Recently, a friend of mine posted a meme from the Prepare to Take America Back Facebook page about a gun dealer who has bacon in his shop, and if a prospective gun buyer intends on buying a firearm from him, you have to eat the bacon. The purpose of course, is to prevent Muslims from buying guns.
A lengthy discussion ensued, so I felt this was a good opportunity to promote skepticism over ideology and point out the flaws in the arguments by analyzing both sides.
The Actions of the Dedicated
If someone is so delusional as to want to murder a number of people at will for their god, it stands to reason they are not subscribing to a rational mindset. They are highly dedicated to an end result, and nothing other than a good person with a gun is likely to stop them. So I’m pretty sure if they’re motivated enough to murder, they could easily justify eating a piece of delicious bacon for the cause. It is likely only rational non-violent Muslims would be restricted from buying guns in this manner.
Remember, they’re not supposed to look at naked women either, but when Bin Laden was killed, he had quite the porn stash.
I should also point out that many gun owners have come out against No-Fly-List restrictions on gun purchases because a few innocent people end up on that list. So preventing law-abiding Muslims from buying a gun just because of the actions of a few violent ones seems rather hypocritical.
The Gun Rights Paradox
Gun rights advocates like myself point out that while it may be true that guns are the #1 tool used to murder people around the world, it’s a flawed argument if you’re using it to argue guns are likely to kill. There are nearly 80 million gun owners in America, but only approximately 32,000 violent incidents are performed by such people. Meaning that for every 100 gun owners, approximately 99.96 of them will harm no one who was of no threat to them.
While the numbers might be slightly different, you could replace the term “Gun Owners” with “Muslims” and make the exact same argument.
Like gun owners, most Muslims are indeed non-violent. So for gun owners fighting for gun rights by pointing to the above statistics to be ideologically consistent, they shouldn’t be promoting anti-Muslim views either.
The Constitutional Argument
The bacon scheme, while clever, many argue is a violation of the 1st amendment that seeks to prevent religious discrimination. But if we look at the verbiage of the first Amendment, it should be obvious it’s not an issue.
Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the government for a redress of grievances.
The first five words are, “Congress shall make no law.” As this is a private business owner, he’s not congress. The first amendment restricts government and protects him, not the other way around.
Aside from the Constitutional issue, the government may not discriminate because we all pay our taxes to it and it governs all of us equally, therefore we deserve equal protection under the law.
But whether it be the KKK, Black Panthers, Westboro Baptist Church, a Christian bakery owner not wanting to make a cake for a gay wedding, or this gun shop proprietor, in a free country, while government may never discriminate, they should never have the power to dictate who you are kind to or do business with. Let the free market sort it out.
The conversation that ensued on this meme was after a mutual friend commented “94% of terrorist attacks in America are committed by non-Muslims, look it up.”
I, of course, pointed out that since he was the one making the claim, the burden of proof was on him to cite his source, it was not our burden to look it up. So he cited this information from Global Research.
Global research sounds mighty official, but then they cited a graph they stated came from Princeton University’s Loonwatch. Princeton university, being a prestigious institution, should lend some credibility as well. But there’s only one problem—Loonwatch’s “About” page only cites Princeton as the source of the definition of the word Loon from Princeton’s WordNet® 3.0. They may have attended Princeton (they don’t say), but there’s no indication this info is from Princeton University in any official capacity.
Since Loonwatch didn’t compile the data, this makes Global Research’s citation of Loonwatch irrelevant.
As you read the about page, it becomes clear, Loonwatch are opinion bloggers just like me, with no intrinsic credibility that comes from being a well-respected institution or peer-reviewed publication.
Opinion writers only get credibility by citing credible sources, as we don’t compile any of the data ourselves, we merely interpret it. But the genetic logical fallacy requires that we not dismiss their opinion, even if they’re not necessarily a credible source, so we’ll soldier on.
Loonwatch made a graph based on this FBI.gov data, which is a credible citation and to be commended. The thing that differentiates me from Loonwatch is that I won’t be pushing a particular narrative. I will present multiple ways to construe the data so no context is missing. Loonwatch failed to do this, and thus why I’d argue my post is more fair in its analysis.
Loonwatch did little to show how they came to their conclusion. The FBI study, cites individual attacks and who was deemed responsible for them, but did not in any way segregate them into the convenient categories Loonwatch used on their graph, so I can only guess that maybe Loonwatch researched each group deemed responsible individually, and categorized them by categories of Loonwatch’s choosing. While there’s nothing wrong with that, it’s important that Loonwatch at least explain their methodology, which they didn’t.
So don’t take anything Loonwatch or I say to the bank. Look at the FBI Data provided, and come to your own conclusions. I just hope to promote critical thinking.
The problems with the 94% statistic are numerous.
The first flaw is that it breaks the groups up into categories that aren’t mutually exclusive. For instance, you could have Latino Communists, so what group do they fall in on the above chart, Latinos or Communists? And wouldn’t Communists be considered an Extreme Left-Wing Group as well?
Second, the caption they have for the graph reads as follows:
Terrorist Attacks on U.S. Soil by Group, From 1980 to 2005, According to FBI Database
But Loonwatch’s groups are not how the FBI classified them. The FBI classified them by name, such as Al Qaeda, versus grouping them as Muslims like Loonwatch did, making the caption dishonest as they aren’t the groups “according to” the FBI as the caption states. That doesn’t mean Loonwatch’s interpretation of the data is inaccurate, but when people make false assertions like that, it talks negatively to their credibility, as they’re either being dishonest or sloppy in their work.
Thirdly: It counts each attack as one incident out of 316, no matter how many were killed or injured in that incident including many that resulted in no death or injury at all.
