Knowledge and Power of Civil Society : an empirical study of Brazilian professionals working in NGOs

This study critically analyzes the way Non-Governmental Organizations (NGOs) operate in Brazil and their contribution to the development of Brazilian civil society. The notion of ‘power fields’ and ‘habitus’ proposed by Bourdieu (1989, 1996), provides the theoretical backdrop to our discussions. This focusses largely on the recursive connection between structure and agency which resonates with the work of Fligstein (2007) and Giddens’ (1984) structuration theory. Although there has been extensive research on NGOs, there has been little academic scrutiny on the nature of outcomes of their interventions. This study therefore seeks to critically examine the learning and social practices developed by professionals in NGOs in their daily activities. Drawing on a qualitative case study approach in Brazil, this study contributes to this debate by providing insight into the nature of the operations of NGOs and the knowledge they produce, through the recursive connection between structure and agency. The results of the study show that the knowledge generated by NGOs in Brazil give them a certain level of power and influence in the country’s civil society. Introduction Many of the issues identified by civil societies today as relevant public problems in contemporary societies are also subject to controversies and debates (Evers, 1995; Salamon & Anheier, 1997; Morris, 2000; Sobottka, 2003; Alves, 2004; Teodósio, 2014). In most cases, it is not uncommon, for example, to see conflicts between green movements, governments, corporations and the citizens regarding sustainable environmental development interventions

We use the notions of 'power fields' and 'habitus' proposed by Bourdieu (1989Bourdieu ( , 1996) ) and the recursive connection between structure and agency associated with Fligstein (2007) and Giddens ' (1984) structuration theory to support our arguments.This theoretical approach permits to us to critically analyze the social learning practice developed by NGO professionals in their day-by-day practices and activities.
Agents or groups can be defined by their relative positions in the social space.This means that the differences among social actors also define their positions in a specific social field.
According to Bourdieu (1989), each position or class is determined by the position it occupies within a region or social space.This allows us to identify power relations that can be attributed to each social actor.The 'habitus' therefore can be recognized as a practical approach created by the social actors in their day-by-day activities inside these social fields that encourage them to act and react without using a specific method (Bourdieu, 1996).These two theoretical approaches are useful to understand the process of knowledge production as a resource of power and a collection of practices by these social actors, particularly the NGO professionals, in the civil society field.In this perspective, we also need to understand that the process of acquiring knowledge is based on practical learning by professionals (Ribeiro, 2003;Santos, 2004).This is largely as a result of interactions and practices shared by people in everyday behavior through the binding process between structure and agency (Lave & Wenger, 1991).This is an area that has not been sufficiently scrutinized from an academic perspective.This paper therefore seeks to address this gap in knowledge by providing insight into the knowledge produced by NGOs as the possibility of expanding its power and influence in the Brazilian civil society.We attempt to do this by adopting the assumption that the generation of knowledge occurs through the predisposed recursion between agents and structure, based on relational and complementary processes enabling the consolidation of power of non-governmental organizations in the civil society.
In analyzing civil society organizations we cannot take a reductionist approach that allows us to treat these organizations as components of the Third Sector.In other words, we cannot take non-profit organizations or Third Sector Organizations as a constitutive part of civil society (Sobottka, 2002;Alves, 2004;Teodósio, 2010).This ideological allusion to the Third Sector evokes two forms of distortion of reality.The first expresses itself in an attempt to 'reduce the political aspect of civil society to transform it into a place of service in order to replace the state' (Alves, 2004 p. 152).Secondly, 'by excluding profit-making organizations from civil society, the new ideologues claims to depoliticize the market, making it only an object of technical considerations' (Alves, 2004 p. 152).In this paper, we regard civil society as a space of action for individuals and multiple types of organizations in a continuum consisting of the state, the market and the public sphere, developing political, economic and cultural frames in the same structure of social action (Fischer et al., 2006;Teodósio, 2010;Alves, 2004).From this perspective we discuss and critically analyze the knowledge produced by Non-Governmental Organizations (NGOs) in order to explore the ways in which they can expand their field of power in Brazilian civil society.
