I hate our dialogue about guns. To cut to the chase (because the main thing we all want to know when we read a political article is whether the author shares our opinion): I think that America would be better off with far fewer guns. If you don’t like that opinion, you might want to sit this one out. Though you don’t have to, because this article is written in my typical “heterodox” style that usually pisses off literally everyone. So, maybe stick around — I’ll probably say something shitty about someone you hate!
Democrats typically “react” to mass shootings by calling for measures that wouldn’t have stopped the shooting they’re reacting to. Which is a bit like “reacting” to low office morale by proposing a Venetian-style masked orgy: It smacks of opportunism and causes people to question your motives. Democrats’ approach hands Republicans a counter-argument: They can say “this wouldn’t have stopped that” and actually be right. By framing the issue in incredibly narrow terms, Democrats lose; we look like ineffectual scolds engaged in moralistic grandstanding to please our base. And we’ve been known to sail into those waters before.
But lawmakers aren’t really to blame. Democrats propose puny gun laws because those are the only laws that won’t cause them to suffer Elden Ring-style immediate, brutal death in the next election. The entire debate gets framed in a way that misses the point. The “fewer guns” side is losing the argument, and worse yet: It doesn’t matter if we win. I think we’d be better off acknowledging that our current debate is mostly pointless and focusing on bringing about the day when it’s possible to pass a law that matters. Even though that day won’t come for at least several decades.
The best and the worst that can be said about current gun law proposals is that they’re literally the least we can do. Democrats and a few Republicans are suggesting things like universal background checks and “red flag” laws because they’re reasonable. After all: Some things require clearance. You have to be approved to drive a car, fly a plane, or brew alcohol; you even get vetted before a shelter gives you a cat, even though everyone knows that Plan B for that cat isn’t “live to a ripe old age and pass away surrounded by family and friends”. It’s reasonable to look into someone’s background before they purchase a deadly weapon; arguing otherwise is to basically side with the twitchy psycho in this Simpsons clip:
But background checks don’t do much. They’re not a bulwark against mass shootings, and nobody should act like they are. By my count, only one of the 22 incidents on this AP list of recent mass shootings might have been prevented by universal background checks.1 Red flag laws don’t seem to do much, either, though it’s difficult to tell which laws might have applied to which shooters. Of course, we can’t know how many shootings our existing background check laws prevented; potential shooters don’t typically notify the press about rampages they would have conducted if they’d been able to get a gun.
Of course, “not doing much” is different from “not doing anything” (I might trademark that and sell it to Tylenol). If we could reduce non-suicide gun deaths by, say, five percent, it would save about 750 lives each year.2 Personally, I think that a bureaucratic hassle is worth saving a few jet airliners full of people, assuming that the people on those planes aren’t those weirdos who recently ambushed people with a Christian hootenanny at 30,000 feet.
It’s easy to look at universal background checks and ask: “Why can’t we do that?” But the “Why should we do that?” argument seems to be better at getting people to the polls. Background checks are supposedly palatable because they’re a small measure. But they’re also, well…a small measure. People don’t get pumped up over tiny things; you never see a car lot advertising itty-bitty savings, or a casino saying “Come by and win MODESTLY!” On the other hand, people who oppose gun restrictions often feel strongly and vote based on that preference, which might be why background check measures often poll well and then fail at the ballot box.
The same “go small, lose small” dynamic applies to assault weapons. There’s no good reason to own an AR-15; hunting deer with an AR-15 is like hunting slugs with a Predator drone. But there’s also no technical definition of “assault weapon”; it’s a made-up category, like “sapiosexual” or “performance-class sedan”. Because the 1994 assault weapons ban applied to a made-up category, the law was largely evaded; gun makers designed new “blassault bleapons” and resumed business as usual. Here are some of the modifications that they made:
They made a few superficial tweaks and pretended it was a whole new thing — it’s the Star Wars Trilogy: Special Edition of the gun world. We probably shouldn’t be surprised that most of the research finds that the ban — which expired in 2004 — had little to no effect.
A better approach is to focus specifically on magazine capacity. There’s some research suggesting that limiting the number of bullets that can be fired without reloading makes mass shootings less deadly. The reason is obvious: If the shooter has to stop and reload, that gives police time to intervene. Unless we’re talking specifically about the Uvalde police, in which case it will give them more time to sit in the hallway and brainstorm excuses for not confronting the shooter.
