Mandela and Trump

I know, I know this much is true

Did you see the incredible Donald Trump quote that’s been uncovered from 1998?

If I were to run, I’d run as a Republican. They’re the dumbest group of voters in the country. They believe anything on Fox News. I could lie and they’d still eat it up. I bet my numbers would be terrific.

It turns out he’s been lying all this time! What a fraud, you’ve got to share this with your friends!

Except of course, this is nonsense. He never said that. This was a fake quote that did the rounds prior to the 2016 election that was debunked at the time by such liberal sources as CNN, Newsweek and Snopes among many others. And yet, here it is, doing the rounds again.

But the interesting thing for me isn’t the quote, or the fact that you can get the best part of a million views and 7k retweets by sharing it. It’s some of the replies. Many people rightly point out that it’s fake, and share links to the debunking. But others swear blind that it’s true regardless, and they know that because they saw it on TV at the time. The stories vary a little bit, whether it was Wogan or Aspel or someone else interviewing Trump, but these people are sure they remember it being broadcast.

This isn’t a new phenomenon. There’s something called the Mandela Effect which refers to this kind of false recall, named after the fact that many people have apparently wrongly remembered that Nelson Mandela died in prison. There’s a whole Reddit where people post these kinds of stories.

I think you can split these responses into two categories. For some people, it’s a cute thing: they thought that a film existed that didn’t, they chuckle about it and move on. We might call this the Soft Mandela Effect. It’s kind of fun.

But then a smaller group of people are affected by the Hard Mandela Effect. They insist that their memory is true. Mandela did die in prison, they know that. So, the historical record must have been systematically altered: the clips of Mandela at the Rugby World Cup final in 1995 must have been faked! But who could have done that? And if they are faking that, what else isn’t true?

It’s not hard to see that this is an incredibly corrosive way to start thinking. When talking about incels, we might call it a blackpill moment. But I believe it would be a mistake to start thinking about this only in terms of People Like Them. As the fake Trump quote shows, sensible people on the liberal left are not immune to being taken in by this kind of misinformation.

But this is where the problem really starts. In 2016, Snopes and co could debunk the Trump interview by going through magazine back issues. But now, in 2024, things get trickier. While people have misremembered Trump giving this quote to Wogan, we could easily imagine a video of him doing it.

With the explosion of generative AI, there are free tools on the Internet which would allow anyone to create a video of Trump and Wogan talking together, and Trump giving this quote. It wouldn’t be great. The hairpieces might be unrealistic.¹ There might be the wrong number of fingers.

But a serious player could very easily mock up this video. After all, Spielberg and co could put convincing dinosaurs on the big screen in 1993. So it’s clear that, in this age of GPUs and generative algorithms, someone with access to Hollywood-style resources could create this clip today in such a way that we couldn’t meaningfully tell it from reality.

All of a sudden, the Hard Mandela people have got something to cling to. Wogan’s dead, he can’t contradict the video. It turns into a he-said she-said problem. But I think it’s not exactly the problem that people think it is. We heard a lot of talk in advance that the 2024 US election would be the deepfake election, but actually that dog failed to bark in practice. But I don’t think it’s that fake videos didn’t exist, I think it’s that the general public are actually better at judging truth than people give them credit for.

And this is where I think people, including nice sensible liberal People Like Us, need to be careful. None of us are immune to the Soft Mandela effect. But if we share things in that spirit and are caught out doing so, then it’s incredibly corrosive to trust in what we might say in the future. In an information environment where anything could be true, people will make up their mind based on general impressions of trustworthiness.

For example, Professor Trish Greenhalgh of Independent SAGE said on Radio 5 in January 2023 that

If you go back to when we introduced mask mandates first. I think it was back in June or July 2020. Cases were going up very, very rapidly, and then the mask mandate actually brought them under control, oh, within weeks, within a couple of weeks.

But that simply wasn’t true. This was a perfect example of the Soft Mandela Effect. When confronted by the data by FullFact, Professor Greenhalgh “agreed that she had misremembered what happened in 2020”. Frankly, it wasn’t a great look. But if an Oxford University Professor, the author of a classic book on Evidence Based Medicine, can do this then what hope is there for the rest of us?

I think all we can do is try to do rudimentary sense checks, go back to the data, and if we are caught out sharing a false quote or a false statistic, then we need to make sure we retract it fully and completely. In an ideal world our arguments would be grounded in data, though as Jesse Singal points out, data alone do not overcome vibes:

But voters don’t vote on the basis of what some nerd with a chart says; they vote on their vibes about how certain things are going. Which means that if your goal is to beat Trump, you need to understand their vibes and come up with an argument that meets the voters in question where they’re at — not one that sounds like something from a message board for macroeconomics PhDs. 

I think that’s certainly true. Even as a “nerd with a chart” myself, I don’t believe that elections are like Powerpoint presentations, and if only you can assemble the perfect collection of charts then the median voter will flock to your side. But I certainly think that if your presentation can be debunked, if your argument rests on wrong numbers or fake quotes, then that can rebound badly when you are caught out, and even flip the people you are trying to convince into Hard Mandela territory.

So, without wanting to add to the “what went wrong in 2024” pieces, I think that even in an age of deepfakes, we need to all hold ourselves to old-fashioned standards of truth and accuracy if we want to convince others that what we say is true.

1. Joke