Is Arming Yourself a Rational Response to a Mass Public Shooting?

I haven’t had time to write this up yet, but an underlying theme in academic studies of guns is that carrying a concealed firearm (and in some cases even owning one) is not a rational decision for most Americans under most circumstances. “Rational” here understood in the classical economic sense of maximizing benefits and minimizing costs.

Even less rational, then, would be arming yourself for self-defense under extraordinary or exceptional circumstances. Like after a mass public shooting. And yet we see interest in concealed carry spike after mass public shootings time and again.

John R. Lott, Concealed Carry Permit Holders Across the United States: 2017, p. 27.

The official statistician of the gun rights movement, John Lott, is not alone in suggesting this connection. For my Sociology of Guns seminar last week we read an article that systematically connected high profile mass public shootings with concealed carry permit applications in Tennessee.

“Reacting to the Improbable: Handgun Carrying Permit Application Rates in the Wake of High-Profile Mass Shootings” was published in 2017 in the academic journal Homicide Studies.

The authors looked at the effect of 10 specific (systematically identified) high profile mass public shootings from 2008 to 2014 on the rate of concealed carry permit applications in 95 counties in the state of Tennessee, controlling for a host of other factors that could effect permit applications.

In the end, they found that these 10 high profile mass shootings were related to an 11% increase in the rate of permit applications (with a 95% confidence interval of 9% to 13%).

Screen cap of “Reacting to the Improbable,” from

Of course, the study is not without limits, and the authors go on for page after page specifying alternative statistical models and acknowledging shortcomings. But the core finding, that some significant part of the population will seek to arms themselves in the wake of a mass public shooting, as opposed to calling for citizens to give up their arms, is significant.

These diametrically opposed responses — disarmament now! vs. arm yourselves now! — reflect the differing views of risk and safety that exist in American society with respect to guns, and different understandings of the role of government (including law enforcement) as opposed to individuals in mitigating risk and ensuring safety (Kahan and Braman).

Rational or not, many people continue to opt for “DIY security citizenship” (Caroline Light) under which they serve as “law-abiding one man armies” or “mobile sovereigns” (Elizabeth Anker).


  1. Great post. THanks for highlighting the study.

    In light of Broward County proving that the police will not protect you, and several court decisions stating that police aren’t obligated to protect citizens, many will choose to select better protection tools. Guns are affordable for most. Personal security details for the better heeled, firearms for the working folks.

    Man may not be the utility maximizing homo economis being that undergrads are taught. Man is interested in surviving and thriving. Most will learn and adapt to their environment to thrive. In today’s environment carrying a gun, saving a little money, learning new job skills, etc all make sense.

    Whike utility maximizing is a useful lens, often assessing incentive and opportunity cost is more useful. In the case of arming oneself the incentive is made clear by recent police actions. And the opportunity cost of not doing so may be extremely high in Black Swan cases like workplace murder attacks. The decision is easily understandable.

    That said, protecting schools should be at least as important as protecting a football game. Physical barriers to control access and in-site armed security are necessary first steps. Yes, they are expensive. If children are the future, then what’s it worth to protect them in a meaningful way?

    Schools are already Gun free zones. More anti-gun laws won’t mitigate the threat.

    Liked by 3 people

  2. Reblogged this on and commented:
    Police are, AFTER THE FACT.
    Don’t forget that there are many high quality used firearms at gun shops. If properly licensed/permit holders, people can find outstanding guns with holster wear and very little use, such as police revolvers, since the semi-auto craze took hold, and the revolvers turned in, are usually cosmetically worn but real sweet shooting and accurate. Another thing is that in hand-to-hand combat, the revolver is able to be used as brass knuckles, with the crane opening browlines and allowing blood to go into an aggressor’s eyes. When the police finally arrive, identify yourself and let them know what happened. Understand this: If you hear the words, “PO-LEECE! DON’T MOVE!”, don’t move. If you are ordered to, “DROP THE GUN!”, drop it from your hand. An old revolver has little worry of another scratch on the surface.

    Liked by 1 person

  3. Am I missing something, or does the first graph you provide fail to show a correlation between googling for “concealed carry” and mass-murder events? In every case except for the Paris attack, the curve is already rising sharply before the line indicating the attack event, and and in those cases, the curve does not appear to rise any faster after, and in the case of Orlando and Newtown even drops off a bit.

    Not that this invalidates the overall research, nor the intuitive response that yes, mass murder events will trigger interest in concealed carry (regardless of the rationality of such an action).

    Liked by 2 people

    • That is a striking observation. It would be interesting to see if the association persists with more events.

      If not random, what could cause this? Some subtle but significant societal stress that triggers both the shooter and an upswing in concealed carry interest that then continues and is perhaps amplified by the mass shooting event?

      Liked by 1 person

    • Good catch. As R in NC suggests, probably need to see what else might be resulting in spikes. Note at other points similar steep spikes occur, then drop to a “normal” rate. Perhaps it’s a matter of timing, a “spiking factor” occurs and then a mass killing “catches the wave” and blows it up higher and creates, as the spike recedes, a new higher mean.

      It’d be interesting to take it back further and pull it forward, see if that “stair step” trend continues each way. Prior to San Bernadino it almost looks like it was falling back to that 1st line mean. Maybe track major news reports to see if their release resulted in increased searches prior to those spikes?


      • I highly doubt that as well. In fact, as MatthewCarberrryBlog pointed out, you can see other spikes of similar curve that don’t have a mass shooting, and then often subside back towards a baseline, while the ones associated with a mass shooting continue to rise to quite high levels.

        It could just be a coincidence. I doubt that there is some shared causal property that drives CC google searches and mass shooting events (although maybe there is).

        Liked by 1 person

      • Could be an error in reporting interval. It makes sense to hypothesize that mass shooting events trigger various responses, whether calls to ban guns or people arming up for Armageddon (or buying another brace of AR’s in fear of gun bans). But those graphs are suspect. Then again, not the first time that Dr. Lott has been in hot water about his research.

        Paper is behind a paywall, so I’m moving on.

        Liked by 1 person

  4. Mr.Pincus has great thoughts re: probable v. possible.

    Is carrying rational? If merely weighing cost/benefit of a highly unlikely event- no.

    What I tell students is
    that I will bet a year’s salary I will never need to pull my gun in defense. What I won’t bet is my daughter’s life.

    It is nothing but catastrophic insurance. The likelihood is roughly nil …but the downside should it occur is too steep.

    Liked by 2 people

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.