Sharing Design Agency with Local Partners in Participatory Design

ORIGINAL ARTICLE Sharing Design Agency with Local Partners in Participatory Design Chiara Del Gaudio 1,*, Carlo Franzato 1, and Alfredo J. de Oliveira 2 1 2 Universidade do Vale do Rio dos Sinos (UNISINOS), Porto Alegre, Brazil Pontifícia Universidade Católica do Rio de Janeiro (PUC-Rio), Rio de Janeiro, Brazil Partnerships with local actors are quite common in social design projects developed in unknown contexts. Several design researchers describe them as positive and supportive elements for project development. However, setting up a partnership may bring several unexpected challenges to the designer’s agency and have strong implications on the design process—with a level of inluence that changes according to contextual conditions too. This paper aims to point out and discuss them. In order to do this, it explores what a local design partner could mean for the designer and the design process, by describing and analysing action research undertaken on a participatory design project conducted in a Brazilian favela in partnership with a local NGO. Three under-discussed issues about local partnerships emerged and are examined through partnership, power, delegation, and agency theory. Lastly, ive strategies to deal with them and to strengthen the designer’s practice are presented. Keywords – Agency, NGO, Participatory Design, Partnership, Slum. Relevance to Design Practice – This paper presents the inluence and risks related to having local partners for the designer’s agency and for project development in participatory design projects, with a special focus on those undertaken in poor, undemocratic, marginalized, and culturally distant contexts. Weak points of the present literature are discussed and suggestions to deal with them are given. Citation: Del Gaudio, C., Franzato, C., & Oliveira, A. (2016). Sharing design agency with local partners in participatory design. International Journal of Design, 10(1), 53-64. One of the fruitful paths that has emerged is to develop participatory design (PD) processes. Actually, in PD the designer opens the process up to users starting from the initial phases of the project (Bjerknes, Ehn, & Kyng, 1987), sharing the decision-making power usually exercised by whoever leads the process and plans the project. This brings great change for users: they participate in all the implemented activities and express their ideas—that is, they gain the power of inluencing the design process and they are empowered by it. Consequently, when used with local groups of people or communities, through participation, the PD approach has the potentiality of fostering more democratic dynamics and attitudes, and the redistribution of power both within the design process and within society too. Changing the design process becomes the irst step towards democratizing society. However, this change in process changes the designer’s agency too. If broadly speaking we can consider the designer’s agency as his or her capacity and possibility of action, in a PD project with a local community, it relates not only to technical project design tasks, but also to his or her capacity and Introduction Over the last 50 years a critical view of design practice has become widespread among designers, both relating to the consequences of its actual application and to the possibilities of playing a different role within society. Recently, due to the pressure of societal issues and to the shared understanding of the relevance of moving towards a more egalitarian society, the interest in applying design skills to improve people’s life and to foster more democratic dynamics and contexts has increased. Designers seeking to act in this way—which means within the so-called social design ield— can choose from among a wide range of different possibilities of action: from the situation to tackle, to the characteristics of the territory, to the approach used, and to the speciic design object and aim. They can target their actions towards social inclusion or social innovation, towards healing problems or towards solving them. In addition, they can work in their context of origin and in already democratic areas, or otherwise. Even if all these possibilities seem worthy of attention, the importance of the designer acting with communities in poor, undemocratic, and marginalized contexts has often been stressed. In such contexts, a social design practice can contribute to resolving social issues and to improving the everyday-life of the local population. Ultimately, the improvement of the social conditions of these areas can be a part of a wider process of integrated development and change in these territories. However, due to the designer’s unfamiliarity with these contexts—often culturally and physically distant from him or her—and the newness of the practice itself, one must understand how designers can design there. www.ijdesign.org Received Sep. 11, 2015; Accepted Feb. 05, 2016; Published April 30, 2015. Copyright: © 2016 Del Gaudio, Franzato, & Oliveira. Copyright for this article is retained by the authors, with irst publication rights granted to the International Journal of Design. All journal content, except where otherwise noted, is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 2.5 License. By virtue of their appearance in this open-access journal, articles are free to use, with proper attribution, in educational and other non-commercial settings. *Corresponding Author: [email protected]. 53 International Journal of Design Vol. 10 No. 1 2016 Sharing Design Agency with Local Partners in Participatory Design Actually, the strategy of setting up a partnership with local actors—among which there are different kinds of organizations too, such as local public administrations, associations, and NGOs—implies power delegation and, in turn, delegating power inluences actors’ agency both positively and negatively— consequently this means that a partnership may increase or reduce the designer’s agency. However, whereas the present literature points out the positive aspects of local partnerships, it does not really detail the overall inluence on the design process, such as their unexpected and intricate aspects. What is the inluence of having a local design partner on the design process and on the designer’s agency? Which are the challenges? And what additional issues relating to local partnerships have to be considered when working on PD projects in conlict-affected and fragile urban contexts? Thus, the main aim of this paper is to discuss agency sharing within a partnership, speciically the implications for the designer’s agency associated with having local design partners in PD projects, and on the strategies to deal with them. This is an interdisciplinary domain. First of all, design projects with and within communities and territories are interdisciplinary in nature: they require bringing together contributions from different ields since their issues cross disciplinary borders (Silva, Morais, & Rubenilson, 2009). Moreover, in the case of setting up a local partnership, this characteristic intensiies as the PD project is conducted by actors belonging to different institutional environments (Dille & Söderlund, 2011). Due to this coniguration, they can be deined as inter-institutional projects, or even inter-organizational ones if one includes all the types of organizations operating with or within a community, as well as informal ones. For these reasons, in order to achieve our aim, in addition to design references, we are also going to use authors from social sciences and organizational theory for deining and understanding the concepts of agency, partnership, delegation, and power. This will allow a deeper understanding and the possibility of improving PD methods, as well as, more broadly, the work of professionals who operate within community projects and in critical contexts. Considering the presented purpose, after identifying the presented issues and deining our theoretical framework, we undertook action research on a PD project aimed at regenerating urban space in partnership with a local organization. Actually, due to the relevance of the contextual dimension in PD and of observing how design happens in the real world in design research, we needed a ield approach to design (Koskinen, Zimmerman, Binder, Redström, & Wensveen, 2011): to listen to people, and to follow and observe design unfolding in practice. Action research served this purpose: it allows theory and real life to be brought closer together, and knowledge to be produced through action (Fals Borda & Rahman, 1991; Brandão, 2005). The subsequent analysis of the project suggested paying attention to four aspects relating to designer’s agency and local partnerships: 1) power delegation; 2) indirect use of power; 3) collaboration and competition dynamics within the partnership; 4) and the partners’ interest in the suggested process. Analysing them based on the theory, we understood possibility of building relationships among the involved actors