Effectiveness in Stopping Public Mass Shootings

What is an assault weapon? What is a mass shooting?

Those two questions are not pedantic questions when we are talking about policy and trade-offs (because policy always has trade-offs). This is of particular interest when we are talking about how to solve the specific problem of public mass shootings, and even more particularly, within schools.

For example on the latter way of defining a mass shooting: Everytown for Gun Safety, the nonprofit organization founded in 2013 to advocate for gun control and against gun violence, keeps a database of mass shootings, and according to its database, since 2009, there have been 274 mass shootings in the United States. From those shootings, 1,536 people have been shot and killed, and 983 people shot and injured.

Everytown defines a mass shooting as, “an incident in which four or more people are killed with a firearm, excluding the perpetrator.” That seems like a good start, but then we learn what is included, “… this analysis includes mass shooting incidents that occur in both public and private spaces, have any number of shooters, and result from a myriad of motives, such as group violence, domestic violence, or terroristic violence.”

Emphasis is mine. You see the issue we’re already running into? From the standpoint of Everytown, an organization that wants to enact gun control against all gun violence, I fully understand this methodology, and including private spaces, group violence, and domestic violence. But when we’re specifically trying to solve the problem of public mass shootings (and again, more particularly within school spaces), its database and that 274 mass shootings claim only serves to confuse the issue.

Because it may be the case that such and such policy aims may work for one set of mass shootings (like gang violence), or such and such policy aims may work for another (like domestic violence), but not for public mass shootings, which understandably recieve the most public outcry and calls for policy solutions.

Again, we are talking about a very specific kind of crime, which requires a specific sort of focus, and as such, we need a specific understanding about what we are even talking about (definitions) in order to present solutions to that problem.

By comparison, for example, Mother Jones magazine, also has a mass shootings database. For one, their dataset covers 1982 to 2022, but even with that expanded data set, Mother Jones found 128 mass shootings (and notes that most of the killers acquired their guns legally).

Their methodology for the dataset is to define mass shooting as “indiscriminate rampages in public places resulting in four or more victims killed by the attacker.” See, I personally prefer that definition because it’s far more specific to the problem we are trying to address, and the problem that receives the most outcry because of its public nature.

[You may be wondering why both databases utilize “four or more victims killed” instead of, say, three, or even two; that’s the Federal Bureau of Investigation standard definition for a mass shooting: four or more.]

So, beyond that point, since we’re talking statistics, let’s just go ahead with another point: Despite the rightful outrage that public mass shootings engender (and I know this isn’t solace to those affected by it), they are a rare form of gun violence relative to overall gun violence. Which again, bolsters the point of why we need to be precise about potential policy applications: Will such and such policy address the broader gun violence problem, or this specific, albeit rare, problem? Like Everytown, and most people of good moral character, you may be interested in both! But that doesn’t mean the former will be applicable to the latter. That’s the problem I’m trying to repeatedly emphasize here. I don’t think you can just apply the broad policy program and hope it works, or trickles down, to the rarer form of gun violence.

According to UC Davis Health, there were 39,707 deaths from firearms in the United States in 2019. The vast majority of those, 23,941 deaths, or 60 percent, were suicides. The other large chunk are the 14,861 people in the U.S. who died from firearm homicide. (Firearms in general make up the vast majority of “means” for homicide deaths, and half of suicides).

The remaining 3 percent of firearm deaths are unintentional, undetermined, from legal intervention (the police), or from public mass shootings. The latter percentage specifically? It’s 0.2 percent.

That’s right: Public mass shootings account for 0.2 percent of firearms deaths in the United States.

I think the issue seems clear then: Trying to address a 0.2 percent problem with broad policy is difficult! Again, I’m not saying it’s not doable or that because it’s difficult we shouldn’t try something that might work (with acceptable trade-offs), but we need to acknowledge the difficulty! And again, by the nature of the crime, even though it is 0.2 percent, that 0.2 percent generally receives far more public outcry to do something about guns than the other 99.98 percent of firearm deaths. (In my experience, most people agitating for gun control also don’t realize how many of the firearm deaths are suicides.)

Tracking that .02 percent based on different definitions and methodologies, makes it difficult to even understand if mass shootings are going up, staying the same, going down, or were affected, for example, by the Federal Assault Weapons Ban in 1994 (which then expired in 2004). In other words, when the figure we’re talking about is that small, it’s hard to draw actionable conclusions from such a dataset (presuming again, that we can agree on how to compile the dataset!).

So, before we can talk about the AWB, let’s go back to my first question at the top: What is an assault weapon? I think in popular nomenclature, when the public hears “assault weapon” or even just when conjuring up an image of the AR-15, they imagine that it’s a machine gun. More than one round per every pull of the trigger, or with “spray fire.” But that’s not the case since machine guns are uncommon in the United States, and have been banned since 1986 (ones prior to the ban were grandfathered in), and were heavily restricted before that going back to 1934.

The AR-15 has a lot of mystique in the American public imagination, which itself is a circular feeding mechanism — it has mystique, therefore mass shooters often use that gun, which then feeds into the mystique, which then … — but it is not an automatic weapon. It is a semi-automatic weapon with select fire. The reason for the mystique is typically cosmetic more than function: They can be customizable to look like an automatic weapon. Being customizable is also why the AR-15 is so popular as a rifle, from what I understand.