These theoretical perspectives are useful for constructing a frame in which the process of knowledge production can be seen as a resource of power recognized through the set of practices developed by social actors, such as NGO professionals in civil society.From this perspective, we attempt to understand the knowledge based on practical social learning that happens through the interactions between NGO organizations and professionals of this field and how this is shared by people through everyday behavior in a social action related to structure and agency at the same time (Lave and Wenger, 1991;Kimble, Hildreth, 2004;Handley, Sturdy, Fincham & Clark, 2006;Campos, Mendonça & Alves, 2012;Yerbury, 2013;Guerra & Teodósio, 2014).
When we consider the connection between structure and agency, we assume that the production of knowledge is generated from the predisposed recursion between agents and structure, based on relational and complementary processes, allowing for the consolidation of the power of non-governmental organizations in the civil society field.In order to shed light on these theoretical positions, we attempt to test out these abstract ideas in the social reality of Brazil.We conducted in-depth interviews with professionals working in three NGOs.In addition to these interviews, we used secondary data gleaned from reports, websites, books and manuals produced by these NGOs.
Our results showed that the knowledge generated by these NGOs led to the consolidation of their field of power in Brazilian civil society.This knowledge production indicated that Brazilian social, political and environmental problems have increased the possibilities for NGOs to interact with other actors in the power fields of civil society in the country.In addition, these interactions also brought more power to these NGOs in terms of capacity for action and mobilization of like-minded contemporaries in the public sphere.
The structure of this paper is as follow.First, we discuss the civil society power field as a sphere of action developed by individuals and organizations for multiple purposes.After that, we discuss the notion of 'power fields' and 'habitus' proposed by Bourdieu (1989Bourdieu ( , 1996) ) and components of Giddens' 1984 structuration theory.In the third part of this paper, we discuss the generation of practical knowledge as a way to change the power relations in the civil society field through the recursive process of the interaction between structure and agency.
Finally, we present the results from the research undertaken on key Brazilian NGOs as empirical illustrations of the theoretical ideas about how knowledge about social, political, cultural and environmental issues is generated.We envisage that our study on the generation of knowledge by Brazilian NGOs will advance our understanding and comprehension of the challenges, perspectives and changes in the power relations in the civil society field and contribute to current studies on studies of knowledge and power in the public sphere.

Civil Society Organizations, Power Field and Habitus
Brazilian civil society is composed of a variety of agents and organizations that historically have different practices, strategies and approaches in attempting to advance democracy, ameliorate poverty, fight against the social inequalities and protect the environment.Brazil's political system was a dictatorship during the 60s, 70s and half of the 80s in the 20th century.
This had a negative effect on civil liberties as well as operations of NGOs.
An important group of Brazilian civil society organizations has since emerged with their political backgrounds based on the ideals of democracy and active citizenship (Fischer et al., 2006;Paula, 2005).Supported by the Catholic Church, many grassroots organizations known as Comunidades Eclesiais de Base -CEBs were created during the dictatorship era, inspired by the ideals of Liberation Theology and popular education (Paula, 2005).Most of these civil society organizations believed in local popular and informal knowledge as a way of disseminating their political and socio-economic policies to the poor and to local communities.This movement was developed as an alternative space for political mobilization to encourage popular participation in the discussion of everyday difficulties that people faced.This led to the formation of popular community leaders (Paula, 2005).Some groups also emerged to fight for citizenship rights, such as the movements protesting against the cost of living, unemployment, political repression and the oppression of women (Gohn, 1995).
In the 80s, some organizations with specialized professionals such as advisors, educators and specialists in popular mobilization appeared in the Brazilian civil society field.This was the beginning of Non-Governmental Organizations in Brazil (Paula, 2005).During this period, these NGOs had a confrontational relationship with the political powers, because they fought against the dictatorship and had support from international NGOs, such as Amnesty International.
