Action research and organisational learning: a Norwegian approach to doing action research in complex organisations

Abstract

The purpose of this article is to present a specific approach to the practice of action research ‘in complex organisations’. Clearly, there are many approaches to the challenge of doing action research in organisations; approaches that are, and also must be, quite context dependent and specific. But my purpose is neither to give an overview nor a recommendation of how action research is or should be done in complex organisations by different schools of action researchers around the world. The approach I will present has grown through practical experience accumulated over many years with doing action research in many different Norwegian organisations with organisational change and development as the specific objective. I will limit myself to an outline of this Norwegian context, and to how I and others have worked specifically with organisational learning both practically and theoretically within or in relation to a broad Norwegian or Scandinavian approach to action research and organisation development represented by many individuals.

Keywords:

• action research
• back stage
• development organisation
• Norway
• organisational learning
• reflective spaces

Acknowledgements

The article is an elaborated version of the author’s keynote speech at the CARN conference in Cambridge, UK, on 5 November 2010. The author has made no attempt to explain different varieties of action research – see Eikeland (2006b), Elden and Chisholm (1993), and Cassell and Johnson (2006). Thanks to Hans Christian Garman Johnsen, Tor Claussen, Carol Munn-Giddings, and Richard Winter for feedback on the text. The responsibility remains the author’s own, as always.

Notes

1. The broader Scandinavian and Nordic (Sweden, Denmark, Finland) context remains mostly implicit in what follows. I am now at the Oslo and Akershus University College(OAUC), outside Oslo. But my personal base for the work presented was the Work Research Institute in Oslo from 1985 until the middle of 2008. I continue my work in a similar vein at the OAUC, trying to establish ‘symbiotic learning systems’ (Eikeland 2009b) between higher educational institutions and the workplaces of their candidates. But this is not the subject for this article. There are other schools of action researchers in Norway more akin to the CARN focus on education, but who barely communicate with or relate to the tradition I present here. See, for example, Tiller (2004) and Furu, Lund, and Tiller (2007). For inexplicable reasons, the silence is mutual.

2. See http://www.arbark.no/diverse/hovedavtalen.htm. For an English version of the agreement itself see http://www.lo.no/Documents/english/Basic_Agreement_2010_2013.pdf.

3. See Thorsrud and Emery (1970). The major inspiration for the semi-autonomous work groups as an alternative to Taylorism and Fordism, came from Trist and Bamforth’s (1951) study. For the continuing discussion about Taylorism and anti-Taylorism, see Pruijt (1997).

4. There is also a challenge of how action research should relate to mainstream social theory. See Berg and Eikeland (2008), and Brøgger and Eikeland (2009).

5. After five years, the SBA (Centre for Better Work Life), established collaboratively in 1988 among almost all employers and employee organisations in Norway to promote workplace democracy, was closed down according to plan. See Davies et al. (1993) and the government report NOU1985:1 (Norges Offentlige Utredninger 1985). A new initiative has been taken through the NOU2010:1 (Norges Offentlige Utredninger 2010).

6. See Brøgger (2007) and Gustavsen et al. (2010) for recent statements of a WRI position within this broader setting. See Ekman et al. (2011) for recent presentations within the broader Scandinavian or Nordic context.

7. Both these and other concepts from Argyris and Schön are, of course, still useful and central; for example, the different modes of communication, espoused theories and theories in use, defensive routines in organisations, reflecting in and on practice, and others. See, for example, Argyris (1990) and Schön (1983, 1987).