Functionality-wise, the AR-15 differs from the automatic weapons in that, like a semi-automatic pistol, they fire one round per trigger pull (my understanding of the difference between a semi-automatic pistol and a revolver is that the former will load the next round with each trigger pull).

All of this might seem superfluous, or pointless, to someone who wants to eliminate all guns, or just the AR-15 specifically, or whatever they mean by “assault weapon.” But I think when we’re talking about policy, and again, the specific problem of public mass shootings, it’s important … to know what we’re talking about! To know what we mean by what we’re saying.

For the record, it’s worth pointing out, too, that handguns are the overwhelming weapon of choice for overall gun violence, not rifles, and so not the AR-15. And in point-of-fact, the semiautomatic handgun, which also also been used in public mass shootings, doesn’t have an appreciable difference in rate-of-fire as the AR-15, as explained here by a constitutional law professor.

Back to the AWB. As mentioned, the AWB was enacted in 1994 and banned certain semi-automatic weapons that were defined as automatic weapons, and certain ammunition magazines defined as large capacity.

Something I’ve seen floating around on Twitter is this picture of a plot of mass shootings since 1982:

People have pointed out that there is a noticeable jump after 2004 when the AWB expired. Aside from making the trite (but still true!) point that correlation isn’t causation, let’s dig into whether the AWB was effective, and crucially, whether reenacting the AWB (or something like it) would stop future public mass shootings.

A study released in March of last year by Northwestern Medicine, a nonprofit healthcare system affiliated with the Northwestern University Feinberg School of Medicine in Chicago, found the AWB to be effective: Researchers estimated that the 10-year ban prevented 11 mass shootings, and if it were allowed to continue, would have prevented 30 public mass shootings that killed 339 people and injured 1,139. I don’t have access to the PDF of the study, so I can’t dig into the methodology, and an article about the study doesn’t elaborate either. Overall, I’m skeptical, primarily because that seems like a tricky thing to nail down, right? That one specific law prevented X amount of crime, and would have prevented X amount of crime if still enacted.

On the other hand, a 2017 study, which reviewed other peer-reviewed articles from 1970 to 2016 to examine the association of firearm laws to firearm homicide, found that certain laws (strengthening background checks, for example) seemed to result in decreased firearm homicide rates, but that specific laws aimed at banning “military-style assault weapons” were not associated with changes in firearm homicide rates. Unfortunate, again, I can’t access the full text to see the explanation, but here is my thought (and which also goes back to why I’m skeptical of the prior study): There are a lot of guns in the United States, so, banning particular types of rifles deemed “assault weapons” is not going to have an appreciable impact on would-be criminals’ ability to acquire semiautomatic handguns, which again, can fired at a comparable rate to an AR-15.

Christopher Koper, in a federally funded study of the AWB released in 2004, which was an update on his prior report examining 1994-1996 for the U.S. Department of Justice and Congress, noted that even prior to the ban (and which mostly still holds true today), “assault weapons” were used in between 2 percent to 8 percent of gun crimes. Even with respect to large capacity magazines, or LCMs as Koper refers to it, it’s not clear, he said, that the outcomes of gun attacks depend on the ability of offenders to fire more than 10 shots without reloading.

The short story on the AWB is: Manufactures just produced guns that loophole around the ones outlawed by the ban.

“Should it be renewed, the ban’s effects on gun violence are likely to be small at
best and perhaps too small for reliable measurement,” Koper said, again pointing to the relatively rare use of such guns in gun crime.

The RAND Corporation, a nonprofit think tank, launched the Gun Policy in America initiative to “create resources where policymakers and the general public can access unbiased information that facilitates the development of fair and effective firearm policies,” has a great article detailing basically what I’ve been outlining in this blog post about the messy thicket of trying to understand mass shootings.

Thing is, I haven’t even spent time on the “fair” part of the equation offered by RAND there, which I would take to encompass rights-based questions. My focus this past week, and in this blog post, has been on the question of effectiveness. Because like everyone else of good moral character, I am also interested and invested in how to stop these public mass shootings, so, effectiveness is an important point!

And as it stands, I’m not convinced of the effectiveness of much of what is on offer. We don’t have a clear, common understanding of what a mass shooting is, and so we don’t have a reliable data set to work with. We don’t have a clear, common understanding of what is meant by an assault weapon (or automatic weapon), and what it meant, and would mean, to ban them. And I don’t think we have a clear, common understanding in appreciating the difficulty in tackling a relatively rare, but publicly well-known phenomenon like public mass shootings with public policy, much less giving any weight or serious attention to potential trade-offs of such policy (again, I reference police reform and mass incarceration as trade-offs to consider).

All of this, and more, makes it difficult to be effective at stopping the next public mass shooting, at least from a policy standpoint, but I think the difficulties in that realm also influence the difficulties in the cultural realm as well. Which makes me frustrated! It feels like repeatedly running into a brick wall labeled “impasse.” But I think if we can just find these commonalities, we can actually have more fruitful conversations and more fruitful, potentially life-saving, outcomes.