Nowadays, these NGOs have gained significance because they are potential partners of current governments, due to their professional ability and structure (Ghon, 1995).Fischer et al. explain that one of the reasons for NGOs being perceived as potential partners is because 'neoliberal discourse presupposes the withdrawal of the State from a number of activities and the primacy of the market' (Fischer et al., 2006).This also has led to a reconfiguration of the funding model of NGOs, with an emphasis on the self-sustainability (Ghon, 1995).
During the dictatorship era and until the end of the 1980s many participative spaces were created by the Brazilian government to discuss public policies with civil society.Thus, the role of many Brazilians NGOs changed during this time.In more recent years, although they continue to work for the poor, the role of the Brazilian NGOs has again changed.They are now focusing on providing professional, technical and managerial support to the disadvantaged members of society on issues regarding public policy and public management as well as on the preparation and implementation of policies that result in better living conditions for the population (Marteleto & Ribeiro, 1997;Campos, Mendonça & Alves, 2012).
According to Fischer et al. (2006), these activities have highlighted the position that NGOs hold within the matrix of the state, market and civil society.The partnership approach is based on an assumption that NGOs can produce better and more useful knowledge and solutions to the social problems than the state and, in many cases, that even the academic world fails to address (Bronzo, Teodósio & Rocha, 2012).So, partnerships with civil society organization could bring more effective technical and social solutions to these problems (Evans, 1997;Marteleto & Ribeiro, 1997;Selsky & Parker, 2005;Ospina, Godsoe & Schall, 2001;Ospina & Saz-Carranza, 2005).In this context, from a Bourdieusian perspective, they are assigned the task of building 'new transversal flows' between fragmented institutional spaces (Meister, 2003;Haigh & Hoffman, 2012), thus enabling, in their field of training and action, symbolic intersections with the political (state), economic (market) and scientific (university and research centers) fields (Marteleto & Ribeiro, 1997;Selsky& Parker, 2005;Adami, 2013;Guerra& Teodósio, 2014).
We can use this perspective to understand the knowledge produced by NGOs from the theoretical background of power fields and habitus, notions proposed by Bourdieu (1989Bourdieu ( , 1996)).According to Bourdieu, habitus, as the word indicates, 'is knowledge acquired' and it can also be seen as a form of capital.In a practical sense, habitus enables agents to act and react without conscious obedience to rules (Bourdieu, 1996).However, it also incorporates 'a principle of generating and unifying knowledge which transforms the intrinsic and relational characteristics of a person's position into a distinct way of life, ie, into a distinct set of choices' (Bourdieu 1996 p. 21-22).This capacity to establish difference is fundamental to the notion of habitus and to the notion of field.
Giddens '(1984) structuration theory states that understanding people's activity is the core purpose of social analysis and he goes on further to argue that it is thus important to study practice because it can make a difference to outcomes.In this regard, the processual concept of structuration brings together 'structure and agency ', which are important for understanding the arguments in this paper.
In a social world constituted by different dimensions, agents define themselves by their relative positions in this space; in other words, by their differences, whether objective or subjective.Each social actor is placed in a position or class within a given region of space, a field, where a set of forces are 'imposed on all who enter this field' (Bourdieu, 1989 p. 134).
These are the different kinds of power (or capital) that occur in different fields.The position of a particular agent in the social space can thus be defined by the position they occupy the different fields, ie by the distribution of powers that operate in each field (Bourdieu, 1989).
Following this theoretical approach, we highlight the relationship between the characteristic habitus of a class, 'which is expressed in terms of universalism and formalism of its intentions' and the logic of the field in which they are generated (Bourdieu, 1989, p. 81).This approach is close to the concept of agency given by Giddens (1984) that defines agents as having the power of intention and the capacity for action.
