| Abstract: | ABSTRACT An enduring concern within management and organization studies (MOS) is how to conduct research from perspectives deemed 'alternatives' to those of functionalism and positivism. Our aim is to address this concern with regard to an approach employed by Karen Legge in research on knowledge workers, namely that of actor-network theory (ANT) (or the 'sociology of translation'). Following an introduction to ANT, the views of some its key proponents, and Legge's own use of the approach, the paper presents critical notes on five issues related to the production of ANT accounts – the inclusion and exclusion of actors; the treatment of humans and non-humans; the nature of privileging and status; the handling of agency and structure; and the nature of politics and power in 'heterogeneous engineering'. We discuss the relationships between these issues and the key ANT goal of achieving a sense of 'general symmetry' in the accounting process. In so doing we note how ANT authors are frequently chastised for either failing to take sufficient account of, or promoting too strong a sense of, analytical symmetry in their writing. It is argued that the primary challenge facing ANT researchers is to produce accounts that are robust enough to negate the twin charges of symmetrical absence and symmetrical absurdity. |
