Latest / Blocked and Reported / Episode 233: I Heard You Like Fake News So I Put Some Fake News In Your News About Fake News (With Dan Williams)

Transcript

Episode 233: I Heard You Like Fake News So I Put Some Fake News In Your News About Fake News (With Dan Williams) · Blocked and Reported

0:07
SPEAKER_00: Hello, everyone.
0:08
SPEAKER_00: Welcome to Blocked and Reported.
0:09
SPEAKER_00: I'm Jesse Singel.
0:10
SPEAKER_00: Katie is off this week.
0:12
SPEAKER_00: She is on the run as the authorities try to track her down for her role in the
0:16
SPEAKER_00: cloud seeding conspiracy that led to Hurricane Milton.
0:19
SPEAKER_00: In Katie's place, I'm joined by Dan Williams.
0:21
SPEAKER_00: Dan is a lecturer in philosophy.
0:23
SPEAKER_00: That's an assistant professor,
0:25
SPEAKER_00: as we Americans call it,
0:26
SPEAKER_00: at the University of Sussex and an associate fellow at the Leverhulme Center for
0:30
SPEAKER_00: the Future of Intelligence at the University of Cambridge,
0:33
SPEAKER_00: which is also where he got his PhD in philosophy.
0:35
SPEAKER_00: He writes, quote, the main focus of my research is on the social functions and causes of beliefs.
0:40
SPEAKER_00: I am especially interested in self-deception,
0:42
SPEAKER_00: religious beliefs,
0:43
SPEAKER_00: political ideologies,
0:44
SPEAKER_00: and delusions,
0:45
SPEAKER_00: end quote.
0:46
SPEAKER_00: Dan also authors the truly excellent sub stack conspicuous cognition at conspicuous cognition.com.
0:52
SPEAKER_00: Dan, thank you for joining us.
0:53
SPEAKER_01: Thank you for the invitation.
0:54
SPEAKER_00: Just to give listeners a bit of a roadmap.
0:57
SPEAKER_00: The first part of our conversation will be a tiny bit wonky just so we can define some terms and so on.
1:02
SPEAKER_00: But then we're going to apply your work and your writing to some real world situations.
1:07
SPEAKER_00: including the rioting in England over the summer,
1:09
SPEAKER_00: this researcher who claims to have invented a vaccine for misinformation,
1:14
SPEAKER_00: and a recent article in The Atlantic by Charlie Warzel.
1:17
SPEAKER_00: So the reason I invited you on is I've grown increasingly concerned,
1:22
SPEAKER_00: not just about misinformation,
1:23
SPEAKER_00: but about the way we talk about misinformation.
1:26
SPEAKER_00: It seems like a lot of these supposed experts in this field are
1:30
SPEAKER_00: are making claims about the nature of misinformation,
1:32
SPEAKER_00: its effects on society,
1:34
SPEAKER_00: and how to fight it that aren't really steeped in solid evidence.
1:38
SPEAKER_00: I forget how I found your work,
1:39
SPEAKER_00: but I just find you to be one of the sharpest people around in terms of explaining
1:42
SPEAKER_00: this issue,
1:42
SPEAKER_00: so I'm glad to have you.
1:43
SPEAKER_00: I do think we should get a little bit of definitional stuff out of the way.
1:48
SPEAKER_00: Hopefully this won't consume the whole podcast, but...
1:51
SPEAKER_00: First of all,
1:51
SPEAKER_00: how would you define misinformation and disinformation,
1:55
SPEAKER_00: which are often used interchangeably,
1:56
SPEAKER_00: but they mean different things,
1:58
SPEAKER_00: right?
1:58
SPEAKER_01: Yeah.
2:00
SPEAKER_01: It would be nice if those were just sort of simple definitional questions.
2:04
SPEAKER_01: But unfortunately,
2:06
SPEAKER_01: the way that these terms get used within misinformation research is just mired in
2:12
SPEAKER_01: confusion and inconsistency and subjectivity.
2:16
SPEAKER_01: At the most abstract level,
2:18
SPEAKER_01: I think the way in which the term misinformation is typically used,
2:22
SPEAKER_01: not always,
2:22
SPEAKER_01: but typically,
2:24
SPEAKER_01: is to refer to false or misleading information.
2:27
SPEAKER_01: It's supposed to be silent on questions of intent.
2:30
SPEAKER_01: It says nothing about the motivations or the goals of the people spreading the information.
2:36
SPEAKER_01: It could be that it's just an innocent mistake.
2:38
SPEAKER_01: It could be that they're trolling.
2:40
SPEAKER_01: It could be that it's a sophisticated propaganda campaign.
2:43
SPEAKER_01: We don't know.
2:44
SPEAKER_01: All we know is that the information is false or misleading.
2:47
SPEAKER_01: That's contrasted with disinformation,
2:50
SPEAKER_01: which is used when we're supposed to be able to have a high degree of confidence
2:55
SPEAKER_01: that the person spreading the information knows that it's false,
2:59
SPEAKER_01: knows that it's misleading,
3:00
SPEAKER_01: and is spreading it with the goal of manipulating or deceiving audiences.
3:07
SPEAKER_01: That all sounds somewhat simple,
3:10
SPEAKER_01: but the problem is if you think about that umbrella category of misinformation,
3:15
SPEAKER_01: there's this term in it which is misleading.
3:19
SPEAKER_01: It's really difficult to say in the abstract what it really means for communication to be misleading.
3:26
SPEAKER_01: When much of the post-2016 panic about online misinformation really rose to prominence,
3:34
SPEAKER_01: Most of the focus was on a very specific kind of misinformation,
3:38
SPEAKER_01: which was fake news in the sort of technical sense of that term.
3:42
SPEAKER_01: So not in the sense in which Donald Trump uses that term for like anything that he
3:46
SPEAKER_01: disagrees with,
3:47
SPEAKER_01: but in a sense of disreputable news outlets online just making shit up.
3:53
SPEAKER_01: So if you take the story,
3:54
SPEAKER_01: you know,
3:55
SPEAKER_01: the Pope endorses Donald Trump for president,
3:57
SPEAKER_01: it's fake news in the literal sense that somebody just made that up and then
4:00
SPEAKER_01: published it.
4:02
SPEAKER_01: With that kind of content, it's pretty uncontroversial that it's misleading.
4:07
SPEAKER_01: So it's not going to be that contentious classifying that kind of content as misinformation.
4:12
SPEAKER_01: But the problem is,
4:14
SPEAKER_01: even though there was a big kind of panic about that sort of online content around 2016,
4:20
SPEAKER_00: So let's pause there because I want to ask you about that.
4:23
SPEAKER_00: You connect the rise of the present misinformation discourse to 2016 and Trump and Brexit.
4:30
SPEAKER_00: And I promise I won't derail this further because I want to hear what you have to say.
4:33
SPEAKER_00: But why is that the place to start in understanding this?
4:37
SPEAKER_01: Well, I should say, I mean, that's not just my take that this blew up in 2016.
4:41
SPEAKER_01: That's sort of really well documented.
4:44
SPEAKER_01: And I mean,
4:45
SPEAKER_01: you can sort of try to psychoanalyze people why online misinformation became such a
4:49
SPEAKER_01: big issue then.
4:51
SPEAKER_01: But clearly you had these two populist revolts.
4:55
SPEAKER_01: You first need Brexit and the United Kingdom, which took many people by surprise.
5:00
SPEAKER_01: And then you had Trump, and that also took many people by surprise.
5:04
SPEAKER_01: And these were movements that are explicitly kind of anti-establishment.
5:09
SPEAKER_01: And that means they're also kind of explicitly anti-establishment, knowledge-generating institutions.
5:17
SPEAKER_01: So like in 2016...
5:20
SPEAKER_01: run up to the Brexit result.
5:22
SPEAKER_01: You had senior conservative politicians who were sort of outlining a very skeptical view on expertise.
5:30
SPEAKER_01: And obviously,
5:31
SPEAKER_01: when it comes to the Trump MAGA movement,
5:33
SPEAKER_01: you've also got this incredible skepticism of these establishment institutions like
5:40
SPEAKER_01: mainstream media,
5:41
SPEAKER_01: and then later on,
5:42
SPEAKER_01: obviously,
5:43
SPEAKER_01: public health,
5:43
SPEAKER_01: and so on and so forth.
5:45
SPEAKER_01: So I think there was an attempt to ask, how did this happen?
5:48
SPEAKER_01: And one of the explanations that became very, very prominent
5:54
SPEAKER_01: had to do with online misinformation.
5:56
SPEAKER_01: So I think that's the sort of connection between the two.
5:59
SPEAKER_00: Okay.
5:59
SPEAKER_00: And well,
5:59
SPEAKER_00: let's return to the sort of backseat psychoanalyzing at the end,
6:03
SPEAKER_00: because I am curious about how that provides a particularly comforting explanation
6:08
SPEAKER_00: for political events that otherwise have confounded a lot of us.
6:11
SPEAKER_00: But okay, so misinformation is...
6:16
SPEAKER_00: doesn't care whether or not it was intentional disinformation.
6:18
SPEAKER_00: A classic example there is like Russian troll farms or whatever,
6:23
SPEAKER_00: intentionally trying to seed false information in the populace.
6:29
SPEAKER_01: Exactly.
6:31
SPEAKER_01: And so,
6:31
SPEAKER_01: as I mentioned,
6:32
SPEAKER_01: a lot of the kind of focus around 2016 and the years thereafter was on a very
6:37
SPEAKER_01: specific kind of content.
6:38
SPEAKER_01: And lots of research emerged around this content, which was
6:42
SPEAKER_01: in the technical sense of that term that I mentioned.
6:46
SPEAKER_01: And the problem is over the past few years,
6:49
SPEAKER_01: there's been quite a lot of research now suggesting that that kind of content,
6:53
SPEAKER_01: it does exist,
6:54
SPEAKER_01: obviously,
6:55
SPEAKER_01: but it's nowhere near as prevalent as many people claim it is.
7:00
SPEAKER_01: So most people don't seem to encounter much of that content online.
7:03
SPEAKER_01: It's massively skewed in the sense that most people don't encounter much of it,
7:08
SPEAKER_01: but there's a minority of extremely online,
7:11
SPEAKER_01: active social media users who encounter quite a lot.
7:15
SPEAKER_01: They've got pre-existing,
7:17
SPEAKER_01: often extreme,
7:18
SPEAKER_01: conspiratorial worldviews,
7:20
SPEAKER_01: and they tend to seek out that content.
7:22
SPEAKER_01: It's congruent with what they already believe.
7:25
SPEAKER_00: I found this striking,
7:27
SPEAKER_00: just this idea that not that many of us are exposed to it,
7:30
SPEAKER_00: because every time I go on Twitter,
7:32
SPEAKER_00: I am inundated with the most insane garbage you can imagine.
7:36
SPEAKER_00: But that's sort of the point.
7:38
SPEAKER_00: I'm the sort of person who goes on Twitter, and that's not a random sample of the population.
7:42
SPEAKER_01: That's part of it.
7:43
SPEAKER_01: But I would also say a lot of this research is on social media pre-Elon Musk taking over Twitter.
7:48
SPEAKER_01: So it could be the case now that there's a hell of a lot more of this content on
7:53
SPEAKER_01: Twitter than there was,
7:54
SPEAKER_01: say,
7:54
SPEAKER_01: in 2020.
7:54
SPEAKER_01: That's fair.
7:57
SPEAKER_01: And the point is still true,
7:58
SPEAKER_01: which is that most people in America or elsewhere are not really engaging with
8:03
SPEAKER_01: political content on Twitter.
8:04
SPEAKER_01: So it's not like a representative sample.
8:07
SPEAKER_01: But it could be that there's more of that really crazy,
8:10
SPEAKER_01: insane content on Twitter now than there used to be.
8:14
SPEAKER_01: But in any case,
8:14
SPEAKER_01: I would say there's something of a consensus now among researchers that if you're
8:18
SPEAKER_01: focusing on that really just clear-cut,
8:21
SPEAKER_01: crazy fabricated content,
8:23
SPEAKER_01: it's not as big a deal as many people thought it was around 2016,
8:27
2017.
8:27
SPEAKER_01: It does exist.
8:31
SPEAKER_01: I think sometimes it does have harmful consequences.
8:33
SPEAKER_01: So later on,
8:34
SPEAKER_01: we might talk about the UK riots,
8:37
SPEAKER_01: which might be a situation where it had bad consequences,
8:39
SPEAKER_01: but it doesn't seem like a driver of like big socio-political events and trends as
8:44
SPEAKER_01: many people thought it was.