With this theoretical background, we discuss the interaction between structure and agency without considering the overlap by hegemonic forces of one field on another.We contend that there is an additional and relational process, allowing for the consolidation of the power field of civil society organizations in the public sphere.We assume too that the knowledge generated by civil society organizations is influenced by social and practical processes with each recursion, and this will be discussed in the next section.

Social and Practical Knowledge produced by NGOs
The creation of force fields is based on power relations with objective and subjective dimensions when we consider the prevailing habitus of each field.From this dimension, 'actors, under both stable and unstable institutional conditions, are not just captured by shared meanings in their fields, but are defined as scripts that must be interpreted by professionals or government bureaucrats' (Fligstein, 2007, p. 66).They emerge as counter narrative to the rationalist conception of action, because we 'cannot process all the sensory data in a rational calculation that use and form concepts, based on their previous experience, in order to make decisions and act accordingly'.(Carvalho et al., 2005 p. 3).Considering this point of view, structure and agency are factors that determine action.They neither exist nor perpetuate themselves.Neither has direct influence over the other, but the interaction between them is influenced by the interpretation of its components under legitimacy conditions (Machado-da- Silva, Fonseca & Crubellate, 2005).Fligstein (2007 p. 61) extends this understanding by developing a sociological view of action which he calls 'social skills'.According to this approach, skilled social actors must understand the perception of other actors in their group, regarding their multiple conceptions of interest and identity in specific situations, and provide an interpretation of the situation and actions that are consistent with the interests and identities of all of the social group (Fligstein, 2007, p. 61).
Machado-da- Silva, Fonseca and Crubellate (2005) argue that it is possible to visualize the interdependence between social structures and actions, without succumbing to the idea of linear causality between them.In other words, this means a dialectical interdependence link between objectivism and subjectivism or between mental and social structures (Misoczky, 2003).This conception brings us to a dialogical and interpretive link between structure and agency, where the NGOs can create knowledge in the civil society sphere.They are influenced by the habitus and the presence of social skills in the social action, given their capacity to address local social demands (Guerra & Teodósio, 2013).
We can observe that knowledge is based on social processes and can be analyzed at individual, psychological, social and cultural levels (Ribeiro, 2003;Santos, 2004;Bauman, 2008, Schommer & Boullosa 2010;Guerra& Teodósio, 2014).This conception is based on the social learning approach or socio-practice approach.In this sense, learning is as a result of interactions and practices shared by people, that we can find in daily behaviors.So we try to understand the cultural and social contexts in which people interact, build their shared practices, their cognitive and emotional repertoires and the necessary analysis of the individual cognitive processes of learning (Schommer & Boullosa, 2010).This notion of learning is based on social practice and is seen as a a praxis of action and reflection at the same time (Freire, 1987;Ribeiro, 2003;Santos, 2004).This process is related to the ongoing restructuring of the meanings and signifiers on which specific systems consolidate learning.This, however, does not take place in an unconstrained context (Schommer & Boullosa, 2010, p. 21).Rather, the social approach emphasizes that learning happens within social and cultural structures, where we can see the relational interdependence between subject, world, activity, meaning, cognition, learning and knowledge (Yerbury, 2009).This is influenced by the historical context that is open to changes in a world that is socially constituted day by day (Lave & Wenger, 1991;Haugh 2005;Leana & Pil, 2006;Yerbury, 2013).Based on these assumptions, we can consider that where there is knowledge production, this is generated by the habitus of agents associated through recursive processes, connecting structure and agency.Silva (2004, p. 145) argues that knowledge consists of 'information, which is relatively stable or static, and subjective characteristics in people's minds'.Knowledge generation is therefore based on a dynamic nature of engaging experiences, context, interpretation and reflection.
