Skip to main contentSkip to navigationSkip to navigation
Information trap … Illustration: Sergiy Maidukov

The disinformation age: a revolution in propaganda

This article is more than 4 years old
Information trap … Illustration: Sergiy Maidukov

Troll farms, bots, dark ads, fake news ... from Putin’s Russia to Brexit Britain, new methods are being used to change politics and crush dissent. It’s time to fight back

Father came out of the sea and was arrested on the beach: two men in suits standing over his clothes as he returned from his swim. They ordered him to get dressed quickly, pull his trousers over his wet trunks. On the drive the trunks were still wet, shrinking, turning cold, leaving a damp patch on his trousers and the back seat. He had to keep them on during the interrogation. There he was, trying to keep up a dignified facade, but all the time the dank trunks made him squirm. It struck him they had done it on purpose, these mid-ranking KGB men: masters of the small-time humiliation, the micro-mind game.

It was 1976, in Odessa, Soviet Ukraine, and my father, Igor, a writer and poet, had been detained for “distributing copies of harmful literature to friends and acquaintances”: books censored for telling the truth about the Soviet Gulag (Solzhenitsyn) or for being written by exiles (Nabokov). He was threatened with seven years’ prison and five in exile. One after another, his friends were called in to confess whether he had ever spoken “anti-Soviet fabrications of a defamatory nature, such as that creative people cannot realise their potential in the USSR”.

Forty years have passed since my father was pursued by the KGB for exercising a citizen’s simple right to read, to listen to what they chose and to say what they wanted. Today, the world he hoped for, in which censorship would end – as the Berlin Wall would fall – can seem much closer: we live in what academics call an era of “information abundance”. But the assumptions that underlay the struggles for rights and freedoms in the 20th century – between citizens armed with truth and information and regimes with their censors and secret police – have been turned upside down. We now have more information than ever before, but it hasn’t only brought the benefits we expected.

More information was supposed to mean more freedom to stand up to the powerful, but has also given the powerful new ways to crush and silence dissent. More information was supposed to mean a more informed debate, but we seem less capable of deliberation than ever. More information was supposed to mean mutual understanding across borders, but it has also made possible new and more subtle forms of subversion. We live in a world in which the means of manipulation have gone forth and multiplied, a world of dark ads, psy-ops, hacks, bots, soft facts, deep fakes, fake news, Putin, trolls, and Trump.

Forty years after my father’s interrogation, I find myself following the palest of imprints of his journey, though with none of his courage, or certainty, and none of the risk. I run a programme at a London university that researches the newer breeds of malign influence campaign across the world – and tries to find ways to combat them. But the language, ideals, tactics and stories that sustained the struggle for democracy in the 20th century are now used by the very forces they were meant to fight.

Consider the Philippines. As my parents were enjoying the pleasures of the KGB in the 1970s, the Philippines were ruled by Colonel Ferdinand Marcos, a US-backed military dictator, who used the army to impose censorship and indulge in spectacular forms of torture, leaving victims’ skulls stuffed with their underpants by the side of the road, so as better to intimidate passersby. Marcos’s regime fell in 1986, when millions came out on the streets demanding an end to the censorship and torture.

Today Manila greets you with sudden gusts of rotting fish and popcorn smells, wafts of sewage and cooking oil. Soon you start noticing the selfies. Everyone is at it: the sweaty guy in greasy flip-flops riding the metal canister of a bus; the Chinese girls waiting for their cocktails in the malls. The Philippines has the highest use of selfies in the world, the world’s highest use of social media per capita, the highest use of text messages: 20th-century style censorship would be near-impossible to impose here. But the new president, Rodrigo Duterte, is rehabilitating Marcos’s reputation; he has also found new ways of exerting oppression.

Glenda Gloria remembers the Marcos years. In the 1980s she was a student journalist covering the regime’s torture of opposition figures. Her boyfriend had been arrested for running a small independent printing press and had had electrodes connected to his testicles.

“The psychological warfare that Marcos mastered is very similar to what is happening now,” Gloria told me. “The difference is, Duterte doesn’t have to use the military to attack the media … How is it made possible? With technology.”

Protesters in Quezon City, the Philippines, last year. Photograph: Jes Aznar/Getty Images

Gloria is managing editor at Rappler, the Philippines’ first online news agency, designed not merely to report on current affairs, but to crowdfund for important causes, and gather vital information to help victims of floods and storms. Experienced journalists like Gloria and editor-in-chief Maria Ressa hired 20-year-olds who knew about social media. When you walk into Rappler’s orange, open plan office you notice how young and largely female the staff are, with a small band of older journalists overseeing them with a hint of matronly severity. In Manila they are known as “the Rapplers”.

