When Lebanese journalist Dima Sadek tweeted footage of chants against the country’s powerful armed movement Hezbollah, she quickly became the target of online abuse.
Sadek was reporting from the scene of protests that have swept Lebanon since mid-October, with demonstrators furious at a government they see as corrupt and incompetent.
But Sadek, who worked for the Lebanese Broadcasting Corporation, quickly became the story herself on social media that evening. “A few minutes later – only a few minutes – I was trending,” she says.
A hashtag attacking her eventually became the most-tweeted that night. “Dima the lowest,” it read. . The sudden organised onslaught was familiar to the journalist, who has since resigned, citing the harassment she received as a factor.
As anti-corruption, anti-austerity protesters across the Middle East rise up against their governments, they are increasingly confronted by a new threat. A powerful storm of disinformation obscures online activity surrounding the protests, and comes on top of internet shutdowns, hacking, censorship and threats of violence.
Online disinformation is intended to trick social media users into seeing and sharing material from fake accounts or influencers at home and abroad. Its aim is to steer discussion on the ground, focusing users’ ire on to targets that fit the agenda of those responsible.
Sadek, an outspoken critic of Hezbollah and its political allies, says she has long been a lightning rod for abuse on social media. “Whatever I write, I receive thousands of insults and threats. I get phone calls, my mum receives phone calls, so do my family.”
She accepts that some of the opprobrium she attracts is from genuine users. “But there is also lots of spam,” she says. “It’s an organised machine, using fake accounts.”
The speedy backlash that Sadek experienced bore the hallmarks of attacks using a combination of political “influencers” on social media and fake accounts that work to stoke organic engagement online.
Researchers tracking disinformation campaigns on Twitter during a brief spate of anti-government protests in Egypt in September identified groups of influencers rallying support for the Egyptian government and praising the country’s military and security forces, often opposed by users in Turkey and Qatar sympathetic to the protests.
“There was one group called Friends of the Egyptian People who would post and say, ‘Now make this hashtag trend,’” said Joey Shea, who studies online activity in Egypt. Shea was referring to a group of accounts sympathetic to Egypt and its regional backers in the Gulf. “It was clear this was a coordinated effort,” she said.
Researchers studying social media activity at the Canadian digital rights organisation Citizenlab found that recent protests in Iraq and Lebanon were accompanied by frenzied campaigns on Twitter in which the views of political social media influencers, often from the Gulf, played a central role.
They examined interactions on Twitter through retweets and found that the main hashtags demonstrators were using to share information about the protests were also prime targets for manipulation. “It seems like each group is trying to manipulate and use hashtags to share their political vision,” said Alexei Abrahams, a research fellow at Citizenlab.
Citizenlab established how some individuals on social media could have a large influence on the online conversation about protests in Iraq, Lebanon or Iran despite being located elsewhere. One example is Amjad Taha, a British-Arab commentator living in London and a fierce critic of the Iranian regime. Observers say his tweets often reflect the perspective of the Saudi Arabian government.
“He is the most retweeted among anyone on the hashtag [#LebanonRiseUp] … the defining hashtag of the Lebanese protests,” said Abrahams. At his peak, Taha’s posts accounted for 6.25% of the retweets on the hashtag; other commentators rarely managed even half that level of influence.
Taha said he was happy about his level of influence. “I would call myself a British-Arab influencer. That’s about it,” he said. He denied he was seeking to spread a pro-Saudi point of view, but added that he agreed with Crown Prince Mohammed Bin Salman’s Saudi Vision 2030 reforms and recent changes allowing women to drive.
Abrahams added that an examination of Iraqi protest hashtags also showed that users outside Iraq, in particular from the Gulf, had an even greater impact owing to repeated cuts to the Iraqi internet and limited use of Twitter among local commentators. “You’re seeing Gulf influencers among the top most retweeted accounts in all of those hashtags,” he said.
Online disinformation campaigns about the Middle East often hew to the regional alliances formed after the 2017 blockade of Qatar, whose support for Islamists across the Middle East drew the ire of a group consisting of Saudi Arabia, the UAE, Bahrain and Egypt.
Researcher Marc Owen-Jones of Hamad bin Khalifa University, who tracks malicious bot activity, has studied the rapid spike in the use of bots and the “weaponisation of social media” since the 2017 blockade. Bots are automated software that can execute repeated tasks in bulk, for good or ill, and can be used to fake engagement with a particular hashtag in order to attract real users to the topic.
“There is almost certainly an influence campaign on Twitter to generate support for an Iraqi revolution. Almost 20% of all accounts tweeting one specific hashtag are likely fake,” he said in an analysis posted online, referring to the hashtag “show your support for the right of the Iraqi people to protest peacefully”.
Researchers such as Owen-Jones emphasise how easy it is for those controlling the bots to mask their location, making it hard to track their true origins even though their targets are clear. “I don’t know who is behind [the influence campaign in Iraq], but I’d caution against talking heads emphasising that this is a leaderless uprising and citing social media as a source of inspiration. I don’t doubt that per se, its just important to be cautious about the veracity of a lot of the content out there,” he said.
Twitter recently banned political advertising, and the platform routinely purges bot and fake accounts, although organic content remains permitted. Facebook has been criticised for its refusal to ban fake political advertising on the platform.
Facebook has also drawn criticism from activists in the Middle East for its inability to police the disinformation that proliferates on its network. In Algeria, where protesters favour Facebook over Twitter to spread news of the demonstrations against the military-backed regime, activists began to notice what they term “electronic flies” shortly after protests began in February.
“Electronic flies are online profiles and pages, whether manipulated by real people or bots, that diffuse propaganda on behalf of the regime. They are often portrayed as the e-component of the counter-revolution,” said Raouf Farrah, an Algerian activist and commentator based in Canada. The problem is so widespread that Facebook groups now exist solely to track the phenomenon.
Facebook has also become a venue for undermining democracy in Tunisia, where it was once held up as a key tool used by pro-democracy protesters who unseated authoritarian leader Zine El Abidine Ben Ali during the 2011 Arab Spring. A report released in September by Democracy Reporting International (DRI), a German research group, found that the run-up to Tunisia’s presidential and parliamentary elections this year coincided with a surge in political content posted by Facebook pages without official links to any party or candidate.
In May, Facebook announced that it was removing 265 Facebook and Instagram accounts, Facebook groups and events “involved in coordinated inauthentic behaviour”. According to Facebook, this malicious activity originated in Israel and affected countries such as Nigeria and Angola, as well as Tunisia. The closure of one of the Tunisian pages demonstrates a hall-of-mirrors quality to the problem of disinformation. The page was called “stop the disinformation and lies in Tunisia”.