For us, the core element in pragmatic AR is the creation of arenas where discussion and collaborative research facilitate cogenerative learning. The encounter between local stakeholders and the professional researcher is the cornerstone on which mutual learning is built. Because we do not begin with a technique defined a priori, and we try to match our approach to the local situation and actors, our AR processes generally have a slow and intensely conversational opening stage. In beginning this way, with patient exploratory discussions, ours differs from most other approaches to AR.
We believe that our obligations as professional researchers are to assist the group to choose and learn whatever techniques and theories are suitable for the process they are concerned with. We do not rely on particular “recipes” that always should be followed. Such recipes mainly serve to lessen the insecurities of the professional researchers about what they should be doing, and they generaUy hamper the growth and development of the strength of the local stakeholder group.
We confront the local problems with all the skills and knowledge we have. What we do not know and the skills we lack may well be detrimental to the particular project, and we always wish we were smarter, more skilled, and better trained. But we advocate facing these existential and epistemological uncertainties as professional action researchers directly, rather than adhering to a particular set of recipes that would lower the demands on us personally. This is a necessary part of our integrity as researchers.
At the same time, we do not advocate casting around blindly for a way of working. We believe that in AR central skills revolve around the ability to understand and interpret social and material contexts, to decide on and configure appropriate arenas for discourse, to lead the interaction process, and to assist the participants in testing their knowledge in action and reflecting on the results. Thus, the action researcher must have the ability to interpret and reason about managing cogenerative learning processes that involve the active testing of the resulting knowledge.
A central ethical and political goal in AR is to achieve liberating outcomes. It is not easy to be precise about what a liberating outcome might look like, but we do not advocate narrowing this down to a recipe either. In our AR practice, determining what is a liberating outcome is an explicit part of the cogenerated learning process.
As a point of departure, it is possible to initiate discussion about liberating outcomes by offering an initial definition as “outcomes where local participants gain greater control over their own situation as a group.” We are not referring to personal liberation or the gaining of individual power by group members, but to the increased capacity of local participants to define and manage their own collective situation.
This kind ofliberation is not an abstractly quantifiable product. In highly coercive situations, a small gain may be intensely liberating, whereas in a more open situation, major changes may not be experienced as particularly significant by the participants. We believe that the change has to be real and meaningful to the local participants as a group. It is not up to the AR professional to decide if people have experienced a liberating outcome; it is up to us to pose this issue for the AR group and to keep the conversation about it going.
Source: Greenwood Davydd J., Levin Morten (2006), Introduction to Action Research: Social Research for Social Change, SAGE Publications, Inc; 2nd edition.