When Duterte decided to stand in the presidential election of 2015 he and Rappler seemed made for each other. A mayor from a provincial town with a reputation for being tough on drug offences, Duterte got relatively little TV time and so focused on social media. When Rappler hosted a Facebook presidential debate, he was the only candidate to turn up. It was an overwhelming success. His message – to vanquish drug crime – was catching on. Rappler reporters found themselves repeating his soundbites about the “war on drugs”. When Duterte later went on his killing spree, they would regret using the term “war”. It helped to normalise his actions: if this was a “war”, then casualties became more acceptable.

His famous “earthy” humour, the trademark politically incorrect jokes that Duterte indulges in much like Trump, Putin, Boris Johnson, Jair Bolonsaro and other showmen “populists”, also took on a darker dimension. When jokes are used by the weak to poke fun at the powerful, they can bring authority figures back down to earth. But when such language is used consistently by men of real power to degrade those who are weaker, this humour grows into something menacing: it lays the linguistic path to humiliating victims in other ways as well, to a space where norms disappear.

No one knows exactly how many have been killed in Duterte’s “war on drugs” since he became president. Human rights organisations estimate 12,000 – the government claims 4,200. At one point an average of 33 people were being killed a day, by police and vigilante gangs riding around on motorcycles. There is little due process to ascertain if the murdered were guilty. There are frequent reports of drugs being planted on the corpses that filled up the alleys of Manila’s slums.

When Rappler began to report on Duterte’s killings the site’s carefully curated online community suddenly turned on it. At one point there were 90 messages an hour: claims that Rappler was making up the deaths, that it was in the pay of Duterte’s enemies, that it was all “fake news”. The messages were like an infestation of insects, swarming into email inboxes and descending like a scourge on to the site’s community pages. Rappler journalists were shouted at in the malls: “Hey, you – you’re fake news! Shame on you!” Hashtags calling for the arrest of Ressa began to trend. The government launched a court case against her. She walked around town with bail money on her. As soon as one case was thrown out another would appear. International human rights groups call them politically motivated.

After several months of this onslaught, the Rapplers dedicated themselves to making sense of the attacks. First to catch their eye were the Korean pop stars. They kept appearing in their online community, commenting on how great Duterte was. How likely was it that Korean pop stars would be interested in Filipino politics? When they checked out the comments the pop stars were making they matched one another word for word: obviously fake accounts, most likely controlled from the same source.

They ran a program that scoured the internet to see who else was using the same language. They found other accounts repeating the same phrases. These looked more realistic, claiming to be real Filipinos with real jobs. The Rapplers began researching each one individually, calling their purported places of employment. No one had heard of them. Altogether they found 24 well disguised but fake accounts repeating the same messages at the same time and reaching an audience of 3 million. This was a coordinated attack. But proving who was behind it was near impossible.

Gloria remembers how in Marcos’s time you could see the enemy. There was a sort of predictability: they could kill you, or you could skip town, contact a lawyer, write to a human rights group, take up arms. You knew who the agents were, who was coming for you, who your enemy was. But now? You couldn’t tell who you were up against. They were anonymous, everywhere and nowhere. How could you fight an online mob? You couldn’t even tell how many of them were real. And of course this allowed the government to claim they had nothing to do with these campaigns. Wasn’t it just a question of concerned citizens exercising their right to free speech?

Democratic Representatives Ayanna Pressley, Ilhan Omar, Alexandria Ocasio-Cortez and Rashida Tlaib speak about Donald Trump’s Twitter attacks against them in Washington DC. Photograph: Jim Lo Scalzo/EPA Photograph: Jim Lo Scalzo/EPA

It is a tactic repeated across the world. Campaigns can be incited by pro-government online rabble rousers as in Turkey, or run through “troll farms” owned by tycoons loyal to the Kremlin in Russia (when one such Russian operation had its “news” pages taken down by Facebook, it sued the company for abuse of its freedom of expression). Nor are democracies immune. As a report by the thinktank the Institute for the Future elucidates, when President Trump identifies critics on Twitter, a cascade of abuse then descends on them. This habit was already apparent in the 2016 presidential campaign, but has now reached new intensity in the Trump supporters’ demands to “send home” opposition congresswomen of colour.

What we are seeing is, in the words of Columbia law professor Tim Wu, a situation where “speech itself is seen as a censorial weapon”. This questions the old idea that the answer to “fallacious speech” is “more speech” – that we live in a “marketplace of ideas” where the best “information products” will win out. What if that market can be rigged?

