1M@n@gement recently published an article by Hélène Peton and Stéphan Pezé, entitled “The Unsuspected Dynamics of the Regulative Pillar: The Case of Faute Inexcusable in France” (2014). Based on a historical study of the legal concept of faute inexcusable (inexcusable conduct) in France between 1898 and 2010, Peton and Pezé identify three main tipping points – in 1898, 1941, and 2002 – in the definition and application of the concept of inexcusable conduct. The authors illuminate the political institutional work at the heart of these institutional changes (Lawrence, Suddaby & Leca, 2009). They show that this multifaceted work is not the privilege of the State alone, but also involves legal professionals and members of civil society. Peton and Pezé propose a three-step dynamic model to account for the production of legal rules: An event triggers a period of advocacy work by various groups of interest; when successful, this leads to a change in the legal texts. By underlining the dynamic underlying the coercive pillar of institutions (Scott, 1995), so far neglected by neo-institutionalist research, the authors “highlight a regulative pillar that is ‘ahead’ of the cognitive and normative pillar” within institutional change (page 174).
2In this commentary, we question some of Peton and Pezé’s empirical results regarding both the actors involved and the primacy of the coercive pillar over the cognitive pillar, and we propose more broadly a reflection on the conditions for studying institutional change over long periods. In particular, we point to the fact that a historical neo-institutionalist approach raises two methodological and theoretical questions. The first concerns the delimitation of the boundaries of the field under study. The second concerns the necessity of taking into account dominant collective representations in the analysis of institutional work. We show that avoiding these two matters in the analysis of institutional change over long periods can lead to biased interpretations of institutional dynamics. To that end, we draw on our recent research into the institutional framing of workplace accidents in the French construction industry since 1902; the definition of inexcusable conduct occupies a central position in this framing (Daudigeos, Boutinot & Jaumier, 2013).
How to delimit an organizational field over a long time period
3The first point concerns the delimitation of the organizational field, here understood as “a community of organizations that partakes of a common meaning system and whose participants interact more frequently and fatefully with one another than with actors outside the field” (Scott, 1995: 56). In their study, Peton and Pezé choose to identify the actors involved in institutional work by giving “priority … to legal sources (laws and regulations) and the judicial sources (past rulings)” (page 151). Yet this choice determines the actors rendered visible by their analysis. As Friedland and Alford (1991) underline, the boundaries of a field are hardly identifiable a priori and evolve over time. This plasticity is especially problematic when it comes to longitudinal studies covering a long time period. A way to tackle this problem is to draw on a publication that occupies a central position in the field and whose development of the content thus reflects the evolution of the organizational-field boundaries and thus the changes in the organizations and interest groups that populate it (Hoffman, 1999; Ventresca & Mohr, 2002). Having adopted this approach, our analysis of the archives of Le Moniteur des Travaux Publics et du Bâtiment  from 1902 onwards reveals that the vision of the institutional field of inexcusable conduct in France proposed by Peton and Pezé may be truncated. Peton and Pezé emphasize the work of legal professions before 1941, either in parliament or in the courts, as well as “advocacy” work by employee and employer unions, in order to explain the birth of the 1898 law and its subsequent interpretation. Yet our analysis also reveals the role played by insurance companies in the production of the 1898 law and its subsequent interpretation, a role that is absent from Peton and Pezé’s analysis. The 1898 law represented not only an exceptional suspension of the application of the civil code of 1804, in particular of article 1832 on the reparation of damages inflicted on a third party, but also an important market opportunity for insurance companies. The corpus of Le Moniteur des Travaux Publics et du Bâtiment testifies to the economic stakes of the 1898 law, especially in the context of the debate that opposes private insurance companies, which generate profit for their shareholders, and the syndicats de garantie, mutual insurance companies that work only in their members’ interests (Moniteur des Travaux Publics et du Bâtiment, 1903 n°9 page 1 and n°14 page 1). Given the financial interests at stake, insurance companies became involved in the political battle around the 1898 law and thus in the political institutional work around the coercive