CAIRN-INT.INFO : International Edition

1 Deinstitutionalization, the process by which a taken-for-granted practice is eroded until it disappears completely, is a key stage in the life cycle of institutions (Lawrence et al., 2001); however, it is relatively rarely studied in the literature (exceptions include Chaudhr y & Rubery, 2017; Clemente & Roulet, 2015; Hiatt et al., 2009; Maguire & Hardy, 2009; Oliver, 1992). When it is studied, the focus is chiefly on its final phase and the exogenous shocks that trigger that phase (e.g., Maguire & Hardy, 2009). Yet, deinstitutionalization is a long, complex process, which is characterized by a battle between different types of institutional work (Lawrence et al., 2009), in which disruptive and defensive forms of action come into conflict. Consequently, its trajector y is chaotic rather than linear, studded with phases of rapid change, institutional reinforcement or slow erosion (Clemente et al., 2016; Hargadon & Douglas, 2001).

2 Some research studies exploring the temporal dynamics of institutions (particularly Lawrence et al., 2001) stress the role of actors in the pace of institutionalization, but to our knowledge, no research study has yet explored their role in the acceleration or deceleration of deinstitutionalization. In this study, we analyze action profiles that differ in the form of agency involved, and the way the action taken combines with other actions in a field. We aim to examine more closely the complex interactions that produce temporal effects by crossing agency with forms of resource mobilization, both central concepts in the dynamics of institutional fields (Battilana & D’Aunno, 2009; Dorado, 2005).

3 This research study specifically concerns the case of asbestos use in France, which offers a good illustration of the long, complex process of a practice’s deinstitutionalization. Asbestos was the ‘star’ mineral of post-World War II reconstruction: with its undeniable insulating capacity and its extremely attractive cost, it was universally used and became ‘the travelling companion of industrial capitalism’ (Malye, 2004, our own translation). The fact that this traveling companion was extremely dangerous was later revealed by a significant social mobilization but disputed and downplayed by the lobby for what was sometimes nicknamed ‘white gold’.

4 To understand how the actions driven by different actors in the field influenced the pace of deinstitutionalization of asbestos, we compiled and analyzed a list of institutional actions that took place between 1970 and 1997 (e.g., taking an approach similar to Toubiana & Zietsma, 2017). Our corpus is drawn from archives (press articles, audiovisual archives, reports on National Assembly (French parliament) debates, books, press releases, pamphlets, and a range of professional reports) and from primary data collected from experts, workers, lawyers, epidemiologists, and trade unionists. A timeline representing the pace of deinstitutionalization is established based on how far the core beliefs about asbestos were weakened (or destroyed).

5 This research article looks at deinstitutionalization from a fresh angle, highlighting the long and short phases that make up this lengthy process until the final crisis. In particular, we bring out the efforts made by actors defending a practice to keep deinstitutionalization on a long timescale. We develop the idea that the acceleration phases of deinstitutionalization prior to the final urgency-driven crisis are often short, explaining this by the disruptive actors’ peripheral position in the field. Finally, the case of asbestos use in France shows the importance of collective reflection groups, which are not the same thing as field-configuring events, as a lever for slowing down and potentially even halting deinstitutionalization.

Theoretical framework

The institutional work of deinstitutionalization

6 Deinstitutionalization, the last stage in the life cycle of institutions, is the process by which practices that were previously taken for granted are finally abandoned (Davis et al., 1994; Maguire & Hardy, 2009; Oliver, 1992). It is often studied not only as a precondition for creation of a new institution (Burns & Wholey, 1993; Leblebici et al., 1991; Rao et al., 2003; Schneiberg & Lounsbury, 2008) but also as a process in its own right (Clemente & Roulet, 2015; Delacour & Leca, 2011; Maguire & Hardy, 2009). The seminal research on this process has concentrated on antecedents and exogenous factors, explaining the decline of an institutionalized practice (Davis et al., 1994; Oliver, 1992). Studies on the institutional work of deinstitutionalization, that is, ‘the purposive action of individuals and organizations aimed at […] disrupting institutions’ (Lawrence & Suddaby, 2006, p. 215), have identified several forms of specific actions: disconnecting the system of rewards and sanctions (Jones, 2001; Leblebici et al., 1991), deconstructing the moral foundations of the institution (Ahmadjian & Robinson, 2011), and undermining the assumptions and beliefs inherent to an institution (Leblebici et al., 1991; Wicks, 2001). Maguire and Hardy (2009) extended this approach by demonstrating the existence of disruptive discursive work.

Action inside an institutional field

7 The institutional field is a particularly useful level of analysis regarding institutional work (Reay & Hinings, 2005; Zietsma et al., 2016). This article understands the institutional field as defined by Hoffman (Hoffman, 1999; Wooten & Hoffman, 2008): an institutional field forms around a common institutionalized practice – for example, the use of asbestos or the introduction of terrestrial television in France (Ben Slimane, 2012). Analytically, an institutional field can be studied by identifying the actors that interact in relation to a common issue, and whose actions may have reciprocal effects. The institutional field is thus a space of rivalries and struggles, occupied by actors pursuing their own interests. Additionally, the institutional field is a space of constraints: interaction capacities, norms, values, and ways of thinking are still essentially shaped by institutionalized factors.

8 Studying an institutional field thus also requires understanding the efforts made by actors to shape the area – and therefore, the space of constraints – in which they operate. At the level of the instrumental field, deinstitutionalization involves divergent efforts: intentional defensive action to maintain the institution, and intentional disruptive action to bring about significant change in an institution, or possibly even destroy it (Lawrence et al., 2009).

9 We use two key concepts to understand the efforts of actors in an institutional field: (1) agency, highlighting the actors’ capacity for action in a space filled with constraints; (2) the resource mobilization process, accounting for the way the resources in a field – cognitive, social, or material – are captured through intervention by certain actors.

10 (1) Agency, in the sense of actors’ ability to question the institution, is central to institutional work (Dorado, 2005; Heugens & Lander, 2009; Hirsch & Lounsbury, 1997; Lawrence et al., 2009; Mahoney & Thelen, 2010), and is often conceptualized by reference to the seminal work by Emirbayer and Mische (1998). They identified three dimensions of agency: the iterational, projective, and practical-evaluative dimensions, depending on the temporal orientation of the actions, towards the past, present, or future. When an effort is past oriented, seeking to revive old schemas of thought and action, the form of agency involved relates to routine. When an effort is more present oriented, focusing on practical judgments to solve the problems that arise in a specific situation, then pragmatic agency is at work, which is aimed at sensemaking. Finally, future-oriented efforts founded on vision and projective construction of action trajectories, involve strategic agency, thus, facilitating action for medium- to long-term purposes.

11 (2) Resources, whether social, cognitive, or material, are crucial for institutional change (Dorado, 2005). In most cases, the actors do not possess sufficient resources to change an institution by themselves. Consequently, institutional change requires mobilization of a number of actors who together possess the necessary resources to achieve change. Dorado (2005) proposed a typology comprising three categories of resource mobilization: leveraging (seeking out support), accumulating, , and convening (creating groups to reflect collaboratively on the issue). The classic form of leveraging starts with a strategic actor, who defines a cause, and then sets out to rally other actors to the cause. This draws on a range of competencies – rhetorical, political, and social – that gather people to a shared vision. Accumulation is obser ved in cases where institutional change is driven by a juxtaposition of varied resources, possessed by uncoordinated actors, who may have multiple and more ambiguous goals. The accumulation process can take place over long periods. Finally, convening occurs when the resources in the field must be pooled to collectively solve complex problems. In such cases, institutional change depends on arriving at solutions that are collectively acceptable in an institutional field. Particular examples are setting up a sector-specific task force, or a body with equal representation of all stakeholders to compare different viewpoints and propose convergent solutions.

The pace of deinstitutionalization: An underexplored dimension

12 Efforts to bring about deinstitutionalization induce uncertainties and discontinuities (Hargadon & Douglas, 2001; Lawrence & Suddaby, 2006; Rao et al., 2003), which call for subtle understanding of temporal dynamics over the long term. Regarding institutional dynamics, in general, the theory of gradual institutional change (Mahoney & Thelen, 2010) and the stalactite model of change (Djelic & Quack, 2003) highlight that change is often incremental and consists of a succession of few-and-far-between episodes of sudden change, alternating with much quieter periods (Lok & De Rond, 2013) when change – if any happens – is particularly slow. McAdam and Sewell (2001) similarly observed that institutional change is marked by long-term change processes and short cycles. Granqvist and Gustafsson (2015, p. 1010) developed the idea of temporal institutional work, exploring how the actors ‘construct, navigate, and capitalize on timing norms in their attempts to change institutions’, and thus, free themselves from temporal embeddedness (Battilana & D’Aunno, 2009).

