Ever since Art Kellerman and Fred Rivara published research in 1993-94 which definitively linked gun access to increased homicides and suicides, the issue of whether guns protect us from harm or create more harm has been the basic dividing-line between the two sides.

              The ‘guns keep us safe’ argument is identified most frequently in the work of our friend John Lott, whose book, More Guns, Less Crime makes the connection between the issuance of concealed-carry licenses and violent crime rates to argue that as the former number goes up, the latter number goes down.

              The ‘guns make us less safe’ argument is identified most frequently in the work of our other friend, David Hemenway, who makes the connection between per-capita gun ownership and violent crime rates to argue that as the former number goes up, the latter number goes up as well.

              I happen to believe that both of my friends know a lot about statistics and how to use data to develop interesting arguments based on regression methodologies. I also happen to believe that neither of my two friends know anything about guns. Because if they did, they would never use only the numbers on murders committed with guns to make their argument, regardless of whether or not they argue pro-gun or anti-gun.

              Why do both John and David restrict their definition of gun violence only to the number of people who end up dying because someone else shot them with a gun? Because they can’t get good data, verifiable data on the number of people who are shot by someone else but still manage to survive the event.

              The CDC never had good numbers on injuries from non-fatal gun assaults, their estimates were, at best, off by more than thirty percent. And a few years ago, they stopped publishing any data on non-fatal gun assaults, so the whole argument about gun violence rests only on the number of people who are shot and killed.

              And this is where the analysis of gun violence, either pro-gun or anti-gun, comes up short when it’s being done by well-intentioned researchers like John Lott and David Hemenway, neither of whom know very much about guns.

              Here’s the bottom line: The only difference, repeat, the only difference between fatal and non-fatal gun assaults is that in the latter instance, the guy using the gun didn’t shoot straight. He didn’t shoot straight because either he was pointing the gun at a moving target, or the event took place at night in low light, or he just hadn’t practiced enough to hit the target where the bullet would hit a vital spot.

              Guys walking around with guns don’t ever pull their gun out with the intention of shooting someone else in the leg, or the arm, or some other non-vital spot. The gun is yanked out, and if it’s a semi-automatic pistol the trigger will be pulled again and again, and either the victim goes down or he doesn’t go down. That’s the end of that.

              Both Hemenway and Lott are making arguments about guns and violent crime that don’t even remotely capture the reality of gun violence in the United States. So, for example, John Lott recently published an op-ed in which he compared the total number of homicides, with or without a gun, against the estimates for defensive gun uses and found the latter to outnumber the former by four or five to one.

              But if John had compared defensive gun events to the total number of gun assaults, fatal and non-fatal assaults, his argument about the value of guns used for self-protection would collapse. By the same token, David’s comparison of fatal gun violence to per-capita gun ownership is equally invalid, for the simple reason that most of the guns in America’s civilian arsenal happen to be guns designed for hunting and sport and never (read: never) end up being used in any kind of gun violence at all.

              If it were the case that the research published by my two friends never went beyond an entry on their CV’s, I wouldn’t really care what they said or didn’t say. But this research is what is used by both sides in the gun debate to promote and/or justify their ideas about what we should do to reduce the number of injuries and deaths cause by the use of guns.

              Know what happens when you create a law to regulate a problem but don’t understand what the problem is all about? The law has no real effect at all.

              Gee – what a surprise that gun violence keeps going up, not down.