We identify two intrinsically-related knowledges.Firstly, subjective knowledge, identified by the skills inherent in a person and the system of ideas and experience and secondly, perception, related to the explicit knowledge, formalized in text, graphics, tables, figures, drawings, easily organized in databases and publications in general (Nonaka & Takeuchi, 1997).These two types of knowledge contribute to the argument that knowledge is produced by social actors.The first, characterized as subjective knowledge, can be shown by means of perceptions and experiences of agents in the fields of power of NGOs (Bronzo, Teodósio & Rocha, 2012;Guerra & Teodósio, 2014).The second one, identified as explicit knowledge, considers that the knowledge can be systematized, formalized verbally and in records, emphasizing the cultural and symbolic capital of the fields which they belong.Albagli and Maciel (2004) highlight the importance of knowledge as the ability to process and recreate meanings, and further, the ability to convert this knowledge into action, or, more specifically, into innovation.Through this theoretical perspective, it is possible to understand the possibilities for reframing diversified knowledge in their fields of power through the knowledge created by NGOs.
NGOs try to engage social actors (agents) for their activities through mental and manual activity, action and reflection.The production of knowledge can therefore be seen as a result of acceptance of certain actions by agents, which are interpreted as legitimate.The knowledge produced by some NGOs is reproduced in social practice and becomes institutionalized, being considered as desirable and an acceptable choice (Bronzo, Teodósio & Rocha, 2012;Guerra & Teodósio, 2013;Guerra & Teodósio, 2014).NGOs are involved in relational processes that allow the transformation of the field in which they operate by means of recursion, when individuals are considered social actors endowed with 'habitus' and interpretative ability.This perspective allows us to understand that 'diversity [is] the power to change and engage in ways not dictated or guaranteed by the interpretive schemes of individuals and action that derives from them' (Carvalho et al., 2005 p. 20).

Methodology
In order to gain a deeper understanding of the way Brazilian NGOs operate, the investigation used an in-depth case study approach (Tellis, 1997;Yin ,2003).Face to face interviews with key informants from the selected NGOS enabled us to identify valuable empirical data on a very complex area of scrutiny.The case study analysis involved a comparative analysis of three cases.Two of the NGOs work nationally and the remaining one has offices in different countries.These were i) Cooperação para o Desenvolvimento e Morada Humana (CDM) ; ii) Grupo de Institutos, Fundações e Empresas (GIFE); i) Associação Voluntários para o Serviço Internacional (AVSI Foundation).The selection of these cases was based on the need to illustrate the importance, diversity and representativeness of the NGOs under scrutiny.These organizations were selected through our knowledge of the sector and social capital among the civil society practitioners.Most NGOs also collaborate with universities and scientific research centers in Brazil and so this assisted us in generating a reliable data base.Secondary data was collected through reports, websites, books and manuals produced by these organizations as well as government institutions.
We conducted in-depth interviews with senior and operational staff working in the NGOs as a way to gain a better insight into the knowledge produced by NGOs.During this process, we also attempted to, identify and explore: i) the field of power in which each NGO is located and its ability to work in this environment, as the habitus and the social skills that each developed in this situation; ii) the recursive connection between staff of NGOs and field structure, aligned to the knowledge produced by them; iii) the impact of knowledge they created to change the Brazilian civil society field.Each interview lasted for about an hour and also involved repeat visits.
The data collected from the interviews was recorded, transcribed and manually analyzed through an inductive process of interpretation.This allowed us to analyze each case, generate relevant codes and identify the emerging themes (Miles and Huberman (1994), Bryman and Bell (2003), Yin (2003), Straus and Corbin, (2008) and Basit (2003)).We were then able to ascertain if there were similar patterns or themes across all three case studies that could be generalised (Agar ,1980;Miles and Huberman ,1994).

Results -Knowledge and Power of Three Brazilian NGOs
CDM is a Brazilian NGO whose main goal to promote the development of the individual, their personal and social development, in order to overcome their vulnerabilities.This organization's projects and social services provision are aimed at poverty reduction in areas marked by extreme deprivation.This organization emerged in Brazil in the mid-1980s, and is based in the cities of Belo Horizonte, Salvador and Rio de Janeiro, which are cities in three different Brazilian states.Its main partners are the local communities in which it is embedded, represented by neighborhood associations and other local organizations.CDM believes that they must develop their projects in communion and harmony with the needs of the communities.CDM has many institutional partners such as the public sector (federal and local government), private sector, universities and other NGOs.Among these partnerships, we highlight the AVSI Foundation and FIAT in Brazil, working together in social responsibility projects for the development of the communities located in the vicinity of the car factory.In all these projects, CDM works to improve urban planning with a particular focus on areas with high concentrations of poverty.