13 These dynamics seem to be particularly important for deinstitutionalization. Disr uptive efforts to challenge a legitimate practice are generally thought to occur over a short period and aim for rapid deinstitutionalization, sometimes presented as a matter of urgency. However, this acceleration dynamic can be blocked by defensive tactics. The existing institutional literature does not yet explain the effects of such interactions on the acceleration or deceleration of deinstitutionalization. How do disruptive actor s tr y to step up the pace of deinstitutionalization to facilitate resource mobilization? In response, how can defensive actors slow down the pace of deinstitutionalization? By coupling the concepts of agency and resource mobilization, we connect the time horizons of actors with forms of interaction in an institutional field and account for the variable pace of deinstitutionalization in a contested field.

Research method

Research setting

14 Falling from its status of ‘wonder mineral’ to that of ‘public enemy number 1’, asbestos was banned in France in 1997 after a lengthy, discontinuous process of delegitimization. The great controversy over asbestos in France makes it a particularly instructive illustration due to the slowness of the process, the intensity of the struggles, and its public nature. This study of a single case is motivated by an exploratory approach (Eisenhardt, 1989; Yin, 2013).

15 The case of asbestos is treated as an institutional field (Hoffman, 1999) formed around asbestos-related practices in France. The method used in this study is summarized in Table 1. We study the institutional actions involved in the deinstitutionalization process for asbestos in France between 1970 and 1997, and changes in beliefs about asbestos over the same period. We collected a rich body of empirical material drawing on many sources of data, both primary and secondary.

16 For secondary data, we referred to a large number of sources (press, public reports, pamphlets, books, reports by the National Assembly, etc.), as detailed in Table 2. We also studied existing writings about the history of asbestos (Chateauraynaud & Torny, 1999; Lenglet, 1996), which includes additional relevant archival data. In addition, we were given certain documents during the interviews. We collected primary data from 27 interviews, including some with contemporary actors in the deinstitutionalization process (doctors, lawyers, trade unionists, etc.; see Table 2).

Data analysis

17 There were two principal objectives to our data analysis: (1) determining the pace of deinstitutionalization, that is, how beliefs about asbestos changed and the speed of those changes; (2) identifying the principal institutional actions that took place during the period studied and examining their possible impact on the pace of deinstitutionalization.

Analysis of the pace of deinstitutionalization

18 We set out to identify the principal changes in beliefs about asbestos and the speed of those changes. We drew on a rich and varied corpus of secondary data (writings about asbestos, press or media coverage, pamphlets, formal complaints instigating lawsuits, union, institutional, and personal archives; see Table 2). Such triangulation of data sources is particularly important in a longitudinal study (Daudigeos et al., 2015; Peton & Pezé, 2015). We also incorporated data collected from interviews, particularly with experts, to refine our understanding of the field (see Table 2). Some of the analysis was assisted by the Prospero lexicometry software, originally developed for longitudinal textual analysis of controversial issues (Chateauraynaud, 2003). The Prospero software has been used in many past research studies (Cointet & Parasie, 2018), notably in sociology (Boltanski & Thévenot, 2006; Chateauraynaud, 2014), political science (Parasie & Cointet, 2012), and management (Blanc & Huault, 2019). Like the IRAMUTEQ software, PROSPERO can track the trajectory of words and their network in a large corpus (in this research, the words concerned were nouns and adjectives related to the term ‘asbestos’). Our lexicometric analysis covered all the documents in our corpus that had a public dimension (press releases, press articles, pamphlets, public reports, and decrees). Given the diversity of the sources used, the aim was to bring out the common issues and shared debates. We were able to identify important changes depending on the period, for example, generalization of use of the term ‘cancer’ or use of the term ‘asbestos fiber’ instead of ‘asbestos’, or the rise of a more medical or scientific register to discuss management of the asbestos risk. This information highlighted beliefs shared by the majority of actors in the field. Where key shifts in these beliefs were obser ved, we conducted a deeper – manual – analysis of the possible explanations for the changes, with particular consideration to any concurrent institutional actions. Sometimes, the links were obvious, for example, when a television report publicly exposed the risks of asbestos, or when the Jussieu anti-asbestos group started a movement that attracted a wide media coverage. When the causality was less obvious, we sought to identify the actions most probably associated with these shifts in beliefs, first by triangulating the researchers’ analyses and second by questioning experts in the field. Nonetheless, in certain cases, the causal links remained unclear: this is an inherent limitation to analysis of change in an institutional field (Hoffman, 1999).

Table 1

Description of the method

Phases of the researchObjectivesData usedData analysis
Determining changes in the pace of deinstitutionalizationTo identify changes in the core beliefs about asbestos
To represent the pace of deinstitutionalization
Secondary data (press, specialist journals, white papers, etc.) Interviews with expertsTemporal segmentation of the study period
Analysis of changes in beliefs using the
Prospero software
Identification of the intensity of the changes
Determining the types of institutional action during the period studiedTo categorize the forms of institutional action
To understand the role of strategic and pragmatic agency
To take the forms of resource mobilization into account
Secondary data (press, specialist journals, white papers, etc.) Interviews with expertsCoding the actions (disruptive vs.
defensive in intention)
Triangulation between researchers
Verification of the validity of the analysis with experts in the field
Summar yConstruction of a timeline representing the pace of deinstitutionalization and the institutional action by the principal actorsConstruction of propositions concerning the pace of deinstitutionalization
Description of the method

Description of the method

Table 2

Data collected from interviews

Data collected
Secondary data
Archival dataPersonal archives of asbestos victims
Formal complaints and archives (lawyers)
Archives of the Confédération Générale du Travail (CGT) union
Archives of the anti-asbestos group Collectif anti-amiante Jussieu
Archives of the standing committee for asbestos CPA (Comité Permanent Amiante) (minutes of all meetings)
Archives of the asbestos victim defense association ANDEVA
Public reports and parliamentary inquiries
Media dataPress: collection through the Factiva database for the period 1970–1997
INA (French national television archives): around a 100 TV programs from the period 1970–1997 were analyzed (from a selection of 500 occurrences)
Books (7)Emmanuel Henry, Amiante, le scandale improbable, 2007
Odette Hardy-Hémery, Eternit et l’amiante 1922–2000, 2005
François Malye, Amiante, le dossier de l’air contaminé, 1996
François Malye, Amiante: 100,000 morts à venir, 2004
Francis Chateauraynaud & Didier Torny, Les Sombres Précurseurs, une sociologie pragmatique de l’alerte et du risque, 1999
Maria Roselli, Amiante et Eternit – Fortunes et forfaitures – 2008
Annie Thébaud-Mony, Travailler peut nuire gravement à votre santé, 2008
Asbestos expert at the Court of Appeal (1)2½ h
Lawyers (2)1½–2 h
Journalists (3)1½ h
Victims (14)Approximately 1 h per individual interview + group interviews + observation of group meetings
Occupational health doctor (1)1h
Doctor who was a member of the CPA (1)1h
Members of the anti-asbestos group Collectif anti-amiante Jussieu (2)1–1½ h
Foreman for asbestos removal at Jussieu (1)1 h + site visit
Manager of an asbestos removal company (1)45 min
Magistrate (in charge of asbestos cases) (1)30 min
Data collected from interviews

Data collected from interviews

19 Based on these analyses, we divided our total study period into seven periods corresponding to significant actions or a series of actions. Through a qualitative analysis of our empirical data, we listed the core beliefs about asbestos (see Table 3) and changes in those beliefs (Table 6).

20 We compared each period with the preceding period. A belief that undergoes no notable change (or is even reinforced) from one period to the next is coded 3. In our corpus, this is obser ved when asbestos shows a continuing consensual, stable association with a certain belief. A belief that is beginning to be questioned is coded 2 (e.g., when the texts in the corpus indicate doubts about a belief, for instance, following new scientific studies, new problematic cases, etc.). The new information is not consensual and is still only mentioned by a small number of actors. A belief subject to strong challenge is coded 1. When this happens, only a few texts continue to support a belief, but they are very much in the minority, and the majority of actors in the field no longer share it. Code 0 is assigned when a belief has become completely untenable, and the texts no longer mention it, or explicitly reject it. The weighting factor (see Table 6) quantitatively translates our interpretation of the importance of a belief, that is, its influence in the naturalization of asbestos-related practices. These factors were determined by reference to the ‘golden age’ of asbestos (before period 1) based on all our secondary data and the discussions with experts. During that period, asbestos was acclaimed for its effectiveness and controllability (we gave both of these dimensions equal weighting). The idea of its effectiveness was supported by three core beliefs we consider comparable in importance: its price, its insulation capacities, and its versatility. Regarding its controllability, the most decisive belief (weighting factor 4) was that the occupational risk to workers can be controlled. This is a central element, naturalizing the use of asbestos. To a lesser degree (weighting factor 1), controllability was also associated with a belief in the absence of any specific risks (the idea that all workers have the same risk exposure) and the fact that asbestos is a naturally occurring material (and, therefore, as harmless as water). Once again, the weighting was based on interpretation of our data. The weighting factors were kept constant throughout the study period, because the golden age of asbestos was the time when this belief system became established, and it went on to structure later debates about changes in institutionalized asbestos-related practices.