and of developing design processes that enable a more democratic development of the local situation. For all these reasons, recently approaches that improve everyday-life conditions through citizen participation, inclusion, and empowerment have gained importance and recognition when it comes to fostering more egalitarian dynamics, processes, and relationships. In this paper we will focus on their application in conlict-affected and fragile urban areas. Conlict-affected and fragile urban areas are what the UNDP (2012) deines as contexts that are chaotic and hyper politicized, whose inhabitants live in fragile social and economic conditions, that feature social divisions and are affected by social exclusion; where there is no public security, formal institutions, or basic infrastructure; and inally that experience struggles for power among actors with conlicting agendas. The interest in this kind of action has increasingly gained strength in the last 10-15 years, and several suggestions for the designer’s agency in these contexts have been given. In this regard, one of them is to set up partnerships with local actors. This is a strategy that, when it comes to design with and within communities and territories, allows combining designer’s skills and expertise with existing local resources and knowledge. Moreover, as emerges from several design projects undertaken by design researchers—such as Winschiers-Theophilus, Chivuno-Kuria, Kapuire, Bidwell, and Blake (2010)—it is a way to facilitate and achieve dialogue and integration with the context, and it becomes a priority and grows in relevance when the designer acts in an unknown, culturally different or critical territory. This strategy has been favourably embraced by design researchers. Manzini, Jégou, and Meroni (2009), for instance, have even extended this idea to building a local design network in order to foster not only synergies between the designer and the local partner’s knowledge and resources, but also among all the involved local actors. Despite the presented interest and contributions, designing with communities, and in a critical context, still presents several issues that have to be better explored (Correia & Yusop, 2008; Emilson, Seravalli, & Hillgren, 2011; Hussain, Sanders, & Steinert, 2012; Mulgan, 2014). This is exactly the case of having local partners, for instance. Chiara Del Gaudio is a designer and post-doctoral researcher in Design at UNISINOS (Universidade do vale do Rio dos Sinos, Porto Alegre, Brazil). She holds a Bachelor degree and an MSc in Design from Politecnico di Milano, and a PhD in Design from PUC-Rio (Pontifícia Universidade Católica do Rio de Janeiro, Rio de Janeiro, Brazil). Her research focuses on participatory design (PD) and social innovation. In this perspective, she investigates the main obstacles and the necessary conditions for the designer’s social action. Carlo Franzato is a designer, holds a PhD in Design from Politecnico di Milano, and Professor in Design at UNISINOS (Universidade do vale do Rio dos Sinos, Porto Alegre, Brazil). His research focuses on complex collaborative design networks developed by designers, other professionals, organisations and institutions, users, and citizens. Alfredo Jefferson de Oliveira is a Professor in Design at PUC-Rio (Pontifícia Universidade Católica do Rio de Janeiro, Rio de Janeiro, Brazil). His research focuses on design for environmental and social sustainability through a CoDesign approach. www.ijdesign.org 54 International Journal of Design Vol. 10 No. 1 2016 C. Del Gaudio, C. Franzato, and A. J. de Oliveira Enderud, 1981). The reasons are several: because of power being the indirect effect of others’ actions; because it can be originated by unknown actors; and because it can be the result of hidden agendas. This latter possibility is better explained by Bachrach and Baratz (1963) who stressed that power can be exercised by actors participating directly in decision-making situations, but also when actors focus their energies in activities that limit the others’ possibility of discussing and dealing with issues that may hinder their aims. Derkzen et al. (2008) afirmed that in participatory and collective decision-making dynamics it is easier for partners to exercise power over each other, and that this inluences power relations in several different ways that depend on each situation and on involved actors. Furthermore, it has already been observed that the processes that take place within partnerships rarely involve full inclusion and equal participation. All of this brings to light further understandings of partnerships within PD processes: when a designer establishes them it increases and facilitates the possibilities of one partner exercising power over the others; moreover, partnerships with local actors might not mean only and exclusively solidary collaborations but have other deeper and wider implications. One of the main reasons of this phenomenon is that a partnership is a situation in which a dynamic of delegation occurs. To delegate is to ask someone to fulil activities. Thus, when partners collaborate to achieve an objective, each of them will undertake some activities or play roles on the others’ behalf. As presented by Lupia (2001), delegation is quite a common situation—due to everyone being limited in resources, time and energies, and seeking to achieve better results—, and it usually happens because of synergies of resources and of surpassing of personal limits. It allows increasing of results both in terms of the amount of tasks accomplished and in terms of quality of the execution. However, delegation brings risks with it because it implies a power transfer too, as Lupia presented: that in this kind of project: a) the partner is a mediator, and this gives it agency and power over the process; b) the partnership is not only collaboration but also competition, and this requires negotiation skills; c) the partner is unfamiliar with the design process and activities, and this inluences its engagement, thus its collaboration. Subsequently, a more in-depth and literature-based relection on these understandings resulted in ive strategies for a designer dealing with partnerships’ implications: 1) design process explanation and training; 2) sharing and negotiating common interests; 3) developing a dialectic and dialogic process; 4) strengthening a supportive local network; 5) managing collaboration and competition. Finally, this paper is structured as follows: a review of the main features of the concepts of partnership, delegation, power, and agency; the description and discussion of the ield study, meaning the action research conducted in a favela—that is, a Brazilian slum—in Rio de Janeiro together with the description of the main issues brought to light; and inally, the results. Background A stakeholder is anyone who is affected by an organization or who can affect it. Thus, if a design project can be understood as an organization (Dille & Söderlund, 2011), all the actors of its design network are its stakeholders. The mutual inluence that exists among a stakeholder and an organization stresses the stakeholders’ interests in the project and their inluence, thus pointing out the relevance of involving them in the design process and of setting up different kinds of agreements and collaborations in order to implement it and manage their inluence. In this regard, one of the several possibilities is to set up a partnership with them. This is what happens and is suggested by some researchers for undertaking PD projects in poor, undemocratic, marginalized, and culturally distant contexts. A partnership is an arrangement that partners, which are the involved actors, set up to cooperate for the achievement of mutual interests. Despite the ostensible collaborative nature, according to Derkzen, Franklin, and Bock (2008), a partnership is not only a harmonic way for different actors to collaborate in a solidary manner, but it is also an instrument used for the unsymmetrical exercise of power. Actually, they deine partnerships as “arenas of power.” The possibility of exercising power—within and because of a partnership—is due to the close interaction that this kind of agreement enables among the partners and it has strong implications on the relationships itself. In this regard, Foucault (1982) himself understands power as a force that deines a relationship between partners and is deined by them. According to him, the partners inluence each other through their actions, and the actions are each one’s exercise of power over the other. Therefore, a partnership can be seen as an expression of the power that each actor exerts over the other, and it is deined by the power relationship that is established between them. However, not always is a partner able to understand when and how the other is exercising power over it. Power exercise may not be visible in origin and shaped by those who are inluenced by it (Borum & www.ijdesign.org While delegation allows lawmakers to beneit from the expertise and abilities of others, it can also be hazardous. The hazards arise from the fact that delegation entails a transfer of power. For example, every time lawmakers delegate to bureaucrats, they give away a portion of their authority to govern. (p.2) According to Lupia, when an actor entrusts another with implementing some activities, it grants to the other power over them and over all related ones. Nevertheless, entrusting and trustworthiness do not always come together, and if not, a situation of power abuse may occur and the former actor may cede more power than expected instead of increasing its potentialities through the other. Everything presented till now shows how partnerships entail two kinds of power-related risks that can affect them and their dynamics: the partners’ inluence on the other’s activities and situations facilitated by closeness and by the setting up of the relationship; and power abuse brought about by power delegation. These risks stand out for relevance because delegation is directly related to agency. 