The Role of the Police in Evacuations

Abstract

Disaster evacuations are stressful events in which citizens and law enforcement frequently interact with each other. Most emergency response plans are based on military strategies that operate independent of the general public, but we argue that the police must be cognizant of several social psychological factors that affect citizens’ behavior during evacuations, including risk perception, social networks, and access to resources. Drawing from social psychological, criminal justice, and disaster research, we propose a model that (a) describes how citizens’ priorities and behaviors change as a disaster evolves and (b) identifies policing strategies that accommodate these changing behaviors and facilitate a successful evacuation. Our model, embedded in how people behave and what police are taught, can increase citizen compliance with law enforcement during disaster evacuations, remove more citizens from harm, save lives, and improve the relationship between communities and the police.

Keywords

disaster evacuation risk perception social influence styles of policing

When disasters strike, both citizens and law enforcement agents face unfamiliar, uncertain, and often frightening circumstances. During hurricanes, floods, and even riots, for example, citizens fear for their families, homes, and personal safety. These are also incidents in which the police face unusual challenges, such as coordinating with other emergency agencies or widespread power failures affecting communication lines and community well-being. During these stressful events, and in particular, with regard to evacuations, citizens and law enforcement agents often interact with each other, raising questions of communication, perception, and misunderstanding.

There are many instances in which citizens have disobeyed orders to evacuate or disperse and subsequently defied police officers’ attempts to evacuate them from their homes (e.g., Hurricane Katrina in 2005, the Iowa floods in 2008; the Vancouver riots of 2011). Hence improving the safety of citizens and officers, as well as enhancing cooperation between these groups, requires a clearer understanding of how the police can best respond to citizen behavior during disasters.

Although the police are ubiquitous in disasters, studies examining how law enforcement interacts with citizens during these critical events are largely absent in the police and disaster literatures. In particular, research has yet to fully examine how law enforcement’s response to disasters is affected by social psychological factors. Therefore, the goal of this article is to broaden our understanding of how the police should respond to citizens’ priorities and behaviors during disasters.

With this objective in mind, this article proceeds in four steps. First, we review the literature on citizen behavior in disasters and evacuations, particularly emphasizing the importance of risk perception, social networks, and access to resources. Second, we explain why disasters and evacuations require police response in addition to responses by other emergency or civil agencies. Third, building on Wilson’s (1968) styles of policing, we present a model for analyzing police response in disaster evacuations and argue that police agencies may need to change policing styles to accommodate both the expectations and behaviors of citizens and the logistical challenges of the disaster. In the final section, we conclude by discussing how future research should examine law enforcement’s response to disasters.

How Citizens Behave During Evacuations From Disasters

Researchers have long debated the definition and operationalization of the term “disaster,” categorizing disasters according to origin (natural vs. man-made) and other characteristics, such as duration of impact or amount of destruction (Quarantelli, 1984). Because law enforcement must respond to all types of disasters, we rely on a broad definition that defines disasters as events that are

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

Citizens do not typically panic when disasters occur, nor do they appear helpless. Instead, people generally behave rationally and attempt to make informed decisions about the best course of action that will protect them and their families (Helsloot & Ruitenberg, 2004; Quarantelli, 1957). Similarly, despite fears to the contrary, citizens generally do not engage in looting and other destructive or criminal behavior during or after disasters (Quarantelli, 1994; Quarantelli & Dynes, 1969). In fact, citizens frequently engage in disaster mitigation or recovery behaviors, termed “situational altruism” (Dynes, 1994), that include searching for disaster victims or distributing emergency supplies.

Although individuals generally remain law-abiding during disasters, citizens do not always fully comply with recommendations or orders from authorities. Nor do citizens always remove themselves from harm. In particular, during evacuations citizens may ignore evacuation warnings and make decisions that place them in dangerous situations, such as taking shelter in an attic when floodwaters rise. Citizens may also refuse to disperse at disaster sites. Because evacuations can remove large numbers of people from harm, this article focuses on how police can effect compliance from citizens during disaster evacuations.

Citizen Behavior During Evacuations

Evacuation refers to the temporary movement of people as a means of coping with a threat, hazard, or disruption affecting the community. There are three types of evacuations: voluntary, recommended, and mandatory (Walshon, Urbina, & Levitan, 2001). During voluntary evacuations, citizens decide whether to evacuate, but the government does not implement widespread traffic control measures. During recommended evacuations, government agencies establish basic traffic control measures because of the high probability of threat to people in at-risk areas. Mandatory evacuations are ordered when a disaster poses the greatest danger to citizens; mandatory evacuations also present law enforcement with the most diverse and numerous challenges. During mandatory evacuation orders, authorities limit access to the at-risk area(s) and implement widespread traffic control measures, for example, opening all lanes to outbound traffic.

The decision to evacuate is a stressful, momentous event for citizens. Although individual characteristics can affect the likelihood of evacuation, there are also psychological processes that likely influence whether people leave or remain in their home during disasters, including risk perception, social influence, and access to resources (Riad, Norris, & Ruback, 1999). These processes are important considerations for law enforcement agents because citizens appear to make the decision to evacuate or not largely independent of police involvement in disasters. If true, police agencies need to be able to change their style of policing to accommodate the risk perceptions, social influences, and resources of citizens considering evacuating from a disaster.

Risk Perception

Recent disaster research has shifted from analyzing individual-level predictors of evacuation decisions to examining factors related to citizens’ perceptions of the disaster’s threat. Risk perception or threat appraisal is one such important factor affecting evacuation decisions. Threat appraisals reflect how an individual evaluates stressors (i.e., a disaster or evacuation warning) and the impact on the person and people important to the individual. People evaluate risks by examining three factors related to the threat: proximity, certainty, and severity (Perry, 1979). Studies examining evacuations from hurricanes have found that high risk areas experienced higher evacuation rates than low risk areas (Dow & Cutter, 1998; Peacock, Morrow, & Gladwin, 1997). These high risk areas were closest to the projected path of the hurricane, more certain to experience damage, and more likely to suffer severe damage. Thus citizens in these high risk areas probably perceived a greater personal risk and evacuated. Notably, the process of risk evaluation de-emphasizes the type of disaster and instead highlights the perceptions of risk associated with the disaster.