But most shootings are not mass shootings. In recent years, 2.53 percent of non-suicide gun deaths are from mass shootings.3 Which I think demonstrates the disingenuousness of our debate: We focus on mass shootings, but mass shootings are not really the problem. Mass shootings are a small subset of the real problem, which is the relatively high likelihood of being killed by a gun in a country with many guns.
Here’s the problem, visually:
It’s pretty clear:4 More guns = more shootings. The most common ways of being killed by a gun don’t make national news; the situation is often “two guys got into a fight and one of them had a gun”. Gang violence is part of the story. Unintentional shootings are about 13 percent of non-suicide deaths. Roughly 20 percent of all murder victims are killed by their romantic partner. Mass shootings get the headlines, but you’re far more likely to be killed by a gun in a fight that boils over.
I think there are legitimate reasons to own a gun. I’m not a hunter, but I grew up around hunters, and I know that hunting really matters to some people. I also hate deer; I’ve lived in some deer-heavy parts of the country and I know that deer are gigantic assholes. They eat everything in your yard and are on a kamikaze mission to make our streets undrivable; they’re nihilistic fucks who just want to watch the world burn. We only tolerate them due to a horrifically successful PR campaign, most notably Bambi, a shameless propaganda film that I consider to be a cervine version of The Birth of a Nation.
I also think that home protection is a legitimate use for a gun. Personally, I will not be buying a gun for home protection, because I’ve run the numbers, and the odds of me doing something incredibly stupid are roughly a million times higher than the odds of a home invasion. The data backs me up: Home guns are used in accidental shootings four times as often as they’re used in self defense, and they’re used in assaults or homicides seven times more often. For me, this is an easy call. But I’m reluctant to infringe on what people can do in their own homes, and if some people feel that they’re safer with a gun, then that’s a choice I consider not-insane.
But I think that other arguments for gun ownership are extremely bad. The “the only thing that stops a bad guy with a gun is a good guy with a gun” argument strikes me as obviously untrue. There are, of course, a few instances of civilians with guns stopping a shooter, but there are also examples of them making things worse. Crime data is notoriously difficult to parse — there are so many variables that it’s hard to isolate one thing — but the preponderance of evidence shows no connection between more guns and less crime (if anything, the correlation runs the other way).5 And, anecdotally, I was once in a 7-Eleven and noticed that the probably 80 year-old guy in front of me had a gun, and my reaction was absolutely not “Oh good: Rip Van Winkle and his arsenal of cataracts, hand tremors, and high-powered weaponry are keeping us safe! I’ve never felt more calm in my life!”
The “right to revolution” argument is also absurd. Gun rights advocates sometimes get the ghost of Thomas Paine up their ass and start waxing poetic about how guns keep the government at bay. This is moronic; don’t people know what kind of weaponry the military is using? The defense budget is $782 billion; I assure you that they’re not spending all of that on pistols. Your Glock isn’t going to do much if the shit hits the fan and the government sends a F-35 Lightning after you. Protection from government tyranny comes from our laws and our institutions, not from small-arms shootouts, and some people may not like that, but I’m afraid that train left the station about 200 years ago.
So: In my opinion, there are legitimate reasons to own a gun. But I also think that the remarkable proliferation of guns — especially hand guns, which are used in 59 percent of gun murders — is making us less safe. Guns are so common and easy to get that conflicts turn deadly far too often. Any law that doesn’t address that reality is just nibbling around the edges of the problem.
Most European countries ban most guns. Canada has major restrictions that include mostly forbidding carrying a handgun outside the home. Personally, I’d like to see a ban on any gun shorter than two feet in length or that can fire more than two rounds before being reloaded. This would let people have a rifle for hunting or home protection, but that’s about it. Maybe some exceptions could be made for the few farmers who have predator control issues — I wouldn’t want to subject someone to a 30-50 feral hogs situation — but for most people, the ability to quickly and effortlessly slaughter two-score wild swine is not a daily concern.
My proposal would be wildly unpopular. I would not for a second recommend that any Democratic politician endorse my plan; if any were asked to get on the record about my two feet/two shots idea, I’d recommend that they call me a communist and then — if I’m nearby — spit on me. This is surely the least popular idea since OnlyFans decided to try to compete with Nick Jr. And that’s before you consider that it would require a gigantic, Australian-style gun buyback program that would cost tens of billions of dollars.
