An emergency is the ultimate test for an organization, and comparative research across nations has uncovered that things go wrong when there is a breakdown in information, communication and coordination (Dykstra, 2003). Our study revealed that the operations on Utøya 22.7.2011 were characterized by similar challenges, which we categorized into four main domains: organizational structure, leadership, information flow and ICT, and coordination. We will establish the rudiments of our theoretical framework, before we analyze these domains and propose ways to optimize their functions by synthesizing findings from leading schools and researchers in organizational science. Our findings are of interest to all organizations operating in environments in which they might face spikes in the degree of uncertainty and interdependence.
According to Scott´s three levels of analysis (2007), this theoretical framework starts with (1) the ecological level, followed by (2) the organizational level, and (3) the social-psychological level.
The Ecological Level
The synthetic organization. The synthetic organization can be seen at both the ecological- and the organizational level; though, we categorize it into the ecological system, for synthetic organizations are born purely due to of environmental factors, and die when these factors disappear. According to Thompson (1967), the synthetic organisation forms in the context of large-scale natural disasters. It is an ad-hoc organisation and thus, by definition, a spontaneously emerging organisation without prior designation of authority, or formal authority that enforces rules, decisions, or control of resources. Yet, the synthetic organisation is characterised by a consensus amongst participants concerning the goal of the operation, which assumes a sense of rational structure.
Emergent response groups. Task-forces are temporary groups that aim to solve critical problems (Thompson, 1967). Members are selected based on expertise, freed from their regular tasks (Scott & Davis, 2007) and are coordinated through mutual adjustment, as dependencies in the disastrous time are highly reciprocal (Thompson, 1967). Majchrzak, Jarvenpaa, and Hollingshead (2007) call these interdependent task-forces emergent response groups. Emergent response groups are groups with no pre-existing structure concerning rules, membership, authority, or expertise. Interdependency within the group is especially high due to constant change of information and resources. Task uncertainty is at its maximum with unstable task definitions, demand for flexibility, fleeting membership, and coordination of multiple, possibly conflicting purposes. Here, the purpose of emergent response group members is seen to be dynamic, and possibly (Majchrzak, Jarvenpaa, & Hollingshead, 2007), as opposed to Thompson’s the shared goal assumption and instrumental rationality of the synthetic organisation (1967). This might indicate an argumentative dilemma between rational and natural perspective. As opposed to the latter, rational view of goal specificity and the possibility to implement structure and rules that contribute to a shared goal, the former, natural perspective rather acknowledges that the organisation is made out of people with differing interests and motivations (Scott & Davis, 2007). This dispute is relevant, as goal setting has consequences on coordination and communication.
Furthermore, in line with the open system perspective, the law of limited variety suggests that emergent response groups are highly diverse, for their environment is diverse and complex (Pondy & Mitroff, 1979). Thus, a disaster results in the emergent response group’s characteristics. This adaption is the sole response, as given the high uncertainty, they ought to tolerate learning, enable flexibility and innovation, and thus minimize routinized, rigid behaviour. Thus, unlike the value adding character of emergent response groups as suggested by Majchrzak, Jarvenpaa, & Hollingshead (2007), an open system analysis, such as contingency theory, would suggest these characteristics simply being a survival mechanism in an uncertain environment.
The Organisational Level
Contingency theory. The nature of an organization is influenced, contingent, on the environment in which it operates. Hence, this theory goes against theories offering an ideal, best way, to operate an organization. Various kinds of external contingencies can influence the organization, like e.g. political, cultural and technological developments.
Path-dependency theory. This theory is level wise in the greyzone, but explains a process through which organizations develop and reinforce historically situated solutions to various problems. It sees its inception when some critical event occurs, with all procedural options available to the organization, but then some critical juncture occurs in which a regime in the organization develop and focalize attention on a select few procedural options. This situation can be self-reinforcing, more and more options are written off, until the organization is completely locked-in on one modus operandi (Sydow, Schreyôgg, & Koch, 2009).
