ABSTRACT
From Human Relations 2001, 54(8): 879-1013 . Copyright © 2001 by The Tavistock Institute. Reproduced by permission of the authors and Sage
Publications.
C om plexity is not only a feature ot the system s we study, it is also a m atter o f the way in which we organize our thinking about those system s. This second-order com plexity invites consideration o f the m odes o f thinking we use to theorize about com plexity, and in this article we develop the idea o f second-order com plexity using Jerom e Brun er’s contrast betw een logico-scientific and narrative m odes o f thinking. Using B run er’s fram ew ork, we exam ine and critique dom inant form s o f thinking about organizational com plexity that are rooted in the logico-scientific m ode, and suggest alternatives based in-the narrative m ode. O ur evidence for the value o f doing this com es from the logic o f com plexity theory itself, which we claim indicates and supports the use o f the narrative m ode. The potential contribution o f the narrative approach to developing second-order thinking about organizational com plexity is dem onstrated by taking a narrative approach to the m atter o f recursiveness. By extension, sim ilar insights are indicated tor other features that logico-scientific thinkers com m only attribute to com plex system s, including non linearity, indeterminacy, unpredictability and em ergence. [ . . . ] '
A central assumption in organization science has been that organization is an intrinsic feature of the social world. Social systems in general, and business organizations in particular, are thought to be organized in one way or another, and it is the task of organization scientists to find out how and why. To this end, two schools of thought can be broadly distinguished. One is sociological-historical-anthropological in orientation; it seeks to produce accounts explaining the specific features of organization^), either employing what Mohr (1982) called 'the variance model' of explanation, or through tracing back the lineage of organizational features to historical-cum-institutional or cultural factors (e.g. Geertz, 1973: Granovetter, 1992). There is a great deal of methodological and theoretical diversity within this school, but there is also a common theme: the social sciences can offer an account of social organization.
