840231

research-article2019
CCJXXX10.1177/1043986219840231Journal of Contemporary Criminal JusticeLankford et al.

Article
Journal of Contemporary Criminal Justice
2019, Vol. 35(3) 315­–341
Are the Deadliest Mass © The Author(s) 2019
Article reuse guidelines:
Shootings Preventable? sagepub.com/journals-permissions
DOI: 10.1177/1043986219840231
https://doi.org/10.1177/1043986219840231
An Assessment of Leakage, journals.sagepub.com/home/ccj

Information Reported to Law

Enforcement, and Firearms
Acquisition Prior to Attacks in
the United States

Adam Lankford1, Krista Grace Adkins1,

and Eric Madfis2

Abstract
This study examined the 15 deadliest public mass shootings in the United States
from March 1998 to February 2018 to assess (a) leakage of violent thoughts/intent,
(b) leakage of specific interest in mass killing, (c) concerning behaviors reported to
law enforcement, (d) concerning interest in homicide reported to law enforcement,
and (e) firearms acquisition. We then compared our findings on the deadliest public
mass shooters with the Federal Bureau of Investigation’s (FBI) findings on active
shooters in general. Overall, the results suggest that most incidents were indeed
preventable based on information known about offenders in advance, and that
the deadliest mass shooters exhibited more warning signs and were more often
reported to law enforcement than other active shooters. Future prevention efforts
should aim to educate, encourage, and pressure the public to report warning signs
to law enforcement, educate and train law enforcement so that they can more
effectively investigate potential threats, and limit firearms access for people who have
admitted having homicidal or suicidal thoughts or being interested in committing a
mass shooting. These relatively straightforward steps could significantly reduce the
prevalence of future attacks.

1The University of Alabama, Tuscaloosa, AL, USA

2University of Washington Tacoma, WA, USA

Corresponding Author:
Adam Lankford, Department of Criminology & Criminal Justice, The University of Alabama, P.O. Box
870320, Tuscaloosa, AL 35487-0320, USA.
Email: Adam.Lankford@ua.edu
316 Journal of Contemporary Criminal Justice 35(3)

Keywords
public mass shootings, warning signs, prevention, leakage, firearms

Introduction
Despite decades of research into the psychology, behavior, and life histories of mass
shooters, many people remain skeptical that it will ever be possible to predict or pre-
vent their attacks. For instance, recent media articles have insisted that “We Can’t
Predict Who Will Commit a Mass Shooting” and that “Predicting a Mass Shooting Is
Impossible” (Resnick & Zarracina, 2018; Singal, 2015).
The phrase “mass shooting” is often used to refer to a more specific type of crime
known as a public mass shooting or an active shooting. These are incidents in which
perpetrators open fire in public places with the intent of harming multiple victims, and
do not include gang conflicts, robberies, or other more conventional crimes (Blair &
Schweit, 2014; Lankford, 2016b). The key distinction is that while public mass shoot-
ings are traditionally defined as cases resulting in four or more victims being killed,
active shootings have no minimum threshold (Blair & Schweit, 2014; Fox & Levin,
2015b; Lankford, 2016b). Past attacks have occurred at schools and colleges, work-
places, malls, movie theaters, churches, government buildings, military facilities, and
other public locations (Blair & Schweit, 2014; Kelly, 2012).
Part of the prediction and prevention challenge is that even in the United States—
where these offenders are most common—there are more than 1 million people who
will not commit a public mass shooting for every individual who does (Blair &
Schweit, 2014; Kelly, 2012; Lankford, 2016b). This makes preemptively identifying
possible offenders seem as hopeless as finding the proverbial needle in a haystack.
Even if the search for potential mass shooters is narrowed to subsets of the population
most likely to contain at-risk individuals, this is not nearly precise enough for success-
ful risk assessment. For example, most public mass shooters are male, but the vast
majority of men and boys would never consider committing a mass shooting (Madfis,
2014c). Similarly, most public mass shooters own guns and struggle with mental
health problems, but the vast majority of gun owners and people with mental illness
are also nonviolent (Madfis, 2014b; Schildkraut & Muschert, 2013; Schug & Fradella,
2014).

Leakage of Mass Shooters’ Violent Thoughts and Intent

Fortunately, the challenge is not necessarily to identify public mass shooters among
millions of law-abiding citizens, because in many cases they actually identify them-
selves. This is often referred to as “leakage.” As O’Toole (2000) summarized in her
threat assessment of school shooters, “‘Leakage’ 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” (p. 16).
Lankford et al. 317

In some ways, this appears similar to the behavior of people who die through indi-
vidual suicides. Crisis intervention guidelines state that “80% of people considering
suicide give some sign of their intentions” (Giffords Law Center, 2018). Although it
might seem irrational for mass shooters to signal their criminal intent, in many cases
they are also suicidal, so they may have a conscious or unconscious desire to be
stopped or saved before they end up dead or serving a life sentence in prison (Lankford,
2015).
Previous research has found that leakage is quite common among similar offenders
(Bondü & Scheithauer, 2010; Hempel, Meloy, & Richards, 1999; Leuschner et al.,
2011; Madfis, 2014a; Meloy, Hempel, Mohandie, Shiva, & Gray, 2001; Meloy &
O’Toole, 2011; Silver, Horgan, & Gill, 2018; Silver, Simons, & Craun, 2018; Vossekuil,
Fein, Reddy, Borum, & Modzeleski, 2002; White, 2017). For instance, Hempel et al.
(1999) found that among 30 adult mass murderers who offended from 1949 to 1998,
67% made specific or general threats about their violent intentions. Similarly, Meloy
et al. (2001) found that among 34 adolescent mass murderers who offended between
1958 and 1999, 58% of those for whom there was sufficient information made threats
about committing mass murder. In addition, a 2002 study on school shooters by the
U.S. Secret Service found that in 81% of cases, “at least one person had information
that the attacker was thinking about or planning the school attack,” and in 59% of
cases, more than one person knew that the attacker was thinking about or planning an
attack (Vossekuil et al., 2002, p. 25). More recently, a study of public mass murderers
from 1990 to 2014 found that 58% engaged in some form of leakage (Silver, Horgan,
& Gill, 2018), and a Federal Bureau of Investigation (FBI) study of active shooters
from 2000 to 2013 found that 56% openly expressed their violent thoughts or intent
prior to attacking (Silver, Simons, & Craun, 2018).
Some of these studies have looked more closely at other aspects of leakage. For
instance, it appears that leakage is more common among younger offenders (Silver,
Simons, & Craun, 2018; Vossekuil et al., 2002), that leakage is sometimes communi-
cated directly to the targets themselves (Hempel et al., 1999; Meloy et al., 2001), and
that leakage comes in many different forms, including verbal statements, written state-
ments (letters, essays, etc.), online posts, and social media communications (Madfis,
2014b; Silver, Simons, & Craun, 2018).
However, questions remain about public responses to leakage. If mass shooters
commonly tell someone that they are interested in committing violent crimes, why are
these attacks still happening? Are members of the public not reporting this information
to law enforcement?

Public Reporting of Potential Mass Shooters to Law

Enforcement
Unfortunately, prior research suggests that people are often hesitant to report potential
mass shooters to law enforcement, even if they observe concerning behavior (Madfis,
2014a; Silver, Simons, & Craun, 2018). This has primarily been explored in the context
of school shootings, with emphasis placed on how school cultures and climates either
318 Journal of Contemporary Criminal Justice 35(3)

encourage or discourage reporting behavior and prevention strategies (see Madfis,