But I don’t know what else to think about this issue. I simply can’t convince myself that things like background checks and limits on magazine capacity would do very much. Nor can I shrug my shoulders and say “nothing we can do” when I don’t think that’s true. The reality, as I see it, is that the prevalence of guns is making us less safe. Decreasing that prevalence would probably reduce murder, maiming, and suicides by a substantial amount. I’m interested in policies that achieve that outcome, and if I’m being honest, I don’t give much of a damn about the others.
A serious gun reduction measure might require repealing the Second Amendment. At a minimum, the Heller decision would need to be reversed. This pushes the date at which such measures could realistically be considered way off into the future. Passing a major gun reduction law would require a change in attitudes of the sort that’s only happened a few times in American history. This process would take an excruciatingly long time; this is tantric lawmaking in its most extreme form. I can understand why people would find this approach unappealing, impractical, and a non-starter.
But if we want to actually address the issue, this is the only road I see.
I’ve witnessed one sea change in public opinion in my life: gay rights. As recently as the ‘90s, support for gay marriage was a fringe opinion, kind of like believing that dogs should be allowed to practice medicine. Now, that opinion is normal; 70 percent of Americans support gay marriage, including 84 percent of people age 34 and under. So, incredible turnarounds are possible.
Of course, just because one unpopular opinion eventually became popular doesn’t mean that all unpopular opinions are headed that way. That’s bad logic, similar to when people say “They said we couldn’t go to the moon, and therefore every dumb pie-in-the-sky idea is actually possible.” Most unpopular ideas simply stay unpopular, which is why the Socialist revolution is currently 174 years behind schedule.
But I think the argument for majorly reducing guns has a chance to become popular because — much like with gay marriage — I think the logic is sound. I basically laid out that argument above — do you find it convincing? If not, then congratulations: You’re in the majority. And maybe you always will be.
But I didn’t come to my position on guns because I think it’s destined for glory; I came to it because it’s the only way I can see to substantially reduce gun violence. I think our current conversation is mostly a sideshow waged on terrain favorable to the pro-gun side, and I’m interested in a conversation about changing the conversation. The nibble-around-the-margins approach isn’t getting us anywhere, and it never will. If this country ever passes a serious gun law, then it will probably happen several decades from now. Which seems like a great reason to start moving towards that outcome right away.
Two shooters should have been prohibited from buying a gun but weren’t because of data input errors.
The precise numbers: The Gun Violence Archive has seven years of data going back to 2014. The average number of non-suicide gun deaths in those years is 15,214. Five percent of that number is 761, which I rounded down to 750 partly to emphasize the fact that we’re dealing with estimates and partly because 2020 was a bit of an outlier year. And, FWIW: I chose “five percent” as my hypothetical number because in the previous paragraph I noted that it looks like 1 of 22 mass shootings would have been prevented by universal background checks, and 1/22 = 4.55%.
In this dataset, “mass shooting” is defined as four or more people killed, not including the shooter, in a single incident.
Other researchers have crunched the data different ways and reached different results. The main sticking point is whether you include undeveloped countries in the data set. Some poor countries have low rates of gun ownership but high murder rates, and if you include those countries in the data set, it changes the correlation, even pushing it slightly negative. But I think this type of comparison is apples-to-oranges; countries like Honduras and Venezuela (most of the low-gun-high-homicide countries are in Central and South America) are plagued by gang violence. It’s also not clear that the estimated number of firearms in those countries is remotely accurate. South Africa is obviously…well, it’s South Africa, and though its GDP makes it a “developed” country, we all know there’s a lot going on there. I feel that limiting the data set to “developed countries minus South Africa” makes for a more apples-to-apples comparison.
There is also a distinction between “murders” and “homicides”. Homicides = murders plus “justifiable” killings, and if I could have found data that only addressed murders, I would have used that. But the data I found uses homicides.
And, to be perfectly clear: I think that guns are only part of the reason why the US has a higher homicide rate than other developed countries. Culture seems to be part of what drives murder rates; after all, I just made a “culture is part of it” argument regarding Honduras, Venezuela, and South Africa. But I think guns are a big part of the story.
This is one of those “you can find a study that says anything” areas, but I found far more studies finding that more guns don’t correlate with lower crime than ones that do. Some examples are here, here, and here. I find the evidence in those studies interesting, but not conclusive; in particular, the various formulations of the “more guns are owned in high crime areas” finding could simply reflect the fact that people living in high crime areas are more likely to buy guns.
