Must disasters be rapidly occurring? The case for an expanded temporal typology of disasters

Introduction

The sociological study of disasters has ever since the 1950s grown steadily and become a well-established area of study, across many social contexts (e.g., Rodriguez et al., 2007). Research in this field has at many points wrestled with the question of “what constitutes a disaster,” as a way of moving the field forward. Numerous works have demonstrated that reaching widespread consensus about how best to define disasters is not such a straightforward task (e.g., Perry, 2007; Quarantelli, 1998), as disasters themselves are varied phenomena that can be studied from a variety of disciplinary and theoretical perspectives (Drabek, 1989: 253; Elliott and Hsu, 2016).

Perry (2005: 316) has noted that a lack in definitional consensus is not necessarily “a major deterrent to the conduct of research” in the sociology of disasters. Other areas of study have flourished even though there has not been widespread agreement about the conceptualization of certain key concepts. For example, Perry (2005: 315–316) has observed that “Cultural Anthropology has long thrived in the face of considerable controversy regarding the definition of culture.”

However, it is still beneficial to interrogate the meaning of disasters because doing so helps to clarify existing research. Coming to a relative consensus about what disasters refer to is fruitful, according to Quarantelli, because it prevents researchers from talking “past one another” (1998: 4). Without a clear and relatively well agreed upon understanding of definition of disasters, this makes it difficult—and some instances inappropriate—to compare findings, which is an important component of assembling a body of knowledge around a common theme or topic (Perry, 2005). Defining disasters also has the benefit of potentially provoking new lines of inquiry. Disasters need to be re-conceptualized because some aspects of disasters can change over time (De Smet et al., 2012) or because the distinction between natural and technological disasters can be theorized in a more nuanced fashion (e.g., Williams, 2008).

The primary goal of this article is to contribute to this definitional debate in the field of disasters research by concentrating on the concept’s temporal dimensions. As the following section reveals, even though there are numerous ways in which to conceptualize disasters, there has been a strong tendency to construe disasters as fast-moving events in the extant sociological literature. By questioning the merits of this way of theorizing disasters, I seek to develop a more multi-faceted temporal account of disasters that raises new problematics and lines of inquiry for the field to consider. I argue in particular for a typology of disasters that encompasses slow moving and recurrent social breakdowns. These type of phenomena are important to consider because they may be in some respects more socially impactful than rapid onset disasters are, as some accounts (e.g., Matthewman, 2015) have noted.

One of the upshots of understanding disasters in this way is that it further develops theory construction and interchange in the field of disasters research. As numerous surveys of the field have observed, disaster sociology has tended to occupy a more marginalized position within the discipline of sociology, despite dealing with a topic of immense importance (e.g., Stallings, 2002). Some accounts have attributed this situation to the reluctance that the field has largely shown to theory development, especially beyond its analytical boundaries (Tierney, 2007, Hsu and Elliott, 2016). By introducing a temporal dimension to the sociology of disasters, my aim here then is to further heighten exchange between disaster scholarship and other areas of sociological investigation. I find that there is much to be gained in theorizing disasters through a temporal lens, as it can for one thing elucidate the complexities of what disasters constitute and involve.

Speed as a key feature of disasters

The idea that disasters are events that must in some respects transpire quickly has a long history in the disasters research field. Some of the earliest and most influential definitions have included speed as one of the key characteristics of disasters. For example, the definition offered by Fritz, one of the pioneers of disasters research in sociology, frames a disaster as an event that is:

concentrated in time and space, in which a society, or a relatively self-sufficient subdivision of a society, undergoes severe danger and incurs such losses to its members and physical appurtenances that the social structure is disrupted and the fulfillment of all or some of the essential functions of the society is prevented. (Fritz, 1961: 655)

Other language used to formulate disasters as fast moving is found in Alexander’s theorization of what constitutes a natural disaster. For Alexander, “a natural disaster can be defined as some rapid, instantaneous or profound impact of the natural environment upon the socio-economic system” (1993: 4). Barton’s definition of disaster also contains a similar caveat; he frames disasters as “sudden and violent changes in the physical environment threatening both life and property” (1969: 53).

The upshot of conceptualizing disasters in these ways has meant that the field has narrowed the kinds of temporal phenomena it tends to study. Hence, this may be why there is a relatively large body of literature around events such as earthquakes, tornadoes, floods, when compared to the amount of research undertaken on slow-moving occurrences such as droughts and the incremental acidification of the world’s oceans.