2014a, 2014b). However, more scholarship is needed to examine reporting of potential
mass shooters in general (Meloy & O’Toole, 2011; Sarteschi, 2016).
The apparent reluctance of people to report concerning information about mass
shooters appears consistent with research on barriers to bystander intervention, which
shows that many people are uncomfortable speaking up against things that seem
wrong, because doing so seems to put the burden of responsibility on them (Madfis,
2014a; Stueve et al., 2006). In addition, the perception of diffused responsibility,
whereby people often assume that others will respond, also leads many to not come
forward themselves (Mathes & Kahn, 1975). On a personal level, it can feel safer to do
nothing. Research on bystander behavior also emphasizes the risks of ambiguity in a
given situation (Latané’ & Darley, 1969). People may not communicate their knowl-
edge about leakage if they are uncertain about a person’s intentions, and they may be
more prone to interpret legitimate threats as innocent jokes or comments—which is
why improved education about warning signs is particularly important (Hoffman,
2017; Madfis, 2014a, 2018).
An additional challenge is that many people seem to believe that calls to “see some-
thing, say something” primarily refer to the suspicious behavior of strangers. Although
they may be willing to report a stranger, they appear more hesitant to report a family
member, friend, or coworker. For example, in his research on averted school massa-
cres, Madfis (2014a) found that students who observed leakage were more likely to
report the threatening behavior of acquaintances than of close friends. More broadly,
some people may believe that reporting a friend or family member would constitute
betrayal (Lankford, 2018).
It should be noted, however, that similar interventions already occur in the area of
suicide prevention. In that realm, some family members and friends have not only
recommended that at-risk individuals voluntarily seek psychological treatment, but in
extreme cases they have also called law enforcement to conduct welfare checks that
can result in the individual being taken into custody. People typically do this because
they realize it will protect the long-term health of the person they care about (National
Alliance on Mental Health, 2018). When it comes to reporting potential mass shooters,
calling the police could save both the at-risk individual’s life and the lives of potential
victims (Lankford, 2018).
An additional consideration is the relationship between law enforcement and com-
munity members. Because positive bystander behavior (such as reporting threats) is
closely correlated with positive perceptions of police legitimacy and procedural fair-
ness (Madfis, 2014a; Tyler & Fagan, 2008), the public’s trust in criminal justice insti-
tutions may be associated with their reporting of potential mass shooters. People may
be more likely to report warning signs to law enforcement if they are confident that
they will be treated with respect, that the at-risk individual will be treated humanely,
and that it will actually make a difference.
How often do members of the public actually report mass shooters to law enforce-
ment? Although this seems like an extremely important question, only one study with
relevant data could be found. A recently released FBI study (Silver, Simons, & Craun,
Lankford et al. 319

2018) indicated that 41% of active shooters who attacked in the United States from
2000 to 2013 were reported to law enforcement due to their concerning behaviors.
This raises other important questions. Why were offenders who were reported to
law enforcement nevertheless able to attack? Is law enforcement responding effec-
tively to information about potential threats?

Law Enforcement Investigation of Potential Threats

Recent history suggests that when law enforcement officials receive a warning about
a potentially dangerous individual, they often respond by checking that suspect’s crim-
inal record and possible affiliations with other criminals, and/or interviewing the indi-
vidual directly (Associated Press, 2015; Cullen, 2009; Webster Commission, 2012;
Wilber, 2016). However, many public mass shooters do not have connections to other
criminals and have not been convicted of violent crimes prior to their attacks (Fox &
Levin, 2015a; Silver, Simons, & Craun, 2018). And when interviewed, many of these
individuals can successfully fool law enforcement investigators into believing that
they are not dangerous (Cullen, 2009; Wilber, 2016). This should not be surprising
because even people with strong suicidal tendencies often deceive their own doctors
about their plans to die (Nock et al., 2010). Potential mass shooters may similarly
recognize that unless they convincingly deny their intentions, they will be prohibited
from accomplishing their lethal goals.
Of course, law enforcement officers have a wealth of professional knowledge and
field experience, and they are accustomed to being lied to. However, most public mass
shooters are psychologically and behaviorally different from the criminals whom
police deal with on a regular basis (Lankford, 2015; Newman, Fox, Roth, Mehta, &
Harding, 2004). This suggests that law enforcement may be susceptible to some of the
popular misconceptions about those likely to engage in mass murder. For instance, if
they assume that someone intelligent, well educated, polite, articulate, or middle class
could not pose a threat due to these particular traits, they would overlook many of the
types of people who have committed public mass shootings in the past (Lankford,
2018).
In addition, it appears that most law enforcement training on active or mass shoot-
ings focuses on how to rapidly respond to an incident in progress (Advanced Law
Enforcement Rapid Response Training, 2018; ALICE Training Institute, 2018;
Department of Homeland Security, 2018; Federal Law Enforcement Training Centers,
2018), rather than how to identify potential perpetrators in advance. However, there is
a lot of scholarship that might be useful for officers to learn. As just a few examples,
Meloy, Hoffmann, Guldimann, and James (2012) have identified eight key warning
behaviors for threat assessment: (a) pathways, (b) fixations, (c) identifications, (d)
novel aggression, (e) energy burst, (f) leakage, (g) last resort, and (h) directly com-
municated threats. Similarly, Lankford (2018) has provided checklists of warning
signs that law enforcement could use to help determine whether an at-risk individual
appears to have (a) suicidal motives or life indifference that might eliminate fear of
consequences, (b) perceptions of his or her own victimization that might rationalize
320 Journal of Contemporary Criminal Justice 35(3)

attacks against targets, and (c) desires to obtain fame or attention through killing. In
addition, recent work by Geurts, Ask, Granhag, and Vrij (2018) makes important sug-
gestions about law enforcement best practices for interviewing individuals who have
threatened to cause serious harm.
A closer look at mass shooters who were known by law enforcement to be potential
threats, but not stopped from attacking, might shed more light on this subject. Much
may be learned from recent failures to prevent public mass shootings so that these
offenders can be more effectively thwarted in the future.

Mass Shooters’ Acquisition of Firearms

A final factor worth considering is mass shooters’ acquisition of firearms. By defini-
tion, firearms are needed to commit mass shootings, and common sense suggests that
easy access to weapons would increase the likelihood of an attack. There is also empir-
ical evidence that globally, a country’s number of civilian firearms per capita is associ-
ated with the number of public mass shooters who attack there (Lankford, 2016b). The
United States is number one by both measures: It had the most public mass shooters
from 1966 to 2012, and it has more firearms in circulation per civilian than any other
nation (Lankford, 2016b).
Furthermore, multiple data sources indicate that many mass shooters and active
shooters in the United States acquire their weapons legally (Buchanan, Keller, Oppel, &
Victor, 2018; Follman, Aronsen, & Pan, 2018; Silver, Simons, & Craun, 2018). U.S.
federal law currently prohibits people from possessing firearms if they have been invol-
untarily committed, adjudicated to be mentally defective, or met several other disquali-
fying criteria (Giffords Law Center, 2018). However, even though a large proportion of
mass shooters struggle with mental health problems, very few have been involuntarily
committed or adjudicated to be mentally defective (Silver, Fisher, & Horgan, 2018). This
suggests that the current criteria for legal purchase of firearms may be insufficient.
Fortunately, many people on both sides of gun legislation debates in the United
States agree that some high-risk individuals should not be allowed to purchase or pos-
sess firearms. In fact, the National Rifle Association’s spokesperson recently sup-
ported “following up on red flags” and “making sure that people who are dangerous
should not have access to firearms” (Tatum, 2018). Both the President and Vice
President of the United States have also made similar statements (Breuninger, 2018).
However, prior research has not explored how these “red flags” affect access. For
instance, is it primarily those mass shooters who have not engaged in leakage or been
reported to law enforcement who are able to acquire their firearms legally? Or are even
perpetrators who have clearly signaled their homicidal intent nevertheless able to pur-
chase their guns without encountering legal barriers?

The Present Study

Past research on active and public mass shooters has typically prioritized either depth
of analysis or breadth of analysis, but not both. For instance, some studies have focused
Lankford et al. 321

on close examinations of perpetrators and the psychological factors that drove them,
and therefore employed relatively small sample sizes (Larkin, 2009; Mullen, 2004;
Newman et al., 2004; O’Toole, 2000; Vossekuil et al., 2002). Other studies have looked
for patterns in more easily coded elements of their attack behavior—such as offenders’
age, sex, weapon use, attack location, and number of victims harmed—and thus
included hundreds of cases (Kelly, 2012; Lankford, 2015, 2016b).
This study prioritized depth of analysis. Rather than selecting a random sample
from the past half century that would result in cases of varying significance being
analyzed, this study focused on the 15 deadliest incidents that occurred in the United
States over a 20-year period, from March 1998 through February 2018.
The 15 qualifying cases were as follows: the 1999 Columbine school shooting,
2005 Red Lake school shooting, 2007 Virginia Tech shooting, 2009 Binghamton
shootings, 2009 Fort Hood Army base shooting, 2012 Aurora movie theater shooting,
2012 Sandy Hook school shooting, 2013 Washington, DC Navy Yard shooting, 2015
Charleston church shooting, 2015 Umpqua Community College (UCC) shooting,
2015 San Bernardino attack, 2016 Orlando nightclub shooting, 2017 Las Vegas shoot-
ing, 2017 Sutherland Springs church shooting, and 2018 Parkland school shooting.
Because public mass shootings have grown more deadly over time, this sample also
happens to include the five deadliest incidents in U.S. history as of January 2019. (No
names of mass shooters are included in this text, in line with the “No Notoriety” cam-
paign and Lankford and Madfis’s (2018) proposal to deny offenders the attention they
often seek.)