55 International Journal of Design Vol. 10 No. 1 2016 Sharing Design Agency with Local Partners in Participatory Design establish relationships with each other to exchange favours and share skills to achieve speciic results, the partnership will move constantly through this spectrum according to the above-presented dynamics. According to Sennett, to guide a partnership towards one of these speciic kinds, the principal has to apply negotiating skills and to be sensitive to the other. Sennett’s exchange spectrum shows us that collaboration and competition are not opposites in partnerships. If we consider the PD process setting, constant negotiation takes place between the involved actors in order to build and move towards a shared vision. This negotiation includes both collaboration and dialogic competition dynamics. The strong relationship between partnership, power, delegation, and agency, as discussed above, raises a question: if designers need to set up a partnership in PD projects—mostly in unknown contexts— how does this inluence their agency and the design process? We undertook action research to understand this issue better. Agency, at its simplest, is an actor’s capacity and possibility of action (Giddens, 1984). The direct relation between delegation and agency is explained by Foucault’s (1982) statement about the exercise of power: it “is not simply a relationship between partners, individual or collective, it is a way in which certain actions modify others” (p.788). Thus, an actor’s capacity and possibility of action is determined by the power exercised by other actors. This means that in a dynamic of delegation—that implies power’s transfer and exercise—whoever receives power can modify the possibilities of action of whoever delegates. Consequently, delegation may involve agency loss (Pollack, 1997). Agency reduction relating to partnership has a two-fold nature: it can be seen as agency sharing when it is planned, as well as agency loss when it goes beyond the transfer planned and it compromises an actor’s agency. According to agency theory’s researchers and their principal-agent models, within a delegation whoever delegates is the principal and whoever receives authority is the agent. Based on these models, agency loss is deined as the gap between the actual consequences of delegation for the principal and the best possible consequences that could happen (Lupia, 2001). In standard delegation dynamics, agents may have interests that differ from the principal’s ones (Da Conceição, 2010). This means that there is no agency loss when the agent acts according to the principal’s interests, but it increases together with the divergences of interests. Drawing from political sciences considering the political dimension of PD, a relection emerges: due to opening up the process, a design partnership in a PD process implies agency transfer and sharing among the partners. However, when the designer is not able to negotiate this sharing and it happens in a higher amount than planned, he or she loses possibilities of action, meaning agency. Thus, translating the previous statement into a design partnership in a PD process, there is no designer agency loss when the designer is aware of agency transfer and the design partner acts according to the project’s main aim, while it exists and increases when the designer is unable to manage it and the partner does not share it. According to Da Conceição, when the principal and the agent do not share the same interests in delegating power, this is like abdication. Method: Action Research on a PD Project The theoretical background has described the power dynamics that exist within a partnership and how they may inluence and reduce the agency of whomever is setting up the partnership itself—in our discussion, this is the designer. It strengthened our conviction about the need of providing enlightenment about the inluence and challenges—for the designer’s agency and the design process— related to having local design partners in PD projects, speciically in conlict-affected and fragile urban contexts. For this reason, we undertook action research on a PD project conducted in a Brazilian favela in partnership with a local NGO. Besides the previously presented relevance of a ield approach and of observing how design happens in the real world, action research was selected also because in Brazil and South America it is considered of great relevance for application in projects concerning social improvement and community emancipation (Thiollent, 1985). This is explained by the fact that, in this speciic geographical context, social sciences and education are the leading areas for the development of this approach. Lupia and McCubbins (1998) presented two situations that reduce agency loss. The irst one is when both partners share common interests—for instance the same design aim. The second situation is one in which the principal is aware of all the possible consequences relating to the agent’s actions. There are several strategies put into action to reduce agency loss and to inluence agents’ behaviour. For instance, McCubbins, Noll, and Weingast (1987) suggested that to reduce agency loss principals have to exert inluence over the actors that share their interests and that, at the same time, are able to inluence agents’ action. During action research, data was collected through a ield research diary, and semi-structured interviews about the project development and the design process, the collaboration with the involved actors, and the design partner. Speciically, the research diary consisted mostly of on-site observations and photos; on the other hand, regarding the interviews, two of the NGO’s directors and four of its employees were interviewed—considering the large amount of data, just the most relevant for this paper will be cited here. According to main principles of action research (Fals Borda & Rahaman, 1991), throughout the research, the researcher’s point of view is as important as those of the community of practice; all point of views have to be considered in project development and knowledge production. Thus, in the presented case, we tried to value the point of view of all the involved actors. However, considering that here the focus is on the implications of local partnership for the designer’s agency, we consequently paid special attention to the designer’s point of Power dynamics implied in a partnership and the principal’s effort to manage them point out that a relationship set up for collaborating may shift constantly between collaboration and competition. As explained by Sennett (2012), exchanges among actors may be of ive different types: altruistic exchange; win-win exchange; differentiating exchange; zero-sum exchange; and winner-takes-all exchange. Therefore, when partners www.ijdesign.org 56 International Journal of Design Vol. 10 No. 1 2016 C. Del Gaudio, C. Franzato, and A. J. de Oliveira view—this means to data contained in the ield research diary. Even if this choice is aligned with our focus, this could be in part a limit of the presented research. Nevertheless, it was not exclusive; the other actors’ points of view have also been considered through data contained in the undertaken semi-structured interviews, as presented in the data analysis section. In regard to analysis, after being collected, the data was divided into analysis units and encoded, and all data relating to local partnerships and the designer’s agency was grouped by semantic coherence. The analysis led to the identiication of four main categories: power delegation; indirect use of power; collaboration and competition dynamics within the partnership; and partners’ interest in the suggested process. Their deeper understanding led to our results. Finally, data analysis will be described in the following section. There are several consequences of this situation on local social fabric: citizens live in fear and insecurity, for instance. This, together with their unattended necessities, lead them not to fulil their duties and not to be interested