Emergency Management


Paper presented at theNATO Advanced Research Workshop on Understanding and
Responding to Terrorism: AMulti-Dimensional Approach. Sponsored by NATO and the
Turkish National Police / Turkish Institute for Police Studies at the University of North
Texas. September 8-9, 2006. Washington, D.C.
Funds for portions of this research were provided by Multidisciplinary Center for
Earthquake and Engineering Research (MCEER) New Technologies in Emergency
Management, No. 00-10-81 and Measures of Resilience No. 99-32-01; the National
Science Foundation; the Public Entity Risk Institute No. 2001-70 (Kathleen Tierney,
Principal Investigator); National Science Foundation No. 0603561 and 0510188 (James
Kendra and Tricia Wachtendorf, Principal Investigators) and the University of Delaware
Research Foundation (Tricia Wachtendorf, Principal Investigator). We are grateful to the
South Street Seaport Museum (Mr. Jeffrey Remling, Collections Director), for access to
interviews with participants in the waterborne operations. Funding to the museum for
these interviews was provided by the National Endowment for the Humanities, and the
interviews were conducted by David Tarnow.
Improvisation, Creativity, and the Art of Emergency Management’
James Kendra, Ph.D.
Emergency Administration and Planning Program
Department ofPublic Administration
University of North Texas
Tricia Wachtendorf, Ph.D.
Disaster Research Center
Department of Sociology and Criminal Justice
University of Delaware
Improvisation is a significant feature of every disaster, and Tierney (2002) has argued
that, if an event doesn’t require improvisation, it is probably not a disaster. Improvisation
has had something of a checkered history in the emergency management field since its
appearance in a disaster response seems to suggest a failure to plan for a particular
contingency. Even scholars who have recognized the value ofthis capacity have tended to
subordinate it to planning. Kreps (1991: 34) for example, who has completed some of the
most detailed studies of organizational improvisation in disaster, has defined
improvisation as organizing “during an event,” while preparedness is organizing “before
an event.” He has stressed that preparedness and improvisation are the “foundations” of
emergency management (31), yet he nevertheless privileges preparedness, especially
planning, as the favored element. Drabek (2001) too emphasizes the need for planning to
reduce the incidence of (the inevitably necessary) improvisation. Thus improvisation
occupies a somewhat conflicted space in the realm ofemergency and crisis management
capacities: we plan in detail so that we don’t have to improvise, knowing that we will
have to improvise.
This paper discusses emerging understandings of improvisation in emergency
management and their relationship to planning as well as to other such noted disaster
phenomena as emergence, or the appearance of new groups of people organized to meet
disaster-related needs. We reconsider the suggestion that improvisation must be
positioned with respect to planning in such a way that it somehow seems to be the weak
link, or an indication ofsome failure or dysfunction. We argue that improvisation is a
distinct capacity that individuals and groups employ, and that while planning
encompasses the normative “what ought to be done,” improvisation encompasses the
emergent and actual “what needs to be done.” The public policy scholar Michael
McGuire has suggested that “plans are hypotheses” about a projected future condition.
Since real conditions are likely to differ, adjustment to plans will always be necessary.
Weick (1998) has drawn on jazz as a model and lens for understanding improvisation in
1 This is an earlier draft of a paper presented at the NATO Advanced Research Workshop on Understanding
and Responding to Terrorism: A Multi-Dimensional Approach, sponsored by NATO and the Turkish
National Police/ Turkish Institute for Police Studies at the University of North Texas. September 8-9,2006.
Washington, D.C.. The draft is made available with permission from the NATO conference organizing
committee. The final paper will be published in the NATO Security through Science series.
organizational settings. Jazz musicians are not censured for their improvisations, nor are
they criticized for not composing their scores in advance. Rather, their extemporaneous
compositions are celebrated-that’s what jazz is, and a successful jazz performance is the
end result of training, knowledge, practice, and experimentation. Similarly,
improvisational comedy and theater are regarded as high expressions of a stage
performer’s art. This doesn’t mean that these performers do not practice or build
repertoires of material that they can draw upon in given circumstances. On the contrary,
they work to build their knowledge across a range of fields, and this knowledge provides
the elements for each improvisational outcome. Improvisation plays an equally important
role in emergency management, where training, practice, and knowledge of both the field
and the community serve as repertoires of material emergency managers can draw upon
in the ambiguous and dynamic conditions of a disaster where not every need has been
anticipated or accounted for (Wachtendorf, 2004, Wachtendorf& Kendra, 2005).
