The aim of this paper is to describe how tools used in action research can promote learning in relation to Vygotsky’s concept of the zone of proximal development (zpd). In the study, Vygotsky’s (2001) theory of the zone of proximal development was used to describe how knowledge development can be manifested in a work of change. The data production took place in a school improvement project where fifty teachers participated over a period of three years. Tools for data production, and at the same time tools for learning, were: facilitation, shadowing and individual log book writing. The tools were initially analysed from three aspects: as used individually, as used in pairs or in a group, and in relation to the zone of proximal development. The responses were divided into confirmations and challenges. Although the participants expressed a need for challenges, they were in practice more prone to confirm one another. This gap showed the importance of analysing and openly talking about the difference between confirmation and challenge in zpd. In the joint analysis work, differences were elucidated between using tools outside zpd and learning through tools within zpd. Knowledge was developed in both situations but at different levels of reflection. In relation to zpd, the importance of colleagues was emphasised. In her/his competence development, a teacher is dependent on one or several critical friends (cf Handal, 2007) to be challenged as a knowledge developer. Giving response proved more difficult in virtual meetings than in physical meetings.

Action research, as used in the present study, is a way of combining the demands for scientific methodology with those for practical relevance, where the striving for shared knowledge formation and democratic interplay constitute central elements. This meant that the teachers in the study were encouraged to reflect on their practice, change habitual patterns and develop teaching theories on the basis of their classroom experiences (Persson & Rönnerman, 2005). In the democratic working methods aimed at in action research, an attitude is required whereby all involved persons respect one another and regards one another’s experiences as being of interest. This means systematically seeing that everybody has a voice and creating a critical constructive spirit (Holmstrand & Härnsten, 2003). In the present study, the emancipatory goal was to release power by first verbalising thoughts and reflections in the group, then developing the formulations into attempts at change in practice. In this way, the participants acted in both conversations and actions. Action research is often described as a meeting between theory and practice (Holmstrand & Härnsten, 2003; Lendahls, Rosendahl & Rönnerman, 2005), and in order not to render the meeting more difficult, Persson (2007) argues for bridging the dualism between theory and practice that is sometimes taken for granted. Teachers most often have little experience of doing research, which does not mean that they lack a theoretical basis or analytical tools for understanding what happens in schools. Like Persson I wish to emphasise that all participants are both theoreticians and practitioners.

Tools based on written texts were the most difficult to incorporate. The teachers had little experience of presenting texts containing reflections on everyday classroom work (Wennergren, 2006). Many of them saw themselves as talking rather than writing persons. It emerged, however, from the teachers’ texts that this was a matter of attitude rather than lack of competence. Handal (2007) expects teachers to be able to take an intellectual attitude to complex tasks and challenges in their profession, a demand that is also reasonable to make on teachers’ improvement work. Facilitation was the most highly valued tool in relation to learning, although it was used only on four occasions. At first sight facilitation might be seen as an oral tool, but since the contents of the conversations were based on written logs or shadowing documentations, a written preparation was also included. This shows that the tools need to be combined in various ways. The most sensitive tool was shadowing. The teachers thought that shadowing required many reflective steps, great trust and confidence in the shadowing person. In Bakhtin’s (1981) terms, the process is about appropriating, making the tool a part of one’s own professional actions, something that is not unreflectingly made to fit into one’s own or one’s colleague’s theory of the practice but rather challenges it (cf. Handal, 2007; Wennergren & Rönnerman, 2006). The teachers’ experience of participation and meaningfulness influenced the oral and written learning dialogues that took place in the communities of practice. The meaningfulness was based on a holistic experience where the teachers saw meaning not only for themselves but above all for the pupils. In order to assume the role of actor in the dialogue, they all needed a shared understanding of what dialogue competence (cf Wilhelmson, 1998; Wilhelmson & Döös, 2005) implied in practice. Each of the prerequisites for a learning dialogue that emerged in the studies does not in itself constitute new knowledge, and they can all be verified by studies based on sociocultural approaches. The new knowledge consists of all the ingredients being formulated and realised in a shared context, and of the most central concepts being realised in parallel both in the communities of practice and in the classroom. This parallelism contributed to curiosity and energy for the work of change in the classroom. A learning organization (Senge, 2006) or a cooperative school culture (Blossing, 2003; Hargreaves, 2004) has several features in common with learning communities of practice (Wenger, 1998). In order to understand school improvement groups from the perspective of adults’ learning in working life, I have drawn parallels to communities of practice. Wenger’s theory points out the importance of experiencing meaning. Learning in school improvement groups has to be experienced as meaningful, since this is the very fuel of the process. Tools contributing to school improvement processes may be seen as an aspect of naming existing knowledge, and, as Wenger emphasises, the driving force for new knowledge is knowledge that already exists. The participating teachers would not be where they are today without the driving force of history. I interpret their statements as implying that new concepts and theories that they did not have before have also served as support for concretising the present and forming the future. Wenger’s theory entails a view of knowledge development showing that the meaning of teachers’ competence development needs widening. When competence development implies learning from one another’s practices, this may mean taking part in shadowing, facilitation or video documentation. Teachers regarding themselves as actors will demand alternative forms of competence development. Only the teachers themselves can give one another legitimacy for being actors on the improvement arenas of schools – going from peripheral to full participation.