« We Can't Handle The T… | Home | Lies, Damned Lies and… »

Lies, Damned Lies and Anecdotes

A lot of people--like this guy--think that Mark Twain has more to say about policy discussions then David Hume, mainly because he was able to coin witty aphorisms like, “There are only three types of lies: lies, damned lies, and statistics.” (Quote behaving badly? Kind of. Twain popularized the phrase in Chapters From My Autobiography, but didn’t coin it. He attributed it to Benjamin Disraeli--On Violence’s fourth favorite British Prime Minister--but the phrase never appeared in Disraeli’s writings. More here.)

How true, and how wrong.

Eric C and I like to consider ourselves as skeptics, towards both sides of the political spectrum. We also love debunking bad quotes, wrongs facts and common wisdom. That is why we think that the quote should read, “There are three types of lies: lies, damned lies, and anecdotes.”

Sure statistics can be misleading. Take the gun control issue, for example. On this episode of Intelligence Squared, “Do Guns Make Us Safer?”, both sides come to the debate armed with statistics; neither has absolute proof. Rachel Maddow summed the situation after the shooting in Tucson last month, “In 2004, a national blue ribbon panel of statistics and criminology experts looked into whether or not ‘right to carry’ laws reduced crime in states that have them. Their conclusion famously was, they don‘t know. They were completely unable to come to any statistical conclusion about it at all.”

Statistics cloud the gun control debate, but anecdotes ruin it. Maddow opened that segment above with an anecdote, responding to a congressman’s anecdote. Or consider this portion of a letter written to the Army Times called, “Let Soldiers Carry Arms”:

       The concealed carry of firearms is against the law on all military bases. Why are those sworn to protect our nation against all enemies prohibited from carrying firearms by shortsighted and risk-averse policies of the past? Our families, units and friends sit defenseless on post. I thought this policy would have changed following the Fort Hood, Texas, massacre, but it has not. Does someone think soldiers are going to shoot at the military police? Are we going to road rage and shoot fellow servicemen and families on post? Are we disarming the whole military to bring the risk of accidental discharge down from .003 percent to zero?

       How absolutely ridiculous do these reasons sound after one looks at the carnage and sorrow caused by just one villainous traitor at Fort Hood? Would 13 have died at Fort Hood if even 5 percent of the soldiers in that room had been armed? Would the shooter even have carried out his crime if he knew that his victims would not be defenseless? Maybe not.

Would the alleged Fort Hood shooter Nadal Hassan have been stopped if soldiers could carry guns on post? Who knows? Again, anecdotes muck up both sides of the debate. For every mass casualty event like Virginia Tech or the Fort Hood Shootings, there is an example of a family member shooting another because they owned a gun. Complex issues like gun control can’t be solved by anecdotes.

Gun violence is an incredibly tricky issue, for both sides. Statistics can help us solve it; anecdotes can’t. Of course, lies, damned lies and anecdotes have crept into the foreign affairs and national security arena too. On Wednesday I’ll tackle that subject.

thirteen comments

I think anecdotes just ruin public debate. If I hear someone arguing something with only anecdotal evidence, I don’t listen.

Statistics can be manipulated, but they can also prove things. There is the possibility that statistics will eventually solve the gun control debate. As one political scientist noted—and we lost the link—many states in America have no gun control laws, meaning eventually the statistics will tell us if more guns make us safer.

Unlike anecdotes, where after every public shooting both sides of the debate claim their side is right.

I wouldn’t trust much that Maddow says.

I remember an email going around of a man (attempting to rob) a convenience store only to have a police officer enter behind him. The accompanying line was stating how stupid the man was and how the event would have gone the same way if a civilian entered with a gun. All I took from the event was that we could use more police on the streets.

That is the question matt, would we be more secure with more armed americans? Tough to say. I think both sides has compelling arguments. I mean, people would be less likely to use violence, but then people at bars who are drunk are stupid, and I don’t want them to have guns. So maybe the same laws that apply to driving while drunk should apply to concealed weapon carriers?

Why did murder rates decrease when CCW laws were passed?

I’d ask you for proof, but I have no doubt you’d come up with a statistic. and I could come up with a statistic disproving that statistic.

I feel both sides have statistics proving gun control works, or doesn’t work. The debate is too entrenched.

I just have FBI crime numbers. Good as any official source I’d say. Specifically, Texas and Florida.

Anti Bill of Rights people said there’d be a bloodbath in FL is CCW was passed… the opposite happened. Whoops!

I’m all for CCW permits on a person by person basis. Just like I’m all for gun ownership on a person by person basis. CCW permits should obviously be more restrictive than normal gun laws because of the inherent potential problems that can arise from concealing a weapon. But suggesting (as in above anecdote) that the presence of a gun makes a situation safer is biased logic and borders on hubris.

Both sides have statistics. Maddow goes on to point out that the states with the highest rates of gun deaths correlates with gun deaths. The FBI’s numbers wouldn’t exclude the overall nation wide decline in crime.

There are numbers on both sides, and flaws with all those numbers.

The problem with the CCW debate is someone can just illegally carry a gun if they really wanted to commit a crime… why go thru all the trouble to get a permit if you’re a criminal?

Also, Lefties have great difficulty explaining how historic highs for firearm ownership and record breaking unemployment and home foreclosures still place murder rates down than where they were 10 years ago.

I think the gun debate is too complicated for simple statistics. I mean, each side has their evidence. Check out the Intelligence Squared podcast, each side has their evidence. The bigger point with this post is that we need to use evidence and statistics, not anecdotes.

Like these statistics?

In the 1980s and ’90s, as the concealed-carry movement gained steam, Americans were killed by others with guns at the rate of about 5.66 per 100,000 population. In this decade, the rate has fallen to just over 4.07 per 100,000, a 28 percent drop. The decline follows a fivefold increase in the number of “shall-issue” and unrestricted concealed-carry states from 1986 to 2006.

[T]he FBI reported a surge to 10 million in 2006, 11 million in 2007, nearly 13 million in 2008 and more than 14 million last year, a 55 percent increase in just four years.


Both sides definitely have statistics.