However, it is important to note that this does not mean that no research has been undertaken on phenomena which lack a sudden quality. Nor does this mean there have not been any notable arguments whatsoever that call into question the definition of disasters as rapidly occurring events. These strands of research do exist and it is possible to find writings for example on phenomena such as climate change (e.g., van Aalst, 2006), antimicrobial resistance (Viens and Littmann, 2015) and droughts (e.g., Smith et al., 2012). But arguments that explicitly state why these phenomena ought to be construed as disasters are rarely—if ever—found in these studies, barring some notable exceptions (e.g., Viens and Littmann, 2015). Additionally, insofar as there have been theoretical arguments which attempt to expand the definition of disasters beyond rapidly occurring events (e.g., Buckle, 2005), it would be an overstep to say that these have been sustained or firmly accepted within the mainstream of the disasters research field in sociology.

Why this has been the case in part is because efforts have been made to regulate what the disaster research field ought to consist of, at least temporally speaking. Where this is especially evident is in the work of Quarantelli, who like Fritz is considered one of the seminal founders of disaster research in sociology. Quarantelli has argued on many occasions that there are good reasons why we ought to “exclude from the concept of ‘disaster’ all very diffused events, including traditional droughts and famines and certain kinds of epidemics” (2005a: 335).

Quarantelli’s concern is that moves to expand the boundaries of what constitutes a disaster run the risk of using the label “to capture too much” (2005a: 333). One of the benefits of holding a more restricted view of disasters is that it allows for a greater amount of analytical precision. Quarantelli (2005b), for example, has advanced a strong case for why disasters ought to be thought of as being distinct from catastrophes. In some respects, catastrophes can be thought of as a subset of disasters, but they differ considerably in terms of scale and intensity. Catastrophes signify that most, if not all, of a particular context has been impacted by some disruptive event, and this includes the capacities of local agencies to undertake any relief/recovery efforts (Quarantelli, 2005b). Disasters, by contrast, are not bound by such characteristics, as they can refer to socially disruptive phenomena that are not as all encompassing.

Correspondingly, this informs Quarantelli’s rationale for why we ought to limit the meaning of disasters to “primarily focused occasions, both in terms of time and space” (2005a: 334). Quarantelli suggests that it is not appropriate to include gradually occurring phenomena such as droughts, famines and the incremental spread of diseases like HIV within the definition of disasters because there are other concepts that are more fit for purpose. Quarantelli specifically asserts that the aforementioned phenomena are better classified as social or ecological “problems” (2005a: 335–336).

The aim of the next section is to explore the benefits and limitations of adopting this line of thinking. Although there has been much fruitful debate regarding a whole host of ways of defining disasters, including whether it is appropriate to distinguish between natural and technological disasters (e.g., Quarantelli, 1987; Williams, 2008), little discussion has been explicitly devoted to ascertaining the value of using rapidity as an essential criteria of disasters. The rest of this article is meant to demonstrate the utility of undertaking such a line of inquiry.

The questionable basis for excluding slow moving disruptive phenomena

In assessing the value of defining disasters as fast moving events, it is firstly important to recognize that imposing limits on the meaning of disasters does have some merit. Quarantelli and Dynes (1970: 328) have discussed how the term, disaster, can be a “sponge” concept, which potentially refers to an innumerable amount of things. However, if left too open-ended, the term runs the risk of becoming less meaningful and it may also prevent insightful empirical comparisons from taking place (Quarantelli, 1998). Therefore, this is why it is worthwhile to offer explicit and carefully delineated definitions of disasters.

But there is some cause to call into question how Quarantelli proposes to do so by equating disasters with rapidly occurring events. For one thing, this is because the distinction between diffuse and focused phenomena itself is not quite as self-explanatory as Quarantelli makes it out to be. The criteria that disasters must be fast moving are not necessarily a clear guide for which to determine the sorts of phenomena that are meant to count as disasters. And this is because the descriptive terms involved (e.g., “sudden,” “immediate,” “fast,” “rapid”) are all relative ones. The time it takes for a tornado to tear through a city may be considered fast when compared with a hurricane that may take longer to make a significant impact. And the same could be said about a hurricane in comparison with the time it usually takes for a devastating drought to take hold. However, more needs to be articulated as to the basis for why it would be appropriate to group tornadoes and hurricanes as being focused while not doing the same for droughts, besides merely asserting that they ought to be labeled as such. If the point of putting forth definitions is to increase the level of consensus there is around what disasters consist of, then more conceptual clarity is needed in terms of what the terms “focused,” “rapid,” “immediate,” and “sudden” actually mean.

