top of page

Four Observations About Mass Shootings


This column does not contain any super-easy super-obvious solutions to the problems of mass shootings. However, it does make four points that are not particularly being made in media discussions of the recent outbreaks of gunmen firing randomly into large crowds.

1. Gun Control Would Do No Harm but It Would Probably Not Reduce the Problem.

There is nothing wrong with gun control. In principle, I am all for it. But I question whether gun control will stop mass shootings.

There is now a substantial literature on the effect of access to guns on homicide rates and violence generically. The consistent finding of these studies is that access to guns neither increases or decreases the rate of violent deaths. U.S. states with gun control do not have lower homicide rates than states without gun control. Countries with gun control do not have lower homicide rates than countries without it.

Consider Japan, Singapore and Switzerland. These three countries have the lowest homicide rates of any country in the world. Japan and Singapore have absolute bans on individual gun ownership. Switzerland requires gun ownership for all males. Serving in the Swiss army is mandatory. All Swiss men keep their guns at home. Moving from zero percent gun ownership to one hundred percent gun ownership seems to make no difference in the homicide rate.

Someone might argue “But Europe Has a Lower Homicide Rate Than the United States.”

True enough. However, Europe has less social inequality than the United States. European nations have more welfare than we do in the United States. There is less poverty in Europe than in the United States. An angry young man who is fired from his job in the United States has no income, no food stamps, no health care, no housing, no nothing. An angry young man who is fired from his job in Germany gets a government check, has all the food he needs, can see a free government doctor any time he wants and can live in a government apartment. His basic needs are essentially covered. This tends to make a young guy who has lost his job a little less angry.

Yes, racist violence and hooliganism is going up in Europe. But welfare payments are also shrinking and poverty rates are going up. Of course, there is “no relation” between these two European trends. (Cough, cough.)

As for the efficacy of gun laws, consider that an awful lot of illegal stuff is sold in the United States. You may know people who have purchased marijuana even when they weren’t supposed to. You may know people who have purchased stronger drugs even when they weren’t supposed to. To be sure, making guns illegal makes them harder to get for the person who is not highly motivated to make the extra effort. Do you think someone who is about to go into a shopping mall, kill dozens and dozens and dozens of people and die themselves in a blaze of gunfire might be motivated enough to risk a jail penalty for buying a gun?

2. More Access to Mental Health Services Might Be Useful Here.

The Substance Abuse and Mental Health Services Administration estimated that in 2016, the majority of people with mental health problems were receiving no form whatsoever of psychological or psychiatric treatment. Over one third of all people with serious mental health problems were receiving no form of treatment. Those are a lot of mental health problems that are going untreated.

To be sure, the overwhelming majority of mental health patients are not at risk of becoming mass shooters. Most of the people who would be being treated are not the particular individuals who are planning mass assaults. In 2019, up through August 4, there were 248 mass shootings in the United States. The National Council for Behavioral Health estimates that in any given year in the United States, there are slightly under 44 million people who are mentally ill.

This means even if we provided treatment to every person with mental health problems, chronic or acute, in the United States, 99.9995 percent of those treatments would have been to people other than the 248 mass shooters. Plus, what are the odds that some mass shooters intentionally avoid treatment, or that they enter treatment that turns out to be ineffective?

Still, not treating over a third of all people with serious mental health problems is not good. Not treating the majority of people undergoing acutely problematic stress is not good either. Being able to reach some of these people would undoubtedly make an impact.

3. Having Internet Settings Where People Encourage Each Other to Commit Acts of Extreme Violence is Highly Problematic.

A consistent finding in criminology is that the probability of committing a deviant act is significantly increased by belonging to a deviant social network with a deviant moral code. One of the reasons that most people do not commit acts of violence is they live in a world in which everyone they know disapproves of such violence. Their families would disapprove. Their romantic partners would disapprove. Everyone at work would disapprove. All of their friends would disapprove. If they were to do something violent, they would be cut off by everyone they love or care about. Life would not be worth living, so they don’t do anything violent.

If you take those same “normal” people, and put them in a social group where it is okay to kill, and then they will kill. If there is a war, and they enlist to serve their country, and they are now on the battlefield facing the enemy, they will do what their new social network wants them to do. They will kill as many of the enemy as they can without putting the rest of the unit at risk.

The differential association school of criminology argued that a great deal of deviance comes from belonging to a social group that encourages deviance. There is a lot of support for the theory, but some of the most useful support comes from the Rochester Gang Study. These researchers found that for adolescents, crime goes up when they join a gang and goes down when they leave the gang. This holds even when you control for poverty, coming from a broken home, past deviance and a whole range of other control variables. Belonging to a group encouraging you to break the law makes you more likely to break the law.

Currently, as far as I know, there are no clubs or gangs where people have face to face meetings encouraging each other to shoot up shopping malls. However, such virtual clubs and gangs do exist on the internet. There all sorts of places on the internet where people of like interests can get together and encourage each other to do whatever. Such virtual meeting spaces did not exist before the development of the internet. Mass shootings have gone up dramatically in the last fifteen years – the same period that has seen the fleshing out and articulation of special interest sites on social media. The mass shooting phenomenon predates Donald Trump and is occurring in nations other than the United States. There may be some other time-based cause that I have missed, but the presence of new violent subcultures on the internet seems to fit the data.

4. The Copycat Effect.

Social behavior occurs in waves. Someone does something new. It gets a lot of attention. It generates a lot of copycats who want to do the same thing.