Factors about disaster warnings can also affect citizens’ perceptions of risk and, in turn, their evacuation decisions. Research finds that the most effective warnings are issued frequently, from a credible and familiar source, contain specific information about the disaster (i.e., location of risk, time to impact, nature of the hazard), and advise citizens to take specific protective action (Drabek, 1999; Sorenson, 2000). Citizens who receive clear and consistent warnings about the proximity, certainty, and severity of the disaster are more likely to accurately perceive the disaster risks and make appropriate evacuation decisions. This point is highlighted in research showing that people chose not to evacuate primarily because they did not believe the disaster posed any danger to them (Perry, 1985). Therefore, although the disaster type is important for estimating the physical impact, citizens’ perceptions of the disaster’s risk are more important for evaluating the social impact and subsequent decision making, though the physical and social impacts of an event are likely correlated.

Social Influence

Social networks and social norms are two key aspects of social influence, a mechanism that strongly shapes both individuals’ behaviors and group behaviors (Aarts & Dijksterhuis, 2003; Goldstein, Cialdini, & Griskeviscius, 2008). Social networks are composed of interconnected individuals tied to each other through some type of interdependence, such as friendship, finances, or information exchanges. Norms refer to explicit or implicit rules of behavior in social interactions. Social influence becomes increasingly important when warnings about a stressful event are unclear or uninformative (Nilsson & Johansson, 2009). Similarly, research has long noted the importance of social influences (i.e., family and friends) when disasters strike (Lindell, Lu, & Prater, 2005; Quarantelli, 1957). Citizens seek information from the media about the disaster but, when faced with the decision to evacuate, citizens frequently rely on their social networks to gather additional information about the disaster, confirm their perceived risks or authorities’ warnings about the risks, and solicit others’ opinions about whether or not to evacuate (Drabek, 2000; Proulx, 1999).

Social networks also are important when citizens begin evacuating. Whether evacuating from a rapidly emerging disaster (e.g., a nightclub fire) or a slowly emerging disaster (e.g., a hurricane), individuals typically evacuate in groups, namely, with their friends, family, or coworkers (Cornwell, 2003; Johnson, 1988; Kirschenbaum, 1992). Researchers studying the 1993 World Trade Center bombing and the September 11 attacks frequently observed this phenomenon (Aguirre, Wenger, & Vigo, 1998; Murphy, 2002). People want to ensure that their friends and family members are safe and accounted for before evacuating from a disaster.

Social networks also reinforce group norms. Citizens frequently use their neighbors or other close network members as referents for their own behavior (Schultz, Nolan, Cialdini, Goldstein, & Griskevicius, 2007) even though individuals often do not recognize this influence (Nolan, Schultz, Cialdini, Goldstein, & Griskevicius, 2008). The powerful influence of group norms or behaviors has direct implications for evacuations. In their study of the Three Mile Island nuclear power plant accident, Cutter and Barnes (1982) found that among citizens who evacuated, 74% reported that their neighbors also evacuated. Among those who did not evacuate, 50% reported that their neighbors also did not evacuate. In addition, social networks provide not only information about the disaster, advice about the decision to evacuate, emotional support, and norm reinforcement, but also access to resources.

Access to Resources

Access to resources includes both physical and intangible resources. During evacuations, physical resources may include transportation, safe shelter, or the ability to borrow money from friends or family. Intangible resources may include information about the disaster, emotional support, and size of the social network. Theoretically, individuals who have smaller social networks have less social support (Kaniasty & Norris, 1995). Therefore, citizens who have limited information about the disaster, no transportation, and no place to seek shelter may decide against evacuating. Yet weak social networks might also have the opposite effect—citizens who lack a strong social network may be more likely to evacuate because there are fewer individuals to influence these citizens against evacuating.

Risk perception, social influence, and access to resources strongly affect individuals’ behavior and decision making during disasters. These factors present challenges for law enforcement agencies charged with warning, evacuating, rescuing, and calming citizens as a disaster evolves. However, by knowing about the social impact of a disaster, police departments can capitalize on these social processes, allowing officers to use risk perception or social influence to gain cooperation from citizens regarding warnings or evacuation orders.

The Role of Police in Disasters

The police are a unique government institution, set apart from other government agencies by their responsibilities, powers, and visibility. Only law enforcement agents are legislatively mandated to preserve public order and safety and legally authorized to use force to achieve these objectives.

To maintain peace and order, the police exercise their unique authority in three primary capacities: crime prevention and response, order maintenance, and public service. First, officers attempt to prevent and control crime by patrolling neighborhoods, conducting stop and frisks, responding to potentially criminal incidents (e.g., fights or disturbances), investigating reported crimes, and identifying and apprehending offenders. Second, officers engage in order maintenance by presiding over civil disputes (e.g., evictions), resolving public nuisances (e.g., noise complaints), and facilitating the movement of individuals and vehicles (e.g., parades, sporting events, protest events, etc.). Third, police officers provide public service to citizens by rendering emergency aid, taking into protective custody the mentally ill or others who cannot care for themselves, and fostering police–community relationships (e.g., participation in neighborhood watch programs). When considering these numerous and diverse responsibilities, society seems to expect that the police will not only protect but also “perform every conceivable kind of service” (Vollmer, 1969, p. 3) and “combat every social ill” (Deltratte, 1996, p. 23). Despite these societal expectations, police departments and officers do not view these responsibilities equally important.

Styles of Policing

Although all police departments engage in some level of crime prevention and response, order maintenance, and public service, law enforcement agencies do not uniformly prioritize these responsibilities. Thus across departments there may be differences in how dutifully officers prevent and respond to crime, preserve order, and provide public service. In his now classic book, Wilson (1968) described three styles of policing (legalistic, watchman, and service oriented) that are important for analyzing how the police fulfill their roles and responsibilities in both normal and disaster situations.

