PLANotes

That homogeneous, egalitarian communities do not really exist, except in the imagination, is not a new insight. The challenge for practitioners of Reflect, as for other participatory approaches, is how to build common ground out of diverse realities. In an urban context, this can be especially difficult. A look at Reflect Projects in Kampala, Uganda and in Dhaka, Bangladesh will highlight some of the problems and also suggest ways forward. The key issue, we argue, is enabling facilitators to respond to the informal, improvised survival strategies of the urban poor.

These two areas share some features that complicate the use of participatory approaches.^[1] Neighbourhoods in Dhaka and Kampala are not separate units with clear physical or social boundaries: city-dwellers’ most important social ties (work, family, friends) often stretch far outside the area where they live. There are strongly marked cultural and economic differences between people living in the same neighbourhood. People may be moving so frequently that their knowledge of their neighbours and immediate surroundings may be limited. Finally, in many rural areas, people’s imagined notion of community incorporates geographical boundaries with strong moral imperatives of reciprocity, trust and solidarity; but in urban areas this idea of ‘belonging’ to a place may be weak, contested or altogether absent.

In Kampala, Reflect has been adopted by a community-based organisation, the Banda Community Development Project (BCDP), with funding from ActionAid. Banda is an informal settlement in a swampy area on the outskirts of the city, with a population of about 10,000. Like most informal settlements in African cities, Banda is made up of migrants from all over the country. Although most people are poor, inequalities among residents are marked. While some people are refugees from the war-torn North, with little or no security of tenure, other long-established residents are teachers, shop-owners and may be property-owners.

In Dhaka, Reflect is directly implemented by ActionAid, together with a micro-enterprise scheme targeted at very poor women, in Mohammedpur thana (district). Mohammedpur has a population of roughly 100,000 people and about half of these are living in squatter areas where eviction is a constant threat. ActionAid concentrates its efforts on Tikkapara, one of 12 squatter bastis (slums) in Mohammedpur. Most of the women who join belong to at least one NGO savings and credit programme (though not always ActionAid’s). Again, as in Banda, the inhabitants of the bastis come to Dhaka from all over and for very different reasons.

While knowing that urban neighbourhoods in Kampala and Dhaka are not villages, the first instinct of many development workers is to try to treat them as if they were, organising activities around wards or other geographical/administrative divisions and trying to generate shared interests from shared residence. Often this results in frustration with ‘ungrateful’ residents who apparently refuse to co-operate or in projects which residents tolerate but never actively make their own.

This was the approach first adopted both in the Banda and Tikkapara Reflect projects, but signs that it wasn’t working came early on, in the form of problems with meeting places, timings and attendance. Beyond the difficulties of finding a private space to meet in a crowded, land-scarce informal settlement, there was also the fact that not everyone wanted to attend a circle near to their home. In Banda, for example, some women attended a circle near their work but lived elsewhere; others thought it more prestigious to learn from the centre attached to the BCDP office; and again others preferred going to a circle where neighbours and friends could not see them. In addition, urban participants have many different occupations, and these jobs tend to be informal and thus unpredictable. This makes it hard to schedule meetings at times when all members can attend. In Tikkapara, women were prevented from attending by evictions and other crises stemming from their insecure foothold in the city; they also were less likely than rural women to have kin or friends nearby whom they could call on for childcare or other domestic help.

These logistical problems might have given warning that the geographically bounded community was a model ill-suited to Banda and Tikkapara but the more serious difficulties emerged in the application of Reflect units. The temptation in urban contexts is simply to adapt the PRA tools and topics that have worked well in rural settings, without rethinking the rationale behind the techniques. For example, the ‘village map’ is enduringly popular, probably because it allows participants to take a familiar reality and view it from a perspective that makes it possible to imagine change and plan action. In urban contexts, it is often ‘translated’ into a household or social resources map. However, if the presumption of an intimately shared reality is missing, the exercise loses much of its point. In Banda, when geographical maps were attempted, the fact that members of any one circle were drawn from all over the settlement made it impossible to produce something that was meaningful to all participants.^[2] In Tikkapara, participants from different neighbourhoods took great interest in the production of detailed maps (perhaps because Tikkapara is less spread out than Banda), but these maps could not easily lead to any kind of joint action.

Another example is the commonly used preference ranking of crops, which works well in settings where most people get their living from tilling plots of similar size and quality, and where a strict set of constraints (soil conditions, rainfall, available technology, etc.) requires each to choose what to grow out of the same limited set of options. In other words, it is a tool for analysing the available options for responding to the very tight constraints which all participants face. In Banda, the ranking of crops became a preference ranking of small business activities, and other units were introduced to help participants apply for jobs and understand their contractual rights as employees.

This was not entirely off the mark: for many women, as in Tikkapara, making money is a huge concern, and many pinned their hopes on Reflect to help them. However, in one circle, it turned out that none of the women had much prospect of starting any sort of business or getting jobs in the formal sector, in part due to their husbands’ determined opposition. Discussions of power relations and economic decision-making within the household would have been more relevant than analysis of small business options or job applications.

