Anatomy of a Scary Statistic

One of the things that I find interesting is the power and resonance a particularly statistic can have on a person based on their preexisting beliefs.  There are two dimensions to this that I’d like to explore over a couple of posts.  The first, and the subject of this particular post, is how a single statistic be used as be used as rather convincing evidence to bolster two opposite positions.

Before we get to the statistic I am referring to, let’s start with the background.  Recently I watched Katie Couric’s new gun documentary, Under The Gun.  For those not aware, there is some controversy surrounding the film.  Essentially, Couric asks members of the Virginia Citizens’ Defense League the following question, and then close to 10 seconds of silence and blank looks were edited in giving the impression that it is a powerful question that they are unable to answer:

If there are no background checks for gun purchasers, how do you prevent felons or terrorists from purchasing a gun?

In reality, the members immediately come up with answers.  Setting aside the ethics of the deceptive edit, one of the benefits of cutting the scene that way is that it allows that question to linger in the viewer’s mind uncontested – What would happen if there is nothing to stop these dangerous individuals from buying guns?  Later in the film they revisit part of this question with a narrative that, if taken at face value, is meant to be absolutely terrifying:

One of the issues that’s come up a lot in the last few years is that anyone who is on the terror watchlist is not prohibited from buying a gun

They then show a narrator asking a series of NRA members how they feel about a person on the government’s terror watchlist not being able to board a plane, but being able to legally buy a gun1.  Of course, the responses they show are from people who are dumbfounded by the question.  After all, what possible response could there be to why extremely dangerous individuals should be able to legally buy deadly weapons?  Finally, set up complete, they hit you with the statistic:

From 2004 – 2014 over 2,000 terror suspects legally purchased guns in the United States

The “over 2,000” turns out to be 2,043 and comes from a March 2015 update to Senate testimony from 2010 by the Director of Homeland Security.  Here is the full statistic which covers February 2004 through December 2014:

Overall, since NICS started checking against terrorist watchlist records in February 2004, FBI data show that individuals on the terrorist watchlist were involved in firearm or explosives background checks 2,233 times, of which 2,043 (about 91 percent) of the transactions were allowed to proceed and 190 were denied.

Taken at face value, it is not hard to see how this could be viewed as an outrage and worthy of fixing.  Reading through the reports, however, I noticed something odd seemed to be missing.

If you are constructing the case for why we need to close the “Terror Gap” as it is sometimes referred to, you would expect the argument to have at least three main parts:

  1. Description of the problem – check
  2. Size of the problem – check (2,043 over 10 years)
  3. Damage the problem has caused

That last part is a fairly key detail.  If we want to prophylactically remove an enumerated constitutional right from people, it is pretty important to demonstrate the harm that the current system is allowing.  In the strongest form of the argument we would be shown that in each of the 2,043 instances someone on a watchlist legally purchased a firearm2 and went on to kill with that weapon.  This would make for an incredibly strong case for new legislation that would be hard to argue against.  If someone went on to kill with the weapon in 50% or more of the instances, the argument would be weaker, but still fairly compelling.  If, similar to the issue of gun ownership in general, only a very small portion of the total instances in question resulted in a death, the argument would be much weaker, but there would still be a justification that could be made.  However, none of that appears to be the case here.  I am not able to find evidence of any harm whatsoever resulting from the 2,043 transactions that were allowed to proceed3.

When you think about, this is really quite remarkable.  If even a mere 1% of the time terror suspect’s background checks were allowed to proceed ended in tragedy there would be 20 examples you could point to of harm that could be prevented.  Instead, there does not seem to be a single one.  Many times when arguing for or against proposed legislation we don’t really know what the result would be if the legislation were in place earlier and are left to speculate.  Here that is not the case.  We know exactly the transactions that the legislation would seek to block.  Given this, the argument for Terror Gap legislation essentially boils down to “This group of people we think is very dangerous should not be allowed  to exercise one of their constitution rights because it will put the rest of us in great peril. Peril so great that in over 2,000 cases, some going back over decade, of transactions it is critical we block, we haven’t been able to identity any harm that has resulted.”  There is crying wolf and then there is going 0 for 2,000.

It is certainly true that just because nothing has happened yet, that is no guarantee that something will not happen in the future.  However, the current situation seems designed to work rather well.  If someone on the terror watchlist attempts to purchase a firearm, the FBI is alerted and can investigate further what danger is posed.  Blocking the transaction would not only cause due process concerns, but would allow a terrorist attempting to orchestrate an attack to find out they were on a watchlist.

And so we are left with a situation where the same statistic is both evidence of the need for new legislation as well as the perfect argument against such legislation.  To those who are in the former camp, a parting question: What other rights should we be able to remove, for the safety of society in general, from people that President Trump’s administration would put on a list of dangerous individuals?

  1. It should be noted at this point that the No Fly List is only one component of the larger Terrorist Screening Database so not everyone on the terror watchlist would actually be prohibited from boarding a plane
  2. The report notes that “The FBI does not know how often a firearm was actually transferred or if a firearm or explosives license or permit was granted, because gun dealers and explosives dealers are required to maintain but not report this information to the FBI.”
  3. Despite the best efforts of organizations such as Center for American Progress and Everytown for Gun Safety to imply otherwise with lists of the attacks that they would like you to falsely believe are part of the 2,043

One thought on “Anatomy of a Scary Statistic

  1. I was thrown by the quote in footnote2 because explosives dealers are required to report sales to limited permit holders. The quote is technically correct, transactions are reported to the ATF and not the FBI, but misleading.

    This got me thinking about explosives license holders, one explosives dealer on the list with 49 employees would show up on (1 license + 49 employee liscense times 4 (licenses must be renewed every 3 years and the report covers almost 11 years) = 200) about 10% of the flagged background checks. While this might not mean anything, the fact that Homeland Security combined firearms and explosives background checks (different forms and procedures for the checks) indicates that Homeland Security could be manipulating the numbers by including explosives license and permit checks.

Leave a Comment