Abstract
Pulling from in-depth interviews with school administrators, counselors, security and police officers, and teachers directly involved in thwarting rampage attacks at 11 Northeastern schools, this study considers the extent to which students have broken through a “code of silence,” discouraging them from informing on their peers. While findings support prior research indicating the vital preventative role of students’ coming forward with information about threats, close scrutiny of averted incidents reveals that scholars and educational practitioners have overestimated the extent to which the student code of silence has diminished post-Columbine. Even in these successfully averted incidents, numerous students exposed to threats still did not come forward; those who did were rarely close associates or confidants of accused students; and some who did ultimately come forward did so as a result of being personally threatened or in order to deflect blame away from themselves, rather than out of altruistic concern for others.
Over the last several decades in the United States, school rampage shootings have taken multiple lives and, as a result, garnered widespread attention among academics and the population at large. These events were highly publicized, as they shocked the American public not only for their brutality, but because of the prior belief that such schools were “safe havens, free of the dangers of street crime” (Lawrence, 2007, p. 147). That such violence could be perpetrated in middle- and upper-class school districts away from the plight of impoverished urban areas was seen as especially perplexing (Kimmel & Mahler, 2003). Perhaps unsurprisingly, a significant amount of empirical research has since been conducted about the phenomenon. During this same period, however, there have also been dozens of averted incidents where student plots to kill multiple peers and faculty members came to the attention of authorities and thus were thwarted. Despite the relative breadth of research on school rampages, few studies have addressed attacks that were planned yet did not come to fruition. Whereas incidents of school violence that have resulted in multiple fatalities and injuries are often extensively investigated by scholars, the government, and the media alike, far less information exists about “near misses” (Verlinden, Hersen, & Thomas, 2000, p. 28). Pollack, Modzeleski, and Rooney (2008, p. 9) noted that: few [schools] track threats made against other students or the school (especially if the event did not result in official law enforcement intervention). The result of this failure to collect and maintain records regarding threats is that very little is known about the extent or nature of the problem.
Though understudied, research on averted rampage has clear practical significance. By learning from the instances in which school rampage threats came to the attention of authorities and were thwarted, there exists potential for future interventions and policies to be modeled on prior successes. In addition, empirical knowledge about how school rampage incidents have been averted is particularly important because many of the most popular and widespread school disciplinary policies and security practices implemented in recent years cannot be reliably linked to preventing rampage but may actually hinder the few preventative measures with an empirical basis.
The findings to follow indicate that it was people coming forward with knowledge about a potential school rampage that has preempted these nearly fatal occurrences. However, even in many of these successfully averted incidents, numerous student bystanders exposed to threats still did not come forward; those who did were often not close associates or confidants of the accused students; and some of the people who did ultimately come forward did so as a result of being personally threatened or in order to deflect blame away from themselves rather than out of altruistic concern for others. While numerous scholars and many of the school and police officials interviewed in this study acknowledge the important role that encouraging positive student bystander behavior plays in averting school rampage, few recognize how seriously ingrained the code of silence is among students. Hence, scholars and practitioners may be overstating the degree to which students, after the widespread reporting of the Columbine attack, now take their peers’ threats seriously and report them to the authorities. Moreover, this article suggests that the lack of positive bystander behavior on the part of students likely reflects increasingly punitive and criminalized school environments that erode trusting relationships between students and school staff members.
Completed and Averted School Rampage
While the term school shooting has been defined and operationalized in a wide variety of ways, Newman, Fox, Roth, Mehta, and Harding (2004, p. 50) can be credited with delineating the term rampage school shootings, which “take place on a school-related public stage before an audience; involve multiple victims, some of whom are shot simply for their symbolic significance or at random; and involve one or more shooters who are students or former students of the school.” Other academic definitions have variously limited school rampages by the type of weaponry, by the offender’s connection to the school as former students, current students, staff members, or strangers, by the relationship between offenders and victims as specific or random targets, and by the number of victims killed or injured (Daniels et al., 2007, 2010; Levin & Madfis, 2009; Moore, Petrie, Braga, & McLaughlin, 2003; Muschert, 2007; Newman & Fox, 2009; Newman et al., 2004; Vossekuil, Fein, Reddy, Borum, & Modzeleski, 2002). In order to preserve as general a concept of rampage as possible, this study defines averted school rampage plots as those cases of any form of fatal violence planned by one or more students targeting multiple specific or nonspecific classmates and/or faculty at their former or current school that was preempted from occurring and causing any death or injury and of which there is any evidence of intent. Thus, the notion of an averted rampage leaves room for a wide range of severity between various cases wherein students’ genuine desire to complete a rampage attack is perhaps problematically assumed (see Madfis, 2012, for further elaboration on differentiating the concepts of threat, risk, plot, and genuine aversion). For example, scrawling a vague threat referencing Columbine on a school bathroom wall and stockpiling an arsenal of weaponry are both actions that could be classified as evidence of an averted rampage, though the actual level of threat present in these two scenarios varies a great deal (Reddy et al., 2001).
O’Toole (2000) first noted the importance of studying averted school rampages, and several scholars (Larkin, 2009; Newman et al., 2004) have since compiled incidents of averted rampages but only tangentially discussed them. The studies that have centrally focused upon averted rampage attacks have been conducted by Daniels and his coauthors (2007, 2010).
Daniels and his colleagues (2007) completed a content analysis of 30 school rampages that were thwarted in 21 states between 2001 and 2004. From newspaper accounts, the authors reported data on the details of the plot, how the plot was discovered, what steps were taken by the school and law enforcement once the plot was revealed, and what the legal outcomes of the incidents were. They discovered that the majority of violent schemes occurred in public high schools, though several took place at elementary and middle schools, and one happened at a private school. One student acting alone was implicated in half of the incidents, while two to six students were accused in the other half. Guns were the most frequent intended weapon for plotters, though bombs, knives, and swords were mentioned in other incidents. The majority of plotters communicated their plans to others, with 30% doing so via e-mail or paper notes, 20% verbally informing others, and 15% admitting guilt when questioned by police. Rampage plots were uncovered in a variety of ways. The most common method was other students coming forward to inform school or police officials. This was often a result of plotters’ informing, and in some cases unsuccessfully recruiting, their peers, but students who had overheard rumors or been personally threatened also came forward. Other plots were averted by school administrators whose suspicions were aroused by rumors or irregular student behavior; staff who overheard the conversations of plotters; police who were alerted to rumors or found notes or e-mails which revealed the threat; and two events were avoided when plots were discovered as a result of the students being investigated for other crimes.