8. See Lahn (2011) for a discussion of learning trajectories. See also the use of ‘trajectory’ in Eikeland (2006b).

9. I started working at the WRI in August 1985, from early 1986 in a nationwide industry-level programme (‘bilbransjeprosjektet’ in Norwegian) involving small car dealers and repair shops lasting until 1990. Although the distinction I am most concerned with here has its general sources in several places (cf. Eikeland 2008a, 2008b), it surfaced in a particular form and setting from within this specific programme (cf. Eikeland 1987, 1989; Paalshaugen 1988; Engelstad 1995). Since 1990 I have worked mostly with public organisations focusing explicitly on organisational learning. Other collaborators from 1985 to 1990 have continued working with private firms without such an explicit focus, more concerned with regional development and networks between enterprises. My personal learning trajectory concerning organisational learning and development started with being a student in the early 1970s, however, at an experimental high school based on radical direct democracy where our own work-organisational experimentation was continuously on the agenda (cf. Angell, Draper, and Eikeland 1975; Eikeland 2008a).

10. The experience from the cases and the programme objectives caught the attention of researchers and organisations abroad even more than in Norway and may be said to have spread immediately to Sweden more than within Norway (see Gustavsen and Sandberg 1984). Partly as a response to the diffusion challenges, some researchers at the WRI turned towards the educational system concerned with school development (see, for example, Blichfeldt 1978; Blichfeldt, Haugen, and Jangård 1979). Gustavsen et al. (2010, 58–63) present a retrospective consideration of this diffusion challenge (cf. 2010, 127ff. and 152).

11. ‘Participative’ should be understood as being able to influence results in most of the phases in Table 2 (see Eikeland and Berg 1997, 18–31). Participatory approaches are important in order to secure ownership of solutions, to transcend the division of labour between intellectual and manual work, to bring all relevant perspectives and aspects into the process for the totality to become visible, to ease organisational learning that involves mutual adjustments by all, to facilitate understanding by starting out from where people are and how they experience things, and more.

12. See Gustavsen (1986a), Gustavsen and Engelstad (1986), and Paalshaugen (1986). The term ‘dialogue-conference’ was not introduced in these texts, however. No texts dated earlier than 1988 utilise this term. The term was introduced after prolonged discussions in the research group about the role and character of ‘dialogue’, and after the introduction of the central distinction discussed under the next heading in this article.

13. The term ‘development organisation’ first appears on print quite casually, once in two short 1986 working papers from the ‘bilbransjeprosjektet’ among other terms, as an alternative designation for ‘project organisation’ and reflecting discussions in our project (cf. Engelstad and Paalshaugen 1986a, 1986b). The concept is still impossible to distinguish from an ordinary temporary project organisation. The distinction discussed in the current text is not present, neither in these documents nor in Paalshaugen (1988). In published or semi-published work, the distinction is introduced into the Norwegian work-life context in Eikeland (1987, 1989). To claim that the distinction emerged as a generalisation from distinctions first made in Engelstad and Paalshaugen (1986a, 1986b) – as done by Engelstad (1995, 176) – is wrong. There is no distinction made in the few pages of these occasional working papers.

14. As Morgan (1986, 13–14) says: ‘One of the most basic problems of modern management is that the mechanical way of thinking is so ingrained in our everyday conceptions of organization that it is often very difficult to organize in any other way’. ‘Lean production’, as a work-organisational outcome, summarises many decades of development work in Japanese automobile industries, at Toyota in particular (cf. Womack 1991; Liker 2004). The danger now is that other countries and industries copy this result as a work-organisational blueprint, overlooking its specific generative processes.

15. The importance of the Aristotelian concord or homónoia as being ‘practically of similar minds’ – consisting in and created through lógos or reasoned speech not only in face-to-face relations but within large linguistic and conceptual communities – seems to be underestimated or ignored when social coordination is attempted reduced to secondary ‘mechanisms’ like markets (trading and bartering), hierarchies (power) or networks (alliances) by Thompson et al. (1991). See Eikeland (2008a, 399ff.) on ‘concord’.

