I want to run something by you that has been on my mind since Saturday because it hits on some of the themes I have been writing about in my journalism and media posts for the past couple of months.
Suspension of disbelief and troll factories
If you recall, I wrote a post about two months ago on suspension of disbelief as critical to processing data. There was some research that suggested we are biased to believing what we hear is true — because “speakers usually do tell the truth.” This was interesting to me because the researchers also said that “an early bias toward believing the speaker is adaptive” – meaning we can change our bias toward believing to one toward disbelief, if the situation warrants it.
And I’ve been thinking about this in the context of propaganda, and how sophisticated propaganda works.
We know now from some insightful reporting on the issue that Russian troll factories aren’t only designed to push an agenda. They are designed to sow discord and create chaos. The troll factories are perhaps more interested in creating inflamed emotions than they are in promoting a specific agenda or point of view. The term troll is appropriate here too because the way it’s used today connotes ‘inflame passion without actually being passionate oneself’. The point of the troll is to push other people’s buttons, not to defend an idea one is passionate about per se.
And the way this works is through exploiting our bias toward truth. The troll factories use the cloak of anonymity in today’s virtual world to mask their true identities and take on a persona designed to engender trust in a target audience. They know that it’s easier in virtual communication – where your five senses are not at play – to fool people into believing they’re someone or something they’re not. The goal is to create a persona that seems trustworthy — and wait for trigger events to deploy it to sow discord. When those events happen, our ‘truth bias’ leads us to believe their comments, statements and posts are the thoughts and beliefs of a real person, when in fact they are not. They are manufactured for a persona and designed to manipulate.
The Serena Williams fiasco
I came figuratively face to face with this on Saturday during the controversy involving Serena Williams and a tennis chair umpire named Carlos Ramos. Shortly after the US Women’s Open Tennis Final between Williams and her opponent Naomi Osaka was over, I posted the following on Twitter:
Carlos Ramos should never oversee a finals match again. He let HIS emotions get the better of him. He should have shown restraint on the third code violation. He didn’t. And the result was chaos. That disqualifies him as a good chair umpire
— Edward Harrison (@edwardnh) September 8, 2018
Williams had received three code violations from Ramos, the match’s umpire. Importantly, two of those violations were entirely discretionary. The first discretionary violation was for ‘coaching’ because her coach had tried to tell her something during the match even though that’s not permitted under US Open rules. Williams, who claims not to have seen the signals her coach made, was incensed because she felt the umpire was branding her a “cheater”. And the fact that this rule is rarely followed or sanctioned probably added to her anger.
Later, when Williams lost a game in the match, she broke a racket in anger. That’s an automatic code violation. And given the previous code, she automatically lost a point. When she was informed that she had been docked a point, she became enraged and berated Ramos for the original coaching infraction, eventually calling him a “thief” for unfairly penalizing her the first time, and thus ‘stealing’ a point from her.
At that point, Ramos made a second discretionary call – at a critical moment in the match – to escalate the feud with Williams by calling a third code violation for verbal abuse. This discretionary call resulted in her forfeiting a whole game, putting her just one game from losing the match altogether.
I saw this refereeing as disgracefully ill-considered given the decisions were made at a critical juncture in the match. As a fan of the winner, Naomi Osaka, it ruined the match for me and – judging from the boos – pretty much everyone watching the match live in the stadium. The scene was ugly. And critically, the opinions were divided.
This was a perfect opportunity for trolling.
I noticed the trolls
So when I posted on Twitter, I got a lot of reactions. Now, I always block people who are uncivil, saying stuff like ‘idiot’, ‘stupid’, or using foul language to belittle someone rather than sticking to a reasoned argument. You have to block people who are antagonistic because they hijack the thread, changing its tone and making the discussion heated and personal for anyone who reads it afterward. Even if it’s not my thread, I block them preventatively. And so as I always do, I checked the profiles of some of the reactions before I blocked them, especially when the posts were excessive. And I found bots, trolls.
They were profiles of American people with normal profile descriptions. But when you looked at “tweets & replies”, you found a hugely disproportionate number of replies, especially to Twitter accounts with a high number of followers. If you are a troll and you want to inflame, you’re going to be more interested in responding than in tweeting. And when you do, you’ll want to make sure you do it in a thread that lots of people will see in as divisive a way as possible, in order to sow discord. These accounts also tended to end in a series of weird numbers.
What I realized after the Serena Williams controversy is that a lot of people probably don’t even know these people are trolls. And what happens is that the trolls hijack the thread by posting inflammatory language that causes real people to respond, outraged that someone could say something like that. In the minds of the respondent, unaware they are responding to a troll, they are countering a wrongheaded person from ‘the other side’, someone who has been crafted to fit the stereotype of a specific type of American. And so the respondent is left believing, ‘the other side’ is craven and deplorable, which is exactly what the troll wants to happen. Afterwards, more people come by the thread, see the back and forth and add to the vitriol. And before you know it, you have an all-out verbal war — all because a troll planted a verbal bomb at the right place and the right time.
That’s what I witnessed on my thread and on multiple other threads during this tennis controversy. And it makes sense to think the Russians were doing it, because it’s a great low cost form of warfare. They are essentially turning us against one another by amplifying divisions and sowing discord wherever they can.
The Internet is their friend in making all this happen.
Propaganda?
That’s how propaganda works, folks. It’s not just about spreading disinformation or controlling media outlets and making them toe the line. It’s also about helping your adversary defeat itself from within.
Anyway, I thought I would share that anecdote with you because it was a revelation for me to see so many trolls and bots trying to hijack the thread of my twitter post on a sports topic, not a political one. Blocking those people was effective in removing the vitriol and reinforced for me that blocking people works.
But on the larger issue of trolling, social media and journalism, the event is alarming. It shows how easily we can be manipulated. I’m talking about you and me, by the way, because I also initially thought these were real people until I dug deeper. The Internet has a lot of positive qualities to it. You’re reading this post only because the Internet exists. I have a following only because the Internet exists. But there’s a very dark side to the Internet as well. And it’s not clear to me which side will win.