Another reason to question the value of theorizing disasters as sudden events and of framing slow moving disruptive phenomena as social/ecological problems has to do with the lack of clarity around why this sharp distinction ought to be made. Quarantelli (1993) rightfully notes that slower-onset phenomena can be as or even much more impactful than occurrences that transpire suddenly. For example, he notes that “the economic costs alone, resulting from chronic exposure to hazardous fumes from chemical plants, would probably dwarf through the years by far what a typical sudden chemical disaster could bring about” (1993: 26). Yet Quarantelli asserts that we ought not to regard slow-moving phenomena as disasters because it is better to label them as social/ecological problems. However, in making this claim, Quarantelli does not ever offer a detailed or explicit description of what a social/ecological problem refers to. And accordingly, this is what makes his argument appear to be a circular one. It is not clear why temporally diffuse disruptive phenomena are best understood as social/ecological problems besides Quarantelli’s unexplained assertion that social/ecological problems are temporally diffuse.

Of course, this is not to say that a case is unable to be made to conceptualize slow-moving phenomena as social/ecological problems. Within sociology, a social problem has been theorized in many ways. For example, it has been suggested that a social problem is “an aspect of society that a significant number of people are concerned about and would like changed” (Henslin, 2000: 3). Along similar lines, a social problem has also been framed as “a putative condition or situation that is labeled a problem in the arenas of public discourse and action” (Hilgartner and Bosk, 1988: 55). If we go by any of these definitions, it is rather easy to comprehend why it would be apt to construe some slow moving socially disruptive phenomena as social problems.

Yet what remains unclear is why we ought to position protracted social disruptions as social/ecological problems instead of as disasters. One reason this approach ought to be called into question is because social/ecological problems and disasters are not conceptually antithetical to one another. As a number of works have argued (Drabek, 1989; Kreps and Drabek, 1996), disasters can be legitimately construed as peculiar types of social problems. Accordingly, this means that it is viable to theorize diffuse and slow-moving phenomena as both disasters and social/ecological problems, instead of having to choose between the two concepts.

Disasters “out of the box”? Disasters as everyday, structural occurrences

If it is indeed problematic to narrowly categorize disasters as fast moving social disruptions, how then to define disasters more adequately? One approach we might take is to broaden our temporal definition of disasters so that it captures how disasters have transformed over time to incorporate new elements.

The account of disasters developed by De Smet et al. (2012) has helped to advance this updated line of thinking. De Smet et al. argue that disasters in the 21st century are novel because they are quantitatively and qualitatively distinct. Disasters differ from those found in previous eras because they appear to be having “a more devastating impact on society, with more infrastructures destroyed and more people affected” (De Smet et al., 2012: 139–140). Also, what is distinctive about 21st century disasters is that “they have become extremely complex events to manage,” in that their temporal and spatial dimensions have changed (De Smet et al., 2012: 140). According to De Smet et al., disasters in the contemporary era tend to have longer-term impacts, even when their triggering events only have minor immediate effects. They cite as an example the 1984 Bhopal gas tragedy, which “caused death to more than 2000 people in the first few days” but caused many more thousands to die in the years that followed (De Smet et al., 2012: 140). De Smet et al. find cause then to advance a more “out of the box” understanding of disasters. Among other things, they question the suitability of a temporally and spatially restricted theorization of disasters, on the basis that the hazard and disaster landscape has evolved to encompass new features. ¹

The work of Matthewman (2015) provides even further impetus to depart from traditional accounts of disasters on temporal grounds. Matthewman is much more detailed about the shortcomings of focusing the field of disasters research on the rapid onset and the spectacular. Matthewman particularly expresses concern about the framing of disasters as events, instead of as processes. What is limiting about an event-based view of disasters is that it can overlook how events emerge out of processes that can run over extremely long periods of time. Matthewman cites as an example Klinenberg’s (2003) study of a heatwave that took place in Chicago in 1995. What happened during the heatwave “was ultimately determined by what had already accumulated across entire lifetimes” (Matthewman, 2015: 136). ² Additionally, there is the question of “when events actually end?”. There are some disasters—such as those involving nuclear radioactive materials—that stand to have an effect on people’s lives far into the future and when they conclude is not easy to mark.

