I’ve been reading some of the commentary on the recent mass shootings in Newtown, Connecticut. I am far from an expert on any of the related issues, but some of the commentary has stirred the following unrelated thoughts.
First, much of the cross-country data being used to make arguments for or against gun control is being misused. I have seen remarks that some countries — usually Switzerland or Israel — have more guns and less homicide. I have seen remarks that other countries, Brazil for instance, has less guns and more homicide. And of course, others point to countries with less guns and less homicide. Using this kind of comparative method, you can pretty much find any country that fits the story you want to tell. But it is the wrong method.
These countries are all majorly different from one another in more ways than just guns. Those major differences all doubtlessly have varying effects on how much homicide there is. The relevant question is whether — all else held equal — introducing more guns will lead to more homicide. So take the Brazilian example. The country has fewer guns per capita than the US but a higher homicide rate. But by itself that tells us nothing. The real question is whether adding more guns to Brazil would make the homicide rate even higher.
That is, does adding more guns create more marginal homicide? It strikes me as pretty probable that if you take any given country and helicopter drop 5 million more guns into it that there will be more marginal homicide. That is an empirical question we probably will not be able to test (unless natural experiments present themselves), but it is the empirical question we are really after. Comparing dramatically different countries to one another on the single-axis of gun ownership will tell you very little about the marginal impact of gun ownership itself on, for instance, the amount of violent crime.
Second, there has been some discussion of race and mass shootings, with a primary focus on white men. According to Mother Jones, over the last thirty years, 42 of the 62 mass shootings have been carried out by white men (only one has been committed by a woman). So white men are the shooters in 67.7% of the mass shootings. This is a sizable majority of them, but it is important to note that — according to 2010 census — non-Hispanic whites make up 63.9% of the population. And this percentage has been declining, meaning it was higher during most of the 30 year period in which these shootings happened. So whites do not appear to be accounting for a disproportionate number of the shootings. Men do, but violent crime is pretty much an exclusively male thing anyways. This presents at least some difficulty for race-based explanations.
Third, focusing on mass shootings by itself seems like a terrible way to figure out their causes. The US certainly leads the world in mass shootings, but even with that, 62 over 30 years is just not a very big sample. If I were trying to understand what is going on here, I would start by looking at related phenomena. For instance, why aren’t we looking at literature (that I assume exists) on suicides and their causes? Just skimming the history of mass shootings, they look way more like suicides than they do the kinds of gun violence people generally compare them to. My suspicion is that suicide research will get you much closer to understanding this stuff than homicide research. But that’s just me, someone who is admittedly not well-versed in these topics.