Data Sources and Collection

Information on each case was drawn from official reports and records, primary source
documents, prior scholarship, and media reports, which have been the most commonly
used data sources in prior studies of mass shooters (Huff-Corzine, McCutcheon,
Corzine, Weller, & Landon, 2014). Whenever possible, we relied on official govern-
ment reports and records, but not all mass shootings are followed by retrospective
investigations and public reports of the type that were demanded after Columbine,
Virginia Tech, Fort Hood, Sandy Hook, and Las Vegas, and for some incidents, addi-
tional evidence was released following the publication of an official report. In cases
where we drew some data from media reports, it was almost always information that
the media had received from government or law enforcement officials.
This study focused on three primary concepts: leakage, information reported to law
enforcement, and firearms acquisition. It appears that only one other study has jointly
analyzed all three of these measures: the FBI’s analysis of active shooters in the United
States from January 2000 to December 2013 (Silver, Simons, & Craun, 2018). Our
study differs from the FBI report in several key ways. For instance, our study period
(March 1998-February 2018) is longer and includes more recent cases. In addition, we
focused on a smaller sample of the most lethal perpetrators, whereas the FBI focused
on a larger sample of perpetrators who averaged far fewer victims killed.
322 Journal of Contemporary Criminal Justice 35(3)

We also added several new measures that the FBI study did not analyze. For
instance, we coded for both (a) leakage of violent thoughts/intent in general (which the
FBI included) and (b) leakage of specific interest in mass killing (which the FBI did
not). We believe the latter measure sheds additional light on the severity of leakage by
identifying perpetrators who specifically signaled their interest in extreme violence. If
there was evidence of either type of leakage for a given case, it was coded “yes”; if not,
it was coded “no.”
In addition, we coded each case for (a) whether the mass shooter’s concerning behav-
ior had been reported to law enforcement prior to the attack (which is a variable the FBI
included in their analysis) and (b) whether concerns about the mass shooter’s specific
interest in homicide (i.e., killing people) had been reported to law enforcement (which the
FBI did not include). Although the FBI researchers used a broad definition of “concerning
behaviors” that included leakage, threats, anger, aggression, mental health concerns, and
use of illicit drugs (Silver, Simons, & Craun, 2018), people rarely report these behaviors
to law enforcement unless they are perceived to be very serious—and even then, they may
not report. By adding a variable for whether the perpetrator’s specific interest in killing
people had been reported, we hope to better understand what information was known by
law enforcement prior to each attack. Each case was coded as “yes” if there was evidence
that this type of information had been reported, and “no” if there was not.
Firearms were coded as being acquired illegally only if the offender stole them or
convinced someone else to commit a crime to obtain them. If the shooter purchased a
firearm “over the counter” just like any other civilian could, that was classified as
legal, even if there were postattack questions about whether some basis could have
existed for prohibiting the purchase. This appears consistent with the FBI report’s cod-
ing method for which perpetrators obtained firearms without stealing them or purchas-
ing them illegally (Silver, Simons, & Craun, 2018). In the FBI’s sample of active
shooters, the legal status of some firearms was unknown; in our sample, that informa-
tion was known for all cases.
Finally, another important difference between our study and the FBI report is that
we provide details on each individual case and how it was coded, rather than publish-
ing only aggregate-level results.

Analysis
In the text of this article, we summarize the leakage, information reported to law
enforcement, and firearms acquisition. Whenever possible, we tried to include direct
quotes from the perpetrators, so readers will be able to see for themselves exactly what
the mass shooter said or wrote. In Table 1, we provide aggregate results from the 15
incidents, and in the appendix, we list how all cases were coded. Our hope is that by
providing this level of detail, we offer a resource for other scholars and practitioners
while also assisting any researchers interested in replicating or extending this work.
We also compared our findings on the deadliest public mass shooters with the FBI’s
findings on active shooters in general (Silver, Simons, & Craun, 2018). Previous
research has shown that mass/active shooters who kill many victims appear to differ
Lankford et al. 323

Table 1.  Frequency of Leakage, Information Reported to Law Enforcement, and Legal
Firearms Acquisition for the Deadliest Public Mass Shooters in the United States.

Factor Frequency (%)

Leakage of violent thoughts/intent 87
Leakage of specific interest in mass killing 80
Concerning behavior was reported to law enforcement 80
Concerning interest in homicide was reported to law enforcement 47
All firearms used in the mass shooting were acquired legally 80

Note. Sample included perpetrators of the 15 deadliest incidents from March 1998 to February 2018.
Firearms were coded as being acquired illegally if the offender stole the weapon or convinced someone
else to commit a crime to obtain the weapon.

from those who kill fewer victims in several ways. For instance, deadlier offenders
appear more likely to seek fame and more likely to commit suicide or “suicide by cop”
(Kelly, 2012; Lankford, 2015, 2016a). Our comparison will help clarify whether there
are also differences regarding their leakage, information reported to law enforcement,
and firearms acquisition.
For each variable, we provide our findings on the deadliest public mass shooters,
the FBI’s findings on active shooters in general, and the FBI’s findings on active
shooters with potentially overlapping cases from the current study removed. The seven
potentially overlapping cases were included in the FBI’s initial report on active shoot-
ings from 2000 to 2013 (Blair & Schweit, 2014) and were the 2005 Red Lake school
shooting, 2007 Virginia Tech shooting, 2009 Binghamton shootings, 2009 Fort Hood
shooting, 2012 Aurora theater shooting, 2012 Sandy Hook school shooting, and 2013
Washington, DC Navy Yard shooting.

Results
As presented in Table 1, this study found that the deadliest public mass shooters com-
monly displayed serious warning signs prior to their attacks. In 87% of the studied cases,
the offenders had previously told others that they had violent thoughts or intentions, and
in 80% of cases, there was leakage of perpetrators’ specific interest in mass killings.
Furthermore, before 80% of incidents, the perpetrators’ concerning behavior had been
reported by law enforcement, and before nearly half of incidents (47%), law enforce-
ment had been informed of the offender’s specific interest in homicide or mass killings.
Despite the frequency of these warning signs, however, 80% of the public mass shooters
in this study were able to acquire all of the firearms for their attack legally.
Below, more details are provided on each case.

1999 Columbine School Shooting

Leakage prior to attack. One offender posted online threats that included the
assertion,
324 Journal of Contemporary Criminal Justice 35(3)

I’m coming for EVERYONE soon and I will be armed to the fucking teeth and I WILL
shoot to kill . . . All I want to do is kill and injure as many of you . . . as I can.

The other offender submitted a creative writing essay about a fictional mass shooting
of “preps” that ended with the assertion, “I understood his actions” (Columbine
Review Commission, 2001; Cullen, 2009).

Concerning behavior reported to law enforcement.  Law enforcement was shown evi-
dence from one offender’s website that he threatened to kill others, including a class-
mate, and had acquired weapons (Columbine Review Commission, 2001). Law
enforcement had made contact with the offenders at least 15 times prior to the school
shooting (Office of the Attorney General, 2004).

Firearms.  The offenders legally received two shotguns and a rifle from a friend who
had purchased them at a gun show on request. They also illegally bought handguns
from two acquaintances (Columbine Review Commission, 2001).

2005 Red Lake School Shooting

Leakage prior to attack.  The offender was expelled a year earlier after threatening to
“shoot up the school” on the anniversary of Columbine. He had also emailed several
friends about his intent to commit a Columbine-type shooting at Red Lake (Meloy &
O’Toole, 2011). In an online forum, he repeatedly posted about mass shootings,
including an animated film he created that ended in the shooter’s suicide. He also drew
pictures of “bloodied bodies and guns” and apparently shared them with classmates,
along with a story about a mass shooting at a school (Davey & Wilgoren, 2005). At
one point, he admitted that things were “kind of rocky right now so I might disappear
unexpectedly.” The offender had also previously been hospitalized for suicidal behav-
ior (Davey & Wilgoren, 2005).

Concerning behavior reported to law enforcement. The offender’s grandfather was a

police officer in the small community of Red Lake, so would have known about the
offender’s threats that led to his expulsion and hospitalization. It was also widely
known that years earlier, the offender’s father had fatally shot himself in the chest dur-
ing a standoff with police (Davey & Wilgoren, 2005).

Firearms.  The offender killed his grandfather and then stole his grandfather’s firearms
to commit the school shooting (Langman, 2016b).

2007 Virginia Tech Shooting

Leakage prior to attack.  Years earlier, the offender had expressed interest in suicide
and homicide in a school writing assignment and suggested that “he wanted to repeat
Columbine.” In college, he gave a class presentation that referred to fellow students
Lankford et al. 325

as “despicable human beings who are all disgraces to [the] human race” and con-
cluded that “I hope y’all burn in hell.” In 2005, he was court-ordered to seek outpa-
tient care at a mental health facility after telling a roommate, “I might as well kill
myself.” He also submitted multiple creative writing assignments to teachers that
focused on mass killing, including a story in which his main character plans a mass
shooting “to kill every god damn person in this damn school” (Virginia Tech Review
Panel, 2009).