With these lessons in mind, BCDP is now moving towards organising circles on the basis of occupation, so that there will be groups of market women, beer brewers, stone-crushers or housewives (the latter will probably be organised on the basis of neighbourhood). This is expected to solve some of the logistical problems mentioned above, but more importantly, circle members (like peasant livestock farmers or maize cultivators) will share a common body of skill and knowledge and will have experienced the same practical dilemmas as they try to make their living.

This makes it possible to base the Reflect curriculum on, for example, income-generating activities, starting with existing practices (including timelines showing different activities that each participant is engaged in during one year/over a lifetime) and then deepening the analysis concerning the origin of raw materials used (maps), barriers to entering or expanding the market such as transport and credit (matrices analysing possible solutions), ways of processing or adding value (flow charts), and markets for the finished product (calendars showing seasonality of demand).

It remains to be seen whether, in the course of analysing common practices, these occupational groups will also discover common interests which they decide to pursue through joint action. Some occupations, such as beer-brewing, already have a strong collective orientation; others, such as stone-crushing, are organised in ways that produce fierce competition.

In Tikkapara it would be more difficult to segment groups by occupation (see Nessa et al., this issue), as most women pursue multiple small-scale income-generating strategies rather than a single occupation (and face strong cultural pressure to downplay these activities, defining themselves instead as ‘just housewives’). In this case, however, the common experiences and constraints are supplied by the women’s interactions with ActionAid and other NGOs active in the area, since most of them are regularly paying money into savings and credit groups managed by NGOs. Strict rules and procedures imposed by the NGO govern women’s access to loans (passbooks, regular meetings with a credit officer, rules about interest, repayment, etc.). Circle members have a common interest in trying to bend these rules where they can; they also face a common penalty if the NGO judges that one or more of them has defaulted. Thus, it makes sense to use Reflect to explore all aspects of loan use, incorporating the analysis of income-generating strategies suggested above, but also engaging in an open dialogue about the NGOs’ role in managing the women’s money, and exploring the feasibility of changing rules that women find unfavourable to them (e.g. increasing the size of loans or relaxing the criteria of acceptable loan use), as well as what could be done to address other constraints on women’s economic independence.

In this case, interventions by powerful outside actors (NGOs) have created interest groups where none ‘naturally’ existed before. In other contexts, Reflect circles might form around other external interventions, such as a public works scheme, a slum resettlement initiative, or a feeding programme. Units could address key features of the intervention that directly affect participants and help them to determine whether benefits are actually flowing to intended ‘targets’. This could develop as a shared learning process, with the implementing agency participating in the Reflect process together with literacy participants.

Another aspect of the model of ‘community’ that informs so much development work is the assumption that a fairly stable set of shared, consensual values and norms governs people’s activities within that community. Participatory approaches, such as PRA, are sometimes used in a way that illuminates the received wisdom on how people think they should act (norms and ideologies) more than how they actually do act (practices). But in urban slums, many people are in situations where cultural rules are muddled, patchwork affairs: either because practice is changing faster than ideology or because cultural differences bring competing ideologies up against one another.

To enable participants to analyse the real constraints that they face, PRA exercises in urban Reflect programmes need to start from the ‘real-life’ practices that participants feel comfortable discussing. Of course, norms and values (such as gender ideologies, for example) need to be brought into this analysis, as they themselves are among the most powerful constraints shaping what is possible and what is desirable. However, seeing them as contested and changing rather than consensual and timeless is essential to the critical dimension that Reflect tries to introduce, and this can best be done by starting from practice.

For example, most Tikkapara women will agree that dowry is a bad thing. But it is only through exploring why dowry is given in particular real-life cases that reasons for the continuing escalation of this practice can be understood, or possible strategies generated for mitigating its negative effects on women. The box below highlights the need to respond more creatively to the reality of survival in the city, which demands temporary, flexible, unofficial (even illegal) arrangements.

How can urban Reflect programmes generate units which help participants and facilitators mine the richness of real, informal practices? We have already argued that it is not enough just to substitute ‘urban’ themes for ‘rural’ ones: a problem situation needs to be defined that involves a tight set of constraints shared by all participants, and a limited set of resources for solving the problem. Focusing on changes in these constraints and in people’s response to them is one recommendation. A focus on change over time may be an ‘entry point’ for facilitators. Both in Banda and in Tikkapara, participants point out significant changes in relations between men and women: changes in household decision-making, in mental and physical violence experienced by women, in gender workloads, in men’s and women’s access to information or participation in social networks, in dowry or bridewealth customs.

BOX

FROM NORMS TO PRACTICES

In Tikkapara, some Reflect circles had discussed the rights of women within marriage. And rightly so, because Tikkapara women are at the forefront of gender change in Bangladesh. Many of them not only work outside the home, in violation of purdah (seclusion) norms, but they are also beginning to demand a greater role in family decisions. This means they are also frequently subject to physical and emotional violence by men resisting these changes. Yet the focus in Reflect circles was on mapping the official channels (the police and the marriage registry office) available to women for seeking protection. However, interviewed participants pointed out that these institutions were actually of little practical use to powerless people like themselves. It is through informal strategies (including enrolment in literacy classes) that many of these women are gaining a surprising degree of independence. However, people do not have free range to invent new strategies; they must use the limited material and cultural resources available to them in their existing social location (gender, class, age, etc.). Exercises to explore these strategies, which might also look at how these have changed over time, strengths and weaknesses, how they could be augmented, which tactics are more effective and why, would probably have been more interesting and useful to participants. These exercises can start off from less controversial and more obvious shared interests at the beginning of the course and move on from there.