The Social – Psychological Level
Transactive memory systems theory. Transactive memory system (TMS) is a shared system for learning, storing, using and communicating, to accomplish individual, group, and organisational goals, and it assumes a shared conception of who knows what in the group (Majchrzak et al., 2007). Research shows that when group members know of another’s knowledge and responsibilities, information coordination is improved, task performance is more effective, and trust enhanced (Hollingshead, 2009).
This circumstance of an elaborate shared knowledge system that has formed over time, is clearly not given in times of disasters. Instead, emergent response groups are spontaneously formed with a lack of rules, procedures, authority, and stability. These conditions make coordination extremely difficult, because no TMS is in place (Majchrzak et al., 2007).
Cognitive synchrony. Due to the nature of disasters and resulting emergent response groups, often no cognitive synchrony is given, meaning that group members have no shared and accurate impression of their own and the other members’ thoughts (Hollingshead, 2009). Therefore, there is no common thinking structure, culture, or focal points that group members can rely on in those uncertain times. As a result, emergent response groups adopt alternative approaches for knowledge coordination, for example learning-by-doing and action-based problem solving (Hollingshead, 2009).
Weick’s sensemaking theory. Human beings have a natural inclination to find structure, especially when confronted with ambiguity, like when we see “figures” in clouds and stellar constellations. Weick proposed a theory to explain the underlying mechanisms of this in the social sphere of life, which he called sensemaking (Scott & Davis, 2007).
Challenges and Proposed Solutions
Challenges and Proposed Solutions
In the following the key challenges that the police faced are brought forward, structured into four main domains: (1) Organizational structure, (2) leadership, (3) information flow and ICT, (4) coordination.
Thompson (1967) defines organization structure as the internal differentiation and patterning of relationships within and between segments or departments, which form major components of an organization.
Organizational structure needs to be more fluid for synthetic organizations or emergent response groups. This can be explained by contingency theory, stating that “design decisions depend on environmental conditions” (Scott & Davis, 2007, p. 103). In case of the disaster, the environment keeps on changing with new information as well as resources coming in. Thus, the emergent response group structure also changes in relation to the environment. It is important to note that this structure will vary from the already established structure within organizations of participating groups as these groups together form a new organization.
Challenges. On 22 of July, as the chain of events unfolded from the first phone call to NBPD to the final rescue and evacuation, we saw the birth and development of an emergent response group. This emergent organization consisted of not only the various police units but also the other citizens as well as health service official as they were all contributing actors in this situation. By nature of an emergent response group, they face the challenges of limited resources, lack of structures etc. As NBPD was probably the most important member of the group as Utøya came under their jurisdiction, we will now look at some of the key structural issues NBPD faced.
In this event, the police force required to move away from its normal structure and prioritize activities based on their relevance. They also had to adapt and change continuously while trying to maximize the efficacy of their effort. Based on the examples below we can see that the Police could have done this better: -They had not given any attention to the event happening at Utøya nor had any plans on how any incidents happening there should be handled (Gjørv et al., 2012). In addition, there appears to be lack of training of the officials as an effort towards cost cutting. There were also no contingency arrangements that provided access to additional staff if required (Gjørv et al., 2012). Also, the two officers on the scene first could have done more than “beyond just observing” (Gjørv et al., 2012).
Solution. The lack of initiative by the first two officers on scene shows the negative effects that path dependence can have. In disaster scenarios a lock in based on path dependence can lead to high inefficiency as neither the normal goals nor the normal procedures are applicable to the situation. In emergent response groups most where environmental factors are changing individuals can provide a stability factor (Majchrzak et al., 2007). Police disaster training should incorporate aspects that research about emergent response groups suggests; for example, efficient utilisation of resources including volunteers. Another way of better preparing for uncertain environment is monitoring the environment and planning responses (Thompson, 1967). This could be done through keeping track of large gatherings within the district of NBPD and other police units.