in the community. An individualistic logic dominates and has weakened community feelings and attitudes (Souza e Silva, 2003, 2010). In order to challenge this situation, it is necessary to stimulate interest in common goods—like public space—and to develop the habit of acting cohesively as a community. Thus, the application of PD processes in such a context seemed to us to be a potential way to address this need. Lastly, Complexo de favelas da Maré is a cluster of 16 small favelas with a population of 130,000 inhabitants. Within the city, it is known for its size, high population density, infrastructural problems, and crime situation. At the time of the project, it was dominated by three local armed criminal groups that struggled for power with great inluence on local dynamics. The project took place in Nova Holanda—one of the sixteen favelas. Data Analysis A praça que nós queremos (“The square we want”) was a PD project targeting the regeneration of an urban context through the fostering of democratic dynamics in the community. It actually occurred in Complexo de favelas da Maré (“Maré slum complex”), and it involved local inhabitants and a local NGO in the collective redesign of a dilapidated public square. In this section, irstly we present the main features of the speciic context of action; secondly, the main phases of project development; then we present the four categories that emerged from data analysis through the description of four situations in which the design partnership impeded the progress of the project and limited the designer’s agency thus exemplifying them. Lastly, we present the main understandings obtained by way of the project. Development The methodological framework of the project was action research on a PD project, and the work was undertaken over eight months, from March to October 2012. It was carried out by a team made up of two designers with signiicant experience and knowledge about the local context and in social and participatory design, three senior design researchers, a junior designer, and a researcher. The researcher, who was the one involved in direct and everyday activities in the favela, met with the others at ad-hoc meetings set up to make decisions about the project. The designer’s presence on the ield was intensive, although it varied throughout the process, in line with the needs of the different phases. At the same time, the design team had a local partner that consisted of a local NGO. Analysis Unit Context The partner was selected from among others by the design team mostly on the basis of its previously developed projects that addressed the improvement of local people’s life through participatory processes. At the same time, the strong presence of local inhabitants among the NGO’s members reinforced the choice—they knew local dynamics and needs. The partnership was established before the beginning of the design project in meetings held to share knowledge and aims, and to deine the partnership main aspects. Actually, the design team and the NGO agreed that the partnership would be for the purpose of undertaking the PD project aimed at promoting community behaviours for the resolution of local life issues. They established that the design object would be determined together after the designer was irst immersed in the territory. Complexo de favelas da Maré is a Brazilian favela, speciically a Rio de Janeiro favela. When we talk about Rio de Janeiro favelas, we are referring to conlict-affected and fragile urban areas. Actually, even if Rio de Janeiro features a geography in which formal and informal urban areas are all mixed together—rich neighbourhoods, medium class ones, and favelas co-exist side by side—the latter territories suffer from a condition of inferiority and invisibility: their inhabitants lack citizenship since the most basic rights are not guaranteed. Social issues like social exclusion, inequity, lack of governmental investments in supplying basic services, criminality, and violence mark the several and many favelas of the city—there are 1094 (Instituto Brasileiro de Geograia e Estatística, 2010). This situation has been caused over time most of all by public repression and inconsistent urban and social policies. At the same time, years of government non-fulilment and absence has allowed criminal groups to grow in power there, to replace public institutions in satisfying people’s basic needs, and now they manage all local dynamics and rule the territory through the power of arms (Souza e Silva, 2004). Thus, at the present moment, the government does not exercise any political or juridical power there (Souza e Silva, 2011). www.ijdesign.org The project consisted of three main phases. During the irst phase, several activities were implemented by the designer to gather knowledge about the context, to become familiar with it, and for integration. This exploratory period lasted for two months, from March to May, and the main activities undertaken were: observation, participation in local activities—both the NGO’s routine and local events—, and unstructured interviews with NGO’s members and local people. 57 International Journal of Design Vol. 10 No. 1 2016 Sharing Design Agency with Local Partners in Participatory Design Afterwards, there was a second phase, in which the design object was selected and the main action strategies deined. The design object was decided in a joint meeting between the designer and one of the NGO’s directors, and consisted of a run-down local square (Figure 1). This meant that the PD process had to be directed towards redeining the use of the square, and towards improving the place through promoting and provoking active citizenship initiatives able to support local inhabitants’ wishes concerning it. This decision was based on a local need that both the NGO expressed—through its employees who were living there—and the designer detected: the need for free, pleasant, open-air public spaces where local people could spend time together. One of the reasons underlying this desire was the small and overcrowded home space. In the favela, people usually live in small houses and have big families. At the same time, broadly speaking, the Complexo was lacking in this kind of outside space. For this reason, the presence of a local run-down square was seen as a relevant possibility of action. The square was almost the only one within the large territorial expanse of the favela, but its conditions prevented many activities such as spending time together, socializing and relaxing, and children playing safely there. Actually, the situation was as follows: garbage on the loor, broken playground facilities (Figure 2), crumbling infrastructure, and lastly a growing number of market stalls within its limits (Figure 3). Moreover, this last element symbolized the inluence of criminal groups on the space: the stalls were built and sold by the local criminal group to people interested in opening a small business. Subsequent to the selection of the design object, the designer tried to set up co-creative meetings with some of NGO’s members to deine action strategies. Actually, the design team thought that encouraging people’s spontaneous participation had to be the irst step. These meetings were intended to generate ideas about how to raise people’s interest in participating in the square renovation. However, they did not happen as planned. Actually, the NGO suggested a different strategy: it proposed organizing a public meeting at which to present the square issue directly to local inhabitants and at which they could express their ideas and needs regarding the place. Vast experience had taught the NGO that this was the best way to involve people and to justify an intervention in a public space to other local actors—such as local associations and criminal organisations. Lastly, in the inal phase, two public meetings were held to promote PD activities related to the selected object. The irst meeting was held at the end of July and it consisted of an opening speech given by the NGO’s members and of an interactive activity with participants promoted by the designer in order to start co-creating (Figure 4). The second meeting occurred at the end of August. The designer tried to promote local inhabitants’ participation and to stimulate the expression of their ideas through some sketches based on people’s previous suggestions. Unfortunately, few inhabitants participated and those who did were mostly shop owners who came with a different aim. They were worried that the project could compromise their businesses and they wanted www.ijdesign.org Figure 1. View of the square. Figure 2. View of the square playground area. Figure 3. The stalls area. Figure 4. The irst meeting: the designer interacting with local people. 