Additionally, Daniels and his colleagues (2010) interviewed school personnel at four schools where rampages were averted. The goals of this research were to better understand what roles school officials felt they played in preventing the rampage, what reasons they attributed to the successful outcome, and what advice they would offer to other schools. In their analysis, the researchers (Daniels et al., 2010, pp. 88) concluded that anti-bullying programs were perceived to be the “most salient” method of preventing future rampages. Many school officials also believed that supportive and strong relationships between school staff and students encourage people to come forward with knowledge about threats, and formal crisis planning that dictates specific roles helps to both prevent and respond to such potentially dangerous incidents.
Contemporary School Rampage Prevention Practices and the Student Code of Silence
Since the Columbine massacre of 1999, numerous shortsighted policies have been proposed and implemented in public schools across the nation. The American response has largely been to increase punitive disciplinary measures, such as zero tolerance policies, surveillance via security cameras and school resource officers (hereafter SROs), and security through target hardening practices such as metal detectors and limited entrances (Muschert & Madfis, 2013). Closer scrutiny, however, indicates that these solutions lack empirical grounding as being genuinely preventative and more likely reflect shortsighted efforts to reduce the anxieties of faculty, students, and parents. For example, schools continue to install security cameras and armed security guards specifically to prevent “another Columbine,” yet both of these measures were already in place at Columbine High School and did not deter or prevent those killings (Kupchik & Monahan, 2006, p. 625). Likewise, Red Lake High School rampage killer Jeffrey Weise walked through the metal detector at his school’s front entrance. When a security officer tried to confront Weise, he immediately shot and killed the man (Meloy & O’Toole, 2011). Zero tolerance school policies were widely adopted by as early as 1993 (Skiba, 2000); yet, they did nothing to curb the increased number of multiple-victim rampage shootings that occurred across the country in the late 1990s. As Cornell and Sheras (2006) have pointed out, excluding students from school via suspensions or expulsions does nothing to resolve the problems or deescalate the conflicts of students, and such punishments have even exacerbated the existing isolation and accumulated anger of previous school shooters (Levin & Madfis, 2009; Madfis & Levin, 2013).
In contrast, empirical scholarship on averted and completed school rampages has begun to formulate a very different approach toward violence prevention. In an early study of targeted school shootings, O’Toole (2000, p. 14) wrote about the prevalence and significance of a concept she called leakage that occurs when “a student intentionally or unintentionally reveals clues to feelings, thoughts, fantasies, attitudes, or intentions that may signal an impending violent act. These clues could take the form of subtle threats, boasts, innuendos, predictions, or ultimatums.” Vossekuil, Fein, Reddy, Borum, and Modzeleski (2002, p. 25) found that at least one person had some prior knowledge about the plans of perpetrators in 81% of targeted school shooting incidents, while multiple people were aware in 59% of their sample. Of those individuals who possessed such important knowledge, 93% were the peers of student perpetrators, such as friends, schoolmates, or siblings (Vossekuil et al., 2002, pp. 25). Meanwhile, a national report (Gaughan, Cerio, & Myers, 2001) found that only 54% of student respondents stated that they would tell an adult if they overheard a school peer discussing shooting someone. These discoveries heightened discussions about a student “code of silence” that stigmatizes students who come forward with pivotal information about their peers’ dangerous intentions (Culley, Conkling, Emshoff, Blakely, & Gorman, 2006; Epstein, 2002; Halbig, 2000; MacDonald & Da Costa, 1996; Merida, 1999; Morris, 2010; Spitalli, 2003; Stancato, 2001; Syvertsen, Flanagan, & Stout, 2009).
The few studies that actually explore how prior incidents of school rampage have been prevented (Daniels et al., 2007, 2010; Larkin, 2009; Newman et al., 2004; Pollack, Modzeleski, & Rooney, 2008) have largely located one common mechanism by which these potentially devastating tragedies have been averted—prospective school shooters revealing their violent intentions to others (i.e., O’Toole’s leakage) who in turn disclosed this concerning information to school and police authorities. As Daniels and his colleagues (2007) concluded, the most common method by which rampage plots have been averted was by means of students breaking through the code of silence to inform school and/or police officials about the leakage of their peers.
As a result, schools have implemented anonymous hotlines and e-mail systems designed to mitigate the problematic code of silence (Teicher, 2006; Wilson-Simmons, Dash, Tehranifar, O’Donnell, & Stueve, 2006; Wylie et al., 2010) and employed bystander prevention programs to encourage young people to be more active about speaking up and stopping the problematic behaviors of their peers (Lodge & Frydenberg, 2005; Twemlow, Fonagy, & Sacco, 2004). In addition, Larkin (2009) has suggested that, post-Columbine, students have become more likely to respond to threats by reporting them rather than acquiescing to the code of silence. In an interview with The New York Times about an averted rampage attack in 2001, criminologist James Alan Fox declared that “it shows that more and more students are willing to come forward and inform on classmates” and that, while snitching was unacceptable before Columbine, it is “now O.K., and is often the only thing to do when you can prevent a tragedy to classmates” (Butterfield, 2001). Likewise, in their study of 15 student bystanders who had prior knowledge of four averted and four competed school rampage attacks, Pollack and colleagues (2008, p. 12) revealed that several students explicitly recognized the Columbine attack as a “wake-up call” after which threats would be taken more seriously by students increasingly willing to break through the code of silence to inform authorities. Even more recently, delinquency expert and director of the Center for the Study and Prevention of Violence, Delbert Elliott, stated, “It’s a very good sign that the norm around ‘snitching’ or being a tattle-tale is changing” (McCrimmon, 2009, p. 5).