Matthewman (2015: 144) thus proposes a number of new ways of thinking about disasters. One of the key insights he develops is to link disasters to Morton’s (2013) theory of “hyperobjects.” This concept of Morton’s (2013: 39) helps us think about how there are some entities, such as climate change, which are so “massively distributed in time and space” that it makes sense to regard them as entities that people must not only live with but also through, as an underlying and unfolding condition of the world in which people find themselves in. Consequently, for Matthewman, what this then calls into question is the modern—and also perhaps social scientific—tendency to regard disasters as being capable of being brought to a well-defined conclusion. Some types of disasters, such as radioactive nuclear contaminations, may be able to be “contained” for some periods of time, but it would be a mistake to regard them as having gone “away.” ³ Disasters, therefore, according to Matthewman, need to be more complexly theorized so that they capture this dynamic.

Relatedly, Matthewman also argues for a theorization of disasters that recognizes disasters as a kind of “slow violence”. As Nixon (2011) has articulated, slow violence describes a type of violence that is less well-understood and studied. It is violence that is “incremental and invisible, deferred and dispersed,” “ambient rather than spectacular,” and the “result of ongoing processes rather than their rupture” (Matthewman, 2015: 144–145). These qualities correspondingly inform the new theory of disasters that Matthewman advances, which diverges sharply from the conventional view of disasters as “contained,” intensive events. Matthewman (2015: 145) develops an understanding of disasters that is more oriented to the “everyday” and to the systematic aspects of social breakdowns that are not easy to detect because of the gradual speeds and expansive timescales at which they operate.

One of the analytical benefits of adopting this “everyday” conception of disasters is that it contributes to a more comprehensive understanding of the kinds of challenges and dangers people encounter. “Everyday” disasters are not just different from “contained” ones, since they can be in some respects even more consequential and severe. It is because they are regarded as “business as usual” that they “invoke no response, much less a recovery” (Matthewman, 2015: 145). And accordingly, this allows them to exact greater tolls than what more noticeable disasters may be able to cause. Matthewman (2015: 145–148) points to a number of cases where this dynamic is present, including the practice of burning of coal as a source of energy, which appears to be an ordinary “everyday” problem people face but in actuality is a slow moving disastrous occurrence if we consider its deleterious effects to public health, the economy, and various life systems.

Matsumoto’s (2013) classification of the 2011 Fukushima nuclear power plant accident as a kind of “structural disaster” engages in a somewhat similar line of thinking. ⁴ Matsumoto does not treat the devastating fallout of the Fukushima nuclear power plant disaster as an isolated socially disruptive incident. Rather, he regards it as the result of a long socio-historical process, stretching back over many decades. This is based on his analysis of what prevented the Japanese government from transmitting critical information about the nuclear accident to those affected in a timely manner. Matsumoto links the Japanese government’s recent failure to communicate to the rise of a culture of secrecy in governmental circles that developed in the lead up to the Second World War—as particularly evident in the concealment of a naval accident in 1937, which stood to reveal a critical fault in the Japanese method of building ships.

When considered in tandem with each other, what these three aforementioned accounts of disasters thus advance is a broader view of the temporalities that disasters encapsulate. De Smet et al. (2012) argue that disasters in the 21st century need to be re-conceptualized so that they capture—among other things—more temporally and spatially diffuse social breakdowns. Matthewman (2015) proposes an “everyday” understanding of disasters that is less fixated on spectacular events and more attuned to the ways disasters can be processual, incremental, silent, and socially invisible. And Matsumoto’s (2013) concept of the “structural disaster” regards disasters as being in some senses cyclical and historically rooted in unfolding social tendencies.

While these works astutely articulate some of the merits of expanding the temporal parameters of disasters, they leave the drawbacks of doing so less well considered. Quarantelli’s (2005a) concern about the definition of disasters becoming overly broad and meaningless especially needs to be addressed. And this is because allowing both rapidly occurring and slow-moving social breakdowns to be counted as disasters potentially runs the risk of ignoring key insights developed in the sociology of time about how fast and slow phenomena relate to one another. In the section that follows, I outline what some of these insights are and additionally I discuss how they can inform a temporal definition of disasters that remains sufficiently complex without being overly open-ended.