Analysis of institutional actions

21 We sought to identify the most important institutional actions during our study period, by referring to our secondary data (press articles, TV reports, and the extant literature on the history of asbestos) and our interviews. We drew up a list of the most significant actions in the institutional field of asbestos, identifying any institutional action that was mentioned by at least two interviewees, or frequently encountered in the secondary data. This list comprised 58 actions representing the institutional efforts undertaken in the field, referenced in Table 7. Each action was categorized into two dimensions: the form of agency involved and the type of resource mobilization. For the first categorization, we only retained strategic agency and pragmatic agency, as routine-based agency (corresponding to Emirbayer and Mische’s ‘iterational dimension’) is not directly related to institutional work (Lawrence et al., 2009). Table 4 shows the criteria used to categorize these actions.

Table 3

Core beliefs concerning the institution in 1970

Core beliefsUnderlying assumptions
Best insulator – best price – versatile
The occupational risk is controllable and non-specific.
Asbestos is a natural, controllable material.
Core beliefs concerning the institution in 1970

Core beliefs concerning the institution in 1970

22 Of the six possible combinations of these two dimensions, we retained four that are conceivable in theoretical terms (pragmatic agency cannot be combined with accumulation [Dorado, 2005]). We then classified the actions involving strategic agency according to their focus on maintaining the existing beliefs about asbestos (defensive work) or on destruction of those beliefs (disruptive work). This gave us seven possible combinations in theoretical terms. Coding from field data reduced this to six combinations (Table 5), which were used to classify the 58 key actions during the study period:

23 The following three combinations of actions focused on institutional maintenance: Leveraging-Strategic agency (LSt), Convening-Strategic agency (CSt), and Accumulating-Strategic agency (ASt);

24 The following two combinations of actions focused on destruction of institutional beliefs: LSt and ASt. It is interesting to note that the CSt combination, although theoretically conceivable, is not found among actions that advanced the deinstitutionalization of asbestos, the result of which is discussed later.

25 Pragmatic agency concerns problem-solving work that changes the institution incrementally without the actors taking a clear position for its defense or destruction. This is coded (ASe).

26 Finally, we constructed a timeline presenting the key points of our analysis (see Figure 1). Although it is complex to interpret because of the many interactions in the field of asbestos, this timeline provides an analytical summary of relations between the referenced actions during the study period and the deinstitutionalization dynamic consisting of long and short phases. This schematization step is useful for sensemaking by researchers (Langley, 1999). Of course, the longitudinal juxtaposition of two dimensions (pace and actions) is not proof of any causality between the two. In each period, we referred to our data to see whether any causality effects were discernible between the institutional actions and accelerations or decelerations in deinstitutionalization.

Table 4

Coding criteria for the actions

Resource mobilizationType of agency
CriteriaLeveraging (L):
An actor tries to achieve his objective by rallying other actors to his cause
Organization of meetings to promote an idea or an aim.
Convening (C):
Efforts to create a collaborative body to debate existing problems
In theory, no dominant aim or idea.
Accumulating (A):
Efforts are focused on personal aims – or the goals of an organization – without joining forces with other actors or organizations.
Strategic agency (St):
Long-term goals
The intent to take action to influence certain aspects of the institution
Reflection on the institutional impact of actions.
Pragmatic agency or sensemaking (Se):
Short- to medium-term goals
Efforts made in response to a complex situation, which may require creativity
The institution itself is not called into question and is not a target for these efforts.
Coding criteria for the actions

Coding criteria for the actions

Table 5

Action profiles

Defense vs. disruptionType of resource mobilizationType of agencyName of action profile based on the combination
Defensive actionLeverage (L)
Convening (C)
Accumulation (A)
StrategicDefensive efforts through leverage (LSt – defensive)
Defensive efforts through convening
(CSt – defensive)
Defensive efforts through accumulation
(ASt – defensive)
Disruptive actionLeverage (L)
Accumulation (A)
StrategicDisruptive efforts through leverage
(LSt– disruptive)
Disruptive efforts through accumulation (ASt – disruptive)
No clearly established intentionAccumulation (A)PragmaticPragmatic problem-solving efforts
Agency Sensemaking/Pragmatic (ASe)
Action profiles

Action profiles

LSt, Leveraging-Strategic agency; CSt, Convening-Strategic agency; ASt, Accumulating-Strategic agency; ASe, Accumulating - Sensemaking.
Table 6

Analytical summary analysis of changes in the core beliefs about the institution

wfPeriod 1Period 2Period 3Period 4Period 5Period 6Period 7
EffectivenessBest insulator23322332
Best price23333331
ControllabilityThe occupational risk is controllable42121230
There are no specific risks11100100
Asbestos is a naturally occurring material13110220
Total institutional changes2.522.
Analytical summary analysis of changes in the core beliefs about the institution

Analytical summary analysis of changes in the core beliefs about the institution

WF: weighting factor (1–4); 3: stability or reinforcement; 2: doubts and questions; 1: disruption; 0: rejection

Results. Understanding the temporal dynamics of deinstitutionalization – The case of asbestos in France

27 Table 6 provides an analytical summary of the changes described in the narrative. The numbers refer to the actions listed in Table 7 and Figure 1.

Period 0 (1945–1970): The golden age of asbestos in France

28 After World War II, western countries were well aware of the economic and strategic benefits of asbestos, the ‘wonder mineral’ (Malye, 1996): ‘From that point on, anything to do with heat or fire was met with the answer ‘asbestos.’ […] The die was cast: this mineral was to be ‘the travelling companion of industrial capitalism’ as the Jussieu anti-asbestos group put it so well in 1977’ (Malye, 1996, p. 29, our own translation). In 1970, there were around 3,000 products containing asbestos in France (from cigarettes to car brakes via a host of other products, such as toasters) (Roselli, 2008). Regarding labor relations, as early as 1945, a specific disease related to asbestos – asbestosis – was formally recognized in France as an occupational illness. In the 1950s, the link between asbestos and cancer was highlighted by several scientific studies, but there was no challenge to its supremacy. The occupational risks of exposure to asbestos were treated as one component of a broader context, as indicated by their inclusion in the same group as all risks related to industrial particles.

Period 1 (1970–1974): The first doubts emerge – A slight acceleration in the pace of deinstitutionalization (Actions 1–9)

29 During this first period, evidence mounted that asbestos is a health hazard. This led to questions about the way asbestos should be used or changes in its uses – but no calls for a ban. The defenders of asbestos have already prepared a strategic campaign to protect their interests and recommended that the question should be addressed through local case-by-case negotiations (Actions 1–9 – Table 7).

30 In London, in 1971, the major industrial players in asbestos held a large conference, where they organized their proactive defense strategy: (1) being well aware of the future legal risks, they structured their defense and developed connections with public authorities.

31 In France, the first pro-asbestos lobby, COFEBRA, [1] was set up (2). There were protests by workers against poor working conditions involving asbestos risks, for instance, at the automotive parts manufacturer Férodo, part of an industr y that used asbestos. In 1973, in line with the recommendations of the London conference, the company responded to a strike by setting up a health and safety committee (3). In fact, the workers at Férodo campaigned for better working conditions rather than actually challenging the use of asbestos. Meanwhile, the International Agency for Research on Cancer (IARC) and the Institut national de recherche et de sécurité (INRS) [2] (French National Research and Safety Institute for the Prevention of Occupational Accidents and Diseases) organized conferences, and published guides and recommendations for protecting people exposed to asbestos (4) (5). The legitimacy of using asbestos was not questioned. The aim was to learn how to use it in a way that reduced the risks of its exposure.