Regarding knowledge generation, CDM, derives knowledge from both the exchange of information with partners and beneficiaries involved in projects such as socialization and through the systematization of knowledge that is produced through team work.This, according to one of the managers interviewed, takes place through opportunities to share practices developed through the organizations` activities.However, this latter form of knowledge is identified by the respondent as one of the greatest challenges of CDM.
We have projects in separate areas with specific teams dedicated to their development.Having the knowledge produced by these teams shared across teams is very difficult because people do not always need to learn about the other projects engaged in by the organization.They are committed to developing activities that people can choose to engage in.So, to get people also to participate in the intellectual heritage of the CDM is a huge challenge for us (Interviewee CDM, 2013).
According to the CDM manager, a significant amount of knowledge is produced by other civil society organizations, international organizations and the third sector.This is largely as a result of various interventions aimed at addressing social, political and environmental problems in Brazil.In most cases, this knowledge is produced by NGOs where the public sector does not develop programs, projects and actions to solve social problems.So, in reality these developments bring greater legitimacy to the actions of the NGOs, because managers of civil society organizations generally have a better understanding of local communities than managers in government departments or even the managers of corporations.This came out clearly from the informant from CDM when he said: What I hear a lot, including from our technical staff, is that we always need to do a conversion, ie, the third sector should exist where the State does not have a presence.And that's a mistake.I try to fight for this sector as a legitimate expression of the society.Like the private sector, it is needed by a particular group, the difference is in the perspective of financial profit as an end that is not part of the values of the third sector (Interviewee CDM 2013).
Another NGO we studied was the AVSI Foundation.It is an Italian civil society organization created in the early 1970s and it currently works in a number of developing countries.AVSI began its activities in Brazil in the 1980s mainly supporting Italian missionaries who were working in the city of Belo Horizonte.Their work was largely focussed on education and human dignity.Guided by the Social Doctrine of the Catholic Church and currently headed by an Italian with significant managerial experience, AVSI supports a network of more than sixty (60) organizations working on different agendas of social and community development in Brazil.In this network, the AVSI Foundation tries to share knowledge, practical learning from its work in poor Brazilian communities and methods of development of social programs and projects.According to AVSI's manager, the organization seeks to achieve its objectives through a comprehensive network of relationships with local partners in different areas of Brazil, a big and diverse country with varying levels of deprivation in its communities.These partners are government institutions, at both federal and regional level; big corporations; educational institutions, including universities, grassroots associations and other NGOs; religious organizations, congregations, parishes and dioceses; and health institutions and hospitals.
Despite the involvement of these partners in the implementation of projects, AVSI faces a major challenge related to the make-up of its international cooperation strategy.This is reflected by an informant from this organization when he said: We can see the reduced role of Italy's International Relations Ministry in their cooperation in development and in addition through the private sector involvement in the social sector.The programs cannot any more be programs offered only by the government or only by NGOs, they have to be joint programs.So the main challenge is to adapt our organization to the point of view of work permeated by the division of responsibilities.This is the new economic, geopolitical and social scenario we are going through (Interviewee AVSI Foundation, 2013).
According to the AVSI Foundation manager many problems in Brazilian civil society organizations largely emanate from the lack of professional skills among the staff of these organizations.He went on to state that the NGOs also suffer from an image problem in Brazil, where most people see them as organizations that are run by inexperienced people.As a result many practices in Brazilian NGOs are largely guided by good will.There is therefore a need to improve the competence of civil society professionals, through effective management and through evaluation of social projects to attain efficiency (Interviewee AVSI Foundation, 2013).In addition, this manager thinks that NGOs in Brazil currently have a bad reputation mainly due to the corruption scandals in the state-civil society partnerships that have occurred over the past 10 years.However, it is clear that both the CDM manager and the AVSI Foundation manager strongly believe that Brazilian NGOs have an important role to play in assisting the poor through the development of social projects.