Scholars and practitioners often define emergency management as both an art and a
science (see, for example, Rubin, 2004), an understanding that applies to other
professions where people interact with the natural environment. For example in medicine,
a specialty routinely also described as art and science, the environment is the highly
localized one of a patient’s body; in navigation and aviation the environment is the
natural one of ocean basins, the atmosphere, land forms, weather, and climate. But in
emergency management, the setting is even more complex, consisting not only of the
earth’s processes but also humanity’s industrial activity and the distribution of people and
their complex social and economic systems, systems that are imperfectly understood even
in normal times yet whose ramifications extend into, and even create, the situations we
colloquially call disasters. For the emergency manager, the science extends from the
social and natural sciences that provide the foundation for understanding the causes and
distribution of hazard. The earth and atmospheric sciences tell us about geologic and
climatic processes; the social sciences tell us about people’s understanding of and
response to those processes and moreover provide insight into the social systems that lead
to exposure to forces of nature, or that lead to mismanagement of our industrial systems
resulting in systems failure, hazardous releases, and environmental
But more can be said about the science of emergency management, for though it involves
applications of principles from various disciplines, its practice is a particular kind of
science, one that leaves behind standard methods and well-defined procedures of its
foundational disciplines to become something more rooted in interpretation, judgment,
and the negotiation of ambiguity. Funtowicz and Ravetz (1992) have identified various
forms of scientific practice. They distinguish between Kuhnian “normal science” where
scholars work within established norms and sets of procedures, and “post-normal
science,” characterized by problems both with high uncertainties and high decision
stakes, where issues are often ill-defined, proper methods unclear, data mixed or
incomplete, and social and political considerations intermixed. Environmental challenges
such as global warming are “post normal,” for example. Funtowicz and Ravetz ( 1992)
also identified a middle ground of moderately high decision stakes and moderately high
uncertainty, a conceptual zone requiring the application of professional craftsmanship, as
is required in the various engineering disciplines.
These are also the characteristics ofemergencies, where the precise unfolding of
circumstances is unknown. As in other disciplines, the art is in the application of
knowledge in irregular circumstances, where we encounter the genuine evolving of actual
events, not an idealized type. “Science deals with regularities in our experience; art deals
with singularities” (Weinberg, 1985: 60). Singularities are the elements of disaster not
predicted in advance. Weick (1998), in reviewing recent management theory, noted that
all management is improvisation to some extent. If that observation is valid for
organizations in commercial enterprises where decision making scenarios that are often
described as “dynamic” are ponderous as compared to crisis management time scales,
then surely improvisation would seem to be elemental in emergency management, too.
Given this, we suggest that improvisation is a high expression of an emergency manager’s
Scholars have defined improvisation variously, and Weick (1998: 546-547) provided an
often-cited definition:
Considered as a noun, an improvisation is a transformation of some
original model. Considered as a verb, improvisation is composing in real
time that begins with embellishments of a simple model, but increasingly
feeds on these embellishments themselves to move father from the original
melody and closer to a new composition. Whether treated as a noun or a
verb, improvisation is guided activity whose guidance comes from elapsed
patterns discovered retrospectively.
Miner, Bassoff, and Moorman (2001: 314) have defined improvisation as ” … the
deliberate and substantive fusion ofthe design and execution of a novel production.” In
other words, theorists place an emphasis on time, and all stress the simultaneous or nearsimultaneous
conception and implementation of action, as in the playing of a note in jazz
or the introduction ofa product design element or process in manufacturing. Miner,
Bassoff, and Moorman (2001) have further identified several distinct improvisational
products. These include “artifactual improvisations,” where the outcome is a tool or
object; “behavioral improvisations,” where the outcome is a new process of set of actions;
and “interpretive improvisations, where the product is a new way of understanding needs,
obligations, or conditions. They note that while much writing about improvisation in the
management literature takes a generally positive stance, outcomes are not always
favorable, and moreover organizations do not always learn from their improvisations or
gain methods or insights with any longevity.
Adopting a different approach, we (Wachtendorf, 2004; Wachtendorfand Kendra, 2005;
Kendra and Wachtendorf, 2006) have postulated a 3-element typology of improvisational
types, based on the emergence of improvisational activity with respect to an existing plan,
model for action, or standard procedures. In reproductive improvisation. improvisers
recreate an existing capacity; in adaptive improvisation, they amend an existing capacity
to match changing demands, producing a new system, and in creative improvisation they
create an entirely new capacity in the absence of an existing model. All of these forms
occur under tight time constraints and with pressing demands for action.
