On interdisciplinary movements: The development of a network of support around Foucaultian perspectives in accounting research

Abstract

This paper seeks to better understand interdisciplinary movements in the making. Our investigation focuses on the processes through which a network of support surrounding Michel Foucault's ideas originally developed in the sociological and organizational stream of accounting research. Drawing on the sociology of translation, we first examine how a network of support emerged around the journal Accounting, Organizations and Society (AOS), which is generally perceived as the main vector of dissemination of sociological and organizational accounting research. Our investigation then focuses on how Foucault's ideas, a few years after the founding of AOS, came to the attention of a group of accounting academics in the UK – a group in which the editor-in-chief of AOS was a key actor. We also examine how a network of support surrounding Foucault's ideas subsequently developed in the greater accounting research community. Our analysis emphasizes the role of epistemological uncertainty in the constitution of networks of support around journals and ideas, and the role of trials of strength (Latour, 1987) in fuelling or mitigating this uncertainty, thereby influencing actors' interests and commitments to particular networks. Our analysis also highlights the critical role that imitation and social differentiation play in the travel of ideas between scientific fields and the creation of scientific knowledge.

Acknowledgements

We would like to acknowledge the comments made on prior versions of the paper by Peter Armstrong, Salvador Carmona, David Cooper, Cameron Graham, Anthony Hopwood, Anne Loft, Kari Lukka, Keith Robson, Brian Shapiro and two anonymous reviewers. We also acknowledge the comments made by participants at the 2002 Critical Perspectives on Accounting Conference (New York), the 2002 World Congress of Sociology (Brisbane), the 2003 Interdisciplinary Perspectives on Accounting Conference (Madrid), and the 2004 Alternative Accounts Workshop (Calgary). Of course, the paper is only reflective of the authors' point of view. Thanks to Jake Frederick for research assistance.

Notes

¹ Several categorizing schemes are used in the literature to classify Foucaultian perspectives. On the methodological level, one may distinguish between archeological and genealogical perspectives (Armstrong, 1994). On the substantive level, Foucaultian perspectives may be distinguished by the extent to which they deal with concepts such as disciplinary power (e.g. Foucault, 1977), technologies of the self and avowal (e.g. Foucault, 1988), and governmentality (Foucault, 1978b).

² Even at the end of the 1970s, when the English version of his most widely known books had just recently been released (i.e. Discipline and Punish in 1977 and History of Sexuality in 1978), Foucault was prominent on A&HCI. For example, Foucault is the 24th most heavily cited 20th-century author on A&HCI for the years 1977 and 1978 (Megill, 1987). For the period 1977–83, Foucault's rank increased, being the 13th most heavily cited 20th-century author on A&HCI (Garfield, 1986).

³ Search on SSCI and A&HCI was performed using the author's last name, and the initial of his first name followed by an asterisk. The results may therefore be overstated due to different individuals having the same last name and first-name initial. However, the impact is likely to be minimal given the notoriety surrounding the intellectuals listed above.

⁴ Given the reservations that Latour (1999b) expressed about the term ‘actor-network theory’, we have decided to use the term ‘sociology of translation’. Latour argues that the recent popularization of the word ‘network’ implies transport without deformation – contrary to his ontological stance. In addition, Latour objects to the use of the word theory. According to Latour, the sociology of translation is a method that can be used to learn from the actors about what they do, how and why they do it, without imposing on them sophisticated a priori categorizations.

⁵ While behavioural papers adhere to the positivist research tradition, sociological/organizational papers are more in line with the interpretive-critical research tradition (Chua, 1986). The sociological and organizational stream of accounting research is also called ‘interdisciplinary research’.

⁶ However, it is worth noting that in the time period investigated we found two behavioural papers (respectively published in 2001 and 2003) that refer to Foucault, thereby pointing to (a modest level of) cross-paradigmatic influence.

⁷ We found several papers that mention Foucault (or ‘Foucaultian’ or ‘Foucauldian’ research) in the text but whose reference section does not include any work of Foucault. Each of these papers was considered as having made one reference to Foucault.

⁸ In order to validate whether our results are representative of the extent to which Foucault is cited in other accounting journals that publish sociological/organizational papers, we performed a citation analysis of sociological/organizational articles published in Accounting, Auditing & Accountability Journal (AAAJ) and Critical Perspectives on Accounting (CPA), for the years 1994, 1998 and 2001. The results show that the proportion of sociological/organizational papers in AAAJ that refer at least once to Foucault was 40% in 1994, 22% in 1998 and 13% in 2001. Corresponding results for CPA were 39% in 1994, 26% in 1998 and 22% in 2001. Overall, these results suggest that in a manner similar to the AOS papers we analyzed, a significant number of authors of sociological/organizational articles published in AAAJ and CPA refer to Foucault. However, it is worth noting that the proportion of papers that cite Foucault in these two journals appears to be decreasing over time, especially in AAAJ. We did not find such a trend in AOS.

⁹ The citation analysis did not allow us to develop a precise idea of the importance ascribed to Foucault in papers that refer to his writings. In this respect, Armstrong (1994) argues that Foucaultian perspectives have played a central role in the formulation of the theoretical frameworks that a number of accounting researchers have employed in their own work.