Concerning behavior reported to law enforcement.  The offender was reported to campus
police multiple times—for stalking female students, for his antisocial behavior in
class, and for his suicidal remarks—but law enforcement was apparently not given his
disturbing writing assignments that signaled his interest in committing a mass shoot-
ing (Virginia Tech Review Panel, 2009).

Firearms.  Legally purchased from a gun store and a pawnbroker, although his having
been court-ordered to receive mental health treatment could have prohibited the pur-
chase (Virginia Tech Review Panel, 2009).

2009 Binghamton Shootings

Leakage prior to attack.  The offender had told a coworker that he would like to shoot
the President. The coworker took the threat seriously and said he would report it to the
FBI, but changed his mind after the offender said he had been joking (Rivera & Schw-
erber, 2009).

Concerning behavior reported to law enforcement.  Years earlier, law enforcement had
received a tip that he was planning a bank robbery (that he never committed) and that
he had a crack cocaine addiction (Rivera & Schwerber, 2009). He also had another
unspecified “criminal incident” in his past and had at least five contacts with law
enforcement since 1990 (Buchanan et al., 2018).

Firearms.  Legally purchased from a gun store (Buchanan et al., 2018).

2009 Fort Hood Army Base Shooting

Leakage prior to attack.  Despite enlisting in the U.S. military, the offender told peers
he considered Sharia law a higher calling than the U.S. constitution. During his fellow-
ship at an Army medical center, he openly expressed support for violent Islamic
attacks in class presentations and written assignments. He made repeated claims that
oppression of Muslims justifies suicide bombings and that suicide bombers are
rewarded in heaven. He also posted online that suicide attacks that kill enemy soldiers
should be considered heroic. Two military officers described him as a “ticking time
bomb” (U.S. Senate Committee on Homeland Security and Governmental Affairs,
2011; Webster Commission, 2012).
326 Journal of Contemporary Criminal Justice 35(3)

Concerning behavior reported to law enforcement.  The FBI obtained a series of emails
between the offender and a foreign Islamic extremist in which he suggested that sui-
cide attacks that kill innocent civilians are heroic and that “Muslims in the U.S. mili-
tary” who have “killed or tried to kill other U.S. soldiers in the name of Islam” should
be considered holy martyrs. However, the officials believed the offender’s communi-
cations were not a matter of concern (U.S. Senate Committee on Homeland Security
and Governmental Affairs, 2011; Webster Commission, 2012).

Firearms.  Legally purchased from a gun store (Buchanan et al., 2018).

2012 Aurora Movie Theater Shooting

Leakage prior to attack.  In an online chat with a female friend, the offender said he
wanted to “kill people of course” and suggested the way to avoid prison is to “kill
many people.” He also had conversations with a classmate about wanting to kill peo-
ple (“Prosecutors,” 2012). In addition, he told one psychiatrist that he would like to
kill people and thinks about it often. When asked, he refused to provide details on his
plans because “you’d lock me up.” He similarly informed another psychiatrist that he
had a Plan A of suicide and a Plan B of mass murder-suicide. This psychiatrist believed
the offender was careful in his statements and probably “filtering a lot [of his answers]”
to avoid getting in trouble (Metzner, 2013).

Concerning behavior reported to law enforcement.  The university’s police department

was contacted by one of the offender’s psychiatrists, who expressed concerns and
inquired whether he had a criminal history. However, the offender dropped out of
school before much further action was taken (Associated Press, 2015).

Firearms.  Legally purchased from gun stores (Buchanan et al., 2018).

2012 Sandy Hook School Shooting

Leakage prior to attack.  According to FBI reports, the offender had made threats to kill
his mother and some children at Sandy Hook, which were overheard and reported by
another person (Associated Press, 2017).1 In addition, a fellow video game player
knew that the offender kept a spreadsheet with details on past mass murders and kill-
ing sprees, and that he was “‘depressed’ and ‘isolated’ and someone who viewed death
as an ‘escape from his joyless existence’” (Associated Press, 2017). The offender also
made many posts in an online forum of people who were fascinated by school shoot-
ings and mass shootings.

Concerning behavior reported to law enforcement.  Police were notified of the offender’s
threats, but claimed there was nothing they could do and that the state police should be
notified instead (Associated Press, 2017).
Lankford et al. 327

Firearms.  The offender broke the law when he stole his mother’s firearms and used
them to kill her and commit the school shooting (Langman, 2016b).

2013 Washington, DC Navy Yard Shooting

Leakage prior to attack.  No evidence of leakage pertaining to violent intentions has
been found; however, the offender reportedly told friends that “he suffered from
depression and post-traumatic stress disorder” (Goode, Nir, & Fernandez, 2013).

Concerning behavior reported to law enforcement.  Years earlier, he explained to police

that he had shot out someone’s tires after experiencing a “blackout fueled by anger,”
and he was arrested and charged with a felony. In a separate incident, police believed
he slashed the tires of five vehicles near his apartment. He was also arrested for ille-
gally discharging a firearm in public (Department of the Navy, 2013). However, in all
of these cases, criminal charges were dropped. More recently, police were called by a
hotel clerk who feared that the offender would hurt someone and found the offender
paranoid and delusional. He had “taken apart his bed, believing someone was hiding
under it,” “taped a microphone to the ceiling to record the voices of people that were
following him,” and complained about a computer chip in his head and being con-
trolled by microwave signals. Police contacted the U.S. Navy to report their concerns:
“FYI on this . . . Just thought to pass it on to you in the event this person escalates”
(Department of the Navy, 2013).

Firearms.  Legally purchased a shotgun from a gun store after being prohibited from
purchasing an assault rifle because he was an out-of-state resident (Buchanan et al.,
2018; Department of the Navy, 2013).

2015 Charleston Church Shooting

Leakage prior to attack.  The offender told a neighbor that “he wanted to open fire on a
school” or shoot up a local college (Reuters, 2015). He later told a friend “that he
wanted to kill black people at a historic African Methodist Episcopal church in
Charleston in order to start a race riot,” that he had been planning his attack for 6
months, and that he would kill himself afterward (Dixon & Sack, 2017). He also
posted a manifesto online, stating that he had “no choice” but to commit an attack
against black people.

Concerning behavior reported to law enforcement.  The offender had been arrested for
drug possession, banned from the shopping mall where the arrest occurred, and then
rearrested 2 months later for trespassing (Buchanan et al., 2018).

Firearms.  Legally purchased from a gun store, although his drug possession charge
could have prohibited the purchase (Buchanan et al., 2018).
328 Journal of Contemporary Criminal Justice 35(3)

2015 UCC Shooting

Leakage prior to attack.  The offender had been discharged from the Army after a
suicide attempt, had once threatened his mother with a shotgun, and had shown his
mother videos he watched of people being killed. His mother said he was “angry at
the world” and “said she thought about calling police. But she didn’t want him to go
to jail . . . worrying that an arrest would make a life already full of challenges all the
more difficult” (Schmidt, 2017). One month before his attack, he posted online
about his excitement from seeing a video of the 2015 Roanoke shooter killing a
journalist and cameraman: “if anyone gets the chance, go on youtube and see the
footage of him shooting those people. It’s a short video but good nonetheless”
(Langman, 2016a).

Concerning behavior reported to law enforcement.  No evidence found.

Firearms.  Legally purchased through a federally licensed firearms dealer (Buchanan

et al., 2018).

2015 San Bernardino Attack

Leakage prior to attack.  The male offender had made previous plans with a neighbor
“to launch deadly attacks on the college they had attended and on a busy California
freeway,” and the neighbor had purchased weapons for them before deciding not to
participate (Nagourney, Perez-Peña, & Lovett, 2015).

Concerning behavior reported to law enforcement.  No evidence found.

Firearms. Legally purchased at gun stores by the male offender and his neighbor
(Buchanan et al., 2018).

2016 Orlando Nightclub Shooting

Leakage prior to attack.  The offender told coworkers he had connections to Al Qaeda
and Hezbollah and “that he wanted to die as a martyr.” He told a member of his
mosque that he had been watching extremist sermons online. He also watched violent
terrorist recruiting videos in front of his wife and asked her, “How bad would it be if
a club was attacked?” (Lotan, 2017; Wilber, 2016).

Concerning behavior reported to law enforcement.  The FBI had investigated the offender
for 10 months and interviewed him multiple times after coworkers and a member of
his mosque reported their concerns. However, the FBI found that the offender had no
terrorist connections and limited ideological knowledge. The offender said he lied
about his terrorist connections to scare coworkers and denied watching extremist vid-
eos (Wilber, 2016).
Lankford et al. 329

Firearms.  Legally purchased at gun stores (Buchanan et al., 2018).