Leadership in an emergent response group is a crucial role since both time and resources are limited; thus, the direction they are pointed at can have a huge impact on outcomes.
Challenges. While various police units and other groups formed a part of emergent response group on 22nd July, the NBPD should have been the natural leaders since the Utøya was under their jurisdiction. The commission felt that “the police operation lacked proper leadership at the operational level from the beginning until the perpetrator was arrested” (Gjørv et al., 2012).
Solution. In emergent response groups, leadership is also emergent, for it is unclear, fluid and dispersed (Majchrzak et al., 2007). The existence of police solves this problem to some extent, as they are expected of fulfil the leadership role in “disaster” situations. Their power is based on authority, authorized by the government (Dornbusch, Scott, Busching, & Laing, 1975). Thus they automatically form a part of dominant coalition. Furthermore, there is a connection between external conditions and the internal distribution of power (Greenwood & Hinings, 1996). Hence, for emergent response group volatile external conditions lead to changing members in the dominant coalition.
If we look at 22 July from Thompson and Tuden’s (1959) typology of preference about outcome vs belief about causation we can see that this situation would require expert judgement as there was overall agreement on goals, but means of achieving them was uncertain. The police being probably the only constant member of dominant coalition needs to be the one applying the expert judgement, if they have the required skill match with the need of the moment or authorise someone who does. In some cases of disaster scenarios the goals might be clear or align, making the task of police forces more difficult. This difficulty can to some extent be reduced by focussing leadership training, focusing on differences between normal- and disaster scenario leadership. This should include experiential learning (Hedberg, 1981; March, 1988) since the goals of the emergent response group might change based on new information inputs and use of interpersonal control systems since structural control systems will not apply outside the formal organisation.
Moreover, the cognitive structure for emergent group centres on action based scenarios including decisions, actions, knowledge, events and feedback (Majchrzak et al., 2007). Thus the leadership orientation also needs to be more action based rather than knowledge or hierarchy based unlike in normal scenarios.
Information Flow and ICT
How information is collected, processed and distributed throughout an organization is contingent on ICT and human processing abilities. It’s under great environmental changes that organizational ability to reorganize internal features, achieving best fit, is put through the ultimate test. This is why various theorists argue that information flow plays the most critical role in interconnecting the elements of a system (Scott & Davis, 2007).
Challenges. Poor information flow between the districts and underlying units is something the commission highlights; they did not have a shared platform for informational exchange. Communication between the Control Room supervisor and the Incident commander was poor, which in turn affected the Operational commander; like whether BT arrived by helicopter and confusion about the meeting point (Gjørv et al., 2012).
Mass notification/mobilization systems were not available to the police, causing them to rely on archaic methods like phone lists to acquire more resources; this was worsened by initial reluctance to even contact other police districts (Gjørv et al., 2012). A tip from a witness detailing a suspicious person driving away from the Government Quarter was received and written down on a post-it by a switchboard operator but left unnoticed on the Control Room supervisors desk for 20 minutes (Gjørv et al., 2012): Breivik passed two police cars and one police station on the way to Utøya (Hildrum, 2012).
Solution. 22 July became a live demonstration of the informational input overload hypothesis (Scott & Davis, 2007). An institutional theorist might hence argue for investing in the police´ ICT due to its crucial role on a systemic level, for outdated ICT and/or poor utilization of it is a common issue for emergency responders (Hollingshead, 2009; Santos, Borges, Canós, & Gomes, 2011). However, ICT implementation has shown to result in a productivity paradox, increasing individual but not collective productivity, when the loosely-coupled nature of the organization is not taken into account.
