across much of sub-Saharan Africa,and Ghana is no exception. In 1960, shortlyafter Ghanaian independence, Gould asksrhetorically, ‘‘If an area is known as a strong-hold of the Opposition, will it get funds fortarring its roads? And those roads that swingaround the villages of Opposition chiefs, was itsimply a matter of terrain and drainage?’’ Thesituation has changed little (Nugent, 2001, p.409). Clearly, roads cannot be built everywhere,but it is important to acknowledge the politicalfactors which influence both road constructionand road maintenance programs. Prioritizationof road improvement in Ghana, according to asenior roads officer, is shaped by reference tothree factors: (a) areas of high agriculturalproduction, (b) construction of ‘‘missing links,’’to ease regional road networks and (c) politics,‘‘since politicians make the final decisions.’’ Inthis context the implications of the currenttrend toward decentralization in sub-SaharanAfrica need consideration.Administrative decentralization has been

promoted by the World Bank as an element ofgood governance in recent structural adjust-ment programs and, in theory, might be ex-pected to improve the lot of less accessible areas,including off-road settlements and their inhab-itants, since it brings government closer to thepeople. But in Ghana, where a decentralizationprogram has been in place since 1988, the evi-dence is not, as yet, encouraging. The District

Assemblies themselves have lacked sufficientfunds to undertake substantial developmentprojects and, where projects eventually emerge,their location may well more strongly reflectlocal power relations than relative need. On thebasis of a detailed study of two Ghanaian dis-tricts, Ayee (1996) suggests, ‘‘decentralisationmay help to augment the dominance of thosewho, because of wealth or status, are alreadypowerful at the local level.’’ Kyei (1999), in arecent study of decentralization in two Ghana-ian districts, one in Ashanti and one in UpperWest, made a comparison of service provisionin on-road and off-road settlements and foundsignificant differences. District Assembly mem-bers are required to meet the people of theirelectoral area once a month, but as both Ayeeand Kyei have observed, many are employedoutside the district altogether and cannot affordtransport fares to their electoral areas: it isparticularly unlikely that they will venture intoareas where road access is difficult. 3

This conclusion is in line with experienceelsewhere: Samoff (1990) observes that ‘‘many,perhaps most, decentralization efforts appearnot to have improved local service delivery orthe general standard of living in rural areas.’’Litvack, Ahmad, and Bird (1998) suggest thatmarket solutions to service delivery need to beconsidered when rethinking decentralization.But given the relative poverty of most off-roadareas, these are unlikely to improve conditions.The World Bank 2000/2001 World Develop-ment Report, ‘‘Attacking poverty’’ (2000,p. 106–108), recognizes that decentralizationcan ‘‘bolster the power of elites in settings withhighly unequal power structures’’ and notesthat transferring power and resources to the‘‘sub-municipal level—such as neighbourhoodsor villages—requires special effort’’: but thespecific case of off-road settlements is not pur-sued. Ironically, while ‘‘off-road’’ vehicles con-tinue to proliferate in the regional capitals ofsub-Saharan Africa, government officials andothers are no keener now than they ever were totake their shiny landcruisers along the bushtracks which were the ostensible reason fortheir acquisition.

5. PROSPECTS FOR CHANGE

(a) Improving physical access

The funds for adequate feeder road con-struction and maintenance by government and

WORLD DEVELOPMENT292

other agencies to serve all current off-road areasin regions like coastal Ghana will not beavailable for the foreseeable future. It is clearlynecessary to explore alternative means bywhich the isolation and associated poverty ofoff-road dwellers can be reduced. The potentialfor improvement in roads, conventional trans-port services and the availability of IMTs areconsidered in turn.An extension of community participation in

road maintenance through unpaid labor con-tributions of the type currently found withinGhana’s coastal settlements could be one wayforward, though as Ostrom and others havepointed out, this approach should be pursuedwith caution, since it is difficult to ensure anequitable distribution of work among commu-nity members, and ‘‘free riders’’ are likely totake advantage of community efforts (Ostrom,Schroeder, & Wynne, 1993, pp. 77, 85). 4 Oneof the questions which needs to be explored isthe equity of current village-level road mainte-nance between men and women and richer andpoorer groups and attitudes within the villagesto extending community maintenance (partic-ularly small-scale spot improvements) to inter-village roads. In the former tin areas of the JosPlateau, Nigeria, in the early 1990s, off-roadvillages were attempting fairly extensive road-works, culvert repair and even bridge con-struction; here the constraint on extension ofsuch activities seemed to be merely absence ofsupport from the Local Government for morecomplex tasks. The fact that many of the Pla-teau routes were originally developed by pri-vate tin mining companies may partiallyexplain why roads there are not perceived assimply belonging to government. Recent re-search in northern Nigeria (Meagher, 1999;Porter, 2001) suggests that community roadwork has been crucial to maintaining accessover the 1990s. In coastal Ghana, however, thecommon perspective seems to be that intervil-lage roads belong to the government (and spe-cifically the Department of Feeder Roads) andit is likely that this inhibits community main-tenance efforts. (This view is common inGhana, according to Ministry of AgricultureVillage Infrastructure Project staff.) Similardifficulties have been reported in East Africa(Airey & Wattam, 1998). Malmberg-Calvo(1998) emphasizes the need to develop an in-stitutional framework for managing and fi-nancing the lowest level of the road/pathnetwork. She stresses stakeholder involvementand the need for a redefined public-private