8:46
SPEAKER_01: But the issue is there's now been this response by people who work on
8:50
SPEAKER_01: misinformation to really expand the meaning of the term,
8:53
SPEAKER_01: a kind of concept tree,
8:55
SPEAKER_01: where because there's this generic idea of misinformation being misleading,
9:01
SPEAKER_01: now there's this idea within this area of research that misinformation
9:04
SPEAKER_01: The term should be used for content from pundits,
9:09
SPEAKER_01: from commentators,
9:10
SPEAKER_01: from mainstream media outlets that is misleading,
9:14
SPEAKER_01: even if the content is true because it's sort of cherry-picked or it's selective or
9:19
SPEAKER_01: it naps appropriate context and so on.
9:21
SPEAKER_01: And you can see once you get into that kind of territory,
9:24
SPEAKER_01: there's going to be massive subjectivity and bias and disagreement when it comes to
9:29
SPEAKER_01: classifying content as misinformation.
9:32
SPEAKER_01: So we moved away,
9:32
SPEAKER_01: I think,
9:33
SPEAKER_01: from this case study where most people are going to agree it definitely is
9:37
SPEAKER_01: misleading to this much more subjective territory where I think it's much more contentious.
9:43
SPEAKER_00: Isn't part of the problem here that – I mean philosophers actually argue over what knowledge is.
9:50
SPEAKER_00: That's hotly debated or it was at one point.
9:53
SPEAKER_00: To apply that to the real world.
9:55
SPEAKER_00: So something like –
9:57
SPEAKER_00: do masks prevent the spread of COVID?
9:59
SPEAKER_00: Which is, at one point, was a hotly, like the policy debate, one of the policy debates.
10:06
SPEAKER_00: The only real answer to that,
10:07
SPEAKER_00: if you know how sort of science and evidence work,
10:09
SPEAKER_00: is that it would be very hard to know with a high degree of certainty whether or
10:14
SPEAKER_00: not masks prevent the spread of COVID.
10:16
SPEAKER_00: And yet, unfortunately, that was treated as...
10:20
SPEAKER_00: a conspiratorial thing to say and in my view the the actions of fact checkers
10:25
SPEAKER_00: during that error i said error freudian slipped era um a corroded trust in experts
10:33
SPEAKER_00: because i i don't know i guess there's a tension here where if you're a public
10:35
SPEAKER_00: health expert and you want people to wear masks and i i wore masks when people wore
10:38
SPEAKER_00: masks um
10:40
SPEAKER_00: you maybe want to shield some of that complexity from them.
10:43
SPEAKER_00: Because in the midst of a crisis,
10:45
SPEAKER_00: you're not going to get up and start droning on about randomized controlled trials
10:50
SPEAKER_00: and levels of evidence.
10:52
SPEAKER_00: But I guess part of your argument is that there has to be a little bit more honesty and humility here.
10:58
SPEAKER_00: Because if you start saying it's misinformation to claim something like
11:03
SPEAKER_00: masks don't prevent the spread of COVID,
11:05
SPEAKER_00: you could easily respond to that by saying,
11:07
SPEAKER_00: I mean,
11:07
SPEAKER_00: why isn't it misinformation to say they do prevent the spread of COVID?
11:10
SPEAKER_00: We don't know.
11:11
SPEAKER_00: There's a lot of we don't knows out there, right?
11:14
SPEAKER_01: Exactly.
11:15
SPEAKER_01: Even these relatively narrow scientific questions are just nightmarishly complicated,
11:20
SPEAKER_01: and there's a range of opinion at a given time.
11:22
SPEAKER_01: It's quite rare when you've got overwhelming expert consensus.
11:27
SPEAKER_01: In some cases, you do.
11:28
SPEAKER_01: I would say if someone's claiming vaccines give you autism or something like that,
11:33
SPEAKER_01: that seems like a case where we just know that that is not true,
11:37
SPEAKER_01: according to overwhelming expert consensus.
11:39
SPEAKER_00: I don't want to get into the territory of mine so open my brain falls out.
11:42
SPEAKER_00: I think something like the COVID vaccine,
11:44
SPEAKER_00: where it's a well understood physical process and high quality data is very
11:49
SPEAKER_00: different from a question about something like the masks,
11:51
SPEAKER_00: I would argue.
11:52
SPEAKER_01: I completely agree.
11:53
SPEAKER_01: But I'd also add that now there's this idea that there are all sorts of claims
11:58
SPEAKER_01: people could have made about the vaccines and did make about the vaccines,
12:01
SPEAKER_01: which I think is sort of uncontroversially misinformation.
12:05
SPEAKER_01: But now there's this big move to say,
12:06
SPEAKER_01: well,
12:07
SPEAKER_01: if there's a mainstream media outlet and it publishes a story about somebody
12:11
SPEAKER_01: getting vaccinated and then dying shortly afterwards...
12:14
SPEAKER_01: Even if every aspect of the story is true,
12:16
SPEAKER_01: and the story just reports that it's being investigated,
12:22
SPEAKER_01: there's this drive to classify that kind of content as misinformation.
12:27
SPEAKER_01: Because even if it's true, people might be misled.
12:30
SPEAKER_01: They might come to think that the dangers of these vaccines are more serious than they in fact are.
12:36
SPEAKER_01: And I would argue once you get into that kind of territory,
12:38
SPEAKER_01: it gets really murky and complicated very,
12:41
SPEAKER_01: very quickly.
12:42
SPEAKER_01: It's really difficult to know how you're going to apply a principle there consistently.
12:46
SPEAKER_00: I don't want to step on my own interview,
12:48
SPEAKER_00: but a lot of the claims you hear when,
12:49
SPEAKER_00: for example,
12:50
SPEAKER_00: the New York Times earlier this year,
12:52
SPEAKER_00: they published a big piece,
12:54
SPEAKER_00: as far as I know,
12:54
SPEAKER_00: responsibly reported on the thousands of people who believe COVID vaccines harm them.
12:59
SPEAKER_00: And the piece was open to the possibility that in some cases these people are right.
13:04
SPEAKER_00: I think a lot of the response to that has embedded in it certain theories about the
13:10
SPEAKER_00: effects of mass media on people's beliefs and behavior,
13:13
SPEAKER_00: such as...
13:15
SPEAKER_00: Well,
13:15
SPEAKER_00: basically,
13:16
SPEAKER_00: that mass media has a predictable effect on people's beliefs and behaviors.
13:19
SPEAKER_00: Like, there's... Oh, the Times has endorsed this idea.
13:22
SPEAKER_00: Therefore, a lot of people will now be infected with the concept of vaccine injury.
13:26
SPEAKER_00: That'll change their behavior in a way that will hurt human society.
13:30
SPEAKER_00: Your argument...
13:32
SPEAKER_00: I'm gathering from reading your work is that,
13:35
SPEAKER_00: I mean,
13:35
SPEAKER_00: first of all,
13:35
SPEAKER_00: there's just all sorts of philosophical reasons to not want to suppress that sort
13:39
SPEAKER_00: of news,
13:39
SPEAKER_00: even informally.
13:40
SPEAKER_00: But second of all,
13:41
SPEAKER_00: we don't know exactly the effect of one or 10 or 100 New York Times articles on the public,
13:47
SPEAKER_00: because this is a very complicated dynamic system.
13:51
SPEAKER_01: It's so complicated.
13:53
SPEAKER_01: We also don't know how people would react to a situation in which they know that
13:59
SPEAKER_01: mainstream media outlets or whatever else are not reporting on certain things on
14:04
SPEAKER_01: the basis that it might mislead people.
14:06
SPEAKER_01: You can see very easily how that might decrease people's trust in these sorts of institutions.
14:11
SPEAKER_01: You're exactly right.
14:12
SPEAKER_01: Many people have this very simplistic
14:15
SPEAKER_01: theory in their head about how it is that information leads to belief.
14:20
SPEAKER_01: It's if people,
14:22
SPEAKER_01: and it's always other people,
14:24
SPEAKER_01: the person who is advancing this theory never thinks about themselves,
14:27
SPEAKER_01: but if people are exposed to some kind of information,
14:32
SPEAKER_01: then that's very likely to lead them to belief.
14:36
SPEAKER_01: There's a linear relationship between information exposure and belief.
14:41
SPEAKER_01: I just think we now know a lot about how
14:44
SPEAKER_01: sophisticated the psychological processes are that underlie belief formation,
14:49
SPEAKER_01: and how complex issues of trust and distrust are,
14:54
SPEAKER_01: and how they're mediated by people's attitudes towards institutions and different
14:59
SPEAKER_01: kinds of sources and so on.
15:01
SPEAKER_01: That sort of simplistic theory just completely misrepresents the complexity that
15:05
SPEAKER_01: you find in those sorts of areas.
15:06
SPEAKER_00: Right.
15:06
SPEAKER_00: And we should separate out two things.
15:08
SPEAKER_00: It's one thing if the processes that answer the question of what we believe and why are
15:14
SPEAKER_00: are sophisticated,
15:15
SPEAKER_00: which surely they are because it's the human brain,
15:18
SPEAKER_00: that doesn't necessarily lead to a sophisticated outcome.
15:22
SPEAKER_00: And a lot of your work,
15:24
SPEAKER_00: sophisticated in the sort of zoomed out nerdy,
15:28
SPEAKER_00: I arrived at this conclusion for the quote unquote right reasons.
15:30
SPEAKER_00: Because a lot of your work is about social epistemology, which tell me if I have this right.
15:35
SPEAKER_00: This is the study of how people learn and convince themselves and convince others
15:40
SPEAKER_00: of things in social environments,
15:42
SPEAKER_00: more or less.
15:43
SPEAKER_01: Exactly.
15:44
SPEAKER_01: Yeah.
16:05
SPEAKER_00: I'm not just expressing my considered view on the subject.
16:08
SPEAKER_00: I'm also displaying something about my own beliefs,
16:12
SPEAKER_00: my own values,
16:13
SPEAKER_00: and this will help me to fit in with tribes that are important to me.
16:16
SPEAKER_01: Exactly.
16:17
SPEAKER_01: So I think human beings, this is sort of banal, but it's important.
16:22
SPEAKER_01: We are ultra social animals and we care massively about winning approval and trust
16:29
SPEAKER_01: and status from social networks that we belong to and identify with.
16:34
SPEAKER_01: This drives lots of weird things that human beings say, do, and think.
16:40
SPEAKER_01: I think in the domain of behavior, that's kind of obvious.
16:42
SPEAKER_01: Lots of human behavior is geared towards signaling that we've got certain kinds of characteristics.
16:48
SPEAKER_01: Why do people waste lots of money on pointless luxury goods is to signal their wealth.
16:55
SPEAKER_00: Why do people use complex- And that's a more convincing signal than just saying I
16:59
SPEAKER_00: have a lot of money because anyone could lie about having a lot of money.
17:02
SPEAKER_01: Exactly.
17:02
SPEAKER_01: It's differentially expensive.
17:04
SPEAKER_01: You can't afford a Rolex unless you are actually quite wealthy,
17:07
SPEAKER_01: so wearing a Rolex credibly signals that you are a wealthy person.
17:12
SPEAKER_01: Many, many examples like that.
17:13
SPEAKER_01: Why do some people all around the world spend lots of time engaging in costly religious rituals?
17:21
SPEAKER_01: Forcibly, part of that is to signal that they're a good, loyal member of their religious community.
17:26
SPEAKER_01: It's a credible signal because only if you really care about the religious
17:30
SPEAKER_01: community would you invest so much time and energy in engaging in these sorts of rituals.
17:35
SPEAKER_01: And as you say, I think that doesn't just shape people's behavior.
17:38
SPEAKER_01: I think it also very plausibly shapes what people believe and the kinds of
17:42
SPEAKER_01: narratives that they sign on to and endorse.
17:45
SPEAKER_01: So,
17:45
SPEAKER_01: you know,
17:46
SPEAKER_01: if I believe that the 2020 election in the US was fraudulent,
17:51
SPEAKER_01: that marks me out as a kind of good,
17:53
SPEAKER_01: loyal MAGA Republican.
17:55
SPEAKER_01: If I believe that the world is a white supremacist,
17:59
SPEAKER_01: patriarchal,
18:00
SPEAKER_01: heteronormative hellscape,
18:01
SPEAKER_01: that signals I'm a certain kind of ultra-progressive,
18:05
SPEAKER_01: and so on and so forth.
18:06
SPEAKER_01: You find that in the political domain and the religious domain.
18:10
SPEAKER_01: I think the important thing to say about it is it's not that it's insincere,
18:14
SPEAKER_01: the claim isn't that people don't really believe this,
18:16
SPEAKER_01: they're just saying it to advertise which group they belong to.
18:20
SPEAKER_01: It's rather the groups that we belong to and these sort of motivations to display
18:25
SPEAKER_01: our allegiances to them end up shifting what we actually believe and think about
18:30
SPEAKER_01: various sorts of topics.
