«See discussions, stats, and author profiles for this publication at: CRISES: EVENTS OR PROCESSES? ...»
See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/254276369
CRISES: EVENTS OR PROCESSES?
Article · January 1998
2 authors, including:
EMLYON Business School
26 PUBLICATIONS 463 CITATIONS
All in-text references underlined in blue are linked to publications on ResearchGate, Available from: Bernard Forgues letting you access and read them immediately. Retrieved on: 14 October 2016 CRISES: EVENTS OR PROCESSES?* Bernard Forgues Professeur en Sciences de gestion Christophe Roux-Dufort EDHEC Graduate School of Management 98 - 71 CRISES: EVENTS OR PROCESSES?* Bernard Forgues Christophe Roux-Dufort IAE Tours EDHEC Graduate School of Management 98 - 71 Hazards and Sustainability Conference, Durham, UK, 26-27 May 1998 Introduction For several years crisis management has become a focus of attention for many scholars in management. Research in this field is numerous but remain very fragmented. This fragmentation is due to the fact that crisis management has been approached by many disciplines, from various viewpoints. Due to the complexity of the subject under examination, researchers conduct analyses at different levels, have to be aware of the inner and outer contexts of the organization, and may observe very different phenomenons gathered under the single concept of ‘crisis.’ Much of the difficulty of conducting research in this domain has to do with empirical observations. Gaining access during a crisis is almost impossible, and would anyway bring more problems than answers. For instance, a great many organizations make decisions and interact with each others during a single crisis situation. This brings practical questions to the field researcher concerning data collection, including very basic ones: where to go, nearby the, say, burning warehouse or in the crisis room; who to observe and interview? The unit of analysis chosen for the research does not always bring a clear answer to those dilemmas.
Collecting data after the facts has other well-known pitfalls, such as recollection, a posteriori rationalization, missing archives, etc.
However, beyond the problems of observing the dynamics of a crisis and the limits of treating it a posteriori, there is a second level of difficulty, namely the integration of research. While the first level is of importance to any one researcher engaging into field research, the second one is of concern to our community as a whole. Integrating research from various disciplines, with different focus and methodologies is no easy task. In fact, the richness of varied approaches brings its own drawbacks, such as having still instable and embryonic theories.
Considering this state of the art, a valuable avenue for crisis management relies in working at building integrative frameworks and theories to gather the parts of this puzzle. It should be * An earlier draft was presented at a research seminar in HEC Paris. We are grateful to participants for helpful and constructive comments.
noted that we don’t aim at imposing a single paradigm for crisis management research. On the contrary, we believe with Van Maanen (1995: 133) that such a goal would be “philosophically indefensible; extraordinary naive as to how science actually works;
theoretically foolish, vain and autocratic; and (…) reflective of a most out-of-date and discredited father-knows-best version of knowledge, rhetoric and the role theory plays in the life of any intellectual community.” Rather, our point here has more to do with a better understanding of the (often) untold assumptions underlying crisis management research.
We argue that the major obstacle for researchers stems from a conceptual ambiguity on the notion of crisis itself. In looking for ‘what’s behind the research’ (Slife and Williams, 1995), we were struck by an apparent contradiction between two views of the crisis, namely crisis as
an event vs. crisis as a process. This issue leads to some contradictions into the literature:
while most of us appear to agree on the fact that crises are processes, we nevertheless often treat them as events. In fact, it's just as if we were all constrained by a sort of “research correctness” imposing us to state (literally) that the crisis is a process. However, this tribute being paid, we keep on working on cases which often are considered as events. This can be viewed for instance in works where the definition of the crisis focuses on processes and evolutions, this being followed by a typology of crises…
Event and process views in crisis management literature
Claiming that the field is fragmented is not new. Numerous authors made this statement before us, and called for integration. Some proposed models or frameworks aiming at unifying our research perspectives. For instance, Milburn, Schuler and Watman (1983: 1143) proposed “a definition and conceptualisation of organization crisis which will: (a) be as inclusive of the crisis relevant organiztional and individual phenomena as possible; (b) be as suggestive as possible for effective crisis management strategies and; (c) be as fruitful as possible for generating hypotheses about organizational crises.” Their proposed conceptualization includes three major aspects, arranged temporally: antecedents, both in the external and internal environment; moderators, both of the antecedent-crisis and of the crisisresponse relationships; and responses, both at the individual and organizational level. The model developed by Shrivastava, Mitroff, Miller and Miglani (1988) is focused on industrial crises, but broader and more precise. Crises are defined as having trans-organizational causes, involving social, political, and cultural variables. They are composed of many loosely coupled interdependent events, each of them setting the stage for the next one to occur in a chain reaction. The authors state that the crisis is triggered by a specific event identifiable according to time and place. Preconditions for this triggering event are created by organizational and environmental conditions. More recently, Shrivastava (1994) proposed a shift in the way we consider crises and consequently conduct our research. The concept of ‘ecologically sustainable economic development,’ he argues, provides a promising base and overcomes the limitations of our trial-and-error way of learning. Finally, Pearson and Clair (1998) offered a framework depicting the crisis management process and allowing for integration of psychological, social-political, and technological-structural research perspectives.