Table 7

List of institutional actions

ActionsDateTypes of action
Phase 1
1First International Conference of Asbestos Information Bodies – London (UK)1971LSt
2Formation of the COFREBA1971ASt
3Start of the strike at Ferodo1973ASe
4WHO international conference – Lyon (France)1972ASe
5Publication by the INRS (guide to safe use of asbestos)1972ASt
6Lawsuit against Johns Manville (USA)1973ASe
7ILO international conference – Geneva (Switzerland) ‘Asbestos: Health risks and their prevention’1973ASe
8A special committee for safety is set up at Eternit (for France)1973ASt
9IARC declaration recognizing several forms of asbestos as human carcinogens (subject to some reservations for the most common forms)1973ASe
Phase 2
10Start of the strike at Amisol1974ASe
11Formation of an action group at Paris-Jussieu university1975ASe
12Meeting between Amisol and Jussieu1976LSt
13Mobilization of support for Amisol employees1976LSt
14Start of new negotiations at Amisol1976ASe
Phase 3
15INRS proposal to reduce acceptable exposure levels1976ASe
16Law to protect workers under 181976ASe
17Decree limiting risks of exposure to asbestos1976ASe
18Change in formal recognition of occupational illness caused by asbestosMay 1976ASe
Phase 4
19Media coverage of the Amisol scandalAugust 1976ASe
20Exposé by a consumer association of the presence of asbestos in everyday consumer goodsOctober 1976ASe
21The asbestos producers’ union publishes several advertisements and brochuresNovember 1, 1976ASt
22Conference organized by the asbestos producers’ union (promoting and defending asbestos)November 3, 1976LSt
233 primetime TV programs denouncing the risks of asbestosNovember 29, 1976ASe
24IARC conference on the risks of asbestosDecember 14–17, 1976ASe
25The asbestos producers’ union writes to the French Prime Minister (accusing the Jussieu group of slander)December 20, 1976LSt
26Publication of a book by the asbestos producers’ unionJanuary 1977LSt
27Forceful response by the members of the Jussieu groupApril 5, 1977ASe
28Publication of a leaflet entitled ‘Danger, Asbestos’June 14, 1977LSt
29Recognition by the IARC of the carcinogenic nature of all forms of asbestos1977ASe
Phase 5
30Decree banning asbestos flock in housingJune 29, 1977ASt
31Decree reducing authorized exposure levelsAugust 17,1977ASt
32Decree introducing safety measures for transporting asbestosAugust 29, 1977ASt
33Decree banning asbestos flock (in all buildings)March 20, 1978ASt
34Conference in defense of asbestos – Paris (France)1979 (several dates)LSt
35The asbestos producers’ union becomes an association1980ASt
Phase 6
36Formation of the CPA1982CSt
37Removal of asbestos from school buildings becomes mandator y1982ASe
38World Symposium on Asbestos – Montreal (promoting controlled use of asbestos)1982LSt
39Decree further reducing authorized exposure levels1987ASe
40Ban on all forms of asbestos except chrysotile1988ASt
41Decree further reducing authorized exposure levels1992ASe
Phase 7
42Complaint filed by the ‘widows of Gérardmer’1994ASt
43Start of an investigation by journalists at Sciences et Avenir1994LSt
44Start of an investigation by journalists at Le Monde Diplomatique1994LSt
45Lecture by Julian Peto – Jussieu – Paris1994LSt
46Formal recognition of the dangers of asbestos by a committee of independent scientistsNovember 1994ASe
47Formation of the CAPER1995LSt
48Publication of a French scientific studyJune 1, 1995ASt
49The CPA is disbandedSeptember 1995ASe
50The Envoyé Spécial report ‘Asbestos – the contaminated air scandal’ is broadcastSeptember 28, 1995ASe
51Formation of the ANDEVA1996LSt
52Decrees reducing authorized exposure levelsFebruary 7, 1996ASt
53Decree on protection of workersFebruary 7, 1996ASe
54Law on the requirement to have asbestos surveys of all buildingsFebruary 7, 1996ASe
55Complaints are filed, launching criminal and civil casesJune 25,1996ASt
56Publication of the official INSERM reportJuly 2, 1996ASe
57Announcement of a forthcoming ban on asbestosJuly 3, 1996ASt
58Start of the ban on all forms of asbestos (covering imports and use)January 1, 1997
Decree of December 26, 1996
List of institutional actions
List of institutional actions

List of institutional actions

LSt, Leveraging-Strategic agency; CSt, Convening-Strategic agency; ASt, Accumulating-Strategic agency; ASe, Accumulating - Sensemaking.

32 In the United States, these issues took on a national dimension: a class action was launched against Johns Manville, one ofAmerica’s biggest asbestos producers (6), which subsequently filed for protection under the bankruptcy laws. In 1973 the International Labour Organization (ILO) and the IARC published studies confirming the risks of exposure to asbestos (7) (9). As part of this movement, the French asbestos cement company Eternit set up a safety and working conditions oversight committee (8).

Period 2 (1974–1976): The beginnings of resistance – An acceleration in the pace of deinstitutionalization (Actions 10–14)

33 For the first time, the question of the dangers of asbestos, beyond occupational risks alone, was clearly raised. Its superiority for some uses, such as insulation, was still indisputable. This period was marked by the arrival of peripheral actors in the field (Actions 10–14 – Table 7).

34 In 1974, a strike broke out at an asbestos processing factory, Amisol, in the center of France to save the business (10), but economic reasons forced the company to close down. At this time, the workers – mainly women – had no knowledge of the risks to which they had been exposed. At the same time, staff at Paris-Jussieu university discovered asbestos fibers in their laboratory and wondered whether there might be asbestos in their offices too. A group headed by Professor Henri Pézerat was formed (11). This group discovered international studies, and the proven link between exposure to asbestos and cancer. Their reading led them to Amisol (12), where the situation had worsened – about 10 of the striking wor kers had died. The J ussieu group launched an appeal for support to defend Amisol employees’ rights to urgent, free medical treatment or transfers where possible (13). With the backing of the Jussieu group and public opinion on their side, ‘the Amisols’ started new negotiations (14).

Figure 1

Timeline of institutional actions and the progress of deinstitutionalization of asbestos in France

Timeline of institutional actions and the progress of deinstitutionalization of asbestos in France

Timeline of institutional actions and the progress of deinstitutionalization of asbestos in France

Period 3 (January–May 1976): The first political reactions – A deceleration in the pace of deinstitutionalization (Actions 15–18)

35 After the early doubts and the start of the challenge to asbestos in France, the authorities now joined the field, and the debate shifted to control of asbestos-related risks (Actions 15–18 – Table 7).

36 This period was marked by an initial series of steps by the legislator in response to the events of the previous period. Paradoxically, deinstitutionalization slowed down. The vital need for controlling the use of asbestos became the central focus of the debate as public awareness of the problem grew. The authorities thus confined the issue to the question of control. Nonetheless, the controversy brought to light by the J ussieu group, and the embodiment by ‘the Amisols’ of the dangers of asbestos, gave asbestos new connotations in public opinion: it was now associated with fear and death.

37 From the year 1976, the French authorities began to consider the specificities of asbestos-related risks. The INRS suggested setting of maximum exposure levels (15). After that proposal, a law was adopted to limit the asbestos fiber content of air (17) and protect employees aged under 18 years (16). The legislator also formally recognized occupational illnesses attributable to exposure to asbestos fiber, a move that emphasized the specific risks of this industrial particle (18).

Period 4 (August 1976 – June 1977): The first asbestos crisis in France – A substantial acceleration in the pace of deinstitutionalization (Actions 19–29)

38 The arrival of the media in the field and the organization of disruptive work driven by the Jussieu group soon brought a serious challenge to the core beliefs about the use of asbestos: the fact that asbestos is a health hazard was acknowledged, and the struggle now turned to its controllability and effectiveness (Actions 19–29 – Table 7).

39 The media seized on the Amisol scandal, and the Jussieu group’s scientific legitimacy entitled it to challenge the very use of asbestos in France (19). A national consumer association denounced the presence of asbestos in wine and other everyday consumer goods (20). Asbestos producers fought the repeated media coverage by distributing best practice guidelines and preventive documentation (21). They held a conference, declaring that the risks associated with the asbestos fiber were controllable (22), and that they were taking appropriate action against the potential dangers. A popular primetime television program broadcasted three reports on asbestos and its risks (23), thereby openly discussing the responsibilities of politicians and industrials in the prevention and control of this health hazard, whose risks were now considered more than just an occupational health issue. The IARC organized a conference to explain the risks of exposure to asbestos (24). Through their association, asbestos producers grew more active and implemented the recommendations of 1971. They wrote to France’s Prime Minister complaining about scaremongering and questionable allegations by the Jussieu group, and how this was affecting the French economy (25). They also published a book summarizing best practices for asbestos use (26). The economic necessity of maintaining the institution of asbestos was central to their arguments. The debate was clearly structured as a conflict between industrials defending controlled use of a material that was vital to national economic development and critics (embodied by the Jussieu group) complaining about the total lack of transparency regarding asbestos and the undeniable risks of death. The Jussieu group exercised its legal right of reply by publishing a response to the accusations made by the asbestos producers’ association (27) and issued a leaflet entitled ‘Danger, Asbestos’ [3] (28). The IARC formally recognized that all forms of asbestos cause cancer (29).

40 Some central aspects of asbestos use were challenged during this period, leading to an acceleration in the pace of deinstitutionalization. However, the institution of asbestos was not destroyed completely, because its effectiveness remained to be a central pillar supporting its continued use.