18:31
SPEAKER_01: I don't think that's the only thing going on in these sorts of cases.
18:34
SPEAKER_01: So I think often people are just trying to get at the truth,
18:36
SPEAKER_01: but they're in these really skewed information environments where all of the
18:40
SPEAKER_01: evidence they encounter lends some sort of needs in one particular direction.
18:44
SPEAKER_01: But I do think it's a big...
18:46
SPEAKER_01: aspect of what's going on.
18:48
SPEAKER_01: And you're right,
18:48
SPEAKER_01: it connects to this sort of social epistemology because it's a clear case where
18:52
SPEAKER_01: people's attempt to get at what's true is influenced by this underlying social motivation.
18:59
SPEAKER_00: I both do and don't find it tricky to wrap my head around this.
19:02
SPEAKER_00: So there are these anthropological findings
19:06
SPEAKER_00: of um initiations into cults or fraternities or the military and people are like
19:12
SPEAKER_00: why would you join a cult that puts you through such an insane process uh would i
19:18
SPEAKER_00: pick whatever you want something very painful and and the idea is that is a signal
19:23
SPEAKER_00: of your allegiance to the cult it's like it's not it's you do it because you then
19:27
SPEAKER_00: feel part of the group basically right
19:30
SPEAKER_01: You feel part of the group,
19:31
SPEAKER_01: and also the group is more likely to accept you and approve of you and feel like
19:35
SPEAKER_01: you're a trustworthy and loyal member of that group.
19:37
SPEAKER_00: And it builds social cohesion, which is very evolutionarily important.
19:40
SPEAKER_00: Now, I get that.
19:43
SPEAKER_00: I would like to think that my own beliefs were formed for the right reasons.
19:46
SPEAKER_00: But if I actually reflect on it, I grew up in a very liberal town in Massachusetts.
19:51
SPEAKER_00: I think as soon as I was politically conscious, I understood in this...
19:56
SPEAKER_00: Hard to describe, but potent way what the right views were and what the wrong views were.
20:02
SPEAKER_00: I knew that I was supposed to be, whatever that means, in favor of what we now call reproductive rights.
20:07
SPEAKER_00: I was supposed to be in favor of gay marriage.
20:09
SPEAKER_00: It was one of the first subjects I wrote about.
20:12
SPEAKER_00: And in one of your podcast interviews,
20:13
SPEAKER_00: you said that people maybe don't appreciate how little insight we have into our own
20:18
SPEAKER_00: process of forming beliefs.
20:19
SPEAKER_00: It's easy to point a finger at others and say, well, they got taken in by the MAGA cult.
20:24
SPEAKER_00: They're an SJW.
20:25
SPEAKER_00: They had some failure of logic and reason.
20:28
SPEAKER_00: Your argument, as I take it, is that
20:32
SPEAKER_00: Other than self-flattery, there's no good reason to think that our own processes are more rigorous.
20:36
SPEAKER_01: Yeah,
20:37
SPEAKER_01: it's complicated because even though I think that's true at the level of individual psychology,
20:43
SPEAKER_01: you might think,
20:44
SPEAKER_01: okay,
20:44
SPEAKER_01: but you're embedded in a sort of different informational environment and embedded
20:49
SPEAKER_01: in other kinds of institutions which help to counteract some of those biases.
20:54
SPEAKER_01: That's why I think if you take an individual scientist,
20:57
SPEAKER_01: they're often very biased and dogmatic and overconfident,
21:00
SPEAKER_01: but they're embedded in an institution with norms and procedures which guard
21:04
SPEAKER_01: against those biases and counteract them to some significant degree.
21:09
SPEAKER_01: But yeah,
21:09
SPEAKER_01: I think the more general point is true,
21:11
SPEAKER_01: which is we all think of ourselves as objective,
21:14
SPEAKER_01: disinterested truth-seekers.
21:16
SPEAKER_01: The term that I really like from psychology is naive realism.
21:20
SPEAKER_01: Most of us think the truth, even about complex social, political issues, is self-evident.
21:27
SPEAKER_01: Our beliefs,
21:28
SPEAKER_01: our opinions,
21:28
SPEAKER_01: are just a dispassionate,
21:32
SPEAKER_01: disinterested,
21:33
SPEAKER_01: objective reflection of those self-evident truths.
21:37
SPEAKER_01: If you think the truth is self-evident, then insofar as people don't accept the
21:42
SPEAKER_01: the truth,
21:42
SPEAKER_01: if they disagree with you,
21:44
SPEAKER_01: there's this temptation to think either they're lying,
21:46
SPEAKER_01: they know what the truth is,
21:48
SPEAKER_01: and they're just bullshitting or being deceptive,
21:51
SPEAKER_01: or they're psychotic,
21:52
SPEAKER_01: they just lost their mind,
21:53
SPEAKER_01: or they're post-truth or something like that,
21:56
SPEAKER_01: or they've been brainwashed.
21:58
SPEAKER_01: We tend to have this very Manichean view where there are the
22:02
SPEAKER_01: good, objective, honest, rational people that happen to be on our side of the particular debate.
22:07
SPEAKER_01: And then there are the other people who are just sort of beyond reason or they're deceptive.
22:12
SPEAKER_01: And that itself, I think, is shaped by many of these underlying social motivations and biases.
22:18
SPEAKER_00: One of my favorite examples I've heard you cite,
22:20
SPEAKER_00: there was some survey where – tell me if I'm remembering this right.
22:24
SPEAKER_00: I think a significant percentage of political liberals in the US said they would
22:29
SPEAKER_00: rather have a meteor hit the earth than have Trump be elected.
22:33
SPEAKER_00: And that's one of those –
22:36
SPEAKER_00: examples that is rare in how certain we can be that people don't really think that
22:43
SPEAKER_00: and your argument would be even though this is a survey even though we think people
22:46
SPEAKER_00: are reporting the truth maybe some of them feel it in the moment this is this is
22:50
SPEAKER_00: signaling in its purest form you're saying
22:53
SPEAKER_00: I despise Trump.
22:55
SPEAKER_00: And maybe the analogy there is people who say they feel the election was stolen.
23:00
SPEAKER_00: It could be that with some of them.
23:02
SPEAKER_00: If you had a beer with them or a few beers and you really got into it,
23:05
SPEAKER_00: this is a way of them expressing a more general distrust of authority.
23:09
SPEAKER_00: Isn't it just that a lot of these statements for the individual –
23:14
SPEAKER_00: are sort of low stakes in terms of real world import.
23:17
SPEAKER_00: It's easy to send that signal out and to gain the benefits from it.
23:20
SPEAKER_00: Because at the end of the day,
23:22
SPEAKER_00: you don't actually have the power to have a meteor is not going to hit Earth.
23:26
SPEAKER_00: There's not much you can do about whether or not the election was stolen.
23:28
SPEAKER_00: So it's just a convenient thing to say.
23:31
SPEAKER_01: Yeah, I think that's probably right.
23:33
SPEAKER_01: The incentives here don't reward people for being incredibly intellectually rigorous.
23:39
SPEAKER_01: So if you put a lot of effort into trying to get things right and you're really
23:42
SPEAKER_01: dispassionate and thoughtful and so on,
23:46
SPEAKER_01: even if you get things right,
23:47
SPEAKER_01: the payoff for you as an individual is not going to be that much.
23:50
SPEAKER_01: But if you form beliefs that deviate from those that are prevalent within your
23:55
SPEAKER_01: political or cultural tribe or whatever,
23:57
SPEAKER_01: that can be really,
23:58
SPEAKER_01: really materially and socially costly.
24:02
SPEAKER_01: I do think another thing that's going on is,
24:04
SPEAKER_01: in many of these contexts,
24:06
SPEAKER_01: one of the motivations is this kind of signaling motivation.
24:10
SPEAKER_01: We want to be looked upon favorably by members of our political or cultural faction or tribe.
24:16
SPEAKER_01: There's this other thing that happens as well, which is related to it.
24:21
SPEAKER_01: what I sometimes have called advocacy-biased cognition,
24:26
SPEAKER_01: which is often when you're really motivated to advocate for a particular position
24:31
SPEAKER_01: or claim or narrative,
24:33
SPEAKER_01: what tends to happen is your beliefs shift in the direction of that claim or that narrative.
24:39
SPEAKER_01: So if you're really motivated to demonize the elites, let's say, as many conspiracy theories are,
24:45
SPEAKER_01: A conspiratorial narrative might perform that function well,
24:48
SPEAKER_01: so you're going to be motivated to spread it and advocate for it and so on.
24:52
SPEAKER_01: But in the process of advocating for it,
24:54
SPEAKER_01: what often happens is your beliefs shift in the direction of that narrative.
24:58
SPEAKER_01: I think a lot of politics also has this character.
25:02
SPEAKER_01: It benefits people who are really politically engaged and really care about their team winning.
25:07
SPEAKER_01: to demonize the other team in a really extreme way because they want to mobilize
25:11
SPEAKER_01: support for their coalition.
25:13
SPEAKER_01: They want to discredit their rivals and so on.
25:15
SPEAKER_01: But in the process of being an advocate,
25:18
SPEAKER_01: being a kind of propagandist,
25:20
SPEAKER_01: people end up shifting their beliefs in a direction of the things that they're
25:24
SPEAKER_01: motivated to spread or advocate for.
25:27
SPEAKER_01: I think that's another really big source of bias in these cases,
25:31
SPEAKER_01: which cuts against this goal of forming accurate beliefs.
25:35
SPEAKER_00: In other words, just the act of advocating for a position will nudge you more in that direction?
25:40
SPEAKER_00: Or am I oversimplifying that?
25:41
SPEAKER_00: I think that's right, yeah.
25:43
SPEAKER_01: My sense is,
25:43
SPEAKER_01: I mean,
25:44
SPEAKER_01: with all of these things,
25:46
SPEAKER_01: we're dealing with psychology here,
25:47
SPEAKER_01: so the empirical literature should always be taken with a grain of salt,
25:51
SPEAKER_01: to put it mildly.
25:51
SPEAKER_01: But yeah,
25:52
SPEAKER_01: I think there's relatively good empirical research suggesting that advocating for
25:58
SPEAKER_01: things or being motivated to push out or spread a particular kind of narrative is
26:03
SPEAKER_01: tends to have the effect of people internalizing that belief or that narrative.
26:07
SPEAKER_00: So I want to move on to some sort of real-world application of these ideas.
26:10
SPEAKER_00: But one other sort of nerdy thing I wanted to surface is this idea of,
26:16
SPEAKER_00: what do you call it,
26:18
SPEAKER_00: relational versus epistemic goals or types of thinking.
26:22
SPEAKER_00: Is the basic idea here...
26:24
SPEAKER_00: Just that a lot of thinking and a lot of expressions of opinion is simply not
26:28
SPEAKER_00: motivated by getting at the truth,
26:30
SPEAKER_00: whereas other times it is motivated by getting at the truth.
26:32
SPEAKER_00: And most of the time it's somewhere in the middle,
26:33
SPEAKER_00: but we should recognize these as separate impulses that will often lead us in
26:37
SPEAKER_00: different directions.
26:38
SPEAKER_01: Exactly.
26:39
SPEAKER_01: Yeah.
26:39
SPEAKER_01: And I'd say, I mean, it depends on the domain.
26:41
SPEAKER_01: So in everyday life, for the most part, we want to form accurate beliefs.
26:45
SPEAKER_01: Like if I'm going to cross the road, I want to know whether there's a car coming down and
26:48
SPEAKER_01: If I'm going to make dinner, I want to know what food I've got in the fridge, etc.
26:52
SPEAKER_01: So in those sorts of cases, your goal is to figure out how things are and to form accurate beliefs.
26:57
SPEAKER_01: But in domains like politics and cultural wars and forming beliefs about the supernatural and so on,
27:04
SPEAKER_01: In those kinds of contexts,
27:05
SPEAKER_01: often our goals are non-epistemic in the sense that they're not to do with
27:09
SPEAKER_01: acquiring knowledge.
27:11
SPEAKER_01: I think the two big ones,
27:12
SPEAKER_01: as I've already mentioned,
27:13
SPEAKER_01: are either signaling,
27:15
SPEAKER_01: marking ourselves out as somebody who has good traits,
27:18
SPEAKER_01: winning approval from others,
27:20
SPEAKER_01: or to do with advocacy and propaganda in the sense that we're motivated to push out
27:26
SPEAKER_01: or spread particular kinds of messages.
27:28
SPEAKER_00: So if we imagine the hypothetical caveman out on the prairie or whatever – I guess prairie.
27:34
SPEAKER_00: Were there woolly mammoths on the prairie?
27:36
SPEAKER_00: I want this story to involve a woolly mammoth because those are cool.
