Yes, another gun post. I have more stuff in the queue, I promise.
Vox and other liberal websites have been posting a chart from Tewksbury that shows that the amount of gun violence in a nation rises with the number of guns. If you look at their first chart, you’ll see a bit of a trend. But their second plot just shows countries with a very high Human Development Index and the trend becomes clear:
There is no deception going on here. The plot is accurate, to the extent that the data are. My problem, however, is the reasoning. Note what is being plotted: gun deaths. Not violent deaths, suicide or homicides … only those specifically with guns. But that begs the question: what is this actually telling us? Is it telling us that fewer guns would mean fewer deaths? Or does it just mean that fewer guns means more people dying by other means?
Vox clearly believes the former, having made the claim that if our levels of gun violence were the same as the UK’s, that would save 20,000 lives a year. But I’m dubious. In my previous debunking of some myths by Mother Jones, I noted that they did the same thing: plotted gun deaths against gun ownership state by state and claimed that more guns resulted in more deaths. But if you looked at total suicide and homicides, the picture was far murkier:
I can’t embed the graphic but when you look at the total violence rate from all methods of killing — using the same sources they link — the correlation is not nearly as strong (R^2 of .13) The trend is 0.10 for every percent. So eliminating ALL guns — even if you assume that there is no increase in criminality — would reduce the death rate to about 14.8 or basically as peaceful as Iowa with its 44% ownership rate and Rhode Island with its 13%.
Update: Eugene Volokh revisits the issue here and comes to the same conclusion.
Well, I can embed graphics here. So here is what you get if you plot up the rate of violent death against gun ownership for all the countries with a high Human Development Index (except Liechtenstein, Andorra and Hong Kong, for which I could not get complete data)
Now you see my point. There is no correlation whatsoever between gun ownership and violent death. If anything, there is a slight anti-correlation (which is basically caused by a few Baltic countries having low gun ownership rates and astronomical suicide rates).
Suicide is the key here. The suicide rate in the developed world is six times the homicide rate. In the United States, twice as many people take their own lives with a gun as take someone else’s. And more honest gun control advocates will tell you that, even if gun control didn’t stop murders, it could prevent suicides. Suicide tends to be impulsive and guns are a much more lethal method than just about anything else.
But the number belie the assumption that more guns automatically mean more suicide. When looking at this data, it was blindingly obvious that there is no correlation, on a country-by-country basis, between suicide rate and gun ownership. No, the difference driving suicide rates is cultural. Lithuania and South Korea, for example, have minuscule rates of gun ownership and very high rates of suicide. Many Middle Eastern countries have high rates of gun ownership but extremely low rates of suicide. That’s not because Kuwaitis are practicing great gun safety and South Koreans are careless. That’s because South Koreans are massively more likely to kill themselves. The gripping hand is that 99.8% of the time, they use something other than a gun. So the plot favored by the gun control advocates literally ignores 99.8% of South Korea’s problem with violent death.
Does that seem reasonable to you?
(Homicide rates show a slight trend of increased rate with increased gun ownership. But it is extremely weak, with an R^2 of .007. That’s noise. My data do not include accidental deaths but those constitute a very small percentage of gun deaths, about 1.5% in the United States.)
I want to be very clear on this: my point is not that there is no correlation between the presence of guns and violent death. If we had fewer guns, maybe fewer people would kill themselves. If Japan had lots of guns, maybe their suicide rate would be even higher. I’m not addressing that. My point is that this specific talking point is irrelevant when it comes to gun control. Guns are not the biggest factor in violent death. Culture is the biggest factor and it’s not even close.1
And that brings me to my point. People keep asking me why the United States is such a violent place compared to say, Canada. Surely, they say, it must be because of our gun culture. I think they’re right about the second word. It is culture. We have a culture that glorifies violence. I don’t mean necessarily in video games or rap songs. I mean in real life. I mean in inner cities, where violence is ubiquitous and role models are non-existent. I mean on an institutional level, where two million people are in prison, 80,000 SWAT raids are launched very year, a thousand people are killed by police and God knows how many are roughed up. I mean on a commentary level where we are very casual about just how many people get killed in a war. I mean on every level. Every day. We act as though human life is cheap. And then we act all surprised when young mean act as though human life is cheap.
Guns and gun violence are a symptom. The disease is our culture. It’s a disease that’s getting better. As I’ve noted many times, violence is way down from it’s awful peak in the 90’s. There are brave people trying to bring sanity to the war zones in our inner cities. As I said in my Sandy Hook post, there are probably dozens of mass shootings prevented by a trouble man getting help. But the amount of violence in our society is still way too much. I think there is a lot we could do to help (better mental health services, more community-oriented policing, ending the War on Drugs). But grabbing guns would be very low on my list.
1. Here’s a thought experiment to illustrate this. Ask yourself this question: would you rather be locked in a room with ten unarmed convicted murderers or ten armed law-abiding NRA members?↩
Update: German Lopez graciously responded to my criticisms on Twitter. You can see the exchange here.
His substantive criticism is that there is a lot of academic research claiming this (some of which I’ve addressed before) and that if you look at only the 25 highest HDI countries, you see a clear trend of homicide rate with gun rate. The latter is true … mostly because of the United States. If you remove the US from the trend, it mostly disappears. So this amount to basically saying the US has a lot of guns and a lot of murders. That’s true enough.