03/10/2022
By Katharine Webster

How many law enforcement officers, military veterans and active duty service members are affiliated with far-right groups that commit acts of domestic terrorism? Are people who post violent thoughts online likely to act on them?

Three professors in the School of Criminology and Justice Studies have won three grants, together and separately, to study these questions.

All three are collaborating on a two-year, $278,425 grant from the U.S. Department of Homeland Security to research the “insider threat” posed by law enforcement officers who join militant far-right, white supremacist and anti-government groups.

That grant is led by Assoc. Prof. Christopher Harris, an expert in policing who will also look at whether law enforcement agency policies, practices and culture deter officers from joining extremist groups.

“We want to see if there are commonalities, psychologically or regionally, among the officers who belong to these groups, and what kinds of agencies the officers tend to belong to,” he says.

Harris is joined on the grant by Prof. Arie Perliger, director of the graduate program in Security Studies, who maintains the largest known database of violent far-right incidents in the United States, and Assoc. Prof. Neil Shortland, a forensic psychologist who directs the university’s Center for Terrorism and Security Studies.

“It’s the ideal team,” Harris says. “Far-right extremism is Arie’s area of expertise, Neil has expertise in psychology and police work, and police are in my wheelhouse.”

Perliger has obtained another grant from the Department of Defense for about $200,000 to do a similar insider threat study on military veterans and active-duty service members who are affiliated with the same kinds of extremist groups.

And under a third grant from the National Institute of Justice, Shortland will be researching whether there is a way to predict from someone’s extremist or hateful posts online whether they will go on to commit violence.

Assessing Insider Threats

Image by Katharine Webster

Harris says that the insider threats team will first mine Perliger’s database for information about law enforcement officers who have taken part in right-wing violence from 1990 to the present day. They will also add to the data by scouring news reports, court records and other sources.

Those cases will be analyzed for the officers’ individual psychological traits and circumstances, as well as by the type, size, location and policies of their law enforcement agencies.

At the same time, the team will survey 2,000 law enforcement agencies about their policies and practices involving membership in extremist groups. The survey will go to all 50 state police agencies; every police and sheriff’s department serving more than 100,000 people; and a representative sample of smaller law enforcement agencies across the country.

“We’re going to look at whether agencies have policies against belonging to these groups,” Harris says. “Do they have policies about what can be posted on social media? Have they uncovered any insider threats?”

The researchers will then do in-depth surveys of individual officers in three agencies at high risk of having members join extremist groups as well as officers at three low-risk agencies, to drill down further into the agencies’ cultures.

They hope to continue working with DHS to integrate their findings into training materials while updating best practices for agencies to adopt, Harris says.

“I’m really big on translating research into practice,” he says.

New Focus on Domestic Terrorism

Image by Courtesy

Perliger says the Department of Defense grant will use similar methods to look at affiliation with extremist groups by active-duty and veteran service members.

Both grants are the result of government agencies increasing their focus on insider threats and domestic terrorism after the storming of the U.S. Capitol on Jan. 6, 2021. More than two dozen members of law enforcement and at least as many military veterans were involved, according to arrest and court records.

Between the two grants, Perliger says he expects to find hundreds of police and current and former service members who are affiliated with extremist groups.

“January 6 was one of the triggers that pushed the Department of Homeland Security to do research and look at the presence of law enforcement officers in that crowd,” Perliger says. “They want more accurate, empirical data on this threat and how we can respond to it.”

On the military grant, Perliger will be working with Mia Bloom, professor of communication and Middle East studies at Georgia State University, and UML Ph.D. student Hope LaFreniere.

They will survey various military departments and agencies and profile service members and veterans who have joined extremist groups and acted violently, as well as those who have joined but not engaged in violence.

Do Words Matter?

Image by Katharine Webster

Shortland is co-principal investigator on the $1.13 million National Institute of Justice grant to investigate whether extremist and hateful postings on social media can be analyzed to predict whether the writer will go on to commit an act of violence.

Shortland and his partners will analyze hate-filled screeds and political manifestos posted on Twitter, Facebook, blogs and web forums by known terrorists and mass shooters, as well as by people who posted similar extremist or violent narratives but did not commit violent acts.

Often, people who do end up committing acts of mass violence and terrorism were previously “known to law enforcement” but were judged not to pose a major threat, Shortland says.

“We want to see if we can develop a predictive tool, based just on words, to know who actually poses the greatest risk in the real world,” Shortland says. “In other words, do words matter?”

Preliminary results are promising, he says. The grant builds on previous research conducted under a university seed grant by Shortland, Psychology Prof. Alyssa McCabe and Computer Science Assoc. Prof. Anna Rumshisky.

In that research, students hand-coded 185,000 words of online postings by 30 known mass shooters and lone-actor terrorists, as well as about 50,000 words of similar postings by people who did not engage in violence. They found some patterns, but they need more data.

Shortland is working on the current grant with Michael Sofis, senior scientist at Advocates for Human Potential Inc., and James Pennebaker, chairman of the Psychology Department at the University of Texas at Austin. The team will use a linguistic analysis tool developed by Pennebaker.