globalizing development in tanzania

7/28/2019 Globalizing Development in Tanzania

1/22

http://coa.sagepub.com

Critique of Anthropology

DOI: 10.1177/0308275X030230020012003; 23; 123Critique of Anthropology

Maia GreenParticipatory Project Management

Globalizing Development in Tanzania: Policy Franchising through

http://coa.sagepub.com/cgi/content/abstract/23/2/123The online version of this article can be found at:

Published by:

http://www.sagepublications.com

can be found at:Critique of AnthropologyAdditional services and information for

http://coa.sagepub.com/cgi/alertsEmail Alerts:

http://coa.sagepub.com/subscriptionsSubscriptions:

http://www.sagepub.com/journalsReprints.navReprints:

http://www.sagepub.co.uk/journalsPermissions.navPermissions:

http://coa.sagepub.com/cgi/content/refs/23/2/123Citations

at Inst Univ dEtudes on May 27, 2009http://coa.sagepub.comDownloaded from
http://coa.sagepub.com/cgi/alertshttp://coa.sagepub.com/cgi/alertshttp://coa.sagepub.com/subscriptionshttp://coa.sagepub.com/subscriptionshttp://coa.sagepub.com/subscriptionshttp://www.sagepub.com/journalsReprints.navhttp://www.sagepub.com/journalsReprints.navhttp://www.sagepub.co.uk/journalsPermissions.navhttp://www.sagepub.co.uk/journalsPermissions.navhttp://coa.sagepub.com/cgi/content/refs/23/2/123http://coa.sagepub.com/http://coa.sagepub.com/http://coa.sagepub.com/http://coa.sagepub.com/http://coa.sagepub.com/cgi/content/refs/23/2/123http://www.sagepub.co.uk/journalsPermissions.navhttp://www.sagepub.com/journalsReprints.navhttp://coa.sagepub.com/subscriptionshttp://coa.sagepub.com/cgi/alerts


2/22

Globalizing Development in

Tanzania

Policy Franchising through Participatory Project

Management

Maia Green

Department of Social Anthropology, University of Manchester

Abstract This article explores the social processes through which projects andprogrammes in international development become standardized in the contextof global social policies. The striking similarity in development projects in diversesettings occurs despite the introduction of participatory planning methodolo-gies which are intended to constitute the mechanism through which benefici-aries can become involved in the design and implementations of interventionswhich affect them. An anthropological account of stakeholder workshops in the

development sector of Tanzania shows how the workshops facilitated by thosedefining themselves as development professionals create the social space ofprojects as envisioned in documentation and how the management modelswhich inform development planning create development, not as failure as otheranthropological accounts would have it, but as success in relation to the achieve-ment of their planned objectives.

Keywords development management international development Tanzania

Why are development projects so similar in so many places? And why,despite the rhetoric of participation in which the poor apparently con-

tribute to the design of strategies to lift them out of poverty, is the currentincarnation of development so uniform wherever it appears? Can it reallybe the case that the intended beneficiaries of the international develop-ment effort are universally committed to the managerial models currentlypresented as solutions to the problem of poverty and to the neo-liberalpolicy prescriptions they entail? This article offers some provisional answersto these questions through an examination of the planning practicesthrough which development, defined in practical terms as what develop-ment agencies do, actually takes place. I argue that while the standardiza-tion of development globally is partly a function of the standardization of

development problems and solutions (Arce, 2000; Ferguson, 1990), it is alsoa consequence of the kinds of practices used to plan and implement

Article

Vol 23(2) 123143 [0308-275X(200306)23:2; 123143;029008]Copyright 2003 SAGE Publications(London, Thousand Oaks, CA and New Delhi)

http://www.sagepublications.com/http://coa.sagepub.com/http://coa.sagepub.com/http://coa.sagepub.com/http://coa.sagepub.com/http://www.sagepublications.com/


3/22

development as a process of transforming policy visions into manageablerealities through the social constitution of projects subject to specific tech-niques of audit, organization and control (Strathern, 2000: 2). Develop-

ment does not produce unanticipated effects in the sense proposed byFerguson (1990). Although the long-term impacts rhetorically claimed bydevelopment are unlikely to be achieved, for reasons touched on below,those involved in planning development are well aware of the limitationsof what they are trying to achieve and their short-term objectives are real-istic in relation to the particularities of development management.

Development and its side-effects

International development is currently being reformulated. Assumptionsabout economic growth and modernization theory are being supersededby paradigms emphasizing the multifaceted nature of poverty and theimportance of human rights. International Development Targets nowinform the policies of the whole hierarchy of institutions involved indevelopment implementation,1 from the lowest-tier community-basedorganizations to the multilateral and bilateral agencies that provide thebulk of development funding. Consensus is not absolute. Critiques of thecontemporary development agenda, and of the globalization with which it

is apparently aligned, are taking place on the margins of the formaldevelopment sector. For all their radical posturing, those talking againstdevelopment remain committed to something of a developmentalist2

perspective in relation to the South, and their positions are increasinglyincorporated within the development mainstream via international civilsociety organizations.3 While development represents itself in terms of thebeneficial intentions of policy content, it is viewed in a less positive light bythose who choose to position themselves outside its institutional structure.Development as an institution is seen by writers such as Crush (1995),Escobar (1995) and Ferguson (1990), among others, as a bureaucratic force

with global reach and an explicitly pro-capitalist agenda, operating as a toolof regimes that seek to perpetuate relations of inequality and dependencebetween the West and the rest and, through their representation, to per-petuate the construction of others as postcolonial subjects. Recent writingson globalization have attempted to reconceptualize the geographical basisof this relation, in the process acknowledging the re-siting of dependency,in which the transnational person simply relocates their subjugationthrough economic migration (cf Arce and Long, 2000: 15). The Mexicanmigrant cleaning in the rich households of California may be a post-peasant in the new categories of postmodernity, but he or she remainspoor.4 Moreover, this persistent poverty is a direct consequence of whatdevelopment has achieved. In the words of Michael Kearney, developmentprojects implemented by state agencies and civil society organizations alike,

124

Critique of Anthropology 23(2)

http://coa.sagepub.com/http://coa.sagepub.com/http://coa.sagepub.com/http://coa.sagepub.com/


4/22

in Mexico and elsewhere, actually function to reproduce rather thanimprove existing social, political, and economic problems (1996: 143).