Public officials, especially those accustomed to highly controlled and regimented
organizational structures, tend to be discomfited by the prospect of improvisation in their
environs, since improvisation suggests not only novel, untested, and perhaps unexpected
actions, but also actions that may be taken unbeknownst to other participants in an
emergency response. The result of such autonomous activity is likely to be confusion,
waste, and poor delivery of emergency management services. Indeed, for some officials,
and even for some scholars, the image of improvisation is of independent, disconnected,
and chaotic activity-the kind of activity that emergency plans, and management
structures such as the Incident Command System, were developed to prevent. Such
images are, at best, a caricature of improvisation and at worst reinforce the perception
that improvising is an also-ran to detailed planning. Improvisation, if it is truly a set of
individualistic acts, can go badly and yield the malfunctional outcome feared by
emergency managers and other officials. However, group improvisations, the emergent
products of collective problem-solving activity, can be highly effective responses to
unusual situations and novel demands.
Other works explore in detail the numerous examples of improvisation during the World
Trade Center response (Kendra & Wachtendorf, 2003a; Wachtendorf, 2004; Wachtendorf
and Kendra, 2005; Kendra and Wachtendorf, 2006). We review some of them here to
illustrate the important role improvisation plays in disaster management. On the morning
of September 11,2001, Office of Emergency Management (OEM) officials were at their
emergency operations center (EOC) preparing for a bio-terrorism exercise. The EOC and
OEM offices were located at 7 World Trade Center, adjacent the Twin Towers. This
state-of-the-art facility was constructed in the late 1990s and was equipped with
sophisticated monitoring equipment and designed in such a way, in work pods comprised
of specific organizational representatives, to maximize coordination between the
numerous agencies that would need to respond to various emergency support functions. 7
World Trade Center was damaged by debris from the Twin Towers and collapsed later
that afternoon.
With no back-up facility in place, OEM staff and city agency representatives needed to
quickly improvise a new site of central coordination. Eventually that day, a temporary site
was established in the police academy library, and within several days the operations were
shifted to a large shipping pier along the Hudson River (the intended site for the next
day’s bio-terrorism drill). The initial goal was clear: improvise in such a way as to closely
reproduce the 7 World Trade Center EOC. The original EOC would have proven a
formidable site from which to launch a response of this kind. In other words,
improvisation manifested itself so as to employ substitutes in an effort to replicate the
original facility.
In a like manner, the overwhelming involvement of local agencies and volunteers as well
as the widespread convergence of people and organizations from across the country to
help with the response generated a demand for adaptive improvisations with respect to
credentialing. The standard operating procedure of relying on agency badges for entry
into areas associated with the response efforts was inadequate to limit the number of
personnel and volunteers from entering secured areas. Indeed, the emerging response
network included key participants with much needed expertise, but who were either
unaffiliated with a recognized agency, associated with an agency from outside the
traditional response network, or affiliated with an organization from outside the greater
New York metropolitan area. At the same time, allowing every city worker with an
agency badge to have site access would have been equally unmanageable. A new
credentialing system consequently had to be adapted from the original protocol. Then, as
response locations changed, as badges needed to be accounted for, as access needed to be
tightened – for example, for safety reasons at Ground Zero– and as more sophisticated
credentialing equipment became available, adaptations were made to the badges
themselves. Even as a new credentialing system was improvised, those in charge of site
security found that additional adaptations were needed to account for new circumstances
or unintended consequences of the new badges themselves.
The waterborne evacuation of Lower Manhattan is an exemplar of creative improvisation.
During that event, several hundred thousand people were evacuated in a spontaneous fleet
ofassorted vessels: towboats, dinner cruise boats, tour boats, yachts, and other craft
converged on Manhattan and shuttled evacuees to Staten Island, Brooklyn, or various
points in New Jersey. Upon disembarking people, the boats carried supplies and rescue
workers into the city, an operation which for some vessels lasted several days. There was
no pre-planning for this kind of event, although the Coast Guard personnel who helped to
coordinate this event drew on elements of search and rescue and crisis management plans
that presupposed a much smaller incident. Instead, participants in the evacuation and
supply lift operation responded to cues within their environment based on their own
repertoire of knowledge and experience to determine that a response was necessary. Such
cues included sightings of individuals gathering at the waterfront as well as sightings of
and radio transmissions from other vessels. Their repertoire included an understanding of
the harbor, knowledge of the commuter population that comprised the daily population of
Lower Manhattan, training in spills or search and rescue operations, and an occupational
ethos of rescue, albeit one typically associated with rescue-at-sea operations.