partnership, whereby local governments ortheir agents manage the core rural roads andcommunities and farmers’ associations choosewhich roads and paths they will own. Definingownership is clearly crucial to improvements inroad and path management, but the process islikely to be both difficult and time-consumingin practice.Very few individual off-road inhabitants can

afford to buy conventional motor vehicles andwhen they buy vehicles (generally saloon carsused as taxis), evidence from coastal Ghanasuggests the vehicle is usually kept elsewhere, atthe paved road, because of local road condi-tions. There may be potential, however, forcommunity ownership of more robust (but ex-pensive) lorries and smaller four-wheel drivevehicles in settlements with basic motor access.There are few cases reported in the literatureand very few examples of local initiatives byoff-road villages have emerged from my fieldstudies in Ghana: just one community busservice operated by a church group for a shortperiod, and one vehicle operated by a familygroup. In the former case, once the bus re-quired major repairs it was rerouted to runfrom a paved road base, reportedly because thiswas the only way the villagers could recoup thecost of repairs. In the latter case (a village inGreater Accra Region) a satisfactory arrange-ment seems to operate whereby the chief runs apickup for general use by the village. But, this isessentially a one- family village. Most villageshave more complex family structures and theremay be the additional complicating factor ofrecent/temporary immigrant populations.For some off-road villages the capital cost

even of community vehicle ownership, whencompounded by very high maintenance costsalong poor access routes, would be too high.Nonetheless, the potential for cooperativetransport needs further research. Groups suchas off-road women marketers may obtain par-ticular benefits from cooperative arrangements.Across sub-Saharan Africa, the majority oftransport services are owned and operated bymen. In areas where women are major cropmarketers it would seem advantageous forwomen to obtain greater control of transport.Support for expanded ownership of transportby such women, whether as individuals or asgroups, through improved credit arrangements,could be important in providing transportservices more precisely tailored to women’sneeds. 5 An alternative or complementaryapproach could be to explore the potential for

LIVING IN A WALKING WORLD 293

subsidised off-road transport, with subsidiesprovided to transporters prepared to provide aregular service along routes to selected off-roadsettlements. The approach of targeting subsi-dies by route has been recommended by theWorld Bank (1996, cited by Booth et al., 2000,p. 81 ) in an urban context.Given the difficulties in obtaining access to

conventional transport in sub-Saharan Africa,programs to introduce cheaper, more sustain-able, intermediate transport technologies whichcan operate over suitable tracks seem to be themost obvious route to improving access in mostoff-road villages. Use of bicycles and other In-termediate Means of Transport (IMT) varieswidely across the continent, though commonlyownership and use is much higher among menthan women. In addition to a number of recentinitiatives by donors (including the WorldBank’s Village Infrastructure Project now un-der way in Ghana), nongovernmental organi-zations (NGOs) are now playing a small butsignificant role, in some regions of Africa, notonly through provision of IMTs, but alsothrough initiatives such as ‘‘Riders for Health’’which provides training in motorcycle mainte-nance and driving (Scholten, 1997). Motorcy-cles are often perceived as a particularly usefulIMT because they combine relatively highspeed with reasonably modest cost. In Nigeria,motorbike taxis have played a growing role inimproving access in rural areas in recent years(Porter, 2001; Yunusa, 1999). Barwell (1996),among others, points out, however, that IMTs,though cheap by comparison with a motor ve-hicle, are still expensive for poor farmers. Inrural Tanzania in the early 1990s the price of anew bicycle was the equivalent of almost ayear’s minimum wage (Bryceson & Howe,1993). Barwell (1996, p. 62), argues that there isneed ‘‘to establish rural credit systems for IMTwhich are more widely distributed, impose lessrigorous conditions and have more appropriateadministrative procedures than those whichtypically exist at present, and accept the IMT assecurity for the loan.’’In coastal Ghana bicycle and other IMT