pillar, in a similar way to employer and employee unions (Gibaud, 1998). Several secondary sources also confirm the influence of insurance companies on the production of law during this period, in particular in the Social Security and Welfare Commission of the Chamber of Deputies (Commission des assurances et de prévoyance sociales de la Chambre des deputés) (Ewald, 1986; Omnès, 2009). The identification of this new actor is problematical, since Peton and Pezé explicitly formulate their research question as the understanding of “who are the actors who shape the regulative pillar” (page 146). For them, institutional change draws on the contrast between capital and labor (employer unions versus employee unions), and on the dynamics of legal professions and public authorities. The data they choose to analyze lead to a predefinition of the actors involved. In comparison with the dominant understanding of political institutional work, which reduces legal production to two actors (the State and civil society), whose components generally remain fuzzy, this certainly allows them to convincingly open the black box of the State (see figure 8, page 168). However, the authors offer a rather limited vision of the components of civil society. In that respect, taking account of the lobbying of insurance companies in the production of law adds other economic and professional dynamics to their interpretation. The historical study of institutional change thus invites the more systematic mapping of institutional workers, for instance by drawing on a publication that is central to the field, or by diversifying the sources of primary data.
How to take into account the influence of historical collective representations in analyzing institutional work
4Our second critique concerns Peton and Pezé’s insufficient consideration of dominant collective representations in the organizational field, which can influence the institutional dynamics. By “collective representations,” we mean the larger cognitive frames that structure the behaviors and interactions of actors in the field (DiMaggio & Powell, 1983; Mohr, 2000). By focusing more on the coercive pillar of institutions, Peton and Pezé downplay the cognitive pillar, which leads to a biased interpretation of the underlying dynamics of institutional change. In drawing on the results of our study, we show first that taking into account such collective representations minimizes the importance of power struggles between actors involved in institutional work, and second that the effects of these power struggles can be understood only through their embeddedness in larger societal changes. Drawing on our analysis of the historical collective representations at work in the field of inexcusable conduct in France, we offer in the following paragraphs two illustrations of the limits of Peton and Pezé’s proposed interpretation.
5First, our results question Peton and Pezé’s analysis of the stabilization of the definition of inexcusable conduct in 1941 by the “Veuve Villa” ruling from the Court of Cassation. According to them, the “advocacy” of employee unions and accident victims, together with the legal uncertainty created by the 1898 law, would eventually trigger a work of definition by the Court of Cassation (see figure 5, page 160).
6However, this explanation is contradicted by the very low number of cases of inexcusable conduct held after 1941, which seems to point to limited social attention to this question (Jaillet, 1980; Sargos, 2003). If the Veuve Villa ruling were the result of a social struggle, one would instead expect this apparatus, once clearly defined, to be widely used. Our results in fact suggest another explanation for the 1941 turn: This date also corresponds to the stabilization of new dominant collective representations within the field. The spirit of the time became about universal protection of workers and massive investment into technical prevention, rather than about individual responsibility. Workplace accidents were no longer understood as resulting from individual wrongdoings, but as the outcome of complex processes in which work organization and hazards played a large role (Caloni, 1952). This allows us to advance another hypothesis to explain the stabilization of the definition of inexcusable conduct through the Veuve Villa ruling in 1941: It was because the definition of responsibilities became less a question of power between the different parties involved (employee unions, victims, employers, and the State) that the Court of Cassation reached consensus. It was not a power struggle taking place in a context of legal uncertainty that explained the institutional change, but rather the diminution of the political stakes around the attribution of individual responsibility. Our analysis thus leads us to an explanation that is different from that of Peton and Pezé, for whom institutional change comes about only when, “after a successful work of advocacy, the actors implement a political institutional work of vesting or defining” (page 173).
