Mixing methodologies and paradigmatic commensurability

My first critical awareness of the tensions between different methodologies arose 5 years ago when preparing the materials for two undergraduate courses: ‘Operations Management’ and ‘Strategy’. The former was concerned with explanation in which there is the assumption of causality; the operations environment could be designed in an optimal manner with predictable behaviour—the domain of those following an ‘objectivist’ epistemology. The latter relied upon interpretation and understanding, with the only certainty being that you would never know if you had a ‘good’ understanding and had made the best decision, but you would always know when you were wrong when everything fell apart, unless you were lucky— the domain of those following an ‘constructivist’ epistemology. The second event that resurrected this awareness of tensions occurred when attempting to write about the different approaches to organisational change and the use of ‘soft’ approaches to problem solving. Having been grounded in the work of Stafford Beer (Viable Systems Model [VSM]; Beer, 1979, 1985) and Raul Espejo (The Cybernetic Methodology; Espejo, 1988, 1990, 1992) I have found myself, over the last decade, tacitly using this work in a variety of organisational change programmes. Thus, in the process of writing about this, I have been challenged by the need to reconcile what are viewed as epistemological or paradigmatic incommensurabilities. This has attracted me to a series of discussions within Journal of the Operational Research Society, which appear to encapsulate this argument. Drawing upon these as well as other arguments, I will attempt to explain my own reconciliation of this debate. The conventional approaches to change management can be viewed as ‘hard problem’ orientated; the problem is clearly established, with the concern being about which course of action should be adopted. The context of implementation is viewed as unproblematic. However, it is proposed here that change management efforts, both large and small, entail a level of complexity which may be tacitly acknowledged and even addressed, but not in an effective manner. This complexity arises due to the social context within which change takes place and the presence of a variety of stakeholders. The inadequate handling of this complexity increases the likelihood of hindrances to change and a less than acceptable outcome. From the viewpoint of the management of change, this complexity raises itself as problematic, with the dilemma that it is not clear what the issues are. The revealed complexity associated with the intended change shifts its handling from a ‘hard’ orientation to a ‘soft’ orientation as discussed by Checkland (1981). This invites ‘soft’ orientated problem structuring methodologies (eg Soft Systems Methodology [SSM], Critical Systems Thinking, Cybernetic Methodology) as a complement to more conventional approaches. However in doing so, this raises an issue pertaining to the change manager and whether it is commensurate to mix ‘hard’ and ‘soft’ approaches. The incommensurability of methodologies is raised by those purists who argue that a given epistemological stance translates into an appropriate methodology (Jackson and Carter, 1991). Thus, the positivist tradition that is associated with ‘hard’ approaches contrasts with the phenomenological tradition of the ‘soft’ approaches. In other words, the way we think about society and how knowledge comes into being determines what we can observe and know about society. Thus, it is inappropriate to use methodologies from different epistemological traditions together. This message has been clearly articulated by Burrell and Morgan (1979) who call for ‘paradigmatic closure’ (ibid.: 398) in the context of their description of four idealised sociological paradigms, where a paradigm is defined by its meta-theoretical assumptions and is thus exclusive. Indeed one summation from a two day ‘Alternative Paradigms Conference’ held in 1989 was that ‘the participants agreed that homogenisation of paradigms should not and could not occur’ (Dobbert, 1990, p 288). Guba (1990, pp 370–371), in his summation of this conference, states that ‘paradigms . . . are basically incommensurable. That is, there probably does not exist some fundamental, rational framework to which all paradigms can be reduced so that conflicts and inconsistencies can be resolved’. He also postulates that further discussion may give rise to ‘an as yet unimaginable paradigm (dare I say, metaparadigm?)’ (ibid.). However, the view that methodologies are incommensurate is questioned. Gioia and Pitre (1990), while acknowledging the incommensurability of core assumptions argue that the boundaries are ‘ill-defined and “blurred” ’ Journal of the Operational Research Society (2011) 62, 806–813 © 2011 Operational Research Society Ltd. All rights reserved. 0160-5682/11