This produces social behavior that occurs in waves. Someone does something. It is a big success, measured by whatever criterion the observer defines as success. Then more people do it and more people do it. This produces vast waves of the behavior being observed.

What stops the wave? The most common cause is the behavior starts failing. People look at the conspicuous failure and decide that continuing to engage in that behavior may not be worth it. The wave dies out, and continues to die out until there is some brand new success.

This does not bode well for mass shootings. Generally, even in “successful” mass shootings, the shooter gets killed. In some ways, the mass shooting is a form of suicide in which the shooter decides to take a bunch of people out along with himself. He gets a lot of attention and he makes a statement. Shooting the killer early does not “make a failure” that could end the wave. In the Dayton killing, the shooter was killed almost immediately after the arrival of the police. He still killed nine people and became a headline story in the national news. Stopping the killer before he can start shooting does not count as much of a failure. This is not headline news material – and it is unclear either in a court of law or in the court of the mass media if one can prove that he really intended to go through with the act.

Strike waves stop when a big strike loses.

Riots stop when rioters are overwhelmingly repressed by the police before they can do much damage.

This type of incident produces a huge media worthy “failure”.

It is hard to imagine what a big media-worthy “failed” mass shooting would look like.

So, unless some sort of ennui or “going out of style” occurs, I can’t see the type of failure occurring that would discourage mass shooter wannabes.

*  *  *

So, what would stop the mass shootings?

This ain’t an easy question. Most of the solutions are magic pixie fairy dust.

Providing a massive increase in the nation’s mental health provision? Nice as hell but ain’t gonna happen.

Reducing the ethnic tensions in our society? Nice as hell but ain’t gonna happen.

Reducing the social isolation that allows deviants to create violent subcultures without more mainstream members of society intervening with their own social network pressure? Nice as hell but ain’t gonna happen.

Putting in gun control. This might even happen. It probably won’t help.

What does that leave?

Monitoring and/or closing the internet sites where extremists cheer each other on. This is practical and this can be done.

What else does this leave?

Better policework.

We don’t like the intrusions of a surveillance state. We don’t like the reduction of privacy that has come from security cameras everywhere, and everything we do on the web being watched. But if society is trying to stop something bad from happening, surveillance is one of the few practical security measures that exist.

Mass shootings are similar to terrorism either in other countries or our own. There is not that much that can be done to stop “root” causes. But you can try to identify likely terrorists and disable them before they get started.

Sometimes there really are situations where “This is a Job for the Cops.”.

As I said before, I don’t have any magic-wand easy-solution cures.

But it does seem sensible to try to increase the general provision of mental health services. That means reducing the number of people with no health coverage whatsoever, and making sure that people with health coverage get mental health services in the mix.

It does seem sensible to monitor or even close down the internet locations where people brag about and plan their latest potential mass shooting. Sure, people can always open up more sites; but we can make the job more difficult. And it certainly makes sense for the authorities to monitor the people who are talking up violence in a big way.

We might want to do something to reduce the amount of ethnic hostility that is out there in the world. (This is admittedly a tough project.)

We might want to provide better economic and social support for those people who find themselves economically displaced so they are not facing such catastrophic falls.

We might want to think about the increasing isolation of modern life, that allows people to live their lives only texting with dangerous deviants hundreds of miles away because they have no normal friends or associates at home.

Solving the isolation of American society is no easier than getting rid of ethnic hostility – or for that matter paying for a vast expansion of the mental health network.

But isolation, legitimate economic problems, unresolved stress or aggression issues, and facilitators sitting by computers willing to egg them on is not a healthy combination. Any improvement on any of these indicators would make some sort of positive difference.

For more on this topic see: Postscript to “Four Observations About Mass Shootings”

For more on the non-relationship between gun control and homicide rates, see anything written by Gary Kleck. He has made a lifetime of researching the subject. Often in academia, for any article on a position, there is an equal and opposite article on the other side. But Kleck’s critics have had a hard time coming close to matching his impressive stock of empirical studies. One introduction to his work is his Point Blank: Guns and Violence in America.

The non-treatment of mental illness statistics come from Substance Abuse and Mental Health Services Administration (SAMHSA). (2017). Key Substance Use and Mental Health Indicators in the United States: Results from the 2016 National Survey and Drug Use and Health (HHS Publication No. SMA 17-5044, NSDUH Series H-52). The statistics on the extent of mental illness in the United States come from

The criminologist who first laid out the theory of differential association is Edwin Sutherland. If you don’t like reading oldie moldies, a more up to date treatment can be found in Donald Cressey’s Delinquency, Crime and Differential Association. The Rochester Gang study referred to above is Thornberry, T. P., Krohn, M. D., Lizotte, A. J. and Chard-Wierschem, D. (1993). The role of juvenile gangs in facilitating delinquent behavior. Journal of Research in Crime and Delinquency 30, 55–87. One objection to the above is that maybe only really really bad kids join gangs and when they turn good they want to leave. A Norwegian team of researchers studied Norwegian gangs (yes, that’s a thing.) and found that even allowing for that specific objection, joining gangs still had an independent effect on increasing crime. See Bendixen, Mons, Inger M. Endresen, and Dan Olweus. “Joining and Leaving Gangs: Selection and Facilitation Effects on Self-Reported Antisocial Behaviour in  Early Adolescence.” European Journal of Criminology 3, no. 1 (January 2006): 85–114.

On what stops and starts behavior that occurs in waves, see the article I wrote with Carol Conell in the 1995 American Journal of Sociology: “Learning from Other People's Actions: Environmental Variation and Diffusion in French Coal Mining Strikes, 1890-1935”.

bottom of page