ownership is not common, though it is muchhigher in roadside than off-road settlements. Inthe five off-road study villages only three menin total owned handcarts in the late 1990s.Between two and six men owned bicycles ineach village, but there was only one womanbicycle owner in all the villages (and the bicyclewas acquired through her job with an NGO). Afive-year old second-hand bicycle could be

obtained (at a roadside center) in this region, atthat time, for around 50,000 cedis (under £20).My research uncovered no cultural restrictionson women riding bicycles (by contrast withearlier fieldwork in northeast Nigeria) and mostof the women interviewed—especially youngerwomen—seemed receptive to the use of bicycletransport. Few women can afford to buy bicy-cles, however, and in the study villages only onehusband was prepared to loan his bicycle to hiswife. A similar reluctance among men to loantheir bicycles to wives was noted in a Ugandanstudy, where the reason given was fear ofdamage (Bryceson & Howe, 1993; also seeMalmberg-Calvo, 1994b, who reports a rangeof cultural, educational and economic con-straints which inhibited cycle use amongUgandan women). One of the factors whichmay also deter bicycle use in off-road settle-ments in coastal Ghana is that fact that re-pairers and bicycle repair equipment such aspuncture kits are usually only found at majorroadside settlements.A project in northern Ghana in the 1980s

which included IMTs, funded by the WorldBank, provides evidence of some of the diffi-culties IMT projects may encounter (Buabeng,Sarfo-Mensah, & Dennis, 1995; Salifu, 1994;White, Erlank, & Matthews, 2000). An ar-rangement was made with a local manufacturerto construct bicycle trailers, but these were tooheavy and needed redesigning. Women in-volved in the labor-based road constructioncomponent of the project were expected topurchase the trailers. Unfortunately, the pro-gram assumed prior bicycle ownership but fewpeople could afford to purchase both bicycleand trailer. Attempts at community trailerpurchase apparently failed because of argu-ments concerning maintenance.There may also be some resistance to IMTs

among policy makers. In Ghana’s CentralRegion there has been some pessimism aboutthe value of an IMT program among govern-ment staff, based on the argument that theterrain is unconducive, the heavy loads in cashcrop areas would be unsuitable for IMTs, and‘‘the culture of the people here may not evenpermit it’’ (interview: feeder roads engineer).Government officials in Accra similarlysuggested that bicycles are more appropriatefor the north (i.e., deprived areas) of Ghana,‘‘in the south people want to buy a car, bicyclesare generally a northern thing.’’ In discussionswith many men in the survey villages, there wasa clear view expressed that a conventional

WORLD DEVELOPMENT294

motor vehicle would be of far more use thanany sort of IMT. Further north, in AshantiRegion, bicycles are apparently looked downon as a mode of transport: according to oneurban Ashanti informant, ‘‘a bicycle makes youlower class.’’Because of the difficulties experienced in

many IMT projects, there is an ongoing debatein transport circles about how to stimulate IMTadoption (for instance, in an e-mail discussionon rural transport services organized by theInternational Forum for Rural Transport andDevelopment, October/November 2000). Astrong emphasis on the concept of ‘‘criticalmass’’ is now emerging. This stresses thattransport technologies of all types need a viablesupporting infrastructure, especially in terms ofrepair facilities, and recognizes that there areother components to critical mass, includingthe erosion of sociocultural inhibitions. Exam-ples such as the spread of ox-carts in northwestZambia (presented by Paul Starkey) appear toprovide excellent illustration of the relevance ofthis approach in a sub-Saharan context.

(b) Nontransport interventions: from processingtechnologies to cyberspace

In addition to improving access to the pavedroad system there may be complementarymeasures which could be implemented to im-prove value- added to local production, such asimprovements in crop storage, expansion oflocal processing technologies, and assistancefor women in acquisition of grinding machinesand similar equipment (those I encountered inoff-road villages in coastal Ghana were allowned by men). The development of specialisthigh-value agricultural products would also beadvantageous. There may be particular oppor-tunities for the development of organicallygrown products for export, given the high costof transporting inputs such as chemical fertil-izer and insecticides to off-road areas.A particularly interesting question is the

potential role of electronic communications inimproving rural access. Hanson (1998) con-trasts conventional life ‘‘on the road’’ with anew kind of ‘‘off the road’’: the ‘‘frictionlessrealms of cyberspace,’’ which can bring newkinds of interaction, increasing access andequity through the substitution of telecommu-nication for travel. In rural Africa, the potentialof telecommunications, for instance, to providedistance learning, advice, market price infor-mation and a host of other services, is ex-

