Activity theory is essentially rooted in social constructionism and social practice theory – broadly categorised as socio-cultural theory. This is traceable to the pragmatism of Dewey and Mead and in strands of recent Marxian theory, including the Soviet school of psychology associated with Vygotsky. Engestrom’s third generation of Activity Theory (2001) also incorporates collaborating activity systems, and further develops the concept of expansive learning, which Engestrom (op.cit.) equates to Bateson’s (1972) ‘Learning III’, ‘where a person or group begins to radically question the sense and meaning of the context and to construct a wider alternative context’. It provides in this research context a means of conceptualising transition and development within the practices of the collaborators. The methodology is therefore interventionist rather than traditional. By this I mean the research process and the researcher are active in the development.
I see developmental work research and other serious attempts at employing activity theory as a laboratory where new theoretical concepts and methodological principles are created, not only tested. (Engestrom, 1990:105)
The roots of the ‘third generation’ are in Engestrom, Engestrom and Karkkainnen’s (1995) recognition of emerging multi-professional working practices, a ‘new work paradigm’, when they developed the notion of a horizontal sharing of expertise across activity systems. ‘Polycontextuality’ recognised that
experts operate in and move between multiple parallel activity contexts…(which) demand and afford different, complementary, but also conflicting cognitive tools, rules and patterns of social interaction…Experts face the challenge of negotiating and combining ingredients from different contexts to achieve hybrid solutions.
The issue then became one of crossing boundaries in complex interacting activity systems.
More recent developments increasingly stress the ‘socio-spatial’ as well as the ‘temporal-longitudinal’ dimensions of expansiveness (Engestrom, Engestrom and Kerosuo, 2003), and for expansive learning to ‘take shape as ‘re-negotiation’ and ‘re-organization’ of collaborative relations and practices between and within the activity systems involved’ (Engestrom, 2004). In this article Engestrom claims that in developing practices between activity systems, expansive learning should be ‘reformulated’ as ‘boundary crossing actions’.
The notion of ‘unit of analysis’, the context, and the inseparability of individual and collective levels of analysis in sociocultural theory is problematic: learning involves the transformation of the social practice of the group, and therefore analysis cannot in purist terms be reduced to what one person does or knows. Tuomi-Grohn and Engestrom (2003: 30) assert that ‘the individual’s learning is understandable only if we understand the learning of the activity system.’ The focus is on process and activity. However, Sawyer (2002) also notes a spectrum of perspective, and more significantly methodological practice, in socio-cultural theory, ranging from strict inseparability theorists (Lave and Wenger, 1991, Hutchins, 1995, Rogoff, 1997,1998) to positions such as Valsiner’s (1991, 1998) and that of Wertsch, who argues that individual and society are ‘analytically distinct, yet inherently interrelated levels of analysis’ (1994: 203).
Sawyer (op.cit.) notes Valsiner’s observation of Rogoff’s empirical work; that it focuses on ‘a person’s contribution to socio-cultural activity, responsibility, ownership of the activity, (and) relations with other people’ (1998: 353). Sawyer’s point is that the empirical work of inseparability theorists is successful because it implicitly accepts analytical dualism: analysis of properties of individuals, of contexts, and the micro-sociological practice that mediate between the two. Engestrom’s notion of context, and the activity system as a unit of analysis, in their purest forms are likely to cause methodological and analytical difficulties, particularly in attempting to address processes of recontextualisation.
Yet in 2001 Engestrom wrote:
As the contradictions in an activity system are aggravated, some individual participants begin to question and deviate from its established norms. In some cases this escalates into collaborative envisioning…
With Tuomi-Grohn he explains:
Expansive learning is initiated when some individuals involved in a collective activity take the action of questioning the existing practice. (2003: 30)
What roles do these agential individuals have, and who do they represent? What is their relative influence within the ‘collective’, and which discourses do they employ, particularly where boundaries are being crossed? How are boundary objects decided, and what discursive significance do they have? How does such an explanation square with the notion of the ‘unit of analysis’? There is an implied duality between the individual and the collective in Engestrom’s position, which relates to the point made by Sawyer in arguing for analytical dualism.
Perspective, linked to affordances of particular contexts, serves to structure the object. Indeed this could represent the movement from the abstract to the concrete that Engestrom stresses (2001). The generalities at the knot have to be re-situated. Conceptually, this is what I mean by ‘recontextualisation’.
Sawyer’s (2002) work suggests that analytical dualism, at least at the heuristic level of analysis, provides a means of distinguishing while relating individual and collective perspectives and learning, and a means of empirical analysis of development at collective and individual levels, both at the knot, and away from it. Boundary crossing is not simply a recording of what happens in the collective context (this in itself cannot be assumed to be consensual – the outcomes of contestation are likely to be subject to contested hegemonies of discursive practices, some of which may not even be articulated by partners at the knot, owing to social and professional identities and role differentiation); it is also investigating the interpretations back across the boundary and its recontextualisation within the affordances of a different activity system. This position provides a way forward in using Activity Theory to research collaborative development, while accessing the individual aspect of the dualism in partner perspectives on priorities and their development.