Mass Shootings, Part II
I began this series of newsletters in response to the Uvalde massacre, which, as of yesterday, is no longer the most recent mass shooting in the U.S.
Events like these touch all of us – the victims and their loved ones, the survivors, and everyone else who grieves, fears for their safety, and quarrels with political adversaries about gun control.
My original plan for this week was to discuss causes and remedies. After feedback from several of you, I've decided to focus on the causes of mass shootings. Next week I'll take a close look at potential remedies. Statistics have a lot to tell us about both.
Recap
Here are four of the main points I raised last week (see here for details):
1. Mass shootings in the U.S. are abundant. We've experienced roughly 1 per month, or between 1 and 2 per day, on average, for the past decade, depending on how "mass shooting" is defined.
2. The U.S. is a world leader in mass shootings. From 1966 through the present, we've had the most mass shooters and shootings. We rank second to Yemen in mass shooters and shootings per capita, and our mass shooting rates are much higher than those of any developed country.
3. The frequency of mass shootings in the U.S. is increasing. Over a third of all mass shootings since 1966 have occurred in the past decade, and the incidence spiked during the pandemic. In 2022, there have been 233 so far, which is more than 10 per week on average.
4. Mass shootings in the U.S. are distinctive. Compared to those in other countries, our shooters are more likely to use multiple guns, and more likely to commit shootings in school setting.
Explanations for mass shootings
Why would anyone perpetrate a mass shooting? And, why are mass shootings so common in the U.S.?
Roughly speaking, experts have proposed two kinds of explanations: psychological and cultural. Psychological explanations focus on the mental health and motives of the shooters. Cultural explanations focus on social deviance, attitudes toward violence, and access to guns.
The distinction between psychological and cultural explanations is highly politicized. Although this is a bit of an oversimplification, gun rights advocates tend to blame psychology, while gun control advocates blame culture.
I don't believe it's necessary or even helpful to politicize these explanations. Instead, I think we should just look at the data.
Psychological explanations
The main psychological explanations point to mental health issues, copycatting, and motives such as a desire for revenge, notoriety, or the shooter's own demise. Most shooters seem to be driven by more than one psychological factor, and so, at the end of this section, I'll discuss mass shooter profiles.
1. Mental health issues.
Mass shootings have been attributed to mental illness (e.g., schizophrenia, depression, PTSD) and to transitory but severe disturbances (e.g., suicidality and/or other emotional crises at the time of the shooting).
One way to evaluate this kind of explanation is to look at the characteristics of individual shooters. For example, the largest study conducted to date shows that roughly two-thirds of U.S. mass shooters since 1966 had experienced prior mental health problems, and 71% expressed suicidality before or during their attack. Although only 16% exhibited signs of a serious mental disorder such as schizophrenia, all of them were experiencing some sort of emotional crisis when they acted out.
Statistics aside, one could argue that perpetrating a mass shooting is, by itself, proof that a person suffers from mental illness and is experiencing a crisis. In other words, one could argue that every mass shooter is, by definition, mentally ill.
Even if you accept this argument, it doesn't lead anywhere useful. Roughly half of Americans seek mental health support at some point in their lives, and the percentage who have considered suicide, on at least one occasion, at least for a moment, is supposedly higher (and probably underestimated, because not everyone would admit to that on a survey). Whether or not we experience mental health issues, only a tiny fraction of us become mass shooters.
We can also evaluate the mental health explanation in a correlational way, by asking whether higher rates of mental illness are associated with higher rates of mass shooting. This association is rarely seen. For example, an influential 2015 study found no association between state-level prevalence of mental health issues and either mass killings or school shootings in each state.
Likewise, although some experts argue that mass shootings are more common in the U.S. because we have the highest rates of mental illness, this argument is based on a flawed premise. According to WHO data, the U.S. ranks 3rd internationally in per capita rates of affective disorders like depression and anxiety, but much lower for others, such as schizophrenia. (In any case, I don't believe we can trust cross-national comparisons of mental illness, owing to cultural differences in diagnostic practices, even among countries that rely on the same authoritative source – i.e., the DSM-V.)