The legalistic style of policing is characterized by a hierarchical and centralized command structure, a formalistic role structure, meaning that rules and regulations are used to control officer discretion so that laws can be impartially enforced, and “functional differentiation,” or a division of labor within the police department. The watchman style of policing generally emphasizes order maintenance goals and, in contrast to the legalistic style, is characterized by fewer rules and policies, less specialization, and greater officer discretion. Officers may ignore minor law violations and make arrests only for flagrant or major violations, or for crimes that pose serious threats to order in a community. The service-oriented style of policing combines aspects from both the legalistic and watchman styles but emphasizes the service responsibilities of police work. Serious crimes assume precedence over minor crimes, and officers frequently practice informal, nonarrest sanctions. This style is also characterized by officer participation in community-oriented policing programs that build cooperation and collaboration between the police and the public.

Wilson’s description of styles of policing makes two important points: (a) styles of policing can vary across officers and police departments, and (b) styles of policing can vary across situations and geographic areas. Variations in policing styles are produced by different citizen expectations of law enforcement, by different police officers’ views of their roles and responsibilities, and by a city’s political culture, including its socioeconomic characteristics (Wilson, 1968). This variation means that officers and departments may not approach every situation using only one style of policing. Instead, styles of policing may be more flexible if the situation, like a disaster, demands that officers reprioritize community needs.

Although disasters often present law enforcement with novel and unpredictable circumstances, the role of the police in disasters is similar to the role of the police in normal situations. That is, although a disaster may threaten a city’s infrastructure and social structure, the police are still primarily responsible for maintaining order and peace, despite assistance from outside agencies, such as the National Guard. The critical difference, however, is how law enforcement prioritizes crime prevention, order maintenance, and public service needs.

The Role of the Police in Disasters and Disaster Evacuations

Disasters present the police with unstable, dangerous circumstances that often frighten and threaten large numbers of citizens. Yet the emergence of a disaster event does not change the societal role of the police—the government and citizens still expect the police to maintain order and peace. Besides fulfilling this legislative mandate, the police must also protect life and property, conduct search and rescue operations, control traffic and crowds, and warn and evacuate citizens (Kennedy, Brooks, & Roncek, 1969).

The police are generally the first organization to respond to a disaster (Quarantelli & Dynes, 1977). Typically, their response involves normal police activity, such as aiding citizens, relaying information to communications centers, and abating the physical effects of the disaster. In addition, the police possess the legal authority, geographic information, and the best immediate equipment to enter disaster sites to conduct reconnaissance or search and rescue operations.

The police are also needed to assist in evacuations from disasters. Among government organizations, the police are best equipped to immediately facilitate the evacuation process. For instance, police officers can go door-to-door to announce and further publicize evacuation orders, help plan exit routes, facilitate departing foot or vehicle traffic, and inspect neighborhoods and houses to assess whether citizens have complied with evacuation orders. When evacuees return, the police may perform several duties, including assisting in the inspection of dwellings for damage, distributing supplies, and alerting citizens of disaster-related dangers (e.g., swimming in flood waters, downed wires, home repair scams, etc.).

Police agencies’ styles of policing may change dramatically as law enforcement prepares for, responds to, and recovers from a disaster even though their larger social role has not changed. For example, during a normal day, police officers may write traffic tickets, patrol neighborhoods, investigate reported crimes, and attend neighborhood watch meetings. Officers operate mostly in the service style of policing during this time. But when disasters approach and then strike, police officers may alert citizens of mandatory evacuation orders, conduct search and rescue operations, or assist in the safety inspection of dwellings. During this time, officers operate in the order maintenance style of policing.

Law enforcement agents may change their style of policing in disasters to respond not only to the physical impact of the event but also the social impact, including the perceptions, beliefs, and behaviors of citizens. Put simply, how citizens respond to disasters and evacuations affects officers’ style of policing. Understanding individuals’ behavior during disaster events, therefore, is critical for law enforcement in gaining citizens’ cooperation and compliance during disaster evacuations.

Authorities can become frustrated when citizens refuse to evacuate. Police officers, in particular, may view refusals as a challenge or threat to their authority during a disaster. However, citizens may disobey the laws and orders of government agents for many reasons: They do not believe the law is just, they believe the law is capriciously or selectively enforced, they do not feel a part of the law-making or law-enforcement process, or they feel disrespected by a government agent (Tyler, 2006). Yet our review of citizens’ behavior during disasters suggests that when citizens disobey a lawful mandatory evacuation order, they generally do so for reasons unrelated to their views of government authority or regard for the law. In other words, during disaster evacuations, citizens are singularly concerned with the welfare of their families, friends, and property, not philosophical issues concerning trust in the government or the legitimacy of laws. Citizens prioritize their immediate need for security and order maintenance over less central concerns, such as community policing initiatives.

If true, law enforcement should focus on adapting to the behaviors and priorities of citizens during disasters, just as the police adapt to the priorities of citizens during nondisaster situations. Therefore, a model is needed to describe (a) how citizens’ priorities and behaviors change as a disaster evolves, and (b) how law enforcement agencies can employ different styles of policing to accommodate citizen needs in every stage of disaster.

Analyzing Police Response in Disasters: A Priority-Based Model

The priorities of the public change as a disaster progresses. Generally, citizens first seek information about the event, then desire emergency response to the event’s physical impact, and finally, need reassurance that their community is returning to normal. Thus a law enforcement agency’s policing styles and strategies must also change, and its effectiveness depends on its ability to successfully transition from one policing style to another. Table 1 presents a model for analyzing the role of the police during disasters and disaster evacuations. This model first describes the stages of a disaster, specifically highlighting evacuations, and then discusses how policing styles might need to change to accommodate the priorities and needs of citizens as well as the physical impact of the disaster.

Table 1.