tremely exciting. In theory, there is hugepotential for IT to transform access, withenormous implications for social equity. Butalthough e-mail has reached rural Mali, Kenyaand South Africa, for example, and Mandelareportedly pledged support for a telecommu-nications development fund that aims to extendcommunications ‘‘to every village in Africa,’’the development of even district-level telecen-tres is probably some far off in most coun-tries. 6 The South African pilot program forrural telecenters has been very slow to take offand has experienced substantial operatingproblems (Schreiner, 1999). The result of manysuch programs when (and if) they arrive, maybe to produce even greater gaps between dis-trict administrative headquarters and margi-nalized off-road centers. Their impact will needto be carefully monitored. Telecommunicationshave the potential to both destabilize and re-formulate, sometimes with important socialconsequences (Hillis, 1998). Graham (1998),reviewing information technology literaturewith a political economy perspective, presentsthe development of new telecommunicationsinfrastructure as ‘‘an asymmetric social struggleto gain and maintain social power, the power tocontrol space and social processes over dis-tance,’’ referring to their potential to extenddomination over excluded groups and so sup-port the production of divided spaces (see alsoMawdsley, Towsend, Porter, & Oakley, forth-coming).

(c) Groups and networks: building social capitalin off-road areas

Given the poverty of many off-road popula-tions, the question of group ownership andoperation has arisen both with reference toIMTs and conventional transport. Community-level action is also an essential component ofmany road maintenance schemes. A group/community approach is now widely favored bydonors and NGOs in Africa as a means ofproviding services following the withdrawal ofmuch public sector service support with struc-tural adjustment. The World Bank in its cur-rent Village Infrastructure Project in Ghana,for example, focuses its support on farmers’groups.We need to look very carefully, however, at

the potential for group ownership/action inspecific circumstances, since there is ample ev-idence of failure of past group/communityprojects across Africa. Ghana might seem an

LIVING IN A WALKING WORLD 295

ideal location for such schemes, since thecountry has a rich heritage of indigenous as-sociations, such as the asafo companies, nnoboalabor groups and susu groups. But a review oftheir activities in the study area and elsewherein Ghana (Lyon, 2000) suggests that the suc-cessful groups are those which have been es-tablished by their own initiative, evolvedslowly, and allow individual members to iden-tify their own objectives and manage their ownincome. They have shared norms of trust andcooperation and are embedded in social rela-tions which have evolved over time. Groupsundertaking joint financial decision-makingand operation, particularly those which havebeen established as a response to donor initia-tives, are far harder to sustain and most disin-tegrate rapidly once external support iswithdrawn. Aryeetey and Appiah (1995) makethe point that the success of groups seems to bevery dependent on ‘‘the extent to which pre-vailing local norms and values regarding groupformation and action are taken into account.’’Group formation is considered in this paper

with specific reference to transport-related in-terventions. Given the disappointing results ofmany externally-initiated group projects, thereis now growing emphasis on the potential forstrengthening existing local social capital. TheWorld Bank thus talks about building socialinstitutions by ‘‘scaling up’’ (World Bank,2000, p. 131). This could have repercussions farbeyond basic transport access. Building orga-nizations and networks can have a remarkableimpact in the negotiation of relationships be-tween localities and the market and the state, asBebbington (1997) has illustrated in a ruralAndean context. He shows how involvement inorganizations and networks can strengthen thehands of individual farm families in accessingand negotiating with non-local actors, institu-tions and organizations, leading to an intensi-fication of agricultural production and localcapital accumulation. Bebbington (citing Ev-ans, 1996a,b) emphasises that social capital canbe constructed and that key individuals—notnecessarily resident—can play a critical role.Evidence from sub-Saharan Africa suggests

that rather than intervening by initiating orsupporting the development of new groups,donors and NGOs need to explore the potentialfor working with and helping to build organi-zations and networks already in existence and,in tandem, promoting policy contexts that willencourage the emergence of indigenous ruralorganizations and networks. The policy context