Towards an expanded temporal typology of disasters

Eliminating the requirement that disasters must be rapidly onset might be considered problematic in a number of regards. One is that it potentially runs the risk of conflating fast and slow moving phenomena with one another. Quarantelli (2005a: 334) expresses this worry by suggesting that disasters can only be meaningfully compared if they are “primarily focused occasions, both in terms of time and space.” This point made by Quarantelli is worthy of further reflection because it touches upon an idea that has recently gained much traction in the sociology of time: the observation that speed matters (e.g., Tomlinson, 2007; Hsu, 2011). Scholarly work around the theme of social acceleration—which refers to the speeding up of social life—has expressly produced insight into why speed influences our understanding of various social phenomena (e.g., Rosa, 2010; Vostal, 2014). According to this growing body of literature, one of the manifold properties of speed is that it affects how people make decisions. In an accelerated social context, people may be more able to reflexively alter the course of their lives, since the capacity to live in a post-traditional social order is linked to higher rates of social change (Hsu and Elliott, 2015). But in an accelerating social world, people may also feel that they are prone to act more recklessly because there is often little time for deliberation and reflection before action needs to be taken (Scheuerman, 2004; Tomlinson, 2007: 44–71). This facet of social acceleration correspondingly mirrors sociological work exploring what it means to institute slower-pace lifestyles (e.g., Osbaldiston, 2013). In a high-speed society, the process of slowing down is meant to help people become more “mindful” of the practices they engage in, such as in the consumption of food (Parkins, 2004). ⁵

These and other insights that have been developed in the scholarly debate surrounding social acceleration all appear to justify in some way Quarantelli’s unease with allowing sudden onset phenomena to be compared with those that are slower moving. At some level and in some instances, it makes sense to bracket rapid onset social breakdowns from those that are more temporally diffuse. However, allowing the label of disaster to encompass these kinds of phenomena does not necessarily mean that the distinction will entirely be effaced. This is because it is possible to develop a temporal typology of disasters that registers how quickly, slowly, or habitually they emerge. ⁶

One viable way in which to preserve the distinction between rapidly and gradually occurring social disruptions under the broader concept of disasters is to appeal to Barton’s (2005) classical typology of collective stress situations. Barton’s account of collective stress situations is not only meant to help “generalize across a broad range of troubling events having different labels” (Stallings, 2005: 245), it also identifies a number of variables that contribute to a more multifaceted theory of disasters. One of the variables Barton uses is the “concentration of time” (2005: 129, emphasis in original). Barton distinguishes between collective stress situations that are “sudden,” “gradual,” and “chronic” (2005: 129). Correspondingly, these labels may also be fruitfully used to describe the different types of disasters that there are, so as to allow more analytically precise and insightful comparisons to take place.

It would be limiting however to simply regard this tripartite understanding of the temporality of disasters as a clear-cut system of differentiation. This refers to the problematic nature of regarding sudden, gradual, and chronic types of disasters as being wholly dissimilar from one another, even if such distinctions can still be in some instances useful to maintain. There is also a need to draw out the interconnections between these three kinds of disasters that I propose. And this has to do with the other additional insights that have emerged in the scholarly debate around social acceleration. Theoretically focused sociological work on speed has recently established how it may be limiting to regard slow and fast temporalities as being diametrically opposed (Hsu, 2015; Rosa, 2010). For one thing, this is because slowness may actually be an intrinsic—though unintended—consequence of social acceleration, as evident in the case of a traffic jam (Rosa, 2010). ⁷ Theoretical accounts of social acceleration have also noted how some slowdowns may actually play a functional role in the intensification of further speed-ups, as apparent in the practice of using naps to improve worker productivity (Hsu, 2014). These insights express why it is worthwhile to adopt a more relational view of speed and slowness, so that they are not only conceived of as polar opposites.

A similar case can be made that a relational approach is also needed to adequately theorize the temporalities of disasters. Disasters that fall into different temporal categories may in fact be inter-connected in some way. Global anthropogenic climate change serves as a good example. It is a phenomenon that is unfolding at a slower rate than many other types of social breakdowns, such as hurricanes and floods, which disaster researchers have traditionally tended to focus on. Yet there is also a sense in which anthropogenic climate change still has a connection to disasters that are more rapidly onset. This refers to the increase in severity and frequency of some temporally focused disasters that climate change may bring about, as some studies have forecasted. ⁸

Dividing disasters into different temporal registers thus not only allows us to understand how disasters vary, it also builds up our understanding of how temporally complex disasters can be. It may be more productive to think of disasters as being temporally hybrid. There is a case to be made that some disasters involve elements of being protracted, suddenly occurring, and structurally recurrent and that these variables cross and tangle in innumerable ways, depending on the disaster in question. It may be, in other words, highly limiting to think of disasters as mono-temporal phenomena.

Conclusion

In this article, I have sought to expand how disasters are temporally conceptualized within sociology. I have done this by offering a critique of the traditional definition of disasters as socially disruptive events that must be temporally focused (e.g., Quarantelli, 2005a). The alternative approach to theorizing disasters that I have put forth encompasses a wider and more hybrid range of temporalities. It holds that disasters can be sudden, gradual, and/or chronic.

