Should Twitter banish Islamic State supporters or track them? A new study investigates

Should Twitter banish Islamic State supporters or track them? A new study investigates

A new study of online support for Islamic State terrorists contains surprising implications for stopping recruitment, as well as for using social media as an intelligence tool.

The paper, published by the Brookings Institution on Thursday, tracks tens of thousands of Twitter accounts over the course of months, achieving the most detailed snapshot of Islamic State activity online to date. During the period covered by the study, Twitter began a new policy of suspending some jihadi accounts, allowing the researchers to study the consequences of extremist censorship in real time.

The lead researchers were J.M. Berger, an open-source terrorism analyst affiliated with Brookings, and Jonathon Morgan, a data scientist. Their findings about these suspensions’ effect on terrorist activity challenges a number of preconceptions about extremism and the internet.

Some terror analysts advocate that companies like Twitter exercise a hands-off approach to jihadi content, claiming that if deprived of a public platform, extremists would simply switch to using other tools, which may be more difficult to track. Others point to the futility of suspending accounts, saying that terrorists would simply start over with fresh usernames in an endless game of “Whac-A-Mole.”

But according to Berger and Morgan, cracking down on key Islamic State supporters has a palpable effect on Twitter activity. Since Twitter’s crackdown on jihadi content in September 2014, they write, “more than 8 percent of online activity by ISIS supporters is now being dedicated to rebuilding the network,” devoting less time to recruitment, propaganda and other activities. And overall, supporters eventually slowed in creating new accounts to replace their suspended ones, showing that they “were wearying of the battle with Twitter.”

In other words, even minor pruning of terrorists’ access to online platforms can hamper the network’s effectiveness. It can also isolate the “base” which the Islamic State hopes to reach: “As suspensions contract the network, members increasingly talk to each other rather than to outsiders.”

Others object to censorship of extremist material on First Amendment grounds, claiming that a sloppy law to remove terrorists from the internet could also be used to restricting other forms of objectionable speech. Berger and Morgan acknowledge that nuanced lawmaking is “something [government] has not traditionally excelled at,” and that social media companies — which already have substantive policies of their own regarding inappropriate content — ought to take the lead in combating terrorist material.

The authors’ research includes substantial data on the sub-network structure of online supporters. They suggest platforms such as Twitter can use their technique to locate key nodes and break the network into smaller, less powerful “clusters.”

This way, the platform can remain a valuable intelligence analysis tool for government and private anti-terrorism efforts, while curtailing the group’s influence.

This report, by Ivan Plis, was cross-posted by arrangement with the Daily Caller News Foundation.

LU Staff

LU Staff

Promoting and defending liberty, as defined by the nation’s founders, requires both facts and philosophical thought, transcending all elements of our culture, from partisan politics to social issues, the workings of government, and entertainment and off-duty interests. Liberty Unyielding is committed to bringing together voices that will fuel the flame of liberty, with a dialogue that is lively and informative.


Commenting Policy

We have no tolerance for comments containing violence, racism, vulgarity, profanity, all caps, or discourteous behavior. Thank you for partnering with us to maintain a courteous and useful public environment where we can engage in reasonable discourse.

You may use HTML in your comments. Feel free to review the full list of allowed HTML here.