The Role of the Police in Disasters

Stages of a disaster
	Alarm stage (approach of disaster)	Acute stage (onset of disaster/immediate aftermath)	Recovery stage (recovery from disaster)
Citizen behaviors	Citizens are likely to:	Citizens are likely to:	Citizens are likely to:
	–Seek information about disaster’s approach	–Seek information about disaster’s onset	–Seek information about disaster’s damage
	–Listen to TV/radio for PSAs	–Listen to TV/radio for PSAs	–Listen to TV/radio for PSAs
	–Gather/consult with social networks	–Fear looters/property destruction	–Consult with social networks
	–Assess disaster risks	–Evacuate at last moment	–Return from evacuation
	–Assess available resources		–Return to homes
	–Evaluate evacuation decisions		–Fear looters/Fear rise in crime
	–Evacuate or not		–Require disaster relief supplies
Citizen priorities	Citizens desire:	Citizens desire:	Citizens desire:
	–Information	–Protection of	–Information about
	about disaster	property	disaster’s damage
	–Information about warnings and protective action	–Search/rescue operations	–Safe, secure, rapid reentry into city
	–Safe, secure, rapid evacuation route/transportation		–Safe, secure, rapid reentry into neighborhoods/homes
			–Disaster relief supplies
			–Protection of property
			–Return to normalcy
Effective policing style	Legalistic	Watchman	Service
	Emphasis: Hierarchical/centralized command; division of labor	Emphasis: Decentralization of command, order maintenance	Emphasis: Service to citizens, informal sanctions, police–community relations
	Advantages: Cohesive, well-organized command structure; similar to ICS; less officer discretion	Advantages: Greater officer discretion, greater flexibility during onset of disaster	Advantages: Alleviate citizen fears, facilitate return to normalcy, rebuild community ties

Stages of a Disaster

Citizen response during disasters can be divided into three stages (Helsloot & Ruitenberg, 2004): the alarm stage, the acute stage, and the recovery or rebuilding stage. Within each of these stages, disasters present different challenges to citizens’ physical and emotional welfare. Table 1 summarizes each stage in a disaster, likely citizen behaviors, expected citizen priorities, and the policing style that best adapts to both the disaster and citizens priorities.

Alarm Stage—Legalistic Style of Policing

During the alarm stage, the disaster initially threatens a community, and individuals are confronted by numerous challenges, including perceiving and processing the disaster and its potential risks, understanding authorities’ warnings, and formulating protective strategies. After learning about the approaching disaster and evacuation warnings, citizens begin gathering information about the potential risks through the media, friends, neighbors, and family members. Consulting with others not only helps individuals assess the disaster risks but also helps individuals assess the available resources needed to evacuate.

It is important to note that the length of the alarm stage may range from a few minutes to a few days. During a spreading toxic cloud following a factory explosion, citizens may have little time to make evacuation decisions. However, prior to a hurricane, citizens have many hours to seek information and make decisions.

Regardless of the length of the alarm stage, the priorities of citizens change as a disaster approaches. Prior to the disaster’s approach, community members may want the police to respond to rising crime rates, traffic issues (e.g., speeding in neighborhoods), or public nuisance issues. However, when citizens learn about an approaching disaster, their priorities narrow considerably. During the acute stage, these priorities include gathering and protecting important social network members, collecting information about how to best achieve this goal and, also, evacuating safely and rapidly away from the disaster.

Law enforcement agencies need to adopt a different policing style to meet these new priorities, but more importantly, to successfully facilitate an evacuation. We believe the legalistic style of policing makes law enforcement agencies the most effective during disaster evacuations, mainly because this policing style emphasizes a hierarchical, centralized command. Other scholars have described the characteristics of this highly bureaucratic management style (Perrow, 1986), and several aspects benefit the police in both the alarm stage and disaster evacuations.

First, a hierarchical, authoritarian management style is able to quickly prioritize, divide, and manage several tasks simultaneously. For example, a centralized command can easily dispatch units of officers to knock on doors and alert residents, conduct traffic control, or conduct reconnaissance of the disaster. Second, the chain of command also ensures that each line-level worker (and supervisor) takes orders from and reports to only one higher ranking official. These specific links are important because during Hurricane Katrina, for example, line-level workers sometimes did not know who was in charge and giving orders (Brinkley, 2006).

Third, a hierarchical command structure is better able to ensure a uniform, consistent response by decreasing officer discretion through explicit written regulations and disaster response SOPs. Moreover, a central command provides officers with identical information to distribute to the public, thereby ensuring a clear, consistent public message about the disaster and appropriate protective actions.

Although police departments engage in the legalistic style of policing during the alarm stages, officers are unlikely to use sanctions (i.e., arrests) to impartially enforce laws. Arresting citizens consumes time and resources that are better spent on other disaster priorities, namely, notifying citizens of mandatory evacuations, facilitating traffic flows, and logistically preparing for the disaster’s onsets. Officers may also be unlikely to use sanctions during the alarm stage because citizens are preoccupied with protecting themselves, their families, and their property and, thus, are committing fewer arrestable offenses. In general, citizens are primarily recovering from the disaster and not committing crimes.

While the legalistic style of policing offers many advantages, law enforcement agencies should consider changing policing styles as a disaster evolves. Wilson (1989) observed that bureaucratic agencies are often rigid, inflexible, unable to adapt to external changes, and may suffer communication breakdowns due to situational factors. In other words, the benefits of the legalistic style of policing in the alarm stage become liabilities during the onset and recovery from a disaster.

Acute Stage—Watchman Style of Policing

The acute stage is characterized by the arrival and immediate aftermath of the disaster. For example, a hurricane may make landfall, a terrorist bomb may detonate, or a toxic cloud from a train derailment may spread through a neighborhood. As with the alarm stage, the length of an acute stage may vary substantially from many minutes to many hours. Often, the police and other emergency responders begin initiating disaster management operations only during the alarm stage (Helsloot & Ruitenberg, 2004).