2017 Las Vegas Shooting

Leakage prior to attack.  No evidence of leakage pertaining to violent intentions has
been found, and his girlfriend claims that he told her nothing about his thoughts or
plans. However, he “made numerous claims to friends and family that he consistently
felt ill, in pain or fatigued.” His doctor thought he may have bipolar disorder, but the
offender refused to discuss the possibility. He declined a prescription for antidepres-
sants but accepted one for anxiety medication (Las Vegas Metropolitan Police Depart-
ment, 2018).

Concerning behavior reported to law enforcement.  No evidence found.

Firearms.  Legally purchased at gun stores and from licensed firearms dealers (Las
Vegas Metropolitan Police Department, 2018).

2017 Sutherland Springs Church Shooting

Leakage prior to attack.  According to a coworker, the offender “displayed a fascination
with mass murders” and was inspired by the 2015 Charleston church shooting. “He
would say ‘isn’t it cool? Did you watch the news?’ . . . He would say he wished he had
the nerve to do it, but all he would be able to do is kill animals.” The coworker warned
supervisors “back off or he would shoot the place up.” She also recounted that the
offender “threatened to kill himself one time if I didn’t let him go see a chaplain” (Kra-
varik, 2017). In addition, the offender showed interest in mass shootings on social media
and sent his mother-in-law threatening text messages up to the day of the shooting.

Concerning behavior reported to law enforcement.  The offender was court-martialed and
convicted for assaulting his wife and breaking his infant stepson’s skull while in the
Air Force and was eventually discharged for “bad conduct.” He was also charged with
a misdemeanor count of animal cruelty for repeatedly punching a dog (Buchanan
et al., 2018).

Firearms.  Legally purchased from gun stores, although his domestic violence convic-
tion could have prohibited the purchase (Buchanan et al., 2018).

2018 Parkland School Shooting

Leakage prior to attack.  The offender introduced himself to an acquaintance by saying,
“Hi . . . I’m a school shooter.” On social media he responded to a mass shooting in
New York by posting, “Man I can do so much better.” In other online posts, he stated,
“I wanna shoot people with my AR-15” and “I’m going to be a professional school
shooter” (Griffin, Glover, Pagliery, & Lah, 2018; “Red flags,” 2018).
330 Journal of Contemporary Criminal Justice 35(3)

Concerning behavior reported to law enforcement.  Both the FBI and local law enforce-
ment were contacted by neighbors, relatives, and other individuals who were con-
cerned that the offender was suicidal, depressed, self-destructive, angry, violent, a
threat who “could be a school shooter in the making,” and someone who showed a
“desire to kill people.” Neighbors reported the offender would sometimes point a gun
at their window. Law enforcement officers had responded to calls from his home many
times in previous years and were specifically told that the offender was in possession
of firearms, although he might try to hide them (Griffin et al., 2018; “Red flags,”
2018).

Firearms.  Legally purchased at a gun store (Buchanan et al., 2018).

Comparisons Between the Deadliest Public Mass Shooters and Other

Active Shooters
As shown in Table 2, beyond its case-by-case analyses, this study also found that per-
petrators of the deadliest public mass shootings appeared quite different from active
shooters who killed fewer victims. Once potentially overlapping cases from the FBI
report were removed, we found that the deadliest mass shooters engaged in leakage
about their violent thoughts/intentions far more often (87%) than other active shooters
(52%). In fact, the most severe form of leakage (signaling specific interest in mass
killing) was more common among the deadliest mass shooters (80%) than even the
baseline level of leakage (signaling violent thoughts/intentions) was among other
active shooters (52%).
In addition, the deadliest perpetrators’ concerning behavior had been reported to
law enforcement prior to the attack much more frequently (80%) than other active
shooters’ concerning behavior had been reported (34%). In fact, for the deadliest mass
shooters, concerns about their specific interest in homicide or mass killing had been
reported to law enforcement more often (47%) than any form of concerning behavior
had been reported for other active shooters (34%).
Finally, there was a small difference in the proportion of perpetrators who were able
to acquire all of their firearms legally: 80% of the deadliest mass shooters and 93% of
other active shooters met this criterion.

Discussion
Lesson 1: Prevention Appears Possible in the Vast Majority of Cases
As this study’s findings have shown, the vast majority of the deadliest public
mass shootings in the United States over the past two decades appear to have been
preventable, based on the information known about offenders in advance. This
assertion is consistent with the FBI’s conclusions in their aforementioned report
(Silver, Simons, & Craun, 2018), although the warning signs among their sample
of active shooters were far less common than those preceding the deadliest
Lankford et al. 331

Table 2.  Comparisons Between the Deadliest Public Mass Shooters and Active Shooters
in the United States According to Leakage, Information Reported to Law Enforcement, and
Firearms Acquisition.

FBI study’s findings on

Current study’s active shooters, 2000-
findings on the FBI study’s 2013, with potentially
deadliest public findings on active overlapping cases from
mass shooters, shooters, 2000- the current study
Factor 1999-2018 (%) 2013 (%) removed (%)
Leakage of violent thoughts/ 87 56 52
intent
Leakage of specific interest in 80 No data collected No data collected
mass killing
Concerning behavior 80 41 34
was reported to law
enforcement
Concerning interest in 47 No data collected No data collected
homicide was reported to
law enforcement
All firearms used in the mass 80 90 93
shooting were acquired
legally (% of known cases)

Note. The current study’s sample included perpetrators of the 15 deadliest incidents from March
1998 to February 2018. The FBI study’s sample included 63 active shooters from January 2000
to December 2013 and was authored by Silver, Simons, and Craun (2018). The seven potentially
overlapping cases all appeared in the FBI’s initial report on active shootings from 2000 to 2013
(Blair & Schweit, 2014) and were the 2005 Red Lake school shooting, 2007 Virginia Tech shooting,
2009 Binghamton shootings, 2009 Fort Hood shooting, 2012 Aurora theater shooting, 2012 Sandy
Hook school shooting, and 2013 Washington, DC Navy Yard shooting. FBI = Federal Bureau of
Investigation.

incidents studied here. As those researchers summarized, “There is cause for

hope because there is something that can be done . . . these observable behaviors
may represent critical opportunities for detection and disruption” (Silver, Simons,
& Craun, 2018, pp. 6, 27).
This hope is even more warranted in the deadliest cases because leakage is so much
more common than among active shooters who kill fewer victims. One potential
explanation for this difference may be the longer planning periods that deadlier mass
shooters typically engage in (Bondü & Scheithauer, 2015). While only approximately
18% of active shooters in the FBI sample planned their attacks for more than 6 months
(Silver, Simons, & Craun, 2018), nearly all of the deadliest public mass shooters in
this study planned for that long or longer. In turn, longer planning times may be associ-
ated with more leakage from potential perpetrators, along with more opportunities to
prevent their attacks.
332 Journal of Contemporary Criminal Justice 35(3)

Even in the unusual cases where the offenders’ violent intentions were apparently
unknown (the DC Navy Yard and Las Vegas shootings), the individuals had admitted
struggling with mental health problems and shown other reasons for concern. The
Navy Yard shooter exhibited signs of schizophrenia, told police he had a “blackout
fueled by anger,” and had repeated encounters with law enforcement over his disturb-
ing behavior (Department of the Navy, 2013). The Las Vegas shooter had been com-
plaining about his health, was believed by his doctor to have bipolar disorder, had been
prescribed anti-anxiety medication, had recently purchased bump stocks and a large
number of firearms, and had sent his girlfriend away to another country for no expli-
cable reason (Las Vegas Metropolitan Police Department, 2018). Overall, it seems that
serious warning signs are often observed prior to the deadliest public mass shootings
and are not only visible in hindsight.