58 International Journal of Design Vol. 10 No. 1 2016 C. Del Gaudio, C. Franzato, and A. J. de Oliveira exemplify this point. He afirmed that the collaboration with the NGO was crucial for acting safely within the territory—every action would fall under the responsibility of the institution, but at the same time, for this reason every designer’s action depended on the NGO’s backing: to defend their territory. This was the last part of the project because the same NGO ended the meeting afirming that these initial representations made by the designer were the project’s inal output. Challenges Experienced While you are operating within the NGO you are relatively safe. Encoding and grouping data collected during the ield study allowed us to identify four categories that gave us insights into the implications of having a local partner. We selected four situations that could exemplify each one of the categories, and that could show how the established partnership inluenced the project and challenged the designer’s agency instead of enhancing it through synergy. They are presented below. Any problem will bounce back to the institution, not to you because you are acting inside the institution. You can stop people on the street, you can call them over them with a microphone, but you are doing so because of the NGO, It is not you trying to reform the square. Your arrival was mediated. [ ... ] But anyway you are limited to the power structure of the institution. You can pick up a microphone only if someone tells you to. You cannot take a microphone out of the drawer and go onto the street and make an announcement, unless you do this with your own microphone, Power Delegation then you have no relationship with the institution. (personal The irst one exempliies the power delegation category. We decided to establish the partnership both for the main reasons that are expressed by the presented literature and because of the project’s contextual conditions. First of all, we made the decision due to the local criminal situation: entering into and even operating within the favela were not a trivial issue. Local criminal organizations required informal permission in order to do so. By working with a local partner, that already had it, this was not necessary. At the same time, the designer was an outsider in the context and having a local partner—in the form of the NGO—with knowledge of it and its rules, could help actions development and activities implementation. Finally, in a place where basic infrastructure is lacking, having a local partner with well-structured headquarters could facilitate the project’s development. Despite all this, the decision of having a local partner had several unexpected implications in terms of the designer’s power and agency that can be better understood considering the process the designer suggested and the one that was carried out. On one hand, there was the designer’s interest in stimulating a local endogenous community process by raising people’s interest in the space irst, and then redesigning the square counting on their spontaneous participation. On the other hand, the partner’s culture revealed itself to be characterized by top-down processes. The result was that none of the implemented activities was actually participative. Even when they occurred—as in the meeting case shown by Figure 4—few people participated and they were not representative of the different types of the square’s users. During meetings, moreover, the designer’s leadership had to be very strong, since the NGO left no time to prepare participants for co-creation. This gap between the designer’s suggestions, the NGO’s suggestions, and the actually implemented activities clearly shows that NGO exercised more power than the designer. In this regard, why did the designer not reject the NGO’s decisions and go further with the original plan? Actually, besides the shared nature of the process, the designer could not act differently because the partnership was necessary both to operate and to be safe in the context. A declaration released by one of the NGO’s employees during the completed interviews can better www.ijdesign.org communication, 13th November, 2012) The presented situation shows that in the project the design partner was the designer’s mediator with the local context. This means that the designer delegated the mediation with local actors to it. This delegation implied the transfer of an amount of power over the design process from the designer to the NGO— meaning the designer’s agency too. When the partner takes on a mediation role with the local context on the designer’s behalf, it has more power than the designer on deciding which actions have to be implemented. Lastly, the designer’s agency is reduced and transferred to the partner to an extent that increases directly with the existence of social and safety issues. Indirect Use of Power The second situation serves as an example for the category of indirect use of power. As previously presented, the design object was the shared redesign of a dilapidated local public square. This topic was suggested by the NGO during the irst meetings with the designer. Even if it was suggested together with other topics, the NGO clearly expressed the relevance it had. During the irst phase of the project, the designer investigated the context, trying to understand its needs—observing and talking with people— and to verify the relevance of what was suggested by the NGO. Conversations with local inhabitants conirmed an unfulilled desire of enjoying the square, of spending time there with friends and families, and that the current conditions were preventing them. Thus, the designer agreed to go further with this topic both because of the information collected and because of the NGO’s interest. The strong interest the local partner had in the object could be useful for the project development because it could result in a stronger commitment and participation on the part of the design partner. However, throughout the process the NGO’s interest in the project revealed itself to be of a different nature. Actually, at the beginning, the NGO afirmed that it was interested in the collaborative renovation of the square because it was a way to give it back to the local population and to foster community feelings. 59 International Journal of Design Vol. 10 No. 1 2016 Sharing Design Agency with Local Partners in Participatory Design interest and relections—such as street art activities, temporary street exhibits, several co-creative dynamics, and ways to divulge the meeting—but the NGO’s prevented them, by claiming various reasons such as lack of funds, lack of time, and lack of people. However, the designer still believed in the importance of a different kind of process and decided to act during the meeting itself. In the succession of ideas suggestion and rejection, we noticed a competitive dynamic between the designer and the NGO. It was competition about who would succeed in applying the desired dynamic during the meeting. Actually, the meeting occurred following the NGO’s standard procedures: some of the NGO’s members conducted it and presented the square conditions and the importance of improving it. However, the designer had prepared—through negotiation with the NGO—visual interaction design tools to stimulate local inhabitant’s expression of their ideas and wishes. The tools were posters intended to provoke people’s interaction and to collect their ideas. They could have been the irst part of a collaborative process. While the designer thought space had been gained and the partnership had shifted more towards collaboration, at the end of the meeting, after the tools application, the NGO invited everyone to a second meeting where the designer would present the new square design based on the expressed ideas. The NGO continued with fostering a different kind of process. The presented situation shows how the partnership changed and shifted in nature: at some points, it was closer to collaboration, at others to competition. This did not happen in a linear way and both the partners experienced the different states. It agreed to conduct a PD process to pursue this aim. By reason of this initial agreement, the designer went forward with the project and the collaboration. However, as presented, during the process the partner never accepted the designer’s suggestions. Even if this attitude on the part of the NGO was initially justiied by its better knowledge of the context, a different interest emerged after the two public meetings were held. The NGO asked the designer to meet the local Secretary of Parks and Gardens to show him some of the sketches developed throughout the process. According to local inancing rules, this would probably be the irst step for the NGO to start a funding application process. Thus, based on a subsequent analysis of the experience and of the NGO’s behaviour, we thought that its real interest was different from the declared one, and that all throughout the project it was addressing it through its actions. In the NGO’s view getting funds to renovate the square was the best way to give it back to local population: this was the quickest and most effective way to succeed. In fact, during a subsequent interview, one of the NGO’s directors stated: I think that the approach you bring requires time. If you really