needs careful attention. The expansion of theformal bureaucratic organization of the statemay be seen as crowding out informal networks(Evans, 1996a), and this seems to have hap-pened in the case of feeder road maintenance inareas like coastal Ghana. Putnam’s synergisticapproach (Putnam, 1993, cited in Evans,1996a) provides a more positive view of civicengagement, but suggests positive state–civilsociety links will need careful nurturing. In thecontext of state decentralization, financial as-sistance to village/area representatives, localadministrators and technical staff (agriculturalextension staff, health workers, etc.) specificallyfor off-road travel, and a policy environmentwhich supports their interactions with off-roadsettlements, (perhaps, for example, through off-road transport subsidies and/or provision ofbicycles or motorcycles), would seem to be anessential step in encouraging state–civil societylinks and thereby reducing the isolation andinvisibility of off-road populations. Unfortu-nately, although the financial costs of such ac-tion could be remarkably modest, particularlyby comparison with road construction, they areprobably still too high for most local govern-ment administrations in countries like Ghanato contemplate. Similarly, while donors seem tohold out hopes that NGO/decentralized gov-ernment collaboration could provide a fertileseedbed for development work, through NGOssupporting initiatives at district level, the evi-dence from Ghana and elsewhere in sub-Saha-ran Africa to date suggests frictions commonlyemerge around issues of political popularityand legitimacy. Most collaboration betweenNGOs and local government has consequentlybeen very limited and ad hoc (Clayton, 1998;Kyei, 2000). We might thus be wise to ponderBryceson’s (1999, p. 47) caution regarding theemphasis on building social capital. She sug-gests it may be counter productive, ‘‘repre-senting donors’ attempts at making do withdeclining physical resource transfers, ratherthan reflective of the actual needs of ruraldwellers,’’ and proposes that an emphasis onhuman capital is more critical.

6. CONCLUSION

The poverty and ill-health prevalent amongrural populations across Africa is compoundedby poor physical access. There is ample evi-dence to suggest that accessibility and mobilityare embedded in the development nexus in

WORLD DEVELOPMENT296

complex and far-reaching ways. Inhabitants ofoff-road settlements in coastal Ghana, aselsewhere in sub-Saharan Africa, are fre-quently marginalized and invisible, even—orperhaps especially—to members of local ad-ministrations. Although some internationaland local NGOs appear to be making moreconcerted efforts to reach the hidden ruralpoor, ‘‘tarmac bias’’ is probably as pervasiveamong government officials and their ilk todayas Chambers (1983, pp. 13–16) observed itnearly 20 years ago. Decentralization pro-grams now in place in many African countriesbring some devolution of power down to localadministrative headquarters, but between localgovernment headquarters and men and womenliving in off-road settlements contact often re-mains minimal and opportunities for lobbyingextremely rare.Expanded community track/road mainte-

nance, IMTs, cooperative transport arrange-ments and low-cost off-road crop processingmay all have a role to play in alleviating ac-cess problems in specific locations, but eco-nomic and cultural constraints and the politicsof implementation (including gender relations)need to be very carefully explored prior to andduring the development of new initiatives. AsLeinbach (2000) emphasizes in a recent reviewof mobility/development issues, we need adeeper understanding of personal and family

mobility needs in order to develop more ap-propriate transport policies. I would arguethat, additionally, we need a greater appreci-ation of access quality issues, including a spe-cific recognition of the significance (insymbolic as well as practical terms) of thetarmac road.Finally, while we will need to look to ways of

supporting the growth of social and humancapital in off-road areas through nurturingimprovements in state-society relations, we alsoneed to be aware of the complexities of such anapproach. Effective NGO/Local Governmentcollaboration is probably crucial to this, but isunlikely to be successful without substantialstate transfers of both human and financialresources from the centre to district level: de-centralization in deed as well as in name. Localgovernment staff will need to be more confidentof their professional skills and better financed ifthey are to interact effectively both with inter-national and local NGOs and with the com-munities they are supposed to serve.Strengthened local organizations and networkscould then have a critical role to play in raisingthe profile of off-road areas and of integratingtheir populations into a more democratic, de-centralized Local Government system. Thegains in terms of local democracy, social equityand, not least, improved rural livelihoodswould be very substantial.

NOTES

1. Moore’s (1979) paper is a notable exception (cf.

Moore, 1979). The World Bank’s sub-Saharan Africa

Transport Policy Programme—SSATP—has initiated a

number of village-level travel and transport studies

(including Barwell, 1996) which have begun to draw

greater attention to the critical importance of accessib-

lity. There are, obviously, exceptions to my generaliza-

tion regarding the relative poverty of off-road villages.

For example, Lyon (2000), in a study of Ghanaian

vegetable producers, points to variations in wealth

among villages which may more closely relate to

productivity of the land, and capital and other benefits

amassed during the cocoa boom, than to current road

access.

2. Though see Chambers (1997, pp. 20–21) re overes-

timation of post-harvest crop losses.

3. District Assembly sub-committees such as District

Environmental Management Committees face similar

problems (Porter & Young, 1997).

4. There have been contributions to this debate on the

GREAT network and rural-transport-develop-

ment@mailbase.ac.uk (D. Seddon, M. Wattam, P.

Winkelmann, N. Sieber). Winkelmann provides a case

study in Flores, Indonesia, where self-help initiatives on

village access roads have been successful.