Effective ICT utilization depends on improved human processing ability, and Simon (Scott & Davis, 2007) would argue that a formalized and goal specific organization would lower the taxing of our bounded rationality. Hence, better plans and procedures for how to strategically group and delegate information during crises, would be beneficial as long as the overarching hierarchical structure dynamically adapt to environmental demands; it cannot be too formalized or centralized. Weick would add that the reduction of informational uncertainty (equivocality) depends on a good environment for sense-making, in which perceptions of events are the outcome of a selection and retention process suited for uncertain scenarios with high complexity and interdependence. Recent extension of TMS theory (Majchrzak et al., 2007) found that individuals are the source of stability in a crisis and that the organization should serve as an intermediating role. One of the challenges within TMS theory is; however, to figure out how to reduce communication difficulties in such dynamic organizations, an issue that might be resolved by improved systems for computer-human interaction in order to lower the energy-costs associated with sense-making (Russell, Stefik, Pirolli, & Card, 1993).
Coordination can be defined as “integrating or linking together different parts of an organization to accomplish a collective set of tasks.” (Ven, Delbecq, & Koening, 1967, p. 322). We can distinguish between two modes of coordination: programmed and feedback, the former being more formalized than the latter. Feedback becomes the dominant mode under high task uncertainty, complexity and interdependence (Thompson, 1967).
Challenges. The aforementioned great deficiency in information flow made strategic, operational and tactical coordination excessively difficult, both between the districts and underlying units. The right people were frankly not in the right places at the right time, and some who were failed to take responsibility: e.g. no one requisitioning MS Thorbjørn. First to arrive at Utøya quay mainland were NBPD patrol P-30, at 17:52; 30 minutes after Breivik started shooting. P-30 did not try to get to the island nor adequately plan for BT´s arrival, partly because of conflicting messages from Operational commander. BT´s arrival was delayed because the Control Room supervisor initially directed BT to undertake search and rescue operations at the Government Quarter, making BT momentarily unavailable in the event of a second attack (Gjørv et al., 2012, p. 104).
Solution. Thompson (Thompson, 1967) presents a Gutman-type scale of interdependence (pooled, sequential and reciprocal), and how to increase coordination in the various types: respectively through standardization, planning and mutual adjustment; with increasing coordination costs and use of horizontal communication channels. A break-down of “the system” makes one more reliant on feedback mechanisms like mutual adjustment. It is in such complicated circumstances that we see increased reliance on grass-root movements like emergent response groups (Majchrzak et al., 2007), who are more action than role oriented; e.g. the rescue operations conducted by civilian campers near Utøya.
The police are an autonomous professional organization that strives to rationalize its operations to reduce the already high degree of uncertainty. It was therefore especially vulnerable in an emergency that broke with its preconceptions, and drastically increased interdependence within and between organizations. The strict geographic departmentalization of the police forces was a problem since functional and process-based types, or fewer and bigger districts, would be better suited for resource-intensive tasks. Authority is another coordinating mechanism in which modification could benefit, e.g. better integrating Fayol´s principles of management: only one supervisor and/or manager per set of activities; unambiguous hierarchy; superiors should not have more subordinates than what is manageable; superiors time should be freed-up to deal with exceptional situations (Scott & Davis, 2007).
It is very hard to coordinate effectively in the absence of a focal point, or cognitive synchrony, but that can be improved by better TMS development (Hollingshead, 2009). Supplemental lessons can be drawn from path dependence theory, since it explains how an insufficient historical path of action was reinforced among the police forces. All available options where initially where filtered through preconceived schemas that gradually spiraled into a focused locked-in modus operandi: path breaking interventions comes about by awareness of the historical origin of the path (Sydow et al., 2009). A study (Bechky & Okhuysen, 2011) of how SWAT teams successfully prepare for uncertainty found that shared task knowledge, cross-member expertise and common work flow expectations is essential. It better enabled role shifting, and speedy reorganization of routines and work.
