Icon: enlarge
The FBI asks for help - and gets it from Twitter users
Photo:
FBI HANDOUT / EPA-EFE / Shutterstock
“Our darkest hour,” wrote John Scott-Railton on Twitter when he saw the first pictures of the storm on the Capitol on January 6th.
"At least one of the darkest I can remember." Scott-Railton is a US citizen, but he works at the University of Toronto in Canada as a researcher at the Citizen Lab.
He has achieved a certain level of notoriety in specialist circles by uncovering, together with his colleagues, numerous attempts by governments to monitor journalists and dissidents.
Since January 6th, he's been known for something else: his attempts to identify the Washington rioters using OSINT methods and report them to the police.
The abbreviation from the world of secret services stands for Open Source Intelligence, the extraction of information from freely available sources.
Scott-Railton analyzes image sections that appear on Twitter.
He looks for distinctive tattoos or patches, lettering and other details on the perpetrators' clothing in order to recognize them in other photos or videos.
But he's also looking for social media accounts where the faces from the Capitol can be recognized.
He buys a high-resolution photo from a news agency for $ 435 so that he doesn't just have to rely on blurred social media material.
In at least one case, he says he uses facial recognition software.
Volunteers quickly jumped to his side, tens of thousands, as he himself says, including grandmothers and students, anonymous activists, journalists and other OSINT experts.
"Is there such a thing as ethical doxxing?"
The first Scott-Railton identified is Eric Munchel of Nashville, Tennessee.
The "cable tie type" as the researcher initially only calls it.
Munchel had entered the Capitol masked and had cable ties with him, suitable to tie people up.
“As someone who grew up in the USA and had civics, like everyone else, this hooded man with the cable ties contradicted my idea of how democracy works so much,” Scott-Railton told SPIEGEL, “that I wanted to find out what his plan was.
And for that I had to know who he is. "
He has now reported several people to the FBI.
140,000 people now follow him on Twitter, 130,000 more than in December.
But his work is controversial.
Vivian Schiller, for example, a director at the Aspen Institute think tank, asks on Twitter: "Is there such a thing as ethical doxxing?"
The term describes the publication of personal data without the consent of the data subject.
Harvard researcher Joan Donovan even calls Scott-Railton's detective work "one of the most dangerous uses of social media by a researcher."
Schiller and Donovan touch a sore point in the online hunt: if there are lots of amateurs in front of a large audience, it can quickly lead to mix-ups.
Innocents could be suspected and condemned.
This is what happened after the 2013 Boston Marathon bombing, and now again.
The confirmation error known from cognitive psychology tends to accelerate confusion, which can sometimes be dangerous for those affected: Once there is a suspect, people tend to look for information that supports the suspicion instead of information that refutes it.
"Why do journalists and researchers help the FBI?" Asks the political magazine "New Republic".
The US Federal Police have extensive capabilities to investigate themselves.
"Even if the suspects are clearly villains in this case, it seems strange that people who are known for their civil rights work are now volunteering for the secret police," says the article, although the FBI is not a secret police .
And further: "Experts like Scott-Railton should monitor the authorities rather than relieve them of the work."
Donovan sees it similarly.
In an interview with "Protocol" she says: "These are surveillance techniques, and when the public uses them and thereby turns masses of people into cops, it seems to me a very dangerous impulse".
Scott-Railton is aware of this.
"This tracing in the net is a double-edged sword," he emphasizes.
He worries that someone keeps writing something on Twitter like "He kinda looks like", followed by a name that may or may not be true.
"That is why I am very careful and do not give names until they are officially confirmed." He repeatedly asks those who want to help him to exercise the same restraint.
"All in all, I think it's more of a positive thing."
He receives support for his position from, among others, Aric Toler from Bellingcat, one of the best known and most capable institutions for OSINT research.
Toler is one of the most renowned experts who take part in the identification attempts.
“Crowdsourcing is obviously powerful and can also go in the wrong direction, but all in all I think it's more of a positive thing,” he says in an interview with “Protocol,” especially since a large part of the work consists of “huge amounts of Search photos and videos «.
For a few days now, Scott-Railton and Bellingcat have been collecting information from thousands of recreational detectives together using an online form.
The rationale for this is not that there is too great a risk of dragging innocents into the limelight or of making yourself the target of vengeful right-wingers, as Joan Donovan warns, among others.
But the sheer mass of clues that are the only way to keep track of things.
According to the AP news agency, at least 90 people have so far been arrested who were alleged to have been involved in the storm on the Capitol.
The FBI lists many of them on a website.
Here you can also find the complaint against the "cable tie type" Eric Munchel, with photos and information that Scott-Railton used on Twitter.
With the help of police databases and surveillance videos from a hotel in Washington, the FBI has also identified another person who had entered the congress building at Munchel's side: his mother.
Icon: The mirror