Here's one more limitation of the mental health explanation: More than half of mass shooters create and carry out detailed plans. That is, they're competent enough to to acquire guns, to determine when, where, and how to carry out a shooting, and to conceal their plans as necessary (although, as I'll discuss next week, most shooters exhibit signs of what's to come). This is a reminder that you consider shooters sick, or deranged, or crazy, or whatever, they do have a degree of mental coherence that allows them to develop and carry out their plans.
In sum, mass shooters have mental health issues, but this doesn't allow us much precision in distinguishing those who will vs. will not perpetrate a shooting. Having a mental health issue – at minimum, experiencing an emotional crisis – is necessary but not sufficient for carrying out a mass shooting. (See the Appendix for a discussion of the necessary-sufficient distinction.)
2. Copycatting.
Some shooters appear to be inspired by prior incidents. Anecdotally, we see this in statements they leave behind (e.g., admiring references to the Columbine duo). Studies that have also provided indirect evidence of copycat behavior. A 2015 classic, led by Dr. Sherry Towers at Arizona State University, focused on mass killings (by any means) and school shootings. Probability statistics were used, under the assumption that if copycatting occurs, incidents will cluster rather than being evenly distributed over time. The researchers indeed found evidence of clustering. Each mass shooting inspired additional ones for a period of about 13 days on average.
Copycatting plays a role in some mass shootings, but we should be careful not to overstate its importance. Here are some key qualifications:
(a) Clustering data may overestimate or underestimate the extent of copycatting. Overestimation occurs when one shooting is only coincidentally close in time to another one (as may turn out to be the case for the Buffalo, Uvalde, and Tulsa shootings). Underestimation occurs when the shooter copies a role model, but does so much later in time.
(c) Copycatting is, at most, only part of what motivates shooters. We don't fully understand their motives, whether we're examining individual statements or statistical patterns. After all, not all shooters explain themselves, and those that do may not accurately describe their motives. Some shooters may be copycatting but not acknowledge it. Others may praise former shooters but only to justify violence they would've committed anyway. Consider too the distinction between copycatting and anti-normative behavior. Did so-and-so shoot up a school because he was specifically copying the Columbine shooters, or did he do it because shooting up a school is something that "evil people" do, a category that happens to include the Columbine shooters, among others?
3. Other motives.
Other motives attributed to mass shooters include a desire for revenge (for childhood abuse, for being the target of bullying, for social or romantic rejection, for perceived injustices at work, etc). Also mentioned in some cases is a desire for notoriety – i.e., to become famous for the shooting.
Here, our only data consist of statements shooters themselves have made in videos, texts, manifestos, and so on. A desire for revenge and/or notoriety does seem to partially motivate some shooters, but we can't draw more specific conclusions, because (a) not all shooters explain their motives, and (b) among those that do, it's not clear how strong these motives were relative to others.
Some experts claim that most or nearly all shooters expect to die while carrying out their plans, either by killing themselves or being shot by law enforcement. This seems to be true in some cases, because the shooters say as much, although in other cases, the experts are drawing inferences from details that might just reveal poor planning (e.g., the absence of an escape strategy).
My conclusion here is that the motives of shooters vary from individual to individual, although a desire for revenge, notoriety, and/or self-destruction figure into the plans of at least some of them.
4. Combined factors.
Although I've discussed the psychological contributors separately, what the shooters themselves say often point to more than one contributor. This has led some experts to identify patterns and create profiles. For example, Drs. Jillian Peterson and James Densley of the influential Violence Project have reviewed all mass shootings since 1966 and found that beside being male and having access to guns, the typical shooter has suffered childhood trauma, experiences an emotional crisis and/or grievance, and can articulate a "narrative" that justifies how they feel. In addition, Peterson and Densley identified five distinct profiles, one representing shooters in each of the following settings: K-12, colleges/universities, workplace, places of worship, and commercial locations.
For example, the profile of a K-12 shooter is a white male who is a current or former student at the school. This shooter has experienced a childhood trauma and is currently suicidal. He plans carefully, tips off others, and uses multiple guns stolen from a family member.
In contrast, the profile of a college/university shooter is a non-white male who is currently enrolled at the university. This shooter has a history of trauma and exposure to violence, uses legally-obtained guns, and leaves behind a "manifesto" justifying his actions.