When asked how the knowledge produced by AVSI was measured, the interviewee reported that the generation of knowledge occurs differently from country to country, state to state and city to city, due to the fact that the organization operates in over 30 countries around the world.So, currently, they are exploring how to consolidate and organize the knowledge that is generated in various countries of the world.
We are creating working groups at international level where the manager responsible for each country and each sector has to coordinate and systematize the knowledge generated.This has been carried out internationally and locally.That is, someone has a responsibility in each country and there is someone to coordinate worldwide.It is not an easy process, it is an attempt to respond to global changes, and it has forced us to begin a process of changing the way we manage our own knowledge.(Interviewee AVSI Foundation, 2013) The beneficiaries of the AVSI Foundation projects have a central role in the generation of knowledge, according the manager interviewed.
It is useless for us to generate knowledge from what we think happens in reality.We have to see what the reality shows us.It's pointless to get into a community and carry out some training and attract international resources to activities that the community has no ability to create and sustain itself.From working together with the community, we generate knowledge for ourselves and for others.It is a practice that we have to systematize, to make it more noticeable and develop more consolidated knowledge that can be shared, in the interest of the community as well, drawing links to other projects or similar situations.(Interviewee AVSI Foundation, 2013) The last NGO researched was the GIFE, a non-profit organization, headquartered in São Paulo, created by Brazilian private sector corporations and foundations.GIFE's main objective is to invest in corporate social responsibility and programs to solve social and environmental problems in Brazil.GIFE's origin can be traced back to 1989 when the organization was developed, underpinned by a strong entrepreneurial focus.A group of entrepreneurs and businessmen, concerned about the problems in societies emanating from actions of corporations, began to meet to discuss ways in which these could be addressed.In 1995, the group was formally established by 25 organizations.Today the organization is made up of a variety of businesses and social investors which currently stand at 130 associates operating their own projects or working in partnership with other corporations and NGOs.
When asked about GIFE's understanding of corporate social responsibility, the respondent from the organization pointed out the role that the third sector and civil society organizations play in drawing the attention of companies to the responsibilities they have to society.'So, not only the state will be responsible for this kind of thing, but companies are also beginning to take charge of it, of generating social value for communities'.(Interviewee GIFE 2013).The organization's activities are focused on 3 key issues: mobilization, advocacy and knowledge management.According to the Coordinator of Knowledge, the 'GIFE's DNA is in the mobilization of business men'.
The objective of this area is the articulation between members in an attempt to make them talk to exchange experiences, discuss trends, challenges and also to create links with non-members, who may know other social investors.These may be from civil society, NGOs, local and federal government ... may be other kinds of partner associations, can be their own private sector companies.Then, the intention is to link different community actors and social investors (Interviewee GIFE 2013).
Advocacy is a new activity within GIFE.It is concerned with representing the interests of corporations on social responsibility legislation, monitoring and influencing the regulatory environment of the third sector while at the same time improving the relationship with the Brazilian government and NGOs.
The management of knowledge is developed through strategic projects that perform quantitative research (and sometimes qualitative research) on social investment and related topics.The systematization and dissemination of knowledge produced by GIFE occurs through meetings among members, as well as conferences, publications, policy papers, book chapters, research and annual reports.Such knowledge has the goal of promoting political institutional strengthening and improving the performance of private social investors on the issues of the social field.GIFE also has several partners that create databases that provide valuable information on social investment in Brazil.