5. In the Ghana study I encountered only one woman

transport owner, a farmer and petty trader in one of the

off-road villages.

6. Telecenters have received some attention in the

AFR-FEM Internet Working Group’s online discus-

sions on ICTs (Information and Communications

Technology), with debates about telephone coopera-

tives, use of the internet by women’s groups, and HF-

radio. In Accra a Centre for Women’s Information

Research and Support (CWIRS) has been established.

One interesting community media pilot project by Eco

News Africa, based in Nairobi, will put up FM

community radio stations in three rural communities

LIVING IN A WALKING WORLD 297

in Tanzania, Uganda and Kenya using solar power and

HF- radio/internet links, in areas with little or no access

to electricity and telephone lines (M. Wambui, AFR-

FEM, August 4, 1998). In Bangladesh, the Grameen

Bank apparently provides loans for rural women to

purchase mobile phones but the impact of this pro-

gramme is somewhat controversial (Pramada Menon,

GREAT network, April 9, 1998).

REFERENCES

Airey, A. (1992). The impact of road construction on thespatial characteristics of hospital utilization in theMeru district of Kenya. Soc. Sci. Med., 34(10), 1135–1146.

Airey, A., & Wattam, M. (1998). Community participa-tion in road maintenance. ASIST Bulletin, 7(July), 3–4.

Anyinam, C. (1994). Spatial implications of structuraladjustment programmes in Ghana. Tijdschrift vooreconomische en sociale geografie, 85(5), 446–460.

Aryeetey, E., & Appiah, E. (1995). Associations ofinformal producers in Ghana: report of a study of 6informal associations in the agro-metal and foodprocessing subsectors. FIT working document, No.13, September.

Ashong, E. N. K. (1996). Labour-based road improve-ment in Ghana: the development and utilization ofsmall scale contractors. Accra: USAID/DANIDA/ILO/Government of Ghana.

Ayee, J. (1996). The measurement of decentralization theGhanaian experience 1988–92. African Affairs, 95,31–50.

Barwell, I., Edmonds, G. A., Howe, G. D. F., & DeVeen, J. (1985). Rural transport in developing coun-tries. London: IT Publications.

Barwell, I. (1996). Transport and the village: findingsfrom African village-level travel and transport surveysand related studies. World Bank Discussion Paper344, Washington, DC: World Bank.

Bebbington, A. (1997). Social capital and rural intensi-fication: local organizations and islands of sustain-ability in the rural Andes. The Geographical Journal,163(2), 189–197.

Beenhacker, H. L. (1987). Rural transport services: Aguide to their planning and implementation. London:IT Publications.

Bishop, P. (1996). Off road: four-wheel drive and thesense of place. Environment and Planning D: Societyand Space, 14, 257–271.

Booth, D., Hanmer, L., & Lovell, E. (2000). Poverty andtransport: a report prepared for the World Bank incollaboration with DFID. Final report and Povertyand Transport toolkit. London: ODI, June.

Bryceson, D. F. (1999). Sub-Saharan Africa betwixt andbetween: rural livelihood practices and policies. Af-rika-Studiecentrum Working Paper 43, Leiden.

Bryceson, D. F., & Howe, J. (1993). Rural householdtransport in Africa: reducing the burden on women?World Development, 21(11), 1715–1728.

Buabeng, S. N., Sarfo-Mensah, P., & Dennis, C. (1995).Rural transport in northern Ghana: the impact ofintermediate forms of transport in rural areas. Brad-ford, Development and Project Planning Centre,Discussion Paper 57, Bradford, UK.

Chambers, R. (1983). Rural development: Putting the lastfirst. London: Longman.

Chambers, R. (1997). Whose reality counts? Putting thelast first. London: IT Publications.

Clark, G. (1994). Onions are my husband: Survival andaccumulation by West African market women. Chi-cago: University of Chicago Press.

Clayton, A. (1998). NGOs and decentralised governmentin Africa. INTRAC Occasional Paper Series No. 18,December, Oxford.

Creightney, C. D. (1993). Transport and economicperformance: A survey of developing countries. WorldBank Africa technical department, Technical Paper232, Washington, DC.

Dawson, J., & Barwell, I. (1993). Roads are not enough.London: IT Publications.

Doran, J. (1990). A moving issue for women: islow cost transport an appropriate intervention toalleviate women’s burdens in southern Africa. Gen-der Analysis in Development, Discussion PaperNo. 1, School of Development Studies, UEA,Norwich.

Ellis, S. D., & Hine, J. L. (1998). The provision of ruraltransport services: Approach paper. World BankSSATP Working Paper 37, Washington, DC.