16. This theatre terminology was introduced as part of my personal turn more explicitly towards organisational learning in 1990 within a municipality merger in southeastern Norway. This project lasted for six years. It forms the background for Eikeland and Berg (1997). Others, like Gustavsen (1986a, 151ff.), had sided explicitly with ‘political theory’ against ‘learning’ as a ‘superordinate’ perspective, aligning ‘dialogue’ solely with the former. Simultaneously, the terminological distinction between work organisation and development organisation was brought from the local project level at the WRI into the design of the first large national research programme concerned with enterprise development – the BU2000 or Enterprise Development 2000 – in the Norwegian Research Council by more influential action researchers at the WRI (B. Gustavsen, P.H. Engelstad, Ø. Paalshaugen). Without references, Gustavsen and Mikkelsen (1993, 14) claim that this very distinction had become the more usual or common one at the time, making ‘development organisation’ among the most central concepts in the programme. After the year 2000, the BU2000 programme was continued by other programmes (VS2010 and VRI) increasingly focusing on innovation, regional development, and networking. The concept of ‘development organisation’ was re-baptised ‘development coalition’ and expanded in order to accommodate to this changed context (cf. Ennals and Gustavsen 1999). In this perspective, it comes as an unexplained surprise that Gustavsen et al. (2010) do not mention the concept or distinction at all after more than 20 years of R&D where they have played a most prominent role (cf. Engelstad 1995, 1996; Paalshaugen 1998; Gustavsen 1996).

17. I use this theatrical metaphor for organisational learning purposes without commitment to any kind of general ‘role theory’. The theatrical distinction corresponds to ontological distinctions, however, discussed in Eikeland (2008a, 349ff.).

18. Our development work is now normally initiated with a dialogue conference where most employees in a firm or department gather over one or two days to discuss questions specified according to generalities such as ‘where are we?’, ‘where do we want to be?’, and ‘how do we get from here to there?’, working in different group-work constellations, issuing in a number of substantial sub-projects and a main project to pursue after the conference. Re-organising means basically to sort, reconfigure, and relate activities into: a ‘work organisation’, a ‘development organisation’, a ‘decisive organisation’ (organising collective decisions/decisions concerning many), a ‘service organisation’ (organising, for example, cafeteria, housing, etc.), and different temporary ‘project organisations’.

19. This temporary reorganisation for specific substantial purposes of people who were otherwise peers into strictly command-and-obey relationships was institutionalised in antiquity in both democratic Athens and republican Rome through the war-time role of ‘general’ or stratêgos in Greece and through the temporary privilege of imperium in Rome. The early Roman emperors – imperatores – ruined this institution by arrogating for themselves eternal rights of imperium.

20. See Eikeland (1997, 2008a). This professional and critical peer-talk from within a discipline, or ‘community of practice’ (Wenger, 1998) – developing, transforming, and transcending it – is the historical origin of the mystified term ‘immanent critique’ and of ‘dialogue’. Important distinctions like these have never been addressed through the concept of ‘democratic dialogue’ as promoted by Gustavsen (for example, 1985, 1986a, 1992). As indicated by Table 2, it is possible to be ‘democratic’ doxocratically without broad participation; for example, by letting majority voting in large constituencies – or opinion polls – decide the outcome of Phase 4 but without any participation by the same people (or others) in the other phases. It is also possible to have broad participation in all phases but yet with a ‘monarch’ deciding alone in Phase 4. Other combinations and distinctions are also possible. This all makes ‘democratic dialogue’ an unfortunate and somewhat confused mixture of epistemology and politics as in Gustavsen (1986a), where participative democracy is made into a generative mechanism for local social theory (the only kind claimed worthwhile besides general political theory as generative theory) and also a criterion for its validity (see Eikeland 1995).

21. In Norway, the idea of establishing a public sphere within a private organisation as presented in Paalshaugen (2002) goes back to Kalleberg (1984). See also Eikeland (1985a). It was also an objective in my work immediately before starting my work at the WRI (1982–1986). See Eikeland and Gullichsen (1987).

22. See Eikeland (1989), where a concept corresponding to ‘absorptive capacity’ is called ‘receiver competence’ (‘mottaker-kompetanse’).