In addition to the assertion that peer pressure against “snitching” on or “ratting out” other students has potentially diminished due, at least in part, to recent extreme forms of school violence, Pollack and colleagues (2008) found that another aspect of school culture was influential in determining whether or not students came forward to reveal information to school staff regarding threats. They concluded that bystanders who alerted authorities “were influenced by positive relations with one or more adults, teachers, or staff, and/or a feeling within the school that the information would be taken seriously and addressed appropriately” (Pollack et al., 2008, p. 7). Likewise, those students who demonstrated an unwillingness to come forward indicated that they anticipated negative responses from school officials if they had shared information. One student bystander did not come forward because he found school authorities “too judgmental,” while another student indicated that he did not tell anyone about the gun he knew was on school property because he anticipated getting into trouble or being interrogated for stepping forward (Pollack et al., 2008, p. 12). This finding is backed up by research confirming the positive correlation between cohesive, supportive, and trusting school climates and student willingness to report threats of violence in both middle and high schools (Brank et al., 2007; Brinkley & Saarnio, 2006; Fein et al., 2002; Syvertsen et al, 2009; Wylie et al., 2010) and college settings (Sulkowski, 2011).
However, it has remained unclear to what extent the student code of silence has in fact been diminished in the post-Columbine era and what role positive student bystander behavior and beneficial school climates have played in averting school rampage incidents since that infamous day on April 20, 1999. Likewise, scholars lack a clear understanding about the types of relationships that exist between the students who leak threats and those who come forward (i.e., close friends vs. strangers or acquaintances), know little about the process by which students come forward (i.e., anonymously or openly), and know almost nothing about which authority figures students confide in when they do break through the code of silence (i.e., teachers, administrators, counselors, police officers, etc.). This article explores these questions via the detailed analysis of 11 instances of averted school rampage. Unlike prior studies that have relied upon content analysis of newspapers (Daniels et al., 2007; Larkin, 2009) or interviews with student bystanders (Pollack et al., 2008) or school officials (Daniels et al., 2010), this study entailed triangulating data from media accounts, legal documents, and interviews with school and police officials and therefore is able to compare interested parties’ perspectives on what led to successfully averted incidents with news media accounts and official documentation from the justice system. Such triangulation proves crucial, for officials’ perceptions of events did not always form a consensus with one another, nor with news reporting or legal documentation. Thus, the multiple data sources consulted in this study enable a more nuanced and detailed understanding of the role of bystander involvement in preventing school rampage.
Method
Cases were first located through the LexisNexis newspaper database. Like Daniels et al. (2007), the search terms included “school” and “plot*,” “school” and “rampage*,” and “school” and “shooting*,” though the additional terms “school” and “attack*” and “school” and “threat*” were added. Cases that occurred outside of the United States were excluded, as were any incidents in which adult strangers and former or current school staff were perpetrators, those that focused upon colleges or universities as desired targets, and those that resulted in any injury or death or where only one victim was intended. Additionally, academic and government-sponsored publications (such as Larkin, 2009; Newman et al., 2004; Virginia Tech Review Panel, 2007), popular press sources (Bower, 2001; Lieberman, 2009; Robertson, 2001), and Internet sites, which compile lists of school violence incidents (Brady Center to Prevent Gun Violence, n.d.; List of School-related Attacks Foiled or Exposed Plots, n.d.; Trump, n.d.) were consulted to gather as comprehensive a list as possible and to confirm the accuracy of data across multiple sources. By means of this method, 195 cases of averted school rampage attacks that occurred in the United States from 2000 to 2009 were located.
The next stage of data collection was exploratory qualitative research that involved contacting the middle and high schools where rampage threats were averted in the last 10 years. To make the project a manageable size, and to facilitate as many face-to-face in-person interviews as possible, the sample was limited to schools in the Northeastern United States (a geographical area that includes the New England states, New York, New Jersey, Pennsylvania, and Delaware). Of the 195 cases nationwide, 29 incidents occurred in Northeastern schools. To facilitate access, I contacted the current head principals at these schools to request access to conduct interviews with them and any other faculty or local police who were directly involved in preventing the potential violence.
Of the 29 school principals contacted, 11 consented to the research (henceforth, these incidents will be referred to as Cases A–K). Though 2 of the original 29 schools were private institutions, all of the 11 schools that granted me access were public institutions. Nine were high schools, one was a middle school, and one was a junior/senior high school. Two were located in Pennsylvania, one was in New York, two were in New Jersey, three were in Massachusetts, and three were in Connecticut. Six of these schools were located in suburban communities where the majority of residents were White and middle class, four of them were located in affluent, predominantly White suburban areas, and one school was in a majority White lower-middle-class rural community. 1
Many of the principals who agreed to be interviewed also referred me to other individuals at the local schools or police departments who could provide additional insight into the offenders, their threats, and the reactions of the schools and communities. In all, I spoke to 32 people (17 administrators, 4 counselors, 7 security and police officers, and 4 teachers) associated with these 11 incidents. Of these, 17 were conducted in person at respondents’ schools or police departments, and the other 15 interviews took place over the telephone when this was the respondent’s preference or when on-site interviews could not be coordinated. As the interviews were shaped by each respondent’s experience and willingness to talk, they varied in length from 26 min to nearly 3 hr, with an average length of 64 min. To triangulate the data gleaned from these interviews and confirm the accuracy of what respondents told me, I cross-checked their accounts of an incident against their colleagues’ accounts of the same incident, as well as with newspaper reporting, and whenever possible, court transcripts, legal briefs, and police incident reports. Utilizing all of this data in a grounded theory approach (Glaser & Strauss, 1967), typological categories and themes emerged through the process of coding with the ATLAS.ti software program.
Findings
A Diminished Student Code of Silence in the Post-Columbine Era
Many respondents discussed the student code of silence as a significant factor in rampage violence prevention and violence prevention more generally. In line with numerous scholars (Elliott in McCrimmon, 2009; Fox in Butterfield, 2001; Larkin, 2009; Pollack et al., 2008), this was framed, for the most part, in the context of a new post-Columbine era wherein the code had been vastly diminished as a result of extreme cases of school violence. For example, the head of school security in Case A, Mr. Pullman,
2
stated that: I think one of the things that Columbine has done is it has broken the old adage, you know, don’t rat out your brother … And I think the kids are learning that there is some point when you should be saying something. Obviously, you’re never going to get them to say everything, but I think the point has been made about the more important things.