In spite of those repeated efforts towards integration, literature on crisis management still suffers from conceptual shortcomings that pave the way for contradictions and theoretical ambiguities. Literature on crisis management often eludes the underlying debate on crisis considered whether as an event or as a process. We know that this debate is certainly not the unique issue that still needs to be resolved in the way towards integration but we think it may be a valuable angle of analysis in order to have a proper view in the burgeoning literature on crisis management.
This debate as to consider crisis as an event or as a process has rarely been explicitly tackled in recent works. Shrivastava (1995) is one notable exception in opposing clearly the event versus process approach of crisis. As he states, “Crises are not events but processes extended in times and space.” (Shrivastava, 1995: 2).
However, this distinction is not explicitly established among researchers, what is even more striking, and this is certainly the most crucial point, is that most of the authors understate that crises are processes but still treat them as if they were events. Our work precisely focuses on this apparent contradiction. The objective of the paper is to discuss both approaches and see how they contribute to (the confusion of) the literature on crisis management.
An analysis of crisis definitions
We have based this paper on a content-analysis of 28 definitions of crisis found in the crisis management literature (see Appendix). We started with the hypothesis that every definition could be classified as process-based or as event-based. We had in mind some basic categories that could distinguish appropriately between both approaches. We used these categories to codify the definitions and assess to what kind of approach they could be linked to. The categories were derived from the prior definitions we had adopted about what was a process and what was an event. The definitions we had chosen were based on our prior knowledge of the literature on crisis management. We used them as guidelines for the content-analysis.
The event approach focuses on the notions of incidents or accidents as the unit of analysis.
Incidents or accidents constitute contingent and/or peculiar events as opposed to routines, regularities and experience. They may be a piece of information, a perturbation, a trouble, a tension that disrupt the fragile balance of the organization. Most of the time, definitions focus on the triggering properties of the event (e.g. Shrivastava, 1987). Triggering events may be considered as a sort of active constituent that put the organization to the test and push it to its extreme limits. It may be isolated in space and time (it happens in a particular moment in time and in a particular place) and has often quite distinguishable origins (social vs. technical source, internal versus external origin, etc.).
The process approach refers to a combination of actions, disruptions or to a succession of sequential (causal and linear) or systemic ( mutual causality, interactions, feedbacks) steps or phases that bluntly combines a series of different familiar or unfamiliar stakeholders, issues and resources resulting in a destructuring effect on the organization and its stakeholders. A process approach also suggests the succession of slow and quick evolution of events and actions prior and after the acute phase of the crisis. Crises are thus seen as phenomenon extended in time and space which induce most of the time a transformation of the organization.
In addition, for each definition, we surfaced the angle under which they were expressed. This
analysis led us to distinguish four angles adopted by authors to define crises:
— the nature of the crises, i.e. elements of definitions that intrinsically specify the concept;
— the causes of crises, i.e. elements of definitions that focus on the causes or the origins of crises;
— the consequences of crises, i.e. elements of definitions that focus on the consequences of crises;
— the dynamics of crisis, i.e. elements of definitions that focus on the way crises develop and manifest itself.
Event approaches in definitions Our content analysis suggests that an event approach focuses mostly on the nature and the consequences of a crisis. In this view, crises are explored through the lens of the triggering event, even if authors are not usually explicit on this point. Definitions actually concentrate on the visible part of crises. Authors attempt to define the concept in terms of impacts and damages. This perspective is very helpful to understand the crisis in its acute phase and contributes to nourish the literature on how to react in times of crisis in order to reduce its impact and resume activity as soon as possible. In spite of this contribution, this view privileges a reactive stance amongst managers and is not very helpful to improve prevention measures and learning capacities.
We observed that, in essence, crises were usually considered as negative events. In the event view of crisis management crises are usually defined as damaging and harmful disruptions or perturbations that threaten the very survival of the organization (Reilly, 1993). From another perspective, crises are considered as unanticipated and low-probability events (Shrivastava, 1987; Mitroff, Pauchant and Shrivastava, 1988; Pearson and Clair, 1988) and are often associated with high impacts (Weick, 1988). The very nature of crisis is precisely defined by the inability to plan or to measure the probability of occurrence and the potential risks it could induce should it occur. This is maybe one of the reasons why crisis management is often associated or even confused with risk management which deals basically with probabilistic methods to assess risks.
This risk perspective offers interesting contributions. Should it be considered only through the triggering event angle, a crisis would undoubtedly be a low-probability event. A hostile takeover, a malicious rumor or an industrial accident are very unlikely and may actually constitute a surprise with potential high impact on the organization. Definitions privileging this perspective regularly emphasize other recurrent facets of crises. They suggest for example that crises are surprising events. The seminal definition of Hermann (1963) is typical of such a position. According to Hermann the surprising effect of crisis induces high levels of stress among decision-makers and restricted time of reaction. Even though this feature is well shared among scholars (Smart and Vertinsky, 1984; Phelps, 1986; Reilly, 1993), the question of surprise has been discussed by others who have moderated this stance by introducing the notion of uncertainty (Lagadec, 1991). Perrow (1984) and Forgues (1993) for instance challenge the characteristic of surprise by referring to high risks industries in which crises are almost “normal”. From that point of view crises are surprising not because we don’t know what is going to happen but instead because we don’t know when it is going to happen. For instance, in the airline industry we all know that air crashes will continue to happen, but we can’t anticipate when and where air crashes they will happen.