Lesson 2: The Public Should be Educated, Encouraged, and Pressured to

Report Leakage and Warning Signs to Law Enforcement
Although most of the mass shooters in this study were reported to law enforcement, it
does not mean that the FBI or police departments knew everything the public did. In
many cases, law enforcement only received a portion of the concerning information
that was known prior to an attack. For example, in the Virginia Tech case, the offender
was reported to campus police multiple times—for stalking female students, for his
antisocial behavior in class, and for his suicidal remarks—but law enforcement was
apparently not given the writing assignments that signaled his interest in committing a
mass shooting (Virginia Tech Review Panel, 2009). As this shows, even though some
members of the public have successfully reported mass shooters to law enforcement in
the past, more progress is needed.
The public needs to be reminded that they are the first line of defense against mass
shootings, and they should be more effectively educated, encouraged, and pressured to
serve as the eyes and ears of law enforcement. Not only do regular citizens outnumber
police officers in the United States by a ratio of more than 300:1—which means they
are often present in situations where law enforcement is not—but they also hear leak-
age, threats, and attack intentions that offenders might not express in front of a police
officer.
Many of this study’s cases highlight the aforementioned challenges of getting peo-
ple to report their own friends and family. As noted earlier, the UCC shooter’s mother
did not report her son because she worried that an arrest would make his life more
difficult (Schmidt, 2017). Similarly, the Red Lake shooter and Charleston church
shooter were not reported by their friends, despite making explicit threats about mass
shootings (Davey & Wilgoren, 2005; Dixon & Sack, 2017; Meloy & O’Toole, 2011;
Reuters, 2015), and the Orlando shooter’s wife did not report him even though he
apparently discussed attacking a nightclub (Lotan, 2017).
Public information campaigns should emphasize that failure to report a potential
mass shooter is not only a violation of one’s public and social responsibility, but it can
Lankford et al. 333

also have serious personal consequences. Far too many families, friends, and cowork-
ers have had to live with traumatic guilt after they saw something, but said nothing. In
addition, people who have heard critical information but failed to report it are increas-
ingly facing lawsuits and criminal charges. Families, schools, and government agen-
cies have all been sued for their inaction, and friends of the 2005 Red Lake shooter and
2015 Charleston church shooter, a neighbor of the 2015 San Bernardino shooters, and
the wife of the 2016 Orlando shooter were each arrested after failing to report clear
signs of those impending attacks (Dixon & Sack, 2017; Lotan, 2017; Meloy & O’Toole,
2011; Wilber, 2016).
Retrospective testimonials should be widely disseminated featuring people who
now regret having not reported their friends or family when it could have prevented a
tragic attack. This would help increase public awareness and could promote social
change. Similarly, every effort should be made to foster the kind of positive climates
and police/community relationships which encourage reporting behavior (see, for
example, Bobo & Thompson, 2006; Desmond, Papachristos, & Kirk, 2016; Kidd &
Chayet, 1984; Madfis, 2014b).

Lesson 3: Law Enforcement Needs to be Educated and Trained on How

to More Effectively Investigate Potential Threats
Although in some of the deadliest mass shootings the public failed to report all con-
cerning information to the authorities, there were also some cases where law enforce-
ment failed to effectively investigate potential threats.
For example, although a psychiatrist for the 2012 Aurora movie theater shooter
contacted both the university’s police department and its behavioral evaluation and
threat assessment team with her concerns, no one appeared to push for more action
beyond checking to see whether this individual had a criminal record and limiting his
access to campus (Associated Press, 2015). If police had looked closer into his case,
they might have discovered that he exhibited many of the key warning signs outlined
by Meloy et al. (2012) and Lankford (2018). In turn, this could have justified further
investigations that would have revealed he was acquiring weapons, building home-
made explosives, and running surveillance on potential targets (O’Neill & Weisfeldt,
2015), all of which would warrant significant cause for concern when evaluated from
a threat assessment perspective (Madfis, 2014b; O’Toole, 2000; White, 2017). This
information could have been used to prevent his eventual attack.
Law enforcement officers should be better educated and trained on identifying the
common warning signs in offenders’ lives and evaluating potential threats. Because
not all threats are equally dangerous, the threat assessment approach (Cornell &
Sheras, 2006; O’Toole, 2000) may offer great utility in distinguishing the characteris-
tics of the most serious and potentially devastating threats from those which may
never result in violence of any kind. These forms of knowledge could give law enforce-
ment a more scientific basis for their investigations, so they would know what types of
evidence are most important to look for and what types of questions are most
334 Journal of Contemporary Criminal Justice 35(3)

important to ask. As the aforementioned FBI report states, “a shared awareness of the
common observable behaviors demonstrated by the active shooters . . . may help to
prompt inquiries and focus assessments at every level of contact and every stage of
intervention” (Silver, Simons, & Craun, 2018, p. 27).

Lesson 4: People Who Have Admitted Homicidal or Suicidal Thoughts

or Interest in Committing Mass Shootings Should be Prohibited From
Purchasing or Possessing Firearms
As noted earlier, federal law prohibits people from possessing firearms if they have
been involuntarily committed, adjudicated to be mentally defective, or met several
other disqualifying criteria (Giffords Law Center, 2018). However, as this study has
shown, these standards are insufficient, because most of the public mass shooters
examined here were able to obtain their weapons legally despite engaging in leakage
and having their concerning behavior or interest in homicide reported to law
enforcement.
Looking forward, the best legal mechanism for keeping firearms out of their hands
might be risk protection orders (also known as “gun violence restraining orders” or
“gun violence protective orders”), which have been recently adopted in several states.
As the Giffords Law Center (2018) summarizes, these orders “allow families, house-
hold members, or law enforcement officers to petition a court directly for an order
temporarily restricting a person’s access to guns.”
This study’s findings provide further clarity on which warning signs should be
considered when making the determination of risk. More specifically, if people who
engaged in the most extreme form of leakage—admitting their homicidal thoughts or
interest in committing a mass killing—were prevented from acquiring firearms, many
of the deadliest incidents of the past 20 years may have been avoided. It seems hard to
imagine that most Americans would argue that people have a right to possess firearms
after they have made such overtly concerning statements about harming themselves or
others.
From the perspective of both members of the public who are considering reporting
leakage and law enforcement officers who are investigating potentially dangerous
people, keeping at-risk individuals from obtaining weapons may stop them from doing
something terrible, without requiring them to be imprisoned or institutionalized. False
positives will continue to be a challenge, but at least people who are inaccurately per-
ceived to be threats by those “erring on the side of caution” would only lose their
firearms, not their freedom.

Conclusion
Based on its analysis of the deadliest public mass shooters in the United States over a
20-year period, this study has shown that such offenders often admit that they have
homicidal thoughts or intentions, including specific interest in committing a mass
Lankford et al. 335

killing. It has also offered a series of recommendations for how this information could
be used to improve policy and prevention strategies.
However, the frequency of these types of leakage among the general population
remains unknown. For example, if someone says “I wanna shoot people with my
AR-15” and “I’m going to be a professional school shooter,” as the 2018 Parkland
shooter did (Griffin et al., 2018; “Red flags,” 2018), does that mean he is extremely
likely to follow through on such threats or only moderately likely? Future studies of
people who made similar statements but did not harm anyone could enhance scholarly
understandings of these warning signs and their predictive power.
In addition, more research is needed on the cases where public mass shootings have
been successfully thwarted (Madfis, 2018). Those cases garner far less attention than
the deadly incidents examined in this study, but they could be quite instructive. Perhaps
the best practices from those cases could provide a template for national policy and
prevention efforts.
Finally, it must be acknowledged that even when warning signs are reported by the
public and taken seriously by law enforcement, it does not guarantee that a potential
offender will be stopped. More research should be conducted on how to turn informa-
tion into action so that the prevalence of these lethal incidents can be significantly
reduced.

Appendix
Incident Coding for Leakage, Information Reported to Law Enforcement, and Legal Firearms
Acquisition for the Deadliest Public Mass Shooters in the United States.
Concerning Concerning
Leakage Leakage behavior was interest in All firearms
of violent of specific reported homicide was used in the mass
thoughts/ interest in to law reported to law shooting were
Incident intent mass killing enforcement enforcement acquired legally

1999 Columbine school Yes Yes Yes Yes No

shooting
2005 Red Lake school Yes Yes Yes Yes No
shooting
2007 Virginia Tech Yes Yes Yes No Yes
shooting
2009 Binghamton Yes No Yes No Yes
shootings
2009 Ft. Hood shooting Yes Yes Yes Yes Yes
2012 Aurora theater Yes Yes Yes Yes Yes
shooting
2012 Sandy Hook Yes Yes Yes Yes No
school shooting
2013 Washington, DC No No Yes No Yes
Navy Yard shooting
2015 Charleston Yes Yes Yes No Yes
church shooting

(continued)
336 Journal of Contemporary Criminal Justice 35(3)

Appendix (continued)
Concerning Concerning
Leakage Leakage behavior was interest in All firearms
of violent of specific reported homicide was used in the mass
thoughts/ interest in to law reported to law shooting were
Incident intent mass killing enforcement enforcement acquired legally

2015 Umpqua Yes Yes No No Yes

Community College
shooting
2015 San Bernardino Yes Yes No No Yes
attack
2016 Orlando nightclub Yes Yes Yes Yes Yes
shooting
2017 Las Vegas No No No No Yes
shooting
2017 Sutherland Yes Yes Yes No Yes
Springs church
shooting
2018 Parkland school Yes Yes Yes Yes Yes
shooting
Total 87% 80% 80% 47% 80%

Note. Sample included perpetrators of the 15 deadliest incidents from March 1998 to February 2018. A description of the
leakage and information reported to law enforcement for each incident appears in the text. Firearms were coded as being
acquired illegally if the offender stole the weapon or convinced someone else to commit a crime to obtain the weapon.

