Little guidance has been given to the management of action research and this paper suggests that ideas from the actor-network theory (ANT) may be helpful. Although ANT has been widely used as a tool for interpretation and description of the interplay of the social and technical aspects of organizations, this paper presents a different use of the ideas. It advocates a proactive use, to bring about changes in a situation rather than merely observing and describing that situation disinterestedly from the outside. The case is described of a long-term action research study carried out within a corporate bank, where the ideas of ANT were employed in this way, as a device for understanding the turbulent social and political context of the research, to plan and manage the intervention process, and where ideas such as networks of interest, translation of interests and black boxes were used to practical effect. Copyright # 2007 John Wiley & Sons, Ltd.Keywords action research, actor-network theory, information systems, research methodology
INTRODUCTIONCollaborative forms of research, particularly action research, are well established ways of advancing the theory and practice of systems thinking but their use can be problematical. In such research the researcher becomes necessarily involved in managing relationships and change in the real-world problem situation, for which they may be ill-prepared. For example, setting up any collaboration can be a protracted and difficult process, requiring social and persuasive skills different from those normally expected of an academic researcher. Researchers may be uncomfortable in performing the negotiation rituals and manoeuvres required to initiate the research and the general conduct of the research may require the type of organizational and persuasive skills more normally required of the management consultant (Clark, 1995(Clark, , 2001Wickham, 1999). Further, on occasions an active role in organizational politics may be called for, to enable the research to happen and to ensure that it continues. It is surprising then that the neither the systems literature nor that concerning action research gives much attention to the management of the collaboration and the researcher's role vis-a-vis the real-world situation. It is with this that this paper is concerned, together with the contribution that can be made by ideas taken from a particular sociological stream. The interdisciplinary roots of management research mean that there is continuing interest in the potential for applying sociological theories. However, the marrying of the two is not always without problems. Where the management topic is fairly recent and has not had time to develop a strong theoretical base of its own then discrepancies in scope may be obvious. Oft-used theories such as the actor-network theory (ANT) (Callon, 1986;Latour, 1987) or structuration theory (Giddens, 1984) are noticeably different in kind to, for example, perspectives used in information systems research such as object orientation, or the simple stan...