The organization (Ângkar) as a state of exception: the case of the S-21 extermination camp, Phnom Penh

Abstract

Organization theory, Clegg pointed out, has failed to address the role of organizations in some of the crimes of/against humanity, suggesting that more attention should be given to the case of total institutions. With this paper we respond to Clegg’s invitation and study the S-21 extermination camp, in Phnom Penh, Cambodia. We do so by engaging with the work of the Italian philosopher Giorgio Agamben, with the aim of investigating the organizational patterns that constitute the camp as a ‘State of Exception’. Doing so shows us how organizations can become malign forces for evil. We explore the implications of this case for more general ‘Kafkaesque organization’, that sometimes reproduce, in more benign forms, many of the practices found at S-21.

Keywords:

• total institutions
• evil organizations
• Cambodia
• Khmer Rouge
• S-21
• Kafkaesque organizing
• state of exception
• bios
• zoe
• bare life

Acknowledgements

We are grateful to the Editor and our reviewers for their comments and suggestions. Miguel Cunha gratefully acknowledges support from Nova Forum.

Notes

1. Of course, in the broader social science literature, the work of scholars such as Bauman (1989) and Arendt (1968) is well known. In one of those effects of professionalization and disciplinization, however, the literature of organization studies (defined as the contributions to journals such as Organization Science, Organization Studies, Organization and Administrative Science Quarterly) has been very largely disengaged from accounts of the Holocaust and other genocidal situations. There are exceptions, such as Grey (2005), which explicitly engage with the Holocaust, as well as Madsen and Willert (1996), du Gay (2000), Armbruster and Gebert (2002), and Ten Bos (2007), and more recently, the important work of Stokes and Gabriel (2010) and the unpublished work of Ek et al. (2007), as well as Clegg (2009). The point remains though: study of genocidal organization is not considered a major part of organization studies.

2. The influence of French Marxism on the Khmer Rouge is often mentioned as decisive, although Short (2004) gives short shrift to this idea. Pol Pot was simply too minor a figure and too un-intellectual to be much involved in the labyrinthine politics of the French left (see Badiou 2008). Anyway, most of what occurred in French Marxist circles, such as those the Cambodian nationalists moved in, would best be thought of as dogma rather than theory (Majumdar 1998), despite pretensions. Short sees the ideological influences as far more local and home-grown.

3. To analyse the data, we read the book twice, the first time without any coding attempts, the second for coding purposes. After that, we often returned to the book to check for evidence and meaning. We tried to limit codes to descriptions of S-21 as an organization rather than about other contents of the book, such as chronologies, biographies, political contextualization and so forth. Our intention was to study the nature of this organization, described as Kafkaesque. As such, all the elements that were not descriptive of the organization were not considered. We coded a total of 76 initial entries that were reduced to 20 first-order codes. These codes provide direct descriptions of S-21 as an organization. They offer relevant information leading us to extract second-order themes from constant comparison (Glaser and Strauss 1967), an attempt to make sense of the data at a higher level of theoretical abstraction. The attempt to extract theoretical meaning from comparing codes and aggregating them at a higher level resulted in six second-order themes related in three opposing pairs: sensegiving and senseblocking, purity and disease, omnipotence and powerlessness. These pairs were already a result of the comparison between categories and the text, and of interpretation, with successive readings of the first-order codes making these tensions apparent. In the third stage, we related the tensions emerging from the data into three paradoxes. The first, the sensemaking paradox, collapses two themes: sensegiving and senseblocking. The second, the Manichean world paradox, articulates the tension between a world where purity and disease were viewed as opposing but shifting categories. The third sets in tension the interplay between omnipotence and powerlessness. These three dimensions constitute overarching concepts that provide a theoretically deeper analysis of organizing forces at play in S-21. Semiotic clustering reduces a very complex organization to a number of deep structural characteristics that help to increase abstraction and to gain analytical sophistication.

4. These were derived from second-order themes, which in turn resulted from the first-order concepts that were directly taken from the source. For the sake of theoretical elegance, we assumed that overlaps could be rendered acceptable through the clarification of the meaning of the final categories.

5. The originator of the term ‘sensemaking’ always renders it as a single rather than hyphenated compound (Weick 1993). We shall follow his lead in this and in coining other ‘sense’ based neologisms.