Echoing this same idea, Mr. Sacco, the principal of high school G, pointed out how he believed things had changed since the Columbine attack. He commented: [Columbine] absolutely made kids much more vigilant about things going on around them … I think it made kids less afraid to speak up if something wasn’t sitting right with them. And since that era, I’ve often had a kid come by, and they’re pained and they’re tortured because they’re thinking about that … and they’ll say, “I’ve got to tell you this,” and so I think that’s the real good thing that’s happened.
One principal, Mr. Cooper, even felt that school lockdown drills, where students practice preparedness for a potential intruder or attack, helped to mitigate the code of silence as a constant reminder of potential school rampages. He believed that: Lockdown drills serve as a reminder to students and staff, and students especially, that, unfortunately, in this day and age, we do need to be ready to try to go into a lockdown. If that helps students to be more aware, if they see something that doesn’t sit right with them, then they might report that to somebody.
Thus, it was frequently understood by school and police officials that, in the wake of Columbine, students were much more likely to come forward when they gained knowledge about their peers’ intentions to do something dangerous or destructive.
Breaking the Code With a Positive School Climate
In addition to asserting the significance of Columbine as a watershed event easing the stigma against students who come forward, there was near unanimity about the appropriate methods to weaken the student code of silence; nearly all of the school and police officials expressed a firm belief in forging cohesive, supportive, and trusting school environments—views very much in sync with existing scholarship (Brank et al., 2007; Brinkley & Saarnio, 2006; Fein et al., 2002; Sulkowski, 2011; Syvertsen et al, 2009; Wylie et al., 2010). Almost without exception, respondents noted the importance of fostering an atmosphere where “students care about their school” because they “feel involved and part of the community,” and “students feel supported” and “know they can trust the adults in the building” because of “predictable and consistent” “close relationships” of “dignity and mutual respect.”
This pattern emerged in the abstract, as officials felt confident that a trusting school environment where kids feel safe confiding in the adults in their school building would ultimately mitigate hypothetical threats in the future. For example, Mr. Flaherty, the principal at the high school where Incident K took place, stated that: The best mechanism we have as a deterrent for these sorts of violent acts is good relationships between kids and adults, because kids will tell you. Kids will tell you when something happens. And if they saw a kid with a knife or they heard about it in the locker room … they’ll tell you, and that’s the best security you can have, better than any policeman with a gun, better than metal detectors, better than locks and cameras. That’s the best security … That is the best insurance you could ever buy.
On a more experiential level, this pattern of responses was also evident among officials’ statements explaining why the specific incidents at their schools were averted. They often credited their particular schools’ positive environment with making it easier for student bystanders to break through the code of silence.
In one instance that demonstrates this typical explanation, the vice principal of the high school during Case C, Mr. O’Brien, noted that his current school displays a particularly trusting relationship, in contrast to another school he worked at previously, which had “a less respectful and collaborative environment, where what the principal said, goes.” The principal at Mr. O’Brien’s school, Mr. Fernandez, similarly asserted the vital role he felt school culture played when he described why their incident was averted. He stated that: It was the result of a combination of relationships and taking things seriously and investigating fully that led to us averting what could have been another Columbine, if not worse … So the prevention factors were investigating fully and also having … a positive student culture where kids can talk to us and we can talk to them.
Dr. England, the school psychologist in Case B, shared Mr. Fernandez’s belief that positive relationships between school staff and students were crucial in preventing the incident she dealt with at her school. She shared that: What happened was that students came forward and told us about [the threat], so I think the fact that there are connections, such close connections, between administrators and students here that you’re much more likely to have students share that kind of information with you.
In addition to emphasizing the value of a positive school climate in general, some officials specified how particular aspects of their schools were important in diminishing the code of silence. One house principal, Mr. Harris, emphasized the positive impact that smaller class sizes and lower student/faculty ratios played in building the trusting environment which he felt ultimately encouraged a student to come forward in Case E. Mr. Harris stated, “We feel that smaller feel created better relationships for kids at their school and more of a connection and more care for their school. And that led to one student coming forward.” Several other officials highlighted the rapport that specific individuals, such as certain teachers, counselors, and SROs had with students. For example, when Principal Fernandez discussed the student who revealed information about his friend’s plot to his teacher, he described the teacher in the following manner: The teacher was a favorite among the students. He’s a young teacher … And students report that they like this teacher. They know that he works hard for them and that he cares about the school … I mean this is a guy who will volunteer for anything and go anywhere. He’s a can-do kind of guy, and I’ve been able to depend on him for a number of different things. So he gets relationships with the students, and that’s why he [the student who came forward] knew that he could trust him to do the right thing or genuinely help him when he shared the information.
Similar praise about unique rapport was heaped upon other officials who were the first authority figures entrusted with crucial information about potential attacks. An SRO was described in the following manner: She was not someone you would think of as a police officer. She handed out lollipops. She was very, very easy to talk to. Not scary in the least bit, not strict and formal as you would assume a police officer to be. So she was very caring and kind and very, very different than what you would assume an SRO to be.
Therefore, according to the accounts of nearly all of the school and police officials interviewed in this study, the rampage attacks upon their schools were thwarted largely due to students’ coming forward as a result of positive school climates where staff had forged inclusive communities as well as trusting and supportive relationships with their students. However, closer scrutiny of the cases themselves reveals a more complicated story and casts doubt on the extent to which the student code of silence has actually been diminished post-Columbine.
Breaking the Code—Interventions Through Leakage
Intervention as a result of leakage took four distinct forms in terms of whom the informants were in relation to the rampage threats and how these informants came to be aware of crucial information about these threats. These categories included bystanders who gained information indirectly from students accused of planning a school rampage, trusted confidants who were directly informed of plans by accused students, targets who were directly threatened by accused students, and co-conspirators who were directly involved, at least at some point and at some level, in the plots themselves. The discourse surrounding the code of silence and student leakage has heretofore only recognized the first two categories of student intervention by bystanders and confidants (though these have not been previously differentiated), and so the characteristics distinguishing these groupings warrant additional consideration.