We should be clear when we discuss improvisation that we’re not talking about everyone
“doing his own thing.” Certainly there is room for individual initiative and innovation.
For a particular set of necessary tasks, tools and methods can be assembled from available
materials. Such improvisations, enacted at a local level and at a restricted scale, are not
likely to disrupt action in a larger response milieu. But improvisations that are likely to
ramify throughout organizational space present a different set of challenges. Here
decisions and actions can influence the conduct ofother actors, perhaps by drawing away
resources of personnel or equipment, or by shifting the conditions of the operational
environment that someone else depends on.
It is this outcome-an anarchical collapse of organization-that emergency planners fear
most in improvisational settings and that operational protocols such as the Incident
Command System were developed to avoid. Yet the Incident Command System has its
faults. Buck, Trainor, and Aguirre (2006) noted that ICS functions best in situations that
are familiar, where participants have worked together before, and that are of a limited
scope. It is less effective, according to their findings, in situations that are new,
surprising, or massive-in other words, where its promised benefits are most urgent!
This suggests that there is something about making ICS work that is not inherent in its
design, but rather that there is something that organizations bring into the mix (otherwise
it would always work). Bigley and Roberts (200 l) found that members of incident
command systems indeed did improvise some aspects of it to help them in particular
situations, a finding that again suggests that responders have an “extra-ICS” set of
capacities that they draw on in an emergency but which ICS allows them to focus and
Which, also, supports ICS, too.
Our work on improvisation supports this broad proposition-in particular, that the
individual and organizational properties that make ICS “work” can allow organizations to
cooperate effectively outside the ICS regime and might in fact be the starting point for a
larger organizational development. Stated differently, the same qualities that make ICS
work can allow other kinds of organization to be effective, as well. For example, in the
waterborne evacuation, the participants were familiar with each other. They had worked
together before, and they were familiar with the resources and with the maritime
operating environment. In this event, a loose-fitting organization evolved with a very flat
hierarchy that emerged around coordination and traffic control of vessels and the
establishment ofthree main evacuee marshalling points ashore in Manhattan. A bus
company joined the operation on the New Jersey side, transporting the disembarked
evacuees to mass transit points. Participants in this operation became participants by
identifying and filling needs; organizing around tasks and geography; and providing
information and allowing the persistence of organizational autonomy.
Moreover, this operation occurred almost completely disconnected from the response
operations at the WTC site. It was a separate entity, with no intersection with the
shoreside emergency management function regarding sharing or drawing away resources.
In fact, these vessels brought in personnel, equipment, and fuel, functioning in a modular
organizational structure that carried out collectively-defined tasks in support of overall
response goals.
But, though this event was improvised it was not anarchical. Some participants referred to
“chaos,” and in reviewing certain photographs and video of the event it certainly appears,
at least superficially, to be extremely hectic. But there is other information to bear in
mind when considering whether “anarchy” was involved, or something else. There were
no significant injuries or vessel mishaps of any substance during the evacuation. Boat
operators negotiated with each other for access to docking space, or simply stood off and
waited their turn. Several participants reported that the usual competitive maneuvering
did not occur between captains.
The utility ofcommunity organizations, and community-based participation, should not
be underestimated even in terrorist-attack scenarios. While we lately hear considerable
discussion of the need for “command and control,” the desired outcome of thiscoordinated
use of resources-is achievable through many types of organizational
structures, some of which are likely to be far more appropriate in civil contexts where
command-and-control has little resonance. Instead, preparedness activities should involve
considerable outreach into communities.
We need to be identifying resources in advance, and we need to allow our collective
imaginations to roam over the range of skills and assets that are out there. If there are
obstacles to a particular application of a certain resource, those should be identified and
examined to see if they might be reasonably set aside in a compelling emergency. Our
work in the World Trade Center disaster uncovered numerous such “workarounds” of
greater or lesser scope-for example, boats carried passengers in excess oftheir certified
capacities, or carried passengers with no certificate to do so.
