Constructing a research network: accounting knowledge in production

Purpose - This paper contributes to the sociology-of-science type of accounting literature, addressing how accounting knowledge is established, advanced and extended. Design/methodology/approach - The research question is answered through the example of research into linkages between accounting and religion. Adopting an Actor-Network Theory (ANT) approach, we follow the actors involved in the construction of accounting as an academic discipline through the controversies in which they engage to develop knowledge. Findings - We show that accounting knowledge is established, advanced and developed through the ongoing mobilisation of nonhumans (journals) who can enrol other humans and nonhumans. We show that knowledge advancement, establishment and development is more contingent on network breadth than on research paradigms, which appear as side-effects of positioning vis-à- vis a community. Research limitations - In our analysis, we followed humans and were able to let them share their strategies with us and validate our ex post facto reading of their papers. We were unable to do the same with nonhumans because of their intrinsic properties. Practical implications - This paper provides scholars with analytical tools that could help them position their research projects within a scientific network and understand the need for interactions with other actors in establishing, advancing and developing knowledge. Originality value - The originality of this paper is twofold. Firstly, we apply ANT to accounting knowledge, whereas the accounting literature applies it to the spread of management accounting ideas, methods and practices. Secondly, we develop an original methodology for data collection by inviting authors from the network to give a reflexive account of their writings at the time they joined the network. Well diffused in sociology and philosophy, such an approach is, albeit, original in accounting research.