(see also Nessa and Jellema,)

Health is another area where participants’ shared anxieties could lead to rich discussions, because concerns about the body, disease and reproduction often symbolise deeper concerns about social change and gender conflict. The ownership and development of slum land is a third area where long-term social and economic changes as well as short-term political interventions have had direct effects on Reflect participants. Units could explore layers of ownership and tenancy, the legal and de facto rights that people have in land and how disputes have been settled.

In designing units with clear objectives like those above, the diversity of participants’ experience and the ‘muddiness’ of cultural norms around these issues becomes an advantage rather than a handicap. The more different perspectives circle members bring to bear, the easier it is to understand practices (such as dowry customs, property rights or gender roles) as human inventions that change over time in line with changing power relations. In some situations, it might even be useful to make learners’ own categories of ethnicity, status, etc. an explicit axis of comparison. However, in this case, great care would be needed to avoid simply reinforcing ethnic and other stereotypes.

Making urban Reflect programmes responsive to urban realities will also demand changes in the training and management of facilitators (see Gautam and Cottingham). Training facilitators to use pre-designed Reflect units and standardised PRA tools encourages a preoccupation with the exact application of the units already learned and inhibits people from manipulating exercises to suit the interests and circumstances of the members of their circle. Instead, trainers should introduce, one at a time, the basic ideas of maps, matrices and diagrams (in a ‘real’ rather than a simulated context) and then focus on learning to adapt these to the needs of specific groups.

Changes in teaching and learning methods may also be needed to accommodate the wide range of skill levels and aspirations commonly found in urban programmes. In Banda, for example, many of the participants knew how to read and write, to some extent, before starting the literacy programmes. This has two implications.

First, the usual procedure of constructing graphics with local materials, then picture symbols (only later introducing words), frustrates many participants who already have some literacy. They need others to perceive their activities as serious, ‘academic’ education, and in the highly public context of Banda, picture-drawing is embarrassing. Instead, all graphics should be constructed with whole words from the beginning, only using pictures as backup for those who have trouble remembering the words (even maps should use lots of words, complemented only by simple pictures and lines). Once people get good at recognising the words from the graphics (exchanging their own personal techniques for recalling words, e.g. by shape, length, first letters etc.), then they should quickly move on to writing and constructing phrases.

Second, BCDP has realised that it must accommodate the strong desire of many participants to learn to speak (and to a lesser extent write) English. The pull of the dominant language is likely to be strong in urban settings, where its influence is pervasive and the disadvantage faced by those who don’t speak it is most blatant. As with other elements of an urban literacy programme, there is a temptation to take a purely functional approach to teaching the dominant language, concentrating on the practical needs of city-dwellers. The Banda programme will try to meet these needs by introducing some labelling of graphics in English (in some circles all labelling will be bilingual) and devising oral exercises on the basis of the finished graphics, loosely following a curriculum introducing important basic grammar concepts. At least some of the units will be chosen so as to specifically address situations in which participants want to make use of English.

However, BCDP also sees a need to go beyond a purely functional approach, using language to open up questions of power and identity (see Archer). A language matrix proved a very useful tool in developing a creative approach to English in Banda. Although people began by stating that they ‘needed’ English for practical purposes, such as going to the bank or the clinic, discussions based on the matrix showed that English was actually not used much in these situations, but that it was nonetheless significant. Further discussions led people to conclude that the main ‘use’ of English is actually to ensure and legitimate privileges. Hence Reflect units on the uses of English in the particular situations named by participants will also have to analyse some of the issues around English as the language of (colonial and postcolonial) power.^[3]

As long as development workers are tied to the model of the bounded, homogeneous community operating according to consensual norms, the realities of urban survival for the poor will enter into programme design only in a negative way, as ‘problems’ for programme staff. However, Reflect need not be based on such a model. As a flexible approach which builds on people’s specific experience of their own reality, it easily accommodates changes in project design in order to make the most of diverse, ‘messy’ urban contexts. Many of these changes are still at the experimental stage, but we are hopeful that they will make the urban Reflect programmes an exciting source of new innovation and insights for practitioners in the growing number of contexts (rural as well as urban) where the notion of community is becoming problematic.

^[1]Although these characteristics may seem typically urban, they are also found outside of cities: see Archer and Jellema, ‘Response to Dyer and Choksi’, Compare 28(1): 88-92 (1998) on an experiment with literacy among pastoralists in northern India.

^[2]In this case it might have been that the scale of the maps was too small - an artificial attempt to fence off a bounded ‘community’ - rather than too large. A meaningful map might have included Kampala

^[3] See Fiedrich, ‘Learning English: An uneasy flirtation with the powerful?’, Education Action 7 (January 1997) available from the current editor.