Where do such resources exist? They exist everywhere in our communities. To take the
WTC attack as an example, we have said elsewhere that New York City contained every
skill needed to handle the disaster (Kendra and Wachtendorf, 2003a). But it is important
to bear in mind that those skills and capacities were not all in the formal, pre-established
emergency response organizations. Some of those skills were in other organizations in the
city government. For instance, as has been documented elsewhere (Langewiesche, 2002;
Wachtendorf, 2004; Wachtendorf and Kendra, 2005) the city’s Department ofDesign and
Construction had had no previous disaster management experience, yet emerged as the
lead agency for the long “unbuilding” of the Trade Center. The Department of Health had
no statutory authority for the environmental health aspects ofthe response, yet they took
on that role. The private sector provided personnel, materials, and expertise, as in the
construction trades. Restaurants supplied food (Kendra and Wachtendorf, 2003b), while
community based organizations from the service, advocacy, and faith communities
provided support and comfort. The speed at which OEM was able to reestablish the EOC
after it was destroyed was in no small part due to the tremendous resources at its
fingertips and its knowledge ofthose resources. Expanding the lease on the shipping pier
from a day-long exercise to occupation of several months and transforming the site into
the necessary facility required a flexibility to conceive of that space in non-traditional
ways. Without having an open mind to envision resources in new ways, or without having
an organizational culture to allow for that envisioning, improvisation is improbable.
Without having an understanding ofthe resources available to begin with, improvisation
becomes impossible.
As with any art, the skill of improvisation cannot taught by assigning a script or working
through a check-list of steps. Its principles, however, can be taught and the knack of
improvisation can be developed through practice and exercise. To the extent that acts of
improvisation depend on attributes ofcreativity, improvisation can be enhanced by
removing organizational impediments to creative thinking (Kendra and Wachtendorf,
2003c) and by creating an organizational culture that values improvisation. Again, we
don’t argue that plans or planning should be discarded, only that plans should be seen as
guides, and planning should be seen as rehearsing for later improvising. Simply inviting
representatives of organizations and agencies to meet regularly can be a useful activity, to
build acquaintances, to share information, and to develop norms of mutual interaction.
Drabek (200 I: 11) argued that “the capacity to improvise is greatest when the pre-disaster
response network has been nurtured and integrated.” The extent to which plans do
account for circumstances that arise in a disaster in fact can facilitate the process of
improvisation. That is, when certain elements ofthe response are not disrupted or are
encompassed in prior planning, responders can direct their attention to the unanticipated
elements of the disaster and improvise with a more concentrated focus. It is not a matter
of abandoning planning for improvisation, nor is it a matter of planning with the goal of
eliminating the need for improvisation. Rather, planning and improvisation are important
aspects ofany effective disaster response and are best considered as complementary.
Thus far we have been concerned with a practical or instrumental outcome oftolerance
for, even celebration of, improvisation. But we can argue for a more theoretical
significance for an institutionalized appreciation for improvisation: that it provides
leverage for putting management-eontemporary, updated, current conceptions of
management-into emergency management, at least with regard to the response phase,
the phase that is least amenable to standard management practices familiar in routinelyfunctioning
organizations and situations. As noted earlier, Weick (1998) asserted that all
management is improvisation to some extent. Thus, as a common reference point,
improvisation can provide a connection between emergency management and “regular”
management. Such a connection would be important in the further development and
grounding of the emergency management field.
Weick (1996), in the 40th anniversary issue ofAdministrative Science Quarterly, reflected
on the challenges facing organizational scientists-ehallenges that included
encroachments by the competing discipline of economics, and by a shift in the legitimacy
of knowledge production away from universities and toward the business sector. Weick
developed his argument by allegory, highlighting the failure of’firefighters at Mann Gulch
and South Canyon to “drop their tools.” Burdened by their heavy equipment, they
couldn’t run fast enough and perished in advancing wildfires. For Weick, those tools
emerged as central to the firefighters’ identities as firefighters, and the inability to discard
them was the physical manifestation ofan individual and organizational incapacity to
make new sense of a shifting situation. The tools were a kind of existential compass that
oriented them to a certain view of themselves and their relationship to that place, though
its features shifted around them. Weick thought that organization researchers now faced a
similar crisis of situational awareness, which demanded an ability and a willingness to
drop old tools of research methods and paradigmatic understanding.
An inverse situation is possible-the need for a discipline to pick up its tools (see Beunza
and Stark, 2003). Emergency managers often talk about the “hands on” nature of their
work, especially to distinguish it from research and academia. But the phrase is at best
metaphorical, referring to engagement with real events, because emergency managers do
not really use “tools” in the sense conferred by the phrase. First responders, rescue
workers, and heavy equipment operators use tools. However, the principal tools of
emergency managers are intellectual and conceptual, involving relationships of people,
things, and places. (Quarantelli, 1997; McEntire, 2005; Pine, 2006).