Perceptions such as Kearneys are not unusual in anthropology, which

has viewed itself as a discipline to an extent outside, and hence untaintedby, close engagement with a development sector that has dirty hands.5 Inactuality, this claim to non-involvement cannot be supported. Even if mostanthropologists do not engage in development in the hands on sense ofapplied practitioners, anthropology has long been complicit in constitut-ing development, through contributing to the kinds of representations ofunderdeveloped otherness that make development thinkable, as well asthrough the pursuit of intellectual agendas that have directly or indirectlyserved development interests (Kearney, 1996; Thomas, 1994).6 Anthro-pologists claimed capacity to interpret local situations and grasp the truesignificance of discourses of power have now informed numerous accountsof development as failure, either in relation to meeting policy intentions orreinforcing local poverty.7 In these writings, which predominantly explorean apparent disjunction between development objectives and local realitiesin the context of fairly small-scale projects, development generally fails forthe same few reasons: lack of local knowledge on the part of the imple-menting agency, inadequate participation of beneficiaries in project designor because the representation of the problem which a project set out toaddress had little basis in local realities (Crewe and Harrison, 1998: 16;

Hobart, 1993). According to James Ferguson, in what has become perhapsthe best-known account read by anthropologists of development as failure(1990), the discourse of development functions as a machine whichobviates political understandings of local situations and ensures thatdevelopment, whatever its intentions, always serves primarily to furtherthe power of the state. While development might fail, in the sense of notachieving its stated objectives, Ferguson argues that it impacts on otherareas through side-effects. In the project he describes, which worked withlocal government in Lesotho, these were, not surprisingly, to strengthen theinstitutions of governance in the interests of the state and state officials.

According to Ferguson these side-effects were unanticipated and unin-tended by the people involved with the project as implementers andplanners. The project . . . may be said to have unintentionallyplayed whatcan only be called an instrumental role in strengthening state presence(1990: 2534).

In Fergusons analysis, which relies on a Foucauldian understanding ofthe relation between discourse and power, the good intentions of develop-ment implementers are subverted by the overriding discourse in which theyfind themselves. Even those potentially capable of presenting realistic infor-mation8 are sucked into the system, producing reports, analyses andsolutions which conform to the expectations of development discourse.Ferguson concludes:

125

Green: Globalizing Development in Tanzania



5/22

. . . the development apparatus in Lesotho is . . . a machine for reinforcing and

expanding the exercise of bureaucratic state power, which incidentally takes

poverty as its point of entry launching an intervention that may have no

effect on the poverty but does in fact have other concrete effects. Such a resultmay beno part of theplanners intentions indeed it almost never is but resultant

systems have an intelligibility of their own. (1990: 2556, my emphasis)

In an equally influential book exploring the significance of developmentin general for the making of the Third World, Arturo Escobar (1995)makes a similar argument about how development achieves its effects.These effects are not, of course, the stated objectives of development policy,but the reproduction of the relations of inequality on which global capi-talism and US dominance depend. In an account based on ethnographicresearch amongst civil servants in the Colombian Ministry of Planning,Escobar presents an image of development institutions as bureaucraticmonoliths in which planners and technicians uncritically impose a hegem-onic order on the South through the manipulation of development dis-course and planning certainties and, in so doing, seek to reduce its peopleto objects of intervention and control.

How anthropologists think development works

Both Ferguson and Escobar set out to deconstruct the assumptions thatthey think liberal Westerners might hold about development, forexample that its institutions work to eradicate poverty, through what arerepresented as examinations of what development actually does.Development in these accounts is not presented as a collective represen-tation in the sense of the thoughts development agencies as institutionsthink (Douglas, 1986) in relation to the policy objectives shared by a rangeof social actors, including local political and civil society organizations aswell as, more obviously, national governments and citizens. Nor does it

appear as an analytical category pertaining to certain kinds of outcomesand effects. Rather, development as an institution and as a discourse is pre-sented as a discrete entity whose influence can be considered in isolationfrom the other factors which contribute to its production and as an agentin its own right, a thing with a life of its own. Development, from thisperspective, is confined to planning ministries charged with developmentprogrammes and to the multilateral, bilateral and non-governmentagencies explicitly oriented towards its implementation. Developmentfailure and development power become the achievement and responsi-bility solely of development institutions and discourse, endowing develop-ment with an influence that is quite remarkable considering its propensityto failure. In reality, of course, the side-effects of development onlyappear as such when development is viewed as an institutional structure

126




6/22

external to the states and structures of governance with which it works. Thestrengthening of state power in Lesotho, which Ferguson regards as aninstrumental effect of development, was arguably an intended effect, a

necessary precondition for strengthening the local authority serviceswhich the project set out to support. Its appropriation by a repressiveregime was more a consequence of the political situation in Lesotho thanan automatic effect of development.

Both Ferguson and Escobar present an alienated view of developmentas a reification of institutional practice in which human agency has a limitedrole to play. This stems in part from the anthropological perception ofadministration and governance as externally determined fields, whosecontent is reducible to the play of bureaucrats concerned with passing thebuck and preserving their own positions in what Herzfeld has character-ized as secular theodicy resulting in the social production of indifference(1993). This reification of development as a bureaucratic machine whichoverrides the agency of the individuals engaged in it permits the construc-tion of an anthropological caricature of development as monolith, denyingthe capacities of social actors within and outside it to influence develop-ment outcomes.9 More fundamentally, although planning is central to theoperation of development in both accounts, this process is never exploredethnographically, merely attributed to the routine of bureaucrats and theproduction of documents that reinforce the hegemonic representation.10

Why development bureaucrats should seek to reproduce development inthe interests of global capitalism or state power is not questioned. Thisarticle adopts a slightly different approach to exploring development,defined in terms of what development agencies actually do and the policyobjectives which drive them. An account of the specific practices associatedwith international development processes in Tanzania permits an analysisthat informs and challenges the ways in which anthropology has tended tointerpret the outcome of development interventions. Such an approachhighlights the importance of the social practices involved in planningdevelopment interventions and the particular ways in which development

problems and solutions are represented through simplistic monocausalchains amenable to transformation via the transfer of financial resources.This, rather than naivety about state power, probably accounts for theabsence of any discernable relation between the interventions described byFerguson for the CIDA project in Lesotho and the reduction of poverty.The planners and senior agency staff along with their governmentcounterparts were doubtless committed to strengthening the capacities ofthe different tiers of government to deliver various kinds of services. Theeffects on government were intentional as a means to an end. What wasflawed in the Lesotho project, and indeed in much development practice,was not so much the activities which the project implemented as the pur-ported relation between cause and effect on which programming ispremised.