Evans, P. (1996a). Introduction: development strategiesacross the public–private divide.World Development,24(6), 1033–1037.

Evans, P. (1996b). Government action social capital anddevelopment: reviewing the evidence on synergy.World Development, 24(6), 1119–1132.

Fernando, P., & Porter, G. (Eds.) (forthcoming).Balancing the load. London: Zed.

Filani, M. O. (1993). Transport and rural developmentin Nigeria. Journal of Transport Geography, 1(4),248–254.

Filani, M. O., & Richards, P. (1976). Periodic marketsystems and rural development: the Ibarapa casestudy. Savanna, 5(2), 149–162.

Francis, P., with Akinwumi, J., et al. (1996). State,community and local development in Nigeria. WorldBank Technical Paper 336, Washington, DC.

Gould, P. R. (1960). The development of the transporta-tion pattern in Ghana. Evanston, IL: Department ofGeography Northwestern University.

Graham, S. (1998). Conceptualizing space place andinformation technology. Progress in Human Geogra-phy, 22(2), 165–185.

Grieco, M., Apt, N., & Turner, J. (1996). At Christmasand on rainy days: Transport travel and the femaletraders of Accra. Aldershot: Avebury.

Guyer, J. I. (1997). An African niche economy: Farmingto feed Ibadan 1968–88. Edinburgh: EdinburghUniversity Press.

WORLD DEVELOPMENT298

Hanson, S. (1998). Off the road? Reflections on trans-portation geography in the information age. Journalof Transport Geography, 6(4), 241–249.

Hillis, K. (1998). On the margins: the invisibility ofcommunications in geography. Progress in HumanGeography, 22(4), 543–566.

Hine, J. L. (1984). Some limitations to the opportunitiesfor road investment to promote rural development.Paper presented at the International Conference onRoads and Development, Paris.

Hine, J. L. (1993). Transport and marketing priorities toimprove food security in Ghana and the rest of Africa.Paper presented to the International Symposium:regional food security and rural infrastructure,Giessen, May.

Hine, J. L., & Riverson, J. D. N. (1982). The impact offeeder road investment on accessibility and agricul-tural development in Ghana. Institute of Civil Engi-neers Conference on criteria for planning highwayinvestments in developing countries, May.

Howe, J., & Richards, P. (Eds.). (1984). Rural roads andpoverty alleviation. London: IT publications.

Kyei, P. (2000). Decentralization and poverty alleviationin rural Ghana. Unpublished doctoral dissertation,University of Durham, UK.

Law, R. (1999). Beyond ‘‘women and transport’’:towards new geographies of gender and dailymobility. Progress in Human Geography, 23(4),567–588.

Leinbach, T. R. (2000). Mobility in development con-text: changing perspectives, new interpretations, andthe real issues. Journal of Transport Geography, 8(1),1–9.

Levy, H., & Malone, P. O. (1988). Transport policy issuesin sub-Saharan Africa. EDI Policy Seminar Report 9.Washington, DC: World Bank.

Litvack, J., Ahmad, J., & Bird, R. (1998). Rethinkingdecentralization in developing countries. World BankSector Studies Series. Washington, DC: WorldBank.

Lyon, F. (2000). Trust and power in farmer-traderrelations: a study of small scale vegetable productionand marketing systems in Ghana. Unpublished doc-toral dissertation, University of Durham, UK.

Madulu, N. F. (1998). Changing lifestyles in farmingsocieties of Sukumaland: Kwimba district, Tanzania.Working Paper 27. Leiden: Afrika-Studiecentrum.

Malmberg-Calvo, C. (1994a). Case studies on the role ofwomen in rural transport: Access of women todomestic facilities. SSATP Working Paper 11. Wash-ington, DC: World Bank.

Malmberg-Calvo, C. (1994b). Case study on IntermediateMeans of Transport: Bicycles and rural women inUganda. SSATP Working Paper 12. Washington,DC: World Bank.

Malmberg-Calvo, C. (1998). Options for managing andfinancing rural transport infrastructure. TechnicalPaper 411 Washington, DC: World Bank.

Mawdsley, E., Towsend, J., Porter, G., & Oakley, P.(forthcoming). Knowledge, power and developmentagendas: NGOs North and South. Oxford: Intrac.

McCall, M. K. (1985). The significance of distanceconstraints in peasant farming systems with special

reference to sub-Saharan Africa. Applied Geography,5, 325–345.

Meagher, K. (1999). If the drumming changes, the dancealso changes: de-agrarianization and rural non-farmemployment in the Nigerian Savanna.Working PaperNo. 40. Leiden: Afrika-Studiecentrum.