While it is easy to see the intellectual heritage of the emergency management field, it is
less easy to find a unifying theme for its practice, since emergency management
encompasses an enormous array of activities taking place at many temporal scales and at
multiple organizational levels. Indeed, much of what emergency managers do does not
involve emergencies, but rather involves reducing the likelihood of emergency
(mitigation) or preparing to respond to one (preparedness) (McEntire, 2005). These
activities involve management as generally understood. In a recent essay for example,
Pine (2006) discusses how management theory applies to emergency management,
identifying a number of management principles that relate to recognized emergency
management activities. These principles include long-term and strategic management
approaches, sustainability, diversity, systems theory, and flexible thinking and flexible
structures. He stresses the need for “improving the management in emergency
management.” Our assessment is generally congruent with Pine’s comments on the
intersection of management theory with emergency management, and that the emergency
management discipline shares certain analogous historical trajectories, as in both fields’
borrowing of themes and practices from other disciplines. However, although the
applicability ofmanagement theory to emergency management can be seen, application of
management principles is more by the accident of situational necessity than by conscious
effort to incorporate management theories. As a consequence much insight is lost, and
there is a tendency to see emergency management as a separate kind of management,
where ordinary rules don’t apply. But people are people whether the milieu is corporate or
crisis, and there is no good reason to suppose that the social relationships of organizations
in emergency are going to be different from the relationships that function in other
situations. In fact, the guiding idea at the founding of the disasters field was the opposite:
that disasters were an opportunity to examine social relationships in a compressed time
interval (Fritz, 1961: 654), an idea that depended on the congruence of these relationships
in both crisis and normal times. It might be more accurate to argue for the need for
“putting the management into emergency management” because although we can see the
relevance of management science and the appearance of social phenomena that occur in
other settings explicit management applications are sketchy at best in emergency
There is one exception to this observation about the sketchiness of management theory in
emergency management, and that is the near-universal approval of the Incident Command
System, which was developed according to the management principles that were
ascendant in the early to mid-1970s (Irwin, 1989). These principles have, however, been
superseded in many applications, especially those focusing on rapidly-changing
environments (Daft, 2004). Thus management theory’s most significant and enduring
impact is in a largely obsolete organizational scheme (Waugh, 2006).
If we are going to put the management into emergency management, what should it look
like? While it is beyond the scope ofthis paper to offer a complete mapping of
management principles to emergency settings (a task begun by Pine (2006), we argue
that creativity and improvisation are good candidates for comprising an orienting theme
or set of concepts for emergency management in the response phase. If we take the view
that disasters are social events, as Dombrowsky (1995) and others have argued, involving
disrupted social structure, as in Fritz’s definition (1961: 655), we can see that emergency
managers manage social relationships to correct those disruptions. They do this by rapidly
recombining resources, skills, and experience, capitalizing on agreements, expectations,
and norms.
Either for improving the management in the field, or for putting it in, an updated
approach is needed. Pine (2006) suggests the need for “more dynamic organizational
structures” that are responsive to changing needs, especially given the open system
character of emergency management operations. Emergencies are the most open of open
systems; much ofthe management, when an emergency is involved, is of fleeting and
transient things. Individual and organizational participants come and go, so that
emergency managers are really managing a shifting pastiche of relationships and
arrangements. They are managing, or trying to manage, interactions of social and physical
systems, not in the comparatively stable situation of a factory or a firm, but in the
changing circumstances of environmental turbulence.
Harrald (2006) argues that characteristics of both “agility and discipline” are needed so
that organizations can maintain their coherence while at the same time respond to
surprising conditions. Plans and structures that allow for easy modification facilitate both
the shared operational concepts and mutable procedures needed in crisis, but practice
together, as in a jazz band or theater ensemble, is need to fine-tune the artistry of
response. We would argue for a scientific basis of emergency management that
emphasizes management, not emergency, but it should be the management of dynamism,
ofambiguity, and ofchange. Only improvisational skills can provide the necessary
capacity for this spirited approach to crisis.
Beunza, D. and D. Stark. 2003. A Desk on the 20th Floor: Survival and SenseMaking
in a Trading Room. Working Paper Series, Center on Organizational