127




7/22

Development planning is concerned with outputs

Metaphors derived from stereotypical images of industrial production

notwithstanding, development technicians are not mere operativesmechanistically perpetuating the rituals of bureaucratic practice (Escobar,1995: 88). Nor does development planning occur in an instant under thesole control of planners in thrall to the development vision. Developmentplanning involves representatives and professional specialists from donorand recipient organizations, as well as personnel recruited to manageimplementation and representatives of the local institutions with which aproject is to work. Increasingly, representatives of the so-called beneficiarygroups are invited to participate in design and management processes, forexample the users of public services or members of farmers groups whoare the intended targets of development assistance. Despite unequalrelations of power between development agencies and the recipients ofdevelopment assistance, those at the receiving end of development do havethe capacity to influence programming and, on occasion, to reject it alto-gether. Even in the work of the IMF, perhaps the most powerful develop-ment exemplar of them all, acceptance of the shared reality that will forma common ground for policy development and planning is the outcomeof a negotiated process between the multilateral agency and recipientgovernment (Harper, 2000: 30). Bearing in mind the fact that such pro-

grammes are often claimed as the initiative of recipient governments,particularly at national level, and that the planning process only starts inearnest once governments have agreed in principle, outright rejectionalmost never happens. Too much is at stake in the proposed transfer ofresources to jeopardize the process, and vested interests on the part ofdonor representatives, as well as recipients, ensure that dialogue continuesover an extended time-frame. Agency staff have their own objectives toachieve in the contracts and performance plans derived from country andsector strategies, which depend on securing the right kinds of agreements.However, other strategies are adopted by stakeholders, including those

within development agencies themselves, to alter or nullify proposed andongoing programmes. These include delaying tactics, insisting that a pro-gramme is redesigned from scratch, and the initiation of changes in direc-tion as a consequence of review missions.

Planning in development, that is in agencies which manage resourcetransfers to support policy objectives, really concerns the shorter-termpurpose of specific funded activities which are intended to producemeasurable outputs at the end of a finite three- to five-year period. It isless concerned with the goal, the higher-tier objective which the purpose isintended to contribute towards. In the logic of the planning techniquesused in development (output-oriented programming or project cyclemanagement for example), outputs are distinct from impacts. Impacts arein fact inferred from the outputs having been achieved through what are

128




8/22

known as output-to-purpose reviews, when at the end of a specified timeperiod assessors try to determine the extent to which the achievement ofthe outputs stated in the original plan has contributed to the purpose of

a project. Despite recent trends towards larger-scale and more flexible pro-grammes, the aggregation of smaller-scale projects and public reforms intooverarching national-level sector programmes, and the stated emphasis onprocess projects in which activities are determined on an ongoing basisrather than specified in advance, the package of activities and fundingoriented towards the achievement of specified outputs and purpose con-stitutes a project, irrespective of scale, and is subject to the same tech-niques of representation and management. The project as matrix of inputsand outputs subject to time and cost is conceptualized as an entity to becontrolled in standard management terms.

Documenting projects

Development management is concerned with the relation between invest-ments and outputs over a specific time-frame. Like other audit techniques,it relies on the production and manipulation of texts through which theserelations can be quantified and controlled. The production of documen-tation at various points in the project cycle is the outcome of specific insti-

tutional practices which have become part of the recognized toolkit fordevelopment planning and management. Project documents use currentpolicy discourse to constitute particular representations of developmentproblems as amenable to particular interventions. Project documentation isnot intended to present an objective account of a social and economic situ-ation and a discursive assessment of how best to address the kinds of issueswhich might come under the ambit of current policy concerns but to supportfinancial transfers with a view to achieving carefully costed outcomes. Suchdocuments support particular positions and courses of action within fundingagencies where different sectors compete for limited amounts of cash, stating

why and how, and what amount of money should be spent by whom to con-tribute toward which policy objective. Consequently, such documents havemore in common with marketing texts than with social and economicanalysis. Given the amorphousness of social reality and the virtual impossi-bility of determining with any certainty the actual relation between what aproject does and other social processes, project documentation is vital in con-stituting the project as a slice of manageable reality where project inputs canbe seen to relate to outputs and, largely by inference, to effects on theground. Projects do not exist only at the level of documentation, althoughthis is the only place where a pure development project, divorced from otherinstitutions, can be represented. The pure project is also periodicallybrought into being as a social institution through the creation of projectspace via the techniques of participatory project management.

129




9/22

Participatory project management

I have been involved in a number of development projects in Tanzania

between 1996 and the present,11 in various capacities: as an itinerantanthropologist, as a low-level development manager responsible forsigning off on projects within a particular sector in a bilateral agency12

and, more recently, as a social development consultant hired to redesignprojects and prepare project documentation. The account below is basedon this experience but applies equally to projects in other sectors financedby other agencies, and to sector programmes. The recent massive chan-nelling of external assistance to Tanzania,13 along with the tendency forinternational development agencies to adopt similar strategies and tech-niques, has contributed to the institutionalization of development practicesacross a range of personnel in development agencies as well as in centraland local government and, to a lesser extent, private individuals living incommunities which have been at the receiving end of development inter-ventions.14 In Tanzanian popular discourse development (maendeleo) is nolonger the prerogative of institutions associated with the state but has cometo refer more narrowly to the projects (miradi), paraphernalia and practicesassociated with international development assistance (Green, 2000). As wellas being associated with institutional presence in the form of offices, per-sonnel and vehicles, which continue to be heavily branded indicators of

donor presence in the country, the category development now conveysthe performance of certain practices in particular institutional forms.These centre on the stakeholder workshop15 as a site for the productionof project reality through what have become recognized as specialist tech-niques of participatory project management. These techniques, promotedby practitioners who have their own professional accreditations and globalnetworks, derive from the general project management strategies used inWestern industry and commerce, combined with the participatorymethods associated with rural development planning but informed by theassumptions of the New Public Management that the quantification of

inputs and outputs can apply to the domain of the social.16Participatory project management explicitly makes the process of

constituting the project a social process through the creation of insti-tutional spaces in which different people can contribute to its construc-tion and, to a lesser extent, control. The documented products ofparticipatory processes feed into the documentation used to chart anddirect the project through its funding cycle from design to appraisal(when the design is assessed by interested parties) and implementation,which is monitored through a series of reviews.17 Mid-term18 and end-of-project reviews are the most important, the former for ensuring that aproject can continue and allowing for some redesign, the latter forallowing funding agencies to determine whether or not a project hassucceeded in achieving its purpose. Different agencies have different