During this stage, the disaster’s risks are most pronounced and visible to individuals. Citizens will likely still seek information about the disaster and its impact from authorities as well as gather and seek information from important social network members. In addition, individuals who did not evacuate prior to the disaster may attempt to flee just prior to the disaster or during the disaster’s onset. During floods, citizens may refuse to initially evacuate for an expected flood, but when the water rises and begins to flood their homes, they may decide to leave.

When the immediate disaster event subsides (e.g., the hurricane passes through, the terrorist bomb destroys a building, or the toxic cloud drifts into the atmosphere), individuals are likely to have priorities different from those in the alarm stage. Citizens may need police (and other emergency responders) to conduct search and rescue operations for victims and render aid to the injured. They may also be concerned about looters (Perry, 1985).

The aftermath of a disaster will likely affect law enforcement logistically. A disaster may destroy power stations, flood police departments and other emergency management buildings, isolate officers or units, damage equipment, and threaten communications networks. More importantly, a centralized command structure may falter when confronted by these challenges. Emergency commanders may not be able to communicate with line-level units, officers may no longer possess functioning equipment, and differently affected parts of the community may warrant different police response. These factors suggest that a different policing approach is necessary during this stage.

The watchman style of policing emphasizes a decentralized command structure, greater officer discretion, and order maintenance. Each of these aspects benefits law enforcement during the acute stage of a disaster. First, a decentralized command structure gives supervisors discretion to assess the situation at their locations and then implement solutions tailored to specific problems. Some supervisors may need to immediately devote resources to search and rescue while other supervisors may need officers to first reconnoiter a neighborhood. Also, a decentralized command structure may be less affected by communications failures because requests, information, and orders do not have to travel a lengthy chain of command. Put simply, the watchman style of policing capitalizes on the flexibility eschewed by the legalistic style of policing.

Second, the watchman style of policing emphasizes order maintenance over the strict enforcement of every law. Citizens are most concerned with their immediate safety, protection of property, and stabilizing the disaster’s effects. By attending to these primary needs, officers can alleviate citizens’ fears of looting and unrest, thereby demonstrating that the government is still maintaining order in the community. Officers may arrest citizens who pose serious threats to life or property, though detaining and housing prisoners adds to the logistical challenges during a disaster’s onset. Officers make better use of their time and resources by simply maintaining order until additional assistance arrives (i.e., the National Guard). After receiving assistance from outside agencies, the police can begin to respond to more specific problems rather than simply focusing on preserving order.

We have argued that the watchman style of policing is most appropriate for the acute stage of the disaster. However, just as the legalistic style of policing struggles to meet the demands of a disaster during the acute stage, the watchman style of policing likewise falters during the recovery stage.

Recovery Stage—Service Style of Policing

In severe disasters, the recovery stage may prove to be the most complex and challenging phase for citizens and police officers alike, in part because this stage may last several weeks or even years. During this stage, evacuated citizens return to the city, their homes, and social networks. Frequently, citizens’ property and homes are damaged or even destroyed. Victims require disaster relief supplies, such as shelter, food, water, clothes, and other necessities. Faced with a wide variety and number of needs, it is not surprising that citizens’ priorities for law enforcement are no less diverse and numerous.

Consistent with the earlier two stages, citizens desire information about the disaster and also consult with their social networks. Evacuated citizens also require safe and rapid passage back into their communities. In addition, evacuated citizens generally want to return home immediately to assess the damage to their property. However, they may be delayed from entering their residences because officials frequently conduct safety inspections on homes and buildings before allowing reentry.

Emergency supplies and access to resources in general are high priorities for citizens, especially citizens with children or elderly family members. Personal protection is also a priority because citizens fear widespread disorder and violent crime (Tierney, Bevc, & Kuligowski, 2006), even if these fears are not always justified (Varano, Schafer, Cancino, Decker, & Greene, 2010). Equally important, most citizens desire a return to normalcy after the stress of a disaster evacuation. In sum, during the recovery stage, citizens’ priorities include safe and rapid reentry, emergency supplies, protection from and reassurance about looting and other crimes, and a return to normalcy.

The service style of policing is best suited for this stage because of its emphasis on police–community relations and establishing community ties. Pursing these goals allows officers to contact, interact with, distribute emergency supplies to, and in general serve individuals or specific neighborhoods. Although the police are still engaging in crime prevention and response, the primary goal is to reestablish strong relationships with community members, similar to what police do during nondisaster situations. Emphasizing community relations and personal contact between officers and civilians addresses several citizen priorities.

First, citizens may be frustrated because of delays returning to their communities and homes. When police officers assist in distributing supplies and personally contact individuals, citizens may feel that authorities are hearing their concerns and addressing problems as efficiently as possible. Second, personal contact by officers through informal neighborhood meetings may alleviate fears of looting or crime as well as reassure the public that the police are capably maintaining order and peace. Third, disasters can be chaotic, unstable, and stressful events for citizens. Interacting with the most visible sign of government service and authority helps restore a sense of stability and initiate a return to normalcy for many citizens.

Case Studies

Our model suggests that successful police response to natural disasters, anticipated terrorist attacks at large scale events, or potential riots depends on adopting and transitioning from one style of policing to another. The following three examples illustrate how law enforcement agencies evaluated, or failed to evaluate, both the disaster’s physical and social impacts and the changing priorities of citizens and their communities.

In 2007, Hurricane Katrina decimated New Orleans, killing hundreds and leaving tens of thousands homeless. Although many were at fault, the New Orleans Police Department (NOPD) failed to effectively respond to numerous challenges before, during, and after the storm struck (Deflem & Sutphin, 2009). In fact, one key lesson learned from Hurricane Katrina is that the police must be able to adapt to evolving changes in duties and operations during a disaster, including warning citizens, rescuing citizens, and controlling crime (Rojek & Smith, 2007).

According to our model, during the alarm stage of a disaster, law enforcement should employ the legalistic style of policing. However, the NOPD operated instead in a more decentralized, watchman style of policing. As a result, when citizens began seeking information about Katrina, they did not receive a clear, consistent message about what protective actions to take or where to evacuate, and thus, officers could not facilitate a successful evacuation.