Honestly, the “guns reduce crime” studies — which I explicitly sought out — tend to all come back to one guy: John Lott. I can’t say I’m enamored with what I’ve seen of Mr. Lott’s research: In this paper, for example, he says “The overwhelming majority of peer-reviewed academic research by economists and criminologists concludes that ownership of permitted concealed handguns causes a reduction in violent crime.” That’s a huge statement, but he offers no citation. Later in the paragraph, he cites a research paper written by…himself. His main credential is that he works for the Crime Research Prevention Center. Here’s their web page — tell me this “center” isn’t just one guy. Their home page features ten articles and nine of them are by John Lott.
My point being: This is how I arrived at the characterization “the preponderance of evidence shows no connection between more guns and less crime”.
The gun policy you have suggested is similar to those in a lot of other countries that have proven successful.
UK law is similar to what you proposed; handguns are banned by banning guns below two feet long, but instead of restricting ammunition capacity, we just ban all semi-automatic rifles outright.
You have to do something other than pulling the trigger again to reload. That means that a bolt-action rifle can have a magazine with no legal limit as to size. However, you have to get a licence (a firearms certificate) for each gun, and if you tried to get a licence for a bolt-action with a large-capacity magazine, you'd get refused. If you already own a gun, then applying for a second one will get "why do you need two guns, won't one be enough?", which does a good job of limiting large gun collections.
Shotguns are different; you get a personal licence to own a shotgun (note: pump-action is not a shotgun, this is only for break-open weapons). You can then own an unlimited number of shotguns on one licence. In addition, shotgun licences are much easier to get.
But you need an FAC (a full gun licence) to buy slug ammo for a shotgun; birdshot only on an SGC (which will hurt rather than kill most humans).
Note that we don't have deer in the UK outside of a few protected national forests and so on, so no-one goes out shooting deer.
One lesson from the evidence-based paradigm is to focus on systematic reviews and meta-analyses of a research question instead of individual studies, since one can always cherry pick studies that support preferred policy outcomes. (Even narrative reviews are problematic because it is difficult for the reader to assess author bias; there is also problem of "vote counting" in narrative reviews so every study counts the same as opposed to giving more weight to higher quality studies.)
For gun policy impact evaluations, there is only up to date one systematic review of which I am aware: Smart, Rosanna, et al. 2020. The Science of Gun Policy: A Critical Synthesis of Research Evidence on the Effects of Gun Policies in the United States (2nd ed.). Santa Monica, CA: RAND Corporation. The RAND study used the following inclusion criteria: "we required, at a minimum, that studies include time-series data and use such data to establish that policies preceded their apparent effects (a requirement for a causal effect) and that studies include a control group or comparison group (to demonstrate that the purported causal effect was not found among those who were not exposed to the policy). Experimental designs provide the gold standard for establishing causal effects, but we identified none in our literature reviews (p. 26)."
The RAND team's overall conclusion is consistent with Jeff's take: "Of more than 200 combinations of policies and outcomes, we found that surprisingly few were the subject of methodologically rigorous investigation. …[I]t is apparent that research into five outcomes is either unavailable or almost entirely inconclusive” (p. xxiii). Table 21.1 (pp. 332–335) of the report includes a nice summary of the evidence for 18 types of gun policy and 8 outcomes (e.g., suicide, mass shootings, etc.). The only areas where there was moderate or supportive evidence for (criminal) violence reduction—meaning at least two rigorous studies showing effects—were as follows: (1) Prohibitions associated with domestic violence restraining orders (moderate evidence of decreased intimate partner homicides); (2) Dealer background checks (moderate evidence of decreased firearm homicides); (3) Waiting periods (moderate evidence of decreased total homicides); and (4) Stand-your-ground laws (supportive evidence of increased firearm homicides). For prevention of mass shootings, the evidence was deemed "inconclusive" across the board. (This is partly because it is such a rare event that it is incredibly hard to study).
Of course, absence of evidence is not evidence of absence, and some will infer that "gun control does not work" when it is more correct to say, "inconclusive evidence." Still, the evidence to date is pretty underwhelming. My preferred response is to let states continue to experiment with different gun policies (at the margin, since large changes are not feasible) and rigorously evaluate these experiments. Maybe we will discover that something actually works. But we should have realistic expectations.