130




10/22

conventions about what can be altered once a project is under way. In theagency with which I worked, the rule of thumb was that the goal andpurpose of a project should remain for the life of a project, but that the

outputs could be redrawn if necessary.Major reviews and appraisals involve an assessment of the routine docu-

mentation of spending and progress which project managers generate,together with field visits by project teams. Such teams typically compriseagency staff, government partners and external consultants specializing inthe particular sector which a project addresses, as well as the field staffresponsible for day-to-day operational management. Given the centrality ofmanaging the spend to the constitution of a project, and the emphasis onbangs for bucks, it is not surprising that the methodology of review is verysimilar to that of IMF audit missions (Harper, 2000). Quantification ofinputs and outputs is critical to assessment of performance, hence theconcern, both at the design stage and throughout implementation, withobjectively verifiable indicators (OVIs) as proxies for assessing impact andperformance. The core team works intensely over several days to collectdocumentation, and holds discussions with a wide range of institutional rep-resentatives who have an interest in the project. These discussions providereviewers with an indication of the possible basis for agreement about thefuture of a project and highlight issues which may need to be addressed fora project to succeed in the quite limited, but realistic, terms of achieving

itspurpose.In addition to project teams, reviews and appraisals involve groups ofpeople from a project area in what are known as stakeholder workshops.Stakeholders, in development parlance, are the individuals and institutionsdeemed to have a stake in a project and who have the potential to impacteither positively or negatively on project outcomes. Agency literatureacknowledges the diversity of interests amongst stakeholder groups andthat funding agencies often have the biggest stake in a project.19 InTanzania such workshops have become a routine part of the design ofdevelopment interventions at all levels, from the senior officials and donor

representatives involved in programming sector reforms to the relativelysmall projects, perhaps financed by international agencies but planned andimplemented by international and national NGOs. Formally such work-shops provide an opportunity to clarify the purpose and outputs of aproject. A more fundamental objective is to build credibility and legitimacyfor the intervention among particular social constituencies recognized bythe agency as having a stake in the project. The stakeholder categoryincludes representatives from donor and recipient organizations as well ascarefully selected representatives of beneficiary communities. It is to anextent created through the selection of participants. In the context ofprojects which are increasingly invisible to outsiders in the form ofspending largely allocated to capacity building and training, workshopsserve two purposes. In constituting projects as social institutions they make

131




11/22

projects tangible, and by creating a category of stakeholders they create asocial group with a stake in perpetuating project space from which moreimmediate benefits of development derive. In Tanzania at least, this

visibility perpetuates stereotypical representations of development asmaterial benefits and as particular techniques of project management.

Workshopping in Tanzanias development culture

Stakeholder workshops in Tanzanias development culture assume aparticular institutional form and are conducted in fairly standard ways, inline with the professional expectations both of the development facilitatorswho specialize in their operation and of the category of what might bereferred to as professional participants, those employed in professionalcapacities in development agencies and the public sector and whose workin maintaining aid-dependent administrations involves participation inworkshops. Participants from within donor agencies also have expectationsabout how workshops should be managed and what they are expected toproduce. A key shared expectation of this category is that workshops areprofessional, that is they conform in their structure and sequence to theproject management expectations of professional participants. Workshops,like projects, are units of management. As packages of inputs and outputs

they run to clear schedules and aim to deliver outputs within a designatedtime period. The production of outputs, in the form of bullet-pointed asser-tions or project logical frameworks, follows clearly determined stages offacilitated group work in which each product is part of a sequentiallymanaged process in the production of specific pieces of project documen-tation.

Professional participants also expect that the location and style of aworkshop conform to their own sense of being a professional involved indevelopment, that is as high-status persons engaged in high-status activities.Professionals expect to be valued and treated in a certain way which indi-

cates their status in relation to other social categories, in this context thepeasants it is the responsibility of public servants to administer. InTanzania professional status, both within the wider public sector andoutside it, is signified through reference to two parallel indices, on the onehand notions about patronage and the power to command others derivedfrom more regionally specific idioms of political relations (cf. Bayart, 1993)and, on the other, the material signifiers of global modernity. Theseinclude style of dress, access to electronic goods and vehicles and, increas-ingly, a professional style of accoutrements associated with management,symbolized by the ubiquitous business card exchanged between officials atall official meetings. The power to exert control, even if symbolic, overothers time or labour by making people wait or sending juniors on errandsare key indices of an officials status, indices shared by the expatriate staff

132




12/22

of development agencies whose own status is marked out vis-a-vis their Tan-zanian counterparts through better offices, luxurious housing, uniformedguards, private vehicles, extra benefits and higher pay. Being driven also

indicates professional status, combining the power to control others labourwith the car as a key signifier of managerial status. National and inter-national officials participating in project events expect to be driven inproject vehicles, even for short distances, as an index of their recognitionas professionals in a project setting, and expect to be able to send projectdrivers on personal errands, even to purchase something from a shop nextto which the vehicle is parked.

The association of externally funded development with signifiers ofmodernity and wealth as apparent characteristics of donor countriesinforms the ways in which development projects are instituted in Tanzaniaand the ways in which development professionals behave. Developmentprofessionals expect to stay at international hotels for example, and areexpected by their counterparts to do so, because such hotels are not merelythe most expensive but conform to shared expectations about internation-ality as outposts of what is taken to be the Euro-American metropolis. Thisstatus is signified by the availability of CNN, international football, dollarcurrency, international food. Similarly, international developmentprojects are expected to have the latest in vehicles and electronic officeequipment, even where these comprise a significant proportion of costs and

are not really essential to the work which a project is intended to support.The development workshops in which professionals participate tend to beheld in explicitly modern-style halls, attached to ex-mission churches orto former parastatal training institutes in regional towns, and in inter-national hotels in Arusha and Dar es Salaam. The expectation that work-shops must be held in the right kind of space, with modern facilities,militates against the utilization of other kinds of available public space forexample primary schools in rural areas for the production of develop-ment, reinforcing the contemporary association of participation indevelopment with professional urban status. It also desituates projects and

stakeholders from the geographical settings in which project impacts areexpected to occur.