We have partner organizations that produce knowledge about corporate social investment through research such as Comunitas, Ethos Institute, some universities, the Brazilian Association of Non-Governmental Organizations (ABONG), the Institute of Applied Economic Research (IPEA) and the Brazilian Institute of Geography and Statistics (IBGE).The IPEA is responsible for writing and publishing reports about NGOs in Brazil.GIFE and ABONG bring the knowledge they have on the civil society sector to support the IPEA in the data analysis.So it's a partnership that has worked very well for over 10 years (Interviewee GIFE 2013).
Our study aimed to investigate the knowledge produced by NGOs as a method with potential to expand its field of power in Brazilian civil society.Qualitative research allowed us to analyze the impact of the knowledge produced by civil society as opportunities for transformation and redefinition of the field in which they operate as organizations of the civil society.

Final Remarks
In our research we can see how three Brazilian NGOs created knowledge initiatives to strengthen their position of power in the civil society field.We can see the 'habitus' and the social skills in continuous action in these three organizations.Although the legitimacy crisis of the civil society organization in the contemporary Brazilian society is evident, it is clear that advances have been made in terms of institutionalization of the role of civil society organizations.These advances come, amongst other things, with the capacity to generate reliable knowledge about Brazilian social problems.CDM and AVSI Foundation organize themselves and developed a kind of social action based on the idea that it is important and possible to overcome human, physical and social vulnerabilities through efficient management of social projects and through the continuous dialogue with local communities.GIFE started with the engagement of entrepreneurs and executives interested in responding to social demands about business, but nowadays it has strong connections with Brazilian government institutions responsible for collecting and studying all the information related to social projects, NGOs, corporate social responsibility actions, fund-raising, impact and efficiency of social and environmental projects.These are themes that concern the media and the public opinion in Brazil and these have improved the position of power of GIFE in civil society.
In the three NGOs, the presence of professionals with social skills were vital to the operations and the expansion of their field of power within civil society as well as reproducing their 'habitus' within an extensive network of partnerships with other stakeholders.We can see the flows among and between fragmented institutional spaces and confluences of knowledge in symbolic intersections among social actors from civil society, state and market through the analysis of the activities and experience of these three Brazilian NGOs similar to those we can find in another studies about NGOs (see for example Marteleto and Ribeiro, 1997;Meister, 2003;Ospina, Godsoe & Schall, 2001;Ospina & Saz-Carranza, 2005;Fredette & Bradshaw, 2012.The analysis of CDM and the AVSI Foundation shows that professional staff tried to align the actions of cooperation and social development work implemented in different countries, thus promoting the visibility of multiple forms of knowledge about social problems that came from different cultures.This finding can be observed in the processes of AVSI Foundation that sought to develop core systematization of knowledge in each country of its operations in order to cope with the related dynamics and geopolitical transformations in which they operate as well as to reflect and learn through practice.According to Fligstein (2007), the ability to influence and lead other actors to the same path is critical to the transformation of structure and this theoretical perspective is confirmed by the empirical data collected on our research.
At the same time that these interactions are developed, operated and supported by the dynamics of knowledge production based through the social practice, we can understand the possibility of recursive connection between agents and structure in a social field of power in the society.This perception can be confirmed by the importance that the three organizations under scrutiny give to the knowledge that is produced from the relations with the public and other institutional partners.Although we found some difficulty in transforming the practices and dynamics developed by CDM into systematized knowledge, we can see that this could be spread across teams among the actors involved (Interviewee CDM 2013).We can therefore conclude that in the three cases that were analyzed, the knowledge generated by their institutional practice has improved their positions of power in Brazilian civil society.
From the empirical analysis undertaken in this research, we found that the knowledge produced by NGOs in the Brazilian context, through the recursive connection between the actors in the fields and structures, contributes immensely to the expansion and transformation of the field in which they operate.The results show that the knowledge generated and disseminated by NGOs allows the consolidation of their power field within Brazilian civil society.The knowledge produced through NGO professional practices combines scientific approaches on social and environmental problems with common sense perceptions and local knowledge as shown by the analysis in this work.