want to achieve the others’ sincere participation you have to make them understand what kind of process this is. So, you cannot just come and say: we will do it this way. No, the person has to participate in the shared creative and implementation process. For this, you need time. So, I think that the idea of changing the situation through a collective process requires a lot of time. A lot of time. (personal communication, 13th November, 2012) Throughout the process the NGO took actions that led the process in this direction and that limited the designers’ possibilities of action. This meant that the NGO indirectly exercised a hidden power over the process and over the designer’s possibilities of action, and that this was related to its hidden agendas. The NGO’s long-term presence in the context, its network, and its knowledge of it favoured the indirect use of power. This inluenced the designer’s agency and reduced it beyond that expected. These dynamics may be quite common and the designer may not be aware of them or may be unprepared to deal with them. Lastly, this situation was ampliied by the partner being a mediator. Not only did the partner have agendas unknown to the designer, but by being a mediator it could ilter information in this regard. Partners’ Interest in the Suggested Process The last situation serves as an example for the category partners’ interest in the suggested process. In the period of time between the irst and the second meeting, the designer tried to involve the NGO’s members more in the square redesign by suggesting several internal activities. One of these was a workshop with the employees who worked in its library that was situated close to the square. Actually, some of them participated in the irst meeting, seemed interested, and were local inhabitants. Due to these factors, and the proximity of the library to the square, involving them could have been useful for the project development. The designer organized a workshop with the employees of the local library who were allocated to the children’s section. The aim of the workshop was to plan together some activities to be undertaken with children in the square and some actions meant to involve children’s parents in the process. This could have been the irst step of a creative participatory process. Due to the reduced amount of time that the employees could dedicate to the workshop, the designer irst interviewed them and, based on this, deined some issues and topics, and organized them into polarities. During the workshop, the designer tried to explore the polarities, and then conduct a brainstorming session to develop ideas that could address the local situation in the desired direction. At both moments several dificulties Collaboration and Competition Dynamics within the Partnership The third situation exempliies the category collaboration and competition dynamics within the partnership. As previously stated, the NGO did not agree with the designer on implementing a process in which ad-hoc planned activities would give rise to people’s engagement irst. Instead, it organized a public meeting at which local inhabitants could express their ideas and needs regarding the place. We observed that during the meeting’s organization, constant negotiation took place between the designer and the NGO about how this meeting had to happen. The designer suggested several activities and actions to stimulate people’s www.ijdesign.org 60 International Journal of Design Vol. 10 No. 1 2016 C. Del Gaudio, C. Franzato, and A. J. de Oliveira These three points provide enlightenment about some challenges to the designer’s agency associated with having local design partners. Relecting on them based on the previously presented theory will lead to a better understanding and show some strategies to deal with them. emerged and few ideas were generated, mostly suggested by the designer. Instead of actively co-creating, the participants acted more like passive informers, and the process itself was laborious. The participants expected answers from the designer more than playing an active role. At the same time, the designer noticed that thinking about something different from what they were used to doing or could do was really hard for them. They were lacking in this habit and expertise. In brief, two kinds of obstacles emerged: the employees had several dificulties in understanding the kind of process and rationale proposed; and they lacked time to do this kind of activity, thus preventing a longer process that could have led to better results. During a subsequent interview, one of the NGO’s members afirmed that her limits in understanding the process led her not to collaborate with the designer: “It is not demagoguery. I mean, I felt that I turned my back on you many times, because I didn’t know what to think about to help you” (personal communication, 12th November, 2012). This situation points out: how in the project the design process was not trivial for participants; the designer’s unawareness and inability to understand the foregoing and to support them; and how this compromised their interest. When participants do not feel useful and potentially active actors in the process, they may easily lose interest in the project and hinder a participatory dynamic. The four presented situations are examples of dynamics that occurred throughout the whole process. Interpreting them through our theoretical background led to some understanding about establishing a partnership with a local actor when conducting a PD project: Discussion and Conclusion The previous section described a PD project undertaken in a conlict-affected and fragile urban context, and pointed out some of the implications of setting up a partnership with a local actor that may be unexpected by the designer. It is an example in which the partnership hindered the project development in several ways and which, mostly, highlights under-discussed issues in current discourse about social design and how they manifest themselves in projects’ development. In the Complexo de favelas da Maré project the designer lost a great amount of agency over the project throughout the whole process. Why and how did it happen? Three understandings emerged about this speciic case: the partner was a mediator and had undeclared agendas that it addressed through indirect use of power; the partnership featured both collaborative and competitive dynamics—the latter being unexpected to the designer; and, lastly, the design partner was unfamiliar with the design process and had dificulties in participating. Some concepts about partnership, power, delegation, and agency are helpful to understand them better—that means how and why they happened, as well as to provide enlightenment about the designer’s responsibilities and possibilities of dealing with them. Within PD, the designer shares his or her power and agency with project participants in order to open and democratize the process. In this setting, a closer partnership with one speciic local actor can be seen as a different way of understanding and practising his or her own agency. Actually, it is a shared and collaborative form of design agency. The designer establishes a local partnership to create synergies among skills and resources that characterize each one of the two different separate entities involved—taking advantage of the different features and skills and overcoming each one’s own limitations. In theory, this situation would only beneit the project. However, designers have to be aware that sharing agency also implies the accomplishment of negotiating tasks, if not, this situation can result not only in the designer’s agency sharing but in its loss too. Actually, as Lupia (2001) explained, in a partnership the designer delegates to the partner the performance of some tasks on his or her behalf. For instance: knowing the context, having the permission to act there, creating relationships with local people, etc. This is clearly an act of agency delegation. Nevertheless, it is an act of power delegation too: by fulilling tasks the partner acquires power over the related activities and actors, thus over the design process. However, a partner may have different and hidden agendas (Bachrach & Baratz, 1963), and this means that in a PD process it may even not share the main principles of the fostered dynamics, and the apparently common design goal. In this speciic case, agency delegation is risky: if not 1. The design partner is a mediator between the designer and the local context. The mediator role gives it a great amount of power over the design process, even more than the designer’s power. Consequently, this has a great impact on the designer’s possibilities of action within the design process and on the project itself. The partner, whose power increases directly in proportion with its knowledge of the context, and the closeness of the relationship with the designer, can act towards the same objective as the designer as well in a different direction, thus inluencing and hindering the design process. 