27:38
SPEAKER_00: Yeah, let's just go with that.
27:41
SPEAKER_00: Upper American Midwest caveman out with his tribe.
27:45
SPEAKER_00: If they're trying to track and kill a woolly mammoth –
27:49
SPEAKER_00: They need to know the truth of where the woolly mammoth went.
27:52
SPEAKER_00: They need to use the latest and greatest caveman tracking abilities to find it and kill it.
27:56
SPEAKER_00: Then that's epistemic, a matter of true knowledge, what the world is like.
28:01
SPEAKER_00: Then when they get the meat and harvest it and they're back in the cave sitting around the fire telling –
28:06
SPEAKER_00: pre-religious stories, the truth doesn't matter as much.
28:10
SPEAKER_00: It'd be a lot more beneficial and evolutionarily adaptive,
28:13
SPEAKER_00: not to be an evo-psych bro,
28:15
SPEAKER_00: but it would,
28:16
SPEAKER_00: to agree with whatever the group's supernatural story is.
28:19
SPEAKER_00: So is that an illustration of these two impulses, would you say?
28:22
SPEAKER_00: Exactly, yeah.
28:23
SPEAKER_01: And it's not just agreeing with the supernatural story.
28:26
SPEAKER_01: Often the supernatural story wears its social function on its sleeve.
28:31
SPEAKER_01: My favorite example of this is Mandir Singh has this research on the Mentawai people in Indonesia.
28:40
SPEAKER_01: They've got a belief in a crocodile-like supernatural spirit that hunts people down
28:47
SPEAKER_01: if they don't share their meat with members of the clan.
28:50
SPEAKER_00: And what I love about it- And you're saying that's literally true.
28:53
SPEAKER_01: Exactly.
28:54
SPEAKER_01: Yeah, that's completely true.
28:55
SPEAKER_01: But what I love about it is it so clearly serves this function of motivating people
29:02
SPEAKER_01: to share with other members of the community.
29:04
SPEAKER_01: And if you believe it and you show that you believe it,
29:06
SPEAKER_01: you signal to other people,
29:07
SPEAKER_01: I'm likely to be sharing and cooperative.
29:10
SPEAKER_01: So it's like there's this social motivation to adhere to this shared narrative,
29:14
SPEAKER_01: but also the narrative itself serves this obvious cooperation enhancing function.
29:20
SPEAKER_00: interesting uh okay let's briefly do housekeeping then we'll move on to the real
29:23
SPEAKER_00: world uh housekeeping we're a podcast blocked or imported.org you can become a
29:30
SPEAKER_00: premium subscriber by going to that website and for just five dollars a month you
29:33
SPEAKER_00: get uh i think we settled on like three and a half we can't really figure out
29:36
SPEAKER_00: exactly what we give our premium subscribers about three and a half episodes extra
29:41
SPEAKER_00: episodes a month uh with me and katie hersog my usual co-hosts
29:44
SPEAKER_00: You also get access to an amazing community,
29:47
SPEAKER_00: great comment sections,
29:48
SPEAKER_00: vibrant debate,
29:49
SPEAKER_00: very few jerks,
29:50
SPEAKER_00: a much lower jerk ratio than most online communities.
29:53
SPEAKER_00: Also,
29:54
SPEAKER_00: please rate and review us on Apple Podcasts and go to BlockedOrImported.reddit.com
29:58
SPEAKER_00: for more amazing conversation.
30:01
SPEAKER_00: All right, Dan Williams, let's move into the real world a little bit.
30:04
SPEAKER_00: Back over the summer, there were these riots in England.
30:06
SPEAKER_00: They were sparked by this horrific knife attack that killed and injured some young girls.
30:11
SPEAKER_00: It was at a dance class, right?
30:13
SPEAKER_01: Yeah.
30:15
SPEAKER_00: So you wrote a piece for your Substack titled, Did Online Misinformation Fuel the UK Riots?
30:20
SPEAKER_00: Here's a little quote from it.
30:23
SPEAKER_00: It is difficult to think of a clearer example of the dangers of online misinformation.
30:27
SPEAKER_00: Nevertheless,
30:28
SPEAKER_00: much of the controversy surrounding its role in the riots has been simplistic,
30:32
SPEAKER_00: uninformed,
30:33
SPEAKER_00: or misguided.
30:34
SPEAKER_00: It exaggerates and misrepresents misinformation's impact and encourages naive policy solutions that might
30:39
SPEAKER_00: This seems quite important, so let's take this on one at a time.
30:42
SPEAKER_00: Remind everyone what the rumors were that arguably helped to kick off the violence and destruction.
30:48
SPEAKER_01: The initial rumor was that the person responsible for these awful stabbings was a Muslim asylum seeker.
30:57
SPEAKER_01: If you know anything about British politics,
31:00
SPEAKER_01: that's the kind of rumor which give them both,
31:06
SPEAKER_01: let's say,
31:06
SPEAKER_01: Islamophobia and hostility towards asylum seekers.
31:13
SPEAKER_01: the whole phenomenon of immigration, you can see why that's going to be a very kind of incendiary rumor.
31:19
SPEAKER_00: And this rumor was false.
31:20
SPEAKER_00: It turned out the young man who did this was Rwandan.
31:25
SPEAKER_00: He was the son of migrants.
31:27
SPEAKER_00: And statistically, demographically, he's most likely Christian, although I'm not sure we know, right?
31:32
SPEAKER_01: Yeah.
31:32
SPEAKER_01: And I should say, I mean, he's British.
31:33
SPEAKER_01: I mean, his parents are Rwandan.
31:36
SPEAKER_01: Yeah, but he was British born.
31:38
SPEAKER_00: It seems like this is a pretty...
31:41
SPEAKER_00: clear-cut.
31:42
SPEAKER_00: I mean, you say it's difficult to think of a clear example.
31:44
SPEAKER_00: These false rumors about who this kid was spread far and wide,
31:48
SPEAKER_00: and then the riots pop off,
31:52
SPEAKER_00: and they were a big deal over there over the summer.
31:55
SPEAKER_00: Why do you think that the narratives about misinformation here, which have prevailed, are oversimplified?
32:00
SPEAKER_00: For many reasons.
32:01
SPEAKER_01: I should say, I mean, I'm not making the claim that the online fake news didn't play a role.
32:06
SPEAKER_01: I think it did in...
32:09
SPEAKER_01: Um,
32:09
SPEAKER_01: initially mobilizing many of the writers,
32:13
SPEAKER_01: but I think there's a lot of kind of context that got lost in the aftermath as
32:18
SPEAKER_01: people really focused in on in.
32:23
SPEAKER_01: as people really focused on online misinformation as this major player.
32:28
SPEAKER_01: One thing is,
32:30
SPEAKER_01: the people who were rioting were a loose coalition of far-right fanatics and thugs
32:39
SPEAKER_01: and people with antisocial personalities.
32:43
SPEAKER_01: Often, drugs and alcohol were playing a big role.
32:47
SPEAKER_01: Now, the fake news was that this was a Muslim asylum seeker.
32:52
SPEAKER_01: The real news, the person who was charged was a black teenager and son of Rwandan immigrants.
33:01
SPEAKER_01: I think the idea that things would have played out very differently
33:05
SPEAKER_01: in response to the true news relative to the fake news in this case is not that plausible,
33:11
SPEAKER_01: given that even once the initial fake news was debunked,
33:16
SPEAKER_01: the riots spread and people still viewed it as indicative of general problems with
33:21
SPEAKER_01: UK immigration and so on.
33:23
SPEAKER_01: I think in large part because the suspect was black and was the son of Rwandan immigrants.
33:29
SPEAKER_01: The second thing is, it's a weird thing to blame online misinformation in this case.
33:34
SPEAKER_01: In the sense that even if the fake news had been true,
33:39
SPEAKER_01: the riots would have been equally appalling,
33:42
SPEAKER_01: irrational,
33:43
SPEAKER_01: unjustified.
33:44
SPEAKER_01: Which is not to say that they can't have played a role,
33:47
SPEAKER_01: but it's a sort of odd situation where normally we think when we blame
33:50
SPEAKER_01: misinformation for things,
33:52
SPEAKER_01: it's because if the misinformation had been true,
33:55
SPEAKER_01: the actions would have made sense.
33:56
SPEAKER_01: Whereas in this case, if the information had been true, the actions would have been similarly awful.
34:03
SPEAKER_01: That's especially important because even though in this case,
34:06
SPEAKER_01: the fake news that this was a Muslim asylum seeker responsible for these killings,
34:11
SPEAKER_01: that was false,
34:13
SPEAKER_01: there have been cases in recent years where asylum seekers have engaged in murders.
34:18
SPEAKER_01: It's not as if this is a crazy outlandish scenario.
34:21
SPEAKER_01: It suggests that the real problem is there are people in society in the UK who are
34:26
SPEAKER_01: willing to engage in these awful riots and attacks and so on.
34:32
SPEAKER_01: I also just think the whole framing of this in the aftermath was social media
34:37
SPEAKER_01: played this incredibly consequential role and people came across this
34:41
SPEAKER_01: misinformation and it sort of infected their brains and they engaged in this bad behavior.
34:46
SPEAKER_01: Whereas I think a lot of what was going on is you had a segment of the population
34:49
SPEAKER_01: with pre-existing very racist views who actively sought out and amplified this content.
34:56
SPEAKER_01: which you see throughout human history.
34:59
SPEAKER_01: Race riots and false inflammatory rumors that precede them long,
35:04
SPEAKER_01: long predate social media,
35:05
SPEAKER_01: suggesting that even if social media did play a role,
35:09
SPEAKER_01: it wasn't a necessary factor for the general phenomenon that we observe.
35:13
SPEAKER_01: You get these kinds of riots in the absence of social media.
35:17
SPEAKER_01: That's all my way of saying that even though I'm not denying that I think the fake
35:20
SPEAKER_01: news played a role,
35:22
SPEAKER_01: at least initially,
35:23
SPEAKER_01: much of the commentary around it and in the aftermath,
35:26
SPEAKER_01: which really focused on fake news as a big player,
35:29
SPEAKER_01: I think was pretty simplistic.
35:31
SPEAKER_00: You're a philosopher, so I take it you won't be put off by a ridiculous thought experiment.
35:36
SPEAKER_00: But imagine we could clone our world exactly as it is.
35:40
SPEAKER_00: and take away social media?
35:41
SPEAKER_00: Just on the question of ethnic strife and race riots,
35:46
SPEAKER_00: does Earth 2 without social media have more or less of it?
35:51
SPEAKER_01: I mean, the honest answer is nobody knows.
35:53
SPEAKER_01: I mean,
35:54
SPEAKER_01: we can look historically at the rate and the severity of race riots before the era
36:00
SPEAKER_01: of social media and after social media,
36:02
SPEAKER_01: or that is during the era of social media.
36:05
SPEAKER_01: And of course,
36:06
SPEAKER_01: I mean,
36:06
SPEAKER_01: there's a great book by,
36:07
SPEAKER_01: I think it's Donald Horowitz,
36:09
SPEAKER_01: called The Deadly Ethnic Riot,
36:10
SPEAKER_01: where he documents ethnic riots across loads of different countries throughout
36:14
SPEAKER_01: human history,
36:15
SPEAKER_01: many of which are far,
36:17
SPEAKER_01: far worse than what we saw recently in the UK,
36:19
SPEAKER_01: and an overwhelming majority of which happened prior to social media.
36:24
SPEAKER_01: In terms of that counterfactual, it's really difficult to say, I think.
36:27
SPEAKER_01: Social media clearly played a role in the sense of that online fake news was circulated on social media.
36:34
SPEAKER_01: People who participated in the initial riots had seen it.
36:37
SPEAKER_01: But that counterfactual, I think, is really, really difficult to evaluate.
36:42
SPEAKER_01: Yet much of the discourse,
36:43
SPEAKER_01: I think,
36:44
SPEAKER_01: assumes we know the answer to that counterfactual,
36:46
SPEAKER_01: which is that in a world without social media,
36:49
SPEAKER_01: we wouldn't see this kind of thing.
36:50
SPEAKER_01: That is what strikes me as really implausible.
36:52
SPEAKER_00: Well, yeah.
36:53
SPEAKER_00: I mean the argument we wouldn't have ethnic – I mean of course we would have ethnic
36:56
SPEAKER_00: riots without social media.
36:57
SPEAKER_00: But when I think of – in the United States,
37:00
SPEAKER_00: one of the worst outbreaks of violence surrounded the shooting of a guy named Jacob Blake.
37:06
SPEAKER_00: It led to rioting in Kenosha.
37:08
SPEAKER_00: It led to deaths.
37:09
SPEAKER_00: It led to several blocks getting burned.