Peterson and Densley, as well as other respected profilers, are careful not to make absolute statements. Their profiles are full of hedge words – they say that "many" or "most" people who fit each profile "often" or "typically" or "tend to" have certain characteristics. All the same, I find this kind of profiling troubling, because the hope is clearly that if we can just identify the right mix of necessary conditions, we'll end up with a sufficient one.
In other words, if you spot a middle school boy who's caucasian, frequently bullied, and depressed to the point of contemplating suicide, and you know he experienced trauma as a child and that his parents keep guns at home, well....
Well what? All we know for sure is that this student needs help, because he's getting bullied and feeling suicidal. He's not necessarily going to commit a mass shooting. If you intervene because you think he's a risk, and he realizes that's why you're intervening, you might make his mental health worse (or, suggest an idea he hadn't considered before).
There's also the potential here for racism. The fact that mass shooters in K-12 schools have tended to be white, while mass shooters in colleges and universities have tended to be non-white, might mean something, but it might also be a coincidence. We don't want disturbed undergraduates of color to be scrutinized differently from their white classmates. Anyone who's disturbed needs help.
Profiling is different from the other psychological explanations I've discussed, because the other explanations aren't presented as anything more than necessary conditions, and they don't single people out quite as pointedly.
Cultural explanations
Psychological explanations are meant to tell us why individuals become mass shooters. Cultural explanations are also applied to individual cases (e.g., so-and-so acted out because he grew up in a violent neighborhood and had access to guns), and, as explanations, they reflect some of the same limitations (e.g., the fact that someone grows up in the midst of violence and has access to guns doesn't guarantee they'll shoot anyone). However, cultural explanations are also meant to show why there are so many mass shootings in the the U.S., These explanations have focused on deviant behavior (crime and violence), attitudes toward violence, and access to guns. (References are sometimes made to "gun culture" – in other words, a combination of the variables I just mentioned – but I think it's important to distinguish them, at least for the moment, because some are more influential than others.)
Deviant behavior
Some experts have argued that because crime is so prevalent in the U.S., there's nothing "special" about our high rate of mass shootings. This is just one more way our crime rates exceed those of other countries.
This idea is not supported by the data. Beginning in the 1990's, studies have shown that for most types of documented crimes, crime rates in the U.S. are comparable to those in other Western industrial nations, whereas violent crimes per capita are much, much higher here. In other words, we don't have more crime; we have more violent crime. More violent crime overall, more violent crimes that specifically involve guns, and more lethal, gun-related crimes.
So, if our high rates of mass shootings don't reflect higher crime rates overall, do they reflect our high rates of violent crime? Not really. A carefully-conducted, influential 2015 study of 171 countries showed that mass shootings were not correlated with violent crime in general or homicide rates in particular. As for the U.S., our rates of gun violence are only strikingly high compared to other Western industrialized countries. Internationally, we've never been among the 10 countries with the most per capita gun violence, though we have the 2nd most mass shootings and shooters per capita.
Bottom line: the unusual prevalence of mass shootings in the U.S. can't be attributed to our rates of deviant behavior (overall crime, violent/lethal crime, gun violence).
Attitudes toward violence and guns
Americans are often described as having distinctive attitudes toward violence. To a greater extent than in other countries, we are said to
—glorify violent behavior (in movies, video games, popular music, sports like pro football, and so on).
—expect violent behavior, owing to to how much actual violence we see on the news, in YouTube videos, etc. (In other countries, public dissemination of violent images is often more restricted.)
—tolerate violent behavior. (Not that we consider extreme violence acceptable, but that it is becoming fractionally less unacceptable.)
I would add that in the U.S., mass shootings have become conceivable. If Sean is outraged and contemplating violence, he has countless options. Shooting people indiscriminately is an option, but it may not have occurred to Sean if nobody had done it before. In other words, Sean might not be copying a specific shooter, but simply choosing among known strategies for acting out.
Finally, attitudes toward guns and gun control deserve mention. Although our attitudes are diverse, some experts argue that compared to other countries, a greater percentage of Americans support a greater extent of individual access to guns.