Indirectly Informed Bystanders
The most common form of intervention, which occurred in five instances in the sample (Cases E, G, I, J, and K), was through innocent bystanders who were not intentionally entrusted with secretive information by the accused students themselves, but who came to know details about a rampage threat indirectly by inadvertently coming across web pages, hit lists, or threatening graffiti written on school property. In Case E, for example, a student who was a casual acquaintance of the students accused of plotting a rampage attack reported the existence of disturbing posts and images on their social network website profiles to the SRO and an assistant principal. Case G came to light in a fairly analogous manner, when several students expressed their concerns to school administrators about one of their peers having what they perceived to be a hit list of names written in his notebook. Likewise, in Case J, a teacher saw one of her students writing out a hit list on a piece of paper at the girl’s desk and subsequently notified the school principal. In Case I, a teacher and several students noticed a threatening message scribbled onto a wall in a boys' bathroom stall and informed the school principal. Similarly, in Case K, students who noticed a threatening message written on a wall in one of their boys' bathroom stalls informed teachers who in turn told the school principal and SRO.
Directly Informed Confidants
The second most common category of intervention involved people who had gained direct information about threats as a result of being confided in by the students accused of plotting attacks. In Case A, a student who had been sent threatening messages and video clips through the Internet informing her of her friend’s desire to attack his school led her to tell her mother who then called the local police department. When questioned by police, the accused student confessed his plans, and an arsenal of weaponry was discovered in his bedroom. In Case B, several students came forward to tell administrators that a peer had brought explosives to school with him, which administrators then found in his backpack. In Case C, John, a high school student, stole guns from his father and gave them to his friend Tim to hold onto until they were needed to carry out the attack. John promised Tim that if he held onto the guns, he would not harm Tim or Tim’s girlfriend. Tim ultimately revealed John’s plans to his parents. Tim’s mother helped Tim get rid of John’s weapons by throwing them into a local stream (an act for which she would later be prosecuted herself), and Tim’s father later informed the school’s principal and vice principal about John’s plans to commit a rampage attack at the school. Shortly after being questioned by the school administrators and denying knowledge about the plot, Tim took it upon himself to tell one of his teachers about his complicity in John’s plot. This teacher then immediately informed the school administration. In Case F, the girlfriend of one of the accused students alerted school officials about the numerous detailed threats made by one of her boyfriend’s friends, though she did not come forward until after her boyfriend had already been accused by several of his co-conspirators of planning the attack.
Threatened Targets
A third form of intervention occurred when people who had been threatened by students came forward. Numerous scholars (Pollack et al., 2008; Reddy et al., 2001; Vossekuil et al., 2002) have noted that targeted attackers who pose serious risks rarely threaten their victims in advance, but Meloy, Hempel, Mohandie, Shiva, and Gray (2001) found that the majority of the adolescent mass murderers studied in their sample did, in fact, directly threaten their targets before committing crimes. In two incidents (Cases D and H), people who felt directly threatened by interactions with students accused of plotting school attacks alerted authorities, though only one incident of threatening occurred in advance and entailed a student coming forward. The students in Case D tried and failed to carjack a vehicle, and police later arrested them while they were walking down a street with an arsenal of weapons. It was their potential carjacking victim, an adult stranger with no connection to the students or school, who alerted police. Perhaps as a result of the student code of silence, it is well established that juvenile victims in general are far less likely to report crimes committed against them to police than are adult victims (Watkins, 2005). However, in Case H, it was one of the numerous students who had been threatened by name on a social networking website by several other students who informed one of his parents of the troubling website exchange. This parent, in turn, called the school board president to express concern.
Involved Co-conspirators
Academic and public discourse has generally neglected to consider the role of co-conspirators who participate in developing a school rampage plot at some point and at some level, but who later come forward to prevent the attack from taking place. The one exception is that of Larkin (2009) who noted that one of the participants in a rampage plot that was averted in New Bedford, Massachusetts, in 2001 came forward because she feared for the safety of her favorite teacher. While incidents such as this, where co-conspirators come forward, may be more uncommon than any of the prior categories of intervention through leakage, Case F followed this pattern. The sequence of events in Case F is rather complicated, but as this particular incident seriously complicates and extends existing narratives about student bystander behavior and the code of silence, it warrants extensive elaboration.
The facts of this case begin when Tommy, a student who would later be charged and convicted of conspiracy to commit murder and threatening to use deadly weapons, is discovered hiding a knife from his father during the course of an emotional outburst. Fearing that Tommy is suicidal, his father subsequently has him hospitalized at a psychiatric facility. Matthew, the other student later convicted on conspiracy charges, visits Tommy in the hospital. While there, Matthew becomes convinced that Tommy is being “brainwashed” and has abandoned their joint plan to attack the school along with two other students, Sean and Justin. Matthew makes an angry scene at the hospital in front of Tommy’s family and girlfriend. After Tommy is released from the hospital, other students observe a positive change in his behavior. Eventually, Tommy and Matthew get into a fight with one another at a party, and Tommy tells Matthew never to speak to him again. According to Tommy, Tommy also tells Matthew at this time that he hopes someone finds out about Matthew’s plot.
Matthew then takes Sean and Justin, the other two student co-conspirators, with him to go speak to the adjustment counselor and later one of the SROs. Matthew tells both Sean and Justin not to speak during these meetings because he should do the talking, since he knows how to handle cops due to the fact that his father is a police officer. Matthew then tells these school officials that they are all afraid of Tommy who has been threatening them, mentioning how he wanted to attain weapons and commit a rampage attack against the school. Both Sean and Justin go along with this plan out of fear of Matthew who had assaulted Sean shortly before this at a party and also because Matthew had threatened to cut out their tongues if they told anyone. Matthew subsequently blames everything on Tommy, while Sean and Justin concur in relative silence. During this time, Matthew assures Sean and Justin that they will get away with concealing their own involvement, and Matthew destroys evidence linking them to the plot. Police search Tommy’s home and find maps and lists of weaponry in his bedroom.
Much later, in front of the grand jury, Justin finally has a conversation with police away from Matthew and informs them of Matthew’s actual deep involvement in the conspiracy. Sean and Justin are granted immunity in exchange for their testimony in the trial of Tommy. The prosecuting attorneys, arresting officers, SROs, and adjustment counselor all felt that Matthew went to the police because he believed he might get caught after his relationship with Tommy had deteriorated, and he no longer trusted that Tommy would keep silent about their plan. In contrast, Matthew’s defense attorneys argued that he abandoned his initial participation in the plot and actually stopped the attack from taking place (though they also denied that any of the accused students ever intended to carry out the attack in the first place). The defense’s argument was ultimately unsuccessful in court and was also not believed by any of the school or police officials interviewed in this study because of the fact that, when Matthew initially came forward, he had clearly lied to everyone in order to hide his true involvement in the plot.
