One of the most thoughtful and numbers-driven posts I’ve read on gun control is Guns and States from Slate Star Codex. It’s a long article, but well worth the read because it shows how not simple this issue is – yet we can draw some tentative conclusions.

Here are his conclusions, with my headers and commentary:

1. Suicides drive most graphs but should be handled differently than homicides.

Scatterplots showing raw correlations between gun ownership and “gun deaths” are entirely driven by suicide, and therefore dishonest to use to prove that guns cause murder (~100% confidence)

MY COMMENT: This skewing of the data muddies the discussion about homicide. We want to reduce both, but analyzing them together makes it harder to come up with a real understanding of the relationship between guns, culture, mental health, and death. Since 2/3 of gun deaths are suicides, we should look into that in parallel, not all bunched up with homicide stats (NOTE: the image to the far right makes it look like homicides go DOWN with gun ownership – but this is an artifact of some of the bad data bandied about – but it also shows how much suicides skew the murder rates).

2. But to a moderate (not high) extent, more guns = more homicides

But if you adjust for all relevant confounders, there is a positive correlation between gun ownership and homicide rates (~90% confidence). This relationship is likely causal (~66% confidence).

MY COMMENT: By “causal,” he means that people with guns are more likely to kill someone, as opposed to the opposite – that people in dangerous areas buy guns to protect themselves, and are therefore more likely to shoot someone (not that this may also be true, but not sure which is more important).

3. The major factor in the US is not guns, but a culture of violence especially among urban blacks (and worse in the South)

Murder correlated with black populations in the US

The majority of the difference between America’s murder rate and that of other First World countries is not because of easier access to guns in America (~90% confidence).

MY COMMENT: This is an incendiary remark, but again, is anyone really surprised by this? This doesn’t mean there aren’t white, violent racists out there, but it does mean that urban blacks have a significantly worse problem than other American populations, and their violence, not gun ownership, primarily drives high murder rates in the US.

In trying to pinpoint possible causes of violence, many conservatives point to the idea that roughly 70% of mass-shooter crimes are done by men with no father, noting that “There’s a direct correlation between fatherless children and teen violence.” 1 2

Marriage v. Childbearing

And no other subpopulation I know of has such awful stats regarding the family – blacks during the 60’s had much stabler families, even though racism was much worse then. Currently, 77% of black children are born out of wedlock (which correlates with 54% of black children living in a single parent household), and fully 41% of black pregnancies end in abortion (up to 70+% in NY and TX). Again, this may be driven by poverty, but let’s not hide in that fact and become passive – we all must take action, not just the government, not just whites, and not just blacks. 3 4 5

4. Easy access to guns increases murders, no doubt

But some of it is due to easier access to guns. This is probably about 0.5 murders/100K/year.

MY COMMENT: There is no doubt that increased gun control would reduce gun deaths, of course. But how much is the question. It is not the most powerful tool we have – since black urban youth are the main murder offenders (white mass shooters are more visible but actually kill many less victims), our best efforts might be spent on these interventions: 6 7

  • Probation strategies – increased contact with police, probation officers, and social workers
  • Public reports of gun stores whose sales were linked to crimes
  • Smart gun technology (weapons that only fire with a PIN or fingerprint)
  • Microstamping of bullets to allow bullet casings to be traced
  • Liability insurance for guns
  • Supporting fatherhood and marriage in society – most of the mass shooters come from homes of divorce.

5. Australia’s buyback was partly successful

An Australian-style gun control program that worked and had no side effects would probably prevent about 2,000 murders in the US. It would also prevent a much larger number of suicides. I am otherwise ignoring suicides in this piece because discussing them would make me too angry.

MY COMMENT: 2000 murders per year and about twice as many suicides are worth preventing, but at what cost? There is some indication that this approach exchanges a lot of money for a drop in the bucket (reminds me of climate change legislation), and may not even work in the US. Criticisms of Australia’s buyback program include:

  • Reducing mass shootings to 0 does not affect the overall murder rate greatly ( 8
  • Buybacks barely move the needle, if at all (esp. since the rates were already falling)  9
  • People sell non-working guns to buy more guns!
  • Buying back 1/3 of the guns in the US would be expensive and practically impossible 10 11
  • A violent black market in guns was created by the Australian approach ( 12
  • Buybacks are soon replaced by more gun purchases, returning the gun ownership rate to what it was before buybacks (The Atlantic) 13

6. Americans enjoy guns – too bad about their loss of pleasure (and liberty)

Probably the amount of lost gun-related enjoyment an Australian-style gun control program would cause do not outweigh the benefits.

MY COMMENT: This is a very subjective point he makes, but to some extent, I agree. IF gun control can make a significant difference. It can make some. Not sure how much, really.

7. Sometimes correlation is better than experimental evidence

Thomas Malthus, father of catastrophic thinking

This is not really enough analysis to make me have a strong opinion about gun control, since this just looks at the correlational evidence and doesn’t really investigate the experimental evidence. Contrary to what everyone always tells you, experimental evidence doesn’t always trump correlational – there are cases where each has its strengths – but it wouldn’t be responsible to have a real opinion on this until I look into that too. Nevertheless, these data are at least highly consistent with Australia-style gun control being a good idea for the US.

MY COMMENT: Well, now he’s backpedaling a little – who cares about causation, or real world effects of expensive programs? Again, this sounds like climate change proponents – “Sure, it may not be a crisis, and these draconian measures will probably do very little, but let’s do this anyway because we’re scared, and doing nothing is worse.” But is it? Maybe draconian measures ARE worse than the calamity you suppose because people die when you divert resources to chicken-little Malthusian catastrophes!


If we are really concerned about gun violence and deaths in America, we will probably avoid less effective methods like gun buybacks and increased gun control, and focus on interventions that work. The culture of violence in poor, especially black communities. The epidemic of single parent and fatherless homes. Prison and parole reform. And perhaps even requiring annual licensure, training, and certification for gun ownership. And if you want to ban the most dangerous guns, ban handguns. But don’t.