This means that they count the 9/11 attacks which officially killed 2972 people and injured an estimated 12,000 others as if they’re somehow one unit equal to the November 11th 2005 Hagerstown, MD arson which killed no one.
Loonwatch headed their post by saying, “Terrorism Is a Real Threat … But the Threat to the U.S. from Muslim Terrorists Has Been Exaggerated.” As such, including incidents which resulted in no human harm, is certainly a bit misleading. The FBI was simply tracking terror attack numbers, but Loonwatch used that data to argue threats to the U.S., which aren’t quite the same thing. The non-injurious attacks may or may not have been intended to harm anyone (a threat), but only scare people into compliance (terrorism). It’s entirely possible those attackers purposefully sought to avoid being a threat to life and limb by bombing unoccupied property, effectively making them non-threats.
Analyzing the data myself
First, let’s eliminate the aforementioned incidents that resulted in no harm to anyone and we’re left with 44 attacks versus 316 to analyze. I’m eliminating these because the narrative is about who is a threat to Americans, so incidents which resulted in no harm should be irrelevant. I researched every group responsible individually to categorize them myself and determine which were Muslim and not.
Muslims committed 6 of those 44 attacks, or 14%. This is more than double the 6% Loonwatch presented, using their same metric. But, it still supports their underlying argument that non-Muslims committed more attacks than Muslims, by far.
After I had done that, instead of treating each incident as if they’re the same, I’m going to categorize them by how many were killed at the hands of terrorists, which is more relevant to the narrative of the threats to Americans.
Of 3,178 terrorist murders, Muslims committed 2,982 of them (94%), which is ironically (and completely coincidentally) the same percentage, yet polar opposite, of the narrative Loonwatch portrayed. There were approximately 13,048 Muslim-committed injuries out of a 14,017, (93%) as well.
Now that may seem like I’ve refuted Loonwatch’s argument since that’s a 188% swing, but I haven’t. I’ve merely presented the same data in a different light.
To be fair, I will also point out that almost all of them are from the September 11th attacks. So one incident of 44 is severely skewing the data. But nonetheless, while Muslims don’t account for most of the incidents, by a landslide they account for the most deaths.
Using the same data Loonwatch did, I could make that argument, leave out the context I gave you, and give a conversely biased opinion to Loonwatch. It’s a lesson in how people leave out info without lying to lead you into a false impression.
What’s This Puerto Rico Stuff?
While we’re on the subject of skewing the data, I could eliminate the events in Puerto Rico as well.
While Puerto Rico is a U.S. Property, I think if you asked both Americans and/or Puerto Ricans whether they consider Puerto Ricans to be Americans, most would say no. They’re not a state, plus they’re not even allowed to vote in U.S. general elections. Again, the narrative was whether Americans are mostly under threat from Muslims, so adding Puerto Ricans to the list is a bit misleading to that narrative for most Americans
Eliminating non-injurious and now Puerto Rico attacks, I have 35 remaining incidents, of which Muslims were responsible for 6, or 17%, which still supports Loonwatch’s claim that non-muslims are responsible for more attacks.
We can agree to disagree on whether Puerto Rico should be excluded from this list or not, but at least I’m telling you I’m doing it, so you can make up your own mind.
Where’s the Current Data?
The FBI Crime Data table cited was 1980-2005. This is data that ended early in Bush’s second term. Click here for what the FBI gives for data after 2005. It’s vague at best, and not in a nice table like the 1980-2005 report, making it difficult to compile any data from it. Maybe the FBI has this info hidden away somewhere convenient for some reason, maybe they’re just lazy. But nonetheless, the data used for the argument is 11 years old.
But scrubbing through this less-than-helpful timeline from the FBI, while there were several terror attacks thwarted two were successful which killed thirteen people and injured thirty more, all committed by Muslim extremists. Add in the recent Orlando attack that happened after the Loonwatch study, there are 49 more deaths on that list, and you realize for the last decade, the only terror threat to Americans, if we’re going by recorded incidents, has been from Muslims.
I’m atheist, and thus against all religion, because I think religious extremists of any faith are capable of doing heinous things. But in the modern era, I do not think anyone could reasonably argue that most ideological unprovoked violent acts in the modern era are not committed by people who claim to be doing those acts in the name of Allah.
But it is important to understand that just because they are responsible for such violence, it does not in any way mean that a majority or even a disproportionate amount of Muslims are violent. Arguing the converse is pure bigotry. But the evidence is clear that for every one American killed or injured in a terrorist attack by non-Muslims, there have been approximately 93-94 who were harmed or killed by Muslims. A narrative that is rather different from the one made by Loonwatch, yet also entirely true.
I have no animosity towards Muslims that I don’t equally have against all religion, my only issue is with misleading stats to push a particular narrative. Whether someone is killed by a religious extremist, or killed by a gang member robbing a store, the end result is identical. As with anything in life, I believe it is important to remain skeptical and question everything, because data can always be presented in a quite misleading manner to serve someone’s agenda. I hold myself to a higher standard, but you can’t possibly know that. And you can’t know it about any other op-ed write either.
2 thoughts on “Misleading Stats, Bad Sources, and the Threat of Radical Islamic Terror”
Just bumped into your site through some Google magic and took me to this page. Though I didn’t agree with your final conclusions, I really appreciated the time and effort you put into the data analysis. You demonstrated very clearly a point I often raise to others; data can often be manipulated to push a narrative. One quibble I had with your blog was you give a pass to the original article,. In my opinion, to not even acknowledge the number of deaths is such a glaring misrepresentation I would have to conclude the author is either 1) an idiot 2) purposely biasing the conclusion to further some position. Anywho, keep up the great work!
Thanks for the kind words, Bill. Glad you liked the article.