Talking manifestos
Plus: The 'both sides trap,' good and bad news for YouTube, a link (?) between depression and political violence, and more
I joined the Tech Against Terrorism podcast to discuss the evolution of extremist manifestos and the impact of online platforms on their prominence.
Specifically, we discussed the lower threshold for dissemination. The Unabomber had to bargain with major media outlets, promising to cease his campaign of violence if they would publish his 35,000 word manifesto, which they did with the assent of the Justice Department, which hoped someone would recognize his language. (It worked.) Today’s extremists can find an audience online much more easily, but the police and the media are starting to learn to deal with these documents more responsibly.
All that and more, on the Tech Against Terrorism podcast.
See also: The Atlantic: The Dangerous Spread of Extremist Manifestos
Next week: Lawful Extremism launches
We’re less than a week out from the release of my new paper, Lawful Extremism: Extremist ideology and the Dred Scott Decision. Next week’s newsletter will ship early, including a brief discussion of the paper and, of course, a link to where it appears on the Center on Terrorism, Extremism, and Counterterrorism website. I’m very excited to put this out into the world and see how people respond.
In the meantime, here’s a brief look at some other recent work of note from around the world on extremism and adjacent issues.
Latest research
Media platforming and the normalisation of extreme right views
This might fall under “things we already suspected to be true” but it’s still an important piece of work. When mainstream media gives right-wing extremists airtime and without assertively challenging their views, viewers are more inclined to agree with those views and to believe that those views are widely held. Platforming such figures in a critical, challenging way still moved viewers to the right, but less so. In short, and in my words rather than the words of the authors, the decision to consistently platform far-right figures and extremists in the media as part of a “both sides” paradigm of journalistic objectivity is a disaster for democracy.
The Institute for Strategic Dialogue tracked the rise in anti-Semitic content online after the Hamas attack on Israel, including on YouTube. The authors attempted to disaggregate anti-Semitic comments from criticism of Israeli policies.
So that was the bad news for YouTube. This is the good news: A second recent study suggests that YouTube algorithms don’t contribute meaningfully to radicalization.
This details quietist Salafi movements in Algeria and Libya and attributes some differences between quietist and revolutionary movements to political opportunity structures.
Who could be behind QAnon? Authorship attribution with supervised machine-learning
Scholars used machine learning to analyze Q drops, with findings that seem to confirm two authors cited in previous reporting by NBC.
You may recall some research I cited a few weeks ago that was interesting but whose significance was hugely overblown by the authors to the detriment of its findings. This paper features some very interesting findings, but maybe not as world-changing as the authors argue. A team of researchers from Harvard, Northwestern and the Hebrew University of Jerusalem have a paper out with a pretty bold claim that when depression, conspiracy beliefs and participatory inclinations are combined, they strongly correlate with (and perhaps cause) people to declare more support for political violence—moreso than people with just conspiracy beliefs and participatory inclinations but no depression. The authors argue that addressing depression could lead to a reduction in political violence—too forcefully, I think, based on the current state of research into this kind of correlation, and also considering the how the media is likely to misread this based on simplistic way most people talk and think about mental health. Also, last I checked, treating depression was far from a slam dunk, so at best, taking this approach means replacing one problem set with another, equally complicated problem set.
Tune in next week!