¹⁰ We have assumed that the first author of a jointly authored article is generally the author who contributed most to the study. Our assumption obviously hides the contribution of co-authors whose names are not listed as first author. However, given the relatively small percentage of articles that refer to Foucault where the co-authors are from different countries (i.e. 12.1%), it is unlikely that our results would be changed to a significant extent if we had relied on other calculative assumptions.

¹¹ The field of financial accounting research provides a good illustration of the flow of imitation that takes place within a field once an innovative paper is published. Numerous accounting researchers became involved in capital market research after the publication of the study by Ball and Brown (1968), who relied on the capital asset pricing model originally developed in finance to investigate the relationship between unexpected earnings and the stock price reaction around the time of earnings announcement (Watts and Zimmerman, 1986).

¹² We do not focus in this paper on the institutionalization phase. Our empirical inquiry relates instead to the time period in which a network of support in accounting research around Foucaultian perspectives emerged and initially consolidated.

¹³ For critical (and asymmetrically predicated) analyses of Foucaultian research in accounting, see Neimark (1990) and Armstrong (1994).

¹⁴ The memory-transcript method consists of noting key aspects of the interviewee's answers during the interview. Then, drawing on these notes, a detailed recollection of the interview is written shortly afterwards.

¹⁵ According to interviewees, the newsletter ceased publication in 1978.

¹⁶ Dyckman and Zeff (1984) refer to this prior paradigm as normative research, in which researchers were concerned, for example, with policy prescription.

¹⁷ We find it ironic that one of the journals that is nowadays commonly perceived in accounting academia as a ‘top’ outlet (Ballas and Theoharakis, 2003) owes its birth to a person who would later be involved in a significant corporate scandal, having secretly diverted millions of dollars from two of his companies and from employee pension funds in an effort to keep his business solvent (www.infoplease.com, 2004).

¹⁸ It seems that all editors of the journals owned by Pergamon Press were required to write an editorial that highlighted the 40th anniversary of the company and the 65th birthday of its founder.

¹⁹ TAR's reluctance to publish certain types of research around the turn of the 1970s is discussed in Dyckman and Zeff (1984).

²⁰ Lukka and Kasanen (1996) maintain that an accounting research élite centred around AOS emerged over the period 1984–93. They also argue that the presence of the AOS élite may benefit the accounting research community – in that it offers an alternative to the epistemologies sustained by US-based accounting journals.

²¹ One interviewee mentioned that the tensions between the two streams of research have mitigated over time, because both psychologists and sociologists realized that there are ‘bigger enemies to worry about in the development of accounting ideas’ (Interviewee B, North American academic, May 2003).

²² An understanding of the emergence of Marxist accounting research is available in Roslender and Dillard (2003). When appropriate, we refer below to points of intersection in the development of networks surrounding Foucaultian and Marxist research in accounting.

²³ Our data indicates that some members of the journal's editorial board had contacts with people working with Foucault or Foucault himself.

²⁴ Two papers published in AOS resulted from the work of this team (i.e. Burchell et al., 1980, 1985) – each of these papers made one reference to Foucault. Furthermore, Hopwood (1987) acknowledges that his paper benefited ‘enormously’ from discussions with members of the team. It is worth noting that one team member mentioned that the reference to Foucault as found in Burchell et al. (1980) does not ‘particularly prioritize’ Foucault, being included in a footnote. At the time, members of the multidisciplinary team were apparently in the process of familiarizing themselves with his work: ‘We were having reading seminars ourselves, reading Popper as well as Foucault and looking at some of the similarities as well as their differences.’

²⁵ Informants mentioned that participants to the workshops included: Paul Bircher, Philip Bougen, Stuart Burchell, David Cooper, Jeremy Dent, Anthony Hopwood, Anne Loft, Brendan McSweeney, Peter Miller, Ted O'Leary, Alistair Preston, John Roberts, Keith Robson and Grahame Thompson. It seems that some of these individuals participated quite regularly over a span of time, or throughout all the time in which the workshops took place. Others were involved on a more irregular basis.

²⁶ According to informants, participants in the seminars included: Stuart Burchell, Colin Clubb, Anne Loft and Ted O'Leary.

²⁷ However, one participant mentioned ‘resisting’ using Foucaultian perspectives in his own research for several years.

²⁸ The degree of influence exerted by the editor of AOS that we document in the present paper is not unique. Dyckman and Zeff (1984) argue that proprietary journals (like JAR) have powerful editors and transmit signals concerning the type of papers that they prefer to receive.

²⁹ The quotation marks surrounding the word ‘fact’ reflect our fundamental doubt concerning the epistemological appropriateness regarding the selection of a set of journals that supposedly includes most of the relevant knowledge in a given area, thereby marginalizing publications in journals unlisted in the database.

³⁰ Data obtained from the American Accounting Association indicates that total membership of the Accounting, Behavior and Organizations Section comprised, at the time in which it was created (May 1982), ‘over 200 members’. Membership grew to ‘over 300’ in October 1982, 810 in 1988 and 1,252 in 1990.

³¹ For example, prior versions of Hopwood (1987) and Miller and O'Leary (1987) were presented at the 1984 University of Wisconsin Conference (AOS, 1987). The materials distributed to participants during the first IPA Conference indicate that a prior version of Loft (1986) was presented.

³² Mayer Zald was on the editorial board of AOS from 1985 until 1993.