Professionals attending workshops do so as part of their professionalvocation, but low levels of public sector pay necessitate compensation forthe demands on their professional time that participation entails. Dailyallowances (posho) are paid to workshop participants, often with differen-tial rates for government employees and beneficiaries.20 Rates for partici-pation generally exceed rates of daily pay for public servants, who have asignificant incentive to attend workshops and seminars wherever possible.The centrality of the workshop to the constitution of the project now meansthat attracting participants is essential for development managers who needparticipants to constitute their project to be managed. Developmentagencies in Tanzania, as elsewhere (Crewe and Harrison, 1998), pay

133




13/22

attendance allowances even to salaried persons because they know that,without pay, attendance will tail off and the credibility of a participatoryproject will decline. In some instances different agencies compete by

offering higher rates to attract participants involved in several projectssimultaneously. Substantial proportions of project budgets are spent onachieving this kind of professional participation, at the same time as theparticipation of beneficiaries in projects through what are euphemisticallyreferred to as contributions (michango) of labour or time are taken forgranted by the elite called to speak on their behalf. Stakeholder workshopshave become so essential to the constitution of development that workshopshave become proxies for project outputs, particularly in relation to intan-gibles like capacity enhancement and training. The ongoing Local Govern-ment Reform Programme, financed largely by multilateral and bilateraldonors and designed by an international audit consultancy, essentially com-prises a series of workshops through which local authorities are expectednot merely to reform themselves but to attain appropriate levels of compe-tency for programme implementation.

As well as allowances, participants are issued with notebooks and pens,and, as befits professionals, are served with elaborate meals featuring choicecombinations of high-status foods, including fish, meat, chicken and rice,in contrast to the average daily diet that a peasant would eat of maize andvegetables. Styles of clothing and self-presentation are also indicators of the

professional and international status associated with development both asproject and as progress. Project workshops are occasions for public servantsto dress in professional style and for representatives of local communitiesto dress in prestigious clothing to mark what have become prestigiousoccasions for participants, often reported in the national media wheredevelopment events are news and conveniently recorded by photogra-phers who are on hand to sell pictures to participants. The professionaliza-tion of participation in workshop settings and the association of workshopswith status means that participants, even in smaller NGO projects, are over-whelmingly drawn from among the local elites who come to the attention

of project staff and of public servants and whose levels of education andfamiliarity with the group work required of the workshop make them idealparticipants. The relativity of poverty and of developed versus developee ascategories of beneficiary and recipient (cf Pigg, 1992) means that these rep-resentatives of local communities are perceived by largely elite project staffand expatriates alike as poor, while communities themselves perceivethem as lower-tier officials and as patrons, who are at the same time clientsof those with higher-tier positions in the public administration and widersocial networks beyond the immediate area (cf Haugerud, 1995). The realpoor are invisible. They dont come to meetings, a project manager fromDar es Salaam told me, an invisibility reinforced by the exclusivity the pro-fessionalization of such workshops creates.

Although stakeholder workshops are concerned with participation,

134




14/22

they are not necessarily participatory (cf Pottier, 1997). The structure ofdevelopment workshops, in any event, probably precludes the kind of inputfrom stakeholders that would significantly alter the conception of a

project once the workshop has been convened. This is not to suggest thatparticipantscannot influence workshop outcomes, but to point out that inorder to do so they would need to be in a position from where they havethe authority to influence, and to be conversant with the skills of manipu-lating discussion and its representation on which the role of facilitatordepends. Even where representatives of beneficiary groups are present ina workshop setting, they are likely to be fewer in number than the pro-fessional groups and, given the etiquette of hierarchy and power inTanzania, will be less likely to speak critically before those representingthemselves as government. The tight organizational structure of facilitationand the construction of workshops as a site for the management of outputsensure that workshops produce highly limited visions.

Facilitating uniformity

Workshops derived from management practices reproduce the kind of text-based approach to the production of social outcomes on which projectdocumentation is based.21 Facilitators, generally self-employed consultants

with backgrounds in development management, that is in the productionand manipulation of project texts, adopt similar approaches to workshopstructure based either on the requirements of the project cycle or derivedfrom their own experience of similar workshops. The detachment of facil-itators from specific projects and the professionalization of facilitationdoubtless contribute to the standardization of workshop products and ofparticipants expectations. Facilitators are generally hired specifically to doworkshops and possibly to prepare key components of project documen-tation, usually centred around a logical framework which it is the objectiveof a workshop to either refine or produce. A facilitator charged with the

preparation of project documents knows that he or she22 has to get theworkshop to produce the kinds of material they need to make an accept-able project, that is, to produce the kind of analysis which will be acceptedby the agency as presenting the right kinds of relationships betweendifferent inputs and outputs and which is supported by the right kinds ofindicators which are realistic enough to be convincing and which are con-sistent with agency policy priorities.23 Facilitators must also create apparentconsensus amongst participants, who should feel as if they have all con-tributed to the design of the project through what is represented as a dis-cursive analysis of key problems and solutions.

Facilitation in such settings requires good people skills to retain adegree of control over the direction of debate in a room, as well as anunderstanding of the logic of project management. The latter is

135




15/22

graphically represented through what has become the basic tool ofdevelopment management, the logical framework. The creation of itsdifferent components through group work allows a facilitator to structure

a workshop. The logframe is in essence a matrix representing chains ofcausality in such a way that certain inputs, cash and activities, producecertain outputs which achieve a specific purpose which contributestowards a higher-order goal. Conversely, a representation of the hierarchyof causes contributing to a particular problem, such as local poverty dueto poor farmer prices due to lack of access to markets due to poor roads,for example, can be reversed to produce a project once priorities forintervention have been agreed.24These priorities are not primarily deter-mined though research or analysis but by the pragmatic requirements ofsector budgets and policy objectives.