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

Note
1. This information was not included in 2013 “Report of the State’s Attorney for the Judicial
District of Danbury” or the 2014 “Report of the Office of the Child Advocate,” and the
Federal Bureau of Investigation did not release it publicly until 2017.

References
Advanced Law Enforcement Rapid Response Training. (2018). Course catalog. Retrieved from
https://alerrt.org/Course-Catalog
ALICE Training Institute. (2018). ALICE training police/LE program. Retrieved from https://
www.alicetraining.com/our-program/alice-training/law-enforcement/
Associated Press. (2015, September 2). Psychiatrist, campus police officer knew James Holmes
wanted to kill. USA Today. Retrieved from https://www.usatoday.com/story/news/nation/
2015/09/02/james-holmes-university-records/71603148/
Lankford et al. 337

Associated Press. (2017, October 24). FBI: Police were told Sandy Hook shooter threatened
to kill mom, students. CBS. Retrieved from https://www.cbsnews.com/news/sandy-hook-
shooting-investigation-fbi-documents/
Blair, J. P., & Schweit, K. W. (2014). A study of active shooter incidents, 2000-2013. Washington,
DC: Texas State University and Federal Bureau of Investigation, U.S. Department of Justice.
Bobo, L., & Thompson, V. (2006). Unfair by design: The war on drugs, race, and the legitimacy
of the criminal justice system. Social Research, 73, 445-472.
Bondü, R., & Scheithauer, H. (2010). Explaining and preventing school shootings: Chances and
difficulties of control. In W. Heitmeyer, H. Haupt, A. Kirschner, & S. Malthaner (Eds.),
Control of violence: Historical and international perspectives on violence in modern soci-
eties (pp. 295-314). New York, NY: Springer.
Bondü, R., & Scheithauer, H. (2015). Kill one or kill them all? Differences between single and
multiple victim school attacks. European Journal of Criminology, 12, 277-299.
Breuninger, K. (2018, February 28). Trump: Take guns “early” without going to court first.
CNBC. Retrieved from https://cnb.cx/2FG9tqT
Buchanan, L., Keller, J., Oppel, R., & Victor, D. (2018, February 16). How they got their guns.
The New York Times. Retrieved from https://nyti.ms/2nutvOd
Columbine Review Commission. (2001). The report of governor Bill Owens’ Columbine
Review Commission. Retrieved from http://hermes.cde.state.co.us/drupal/islandora/object/
co%3A2067
Cornell, D. G., & Sheras, P. L. (2006). Guidelines for responding to student threats of violence.
Longmont, CO: Sopris West.
Cullen, D. (2009). Columbine. New York, NY: Twelve.
Davey, M., & Wilgoren, J. (2005, March 24). Signs of danger were missed in troubled teen-
ager’s life. The New York Times. Retrieved from https://nyti.ms/2lTQ8cB
Department of Homeland Security. (2018). Active shooter preparedness. Retrieved from https://
www.dhs.gov/active-shooter-preparedness
Department of the Navy. (2013). Investigation into the fatal shooting incident at the Washington
Navy Yard on September 16, 2013 and associated security, personnel, and contracting poli-
cies and practices. Retrieved from https://archive.defense.gov/pubs/Navy-Investigation-
into-the-WNY-Shooting_final-report.pdf
Desmond, M., Papachristos, A., & Kirk, D. (2016). Police violence and citizen crime reporting
in the black community. American Sociological Review, 81, 857-876.
Dixon, C., & Sack, K. (2017, March 21). A friend lied about Dylann Roof’s massacre plan. Now
he’ll go to prison. The New York Times. Retrieved from https://nyti.ms/2nxd5ES
Federal Law Enforcement Training Centers. (2018). Active shooter threat instructor training
program. Retrieved from https://www.fletc.gov/training-program/active-shooter-threat-
instructor-training-program
Follman, M., Aronsen, G., & Pan, D. (2018, February 23). U.S. mass shootings, 1982-2018:
Data from Mother Jones’ investigation. Mother Jones. Retrieved from https://www.moth-
erjones.com/politics/2012/12/mass-shootings-mother-jones-full-data/
Fox, J. A., & Levin, J. (2015a). Extreme killing: Understanding serial and mass murder.
Thousand Oaks, CA: Sage.
Fox, J. A., & Levin, J. (2015b). Mass confusion concerning mass murder. The Criminologist,
40, 8-11.
Geurts, R., Ask, K., Granhag, P. A., & Vrij, A. (2018). Interviewing to manage threats: Exploring
the effects of interview style on information gain and threateners’ counter-interview strate-
gies. Journal of Threat Assessment and Management, 5, 189-204.
338 Journal of Contemporary Criminal Justice 35(3)

Giffords Law Center. (2018). Extreme risk protection orders. Retrieved from https://lawcenter.
giffords.org/gun-laws/policy-areas/who-can-have-a-gun/extreme-risk-protection-orders/
Goode, E., Nir, S., & Fernandez, M. (2013, September 19). Signs of distress multiplied on
killer’s path to Navy Yard. The New York Times. Retrieved from https://nyti.ms/2Ga2Koe
Griffin, D., Glover, S., Pagliery, J., & Lah, K. (2018, February 16). From “broken child” to mass
killer. Cable News Network. Retrieved from https://www.cnn.com/2018/02/16/us/shooter-
profile-invs/index.html
Hempel, A. G., Meloy, J. R., & Richards, T. C. (1999). Offender and offense characteristics of
a nonrandom sample of mass murderers. Journal of the American Academy of Psychiatry
and the Law, 27, 213-225.
Hoffmann, J. (2017). The Virginia Tech massacre as a starting point for threat assessment pro-
grams in European universities. Journal of Threat Assessment and Management, 4, 112-117.
Huff-Corzine, L., McCutcheon, J. C., Corzine, J., Weller, M., & Landon, M. (2014). Shooting
for accuracy: Comparing data sources on mass murder. Homicide Studies, 18, 105-124.
Kelly, R. (2012). Active shooter: Recommendations and analysis for risk mitigation. New York:
New York City Police Department.
Kidd, R. F., & Chayet, E. (1984). Why do victims fail to report? The psychology of criminal
victimization. Journal of Social Issues, 40, 39-50.
Kravarik, J. (2017, November 9). Texas shooter claimed to buy animals on Craigslist for “tar-
get practice,” former colleague says. Cable News Network. Retrieved from http://cnn.
it/2AnScPg
Langman, P. (2016a). Chris Harper-Mercer’s blog. Retrieved from https://schoolshooters.info/
sites/default/files/harper-mercer_blog_1.0.pdf
Langman, P. (2016b). The origins of firearms used in school shootings in the United States.
Retrieved from https://schoolshooters.info/sites/default/files/origins_of_firearms_1.1.pdf
Lankford, A. (2015). Mass shooters in the USA, 1966–2010: Differences between attackers
who live and die. Justice Quarterly, 32, 360-379.
Lankford, A. (2016a). Fame-seeking rampage shooters: Initial findings and empirical predic-
tions. Aggression and Violent Behavior, 27, 122-129.
Lankford, A. (2016b). Public mass shooters and firearms: A cross-national study of 171 coun-
tries. Violence and Victims, 31, 187-199.
Lankford, A. (2018). Identifying potential mass shooters and suicide terrorists with warning
signs of suicide, perceived victimization, and desires for attention or fame. Journal of
Personality Assessment, 100, 471-482.
Lankford, A., & Madfis, E. (2018). Don’t name them, don’t show them, but report everything
else: A pragmatic proposal for denying mass killers the attention they seek and deterring
future offenders. American Behavioral Scientist, 62, 260-279.
Larkin, R. (2009). The Columbine legacy: Rampage shootings as political acts. American
Behavioral Scientist, 52, 1309-1326.
Latané, B., & Darley, J. M. (1969). Bystander “apathy.” American Scientist, 57, 244-268.
Las Vegas Metropolitan Police Department. (2018). LVMPD criminal investigative report of
the 1 October mass casualty shooting. Retrieved from https://www.lvmpd.com/en-us/
Documents/1-October-FIT-Criminal-Investigative-Report-FINAL_080318.pdf
Leuschner, V., Bondü, R., Schroer-Hippel, M., Panno, J., Neumetzler, K., Fisch, S., . .
.Scheithauer, H. (2011). Prevention of homicidal violence in schools in Germany: The
Berlin Leaking Project and the Networks against School Shootings Project (NETWASS).
New Directions for Youth Development, 129, 61-78.
Lankford et al. 339