But the temporal typology of disasters that I have developed may by extension also open up the field of disasters research to new spatial dimensions. This has to do with the intersections between time and space. ⁹ The very same grounds used to justify why disasters must be temporally focused have also been used to substantiate why disasters must be spatially concentrated. ¹⁰ Accordingly, it stands to reason that if there are limitations with the former criteria, then the latter one might need to be called into question as well. This then prompts us to arrive at a more “mobile” theory of disasters, which is informed by the theoretical and empirical insights developed in the “new mobilities paradigm” (Sheller and Urry, 2006). Work produced in this rapidly expanding area of study has emphasized how it is fruitful to give movement more of an analytical focus. In the case of disasters, this means that it is not enough to investigate disasters that are spatially restricted. Rather, there is a need to understand how disasters circulate around the world and how they can be spatially diffuse. However, this is an implication that is in need of further theoretical, methodological, and empirical elaboration, ¹¹ since I have only been able in this article to directly address the way disasters have been temporally conceptualized.

Another issue that falls beyond the scope of this article, which remains important to consider, concerns how disasters are defined in other settings. Thus far, I have only dealt with the sociological definition of disasters as an analytical tool utilized by researchers in the field. However, disaster definitions originate and circulate in other sectors beyond the Academy and they are “of crucial importance” in a multitude of ways, as Buckle (2005: 177) has noted. How much does the argument advanced in this article—that we ought to expand how disasters are temporally theorized—also apply to other domains of society? This query is worthwhile to pursue because disasters also appear to be regarded as rapidly occurring events in other realms outside of the Academy. The United Nations (2004: 43), for example, formulates a disaster as a “a sudden, calamitous event that causes serious disruption to the functioning of a community or a society.” And according to the American Red Cross, “a disaster produces a range and level of immediate suffering and basic human needs that cannot be promptly or adequately addressed by the affected people” (as quoted in Strangeland, 2010: 422). If these and other organizations elect to adopt a broader temporal view of disasters, what then might the consequences of this decision be? Might it potentially direct more awareness and resources to social problems that have been overlooked? Or might we wish to refrain from extending the temporal definition of disasters outside of the Academy because the overuse of disaster terminology to marshal action around a problem may have the opposite effect (Viens and Littmann, 2015: 261)? Clearly, this is an issue that needs to be further investigated and given the multitude of domains where definitions of disasters are found, a “one-size-fits-all” approach will not likely be the solution.

And lastly, there is also the matter of how slow moving and chronic types of disasters can be adequately studied. When it comes to analyzing a broader temporal range of disasters, what new methodological tools and conceptual lenses, if any, need to pioneered or further developed? This question is worth asking since disaster research methods can in some respects be quite distinctive in terms of how they are deployed. In discussions about the unique methodological challenges that disaster researchers face, “time” tends to be framed as “a critical variable” (Norris, 2006: 173). There is a tendency in particular to identify the timeliness of fieldwork as a major obstacle. As Killian (2002: 53) notes, “the longer [the disaster researcher] takes to get into the field the more remote the disaster becomes for his [sic] subjects.” But prolonged and recurrent kinds disasters raise other sets of research issues that need to be attended to. As the influential work of Pierson (2003) has established, shifting focus away from the immediate toward the protracted involves the development of new analytical frameworks and instruments in the social sciences. This insight has in turn encouraged some working in the burgeoning field of crisis research to start thinking about what it means methodologically to investigate crises that are slow moving, resulting in a call for more longitudinal studies and a shift toward “conjunctural” explanations that depart from “linear, reductionist, variance-based causal thinking that focuses on proximate factors” (Buchanan and Denyer, 2013: 217) However, more work is needed in the field of disasters research to capitalize and build upon these reflections, partly since the concepts of disaster and crisis do not map exactly onto one another (Boin, 2005).

The expanded temporal typology of disasters I have put forth in this article should thus be thought of as a starting point instead of as a fully formed and rigid systematic account of the temporality of disasters. As I have briefly indicated, there is still quite a lot of terrain left to cover before we are able to arrive at a truly comprehensive theorization of disaster temporalities. Why we ought to keep pursuing this project is because it can serve to further connect disasters research with other key lines of inquiry in the social sciences, which in turn can help to bring disasters from the margins to the mainstream of sociology (Tierney, 2007; Hsu and Elliott, 2016).