Though the alarm stage of the disaster may have been characterized by the watchman style of policing, the acute stage witnessed a move toward the legalistic style of policing by the NOPD, a desire for a strong chain of command and specific directives for responding to the disaster. The destruction wreaked by the hurricane made this strategy an erroneous and impossible option. When the hurricane made landfall during the acute stage, New Orleans police officers suffered breakdowns in communication and lack of transportation (Faulk, 2007; Roman, Irazola, & Osborne, 2007). The NOPD received hundreds of emergency calls immediately following Katrina’s arrival, but the hurricane had already severely damaged law enforcement’s infrastructure (Connolly, 2005). In sum, NOPD’s disaster response failed because their plan did not match different policing styles with the disaster’s impacts and the priorities of citizens.

The flooding of Cedar Rapids, Iowa provides a compelling contrast to NOPD’s disaster response. In 2008, Cedar Rapids suffered flooding so severe the event was dubbed “Iowa’s Katrina” by the media (Minneapolis-St. Paul Star Tribune, 2008, June 16). During the alarm stage, the Cedar Rapids Police Department (CRPD) operated in the legalistic style of policing. There was a strong command structure, a uniform response to the emerging flood threat, little discretion when conducting evacuations, and the prioritization of numerous tasks simultaneously. When mandatory evacuations were ordered, law enforcement notified and assisted residents in evacuating (Cedar Rapids Gazette, 2008b). Officers also ordered citizens to stay out of evacuated neighborhoods and advised that violators would be issued trespassing and curfew citations (Cedar Rapids Gazette, 2008a).

During the acute stage, the flood drastically strained response by law enforcement. CRPD was flooded, much of its equipment was ruined, and 911 calls were rerouted to other agencies. Confronted with these challenges, the department transitioned to a watchman style of policing. Decentralized groups of officers teamed up with firefighters to patrol the city in boats and respond to citizen requests for aid (Cedar Rapids Gazette, 2008c). Instead of relying on a hierarchical command structure, officers were given greater discretion when handling citizens’ calls for service. Overall, CRPD’s response adopted different policing styles as the disaster progressed and was therefore able to accommodate the impact of the flood as well as the changing priorities of citizens.

Finally, our model can also be applied to potential disasters, such as anticipated riots at protests or potential terrorist attacks at large events. In 2002, Salt Lake City hosted the Olympic Winter Games, a prime target for terrorist attacks, as well as a subsequent disaster or evacuation. A review of security at the Salt Lake City games (Decker et al., 2005) revealed several findings that support our model’s basic tenets. Researchers suggested that centralization of security and a clear command structure were critical in preparing for the event. Second, they found that unambiguous directions were crucial for public cooperation and collaboration. Third, researchers recommended viewing the event as an “integrated theater” rather than several loosely connected locations. These features closely resemble the key features of the legalistic style of policing in the alarm stage of a disaster or, in this case, the preparation for a large-scale event. Although Olympic security did not have to respond to a disaster (i.e., a terroristic attack), the Salt Lake City experience demonstrates not only the importance of uniform, hierarchical planning and communication but also understanding when and how to transition to a different response style that addresses the challenges of a disaster and evolving citizen priorities.

Discussion: Studying Police Response in Disasters and Evacuations

We have argued that during disasters, law enforcement must adopt and transition from one style of policing to another to effectively respond to the disaster’s physical and social impacts, including the changing priorities of citizens and their communities. During the alarm stage of a disaster, law enforcement can best meet the challenges of the disaster by employing the legalistic style of policing. During the acute stage, the community suffers the actual impact of the disaster, and law enforcement is most effective operating in the watchman style of policing. Finally, law enforcement engages in the service style of policing during the recovery stage of a disaster, a period in which police–citizen interactions may alleviate fears of looting/crime and help the community return to normal.

This model has important implications for theory, research, and policy. Social psychological theories concerning risk perception, social influence, and access to resources are glaringly absent in the police and disaster literatures, and our framework presents a template for examining interactions between citizens and the police in disaster evacuations. This model demonstrates that the social impacts of a disaster are as important to police departments as the physical impacts. Hence, police departments should view disasters as social events, not simply physically destructive incidents. Law enforcement agents must be prepared to manage several critical social processes inherent in any disaster, including individuals’ perceptions of the event and authorities’ warnings, evacuation decisions, and citizens’ postdisaster anxieties.

The scope of this model is not limited to weather-related disasters but also pertains to other exceptional emergencies, such as blackouts, wildfires, and earthquakes. While the length of each disaster stage may vary according to the event, the priorities of citizens will undoubtedly change as a disaster progresses, and therefore, police response must also change. Moreover, this model also pertains to large-scale events that may become mass emergencies due to terrorist attacks or fires, such as sporting events or protests. Although our case study of Salt Lake City reveals that planners devote most attention to the alarm stage, or preparing for the event, equal attention must be paid to how emergency responders, in particular, law enforcement, will adapt if a disaster strikes.

The social psychological aspect of this framework offers important suggestions for police response operations and police–community relationships. During evacuations, if the police personalize evacuation orders by going door-to-door to contact citizens, then more citizens should evacuate. Likewise, if police contact neighborhood groups or leaders to advise that most of their neighbors are evacuating, then more citizens should evacuate. Disasters also represent important opportunities to strengthen police–community relationships. Citizens normally do not interact with the police, and the majority of citizen contacts are negative encounters, such as traffic stops or arrests (Langan, Greenfield, Smith, Durose, & Levine, 2001). However, if law enforcement agencies are able to evacuate more citizens and to respond more effectively to citizens’ changing priorities during a disaster, then police departments may be able to enhance their public image and foster community relations.