37:11
SPEAKER_00: And at the time, because of social media, video of this police shooting spread far and wide.
37:18
SPEAKER_00: And it left such a mark on society that the Democratic presidential candidates flew
37:22
SPEAKER_00: to be beside the guy who got shot.
37:26
SPEAKER_00: And it led to this chain of events of violence.
37:28
SPEAKER_00: rioting and death and destruction.
37:29
SPEAKER_00: And only later,
37:31
SPEAKER_00: I was taken in by the misinformation only later,
37:35
SPEAKER_00: after a couple independent reviews or one independent review,
37:37
SPEAKER_00: it looks like this was a justified police shooting,
37:40
SPEAKER_00: at least in the US where everyone has guns and police shoot people a lot very
37:44
SPEAKER_00: different over there.
37:46
SPEAKER_00: To me,
37:46
SPEAKER_00: that's the kind of case where it's hard for me to imagine that going down the way
37:51
SPEAKER_00: it is if there hadn't been this very scary,
37:55
SPEAKER_00: horrifying,
37:56
SPEAKER_00: out-of-context video that seeded the public.
37:58
SPEAKER_00: And now I'm falling into the language we're about to partially debunk,
38:01
SPEAKER_00: but the reason I thought this guy had gotten cruelly,
38:06
SPEAKER_00: unjustifiably shot in the back by cops is because I saw a video that seemed to show that.
38:10
SPEAKER_00: And that video was getting passed around, and that led to
38:14
SPEAKER_00: I think death and destruction to me,
38:18
SPEAKER_00: those are the situations where it's sort of hard to,
38:21
SPEAKER_00: we're all speculating here,
38:22
SPEAKER_00: but it's hard to discount the role of social media.
38:25
SPEAKER_00: I think in setting off individual instances,
38:29
SPEAKER_00: I guess what you're saying,
38:31
SPEAKER_00: which is true,
38:31
SPEAKER_00: but feels unsatisfying is like,
38:33
SPEAKER_00: there's just no way to know.
38:34
SPEAKER_00: Cause we don't have the, the counterfactual, the Jacob Blake shooting with no social media.
38:40
SPEAKER_00: Yeah.
38:40
SPEAKER_01: Yeah,
38:41
SPEAKER_01: I think we can't know,
38:42
SPEAKER_01: but I also think if we look at history,
38:44
SPEAKER_01: we see cases where inflammatory anecdotes,
38:49
SPEAKER_01: rumors,
38:50
SPEAKER_01: cherry-picked events- Wishing,
38:51
SPEAKER_00: Rodney King,
38:52
SPEAKER_00: yeah,
38:52
SPEAKER_00: everything else.
38:53
SPEAKER_01: Right, these things circulate really rapidly.
38:55
SPEAKER_01: But the other thing I would say is, okay, social media is playing a role.
39:00
SPEAKER_01: There's this mass communication, often spreads quite quickly.
39:04
SPEAKER_01: There's nothing we can do about certain aspects of that problem.
39:07
SPEAKER_01: We now exist within this era of social media.
39:11
SPEAKER_01: The days when there was a kind of elite monopoly on information dissemination are
39:16
SPEAKER_01: just categorically over.
39:17
SPEAKER_01: And there are so many ways now where people,
39:20
SPEAKER_01: so many means by which people can communicate across lots of different social media
39:24
SPEAKER_01: platforms and private messaging and so on.
39:27
SPEAKER_01: So you can have censorship, content moderation decisions, which might improve things at the margin.
39:33
SPEAKER_01: They might also have all sorts of negative consequences.
39:36
SPEAKER_01: But the more general phenomenon of stuff circulating rapidly across big communities,
39:41
SPEAKER_01: that's just the era that we're in now.
39:44
SPEAKER_01: It's important to recognize that when we now try to analyze these sorts of events,
39:48
SPEAKER_01: there are certain things you can intervene on and try to change and improve,
39:52
SPEAKER_01: and there are certain things you can't.
39:53
SPEAKER_01: And the era of mass communication is the era we're in, and we're not going to be able to change that.
39:58
SPEAKER_00: So in other words, it's almost irrelevant whether or not social media causes this or that.
40:02
SPEAKER_00: Our only hope is to study it and understand it and try to figure out which
40:06
SPEAKER_00: interventions are worth engaging in.
40:08
SPEAKER_01: Yeah, that would be the idea.
40:09
SPEAKER_00: So, okay, this sort of gets into your gripes with the modern anti-misinformation industry.
40:16
SPEAKER_00: What do you call it?
40:17
SPEAKER_00: Do you have a more graceful name for it?
40:19
SPEAKER_01: Sometimes people talk about big disinfo,
40:22
SPEAKER_01: which,
40:23
SPEAKER_00: you know,
40:23
SPEAKER_01: it sounds a little bit conspiratorial and a little bit like...
40:26
SPEAKER_01: The people within it are being sort of discredited or demonized,
40:31
SPEAKER_01: so I've got mixed feelings about it as a term.
40:33
SPEAKER_01: But whatever you call it,
40:35
SPEAKER_01: there's this constellation of academics,
40:37
SPEAKER_01: people in mainstream media,
40:38
SPEAKER_01: think tanks,
40:39
SPEAKER_01: non-governmental organizations that are concerned with identifying and combating misinformation.
40:45
SPEAKER_00: Gotcha.
40:46
SPEAKER_00: And I'm going in pretty biased.
40:48
SPEAKER_00: I think the closest I've looked into one of these outfits is the Center for Countering Digital Hate.
40:53
SPEAKER_00: I don't even remember the details,
40:54
SPEAKER_00: but I looked into the specifics of how they prepared this one report on online harassment,
40:59
SPEAKER_00: and it was ridiculous.
41:01
SPEAKER_00: It was just completely methodologically broken.
41:04
SPEAKER_00: And yet,
41:04
SPEAKER_00: because they're an organization called the Center for Countering Digital Hate,
41:08
SPEAKER_00: and who wouldn't want to counter digital hate?
41:11
SPEAKER_00: It was just carried very credulously by major outlets, including the Washington Post.
41:15
SPEAKER_00: So on that subject,
41:17
SPEAKER_00: last year in the Boston Review,
41:19
SPEAKER_00: you wrote what I thought was a very convincing review essay of a book called Foolproof,
41:23
SPEAKER_00: Why Misinformation Infects Our Minds and How to Build Immunity by Sander von der Linden.
41:29
SPEAKER_00: Van der Linden, probably mispronouncing that, is a Dutch social psychologist at Cambridge.
41:33
SPEAKER_00: That's where you got your PhD.
41:34
SPEAKER_00: He's seen as one of the most celebrated gurus for fighting misinformation.
41:38
SPEAKER_00: So his book,
41:39
SPEAKER_00: which to be clear,
41:39
SPEAKER_00: I haven't read,
41:40
SPEAKER_00: basically touts the fact that he's come up with an experimental intervention that
41:45
SPEAKER_00: can help inoculate people against misinformation.
41:48
SPEAKER_00: These are his words, not yours.
41:50
SPEAKER_00: This got a lot of favorable coverage.
41:51
SPEAKER_00: You tick off some of the headlines.
41:52
SPEAKER_00: CNN researchers have created a quote unquote vaccine for fake news.
41:57
SPEAKER_00: I'll try to read these without cracking up.
41:59
SPEAKER_00: Scientific American,
42:02
SPEAKER_00: as scientific as always,
42:03
SPEAKER_00: they write,
42:03
SPEAKER_00: there's a psychological vaccine against misinformation.
42:07
SPEAKER_00: Rolling Stone at least did the question in the headline thing rather than stating it as a fact.
42:10
SPEAKER_00: That headline reads, the disinformation vaccine.
42:13
SPEAKER_00: Is there a cure for conspiracy theories?
42:16
SPEAKER_00: Wouldn't that be nice?
42:16
SPEAKER_00: So quickly walk us through Vanderlinden's intervention.
42:21
SPEAKER_00: It's basically this pretty straightforward online game you can play, right?
42:24
SPEAKER_00: Yeah.
42:25
SPEAKER_01: Maybe I'll just say two quick things before that.
42:28
SPEAKER_01: Firstly, I don't know Sander van der Linden personally, but I respect him as a person and as a scholar.
42:35
SPEAKER_01: It's just that I see things very, very differently to him.
42:39
SPEAKER_00: It would be a better interview if you said you disrespected him as a person, but I'll let that slide.
42:44
SPEAKER_01: The second thing to say is,
42:46
SPEAKER_01: you quoted this favorable coverage,
42:49
SPEAKER_01: but the other bit of important context is,
42:52
SPEAKER_01: it's not just that it's been covered favorably.
42:54
SPEAKER_01: The interventions based on the work reported in that book have been massively influential.
43:00
SPEAKER_01: They've reached millions of people, funded and supported by tech companies, Google.
43:06
SPEAKER_01: meta supported by governments, intergovernmental organizations and so on.
43:10
SPEAKER_01: So that's why I think I was interested in trying to debunk this set of ideas
43:16
SPEAKER_01: because it's not just favorably viewed by many,
43:19
SPEAKER_01: it's also massively impactful in terms of how governments and tech companies try to
43:23
SPEAKER_01: deal with the issue.
43:26
SPEAKER_01: Now, the intervention is, as you mentioned, based on this analogy with inoculation.
43:32
SPEAKER_01: And the idea is just as traditional vaccines would expose people to a weakened dose
43:38
SPEAKER_01: of the virus so that their immune system would learn to identify it and they would
43:44
SPEAKER_01: develop antibodies against it.
43:46
SPEAKER_01: The idea behind this intervention is if you expose people to a weakened dose of
43:52
SPEAKER_01: what van der Linden calls the DNA,
43:54
SPEAKER_01: or sometimes referred to as the fingerprints of misinformation,
43:57
SPEAKER_01: either by playing an online game,
43:59
SPEAKER_01: as you mentioned,
44:00
SPEAKER_01: where you take on the role of a fake news tycoon using various misinformation techniques,
44:06
SPEAKER_01: or through short videos,
44:08
SPEAKER_01: short animations,
44:10
SPEAKER_01: If you're exposed to the sort of alleged DNA of misinformation in those contexts,
44:14
SPEAKER_01: then you build up quote-unquote mental antibodies against misinformation,
44:19
SPEAKER_01: so you're better at detecting it in the wild.
44:21
SPEAKER_01: That's, roughly speaking, the core idea.
44:23
SPEAKER_00: So they take this inoculation idea pretty seriously.
44:26
SPEAKER_01: Right, yeah.
44:27
SPEAKER_00: So I guess we got to go through this depict thing.
44:31
SPEAKER_00: Again, I haven't read his book yet.
44:35
SPEAKER_00: He basically says that misinformation,
44:37
SPEAKER_00: but not accurate information,
44:38
SPEAKER_00: has certain fingerprints in how it's presented.
44:41
SPEAKER_00: Depict, D for discrediting, E for emotion, P for polarization, I for impersonation.
44:48
SPEAKER_00: I think you didn't get to that one in your book review.
44:51
SPEAKER_00: What's impersonation?
44:52
SPEAKER_01: It's just what it sounds like.
44:53
SPEAKER_01: So if you're on Twitter and you see an account called Barack Obama and it's got 12
44:57
SPEAKER_01: followers and it hosts chewing on content,
44:59
SPEAKER_01: it's probably not Barack Obama.
45:01
SPEAKER_00: You're saying these are hints that this might not be the real Barack Obama.
45:05
SPEAKER_00: Yeah.
45:05
SPEAKER_00: Yeah.
45:05
SPEAKER_00: Okay.
45:06
SPEAKER_00: C for conspiracy, T for trolling.
45:10
SPEAKER_00: I guess I was trying to do a good journalistic interview here, not betray my biases.
45:17
SPEAKER_00: I was shocked that
45:21
SPEAKER_00: this was what he came up with like okay so d for discrediting someone it's a sign
45:27
SPEAKER_00: of misinformation if the person making that argument tries to discredit those who
45:31
SPEAKER_00: agree with him i'm at least describing it right correctly yeah so the idea is fake
45:36
SPEAKER_01: news purveyors will try to discredit reliable sources of information as a way of
45:42
SPEAKER_01: getting people to accept their content
45:45
SPEAKER_00: But as you point out, everyone does this.
45:48
SPEAKER_00: This is going on in the very book explaining the depict model.
45:52
SPEAKER_00: Similar with E for emotion, people make emotionally charged appeals.
45:56
SPEAKER_00: Yeah, no shit.
45:57
SPEAKER_00: I mean, go back to Aristotle.
46:02
SPEAKER_00: Sorry, I'm not trying to be a jerk about this.
46:03
SPEAKER_00: So basically, your argument is all of these are not, in fact, reliable or valid signs of misinformation.