The manner in which this case came to light does not fit with the typical rhetoric of an emergent willingness to break the code of silence as a result of Columbine or a positive school climate. However, several school officials who handled this case still credited their healthy school environment with averting the incident. Mr. McGowan, the principal in Case F, noted that: The bottom line is, it’s about relationships. And when your kids feel comfortable being able to come and talk to you about stuff. We knew about [the rampage plot] because of [the trust we built at a school-sponsored leadership summit], and the fact that that kid trusted [the SRO] because she was with him for those two or three days, and he felt comfortable being able to come and say “I don’t know if this is good.”
One of the arresting officers in this case, Captain Dante, shared this sentiment. He stated that the school counselor and SROs: were close enough with the kids and built a relationship with the kids that they felt like they could sit down and tell them what was going on … when they thought twice, they went to somebody they felt comfortable talking to because they had built a relationship with them. And then I think that’s the biggest thing.
However, it was not the case that students ultimately decided to turn to adults whom they trusted and reveal hidden plans to them. In fact, the plotters lied to the school officials they approached when they came forward and denied their own involvement. This was not an example of a trusting environment encouraging innocent bystanders to come forward, but one in which three members of a school rampage conspiracy came forward to inform on the fourth member of their group in order to divert attention and blame away from themselves.
How Interventions Occurred and Who Was Trusted?
All available information about these incidents indicates that when people came forward, they did so openly rather than anonymously. To some extent, this may reflect the fact that the incidents under investigation in this study ranged from the years 2000 to 2009, and anonymous reporting systems have only been implemented in the last few years (McMillin, 2009; Payne & Elliott, 2011).
When people did come forward to reveal information, they were typically students (as in Cases A, B, C, E, F, G, H, I, and K), though it was an adult who notified the police in Case D, and teachers alerted school administration in Cases I and J. This speaks to Stueve and colleagues’ (2006) call for scholars to expand the definition of bystanders beyond students to include parents, teachers, and school staff members, because they may also possess information about potential violence. In this vein, parents also deserve credit for approaching the authorities in three cases (A, C, and H), as students in these cases revealed information to their parents before or instead of school or police officials.
More generally, beyond the three instances where parents were alerted, students informed SROs in two cases (E and F), administrators in four cases (B, E, G, I), counselors in one case (F), and teachers in two cases (C and K). It is only the instances in which students came forward to school or police officials that technically qualify as breaks in the code of silence. While these criteria apply in 7 of the 11 cases (B, C, E, F, G, I, and K), the adult victim and teacher involved, respectively, in Cases D and J do not qualify, nor do the two instances (Cases A and H) in which students informed only their parents rather than school or police officials. This suggests that, in addition to the expanded notion of bystanders proposed by Stueve et al. (2006), perhaps scholars should similarly contemplate whether or not students who reveal information to their parents technically qualify as breaching the code of silence.
Following the Code—The Persistence of Bystander Inaction
Even in the cases in which some students came forward—and that could subsequently be deemed successful in terms of illustrating positive examples of a broken code of silence—such an unequivocally rosy interpretation oversimplifies the truth. While student bystanders, trusted confidants, and targets deserve credit for individually coming forward and valuing school safety over student norms and concerns about status, positive bystander behavior on the part of a few students often occurred in the larger context of many more students keeping silent about their own knowledge of threats.
Case A perhaps best illustrates this pattern; while it is true that a female student is responsible for revealing the threatening behavior of one of her friends by informing her parents and then the police, many more friends and acquaintances of this accused student had been exposed by varying degrees to his disturbing comments and actions. In fact, the accused student in this case, Jessie, leaked various elements of his school rampage plot to a number of his peers. When later questioned by the police, many of his fellow students acknowledged being aware of his deep fascination with the Columbine attack and had frequently heard him discuss his desire to shoot up and bomb the school. He had posted numerous references to Columbine on his social networking profile page, which many of his friends and school acquaintances read.
As Detective Brown, the SRO directly involved in this incident, commented, “These are all different people, interviewed separately, that are saying the same thing … which makes you wonder why no one said anything.” At least four students entrusted with intimate knowledge about Jesse’s intentions as his confidants did not come forward to reveal what they knew until after Jessie had been arrested. When only one in five students (and these four were merely those that the police had extensive knowledge about because they were willing to provide testimony) came forward to the authorities, this is perhaps more of an indication of the continued resilience and power of the student code of silence, rather than a success story of a school somehow breaking through the code via a positive school atmosphere.
Case A was not alone in illustrating this problematic trend, as many other instances that could potentially be praised for illustrating positive bystander behavior or broken codes of silence actually revealed, upon closer inspection, more examples of leakage that had gone unnoticed, dismissed, or ignored. In Case E, one student came forward to the SRO and then to a house principal in order to inform them of his friend’s nefarious plans. However, several other students had been asked to join in on the murderous plot; they declined not only to participate in it but also to inform any adults about its existence.
In Case F, the four student plotters attempted to recruit at least two other students to participate in their conspiracy to attack the school. While these students rejected the offer to join the group of student plotters, none of the reticent would-be recruits came forward to reveal what they knew. One of the SROs involved in this case, Officer Dudley, stated that: We had at least 15 witnesses that did statements in this case … All of a sudden you’ve got 15 kids that knew about this for the last year and a half. And it turned out that a couple of the kids that were key to the verdicts failed to be recruited … and some of those kids turned out to be our best witnesses … But nobody came forward until these boys came forth [to blame their conspiracy on one of their accomplices].