Alberto Escorcia is a social media wizard and has helped coordinate some of the largest anti-corruption protests in Mexico over the last half decade. He and his friends started with pranks to provoke the police: after students were beaten by police officers, they went on silent marches and staged lie-ins, where they stretched out supine on the street, blocking the road. In time, he realised that if he knew in advance which subjects brought people together, and which words strengthened the interconnections between people, he would be able to “summon up” and strengthen protests. He had long believed that, at its best, the internet can connect society with its deepest needs for social change. He was inspired by how Google managed to predict and nip in the bud a flu epidemic when the company saw how many people were looking up flu symptoms at the same time in one place. Something similar, Alberto argued, could be done with political issues. You can tell what people really care about from their searches and online conversations.

But by the time I met Escorcia in Mexico City he looked too tired even to be frightened any more. Someone had been ringing his doorbell then running away again so he couldn’t sleep at night, shining acid-green lasers into his bedroom, pinging online death threats with his name spelt out in bullets – thousands every day so that his phone vibrated with alerts 24/7, turning it into an instrument of psychological torture. One takes such threats seriously in Mexico. During my visit I was told the story of a social media activist who had run an anonymous Twitter account cataloguing crimes committed by narcos. When the narcos found out who she was they first shot her and then posted her blasted off face on her own Twitter feed: “TODAY MY LIFE HAS COME TO AN END. DON’T MAKE THE SAME MISTAKE AS I DID … I FOUND DEATH IN EXCHANGE FOR NOTHING.”

It wasn’t just the personal threats that Escorcia was worried about – he feared troll farms were doing something more fundamentally damaging. The government was intervening in the relationship between people and their own desire for social change, spamming the internet with messages from fake accounts impersonating support for the government, using part-automated, part-human “cyborg” accounts to distract protesters from organising, sending in social media sock-puppets who pretend to support protesters, and then encourage violence to discredit movements. For 70 years, during the 20th century, Mexico had been a one-party state in which “truth” had been dictated top down. Today bots, trolls and cyborgs could create the simulation of a climate of opinion, which was more insidious, more all-enveloping than the old broadcast media – as it wormed its way into the feeds on your phone and you couldn’t tell whether it was coming from a friend or propagandist. This simulation, Escorcia worried, would then become reinforced as people modified their behaviour to fall in line with what they thought was reality. “Dark days are coming, Peter,” Alberto told me, “a new generation of bots and trolls are pushing us further and further into a world of pure simulation.”

In their analysis of the role of bots, researchers at the University of Oxford called this process “manufacturing consensus”. Similar techniques are being used across the world. Researchers at Harvard have shown how the Chinese government posts 448m social media comments a year, the aim of which is not to engage but to distract, as critical topics are replaced with positive ones.

It’s not just the belief that “more information” leads to better democracy that has been undermined by the digital era. Something more insidious is going on, in relation to our very idea about what a free person is.

Illustration: Sergiy Maidukov

Back in 1977, as my father was being interrogated by the KGB, the conversation would often turn to literary matters. The KGB agents would chide him for his love of modernist authors who, instead of writing in the impersonal, officious language of Soviet socialist realism captured their individual impressions of life. “Is there any joy greater than writing in the first person?” my father had written, in a novella published that year. “I feel any good book which acknowledges the human being, individuality, uniqueness is also anti-Soviet because the state dictatorship is directed against the human being as an individual.”

Today social media puts absolutely no constraints on our limits of self-expression, each of us can be a modernist author on our Facebook feed. But this self-expression is then transmuted into data: the language we use, our likes and shares, are all passed to data brokers and then on to advertisers and spin doctors who target us with specially tailored campaigns we might not even be aware of. The more we express ourselves, the less power we have.

This has produced a propaganda model that is very different from the 20th century. Instead of stuffing an ideology down people’s throats via TV and radio, a spin doctor has to tailor different messages to different social media groups. A country of 20 million, the chatty digital director of Vote Leave, Thomas Borwick, told me, needs 70 to 80 types of targeted message. Borwick’s job is to connect individual causes to his campaign, even if that connection might feel somewhat tenuous at first.

In the case of the vote to leave the EU, Borwick, who seems to approach such challenges like a Rubik’s Cube, claimed that the most successful message in getting people out to vote had been about animal rights. Vote Leave argued that the EU was cruel to animals because, for example, it supported farmers in Spain who raise bulls for bullfighting. And within the “animal rights” segment Borwick could focus even tighter, sending graphic ads featuring mutilated animals to one type of voter and more gentle ads with pictures of cuddly sheep to others.