46:13
SPEAKER_00: They're just how people make arguments.
46:15
SPEAKER_01: Yeah,
46:15
SPEAKER_01: I mean,
46:15
SPEAKER_01: the way I would put it is there's an argument,
46:19
SPEAKER_01: which is not just unique to Vanderlinden,
46:21
SPEAKER_01: I think is quite popular in a positive misinformation research.
46:25
SPEAKER_01: There's something like emotional language is diagnostic of misinformation.
46:29
SPEAKER_01: And in defense of this,
46:30
SPEAKER_01: people will often point to the fact that loss of misinformation features emotional language.
46:35
SPEAKER_01: The thing that I point out in the review is,
46:38
SPEAKER_01: well,
46:38
SPEAKER_01: that's true,
46:38
SPEAKER_01: but it's also true that loss of communication featuring emotional language is
46:43
SPEAKER_01: totally legitimate.
46:46
SPEAKER_01: It's also the case that some of the most consequential propaganda deliberately uses dispassionate,
46:53
SPEAKER_01: neutral language.
46:54
SPEAKER_01: It's one of the oldest propaganda tactics.
46:56
SPEAKER_01: If you describe torture as enhanced interrogation,
46:59
SPEAKER_01: that's a way of misleading people by avoiding activating certain kinds of emotions.
47:05
SPEAKER_01: It's even worse than that because sometimes the very act of describing emotional
47:09
SPEAKER_01: language and emotional content as unreliable can itself be a propaganda tactic
47:15
SPEAKER_01: because it's a way of delegitimizing people who are criticizing a system as being
47:20
SPEAKER_01: hysterical and so on and so forth.
47:22
SPEAKER_01: And for me, the lesson there is just figuring out what's true is really, really complicated.
47:28
SPEAKER_01: And it's very context sensitive.
47:29
SPEAKER_01: And people communicate with different kinds of motivations and interests and lots of different contexts.
47:35
SPEAKER_01: And it would be nice if misinformation had these kind of alarm bells on it, these really simple tells.
47:41
SPEAKER_01: But when you go through all of these cases,
47:44
SPEAKER_01: even just reflecting upon it from the armchair,
47:47
SPEAKER_01: it's clear that it's so much more complicated than that.
47:51
SPEAKER_01: I should say,
47:52
SPEAKER_01: one of the bits of pushback I got after publishing this review is there are some
47:58
SPEAKER_01: content analyses where people look at reliable information or ostensibly reliable information.
48:04
SPEAKER_01: They look at misinformation,
48:06
SPEAKER_01: the database of misinformation,
48:08
SPEAKER_01: and then extract certain kinds of features that distinguish the two.
48:11
SPEAKER_01: There's one study called the Fingerprints of Misinformation where
48:16
SPEAKER_01: What happens is there's a big data set and in the reliable information category,
48:21
SPEAKER_01: you've got articles by the New York Times,
48:24
SPEAKER_01: the Guardian,
48:24
SPEAKER_01: and I think it's the Wall Street Journal.
48:27
SPEAKER_01: And it's assumed that 100% of the content from these outlets is reliable.
48:31
SPEAKER_00: Wait, sorry.
48:32
SPEAKER_00: It's just assumed as like part of the methodology that 100% of them are reliable?
48:36
SPEAKER_01: Yeah.
48:36
SPEAKER_01: And this is very, very common in misinformation research.
48:38
SPEAKER_01: So defining- That's insane.
48:40
SPEAKER_01: I think it's insane.
48:42
SPEAKER_00: Yeah.
48:43
SPEAKER_01: And then there's a fake news corpus where this includes things like Infowars,
48:47
SPEAKER_01: like there's a British tabloid newspaper in there,
48:49
SPEAKER_01: and it's crazy conspiratorial fringe websites online.
48:53
SPEAKER_01: And if you do that content analysis,
48:56
SPEAKER_01: what you end up learning is content from Infowars features more emotional language than,
49:02
SPEAKER_01: say,
49:02
SPEAKER_01: content from the New York Times,
49:04
SPEAKER_01: on average.
49:05
SPEAKER_00: Oh, wait, that's actually...
49:07
SPEAKER_01: The study?
49:08
SPEAKER_01: Yeah.
49:09
SPEAKER_01: And there are content analyses like that.
49:12
SPEAKER_01: And I would say it's not as if we need science really to establish that kind of lesson.
49:16
SPEAKER_01: I think for most people, that's going to be fairly obvious.
49:20
SPEAKER_01: To see how problematic it would be to take the findings from that study and
49:26
SPEAKER_01: generalize them to,
49:27
SPEAKER_01: we've discovered the markers of misinformation in general,
49:31
SPEAKER_01: one of the things they find is that misinformation uses smaller words and a smaller vocabulary.
49:41
SPEAKER_01: If you read articles from Infowars,
49:44
SPEAKER_01: you've got smaller words,
49:44
SPEAKER_01: smaller vocabulary than if you read articles from the New York Times,
49:47
SPEAKER_01: which again,
49:48
SPEAKER_01: I think is going to be fairly obvious to most people.
49:51
SPEAKER_01: But think about how problematic it would be to treat a small vocabulary or the use
49:57
SPEAKER_01: of short words as a marker of misinformation.
50:01
SPEAKER_01: There's no inherent connection at all between how big your vocabulary is and how
50:05
SPEAKER_01: long your words are and misinformation.
50:08
SPEAKER_01: This is the kind of research within this area which I think is often given much
50:14
SPEAKER_01: more credence than it deserves.
50:17
SPEAKER_00: I mean, the problem is to run an actual study on this...
50:21
SPEAKER_00: I mean, it's tricky.
50:22
SPEAKER_00: If you want to do a study explaining the current tone of online disinformation or
50:27
SPEAKER_00: misinformation purveyors like Alex Jones,
50:29
SPEAKER_00: it's valid to point out they use emotional language.
50:31
SPEAKER_00: It's hysterical.
50:32
SPEAKER_00: It's a very specific style.
50:33
SPEAKER_00: But I think you're alluding to this.
50:34
SPEAKER_00: That doesn't actually provide you with a useful tool belt for identifying misinformation in the wild.
50:41
SPEAKER_00: I could say...
50:43
SPEAKER_00: The United States was involved with drug running that funded death squads in Central America.
50:50
SPEAKER_00: I'm stating it very calmly.
50:53
SPEAKER_00: It sounds psychotic.
50:54
SPEAKER_00: It also happens to be true.
50:55
SPEAKER_00: Similarly, I could make a very emotional appeal that happens to be false.
50:59
SPEAKER_00: It's just – I don't know.
51:01
SPEAKER_00: This shouldn't surprise me.
51:02
SPEAKER_00: I wrote a book called The Quick Fix that was about –
51:06
SPEAKER_00: shoddy social psychology,
51:07
SPEAKER_00: but the research you've described so far,
51:11
SPEAKER_00: I'm still a little bit shocked that this is what counts for sort of mainstream
51:15
SPEAKER_00: solid research on misinformation.
51:17
SPEAKER_00: But let's get back to your book review essay,
51:18
SPEAKER_00: because this is,
51:20
SPEAKER_00: Vanderlinden is like one of the guys,
51:21
SPEAKER_00: and he supposedly has research
51:25
SPEAKER_00: backing up his approach.
51:26
SPEAKER_00: He claims,
51:28
SPEAKER_00: probably not as strongly as those headlines,
51:29
SPEAKER_00: but he's more or less claiming that he has come up with at least a partial vaccine
51:33
SPEAKER_00: for misinformation,
51:33
SPEAKER_00: right?
51:34
SPEAKER_00: Yeah,
51:34
SPEAKER_01: I think that the charitable way of interpreting it would be,
51:40
SPEAKER_01: okay,
51:40
SPEAKER_01: this is not going to perfectly,
51:42
SPEAKER_01: quote unquote,
51:42
SPEAKER_01: inoculate people against misinformation,
51:44
SPEAKER_01: but it's going to improve their ability to distinguish good from bad content in the wild.
51:49
SPEAKER_01: That's, I guess, the charitable way of describing it.
51:51
SPEAKER_01: Yeah.
51:52
SPEAKER_00: And what are the problems you found in his research and his justifications in the book?
51:56
SPEAKER_00: Well,
51:57
SPEAKER_01: in terms of the experimental data,
52:00
SPEAKER_01: so one issue is there was a meta-analysis that came out and what it shows is that
52:06
SPEAKER_01: the intervention just makes people more skeptical across the board.
52:10
SPEAKER_01: So they're more likely to rate misinformation as is classified within the
52:14
SPEAKER_01: experiment as misinformation,
52:16
SPEAKER_01: but they're also more likely to characterize true or reliable content as misinformation.
52:22
SPEAKER_01: And if you think,
52:22
SPEAKER_01: as I think,
52:23
SPEAKER_01: that many issues associated with misinformation are sort of downstream of distrust,
52:28
SPEAKER_01: making people more distrustful or more skeptical across the board is probably not
52:31
SPEAKER_01: going to be a great thing.
52:33
SPEAKER_01: Now,
52:33
SPEAKER_01: there is some controversy about that because people will point to studies where
52:37
SPEAKER_01: that doesn't seem to be the case.
52:39
SPEAKER_01: But then I think the second much more fundamental issue is the examples of
52:44
SPEAKER_01: misinformation which are used within the experiments are synthetic.
52:48
SPEAKER_01: And what that means is the experiments just make them up.
52:52
SPEAKER_01: which gives you lots of degrees of freedom to come up with examples of misinformation,
52:57
SPEAKER_01: which are going to maximize the chances the intervention would look successful.
53:02
SPEAKER_00: Let's just pause on that.
53:03
SPEAKER_00: In research,
53:04
SPEAKER_00: a degree of freedom is basically the choices researchers have for designing their experiment.
53:09
SPEAKER_00: I'm oversimplifying this a little,
53:10
SPEAKER_00: but if a researcher has a million degrees of freedom and can tweak things endlessly
53:15
SPEAKER_00: all they want,
53:16
SPEAKER_00: they will be able to,
53:17
SPEAKER_00: consciously or unconsciously,
53:20
SPEAKER_00: develop a study that proves what they want to prove.
53:22
SPEAKER_00: It's sort of a version of torture numbers, and they will confess to anything.
53:25
SPEAKER_01: Yeah, and to be clear, I'm...
53:27
SPEAKER_01: Categorically not saying there's anything deliberate or conscious or fraudulent.
53:31
SPEAKER_01: I don't think that's the situation.
53:32
SPEAKER_00: No, it doesn't have to be.
53:33
SPEAKER_01: It's just, as you say, there's every incentive in the world to get positive findings.
53:38
SPEAKER_01: And the more degrees of freedom there are in the experimental design,
53:41
SPEAKER_01: the more likely it is people are going to end up generating positive findings,
53:45
SPEAKER_01: even if there's nothing
53:47
SPEAKER_01: malicious or deliberate or fraudulent going on.
53:49
SPEAKER_01: I think in the overwhelming majority of cases, there isn't any of that going on.
53:54
SPEAKER_01: It's just that we tend to be biased towards things which generate the results that we want.
54:01
SPEAKER_01: That's to describe things abstractly, but what they do in these experiments is
54:08
SPEAKER_01: What they're supposed to show is by learning to detect the quote-unquote
54:12
SPEAKER_01: fingerprints of misinformation,
54:14
SPEAKER_01: people become better at detecting misinformation.
54:17
SPEAKER_01: But the examples of misinformation they use in the experiments only qualify as
54:22
SPEAKER_01: misinformation because they feature the alleged fingerprints.
54:27
SPEAKER_01: And that might seem like a sort of dry technical point,
54:29
SPEAKER_01: but what it shows is,
54:31
SPEAKER_01: unless you've got independent reason to think that these fingerprints are actually
54:35
SPEAKER_01: diagnostic of misinformation in the wild,
54:37
SPEAKER_01: there's no reason at all to think the success in the experiment will generalize to
54:42
SPEAKER_01: a better ability to detect misinformation in the wild.
54:46
SPEAKER_00: So he's saying depict is a useful diagnostic or predictive framework and then
54:52
SPEAKER_00: basing his success on that without ever proving that the depict thing works for
54:58
SPEAKER_00: identifying misinformation in the wild.
55:00
SPEAKER_01: Yeah, so the presupposition is precisely that these are the fingerprints of misinformation.
55:05
SPEAKER_01: If you don't buy into that,
55:06
SPEAKER_01: then the whole experimental design and the results of that experimental design
55:10
SPEAKER_01: don't provide you with any reason to think that people who improve as a consequence
55:15
SPEAKER_01: of the intervention within the experiment are going to be better at detecting
55:18
SPEAKER_01: misinformation in the real world.