Moore, M. (1979). Beyond the tarmac road: a (nut)shellguide for rural poverty watchers. RRA conferencepaper. Brighton. Sussex: Institute of DevelopmentStudies.

Nikoi, G. (1996). The experiences of rural banks inenterprise credit: Akuapem rural bank case study. InE. Aryeetey (Ed.), Small enterprise credit in WestAfrica. Accra: The British Council/ISSER.

Nugent, P. (2001). Winners losers and the rains: moneymoral authority and voting patterns in the Ghana2000 election. African Affairs, 100(400), 405–428.

Ostrom, E., Schroeder, L., & Wynne, S. (1993). Institu-tional incentives and sustainable development: Infra-structure policies in perspective. Boulder, CO:Westview.

Platteau, J.-P. (1996). Physical infrastructure as aconstraint on agricultural growth: the case of sub-Saharan Africa. Oxford Development Studies, 24(3),189–219.

Plumbe, A. J., Bascou, J., Etounga-Manguelle, D., &Maiga, M. (1995). Sub-Saharan Africa transportpolicy programme (SSATP): Prospective Review.Commission of European Communities/WorldBank. November.

Porter, G. (1988). Perspectives on trade, mobility andgender in a rural market system: Borno North-EastNigeria. Tijdschrift voor economische en socialegeografie, 79(2).

Porter, G. (1995). The impact of road construction onwomen’s trade in rural Nigeria. Journal of TransportGeography, 3(1), 314.

Porter, G. (1997). Mobility and inequality in ruralNigeria: the case of off-road communities. Tidjschriftvoor Economische en Sociale Geografie, 88(1), 65–76.

Porter, G. (2001). Rural access issues and the supply ofurban food markets in Nigeria: focus on market accessfor smallholder vegetable producers on the Jos Pla-teau. Final Technical Report to the UK Departmentfor International Development, May.

Porter, G., & Young, E. (1997). Decentralised environ-mental management and popular participation incoastal Ghana. Journal of International Development,10, 515–526.

Richards, P. (1985). Indigenous agricultural revolution:Ecology and food production in West Africa. London:Hutchinson.

Riverson, J. D. N., & Carapetis, S. (1991). Intermediatemeans of transport in sub-Saharan: Its potential forimproving rural travel and transport. Technical Paper161. Washington, DC: World Bank.

Riverson, J. D. N. et al. (1991). Rural roads in sub-Saharan Africa: Lessons from World Bank experi-ence. Technical Paper 141. Washington DC: WorldBank.

Salifu, M. (1994). The cycle trailer in Ghana: a reason-able but inappropriate technology. African Technol-ogy Forum, 7(3), 37–40.

LIVING IN A WALKING WORLD 299

Samoff, J. (1990). Decentralization: the politics of inter-ventionism. Development and Change, 21, 513–530.

Sarris, A., & Shams, H. (1991). Ghana under structuraladjustment in agriculture: The impact on agricultureand the rural poor. New York: New York UniversityPress.

Scholten, B. A. (1997). Case study of a grass-rootscharity/NGO: riders for health (Rfh). Paper pre-sented at the annual conference of the DevelopmentStudies Association, September. University of EastAnglia.

Schreiner, H. (1999). Rural women, development, andtelecommunications: a pilot programme in SouthAfrica. Gender and Development, 7(2), 64–70.

Steele, P. (1993). Shipping on the cheap: a brief look atsome lowcost transportation options. Ceres, 140, 28–30.

Stock, E. A., & de Veen, J. (1996). Expanding labour-based methods for road works in Africa. TechnicalPaper 347. Washington, DC: World Bank.

Taffe, E. J., Morrill, R. L., & Gould, P. R. (1963).Transport expansion in underdeveloped countries.Geographical Review, 53(4), 503–529.

Urasa, I. (1990). Women and rural transport: An assess-ment of their role in sub-Saharan Africa. Geneva:International Labour Office.

White, P. S., Erlank, G., & Matthews, M. (2000).Promoting Intermediate Means of Transport in Ghanathrough a South–South cooperation. Afribike,June.

Wilbur Smith Associates (1998). Road condition study,Final Report. Accra, May.

Wolff, J. (1995). Resident alien: Feminist cultural criti-cism. New Haven: Yale University Press.

World Bank (1994). World development report 1994:Infrastructure for development. New York: OxfordUniversity Press.

World Bank (2000).World development report 200/2001:Attacking poverty. New York: Oxford UniversityPress.

Yunusa, M. -B. (1999). Not farms alone: a study of rurallivelihoods in the Middle Belt of Nigeria. WorkingPaper 38. Leiden: Afrika-Studiecentrum.

WORLD DEVELOPMENT300