In addition to the students they tried and failed to recruit, one member of the conspiracy told a female student that if anything were to happen, she would be safe because he considered her a friend, and several other female students were explicitly informed about their plans to shoot up the school without expressly being invited as participants. As previously mentioned, only one of these female students came forward to school officials, and this only took place after initial accusations had already been made. 3
Discussion
Ultimately, the findings above confirm previous research and reporting about the manner in which people’s coming forward with important information gained via leakage leads to rampage attacks being averted. They also reveal that the dominant approach to school violence deterrence via punitive disciplinary policies and enhanced security did not play a prominent preventative role. These events were not deterred due to the presence of metal detectors, locked doors, security cameras, or SROs. In fact, many of the student plotters considered these developments to be minor stumbling blocks easily resolved through additional preparations among already detailed plans. For example, some plots entailed killing or disarming officers and bringing weapons into the school buildings before entrance doors would be locked for the day. Additionally, though the logic of zero tolerance is similarly deterrence based, wherein strict punishments for threatening behavior or weapons’ possession in school would dissuade students from these activities, most of the schools where rampages were averted displayed some elements of zero tolerance (though support for the approach varied widely), and all of them were subject to the mandate of the 1994 Gun Free Schools Act wherein bringing a firearm to school warrants a mandatory yearlong expulsion. Thus, these myriad deterrence-based strategies cannot be credited with thwarting these homicidal threats. However, the data do indicate that SROs can play a preventative, if not deterrent, role in averting rampage, as it was SROs whom students trusted enough to approach in two cases (E and F, though the latter incident is less indicative of a genuine display of trust), and they did help garner vital confessions from accused students in Cases A, C, and E.
In addition, the findings suggest that numerous potentially lethal rampage attacks were in fact prevented, at least in part, as a result of school cultures sufficiently positive for students to feel comfortable telling school authorities about threatening behaviors. However, in Case F, school officials credited a positive school culture for averting the attack, even though this was undoubtedly not how the incident played out. The same could be said for Incident A, where the head of school security praised his own school’s positive school culture, when it was actually a student from another town and school district who actually came forward to the authorities, while none of the students at his school who were exposed to leakage did so. Thus, there may be some tendency on the part of officials to attribute undue credit for averting these incidents to positive school climates.
Certainly, students breaking through the code of silence still constituted the primary manner by which most events have been thwarted. That said, close inspection of the details of these cases indicates a murkier and less optimistic picture than the one painted by scholars (Elliott in McCrimmon, 2009; Fox in Butterfield, 2001; Larkin, 2009; Pollack et al., 2008) and generally recognized by the school and police officials interviewed in this study, many of whom agreed that the student code of silence drastically diminished in the wake of numerous highly publicized rampage shooting attacks such as the one at Columbine High School. It is likely that both scholars and practitioners came to this conclusion through media reporting (such as Bower, 2001; Butterfield, 2001; Robertson, 2001) upon recent incidents where rampage plots were averted as a result of students coming forward with information gained through leakage. However, by carefully investigating incidents of averted rampage in detail, two distinct elements of the findings of this study suggest previous conclusions based upon superficial accounts in the media are at best premature.
First, those who came forward were not uniformly close friends and confidants of student plotters; they were also acquaintances, victims of threats, and even conspirators themselves. MacDonald and Da Costa (1996) found that students were more willing to report victimization if the perpetrator was not a friend, and Pollack and colleagues (2008, p. 13) noted that one of their student respondents came forward in part because of the fact that “he was not close friends with the potential attackers, so that allowed him to be more objective when he learned of a possible attack plan.” Thus, it should not be entirely surprising that authorities are often alerted to the presence of rampage threats by people less close to the accused perpetrators. This does suggest, however, that a student code of silence among close friends remains a significant challenge. Second, this concern about students maintaining secrecy in their most personal friendships is exacerbated by the fact that many of the adolescents exposed to leakage who were the closest to the students accused of plotting attacks did not in fact come forward to authorities. More generally, several cases entailed more students exposed to the leakage of their peers who did not ultimately come forward than the number of students who did in fact do so. Thus, while some scholars and media coverage have depicted even one student who comes forward and has any type of relationship to an accused plotter as evidence that the code of silence has diminished, this narrative is complicated when one considers the actual relationships between student bystanders and plotters as well as the rarely reported on but numerous students who have not come forward even in these averted incidents.
Therefore, if scholars want to interpret averted incidents as an indication that the student code of silence has diminished, it is worth considering more specific measures or levels of both leakage and, for lack of a better term, code breaking (i.e., positive bystander intervention by students). Meloy and O’Toole (2011, p. 525) suggested that researchers need more sophisticated studies of leakage that explore the various forms it takes in order to ultimately discern which, if any, are the most “predictive of actual targeted violence.” In the same way, breaks in the code of silence are not uniform and should not be depicted as such by scholars proclaiming their significance and increased occurrence. It may be the case that students who come to be directly informed of threatening information by their closest friends are exposed to the most profoundly significant form of leakage. It is almost certainly the case, however, that these close confidants who then share that knowledge with authorities engage in a far more substantial breech in the code of silence than students who come forward after being targeted themselves or who only know the students they are accusing as distant acquaintances. Thus, instances in line with the former type (such as in Cases A, B, and C) constitute a far better indicator of a broken student code of silence than instances of the latter types (such as E, G, H, I, and K), 4 though all of these latter cases do demonstrate some indication of a willingness to break through the code. At the same time, when only one student comes forward though many more possess information about a threat of violence (as was documented in Cases A and E), this should not be counted as equivalent evidence of a broken code of silence as would an incident in which multiple or even all students with relevant information came forward. Thus, future scholarship on both leakage and the code of silence must consider these nuanced distinctions in order to properly measure and understand the role that both phenomena play in averting school rampage and other forms of targeted school violence.
Conclusions
The findings of this study ultimately suggest that, while students’ coming forward with important information about threats constitutes the key manner by which rampage attacks are averted, a student code of silence persists beyond the previous assumptions and expectations of academics, police officers, and school officials. This confers with Wylie and colleagues (2010, p. 351) assertion that “[a]lthough policies aimed at improving school climate may increase a student’s willingness to report and are important in their own right, improving a school’s climate may be a daunting task.” Syvertsen, Flanagan, and Stout (2009, p. 230) point out that there are potentially “scores of reasons” suggested by decades of bystander research that may explain why students do not intervene in dangerous situations, beyond that of a strict adherence to the code of silence. Bystander scholarship highlights the importance of diffused responsibility, the disinclination to intercede while part of a group because individuals anticipate that others will respond instead (Mathes & Kahn, 1975). As numerous students were aware of the threats made in several of these cases, it is feasible that some students neglected to come forward because they expected their peers to do so. The literature on bystanders also considers the role of ambiguity about the situation (Latané & Darley, 1969). Students may not approach adults if they lack certainty about their peers’ intentions, and perhaps interpret legitimate threats as innocent jokes or comments. Though inaction on the part of students exposed to leakage likely results from a combination of these forces (Pollack et al., 2008), and much work remains to be done in the realm of student bystander awareness and education generally, the findings of this study reveal an important misunderstanding in how both many academics and practitioners discuss healthy and successful school environments.