Once timetables have been agreed for workshops, with necessarybreaks for tea and meals, participants are taken through sequences ofplenary discussions and group work around the management stages largelydetermined by the logic of the framework. Groups may be asked to producelists of key problems using marker pens and flip charts, or state reasons whycertain things occur. These are then reported back to the wider group andstuck on the wall for people to look at, literally enveloping participantsinside the project text. Reporting back provides an opportunity for the facil-itator, standing in front of the seated participants, to verbally edit the sug-

gestions and analytical contributions from groups and to obtain support forhis or her editorial decisions. This is achieved through rhetorical ques-tioning of the point which he wishes to reject and seeking the support ofprofessionals in the audience, who may also need to be edited out wheretheir own interests would threaten the policy objectives of a project (suchas working with the poor). The facilitator writes up the new points on flipcharts. These will form the basis of the written-up account of a workshopcirculated to participants and kept in project files, and inform the produc-tion of project management documentation. The debates around whycertain points are rejected or retained are not recorded.

The analytical constraints of the logical framework and the problemtree in demanding simplistic causal hierarchies produce simplisticstrategies to address complex multifaceted realities. This limitation is notperceived as a problem in the workshop setting because what matters, tofacilitators and participants alike, is what is made to matter by the consti-tution of projects as outputs to be managed. Achieving the outputs whichconstitute project success does not depend on an analysis of causality, buton the production of a hierarchy of causality. Effort is expended to producequality documents and logframes, rather than quality understanding oranalyses (cf Stirrat, 2000). Perhaps the most significant aspect of this kindof analysis and the planning it informs is that the problem analysis containsthe possible solutions likely to be accepted by project teams and partici-pants alike. These solutions, like the analysis itself, are apparently

136




16/22

straightforward inputs that will impact in a management sense on theproblems they are intended to address. The selection of solutions, that isinputs and activities, creates the project as a manageable entity subject to

audit and assessment. Activitiescan be performed, funded and evaluated,inputs measured as related to outputs. A structure for success is in place.

Project propositions and the practice of management

I have suggested that the social constraints on workshop formats and therules of participatory engagement and facilitation serve to reproduce thepure project as a particular kind of social institution which conforms tothe representations contained in project texts through which select aspectsof social reality become manageable. In collaborating to make reality man-ageable, stakeholders contribute to project management, although thisparticipation has little impact on the ways in which a project is conceptu-alized. This is partly due to the project politics around the selection ofstakeholders and the various professional interests in the perpetuation ofdevelopment as this kind of management process, but also because of theconstraints on thinking and on argument imposed by the dual frameworksof development and management.25 Some of these assumptions concernthe necessity of continuing to practise development and its administration

in their current forms. Others centre on the apparent truisms of the NewPublic Management concerning the means through which efficiency andeffectiveness can be achieved in public services, or that better governmentor female participation in education will lead to poverty reduction. In suchcontexts, where truths about development are not negotiated butaxiomatic and where expertise seems to be on hand to make developmentwork, certain things become simply unthinkable, or, if thought, are soondiscarded as untenable (Douglas, 1986: 76), while others are thoughtwithout thinking. The current management paradigm which underpinsthe Local Government Reform Programme being implemented in

Tanzania at present emphasizes privatization as the key to efficient publicservices. In the neo-liberal logic of New Public Management, efficientpublic services mean reduced public spending. In a quite unspecified waythis is assumed to lead inevitably to increased services for the public and,in the development context, reduced poverty. The restructuring manual,prepared by an international audit consultancy, around which localauthorities are expected to implement the reform process, encourageslocal authorities to privatize public toilets in order to achieve a quick winfor reform, which will improve services, raise revenue and impress thepublic. Of course, provision of public toilets is not a core activity of localgovernment in Tanzania and small savings will be made from the sale ofa run down pit latrine or two near a bus stand, yet not one of the localgovernment officers or development professionals participating in

137




17/22

workshops around public services challenged the proposition that thismight be a useful activity.

The propositions put forward in project settings acquire a degree of

plausibility which participants appear to accept, at least for the duration ofthe workshop, where the diagrammatic representations of cause and effectcan represent a social reality detached from both macro-politics and themicro-universes of individual decision-making. The conviction of themanagement models and the facilitators and the assent of experts andprofessionals give the projected project credibility. Logical frameworksseem logical, analytical. Projectsdoproduce anticipated outputs, at least inthe short term. What is less certain is the extent to which the outputs willinform the purpose of a project or whether the purpose of a project willinform the ultimate goal. However, from the perspective of developmentmanagement a project exists as a time-bound grid of relations betweeninputs and outputs subject to assessment and control which has beenbrought into being through the matrix of workshops and documentationthrough which this control is exerted.

Project documentation and, to an extent, project workshops, create theproject as a kind of entity separated off from other kinds of social realitiesin order to make it manageable. This separation is achieved throughspecific styles of documentation and analysis, as well as through thedynamics of workshopping where management logic ensures that partici-

pants produce standard project documentation as the output of a restrictedstyle of logical reasoning which is diagrammatically represented. Projectworkshops bring together different representatives of those deemed bydevelopment agencies to have a stake in the project, dramatizing a particu-lar representation of development as a matter of professional style and theperformance of management practices. The agency staff, project managersand professional facilitators who meet in project spaces come from a widevariety of countries and backgrounds, but all subscribe to the conventionsof development as the performance of management practices with whichthey have expertise. This expertise is not viewed as location- or country-

specific, but as generic to development. In terms of the production ofprojects in the sense described in this article it is.26Location and local know-ledge are simply not relevant to the construction of chains of causality andindicators of assessment that development constituted as project entails.

Project spaces and non-places

The social reality of project space as detached from wider social processeswhere the usual conventions of political engagement, and opposition, donot apply, and where different stakeholders may have completely differentunderstandings of what the issues are and why they matter brings to mindthe vision of non-places proposed by Marc Aug (1995, 1998). Non-places

138




18/22

are, for Aug, sites constituted by contemporary social relations which aredetached from the territorial place and social contexts that comprise whathe calls anthropological place (1995: 101). Paradigmatic non-places are

places through which people pass in their capacity as individuals detachedfrom social networks and obligations. Such places include the airports andfreeways of contemporary travel, constituted through text instructing theuser how the space should be used. The idea of non-places is similar to thenotion of hyperspace proposed by Jameson (cited in Kearney, 1996: 118),premised on the deterritorialization of place, where particular kinds ofplaces as sites for the constitution of particular kinds of relations are freedfrom geographic location.