Period 5 (June 1977–1982): Quietening down – A deceleration in the pace of deinstitutionalization (Actions 30–35)

41 This intermediate period marked a turning point: after the first asbestos crisis, measures were taken by the State, which became an actor for compromise. The actions of the State confirmed the dangers of asbestos; however, it reinforced the idea of its controllability and upheld the necessity of using asbestos. The defensive action during this period took place in the background but was extremely effective, ushering in a subsequent period of silence (Actions 30–35 – Table 7).

42 In the sparring between the industrials and the J ussieu anti-asbestos group, the French public authorities were treated as both allies and enemies. This put them in a mediator role that slowed down or arguably actually halted the deinstitutionalization process. Asbestos flocking for insulation was banned in 1977 for all housing (30), the exposure limits for employees were reduced (31), and special requirements were introduced for transporting asbestos (32). In 1978, asbestos flock was banned for all buildings in France (33). This series of decrees underlines the urgency of the action, but its scope was limited.

43 ‘The only achievement was the decree of June 29, 1977 banning asbestos-based coatings in residential buildings for which planning permission was given after the date of that decision’ (Journalist in voice-over). ‘The decree of June 29, 1977 instead of the proposed law which was much better for people’ (Extract from the report ‘Amiante dioxine’ broadcasted on September 26, 1977 on French TV channel TF1 in the news magazine program ‘A la Bonne Heure’).

44 ‘ The decree of 1977 actually lowered vigilance… a certain number of people protested, saying this isn’t enough, we’re heading for another disaster, but they were voices in the wilderness, and with that the whole asbestos battle died down, until it resurfaced in the 1990s.’ (Interview with Maître Teissonière, lawyer)

45 Meanwhile, the Jussieu group was content with the fact that the matter had been taken up by the authorities, and the asbestos producers’ union organized a cycle of conferences promoting the use of asbestos (34).

46 ‘The spokesmen, campaigners and defenders of the cause had better things to do than keeping up the pressure at great personal cost. What we have here is the classic schema, in which mobilization comes to an end once it has succeeded in getting a collective cause put on the political agenda.’ (Chateauraynaud & Torny, 1999, quoted in p. 75 of the Report to the Senate, 2005, our own translation)

47 In 1980, the asbestos producers’ union became more neutral actor in the field (35), and thus, the disruptive work carried out by the Jussieu group stopped. As the TV debates of the period show, the asbestos producers’ association successfully established itself as the go-to expert on the topic of asbestos. Mobilization declined:

48 ‘why did the crisis see a drop into intensity from 1980 to 1993? That’s explained by the measures taken to limit asbestos in the industrial environment. I’m thinking particularly of the decree of 1977. […] There was no point carrying on the work of the ‘ J ussieu group’ we’d set up. We felt like nobody was listening to us anymore.’ (Extract from the interview with Professor Pézerat for the Report to the Senate, 2005, p. 88, our own translation)

Period 6 (1982–1994): Organized silence – A pause in the deinstitutionalization process (Actions 36–41)

49 In this period, the previous disruptive work was obliterated by the creation of a new organization – the Standing Committee for Asbestos, CPA [4] – under the authority of France’s Ministry for Health. This committee’s missions reinforced asbestos-related practices by promoting the idea of its controllability (Actions 36–41 – Table 7).

50 Building on the work of the asbestos producers’ association, the CPA was set up in 1982 (36), which took full charge of asbestos-related matters in France. A decree was issued requiring the removal of asbestos from school buildings (37). From 1982 to 1988, the CPA was the only official authority on asbestos in France. It defined a new doctrine, that is, controlled use of asbestos, and promoted it in France and worldwide (38). The members of the CPA included asbestos industrialists, representatives of the industries that used asbestos, union representatives, and occupational doctors, and its mission was to ensure that the use of asbestos would continue. Officially, France no longer had an asbestos issue, and the deinstitutionalization process was halted, in contrast to the situation in other western countries, such as the USA, the UK, and Germany. Under pressure from the European institutions, the French government adopted the following new control and restriction measures in the late 1980s: lowering the maximum exposure levels, banning the rarest and most expensive forms of asbestos, and introducing new measures regarding the air dust content (39) (40) (41).

51 This period is specific to France. The almost 10-year halt in the deinstitutionalization process resulted from great strategic intelligence on the part of the asbestos industrials, who successfully mobilized resources and political support. It was a period of silence, a time when officially France no longer had asbestos concerns, as a journalist who investigated the asbestos issue told us: ‘In 1994, two journalists from [consumer affairs magazine] ‘50 millions de consommateurs’ are given the assignment: we’re going to do an investigation into asbestos. They came back 3 weeks later when I was there: they didn’t have anything. The first thing they’d done was go and see the experts at the INC [5] (Institut National de la Consommation) [who worked in the same building], to ask questions: ‘no, there’s no risk’. That guy sat on the CPA (…) and they had the INC all sewn up.’ (Interview with François Malye, our own translation). There was nothing in the press or other media about asbestos-related concerns or dangers: ‘It’s surprising but honestly at the time, everyone felt that the asbestos question had been settled. There was nothing more to be said about it. Measures had been taken, and the problem had been dealt with.’ (Interview with Patrick Herman, journalist). The controllability belief had become an indisputable core belief.

Period 7 (1994–1997): From scandal to ban – A sharp acceleration in the deinstitutionalization process (Actions 42–58)

52 To end the silence, the old actors from the first crisis now joined forces with new actors and undertook disruptive action that totally undermined the beliefs of controllability and effectiveness. As one of the first journalists to write about the asbestos question told us: ‘We weren’t going to do another ‘77 and say, everything’s settled’! (Interview with Patrick Herman, journalist and member of an association named Ban Asbestos) (Actions 42–58 – Table 7).

53 In 1992, a group of widows of technology teachers who all had worked at the same French high school filed a complaint leading to court action against the national education system (42). These women and their lone campaign caught the attention of a journalist from the popular French science magazine Sciences et Avenir (43). In parallel, a journalist from French current affairs broadsheet Le Monde Diplomatique took interest in an asbestos warehouse in the south of France (44). The work of the British epidemiologist Julian Peto also had a notable impact in France. Peto published a study in The Lancet in March 1995, which predicted that the UK would report 10,000 deaths a year caused by asbestos. He was invited by Professor Henri Pézerat to present his work at Paris-Jussieu university (45). The results of Peto were particularly worrying and were validated in published work by French researchers who were not members of the CPA (46). This marked the renaissance of the movement that begun in the 1970s. Henri Pézerat rallied a group of experts (lawyers, journalists, and academics) who relaunched and steered the debate. They formed a committee named as the Committee for Action, Prevention, and Reparation (CAPER) [6] (47). The investigation by Sciences et Avenir was published, with the particularly compelling cover headline: ‘Asbestos – the contaminated air scandal’ [7] (48). In addition to the dangers of asbestos, the debate was now centered on political and industrial manipulation: ‘they’ knew all along, and‘they’ had put us in danger to protect the interests of the industrial players. The CPA was disbanded in September 1995, 3 months after this investigation came out (49). This event is only mentioned in passing in the meeting minutes. By this time, only France’s National Academy of Medicine [8] was still expressing doubts about the real future risks of asbestos, especially the 100,000 deaths projected by Julian Peto’s results. In late September, Envoyé Spécial, a respected primetime current affairs program on a national TV channel, broadcasted a critical investigation into asbestos (50). The asbestos victim defense association ANDEVA [9] was formed in February 1996 (51): ‘So the victims themselves weren’t aware of what was going on, and in the end an association was necessary – the ANDEVA, founded in 1995–1996.’ (Interview with Maître Teissonière, lawyer)

54 Another series of official measures were introduced by a decree in February 1996 (52) (53) (54). Formal complaints starting further legal action were filed in J une 1996 (55). In J uly, the French public human health research organization, INSERM, [10] published a study denouncing the risks of exposure to asbestos (56). The following day, July 3, 1996, a forthcoming total ban on asbestos in France was announced (57). In just a few months, after a long period of silence, asbestos had been totally delegitimized and the debate on the economic consequences of a ban had been swept aside. Asbestos was now entirely associated with the ideas of political manipulation, economic cynicism and sickness or even death. On January 1, 1997, in application of the decree of December 26, 1996, asbestos was completely banned in France (58).


55 Studying the deinstitutionalization process of asbestos in France through the lens of agency and resource mobilization contributes especially to knowledge about the temporal dynamics of deinstitutionalization. The results of this case study reveal four propositions, which are discussed below.

