Earlier this month, the Government Accounting Office released a report entitled Countering Violent Extremism”. Its a great example of how to outright lie using data. (I note that the, um, “analysis” was performed between October 2015 and April 2017, and bears the previous administration’s imprint. The current administration’s inanities lie in an orthogonal direction.)
The upshot of the report is:
GAO recommends that DHS and DOJ direct the CVE Task Force to (1) develop a cohesive strategy with measurable outcomes and (2) establish a process to assess the overall progress of CVE efforts. DHS and DOJ concurred with both recommendations and DHS described the CVE Task Force’s planned actions for implementation.
All well and good, once you get through the acronymese. But you cannot tackle something you don’t understand, or pretend not to understand. (Sun Tzu’s dictum about the need to know the enemy and know oneself comes to mind.) And this is how the authors of the piece understand violent extremists:
Of the 85 violent extremist incidents that resulted in death since September 12, 2001, far right wing violent extremist groups were responsible for 62 (73 percent) while radical Islamist violent extremists were responsible for 23 (27 percent). The total number of fatalities is about the same for far right wing violent extremists and radical Islamist violent extremists over the approximately 15-year period (106 and 119, respectively). However, 41 percent of the deaths attributable to radical Islamist violent extremists occurred in a single event—an attack at an Orlando, Florida night club in 2016 (see fig. 2). Details on the locations and dates of the attacks can be found in appendix II.
The usual narrative we’ve seen in the past decade and change in this country is that right wing extremists are more dangerous than Islamic extremists. But bad as the right wing crazies are, the narrative is getting a bit hard to sustain, what with the internet being so easily accessible. So the rear-guard action now seems to be to say that radical extremists at least aren’t any worst than the people we actually are allowed to think of as villains, and maybe better if you ignore that Mateen fellow.
But going to appendix II, where the data, such as it is, sits, is eye opening. Appendix II is basically a collection of sordid acts, described in short blurbs. Some are well-known, such as the aforementioned Mateen case: “Orlando Night Club shooting. Omar Mateen killed 49.” Some are oddly described. The John Allen Muhammad – Lee Boyd Malvo sniper attacks are broken up into 15 separate incidents, each with one dead victim. (This seems shy of the 17 deaths attributed to them in other sources, but that’s a quibble.)
But most of the data consists of onesies and twosies, of which there are many. Examples include: “Far rightist murdered a homeless man” which occurred on 1/19/2009 in Woodstock, IL. Or “Justin Nojan Sullivan, a convert to Islam, murdered his neighbor.” However, there are an awful lot of instances that are not quickly and easily found online and/or are hard to judge exactly how much ideology was involved. Examples include “White supremacist murdered his stepfather to gain “street cred”” and “Neo-Nazi killed sex-offender priest.” (More info on the latter here.)
But those also are quibbles. A bigger issue is that some obvious cases are not included. For example, the data set doesn’t include a single Islamic extremist killing in 2012. And yet, I remember this happening in 2012:
Gelareh Bagherzadeh and Coty Beavers were both shot to death in 2012 for their relationship with Nesreen Irsan, a young Muslim woman who left her home and faith to marry Coty, a Christian. Nesreen’s father, Ali Irsan, was upset with her decision and decided to punish her by killing her husband and the woman he blamed for encouraging Nesreen to convert to Christianity.
As of now the murder trial still doesn’t seem to have begun, but Ali Irsan and his family members seem to keep getting convicted of other offenses. Perhaps if and when Irsan does get convicted they’ll issue an amendment.
But it turns out that the omission of extremist Islamist violence is systematic, and in my opinion, deliberate. Here’s an entry that illustrates my problem with the data: “Prison gang white supremacists murdered another inmate for not objecting to having an African American cellmate.” A google search turns up more here.
Now, what this crime comes down to is the perpetrators’ anger at the fact that the victim was associating with someone outside their group. With a different cast of characters, the same crime committed for the same reason would be called an honor killing. Which is to say, something bearing a close resemblance to an honor killing falls into the range of activities that qualifies it for the list of extremist activities in the GAO’s paper, at least when it is performed by a violent far right extremist. But then it is reasonable to expect that something very much like an honor killing – say, an actual honor killing – should be included on the list if it is performed by an Islamic extremist as well. And that doesn’t seem to be the case.
As luck would have it, the Justice Department commissioned a study entitled Report on Exploratory Study into Honor Violence Measurement Methods in 2015. The report tells us:
The four types of honor violence discussed frequently are forced marriage, honor-based domestic violence, honor killing, and female genital mutilation.
The report goes on:
Using a method that applies U.S. demographic characteristics to another country’s honor violence statistics, researchers estimate that between 23 and 27 honor killings occur annually in the United States (Curtis et al., unpublished).
This extrapolation may not be very precise – the report cites another paper that looked at actual cases and concluded:
Murder by the family of origin is highest in Europe (72%) and lowest in North America (49%).
The report also indicates that while honor killings occur in a wide variety of religions, it states
In Europe, 96 percent of honor killings were committed by Muslims.
I think it would be more than fair to change the last word to “Islamic extremists.” It would also probably be fair to assume the 96 percent figure is more or less ballpark for the US.
Now, assuming the extrapolations given are squirrelly, let’s cut the numbers by a lot. Let’s assume that Islamic extremists commit 10 honor killings a year in the US, with the balance of the 23 to 27 going uncommitted or committed by another group, say, Amish fundamentalists or Quaker separatists. In that case, we would expect that over the 15 years in question, the total deaths to Islamic extremists in the US would more than double from 119 to (119 + 150) = 269. Additionally, the number of incidents would more than double from 23 to 173. In fact, you can leave out Mateen, Nidal Malik Hasan, the Beltway snipers, the Tsarnaev brothers, the San Bernardino killers, and every other case that is actually mentioned in the report, and the honor killings alone would swamp all the deaths inflicted by the right wing extremists.
From what I could tell, no attempts were made to include Islamic extremists killings that fell under the category of “honor killing.” I did a quick google search and found several murders in the US that fit the “honor killing” description. I didn’t find any in Appendix II of the GAO report. Its an easy enough exercise. Try it yourself. You can even compose a one sentence blurb to see how little effort it would have taken the GAO to do this right. I can only assume someone at the GAO really, really didn’t want it done the right way.
There are other problems with the GAO report, but this post is already long so I’ll close with what I really to understand. Does any of this excuse violent right wing extremists? Heck no. But what is with the attempts to make Islamic extremism look less dangerous?