Both non-places and hyperspace are real places, in that they are terri-torially situated. Their existence as such is relationally determined whetherone is working there or passing through. For Aug they are not home butthey are not abroad either. They are the others space, without the othersin it, space constituted as spectacle (1998: 106). As non-places are consti-tuted though social relations, not location, the constitution of others asrepresentations and as strangers eliminates the possibility of dialogue ordebate. Strangers stare at each other as spectacles across a vacuous chasm,defined by the impossible nature of dialogue and encounter (1998: 107).While some of Augs wider theoretical claims about non-places as charac-teristic of what he calls supermodernity are empirically and theoretically

problematic, the notion of non-place can provide insights for thinkingabout development spaces like projects and workshops. Developmentspaces are similar to Augs non-places in that they are constituted throughsocial relations, texts and deterritorialization in two senses. First, in thatglobal templates are imposed on local realities in a kind of franchise ofsocial and economic policies, creating a kind of hyperpolicy for the globalnon-places that recipients of development transfers become. Second, inthat those coming together for a project only do so through the projectitself, and represent not themselves but a professional position for anagency which is explicitly operating not only outside its place of origin but

to agenda determined in the national and multilateral policy spaces wherethe content of international development is negotiated.

Agency staff do not often engage with locations outside the non-placesof the projects which they are hired to implement, living their lives not ascosmopolitans partaking of hybrid cultures but as metropolitan localsseeking home plus27 (Hannerz, 1996: 107) in the shopping malls, gymsand cappuccino bars that international development assistance produces asa side-effect of massive spending in the capital cities of poor countries. Thetrue scale and socio-economic consequences of this are unknown but someindication of the enormity of developments side-effects on nationaleconomies is provided by the example of Kenya. Of the US$350 million thatthe United Nations alone brought into the Kenyan economy in 1998, lessthan half comes from direct programme assistance. The rest accrues from

139




19/22

myriad spin-offs from transport, salaries, international conferences andsecondary employment (Turner, 2000: 5). UN agencies in Kenya spent$130 million on the salaries of around 2000 staff, of whom 922 were

international.28 Kenya now ranks among the most unequal countries in theworld in terms of per capita income. The detachment of agency staff isshared to an extent by all workshop participants in the deterritorializedsetting of workshops as project space, held in locations detached fromeveryday social processes, involving strangers and divorced from localpolitical processes. Being in a non-place serves, I think, to make the man-ageable realities of projects seem plausible and amenable to control thoughthe global tools of the development profession in policy hyperspace.

The poverty of management and the management of poverty

Development management entails the social constitution of projects asslices of manageable reality in which, in the short term, outputs can beachieved. Development technicians are not mere bureaucrats perpetuatinghegemonic representations of the Third World through projects that per-petuate poverty. Stereotypical representations of the targets of develop-ment assistance are invoked in project documentation as a justification for,rather than as a consequence of, development spending. Designing

development interventions is a contested process involving groups ofpeople representing different social positions at different points in aproject cycle. The facts that participation has little impact on developmentoutcomes and that outputs have a weak relationship with impacts are partlyexplained by the limitations of the management paradigm for dealing withthe social where the relationships between development spending andsocial outcomes are not well understood. As long as development is con-ceived primarily in terms of managing finance, rather than supportingchange, vast sums of public money will continue to be wasted and develop-ment spending will continue to finance management and its associated

spin-offs.

Notes

This paper was presented at the Association of Social Anthropologists Conferenceat Sussex in 2001. Thanks to John Gledhill, Richard Sherrington, Clare Fergusonand Ray Abrahams for their helpful comments on aspects of this paper.

1 For a list of the targets see Development Assistance Committee (1996). Thetargets have since been superseded by the Millennium Development Goals.

2 After Saids notion of orientalist, which entails representing the other andconstituting them as a project, in need of intervention (1995: 7393, but muchof the book has resonance with the argument of this article).

140




20/22

3 This incorporation has been deliberately sought on the part of the multi-laterals. The World Bank, in particular, has sought to include representativesof protest movements in recent talks about development priorities.

4 As well as transitional, in the same way as the category peasant was transitionalbetween primitive and modern and underdeveloped and developed, accordingto Kearneys own critique (1996). The point about post-peasants surely is thatthey occupy a space between the indigenous person and the non-peasant,continuing to carry reference to their peasant heritage in the externallyimposed tag.

5 Note how anthropology differentiates between applied that is, developmentanthropology in the USA and anthropology, while in the UK anthropologistsworking in development tend to do so as policy analysts in the emerging domainof social development (see Green, 2001).

6 The same argument applies of course to colonialism. Considered thus thequestion becomes not whether anthropology was a handmaiden of colonial-ism, but the extent to which the interests of colonialism and anthropologydiverged or converged during and after the colonial period.

7 There are exceptions, such as McMillans excellent (1995) account of thecampaign against river blindness in the Sahel, and Crewe and Harrison (1998).

8 Such as academics and anthropologists!9 Gardner and Lewis challenge this representation in an article in which they

point to the contested nature of development discourse, even within develop-ment agencies (2000). Similarly, Phillips and Edwards (2000) show howdevelopment is negotiated at project level through the strategies of evaluatorsand evaluated at reviews and appraisals.

10 Both accounts are based on fieldwork in the early to mid 1980s.11 I have been involved in ethnographic research in Tanzania since 1989.12 For an account of signing off in relation to delegated responsibility within

public sector agencies see Harper (2000).13 Tanzania received US$1 billion annually in aid throughout the 1990s and 64

million from the UK alone in 1999/ 2000 (Economist, 24 February 2001).14 And internationally. Similar techniques are used in other countries in Africa

and in Asia, as well as in community development settings in, for example, theUK.

15 Workshop in Tanzanian Kiswahili is warsha and stakeholders, a comparativelyrecent word I am told, washikadau, which could refer to, according to adevelopment colleague, holding on to a boat (from shika, to hold on to, anddau, the Arab dhows or sailing vessels which ply the East African coast). RayAbrahams suggests that the phrase may, more accurately, refer to those with ajoint stake in the ownership of a vessel.