Plentiful empirical evidence (such as Brank et al., 2007; Brinkley & Saarnio, 2006; Eliot, Cornell, Gregory, & Fan, 2010; Fein et al., 2002; Sulkowski, 2011; Syvertsen et al., 2009; Wylie et al., 2010) exists to indicate the relationship between cohesive, supportive, and trusting school climates and positive student bystander behavior, and so the problem is not so much that scholars and educators mistake the importance of a beneficial school climate. Rather, the problem is that they are largely mischaracterizing the current atmosphere of American public schools. Despite institutional objectives of inclusiveness and the significance of certain especially kind and empathetic school and police officials, any honest examination of recent developments in American schooling reveals a significant trend away from inclusivity, empathy, and supportiveness, and toward the punitive discipline and enhanced security that Hirschfield and Celinska (2011) label “school criminalization” (see also Hirschfield, 2008; Kupchik, 2010; Kupchik & Monahan, 2006; Lyons & Drew, 2006; Monahan & Torres, 2010; Muschert & Peguero, 2010; Nolan, 2011). The increased use of law enforcement mind-sets, personnel, and technologies in schools, as well as the increasing transfer of school discipline to the juvenile and criminal justice systems (Advancement Project, 2005; Kim, Losen, & Hewitt, 2010; Wald & Losen, 2003), have direct consequences upon the overall school climate. As Casella (2001, p. 35) aptly described, such “policies … bolster punishment in favor of pedagogy, control in favor of understanding.” The result is that many of these features erode trusting relationships between school staff and students (Lintott, 2004; Noguera, 1995; Watts & Erevelles, 2004). This has direct consequences for the prevention of school rampage. Syvertsen and colleagues (2009, p. 229) describe the dilemma eloquently when they conclude that: In this post-Columbine era, public education has seen an increase in zero tolerance policies … It is possible that these policies create an environment that actually discourages students from revealing their concerns to teachers because of the increased “costs” of revelation … For most adolescents, divulging a peer’s confidence is a difficult decision that may be intensified by a zero tolerance climate … [I]n our findings, the more students believed that going to a teacher or principal would result in trouble, the more likely they were to ignore a peer’s dangerous plan or to simply tell a friend (but not an adult).
School criminalization diminishes positive bystander behavior not only as a result of lost trust due to excessive punitiveness, but the perception that such discipline is prejudicial or unfair may similarly lead to fewer students coming forward. Morris (2010, p. 258) found a dominant code of silence at both of the high schools he studied, one a predominantly African American urban school and the other a predominantly White rural school. During his fieldwork, the urban school installed security cameras and began a practice where principals, hall monitors, and SROs “scoured school grounds to ensure student rule compliance.” Reflecting on this development, Morris (2010, p. 270) proposed that: such criminal justice inspired policies would exacerbate the code of silence … Because stop snitching [culture] emerges … based on ambivalence to authority and an emphasis on independent conflict resolution, cracking down with more authoritarian or invasive measures would only increase the code’s strength. Particularly, if students perceive strict or invasive school discipline as biased, they might resist school authority more vehemently, increasing the social distance between students and the school.
It has been widely documented that African American youth are vastly overrepresented in school suspensions and expulsions (Casella, 2001; 2003), even when socioeconomic indicators are held constant (Skiba, Michael, Nardo, & Peterson, 2002), and when their punishments are compared to White students who committed identical infractions (Fenning & Rose, 2007). A recent government report (Civil Rights Data Collection, 2012) indicated that, during the 2009–2010 school year, over 70% of students involved in school arrests were African American or Hispanic, while African Americans accounted for 46% of those suspended multiple times and 39% of expulsions. Even zero tolerance policies, which were formed at least in part with the goal of objectivity in mind, have resulted in disproportionate application toward minority students (Kupchik, 2010; Robbins, 2005). Prolific scholarship indicates that perceptions of racial discrimination erode confidence in the police and criminal justice system, especially among minorities (Bayley & Mendelsohn, 1969; Russell-Brown, 2008; Weitzer, 1996). This often results in members of these communities adopting what Anderson (1999) calls a “code of the street,” one component of which is a reluctance to notify or help the police. Morris (2010) argues that the student code of silence is essentially the adolescent version of Anderson’s code. This suggests that, to the extent that students perceive school discipline as unjust due to racial prejudice, they will be less likely to entrust school and police officials with important but potentially incriminating information.
Therefore, even though many of the recent measures focusing upon punitive discipline and enhanced security were designed specifically to improve school safety in the event of a rampage attack (Muschert & Madfis, 2013), these very practices significantly hinder the one means by which nearly all rampage attacks are actually thwarted—positive bystander behavior on the part of students aware of leakage. Kohn (2004, p. 26) clarifies the most salient point here when he states that such punitive policy “isn’t merely ineffective—it’s actively counterproductive.” That is, doing nothing in response to school rampage fears would have been better than many of the authoritarian practices schools have put in place.
In order to avert future attacks from occurring, schools must abandon these punitive measures and improve upon the means by which these events are actually prevented, via forging genuinely positive school climates that encourage student bystanders to intervene in a responsible manner. In this regard, several innovative programs stand out as instructive guides. For example, Cornell (2013) has long argued for a preventative approach to threat assessment, and has recently developed, implemented, and tested his own model for student intervention in schools, called the Virginia Threat Assessment Guidelines. Likewise, Network against School Shootings, also known as the Berlin Leaking Project, has not only been a vital source of German data on school threats of impending violence, but it has developed and implemented an empirically driven school violence intervention program that trains teachers to recognize student leakage behaviors (Bondü & Scheithauer, 2010; Leuschner et al., 2011). Due to their nuanced conflict resolution approach, both of these examples constitute significant improvements over the simplistically punitive zero tolerance approach which not only fails to deter rampage or to accurately assess threats but which may exacerbate existing student conflicts and actively discourage positive bystander behavior.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The author would like to acknowledge generous research funding provided by Northeastern University's Brudnick Center on Violence and Conflict.