2. The partnership with a local actor is not solidary collaboration. It shifts constantly between collaborative and competitive dynamics. This creates antagonistic tension within the design network that if managed through negotiation will have constructive results. The designer has to understand that the partner is a separate entity with its own beliefs, interests, and ways of action, and has to negotiate with it to accomplish the design process and achieve results. 3. The design partner is unfamiliar with design dynamics and this could undermine its interest in the process, its contribution, and trust in its value. If the partner does not believe in the effectiveness of what is suggested, it will not support the process and will oppose it. www.ijdesign.org 61 International Journal of Design Vol. 10 No. 1 2016 Sharing Design Agency with Local Partners in Participatory Design considering the NGO’s individual members, the dificulties they had in understanding the design process may have undermined their participation and consequently had a negative inluence on the project. Theory and experience gained from this study has led us to identify ive strategies to deal with the challenges and implications of local partnerships: appropriately managed the partner can abuse this power (Lupia, 2001) to the detriment of the designer’s agency and that of the others involved. This risk is not trivial considering, as explained by Broum and Enderud (1981), that the partner can exercise its power in an indirect and invisible way. This implies that the designer may not be aware of it and of who is inluencing the project, even in non-participatory ways. In brief, if the partner has hidden and different agendas, it can invisibly exercise power in the direction of limiting the designer’s possibilities of fostering the desired participatory process (Bachrach & Baratz, 1963)—as happened in the ield study described above. This is a situation that limits the designer’s agency. Hence, as previously stated, in a PD project, whose process openness increases the risk, a partnership requires the designer to engage in intensive and constant negotiations. If not, delegation becomes total, and the designer loses any possibility of acting within the project. However, as the presented case and the present literature show, it seems that the designer could both not be aware of it and lack the negotiation and agency sharing skills necessary to set up and deal with local partnership. What should the designer do or what can he or she do in this case? The Complexo de favela da Maré project will help in this understanding. According to agency theory, the designer was the principal while the NGO—both as an institution and through its individual members—was the agent. As Lupia and McCubbins (1998) explained, when a designer sets up a local partnership, agency loss is zero when the local partner shares the same interest, which in a PD project means sharing the interest in promoting more democratic dynamics and the jointly elected design goal— the collaborative redesign of the square in the Maré case. At the same time, according to them, the designer’s agency loss is limited when the designer knows well the partner, its political position, local dynamics, and actors. If this precondition is lacking, the partner may address hidden agendas that in radical situations may mean manipulating the designer for its own purposes. In this regard, it is important to observe that these are extreme situations. Actually, rarely do two partners, or the different actors of a design network share the same aims. According to the nature of the network form of organization, the aims of the involved actors are convergent with the possibility of synergies. Only in few situations are they actually shared with full agreement (Castells, 2009). Moreover, actors’ aims and interests are never ixed but rather evolve constantly due to several contextual factors and the on-going dynamic network of relations that constantly stimulate change in their perceptions and interpretations (Stacey, Grifin, & Shaw, 2000). Therefore, considering the described case and the three, presented issues, the irst one was due to the designer’s lack of knowledge about the design partner and about the implications of its actions and activities, while the second and the third ones were due to not being aware of the dynamic nature of each actor’s interests and of the factors that sustain and inluence them. The designer did not know the partner and the context well—thus the implication of the partner’s actions—and this allowed the partner to inluence the project signiicantly. Furthermore, www.ijdesign.org 1. Design process explanation and training. For the development of a project, it is relevant that the design partner fully understands how the process works and how it can contribute. It has to feel able and apt to collaborate. If the partner does not feel capable of contributing, it will perceive the participation as a waste of time (Correia & Yusop, 2008) and hindering dynamics will emerge. Thus, to avoid them, at the beginning of the project, the designer has to implement activities to train the partner. 2. Sharing and negotiating common interests. The local design partner may be not only unfamiliar with design process, but unaware of the potentiality of design and the opportunities it can add to its activity too. Should the partner not automatically collaborate, then the designer has to implement activities to stimulate the partner’s interest in the design process, to share and constantly negotiate common objectives, and to work together towards achieving them. 3. Developing a dialectic and dialogic process. As previously stated, at the beginning of the project it is important to know the partner well and the implications of its actions. Usually an initial immersion in the context is considered enough for this purpose. However, due to the designer being an outsider, this may not be suficient. The designer and the partner are separate entities with backgrounds and organizational dynamics unknown to each other. Thus, it is important for both of them to engage together in a discovery process and in building a shared vision of the future. We suggest that the designer should conduct both a dialogic and dialectic process. Its dialogic phase will be a moment in which the designer and the partner become more aware about what they can do together in the speciic situation and understand better each other through an exchanging ideas process; while in the dialectic phase they gradually come closer and build a common vision. This kind of process has to be planned and stimulated by the designer. 4. Strengthening a supportive local network. Agency theory literature points out that to limit power abuse it is important to design ways to affect the agents’ future actions. McCubbins et al. (1987) suggested that the principal has to ensure that people who share its interest will be able to inluence what agents do. In the design case, this means that the designer has to develop strategies to foster and strengthen a network of supportive local social 62 International Journal of Design Vol. 10 No. 1 2016 C. Del Gaudio, C. Franzato, and A. J. de Oliveira designers ind in the elaboration of design scenarios (Manzini, 2003) and prototypes (Björgvinsson, Ehn, & Hillgren, 2010)— whose application has the potentiality to promote awareness, empowerment, inclusion, and participation—a potential path that needs to be improved by them according to the perspective presented in this paper. actors. He or she has to identify the actors that present similar principles and approach, and may be interested in the achievement of the speciic elected design goal. Then, he or she has to do social networking with them. An appropriate network favourable to the project will more easily avoid possible partner’s hindering dynamics. 5. Managing collaboration and competition. Partnership does not mean solidary collaboration, but creative collaboration as well as competition. Both of them may positively inluence the design project. Thus, it is a designer’s task to manage them towards the project purposes. In order to do this, he or she has to constantly observe the partner’s actions and, through negotiation, orientate the partnership towards one of the ive kinds of exchange dynamics (Sennett, 2012) more suitable for each speciic moment. Acknowledgments The author Chiara Del Gaudio would like to thank CAPES, the funding research agency of the Brazilian Ministry of Education, for the PNPD/CAPES scholarship, a scholarship that is part of the CAPES Post-doctoral National Program. References 1. Bachrach, P., & Baratz, M. S. (1963). Decisions and nondecisions: An analytical framework. The American Political Science Review, 57(3), 632- 642. 2. Bjerknes, G., Ehn, P., & Kyng, M. (1987). Computers and democracy: A Scandinavian challenge. Brookield, Canada: Gower. 3. Björgvinsson, E., Ehn, P., & Hillgren, P.-A. (2010). Participatory design and ‘democratizing innovation’. In K. Bødker, T. Bratteteig, D. Loi, & T. Robertson (Eds.), Proceedings of the 11th Conference on Participatory Design (pp. 41-50). New York, NY: ACM Press. 