Researchers (Majchrzak et al., 2007) are voicing the need for new network models for disaster response, to improve the use of resources and voiding the current organizational inefficiency. Our analysis of the police operation on Utøya 22 July built on these efforts, and focuses on how disaster responders (e.g. police) can increase efficiency when acting as a part of emergent response groups. The paper presented the key challenges faced by the police force at Utøya in areas of structure, leadership, information and ICT, and coordination. These challenges can be mitigated by our proposed solutions, involving improved ICT-infrastructure, more context-specific coordination, and building awareness of how to best make use of emergent response groups, as well as leadership training customized for emergency situations.
Though this paper refers to the Utøya case, our analysis has implications for a variety of organizations that face a high degree of uncertainty and interdependency. Further research is required to shed light on potential responses by organizations and individuals in disasters to enable more efficient management of disasters that otherwise put human lives at risk.
Bechky, B. A., & Okhuysen, G. A. (2011). Expecting the unexpected? How SWAT officers and film crews handle surprises. Academy of Management Journal, 54(2), 239-261. doi: 10.5465/AMJ.2011.60263060
Dornbusch, S. M., Scott, W. R., Busching, B. C., & Laing, J. D. (1975). Evaluation and the Exercise of Authority. San Francisco: Jossey-Bass.
Gjørv, A. B., Auglend, R. L., Gerkman, S., Hjeltnes, G., Bokhari, L., Hagen, T., . . . Straume, K. (2012). NOU Norges offentlige utredninger: Rapport fra 22. juli-kommisjonen. (978-82-583-1148-2). Oslo: Aurskog AS.
Greenwood, R., & Hinings, C. R. (1996). Understanding Radical Organisational Change: Bringning Together the Old and the New Institutionalism. Academy of Management Review, 21, 1022-1054.
Hedberg, B. (1981). How Organizations Learn and Unlearn. In P. C. Nystorm & W. H. Starbuck (Eds.), Handbook of Organizational Design (Vol. 1, pp. 3-27). Oxford: Oxford University Press.
Hildrum, A. (2012, 2012). Terrordøgnet 22. juli. Retrieved 11.10.2013, 2013, from http://prosjekt.tv2.no/22-juli/tidslinjer/terrordognet-22-juli
Hollingshead, A. B. (2009). Communication, Coordinated Action, and Focal Points in Groups: From Dating Couples to Emergency Responders. In C. R. Agnew, D. E. Carlston, W. G. Graziano & J. R. Kelly (Eds.), Then A Miracle Occurs: Focusing on Behavior in Social Psychological Theory and Research (pp. 464). USA: Oxford University Press.
Majchrzak, A., Jarvenpaa, S. L., & Hollingshead, A. B. (2007). Coordinating Expertise Among Emergent Groups Responding to Disasters. Organization Science, 18(1), 147-161.
March, J. G. (1988). Decisions and Organisations (J. G. March Ed.). Oxford: Basil Blackwell.
Russell, D. M., Stefik, M. J., Pirolli, P., & Card, S. K. (1993). The Cost Structure of Sensemaking. Proceedings of InterCHI.
Santos, R., Borges, M., Canós, J., & Gomes, J. (2011). The Assessment of Information Technology Maturity in Emergency Response Organizations. Group Decision & Negotiation, 20(5), 593-613. doi: 10.1007/s10726-011-9232-z
Scott, W. R., & Davis, G. F. (2007). Organizations and Organizing: Rational, Natural, and Open System Perspectives. New Jersey: Pearson Education.
Sydow, J., Schreyôgg, G., & Koch, J. (2009). Organizational Path Dependence: Opening the Black Box. Academy of Management Review, 34(4), 689-709.
Thompson, J. D. (1967). Organizations in Action: Social Science Bases of Administrative Theory. USA: McGraw-Hill.
Thompson, J. D., & Tuden, A. (1959). Strategies, Structures, and Processes of Organisational Decision. Comparative Studies in Administration, 195-216.
Ven, A. H. V. D., Delbecq, A. L., & Koening, R. (1967). Determinants of Coordination Modes Within Organizations. American Sociological Review, 41, 322-338.