Although there's no simple way to characterize American attitudes toward guns, it does appear that these attitudes contribute in case-specific ways to mass shootings, given what some of the shooters say about their plans. All the same, we'll never know which individuals would've acted out if they had been raised in a different cultural environment. Moreover, it's clear from the anecdotal evidence that the majority of mass shooters were not raised in crime-ridden neighborhoods by adults who advocated for violence and unrestricted access to guns.
In sum, American attitudes towards violence, guns, and gun control may influence some shooters, but this alone doesn't account for the prevalence of mass shootings in our country.
Guns
Many experts, journalists, and politicians point to guns as the ultimate source of America's problems with mass shootings and other forms of gun violence. America has, by far, more guns per capita than any nation on earth – more than one gun per citizen, on average. Specifically, as of 2022 there are 120.5 civilian guns per 100 U.S. citizens. That's almost twice as many as the second-place country (Falkland Islands, with 62.5 guns per 100 citizens), and over three times as many as every country below 4th place on this list.
We don't only have more guns. Our guns are relatively accessible; most experts agree that we have the weakest private ownership-related gun laws of any industrialized nation (see here, for example). And, the guns that Americans can legally own are relatively deadly. "Deadliness" here refers not only to type of gun (e.g., semi-automatic), but also type of magazine (e.g., high-capacity), type of bullet (e.g., 9mm), and certain accessories (e.g., bump stocks).
Statistically speaking, it's clear that America has way more guns per capita than other countries. As for the accessibility and deadliness of those guns, it's also clear that we're an international leader, but I'm not sure whether these variables can be measured precisely enough to create a ranking system. Access to guns is influenced by total number of guns, but also by laws and other restrictions, which vary widely by state, as do the deadliness of the guns that each state considers legal. So, the safest conclusion might that we have many more guns than other countries, and that our guns are "highly" accessible and deadly.
Unfortunately, connections between guns and mass shootings aren't always presented in very specific terms. What you often hear is that the problem in American is simply that we have more guns (or more access to guns, or more deadly guns), as if that alone completely explained the problem.
When the argument is presented this way, critics respond: Sure, we have more guns and so on, but that doesn't cause anyone to pick one up and start shooting. In other words, the critics are rightly saying that the abundance of guns (or their accessibility, or their deadliness) is merely a necessary rather than a sufficient condition for mass shootings.
Although these critics are right, I do think that the number of guns in the U.S., their accessibility, and their deadliness, all contribute to our high rates of mass shootings. To make that argument though, the connection needs to be spelled out.
1. As a starting point, it helps to distinguish between shootings in general versus mass shootings. (Apologies in advance for the graphic details in this paragraph.) If Sean wants to murder one individual, a gun may or may not be Sean's preferred method. If Sean owns a handgun but thinks it would be too noisy, a blunt instrument might be preferable. Or, if Sean isn't brave enough to get close to the victim, Sean might use a motor vehicle. In contrast, if Sean wants to attack a large number of people indiscriminately, guns are the most efficient, readily accessible means of doing so. (For most civilians without expertise, explosives will be harder to access and more expensive, water supplies will be more difficult to poison, etc.) And, although it would be relatively easy for Sean to run a vehicle into a crowd, guns would harm a greater number of people with greater efficiency.
The argument here is not that access to guns causes mass shootings, but that it helps enables them. More guns, easier access to guns, and more lethal options all increase the chances of mass shootings, because if mass murder is a person's goal but that person has no particular expertise, guns are the most affordable, accessible, efficient means of achieving the goal.
2. Studies have shown that American mass shooters are much more likely than shooters in other countries to use multiple guns. Although this doesn't prove anything, it hints that relatively easy access to guns in the U.S. helps enable mass shootings here.
3. The abundance, accessibility, and deadliness of guns in the U.S. readily converges with some of the psychological and cultural factors I described earlier. Suppose, for example, that Sean experienced abuse as a child, feels socially rejected at school, and has begun to grow angry at his classmates as well as depressed to the point of contemplating suicide. Sean wants some release from his pain, and shooting at classmates indiscriminately is a known option. Sean has many role models to emulate, and some of the shooters – real ones as well as characters in video games and movies – look pretty cool when firing their weapons. Meanwhile, Sean sees how easy it is to acquire guns, legally or illegally. And, as he learns more about guns from the internet and social media groups, he starts to notice some pretty amazing models available at home, in friends' houses, at stores, etc. Six months later, you read about Sean in the news...