Defensive action essentially involves leveraging efforts that promote long phases and helps to slow down the pace of deinstitutionalization

56 We obser ve that the most visible proactive defensive action combines the strategic agency with a leveraging tactic. The success of this type of action depends on its ability to propose a common goal, which reduces uncertainty and inspires other actors. It provides an impetus that facilitates enrolment of additional support. We observe that professional conferences played an important role in restoring trust in challenged practices (conferences 1 and 22), but this kind of conference is different from the field-configuring events defined in the literature of recent years (e.g., Hardy & Maguire, 2010). It is, in fact, the appearance of new opportunities for interaction that enables certain events to (re) configure the field (Oliver & Montgomery, 2008). In this case, the most effective defensive action was driven by actors who had both significant economic resources and good knowledge of the economic, social, and technical issues concerned. As Stinchcombe (1987, p. 114) writes, such actors ‘get paid for thinking about how to achieve and preser ve the values and interests embodied in an institution’. This and their status, as recognized by experts, enabled them to establish the idea of controlled use of asbestos, an idea devised and spread at pro-asbestos conferences in Paris in 1979 (action 34). This case study thus calls for further examination of an aspect that has been so far underexplored in the literature on field-configuring events: the existing research study currently underlines their role in change, but pays less attention to their role in institutional maintenance, particularly as disruptive actors do not attend certain professional meetings they consider as only there to uphold the status quo. Professor Pézerat from the Jussieu anti-asbestos group is a good example of this, as is the Ban Asbestos association formed after the asbestos scandal in France, who always refused to attend such conferences.

Disruptive action produces slow, incremental effects through marginal integration of changes into existing institutional schemas. The acceleration phase of deinstitutionalization is temporally bounded by the disruptive actors’ resources

Disruptive action involves strategic agency that interacts with pragmatic agency and results in gradual deinstitutionalization through marginal integration of changes into existing institutional schemas

57 Institutional change requires interaction between the strategic agency deployed in disruptive action and the pragmatic agency. Visibly, the disruptive ideas defended by actors with strategic vision are able to achieve institutional impact when other groups, without necessarily sharing those ideas, acknowledge them as a question needing an answer. In practice, the actors in the field have to incorporate a new issue into their practices and thinking (pragmatic agency). When faced with new problems in an environment that is growing more complex, due to an increasing number of scientific or medical results and studies, for example, they seek to make sense of the situation through shared beliefs, and also by absorbing the new ideas put forward by disruptive actors. That leads, for example, to continuing to use asbestos but introducing restrictions, such as the minimum worker age and authorized exposure levels. Without actually challenging the institution as a whole, the actors’ everyday activities – and their discourses – come to integrate new ideas, which, through accumulation, can generate change at an institutional level. Disruptive action, even if it fails to achieve immediate change, thus has after-effects, which can trigger change through the intermediary of other actors, including actors with no direct interest in the institutional change. Change is not solely driven by strategic agency but driven by interaction between strategic and pragmatic agencies.

58 Paradoxically, institutional maintenance logics can be one of the outcomes of interaction between the strategic agency and pragmatic agency. This case study shows the role played, in the move from maintenance to destruction, by actors who are often classified for convenience as ‘neutral’. Researchers and public authorities who clearly did not take sides when the debate first arose ultimately acted in ways that worked in favor of defense strategies, for example, setting ‘danger thresholds’, which responded to the disruptive actors but simultaneously reinforced the key discourse of asbestos risk controllability advanced by the defensive actors. Through their strategic agency, defensive actors are able to instrumentalize this apparent neutrality, and thus, push the disruptive actors to the sidelines. This appears to be particularly effective when those disruptive actors belong to what Wright and Zammuto (2013) call the ‘peripheral elites’, with significant cognitive and social resources.

The acceleration phases driven by peripheral disruptive actors are temporally bounded by their resources, despite their strategic agency. The defensive actors, who have more resources, can bring everyone to a compromise that is in their own favor

59 We observe phases of acceleration, and then deceleration, in deinstitutionalization. Radically, new ideas have more chances of being developed and promoted by non-central actors positioned at the boundaries of an institutional field (Ahmadjian & Robinson, 2011; Leblebici et al., 1991). These peripheral actors then have to persuade other parties occupying a more central position in the institutional field (Clemente & Roulet, 2015). This explains why a disruptive strategic agency is only effective when the work is taken up by other actors in the field. We observe that the work done by disruptive actors tends to be absorbed into the compromises constructed with the other actors in the field. After that, it becomes more difficult for disruptive actors to keep up a high level of reflexivity and involvement. For example, this explains the decline in mobilization in the late 1970s. Making concessions in the struggle enables the defenders of the status quo, who have more resources, to rally the central actors to their cause and sideline the disruptive actors even more. Institutional maintenance, here, takes the form of a rearrangement, which both naturalizes the dropping of some beliefs, and thus, contributes to the deinstitutionalization process, and reinforces other shared beliefs, delaying the total discontinuation of an institutional practice. The work of institutional maintenance could, in fact, be more accurately described as work to stall deinstitutionalization.

The acceleration and deceleration phases of deinstitutionalization hinge on the perception of urgency, which is a factor of instrumentalization for strategic actors

60 While the instrumentalization of urgency supports disruptive action on certain issues, we show that it is not in itself sufficient to achieve rapid deinstitutionalization, because the defensive actors generally claim successfully that the actors in the field can respond to the urgency while maintaining the existing institutional rules. Complementing the propositions of Smets et al. (2012) regarding the link between urgency and institutional change, this underlines the active role played by defensive actors in reducing the scope of the urgency, thus preventing an institutional crisis. This case study illustrates how institutional collapse only begins when the core beliefs appear unable to address the urgency factor. After being neutralized and absorbed in the long phases, the urgency presented by certain disruptive actors resurfaces more powerfully when it becomes evident that the current solutions are ineffective. Indeed, in the case of asbestos in France, when further deaths were revealed in 1994, the idea took hold that the urgent need for a ban had been covered up: this led to a loss of legitimacy for the defensive actors, which precipitated the collapse of the institution in less than 2 years after years of institutional maintenance. The second wave of mobilization was organized around this idea of urgency, amplified by discourses, claiming that the central actors in the field lacked the competence to address it. We also note temporal work, as defined by Granqvist and Gustafsson (2015), intended to set the pace for the actors: in this case, the defensive actors essentially promoted long phases, undermining the arguments of urgency and putting the emphasis on gradual, ‘realistic’ change; while the disruptive actors had a greater tendency to promote the urgency of the situation, and thus, speed things up. These obser vations suggest avenues for research concerning the relative perception of time in an institutional field, depending on the actors’ positions. This is a common problem to many public health issues that involve a trade-off between protection of the public, with arguments emphasizing an idea of urgency, and protection of the economy, with arguments encouraging considered, gradual decision-making.

Convening is a form of mobilization that significantly slows down the pace of deinstitutionalization

61 Setting up a collective reflection group involving a broad diversity of actors contributed to a significant, long-term deceleration in the pace of deinstitutionalization. As Dorado (2005) suggests, this is particularly true in a high-uncertainty setting. Such bodies where defensive actors and disruptive actors meet directly, with no intermediary, are a sign that the different actors recognize the impossibility of immediate conciliation and want to start discussions. In this case study, the formation of the CPA (standing committee for asbestos) – a textbook example of a collective reflection group – coincided with a very long period of stability regarding the use of asbestos that was unique in Europe. This type of body operates as a receptacle for alternative ideas in an institutional field, which are debated and may lead to a (gradual) change. The disruptive ideas are thus confined to a reflection group and have little chance of being debated more publicly. Nonetheless, such debates and adjustments may satisfy the most subversive actors in the short term, because they give the impression that the criticism has been taken on board.

62 Disruptive action in this case study was weakened by the participation of disruptive actors in these compromise arrangements. Despite the initial aim of giving all views equal standing, the consensus established by such collaborative reflection bodies tends to be closer to the existing institutional framework, as the environment in which the alternative ideas are devised and absorbed is made up of existing institutional factors. The example of the CPA advances understanding of a very specific type of institutional maintenance work, consisting of strategic organization of the cohabitation between disruptive action and defensive action. The CPA was a meta-organization, which by confronting and incorporating different viewpoints played the role of a field-configuring body, extending beyond the idea of a field-configuring event (Hardy & Maguire, 2010). Ultimately, its mission was instrumentalized by the dominant actors in the field: in accepting the principal accusations about asbestos, those actors made it easier for controllability to become a core belief. This form of institutional work is a last-resort strategy against a compromised institutionalized practice. It is founded on the actors’ ability to systematically integrate criticisms into concrete proposals that actually conform to existing institutional rules, and thus, delay the institutional collapse.


63 This research study was guided by an empirical question: why was the deinstitutionalization of asbestos so slow in France? The story of asbestos use in France offers a perfect representation of a long, complex process involving a conflict between defensive and disruptive action in a succession of phases. We analyzed these phases of acceleration or deceleration in the pace of deinstitutionalization by studying action profiles.