There are also important policy implications for police disaster training and police actions during disasters. These two areas are in particular need of more research. For example, how well are police officers trained to respond to disaster events? More specifically, how well do law enforcement agencies prepare their officers to navigate the three stages of a disaster, respond to changing citizen priorities, and transition from one policing style to another? Currently, U.S. police officers complete National Incident Management System training (NIC, 2008), sponsored by the Federal Emergency Management Agency. However, research has not yet characterized the nature of disaster training for police officers. In general, there is little information about law enforcement agencies’ plans about, capacities for, and effectiveness in responding to disasters (National Research Council, 2004, p. 75).

Similarly, there is little research on what police actions are most effective in each stage of a disaster. Table 1 suggests that some police actions are more effective than others during the stages of a disaster. During the alarm stage, reinforcement of public service announcements is an important police function. Research indicates that citizens seek information about disasters from credible sources, and police may consider assuming a primary role in disseminating warnings, especially in neighborhoods severely threatened by the disaster. In contrast, police sanctions are not likely to be effective. Arresting citizens consumes police time and resources that are better used addressing more important disaster priorities. Furthermore, because citizens are busy bracing for the disaster and generally not engaging in criminal behavior, arrests are less needed. Sanctions are most likely to be time and cost-effective when officers arrest suspects for looting or other crimes feared by citizens during disasters. Arresting people for these types of crimes reassures the public and consumes officer resources only on crimes meriting arrest, as opposed to arresting citizens for minor infractions like curfew violations.

During the recovery stage, the police serve an important role of facilitating community communication and cohesion. The most effective police actions during this stage should allow officers to frequently interact with citizens and demonstrate that the police institution is an important hub for community reconstruction. The police also have more time for service-oriented activities once the disaster has passed and other agencies are assisting with disaster mitigation and recovery. Conversely, the legalistic style of policing, or any approach that lacks a personalized, community-based policing perspective, is less likely to succeed in the recovery stage. Following a disaster, citizens desire a return to normalcy and a strong connection to public service institutions. Isolating the police institution from the community may also engender public resentment toward government agencies, a sense that the government does not truly care about its citizens. In turn, citizens may be less likely to cooperate and comply with law enforcement or other government officials.

In sum, the police need to view disasters as evolving social events that demand different policing styles and strategies as the incident progresses. If citizens’ needs change as the disaster approaches and subsides, then law enforcement training and policies must adapt to these changing priorities. The framework presented offers police practitioners and researchers alike an important model for thinking about the role and efficacy of law enforcement in disasters. It provides police with specific details about how citizens behave during disasters and suggests how police can more effectively respond during disaster evacuations.

Footnotes

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

The authors received no financial support for the research, authorship, and/or publication of this article.

Bios

Luke Bonkiewicz is a police officer at the Lincoln Police Department in Lincoln, Nebraska, USA. He has previously published research on drug market participation, public perceptions of the police, and law enforcement response in disasters.

R. Barry Ruback is a professor of Crime, Law & Justice and Sociology at The Pennsylvania State University. He has previously published research on decision making in the legal system, causes and consequences of victimization, and environmental psychology.

References

Aarts

Dijksterhuis

(2003). The silence of the library: Environmental, situational norms, and social behavior. Journal of Personality and Social Psychology, 84, 18-28.

Aguirre

B. E.

Wenger

Vigo

(1998). A test of the emergent norm theory of collective behavior. Sociological Forum, 13, 301-320.

Associated Press. (2008, June 18). Teams inspect Cedar Rapids homes to see if they’re safe. Dubuque Telegraph Herald, pp. C6.

Brinkley

(2006). The great deluge: Hurricane Katrina, New Orleans, and the Mississippi Gulf Coast. New York, NY: Harper.

Cedar Rapids Gazette. (2008a, June 12). Flooding Notes. Cedar Rapids Gazette, 11A.

Cedar Rapids Gazette. (2008b, June 13). Flooding Roundup. Cedar Rapids Gazette, 2A.

Cedar Rapids Gazette. (2008c, June 14). Cedar Rapids Struggles to Grasp Losses. Cedar Rapids Gazette, 1A.

Connolly

(2005, September 28). Thousands of Katrina 911 calls went astray. The Washington Post, p. A3.

Cornwell

(2003). Bonded fatalities: Relational and ecological dimensions of a firen evacuation. Sociological Quarterly, 44, 617-638.

10.

Cutter

S. L.

Barnes

(1982). Evacuation behavior and Three Mile Island. Disasters, 2, 116-124.

11.

Decker

S. H

Green

J. R.

Webb

Rojek

McDevitt

Bynum

. . . Manning

P. K.

(2005). Safety and security at special events: The case of the Salt Lake City Olympic Games. Security Journal, 18(4), 65-74.

12.

Deflem

Sutphin

(2009). Policing Katrina: Managing law enforcement in New Orleans. Policing: A Journal of Policy and Practice, 3(1), 41-49.

13.

Deltratte

E. J.

(1996). Character and cops: Ethics in policing (3rd ed.). Washington, DC: AEI Press.

14.

Drabek

T. E.

(1999). Understanding disaster warning responses. Social Science Journal, 36, 515-523.

15.

Drabek

T. E.

(2000). Pattern differences in disaster-induced employee evacuations. International Journal of Mass Emergencies and Disasters, 18, 289-315.

16.

Dow

Cutter

S. L.

(1998). Crying wolf: Repeat responses to hurricane evacuation orders. Coastal Management, 26, 237-252.

17.

Dynes

R. R.

(1994). Situational altruism: Toward an explanation of pathologies in disaster assistance (Report No. 201). Newark, DE: University of Delaware Disaster Research.

18.

Faulk

(2007). New Orleans gets a “do-over”: Storm devastation can’t kill the creative spirit of the New Orleans police department. Law Enforcement Technology, 34(10), 78-83.

19.

Fritz

C. E.

(1961). Disasters. In Merton

R.K.

Nisbet

R.A.

(Eds.), Contemporary social problems (pp.135-169). New York, NY: Harcourt.

20.

Goldstein

N. J.

Cialdini

R. B.

Griskeviscius

(2008). A room with a viewpoint: Using normative appeals to motivate environmental conservation in a hotel setting. Journal of Consumer Research, 35, 472-482.