I’d heard of similarly varied messaging used by spin doctors across the world. The challenge with this sort of micro-targeting is that it requires some big, empty identity to unite all these different groups, something so broad these voters can project themselves on to it – a category such as “the people” or “the many”. The “populism” that is thus created is not a sign of “the people” coming together in a great groundswell of unity, but is actually a consequence of the people being more fractured than ever, of their barely existing as one nation. When people have less in common than before, you have to create a new version of “the people” for every election. As too many concrete policies and coherent ideologies would risk alienating parts, these pop-up people need to be united around a leader’s personality and a vague feeling, such as “take back control” or “optimism”. Facts are a hindrance rather than a help: you are not trying to win a rational debate with floating voters; you want to say whatever gets more attention in fragmented social media groups, where the more outrageous you are the more likes you’ll get. Indeed there is something of a rush in throwing a middle finger up to facts, farting at glum reality. Trump and Johnson are both products of this environment.

But if one sells a campaign in so many different ways to different people, then how does one know what it is meant to represent? Was Brexit about animal rights? Immigration? What did the “people’s will” mean? In the words of the interim report of the UK parliamentary committee on fake news, the new information games “reduce the common ground on which reasoned debate, based on objective facts, can take place … the very fabric of our democracy is threatened.”

Boris Johnson speaks at the launch of the Vote Leave bus campaign in 2016. Photograph: Darren Staples/Reuters

Earlier this year I sat at the back of a room in the Commons as members of the committee where I helped as one of many specialist advisers wrote up its final report. Behind the MPs was a painting that showed a scene from the House of Commons in the 18th century, depicting men in wigs making some sort of elegant argument, while the other side listened attentively. It was unlikely to have been a realistic scene – the Commons has always been a rowdy place full of cads and liars – but it did, at least, envision an ideal, what I imagined the committee meant by “reasoned debate” and “fabric of democracy”. Does it have a chance in an age of infinite digital disinformation?

One reaction from governments, including democratic ones such as Germany, France and maybe soon the UK, has been to try to impose censorship on “disinformation”. It’s an understandable reaction – but the wrong one. It imposes a political logic where censorship is normalised again, and so reverses the triumphs of 1989: no wonder Putin loves to invoke a German law on “fake news” to impose new censorship at home. Even more important, it misunderstands how the internet, as opposed to broadcasting, works.

Back in Soviet Ukraine information was scarce. Every morning, father rose at dawn, gently turned the Spidola radio to “ON”, pushed the dial to shortwave, wiggled and waved the antenna to dispel the fog of jamming from Soviet censors, and climbed on chairs and tables to get the best reception, steering the dial in an acoustic slalom between transmissions of East German pop and Soviet military bands. He pressed his ear tight to the speaker and, through the hiss and crackle, heard the words: “This is the BBC in London”; “This is Radio Liberty in Washington.” He was listening for news about current affairs, banned books, the arrest of other dissidents. Smuggling information about political arrests to the western radios was like being granted a second life; it meant your case would be known, you could be defended. Eventually, after many adventures, he would end up working at the BBC Russian service himself, broadcasting censored art and information into the USSR.

Today there is less censorship of content in many places. But there is a more subtle form of censorship about how the information environment around us is shaped. Instead of closing down rights to receive and impart information, we should demand more. For starters, we should have the right to know whether an account online is a bot or someone genuine, whether content is organic or amplified by trolls, who is behind a “news” site. Anonymity is a necessity for many activists, as I’d learned in Mexico, but we should still understand if an account is using a false identity. We need to know why computer programs show us one piece of content and not another; why any ad, article, message or image is being targeted specifically at me and not you; which of our own data has been used to try to influence us and why. Perhaps then we would become less like creatures acted on by mysterious powers we cannot see, made to fear and tremble for reasons we cannot fathom, and instead would be able to engage with the information forces around us as equals. This, in turn, would become the difference between democracy and dictatorship today, for these are just the sort of digital rights the Putins, Dutertes and Trumps need to deny people to secure their power.

Information was the front line in the battle of democracy and dictatorship in the 20th century, the space where victory was articulated and supposedly secured. Since then it has become the arena where democracy is undermined. But it can also be the place where big words such as “rights” and “freedoms”, words that have been appropriated by a new generation of manipulators and oppressive regimes, are given meaning again.

This Is Not Propaganda: Adventures in the War Against Reality by Peter Pomerantsev is published by Faber (£14.99). To order a copy go to guardianbookshop.com. Free UK p&p on all online orders over £15.

Most viewed

Most viewed