64 We stress that deinstitutionalization cannot be considered simply as a confrontation between two opposing camps: the defenders of an institution and its critics. We show that actors with much more ambiguous and complex motivations regarding the institution can also play a decisive role in institutional dynamics, making more use of pragmatic agency than strategic agency.

65 We have stressed, in this study, on how the disruptive actors were able to mobilize a register of urgency to accelerate deinstitutionalization, while the defensive actors had a greater tendency to use the register of controllability, which implies longer timescales. This research study, following the footsteps of other authors, suggests that moments of interaction in an institutional field should be more taken into consideration. This would advance understanding of how pace and the perception of time are manipulated in an institutional work setting and how proposals of temporality are internalized by the actors in the field.


  • [1]
    Comité français d’étude sur les effets biologiques de l’amiante.
  • [2]
    Institut national de recherche et de sécurité pour la prévention des accidents du travail et des maladies professionnelles.
  • [3]
    Danger, Amiante.
  • [4]
    Comité Permanent Amiante.
  • [5]
    At the time, the magazine 50 millions de Consommateurs was owned by the INC (Institut National de la Consommation), which consisted of market survey experts.
  • [6]
    Comité Action Prévenir et Réparer.
  • [7]
    Amiante, le scandale de l’air contaminé. For French readers, this choice of wording necessarily called to mind the ‘contaminated blood’ scandal of 1991 (concerning transfusions using HIV- infected blood).
  • [8]
    Académie Nationale de Médecine, an academy of doctors which advises the government on public health matters.
  • [9]
    Association Nationale de Défense des Victimes de l’Amiante.
  • [10]
    Institut National de la Santé et de la Recherche Médicale.

Through the controversial history of asbestos use in France, we study a long deinstitutionalization process marked by alternating phases of acceleration and deceleration. To understand these changes of pace, we reveal interactions over the long term between action profiles that differ in terms of the type of agency (strategic or pragmatic) and the resource mobilization process (leveraging, accumulation, or convening) involved. Analyzing a rich corpus of documentary data triangulated with interviews, we draw up a schema of the complex deinstitutionalization process concerning asbestos in France. We then set out four theoretical propositions about the temporal dynamics of deinstitutionalization: (1) defensive action essentially involves leveraging efforts that promote long phases and help to slow down the pace of deinstitutionalization; (2) disruptive action produces slow, incremental effects through marginal integration of changes into existing institutional schemas. The acceleration phase of deinstitutionalization is temporally bounded by the disruptive actors’ resources; (3) the acceleration and deceleration phases of deinstitutionalization hinge on the perception of urgency, which is a factor of instrumentalization for strategic actors; and (4) convening is a form of mobilization that significantly slows down the pace of deinstitutionalization.


  • Deinstitutionalization
  • Agency
  • Resources
  • Institutional work
  • Asbestos
  1. Theoretical framework
    1. The institutional work of deinstitutionalization
    2. Action inside an institutional field
    3. The pace of deinstitutionalization: An underexplored dimension
  2. Research method
    1. Research setting
    2. Data analysis
      1. Analysis of the pace of deinstitutionalization
      2. Analysis of institutional actions
  3. Results. Understanding the temporal dynamics of deinstitutionalization – The case of asbestos in France
    1. Period 0 (1945–1970): The golden age of asbestos in France
    2. Period 1 (1970–1974): The first doubts emerge – A slight acceleration in the pace of deinstitutionalization (Actions 1–9)
    3. Period 2 (1974–1976): The beginnings of resistance – An acceleration in the pace of deinstitutionalization (Actions 10–14)
    4. Period 3 (January–May 1976): The first political reactions – A deceleration in the pace of deinstitutionalization (Actions 15–18)
    5. Period 4 (August 1976 – June 1977): The first asbestos crisis in France – A substantial acceleration in the pace of deinstitutionalization (Actions 19–29)
    6. Period 5 (June 1977–1982): Quietening down – A deceleration in the pace of deinstitutionalization (Actions 30–35)
    7. Period 6 (1982–1994): Organized silence – A pause in the deinstitutionalization process (Actions 36–41)
    8. Period 7 (1994–1997): From scandal to ban – A sharp acceleration in the deinstitutionalization process (Actions 42–58)
  4. Discussion
    1. Defensive action essentially involves leveraging efforts that promote long phases and helps to slow down the pace of deinstitutionalization
    2. Disruptive action produces slow, incremental effects through marginal integration of changes into existing institutional schemas. The acceleration phase of deinstitutionalization is temporally bounded by the disruptive actors’ resources
      1. Disruptive action involves strategic agency that interacts with pragmatic agency and results in gradual deinstitutionalization through marginal integration of changes into existing institutional schemas
      2. The acceleration phases driven by peripheral disruptive actors are temporally bounded by their resources, despite their strategic agency. The defensive actors, who have more resources, can bring everyone to a compromise that is in their own favor
    3. The acceleration and deceleration phases of deinstitutionalization hinge on the perception of urgency, which is a factor of instrumentalization for strategic actors
    4. Convening is a form of mobilization that significantly slows down the pace of deinstitutionalization
  5. Conclusion