7In addition, taking account of dominant collective representations in the field also allows a better understanding of the historical conditions explaining the resonance that some institutional work may find at some point in time. Our data thus bring a counterpoint to the analysis proposed by Peton and Pezé of the tipping point of 2002, with the redefinition of inexcusable conduct. Certainly, the authors do pay attention to the cognitive pillar, but only as a trigger for this new institutional dynamic. Thus, according to them, “the source of this change was external to the regulative pillar: it was the progressive misalignment between the regulative pillar and the cognitive pillar” (page 160). But, for the dearth of an in-depth analysis of this cognitive pillar throughout the period, they propose an interpretation of the institutional dynamic that is to our mind incomplete. They identify several components of this dynamic: the publication by the European Union of a directive on the responsibility of employers in 1989, several reports by public bodies that underline the inefficiency of the prevention system, the inequity of the compensation system and – finally – the advocacy work by two lawyers in order to obtain indemnities for the victims of asbestos (see page 160 and figure 6 page 163).
8But it is the taking into account of collective representations that gives some consistency to these heterogeneous elements by pointing to a change in the dominant representations as early as the 1980s – that is, a return to a conception of workplace accidents based on individual responsibility, as shown by our study. Thus, the contestation of the compensatory amounts put forth by Peton and Pezé is only the manifestation of a much deeper contestation of the principle of the 1898 law and of the exemption of the application of article 1382 of the civil code on liability to one’s own acts. This exemption had been accompanied by a ceiling on financial compensation in case of accident. Thus, its contestation at the same time actualizes the question of full reparation. This is why a double debate appears, dealing at the same time with the questions of the liability of company bosses and the fair amount for compensation. The European directive of 1989 – as well as the reports of public bodies – fit perfectly with this evolution in the way workplace accidents are understood. The judicial strategy in the case of asbestos took place in this exact context, which provides it with extraordinary resonance. It reached its target at the end of the 1990s and led to regulatory change, although other work-related diseases that were just as harmful, such as silicosis (Rosental & Devinck, 2007), had not received the same attention in other periods. The change in collective representations thus influences all steps of institutional change and cannot be ascribed to the sole role of a trigger.
A perspective for the historical study of institutional change over long periods
9This commentary proposes a different interpretation from Peton and Pezé’s and, at the same time, raises the wider issue of the historical study of institutional dynamics. Without taking sides with an a priori primacy of structures over agency, our commentary underlines the necessity of systematically putting the analysis of institutional work into perspective with a historical study of the field and its dominant collective representations, so avoiding the risk of a biased interpretation of causal links. In the case of Peton and Pezé’s study, the absence of a systematic consideration of the changing boundaries of the field and collective representations leads to the actors and their rationalities being arbitrarily defined, and the role of power struggles in institutional change being overemphasized. On the contrary, we concur here with Suddaby, Foster, and Mills (2013) in defending the idea that a historical approach should conceive of institutional change as a phenomenon anchored in a specific time and space, which requires careful consideration of the actors involved and their representations. As the perspective of embedded agency asserts (Battilana & D’Aunno, 2009; Daudigeos, 2013), agency and structure are intimately linked. The cognitive pillar is not an exogenous variable to institutional change, intervening only as the trigger of the process. The three pillars must be thought of as subtly interlinked, albeit sometimes out of step with each other, at all stages of institutional change in order to account for complex causalities within the process. In the case of the present commentary, this is actually the confrontation of two empirical materials – that on institutional work and that on the dominant field representations – which allows to elaborate a historical narrative (Carr, 2008; Suddaby, Foster & Quinn Trank, 2010) and illuminate the institutional dynamic over a long period.
Founded in 1902, Le Moniteur is the most widely distributed trade journal in France. It targets all professionals in both public-sector works and the construction industry, providing them with information covering topics as diverse as regulation, results of architectural contests, invitations to tender, latest technological innovations related to construction, and (more recently) corporate social responsibility and sustainable development.