I'm not saying that "Sean" is a typical mass shooter. I'm just using this fictional character to illustrate some of the ways psychology and culture interface. What I want to emphasize is that when you look at this example, it's hard to argue that any one factor "causes" Sean to become a mass shooter. Rather, it's a combination of psychological and cultural influences.
Conclusion: Some bad news, some good news
Sad to say, my conclusion about the causes of mass shootings is not the one that I, or many of us, want to hear the most.
What we want to hear is what causes mass shootings. We want to know exactly what goes wrong, so we can prevent these tragedies from happening.
I also think that many of us (including myself) would like to hear that what causes mass shootings is consistent with our politics. If you're a gun-rights advocate, you want to hear that it's mental illness, not guns per se. If you're a gun-control advocate, you want to hear that it's guns, not mental illness per se.
You know already I'm going to say that mass shootings, at least in many cases, seem to reflect a combination of psychological and cultural factors. But even that might be a desirable conclusion if we could identify exactly which factors are most important, and how they work together.
Unfortunately, I don't think we can do that. Here's the only conclusion I can float with confidence: Mass shootings are caused by people with mental health problems who live in gun cultures. Mass shootings are more common in the U.S. because our gun culture is more toxic than it is in other countries.
At present, we're not able to explain why the interface between mental health problems and gun culture causes one person to engage in mass shootings, another person to commit suicide, and someone else to merely get drunk and punch a wall.
That's the bad news. The good news is that, historically speaking, people have solved all sorts of problems without fully understanding their causes. Medicine offers some great examples. Acetaminophen, lithium, and penicillin are among the drugs whose mechanisms were almost total mysteries when they first came to market, and the details of how they work still aren't fully understood.
Thus, I have faith that we can reduce the incidence of mass shootings in the U.S., even if we never fully explain why they happen, much less predict with complete accuracy exactly who will or won't become a shooter.
Next week I'll talk about which strategies clearly reduce mass shootings, which strategies might be effective, and which clearly don't work. Thank you for staying tuned.
Appendix: Necessary vs. sufficient conditions for mass shootings
Throughout this newsletter I refer to the distinction that logicians make between necessary and sufficient conditions.
To illustrate, imagine for a moment that you're holding an uninflated balloon. Ordinarily it won't float out of your hands in its current state. If you inflate it with helium, it will float, but only if you release it from your grasp.
In this example, inflating the balloon with helium is a necessary condition for the balloon to float. It's "necessary" in the sense that the balloon won't float unless you inflate it. However, the fact that you inflate it doesn't cause it to float. You would have to release it first. Once the balloon is inflated, releasing it is a sufficient condition for it to float. It's "sufficient" in the sense that letting go of the balloon is sufficient to guarantee its floating.
In this newsletter, I argue that mental health problems are a necessary condition for mass shootings. Nobody would engage in a mass shooting if they weren't struggling with mental illness and/or some current emotional crisis. But mental health problems are not sufficient conditions for shooting, because when people have these problems, even serious ones, they don't automatically reach for guns.
Likewise, access to guns is a necessary but not sufficient condition for mass shootings. You can't shoot anyone if you don't have a gun, but owning a gun guarantee that you'll use it.
The distinction between necessary and sufficient conditions gives us a way of talking about the challenges we face in trying to prevent mass shootings. We want to know what the sufficient conditions are. We want to know what kind of mental health issues, and what kinds of life experiences and exposure to gun culture guarantee that a person will perpetrate a shooting. Unfortunately, what we know best are merely the necessary conditions. (And, even then, these conditions aren't as "necessary" as they are in a logic textbook. For example, all school shooters in the U.S. have been male so far, which might lead one to believe that being male is a necessary condition for committing a school shooting. Strictly speaking, that's not true, because people of any gender can fire a gun. All we can say is that at the moment, maleness functions like a necessary condition for school shootings.)