  • Ahmadjian, C.L. & Robinson, P. (2011). Safety in numbers: Downsizing and the deinstitutionalization of permanent employment in Japan. Administrative Science Quarterly, 46, 622–654. doi: 10.2307/3094826
  • Battilana, J. & D’Aunno, T. (2009). Institutional work and the paradox of embedded agency. In T.B. Lawrence, R. Suddaby & B. Leca (Eds.), Institutional work: Actors and agency in institutional studies of organizations (pp. 31–58). Cambridge University Press.
  • OnlineBen Slimane, K. (2012). Retourner sa veste, toujours du bon côté: Travail institutionnel discursif dans le déploiement de la télévision numérique terrestre en France. M@n@gement, 15 (2), 146–179. doi: 10.3917/ mana.152.0146
  • OnlineBlanc, A. & Huault, I. (2019). The maintenance of macro-vocabularies in an industr y: The case of the France’s recorded music industr y. Industrial Marketing Management, 80, 280–295. doi: 10.1016/j. indmarman.2018.06.004
  • Boltanski, L. & Thévenot, L. (2006). On justification: Economies of worth. Princeton University Press.
  • Burns, L. R. & Wholey, D. R. (1993). Adoption and abandonment of matrix management programs: Effects of organizational characteristics and interorganizational networks. Academy of Management, 36, 106–138. doi: 10.2307/256514
  • Chateauraynaud, F. (2003). Prospéro: une technologie littéraire pour les sciences humaines. In CNRS communication. CNRS. EAN: 9782271061331
  • OnlineChateauraynaud, F. (2014). Trajectoires argumentatives et constellations discursives. Exploration socio-informatique des futurs vus depuis le nanomonde. Réseaux, 188 (6), 121–158. doi: 10.3917/res.188.0121
  • Chateauraynaud, F. & Torny, D. (1999). Les sombres précurseurs: une sociologie pragmatique de l’alerte et du risque. Ecole Des Hautes Etudes En Sciences Sociales.
  • OnlineChaudhry, S. & Rubery, J. (2017). Why do established practices deinstitutionalize? An actor-centred approach. British Journal of Management, 30 (3), 538–557. doi: 10.1111/1467-8551.12264
  • OnlineClemente, M., Durand, R. & Roulet, T. (2016). The recursive nature of institutional change: An annales school perspective. Journal of Management Inquiry, 26 (1), 17–31. doi: 10.1177/1056492616656408
  • OnlineClemente, M. & Roulet, T. J. (2015). Public opinion as a source of deinstitutionalization: A ‘spiral of silence’ approach. Academy of Management Review, 40 (1), 96–114. doi: 10.5465/amr.2013.0279
  • OnlineCointet, J. & Parasie, S. (2018). Ce que le big data fait à l’analyse sociologique des textes. Revue Française de Sociologie, 59 (3), 533–557. doi: 10.3917/ rfs.593.0533
  • OnlineDaudigeos, T., Boutinot, A. & Jaumier, S. (2015). L’étude historique du changement institutionnel sur longue période: Chausse trappes et perspectives. Un commentaire sur l’article de Hélène Peton et Stéphan Pezé. M@n@gement, 18 (3), 254–260. doi: 10.3917/mana.183.0254
  • Online Davis, G. F., Diekmann, K. A. & Tinsley, C. H. (1994). The decline and fall of the conglomerate firm in the 1980s: The deinstitutionalization of an organizational form. American Sociological Review, 59, 547–570. doi: 10.2307/2095931
  • OnlineDelacour, H. & Leca, B. (2011). The decline and fall of the Paris Salon: A study of the deinstitutionalization process of a field configuring event in the cultural activities. M@n@gement, 14 (1), 436–466. doi: 10.3917/ mana.141.0436
  • Djelic, M.-L. & Quack, S. (2003). Globalization and institutions. Edward Elgar Publishing.
  • OnlineDorado, S. (2005). Institutional entrepreneurship, partaking, and convening. Organization Studies, 26 (3), 385–414. doi: 10.1177/0170840605050873
  • OnlineEisenhardt, K. M. (1989). Building theories from case study research. Academy of Management Review, 14 (4), 532–550. doi: 10.2307/258557
  • OnlineEmirbayer, M. & Mische, A. (1998). What is agency? The American Journal of Sociology, 103 (4), 962–1023. doi: 10.1086/231294
  • OnlineGranqvist, N. & Gustafsson, R. (2015). Temporal institutional work. Academy of Management Journal, 59 (3), 1009–1035. doi: 10.5465/amj.2013.0416
  • OnlineHardy, C. & Maguire, S. (2010). Discourse, field-configuring events, and change in organizations and institutional fields: Narratives of DDT and the Stockholm convention. Academy of Management Journal, 53 (6), 1365–1392. doi: 10.5465/AMJ.2010.57318384
  • OnlineHargadon, A. B. & Douglas, Y. (2001). When innovations meet institutions: Edison and the design of the electric light. Administrative Science Quarterly, 46 (3), 476–501. doi: 10.2307/3094872
  • OnlineHeugens, P. P. & Lander, M. W. (2009). Structure! Agency! (and other quarrels): A meta-analysis of institutional theories of organization. Academy of Management Journal, 52 (1), 61–85. doi: 10.5465/AMJ.2009.36461835
  • OnlineHiatt, S. R., Sine, W. D. & Tolbert, P. S. (2009). From Pabst to Pepsi: The deinstitutionalization of social practices and the creation of entrepreneurial opportunities. Administrative Science Quarterly, 54 (4), 635–667. doi: 10.2189/asqu.2009.54.4.635
  • OnlineHirsch, P. M. & Lounsbury, M. (1997). Ending the family Quarrel. American BehavioralScientist,40(4),406–418.doi:10.1177/0002764297040004004
  • OnlineHoffman, A. J. (1999). Institutional evolution and change: Environmentalism and the U.S. chemical industry. Academy of Management Journal, 42 (4), 351–371. doi: 10.2307/257008
  • Jones, C. (2001). Co-evolution of entrepreneurial careers, institutional rules and competitive dynamics in American film, 1895-1920. Organization Studies, 22, 911–944. doi: 10.1177/0170840601226002
  • OnlineLangley, A. (1999). Strategies for theorizing from process data. Academy of Management Review, 24 (4), 691–710. doi: 10.2307/259349
  • Lawrence, T., Suddaby, R. & Leca, B. (2009). Introduction. In T. B. Lawrence, R. Suddaby & B. Leca (Eds.), Institutional work: Actors and agency in institutional studies of organizations (pp. 1–27). Cambridge University Press.
  • OnlineLawrence, T. B. & Suddaby, R. (2006). Institutions and institutional work. In S. Clegg (Ed.), The Sage handbook of organization studies (pp. 215–254). Sage.
  • OnlineLawrence, T. B., Winn, M. I. & Jennings, P. D. (2001). The temporal dynamics of institutionalization. Academy of Management Review, 26 (4), 624–644. doi: 10.2307/3560245
  • OnlineLeblebici, H., Salancik, G. R., Copay, A. & King, T. (1991). Institutional change and the transformation of interorganizational fields: An organizational history of the U.S. radio broadcasting industry. Administrative Science Quarterly, 36 (3), 333–363. doi: 10.2307/2393200
  • Lenglet, R. (1996). L’affaire de l’amiante. La Découverte.
  • OnlineLok, J. & De Rond, M. (2013). On the plasticity of institutions: Containing and restoring practice breakdowns at the Cambridge University boat club. Academy of Management Journal, 56 (1), 185–207. doi: 10.5465/ amj.2010.0688
  • OnlineMaguire, S. & Hardy, C. (2009). Discourse and deinstitutionalization: The decline of DDT. Academy of Management Journal, 52 (1), 148–178. doi: 10.5465/amj.2009.36461993
  • Mahoney, J. & Thelen, K. (2010). A theory of gradual institutional change. In J. Mahoney & K. Thelen (Eds.), Explaining institutional change: Ambiguity, agency, and power (pp. 1–37). Cambridge University Press.
  • Malye, F. (1996). Amiante: le dossier de l’air contaminé. Le Pré aux clercs.
  • Malye, F. (2004). Amiante 100 000 morts à venir. Le Cherche Midi.
  • McAdam, D. & Sewell, W. H. (2001). It’s about time: Temporality in the study of social movements and revolutions. In C. Tilley, D. McAdam, E.J. Perry, J. A. Goldstone, R. R. Aminzade, S. Tarrow, & W. H. Sewell (Eds.), Silence and voice in the study of contentious politics (pp. 89–125). Cambridge University Press.
  • OnlineOliver, A.L. & Montgomery, K. (2008). Using field-configuring events for sense-making: A cognitive network approach. Journal of Management Studies, 45 (6), 1147–1167. doi: 10.1111/j.1467-6486.2008.00786.x
  • OnlineOliver, C. (1992). The antecedents of deinstitutionalization. Organization Studies, 13 (4), 563–588. doi: 10.1177/017084069201300403
  • OnlineParasie, S. & Cointet, J.-P. (2012). La presse en ligne au service de la démocratie locale. Une analyse morphologique de forums politiques. Revue Française de Science Politique, 62 (1), 45–70. doi: 10.3917/ rfsp.621.0045
  • OnlinePeton, H. & Pezé, S. (2015). Reflection on historical studies of institutional change: Small steps are not necessarily missteps. A rejoinder to Daudigeos, Boutinot and Jaumier. M@n@gement, 18 (3), 261–265. doi: 10.3917/mana.183.0261
  • OnlineRao, H., Monin, P. & Durand, R. (2003). Institutional change in Toque Ville: Nouvelle cuisine as an identity movement in French gastronomy. American Journal of Sociology, 108 (4), 795–843. doi: 10.1086/367917
  • OnlineReay, T. & Hinings, C. R. (2005). The recomposition of an organizational field: Health care in Alberta. Organization Studies, 26 (3), 351–384. doi: 10.1177/0170840605050872
  • Roselli, M. (2008). Amiante & eternit fortunes et forfaitures. En bas.
  • Schneiberg, M. & Lounsbury, M. (2008). Social movements and institutional analysis. In R. Greenwood, C. Oliver, K. Sahlin-Andersson, & R. Suddaby (Eds.), The handbook of organizational institutionalism (pp. 648–670). Sage.
  • OnlineSmets, M., Morris, T. & Greenwood, R. (2012). From practice to field: A multilevel model of practice-driven institutional change. Academy of Management Journal, 55 (4), 877–904. doi: 10.5465/amj.2010.0013
  • Stinchcombe, A. (1987). Constructing social theories. University of Chicago Press.
  • OnlineToubiana, M. & Zietsma, C. (2017). The message is on the wall: Emotions, institutional expectations, and the dynamics of emotive institutional work. Academy of Management, 60 (3), 922–953. doi: 10.5465/ amj.2014.0208
  • OnlineWicks, D. (2001). Institutionalized mindsets of invulnerability: Differentiated institutional fields and the antecedents of organizational crisis. Organization Studies, 22, 659–692. doi: 10.1177/0170840601224005
  • Wooten, M. & Hoffman, A.J. (2008). Organizational fields: Past, present and future. In R. Greenwood, C. Oliver, R. Suddaby & K. Sahlin (Eds.), The Sage handbook of organizational institutionalism (pp. 130–147). Sage.
  • OnlineWright, A. L. & Zammuto, R. F. (2013). Wielding the willow: Processes of institutional change in English county cricket. Academy of Management Journal, 56 (1), 308–330. doi: 10.5465/amj.2010.0656
  • Yin, R. K. (2013). Case study research: Design and methods. Sage.
  • OnlineZietsma, C., Groenewegen, P., Logue, D. M. & (Bob) Hinings, C. R. (2016). Field or fields? Building the scaffolding for cumulation of research on institutional fields. Academy of Management Annals, 11 (1), 391–450. doi: 10.5465/annals.2014.0052
Hélène Peton
University Paris Est Créteil, IRG, F-94010 Créteil, France
University Gustave Eiffel, IRG, F-77447 Marne-la-Vallée, France
Antoine Blanc
PSL-Université Paris Dauphine, CNRS, Paris, France
This is the latest publication of the author on cairn.
This is the latest publication of the author on cairn.
Uploaded on on 11/05/2022
Distribution électronique pour AIMS © AIMS. Tous droits réservés pour tous pays. Il est interdit, sauf accord préalable et écrit de l’éditeur, de reproduire (notamment par photocopie) partiellement ou totalement le présent article, de le stocker dans une banque de données ou de le communiquer au public sous quelque forme et de quelque manière que ce soit.
Loading... Please wait