On a clear February day in 2010, Andrew Joseph Stack III pushed his single-engine airplane to its top speed and flew it into an IRS office in Austin, Texas, killing himself and a manager of the federal tax collecting agency.
Stack’s suicide mission was clearly an act of extremist terror, but which kind? Was it right-wing because he attacked a government office? Or was it left-wing because the writing he left behind favorably compared communism to capitalism and railed against the bailout of financial institutions in 2008?
Those seemingly arcane and morbid questions have taken on new urgency since President Trump blamed forces “on many sides” for the violence in Charlottesville, Va. Various news outlets have sought to answer those questions by referring to statistics about the exact number of fatalities caused by “right-wing,” “left-wing” or “Islamic” extremists.
But a closer look reveals that those precise figures involve art as well as science, often requiring judgment calls that lead organizations to come up with very different answers to the same question.
“All of this is inherently fairly subjective at some level,” said Alex Nowrasteh, a policy analyst with the libertarian Cato Institute, “because it depends on what people say, or what they sometimes claim as motivation after the fact.”
The New York Times, for example, ran two stories Tuesday that used very different studies to measure the same thing: extremist violence in the United States. One story cited a study by the Anti-Defamation League that counted at least 372 murders committed by “domestic extremists” between 2007 and 2016.
The second Times story referred to Nowrasteh’s analysis for Cato that counted almost two-thirds fewer deaths: 134 murders during that same 10-year period.
The numbers differ because the studies count different things. Cato focused only on acts of terror as most people commonly understand them – individuals or groups who murder strangers for political or ideological reasons. This includes the horrific nightclub shooting in Orlando that claimed 49 victims, and the Christmastime workplace shooting in San Bernardino that claimed another 14 victims.
The ADL arrived at a much larger number by counting all murders committed by people it considers extremists. There are crossovers. Like Cato, Mark Pitcavage, a senior research fellow with the ADL, said his total includes Dylann Roof’s murder of nine African-Americans in a Charleston, S.C., church in 2015. But it also counts a group of white supremacists who killed three of their own: a member they suspected of being a police informant, the alleged informant’s girlfriend, and the wife of the group’s leader in order to collect insurance money.
That’s three murders committed by “right-wing extremists,” but were they really acts of right-wing extremism as most readers understand the term? “A skinhead killing a skinhead doesn’t make them less dangerous,” Pitcavage said. “A murder is a murder.”
Similarly, The Center for Investigative Reporting’s interactive map illustrating “homegrown terror” includes a white supremacist couple who murdered four people near the California/Oregon border. Yet the victims in the incident were one of the convicted killers’ stepfather and mother, as well as two innocent strangers who gave the killers a ride; in other words, crimes committed by people with reprehensible politics but not specifically political crimes.
The center also treads into gray areas. For instance, it counts a black activist’s killing of three Baton Rouge cops last July, and wounding of three others, as an act of right-wing extremism. The logic there appears to be similar to that applied to the left-wing Austin suicide pilot: He was classified a right-wing extremist because he targeted the government.
Despite these differences, most of the extremist/terror trackers agree on the broader patterns. Radical Islamic terror is by far the leading cause of deaths by extremists, followed by the much smaller body counts blamed on right-wing and then left-wing groups.
Still, when one starts the clock running can make all the difference. For example, a tally of radical Islamic terrorism starting on Sept. 10, 2001 would provide a very different picture from one starting on Sept. 12, 2001.
The Cato Institute reports that right-wing groups have been responsible for 10 times as many fatalities as left-wing groups since 1992. But that large discrepancy is due in part to one grisly act that occurred 22 years ago: the 1995 Oklahoma City bombing, which killed 168 people, accounting for 77 percent of the right-wing total.
Another point relevant to Trump’s claim about violence on many sides: Of the 23 murders Cato attributes to left-wing actors since 1992, 13 have occurred since the start of 2016. During the same period, the right has been responsible for five deaths, including Heather Heyer, the woman fatally struck by a driver in Charlottesville. And that’s killings alone, meaning those figures don’t reflect what occurred just two months ago, when a left-wing sniper opened fire on Republican lawmakers playing softball, seriously wounding Rep. Steve Scalise, R-La., in another ugly incident that prompted many to bemoan violence at both ends of the political spectrum.
While good data can help allocate law enforcement resources, trying to say one group is worse or bloodier than another is ultimately a mug’s game, according to those who spend time compiling and classifying these crimes. What’s more important, they insisted, is acknowledging trouble can brew on the right and the left, along with the threat from radical Islamists.
“This is not some horse race,” Pitcavage said. “The left wants to point to the right and the right wants to point to the left, but the reality is we have to deal with all of those. We must be willing to recognize the dangers that could be coming from our own ideological positions. Instead, too many people want to amplify any sort of danger from their opponents without owning up to what’s coming from their end.”