55:21
SPEAKER_00: I've struggled with the question of political bias in academia,
55:24
SPEAKER_00: because obviously academia is overwhelmingly liberal.
55:27
SPEAKER_00: I think the question of how that bleeds into research is really context-dependent,
55:33
SPEAKER_00: and at least in the work I've done looking at interventions to reduce...
55:37
SPEAKER_00: Inequality,
55:38
SPEAKER_00: poverty,
55:39
SPEAKER_00: racism,
55:40
SPEAKER_00: you could just as much argue that they're biased toward a neoliberal approach than
55:44
SPEAKER_00: a lefty approach.
55:45
SPEAKER_00: This thing defeats the left-right spectrum.
55:49
SPEAKER_00: I will say,
55:50
SPEAKER_00: when I looked into research on why conservatives believe what they believe,
55:55
SPEAKER_00: I found a lot of really biased instruments that,
55:59
SPEAKER_00: for example,
56:00
SPEAKER_00: would like...
56:01
SPEAKER_00: Basically only ask about one side of the ledger like is violence ever acceptable to
56:06
SPEAKER_00: defend the US flag I'm making up that example but if you have an instrument that
56:10
SPEAKER_00: asks is violence only ever acceptable to defend the US flag but not is violence
56:15
SPEAKER_00: ever acceptable to protect someone from racism or some other left coded belief
56:18
SPEAKER_00: that's a broken instrument because only conservatives will be able to score high on it.
56:23
SPEAKER_00: In your deep dive into the papers on misinformation,
56:26
SPEAKER_00: whether Vanderlinden or others,
56:28
SPEAKER_00: have you found this?
56:29
SPEAKER_00: Because I have noticed an annoying tendency
56:35
SPEAKER_00: My own prior on this is that right-wing misinformation.
56:37
SPEAKER_00: If we had to say which is worse,
56:39
SPEAKER_00: which I think is a little dumb,
56:40
SPEAKER_00: if we had to,
56:41
SPEAKER_00: I think they have a more developed fake news ecosystem and more problems with rumor mongering.
56:45
SPEAKER_00: I really genuinely think the left has been catching up in recent years,
56:50
SPEAKER_00: and I'm dismayed by how much that's ignored.
56:52
SPEAKER_00: Jacob Blake is a great example.
56:55
SPEAKER_00: It's an American one,
56:55
SPEAKER_00: so I wouldn't expect non-Americans to know about it,
56:57
SPEAKER_00: but ask most American liberals what happened to Jacob Blake.
57:00
SPEAKER_00: The cops shot him in the back for no reason, basically.
57:02
SPEAKER_00: It was just
57:02
SPEAKER_00: isn't what happened.
57:03
SPEAKER_00: And it's corrosive to society and sometimes leads to rioting for misinformation spread.
57:08
SPEAKER_00: Is this a problem in the research you've seen on misinformation?
57:12
SPEAKER_01: This is such a complicated issue.
57:15
SPEAKER_01: And I think it's sort of important and difficult to keep two ideas in one's head at
57:21
SPEAKER_01: the same time,
57:21
SPEAKER_01: which is
57:23
SPEAKER_01: As you say,
57:23
SPEAKER_01: if you're focusing on things like fake news and really low-quality misinformation,
57:29
SPEAKER_01: as well as just blatant falsehoods from elites,
57:34
SPEAKER_01: My view would be, in America at least, that stuff is more common on the right.
57:39
SPEAKER_01: But I think it's also true that this whole misinformation industry is biased in the
57:47
SPEAKER_01: kinds of misinformation that it focuses on.
57:50
SPEAKER_01: So if you ask,
57:51
SPEAKER_01: for example,
57:52
SPEAKER_01: how many articles are there from misinformation researchers exploring
57:57
SPEAKER_01: misinformation associated with
58:00
SPEAKER_01: anti-racist movements,
58:01
SPEAKER_01: climate activists,
58:03
SPEAKER_01: feminists,
58:05
SPEAKER_01: anything to do with these issues of gender identity when it comes from progressive movements,
58:09
SPEAKER_01: there's hardly any compared to the enormous amount of attention and focus on
58:15
SPEAKER_01: misinformation when it comes from the political right.
58:20
SPEAKER_01: In that sense,
58:20
SPEAKER_01: just in terms of the amount of energy and attention poured into different kinds of misinformation,
58:26
SPEAKER_01: I haven't studied this rigorously,
58:28
SPEAKER_01: but my overwhelming impression is there is that kind of bias.
58:33
SPEAKER_01: I think that should cast a shadow over many of the alleged findings within the field.
58:39
SPEAKER_01: For example, the idea that conservatives are more susceptible to misinformation.
58:45
SPEAKER_01: That kind of finding,
58:46
SPEAKER_01: unless you're extremely careful to make sure that you've got a politically unbiased
58:51
SPEAKER_01: assessment of misinformation across all possible issues,
58:56
SPEAKER_01: is not going to be reliable.
58:57
SPEAKER_01: Whenever I've looked into that,
58:58
SPEAKER_01: I think there's no reason to think that the research has really established that
59:02
SPEAKER_01: representative treatment of all relevant issues in politics.
59:06
SPEAKER_01: So I think it is complicated in the sense that I do think there are aspects of the
59:10
SPEAKER_01: political right today,
59:11
SPEAKER_01: which are just a kind of epistemic dumpster fire.
59:14
SPEAKER_01: But I think it's also true that the people that study misinformation are
59:18
SPEAKER_01: politically biased in terms of the kinds of misinformation that they focus on.
59:22
SPEAKER_00: What theory and I guess some empirical work would tell us in general is that we
59:26
SPEAKER_00: should expect that overall people are most susceptible to misinformation when the
59:32
SPEAKER_00: subject is something they're emotional and passionate and tribal about,
59:37
SPEAKER_00: such as a police shooting on the left or,
59:39
SPEAKER_00: I don't know,
59:39
SPEAKER_00: illegal immigration on the right and sort of what Kahneman and Tversky would call
59:44
SPEAKER_00: system one thinking.
59:45
SPEAKER_00: kick in now even here you'd have to test it in an individual context but in terms
59:50
SPEAKER_00: of like what listeners can take away from this and improving their own epistemic
59:54
SPEAKER_00: hygiene as it were is that an insight that you would endorse
59:58
SPEAKER_01: Yeah.
59:59
SPEAKER_01: I mean,
59:59
SPEAKER_01: there's a funny thing actually within the field where to explain why certain people
1:00:03
SPEAKER_01: are susceptible to misinformation,
1:00:04
SPEAKER_01: they'll appeal to things like groupthink and cognitive biases and identity
1:00:08
SPEAKER_01: protective cognition,
1:00:09
SPEAKER_01: all of these things.
1:00:11
SPEAKER_01: But there's never really an appreciation that that might also influence the
1:00:15
SPEAKER_01: research among the people doing the relevant research.
1:00:19
SPEAKER_01: So it's like...
1:00:20
SPEAKER_00: But this is part of the broader problem I'm trying to get at,
1:00:22
SPEAKER_00: which is like I think there's still some chauvinism where – when I was 20,
1:00:27
SPEAKER_00: I'm sure I thought liberals are scientific and rational and logical.
1:00:31
SPEAKER_00: Conservatives are mired in groupthink and dogma.
1:00:36
SPEAKER_00: I get the sense it's not that severe.
1:00:38
SPEAKER_00: They're not angry 20-year-olds.
1:00:40
SPEAKER_00: They're more sophisticated than that.
1:00:41
SPEAKER_00: But I think some misinformation researchers might – could be accused of believing a
1:00:46
SPEAKER_00: version of that still.
1:00:48
SPEAKER_01: I'll also add to this as another bit of nuance that goes back to what we talked about at the beginning.
1:00:57
SPEAKER_01: What is misinformation?
1:00:59
SPEAKER_01: If you're focusing on really clear-cut,
1:01:03
SPEAKER_01: blatant falsehoods,
1:01:05
SPEAKER_01: fake news and so on,
1:01:07
SPEAKER_01: then I do think that kind of content disproportionately exists on the right.
1:01:11
SPEAKER_01: It's not exclusive to the right.
1:01:13
SPEAKER_01: By any means, you find it on the left.
1:01:15
SPEAKER_01: At least my own opinion would be,
1:01:16
SPEAKER_01: in America at least,
1:01:17
SPEAKER_01: you find that kind of content within the right-wing media ecosystem more than the left.
1:01:22
SPEAKER_01: The thing is,
1:01:23
SPEAKER_01: if you look at,
1:01:23
SPEAKER_01: say,
1:01:23
SPEAKER_01: the New York Times,
1:01:25
SPEAKER_01: the New York Times is never,
1:01:26
SPEAKER_01: almost never,
1:01:27
SPEAKER_01: I shouldn't say never,
1:01:28
SPEAKER_01: very rarely going to publish fake news in the technical sense of that term.
1:01:32
SPEAKER_01: It's going to be factual.
1:01:33
SPEAKER_01: They're not just going to make things up.
1:01:35
SPEAKER_01: Does that mean that an outlet like the New York Times is objective,
1:01:39
SPEAKER_01: unbiased,
1:01:40
SPEAKER_01: leading its audience to an accurate perception of various complex issues?
1:01:44
SPEAKER_01: Of course not.
1:01:45
SPEAKER_01: It's just that the way in which
1:01:47
SPEAKER_01: narrative creation and misleading ideas arise within liberal establishment
1:01:55
SPEAKER_01: institutions with more highly educated people is totally different from what
1:01:59
SPEAKER_01: happens if you're looking at like info wars and low quality websites online.
1:02:04
SPEAKER_01: So it gets back to this broader issue of what are we talking about when we're
1:02:07
SPEAKER_01: talking about misinformation?
1:02:08
SPEAKER_01: If we're talking about misleading communication, I would argue there's a lot of misleading communication
1:02:14
SPEAKER_01: within liberal media, within science.
1:02:17
SPEAKER_01: We've already talked about misinformation research,
1:02:20
SPEAKER_01: but it's not going to qualify as quote-unquote misinformation if you understand
1:02:23
SPEAKER_01: misinformation very,
1:02:24
SPEAKER_01: very narrowly.
1:02:25
SPEAKER_00: I'm just thinking of specific controversies I've written about where someone claims X is true.
1:02:31
SPEAKER_00: If you read the abstract of one study, X appears to be true.
1:02:35
SPEAKER_00: There's three other studies over here, more rigorous studies, showing X isn't true.
1:02:40
SPEAKER_00: It is very hard to squeeze a situation like that into the
1:02:44
SPEAKER_00: misinformation box because like it requires a lot of
1:02:50
SPEAKER_00: other knowledge about how science is produced and why you should trust these three studies,
1:02:55
SPEAKER_00: but not this one study.
1:02:56
SPEAKER_00: And it's a mess.
1:02:59
SPEAKER_00: All of which is why I think I share your skepticism that anyone can like easily and quickly.
1:03:04
SPEAKER_00: I mean,
1:03:04
SPEAKER_00: this is one of your most important points is there's not going to be a simple
1:03:09
SPEAKER_00: acronym like to pick.
1:03:10
SPEAKER_00: There's not going to be some quick fix for figuring out how to identify
1:03:14
SPEAKER_00: misinformation in the vast majority of the interesting cases that don't involve
1:03:20
SPEAKER_00: Marjorie Taylor Greene and Jewish space lasers,
1:03:22
SPEAKER_00: there's not really a shortcut to figuring out what's true and what's not.
1:03:25
SPEAKER_00: And I think a lot of people are peddling shortcuts right now.
1:03:31
SPEAKER_00: Last thing is right as we were preparing this interview,
1:03:33
SPEAKER_00: a big Atlantic article went up that I thought was right in our wheelhouse.
1:03:37
SPEAKER_00: This was by Charlie Warzel,
1:03:39
SPEAKER_00: who I think is one of the more smart and sophisticated journalists worrying about misinformation.
1:03:47
SPEAKER_00: Headline of this Atlantic piece, I'm running out of ways to explain how bad this is.
1:03:50
SPEAKER_00: Subhead, what's happening in America today is something darker than a misinformation crisis.
1:03:54
SPEAKER_00: Let me just read the top of this because it is pretty jarring.
1:04:00
SPEAKER_00: The truth is,
1:04:00
SPEAKER_00: it's getting harder to describe the extent to which a meaningful percentage of
1:04:03
SPEAKER_00: Americans have dissociated from reality.
1:04:06
SPEAKER_00: As Hurricane Milton churned across the Gulf of Mexico last night,
1:04:09
SPEAKER_00: I saw an onslaught of outright conspiracy theorizing and utter nonsense racking up
1:04:13
SPEAKER_00: millions of views across the Internet.
1:04:14
SPEAKER_00: The post would be laughable if they weren't taken by many people as gospel.