21.

Helsloot

Ruitenberg

(2004). Citizen response to disasters: A survey of literature and some practical implications. Journal of Contingencies and Crisis Management, 12(3), 98-111.

22.

Johnson

N. R.

(1988). Fire in a crowded theater: A descriptive investigation of the emergencies of panic. International Journal of Mass Emergencies and Disasters, 6, 7-26.

23.

Kaniasty

Norris

F. H.

(1995). In search of altruistic community: Patterns of social support mobilization following Hurricane Hugo. American Journal of Community Psychology, 23, 447-477.

24.

Kennedy

W. C.

Brooks

T. M.

Roncek

D. W.

(1969). The police department in natural disaster operations (Report Series 6). Newark, DE: University of Delaware Disaster Research Center.

25.

Kirschenbaum

(1992). Warning and evacuation during a mass disaster: A multivariate decision making model. International Journal of Mass Emergencies and Disasters, 10, 91-114.

26.

Langan

P. A.

Greenfield

L. A.

Smith

S. K.

Durose

M. R.

Levine

D. J.

(2001). Contacts between police and the public: Findings from a 1999 national survey. Washington, DC: U.S. Department of Justice.

27.

Lindell

Prater

(2005). Household decision-making and evacuation in response to Hurricane Lili. Natural Hazards Review, 6, 171-179.

28.

Minneapolis-St. Paul Star-Tribune. (2008, June 16). Iowa’s Katrina? Not if State Does its Charity Helps Now, but Eventually State must Pay. Retrieved from http://www.startribune.com/opinion/editorials/19989179.html? location_refer=Editorials:highlightModules:1

29.

Murphy

D. E.

(2002). September 11: An oral history. New York, NY: Doubleday.

30.

National Integration Center (NIC). (2008). National Incident Management System (NIMS): Five year NIMS training plan. Retrieved from http://www.fema.gov/emergency/nims/NIMSTrainingCourses.shtm#item5

31.

National Research Council. (2004). Fairness and effectiveness in policing: The evidence. In Skogan

Frydyl

(Eds.), Committee to review research on police policy and practices (pp 75-76) Washington, DC: The National Academic Press.

32.

Nilsson

Johansson

(2009). Social influence during the initial phase of a fire evacuation Analysis of evacuation experiments in a cinema theatre. Fire Safety, 44(1), 71-79.

33.

Nolan

J. M.

Schultz

P. W.

Cialdini

Goldstein

Griskevicius

(2008). Normative social influence is underdetected. Personality and Social Psychology Bulletin, 34, 913 923.

34.

Peacock

W. G.

Morrow

B. H.

Gladwin

(1997). Hurricane Andrew: Ethnicity, gender and the sociology of disasters. London, UK: Routledge.

35.

Perrow

(1986). Economic theories of organization. Theory and Society, 15, 11-45.

36.

Perry

R. W.

(1979). Evacuation decision-making in natural disasters. Mass Emergencies, 4, 25-38.

37.

Perry

R. W.

(1985). Comprehensive emergency management: Evacuating threatened populations. Greenwich, CT: Jan Press.

38.

Proulx

(1999). Occupant response during a residential highrise fire. Fire and Materials, 23, 317-323.

39.

Quarantelli

E. L.

(1957). The behavior of panic participants. Sociology and Social Research, 41, 187-194.

40.

Quarantelli

E. L.

(1984). Perceptions and reactions to emergency warnings of sudden hazards. Ekistics, 309, 511-515.

41.

Quarantelli

E. L.

(1994). Looting and antisocial behaviors (Report No. 205). Newark, DE: University of Delaware Disaster Research.

42.

Quarantelli

E. L.

Dynes

R. R.

(1969). Dissensus and consensus in community emergencies: Patterns of looting and property norms. Il Politico, 34, 276-291.

43.

Quarantelli

E. L.

Dynes

R. R.

(1977). Response to social crisis and disaster. Annual Review of Sociology, 3, 23-49.

44.

Riad

J. K.

Norris

F. N.

Ruback

R. B.

(1999). Predicting evacuation in two major disasters: Risk perception, social influence, and access to resources. Journal of Applied Social Psychology, 29, 918-934.

45.

Rojek

Smith

M. R.

(2007). Law enforcement lessons learned from Hurricane Katrina. Review of Policy Research, 24, 589-608.

46.

Roman

C. G.

Irazola

Osborne

J. W. L.

(2007). After Katrina: Washed away? Justice in New Orleans. Retrieved from http://www.urban.org/uploadedpdf/411530_washed_away.pdf

47.

Schultz

P. W.

Nolan

Cialdini

Goldstein

Griskevicius

(2007). The constructive, destructive, and reconstructive power of social norms. Psychological Science, 18, 429-434.

48.

Sorenson

J. H.

(2000). Hazard warning systems: Review of 20 years of progress. Natural Health Review, 1, 119-125.

49.

Tierney

Bevc

Kuligowski

(2006). Metaphors matter: Disaster myths, media frames,and their consequences in Hurricane Katrina. ANNALS of the American Academy of Political and Social Science, 604(1), 57-81.

50.

Tyler

(2006). Why people obey the law. Princeton, NJ: Princeton University Press.

51.

Varano

Schafer

J. A.

Cancinco

J. M.

Decker

S. H.

Greene

J. R.

(2010). A tale of three cities: Crime and displacement after Hurricane Katrina. Journal of Criminal Justice, 38, 42-50.

52.

Vollmer

(1969). The police and modern society. College Park, MD: McGrath.

53.

Walshon

Urbina

Levitan

(2001). National review of hurricane evacuation plans and policies. Baton Rouge, LA: Louisiana State University Hurricane Center.

54.

Wilson

J. Q.

(1968). Varieties of police behavior. Cambridge, MA: Harvard University Press.

55.

Wilson

J. Q.

(1989). Bureaucracy. New York, NY: Basic Books.