Lotan, G. T. (2017, March 10). Judge: Pulse shooter Omar Mateen’s widow, Noor Salman, will
stay in jail. Orlando Sentinel. Retrieved from http://www.orlandosentinel.com/news/pulse-
orlando-nightclub-shooting/os-noor-salman-pulse-details-20170310-story.html
Madfis, E. (2014a). Averting school rampage: Student intervention amid a persistent code of
silence. Youth Violence and Juvenile Justice, 12, 229-249.
Madfis, E. (2014b). The risk of school rampage: Assessing and preventing threats of school
violence. New York, NY: Palgrave Macmillan.
Madfis, E. (2014c). Triple entitlement and homicidal anger: An exploration of the intersectional
identities of American mass murderers. Men and Masculinities, 17, 67-86.
Madfis, E. (2018). Insight from averted mass shootings. In J. Schildkraut (Ed.), Mass shootings
in America: Understanding the debates, causes, and responses (pp. 79-84). Santa Barbara,
CA: Praeger.
Mathes, E. W., & Kahn, A. (1975). Diffusion of responsibility and extreme behavior. Journal of
Personality and Social Psychology, 31, 881-886.
Meloy, J. R., Hempel, A. G., Mohandie, K., Shiva, A. A., & Gray, B. T. (2001). Offender and
offense characteristics of a nonrandom sample of adolescent mass murderers. Journal of the
American Academy of Child and Adolescent Psychiatry, 40, 719-728.
Meloy, J. R., Hoffmann, J., Guldimann, A., & James, D. (2012). The role of warning behaviors
in threat assessment: An exploration and suggested typology. Behavioral Sciences & the
Law, 30, 256-279.
Meloy, J. R., & O’Toole, M. E. (2011). The concept of leakage in threat assessment. Behavioral
Sciences & the Law, 29, 513-527.
Metzner, J. (2013). Case Number 12CR1522: Sanity evaluation [Copy Obtained by the Authors].
Mullen, P. E. (2004). The autogenic (self-generated) massacre. Behavioral Sciences & the Law,
22, 311-323.
Nagourney, A., Perez-Peña, R., & Lovett, I. (2015, December 17). Neighbor of San Bernardino
attackers faces terrorism charges. The New York Times. Retrieved from https://nyti.
ms/2kxhkLY
National Alliance on Mental Health. (2018). Calling 911 and talking with police. Retrieved
from https://www.nami.org/find-support/family-members-and-caregivers/calling-911-and-
talking-with-police
Newman, K. S., Fox, C., Roth, W., Mehta, J., & Harding, D. (2004). Rampage: The social roots
of school shootings. New York, NY: Basic Books.
Nock, M. K., Park, J. M., Finn, C. T., Deliberto, T. L., Dour, H. J., & Banaji, M. R. (2010).
Measuring the suicidal mind: Implicit cognition predicts suicidal behavior. Psychological
Science, 21, 511-517.
Office of the Attorney General. (2004). Report of the investigation into the 1997 directed
report and related matters concerning the Columbine High School shootings in April
1999. Retrieved from https://schoolshooters.info/sites/default/files/1997_1998_colum-
bine_report.pdf
O’Neill, A., & Weisfeldt, S. (2015, June 17). Psychiatrist: Holmes thought 3-4 times a day about
killing. Cable News Network. Retrieved from http://cnn.it/1BhGQvr
O’Toole, M. E. (2000). The school shooter: A threat assessment perspective. Federal Bureau
of Investigation. Retrieved from https://www.fbi.gov/file-repository/stats-services-publica-
tions-school-shooter-school-shooter
340 Journal of Contemporary Criminal Justice 35(3)

Prosecutors: Theater shooting suspect told classmate he wanted to kill people. (2012, August
24). Cable News Network. Retrieved from https://www.cnn.com/2012/08/24/justice/colo-
rado-shooting/index.html
Red flags: The troubled path of accused Parkland shooter Nikolas Cruz. (2018, March 10). The
Washington Post. Retrieved from http://wapo.st/cruztimeline
Resnick, B., & Zarracina, J. (2018, February 22). This cartoon explains why predicting a
mass shooting is impossible. Vox. Retrieved from https://www.vox.com/science-and-
health/2018/2/22/17041080/predict-mass-shooting-warning-sign-research
Reuters. (2015, June 20). Charleston shooter originally planned to attack a college. New
York Post. Retrieved from https://nypost.com/2015/06/20/charleston-shooter-originally-
planned-to-attack-a-college/
Rivera, R., & Schwerber, N. (2009, April 4). Before killings, hints of plans and grievance. The
New York Times. Retrieved from https://nyti.ms/2IUXtl4
Sarteschi, C. (2016). An examination of thwarted mass homicide plots and threateners.
Aggression and Violent Behavior, 30, 88-93.
Schildkraut, J., & Muschert, G. W. (2013). Violent media, guns, and mental illness: The three
ring circus of causal factors for school massacres, as related in media discourse. Fast
Capitalism, 10, 23-43.
Schmidt, B. (2017, September 9). UCC mass shooting: Killer “angry at the world,” mom told
detectives. Oregon Live. Retrieved from http://www.oregonlive.com/pacific-northwest-
news/index.ssf/2017/09/ucc_shooting_shooters_mom.html
Schug, R. A., & Fradella, H. F. (2014). Mental illness and crime. Thousand Oaks, CA: Sage.
Silver, J., Fisher, W., & Horgan, J. (2018). Public mass murderers and federal mental health
background checks. Law & Policy, 40, 133-147.
Silver, J., Horgan, J., & Gill, P. (2018). Foreshadowing targeted violence: Assessing leakage of
intent by public mass murderers. Aggression and Violent Behavior, 38, 94-100.
Silver, J., Simons, A., & Craun, S. (2018). A study of the pre-attack behaviors of active shooters
in the United States between 2000 and 2013. Federal Bureau of Investigation. Retrieved from
https://www.fbi.gov/file-repository/pre-attack-behaviors-of-active-shooters-in-us-2000-2013.
pdf
Singal, J. (2015, October 2). We can’t predict who will commit a mass shooting. Gun control is
the only way out. New York Magazine. Retrieved from https://www.thecut.com/2015/10/
gun-control-is-the-only-way-out.html
Stueve, A., Dash, K., O’Donnell, L., Tehranifar, P., Wilson-Simmons, R., Slaby, R., & Link,
B. (2006). Rethinking the bystander role in school violence prevention. Health Promotion
Practice, 7, 117-124.
Tatum, S. (2018, February 22). NRA spokesperson: “Insane monster” shouldn’t have been able
to get a firearm. Cable News Network. Retrieved from http://cnn.it/2CeT3qg
Tyler, T., & Fagan, J. (2008). Legitimacy and cooperation: Why do people help the police fight
crime in their communities? The Ohio State Journal of Criminal Law, 6, 231-275.
U.S. Senate Committee on Homeland Security and Governmental Affairs. (2011). A ticking
time bomb: Counterterrorism lessons from the U.S. government’s failure to prevent the
Fort Hood attack. Retrieved from https://www.hsgac.senate.gov/imo/media/doc/Fort_
Hood/FortHoodReport.pdf
Virginia Tech Review Panel. (2009). Mass shootings at Virginia Tech: Addendum to the
report of the review panel. Retrieved from https://scholar.lib.vt.edu/prevail/docs/
April16ReportRev20091204.pdf
Lankford et al. 341

Vossekuil, B., Fein, R. A., Reddy, M., Borum, R., & Modzeleski, W. (2002). The final report
and findings of the safe school initiative: Implications for the prevention of school attacks
in the United States. Washington, DC: U.S. Secret Service.
Webster Commission. (2012). Final report of the William H. Webster Commission on The
Federal Bureau of Investigation, counterterrorism intelligence, and the events at Fort Hood,
Texas, on November 5, 2009. Retrieved from https://www.hsdl.org/?view&did=717443
White, S. G. (2017). Case study: The Isla Vista campus community mass murder. Journal of
Threat Assessment and Management, 4, 20-47.
Wilber, D. Q. (2016). The FBI investigated the Orlando mass shooter for 10 months—And
found nothing. Here’s why. Los Angeles Times. Retrieved from http://www.latimes.com/
nation/la-na-fbi-investigation-mateen-20160712-snap-story.html

Author Biographies
Adam Lankford is an associate professor of criminology and criminal justice at The University
of Alabama. He is the author of two books and many peer-reviewed journal articles on various
types of criminal behavior, including mass murder, mass shootings, and terrorism.
Krista Grace Adkins is a former graduate from The University of Alabama and a current law
enforcement officer. Her research interests include terrorism, homeland security, and addiction
and recovery.
Eric Madfis is an associate professor of criminal justice at The University of Washington
Tacoma. He has published numerous articles and book chapters about the causes and prevention
of mass murder in schools and other settings. His book, The Risk of School Rampage: Assessing
and Preventing Threats of School Violence (2014, Palgrave Macmillan), explores how threats
of multiple-victim rampage shootings are assessed and prevented in American public schools.