Resistible Force Meets Movable Object

I have several data-based pieces coming over the next several months to address these questions, so I'm not going to do it all in this space, but two charts from my IS monitoring lists provide some preliminary insight into the effect of the terminations.

UPDATE: The most immediate effect can be seen in the composition of my IS monitoring lists. When I set up my monitoring list, for various reasons, I break it up into three equal lists based on the number of followers each account has. Prior to the Twitter suspension campaign, those three equal parts broke down to accounts with less than 250 followers, accounts with less than 800 followers, and accounts with more than 800 followers. As of Oct. 3, to equally distribute all my accounts, I had to set the threshholds at less than 150, less than 500 and greater than 500. This shows that the IS user base is working with much smaller numbers than before. And keep in mind that it's easier to find accounts with many followers, so this breakdown tends to be top-heavy. Here's a chart showing the change:




END UPDATE

The size of the lists change over time, partly due to new users but also due to my ongoing discovery of accounts, so these are based on per-user averages. The first shows how Twitter's most recent suspension of 400 accounts affects the "internal" network in samples taken on Sept. 28 and Oct. 1. 

                         

This includes both internal and external audiences. It also doesn't account for retweeting bots, a new crop of which was recently launched by IS. Subtract the impact of the hundreds of bots (which I don't have time to do right now) and the drop in October would be even greater. This is relevant as it pertains to human behavior, although the bots are currently a legitimate part of the ecosystem. Among other things, the bots drove links and retweets to the IS propaganda film "Flames of War."

There's also content quality. Anyone who follows any reasonable number of IS accounts has no doubt seen that rebuilding their networks now consumes a disproportionate amount of time. In other words, the quality of the interactions has been dramatically impacted, with thousands of tweets devoted to announcing and promoting newly reconstituted accounts and debunking fake ones that pop up while the originals are gone.

Time and energy spent recreating the network is, at this point, a significant portion of what IS does online (a minimum of 8 percent of tweets from September 29 to October 1, and likely higher), drawing focus and resources away from the business of sending ordinary Westerners pictures of severed heads when they're trying to get sports scores or live-tweet Cake Boss.

The suspensions have also caused IS users to think before they tweet; the heads and threats to execute hostages are in decline, though by no means absent, and when they start to pop up again, accounts go down. All of this is on Twitter, of course, it doesn't speak to "availability" of IS content. It speaks to dissemination and reach on one of the best platforms for driving traffic, as well as highlighting the fact that these online communities can be incentivized to change their behavior.

Many suspended accounts return, of course, but they have to rebuild every time, and the data suggests there is good reason to think they will lose ground over the long haul. We'll soon have enough data to talk about more definitively in the IS context. The early data is very encouraging and I will publish more of it when the book comes out.

Fundamentally, however this comes down to the inexplicable argument I've heard time and again, that in a world where we do practically anything to fight terrorists practically anywhere regardless of the costs or collateral damage, and with too little regard for whether what we do works, that kicking a very small number of terrorists off of Twitter for a small- to medium-gain is somehow a bridge too far, and ultimately useless because it doesn't instantly and magically end extremism.

I continue to be unmoved.

More to come, down the road a bit...

Opinions expressed herein are those of J.M. Berger. Buy J.M. Berger's book, Jihad Joe: Americans Who Go to War in the Name of Islam and pre-order his new book with Jessica Stern, ISIS: The State of Terror. 

RECENT PUBLICATIONS