4. Borum, F., & Enderud, H. (1981). Konlikter i organisationer: belyst ved studier af edb-systemarbejde [Conlicts in organisations, illustrated by cases of computer systems design]. Copenhagen, Denmark: Nyt Nordisk Forlag Arnold Busck. 5. Brandão, C. R. (2005). Participatory research and participation in research. A look between times and spaces from Latin America. International Journal of Action Research, 1(1), 43-68. 6. Castells, M. (2009). A sociedade em rede [The rise of the Network Society]. São Paulo, Brazil: Paz e Terra. 7. Correia, A. P., & Yusop, F. D. (2008). “I don’t want to be empowered”: The challenge of involving real-world clients in instructional design experiences. In D. Hakken, J. Simonsen, & T. Robertson (Eds.), Proceedings of the 10th Conference on Participatory Design (pp. 214-216). Bloomington, IN: ACM Press. 8. Derkzen, P., Franklin, A., & Bock, B. (2008). Examining power struggles as a signiier of successful partnership working: A case study of partnership dynamics. Journal of Rural Studies, 24(4), 458-466. 9. Da Conceição, E. (2010). Who controls whom? Dynamics of power delegation and agency losses in EU trade politics. Journal of Common Market Studies, 48(4), 1107-1126. 10. Dille, T., & Söderlund J. (2011) Managing inter-institutional projects: The signiicance of isochronism, timing norms and temporal misits. International Journal of Project Management, 29(4), 480-490. We believe that, by applying these strategies, the designer’s agency loss will be reduced. However, despite what has been presented till now, we think that local design partnerships still have to be better discussed by the design community. The presented strategies are not really ad hoc design processes to be replicated for designing in the social ield. In fact, this paper was intended to discuss and improve the PD approach. Thus, these strategies are just suggestions meant to be better explored and to stimulate further research. In this regard, one of the main understandings that stands out from this paper is that current literature does not prepare designers for partnerships. It does not really present the two-fold nature of partnership and it does not discuss the implications and challenges for the design process, as well as partnership dynamisms and otherness issues. By afirming this, we are not disregarding the relevance of establishing partnership with local actors. On the contrary, we think that having a local partner is important for understanding the context and integration within, as well as for action. Moreover, considering that the partner is one of the actors of the local design and social network, it and its local relationships are important—as well as each of the actors’ relationships in the local social network. Relationships are at the core of PD. Actually, a PD process occurs among, through, and because of the social network that all the involved actors weave during design practice. Emilson, Seravalli, and Hillgren (2011) called this feature of the design process infrastructuring. The -ing form of the verb suggests that such effect of PD is not a inal result of the process, but an on-going and never-ending dynamic that is generated by collaboration and generates collaboration—both dialectically and dialogically. The presented case is an example of how designers may still lack the skills to truly contribute to this process. Finally, we think that the fact of partnerships being arenas of power (Derkzen et al., 2008), not only in conlict-affected and fragile urban areas but in others too, can be welcomed by designers as an opportunity not as an obstacle. Actually, if the possibility of discussing social issues that affect a community is a mandatory feature of agonistic spaces (Björgvinsson, Ehn & Hillgren, 2010), www.ijdesign.org 63 International Journal of Design Vol. 10 No. 1 2016 Sharing Design Agency with Local Partners in Participatory Design 24. Pollack, M. A. (1997). Delegation, agency, and agenda setting in the European community. International Organization, 51(1), 99-134. 25. Sennett, R. (2012). Together: The rituals, pleasures, and politics of cooperation. New Haven, CT: Yale University Press. 26. Silva, H. de A., Morais, M. C., & Rubenilson, T. B. (2009). Planejamento urbano e participação popular: Uma experiência de ensino e extensão no Bairro Ponta Negra, Natal/RN [Urban planning and public participation: A teaching and applied research experience in Ponta Negra neighborhood, in Natal/Rio Grande do Norte]. Risco: Revista de Pesquisa em Arquitetura e Urbanismo, 9, 110-123. 27. Souza e Silva, J. (2003). Sobre a vivência dos moradores das favelas carioca[About the everyday life of the inhabitants of Rio de Janeiro’s favelas]. In E. Sydow & M. L. Mendonça (Eds.), Direitos Humanos no Brasil (pp.171-176). São Paulo, Brazil: Global Exchange.Souza e Silva, J. (2004). Favelas — Além dos estereótipos [Beyond stereotypes]. Democracia Viva, 22, 11-16. 28. Souza e Silva, J. (2010). As unidades policiais paciicadoras e os novos desaios para as favelas cariocas. Observatório de favelas[The peacemaker police units and the new challenges for Rio de Janeiro’s favelas. The observatory of favelas]. Retrieved December 23, 2013, from http:// observatoriodefavelas.org.br/wp-content/uploads/2013/06/ Aspectos-humanos-das-favelas-cariocas.pdf 29. Souza e Silva, J. (2011). Favelas: As formas de ver deinem as formas de intervir[ 30. Favelas: The ways of looking at them deine the ways of intervening]. Revista Econômica, 13(1), 47-57. 31. Stacey, R., Grifin, D., & Shaw, P. (2000). Complexity and management: Fad or radical challenge to systems thinking? London, UK: Routledge. 32. Thiollent, M. (1985). Metodologia da Pesquisa-Ação [Action Research Methodology]. São Paulo, Brazil: Cortez. 33. Winschiers-Theophilus, H., Chivuno-Kuria, S., Kapuire, G. K., Bidwell, N. J., & Blake, E. (2010). Being participated: A community approach. In K. Bødker, T. Bratteteig, D. Loi, & T. Robertson (Eds.), Proceedings of the 11th Conference on Participatory Design (pp. 1-10). New York, NY: ACM Press. 34. UNDP (2012). Supporting capacity development in conlict and fragile contexts. Retrieved November 23, 2014, from http://www.undp.org/content/dam/undp/library/crisis%20pr evention/2012SupportingCapacityDevelopmentinConflict FragileSettings.pdf 11. Emilson, A., Seravalli, A., & Hillgren, P. -A. (2011). Dealing with dilemmas: Participatory approaches in design for social innovation. Swedish Design Research Journal, 11(1), 23-29. 12. Fals Borda, O., & Rahman, M. A. (1991). Action and knowledge: Breaking the monopoly with participatory action-research. New York, NY: The Apex Press. 13. Foucault, M. (1982). The subject and power. Critical Inquiry, 8(4), 777-795. 14. Giddens, A. (1984). The constitution of society: Outline of the theory of structuration. Oakland, CA: University of California Press. 15. Hussain, S., Sanders, E. B.-N., & Steinert, M. (2012) Participatory design with marginalized people in developing countries: Challenges and opportunities experienced in a ield study in Cambodia. International Journal of Design, 6(2), 91-109. 16. Instituto Brasileiro de Geograia e Estatística. (2010). Censo demográico 2010. Aglomerado subnormais. Primeiros resultados. [Demographic Census 2010. Subnormal Clusters. First Findings]. Retrieved March 30, 2016 from http://biblioteca.ibge.gov.br/visualizacao/periodicos/92/ cd_2010_aglomerados_subnormais.pdf 17. Koskinen, I., Zimmerman, J., Binder, T., Redström, J., & Wensveen, S. (2011). Field: How to follow design through society. In I. Koskinen, J. Zimmerman, T. Binder, J. Redström, & S. Wensveen (Eds.), Design research through practice: From the lab, ield, and showroom (pp. 69-87). Waltham, MA,: Morgan Kaufmann. 18. Lupia, A. (2001). Delegation of power: Agency theory. In N. J. Smelser & P. B. Baltes (Eds.), International encyclopedia of the social and behavioral science (Vol. 5, pp. 3375-3377). Oxford, UK: Elsevier. 19. Lupia, A., & McCubbins, M. D. (1998). The democratic dilemma: Can citizens learn what they need to know. New York, NY: Cambridge University Press. 20. Manzini, E. (2003). Scenarios of sustainable well-being. Design Philosophy Papers, 1(1), 5-21. 21. Manzini, E., Jégou, F., & Meroni, A. (2009). Design oriented scenarios: Generating new shared vision of sustainable product service systems. In M. Crul, J. C. Dielh, & C. Ryan (Eds.), Design for sustainability: A global guide (pp. 15-32). Retrieved December 23, 2013, from http://www.d4s-sbs.org/ d4s_modules%20total%20s.pdf 22. McCubbins, M. D., Noll, R. G., & Weingast, B. R. (1987). Administrative procedures as instruments of political control. Journal of Law, Economics, & Organization, 3(2), 243-79. 23. Mulgan, G. (2014). Design in public and social innovation. What works and what could work better. Retrieved February 22, 2016, from https://www.nesta.org.uk/sites/default/iles/ design_in_public_and_social_innovation.pdf www.ijdesign.org 64 International Journal of Design Vol. 10 No. 1 2016