16 For a critical account of the New Public Management approach in a develop-ment context see Minogue et al. (1998).

17 For examples of the project cycle approach see ODA (1995).18 That is, a review midway through the life of a project.19 See, for example, the UKs Department for International Developments

Technical Notes on Stakeholder Participation (www/ dfid.gov.uk).20 Per diem culture also extends to staff of donor agencies, who receive allowances

for overnight stays and out-of-office work, some of which are substantial.21 For a similar perspective on workshops in Tanzania see Pottier (1997).22 In Kenya and Tanzania they are usually middle-aged male professionals.23 Right in the sense of conforming to what counts as current best practice within

141




21/22

development thinking and evidence based (where the selection of evidencetends to be determined by policy priorities).

24 For example through uprooting the problem tree.

25 Ferguson makes a similar point in relation to the limitations of developmentthinking, but he does not connect this to the construction of development asmanagement (1990: 260).

26 This assertion is based on personal experience. The only time when knowledgeof Tanzania mattered to my job as a development manager was when I couldproduce the documents myself, which would have otherwise been contractedout to consultants.

27 For example British emigrants to Spain may be seeking home plus sun.28 The figures given in theFinancial Timesof 16 January 2000 state that:

The UN employs 1291 national and 922 international staff excluding short-term contracts at a salary cost of more than $130 m. Almost $6 m of that

is distributed via secondary employment of UN house staff, includingservants, gardeners, guards, and drivers. (Turner, 2000: 5)

References

Arce, A. (2000) Creating or Regulating Development: Representing Modernitiesthrough Language and Discourse, in A. Arce and N. Long (eds) Anthropology,Development and Modernities: Exploring Discourses, Counter-tendencies and Violence,pp. 3251. London: Routledge.

Arce, A. and N. Long (eds) (2000) Anthropology, Development and Modernities:Exploring Discourses, Counter-tendencies and Violence. London: Routledge.

Aug, M. (1995) Non-Places: Introduction to an Anthropology of Supermodernity. London:Verso.

Aug, M. (1998) A Sensefor theOther: TheTimeliness and Relevanceof Anthropology.Stanford, CA: Stanford University Press.

Bayart, F. (1993)TheStatein Africa: ThePolitics of theBelly. London: Longman.Crewe, E. and E. Harrison (1998) WhoseDevelopment? An Ethnography of Aid. London:

Zed.Crush, J. (ed.) (1995) Power of Development. London: Routledge.Development Assistance Committee (DAC) (1996) Shaping the21st Century: The

Contribution of Development Co-operation. Paris: OECD.

Douglas, M. (1986) How Institutions Think. New York: Syracuse University Press.Escobar, A. (1995) Encountering Development: TheMaking and Unmaking of theThird

World. Princeton, NJ: Princeton University Press.Ferguson, J. (1990)TheAnti-Politics Machine: Development, Depoliticization and Bureau-

cratic Power in Lesotho. Cambridge: Cambridge University Press.Gardner, K. and D. Lewis (2000) Dominant Paradigms Overturned or Business as

Usual? Development Discourse and the White Paper on International Develop-ment, Critiqueof Anthropology20(1): 1529.

Green, M. (2000) Participatory Development and the Appropriation of Agency inSouthern Tanzania, Critiqueof Anthropology20(1): 6789.

Green, M. (2001) Social Development: Issues and Approaches, in U. Kothari and

M. Minogue (eds) Development Theory and Practice: Critical Perspectives, pp. 5270.Basingstoke: Palgrave.

Grillo, R. and R.L. Stirrat (1997) Discourses of Development. Oxford: Berg.

142




22/22

Hannerz, U. (1996) Transnational Connections: Culture, People, Places. London:Routledge.

Harper, R. (2000) The Social Organization of the IMFs Mission Work: An Examin-

ation of International Auditing, in M. Strathern (ed.) Audit Cultures: Anthropo-logical Studies in Accountability, Ethics and the Academy, pp. 2353. London:Routledge.

Haugerud, A. (1995)TheCultureof Politics in Modern Kenya. Cambridge: CambridgeUniversity Press.

Herzfeld, M. (1993)TheSocial Production of Indifference: Exploring theSymbolic Roots ofWestern Bureaucracy. Chicago: University of Chicago Press.

Hobart, M. (ed.) (1993) An Anthropological Critique of Development. TheGrowth ofIgnorance. London: Routledge.

Kearney, M. (1996) Reconceptualizing thePeasantry: Anthropology in Global Perspective.Boulder, CO: Westview Press.

McMillan, D.E. (1995) Sahel Visions: Planned Settlement and River Blindness Control inBurkina Faso. Tucson: University of Arizona Press.

Minogue, M., C. Poldiano and D. Hulme (eds) (1998) Beyond theNew Public Manage-ment. Changing Ideas and Practices in Governance. Cheltenham: Edward Elgar.

Overseas Development Agency (ODA) (1995) Guideto Social Analysis for Projects inDeveloping Countries. London: HMSO.

Phillips, S. and M. Edwards (2000) Development, Impact Assessment and the PraiseCulture, Critiqueof Anthropology20(1): 4767.

Pigg, S.L. (1992) Inventing a Social Category through Place: Social Representationsand Development in Nepal, ComparativeStudies in Society and History34(3):491513.

Pottier, J. (1997) Towards an Ethnography of Participatory Appraisal andResearch, in R. Grillo and R.L. Stirrat (eds) Discourses of Development, pp.20328. Oxford: Berg.

Said, E. (1995) Orientalism. Western Conceptions of theOrient. London: Penguin.Stirrat, R.L. (2000) Cultures of Consultancy, Critiqueof Anthropology20(1): 3147.Strathern, M. (2000) New Accountabilities: Anthropological Studies in Audit, Ethics

and the Academy, in M. Strathern (ed.) Audit Cultures. Anthropological Studies inAccountability, Ethics and theAcademy, pp. 118. London: Routledge.

Thomas, N. (1994) Colonialisms Culture: Anthropology, Travel and Government. Oxford:Polity.

Turner, M. (2000) UN Presence Keeps Kenyan Economy Afloat, Financial Times,London, 16 January.

Maia Green teaches anthropology at the University of Manchester. Her researchcovers such topics as participation in development, the politics of witchcraft sup-pression practices, and gender in east Africa. She is the author ofPriests, Witches andPower: Popular Christianity after Mission in Southern Tanzania (Cambridge UniversityPress, 2003). Address: Graduate School of Social Sciences, University of Manchester,Williamson Building, Oxford Road, Manchester M13 9PL, UK. [email: [email protected]]

143


globalizing development in tanzania

Documents