1:04:18
SPEAKER_00: Among them,
1:04:19
SPEAKER_00: Infowars' Alex Jones,
1:04:20
SPEAKER_00: who claimed that Hurricanes Milton and Helene were weather weapons unleashed on the
1:04:24
SPEAKER_00: East Coast by the U.S.
1:04:24
SPEAKER_00: government.
1:04:25
SPEAKER_00: and truth-seeker accounts on X that posted photos of condensation trails in the sky
1:04:29
SPEAKER_00: to baselessly allege that the government was spraying Florida ahead of Hurricane
1:04:34
SPEAKER_00: Milton in order to ensure maximum rainfall,
1:04:36
SPEAKER_00: just like they did in Asheville.
1:04:38
SPEAKER_00: Um, so this stuff hits close to home for, uh, Katie, for my co-host.
1:04:42
SPEAKER_00: She's from Asheville and, um,
1:04:44
SPEAKER_00: A chunk of her region was destroyed by this.
1:04:46
SPEAKER_00: She talked on this podcast with a couple of guests about how bad the rumor mongering has gotten.
1:04:51
SPEAKER_00: When I read that paragraph, I feel what Marcel is arguing because I've been on Twitter too.
1:04:56
SPEAKER_00: The rumor mongering is out of control.
1:05:00
SPEAKER_00: Isn't this a compelling case that we should worry a great deal about misinformation
1:05:03
SPEAKER_00: in the context of these natural disasters?
1:05:05
SPEAKER_00: Well, yeah.
1:05:06
SPEAKER_01: I mean, I think there's lots to say about that article.
1:05:08
SPEAKER_01: I think we should worry about misinformation and unfounded rumors and so on.
1:05:13
SPEAKER_01: But I think what I find frustrating about the article is a few things.
1:05:17
SPEAKER_01: First of all, it quite literally couldn't be more alarmist in terms of the framing.
1:05:22
SPEAKER_01: I'm running out of ways to describe how bad things are.
1:05:27
SPEAKER_01: It's maximally...
1:05:28
SPEAKER_01: And I think it's important to distinguish the claim that there are genuine problems
1:05:33
SPEAKER_01: here and we should be concerned with these problems from the maximally alarmist
1:05:38
SPEAKER_01: takes that you often get about these problems,
1:05:41
SPEAKER_01: because it's important to have a sense of how prevalent and how impactful the
1:05:45
SPEAKER_01: random rumors are.
1:05:48
SPEAKER_01: The second thing I'll say is,
1:05:49
SPEAKER_00: I mean,
1:05:49
SPEAKER_00: it's assumed- Well,
1:05:50
SPEAKER_00: actually,
1:05:51
SPEAKER_00: let's pause on that for a sec,
1:05:52
SPEAKER_00: because-
1:05:54
SPEAKER_00: It sounds like you're saying this stuff isn't as prevalent as the picture Warzel is painting.
1:06:00
SPEAKER_01: Well...
1:06:01
SPEAKER_01: To be clear, I don't have any special expertise about this specific situation.
1:06:05
SPEAKER_01: I would say reading that piece itself,
1:06:08
SPEAKER_01: there's no effort really to establish that what's happening now in terms of rumor
1:06:11
SPEAKER_01: mongering is worse than things have been in the past,
1:06:15
SPEAKER_01: when you often get rumors circulating among communities in the aftermath of earthquakes,
1:06:22
SPEAKER_01: natural disasters,
1:06:23
SPEAKER_01: and so on.
1:06:24
SPEAKER_01: It might be worse, but from the piece, we've got no reason to believe that it's worse.
1:06:27
SPEAKER_01: I also think in the piece,
1:06:29
SPEAKER_01: there's a real emphasis on social media,
1:06:32
SPEAKER_01: the way it enables people to inhabit these bespoke realities.
1:06:36
SPEAKER_01: And again,
1:06:37
SPEAKER_01: because social media is such a common form of communication today,
1:06:42
SPEAKER_01: any ideas that are circulating,
1:06:44
SPEAKER_01: some of them are going to be on social media,
1:06:46
SPEAKER_01: but it doesn't necessarily mean that social media itself is having this really
1:06:49
SPEAKER_01: impactful role.
1:06:50
SPEAKER_01: And there's no real attempt to establish that in the piece.
1:06:53
SPEAKER_01: And it fits into this template where
1:06:56
SPEAKER_01: First of all, Alex Jones is saying something.
1:06:59
SPEAKER_01: He's been saying insane things for decades.
1:07:02
SPEAKER_01: Some of the things it links to in the piece,
1:07:04
SPEAKER_01: like there's one post by a user on X which has five likes,
1:07:09
SPEAKER_01: and most of the comments were just telling the person what an idiot they are.
1:07:13
SPEAKER_01: It takes these big numbers,
1:07:15
SPEAKER_01: like this particular post was viewed millions of times or whatever it might be,
1:07:19
SPEAKER_01: which is often what people do in this space.
1:07:21
SPEAKER_01: But big numbers are the rule on the internet.
1:07:25
SPEAKER_01: And you can't assume just because lots of people happen to scroll past something
1:07:28
SPEAKER_01: that that means it was particularly impactful.
1:07:32
SPEAKER_01: So none of this is to say that there's nothing to worry about.
1:07:35
SPEAKER_01: But I think the thing that really bothered me about the piece was the analysis
1:07:38
SPEAKER_01: which is given of it,
1:07:40
SPEAKER_01: which is,
1:07:41
SPEAKER_01: in effect,
1:07:42
SPEAKER_01: at the end,
1:07:43
SPEAKER_01: the claim that there's a division in America between those people who basically
1:07:48
SPEAKER_01: want to live in reality and want to form accurate beliefs about the world,
1:07:52
SPEAKER_01: and those people who don't,
1:07:53
SPEAKER_01: who are sort of,
1:07:53
SPEAKER_01: doesn't use this term,
1:07:54
SPEAKER_01: but it's the general vibe,
1:07:56
SPEAKER_01: post-truth.
1:07:57
SPEAKER_00: Americans are divided not just by political beliefs,
1:07:59
SPEAKER_00: but by whether they believe in a shared reality or desire one at all,
1:08:03
SPEAKER_00: which,
1:08:03
SPEAKER_00: as you're saying,
1:08:03
SPEAKER_00: suggests there's folks who don't want to live in a shared reality.
1:08:06
SPEAKER_01: Right.
1:08:06
SPEAKER_01: And what bothers me about that is it's very partisan, right?
1:08:12
SPEAKER_01: There are the good people,
1:08:13
SPEAKER_01: there are the bad people who don't want to believe in reality,
1:08:15
SPEAKER_01: and they're overwhelmingly depicted as being on the right in this piece,
1:08:18
SPEAKER_01: which I think is very typical of these kinds of accounts.
1:08:21
SPEAKER_01: But it's also just such a shallow analysis, I think.
1:08:24
SPEAKER_01: So there are deep problems in America.
1:08:28
SPEAKER_01: And when it comes to this kind of rumor mongering,
1:08:30
SPEAKER_01: which I think is prevalent among segments of the right.
1:08:33
SPEAKER_01: But a lot of that has to do with factors like intense political polarization, institutional distrust.
1:08:39
SPEAKER_01: the way in which those two things interact in particular kinds of cases.
1:08:43
SPEAKER_01: This sort of analysis where,
1:08:45
SPEAKER_01: okay,
1:08:45
SPEAKER_01: there are the people who want to live in reality and the people that don't,
1:08:48
SPEAKER_01: I don't think that provides a sophisticated or nuanced way of trying to really
1:08:51
SPEAKER_01: engage with that.
1:08:53
SPEAKER_01: And I actually think denigrating and demonizing huge statements of America as just
1:08:58
SPEAKER_01: not even being concerned with reality at all is the sort of thing which is likely
1:09:02
SPEAKER_01: to exacerbate the problem rather than help it.
1:09:05
SPEAKER_01: I will say, as I've already said earlier on, motivated reasoning, that is a thing.
1:09:11
SPEAKER_01: That's not unique to the right,
1:09:12
SPEAKER_01: but this idea that sometimes people prioritize goals over truth,
1:09:16
SPEAKER_01: that does occur.
1:09:17
SPEAKER_01: I just don't think it's as extreme as depicted in this piece,
1:09:21
SPEAKER_01: and it's not concentrated among only one segment of the population.
1:09:26
SPEAKER_01: I would also say one aspect of the piece that I do agree with is pointing to the
1:09:30
SPEAKER_01: behavior of elites like Elon Musk,
1:09:33
SPEAKER_01: Donald Trump.
1:09:35
SPEAKER_01: They behave in ways that are contemptible.
1:09:38
SPEAKER_01: I think in Jonathan Rauch's book, he's got this term, epistemic sociopaths.
1:09:44
SPEAKER_01: And I think that's like a nice characterization of the way in which these elites
1:09:48
SPEAKER_01: often behave,
1:09:48
SPEAKER_01: which is they've just got no interest at all in being thoughtful and reasonable
1:09:53
SPEAKER_01: when they're communicating.
1:09:54
SPEAKER_01: And that's a really big problem.
1:09:56
SPEAKER_01: It's not clear to me that that's best analyzed through the lens of social media,
1:10:00
SPEAKER_01: that you've got these just assholes on the right who just
1:10:03
SPEAKER_01: churn out lots of bad content.
1:10:06
SPEAKER_01: But that is a real problem, so I agree with that aspect.
1:10:08
SPEAKER_01: But considered as a whole,
1:10:10
SPEAKER_01: I think it fits into this somewhat smug,
1:10:13
SPEAKER_01: liberal,
1:10:13
SPEAKER_01: alarmist narrative supported by anecdotes and cherry-picked quotes on social media
1:10:21
SPEAKER_01: and pointing to large numbers of views and posts without really contextualizing them.
1:10:28
SPEAKER_00: Just to end on a bright note that ties into your critique of the War cell piece,
1:10:33
SPEAKER_00: there's some evidence suggesting that even in today's supposedly dark age of misinformation,
1:10:39
SPEAKER_00: the vast majority of people consume mainstream news,
1:10:42
SPEAKER_00: right?
1:10:44
SPEAKER_00: Even the biggest folks like Alex Jones,
1:10:46
SPEAKER_00: they have millions of listeners,
1:10:47
SPEAKER_00: which is lamentable,
1:10:48
SPEAKER_00: but they do not represent the median consumer of news.
1:10:51
SPEAKER_01: Yeah, although part of that is because average people who aren't very politically engaged-
1:10:58
SPEAKER_01: They don't really follow the news closely at all.
1:11:00
SPEAKER_01: But you're right,
1:11:01
SPEAKER_01: in terms of most Americans,
1:11:02
SPEAKER_01: my understanding of the empirical literature is they tend to have pretty mainstream
1:11:07
SPEAKER_01: and not particularly skewed media diets.
1:11:09
SPEAKER_01: But I would say...
1:11:11
SPEAKER_01: When it comes to the segment of the population in America who is highly politically
1:11:15
SPEAKER_01: engaged or that is highly politically engaged,
1:11:19
SPEAKER_01: there I think you do see much more fragmentation in media in the sense that people
1:11:24
SPEAKER_01: on the right do tend to consume political content from sources that are very
1:11:29
SPEAKER_01: different from those people on the political left when it comes to the highly
1:11:32
SPEAKER_01: politically engaged segment of the population.
1:11:35
SPEAKER_00: Anything else you wanted to hit on before we wrap this up?
1:11:38
SPEAKER_01: She knows everything.
1:11:39
SPEAKER_00: Okay.
1:11:39
SPEAKER_00: Remind everyone where they can find your work.
1:11:41
SPEAKER_01: Conspicuous Cognition.
1:11:42
SPEAKER_01: I write weekly blog posts about these issues and social epistemology and things like that.
1:11:48
SPEAKER_00: Okay.
1:11:48
SPEAKER_00: Are you on Twitter?
1:11:50
SPEAKER_00: Spreading misinformation?
1:11:51
SPEAKER_00: I am on Twitter.
1:11:51
SPEAKER_01: Yeah.
1:11:52
SPEAKER_00: What's your username?
1:11:53
SPEAKER_01: Dan Williams.
1:11:54
SPEAKER_01: Dan Williams Phil.
1:11:55
SPEAKER_00: Thank you so much for joining us, Dan.
1:11:57
SPEAKER_00: This was a great conversation.
1:11:58
SPEAKER_01: Thank you.
1:11:59
SPEAKER_01: Cheers.
1:11:59
SPEAKER_00: This has been Blocked and Reported.
1:12:01
SPEAKER_00: As always, we're produced with help from Jessica, the 80s baby.
1:12:03
SPEAKER_00: Thank you for listening.
1:12:05
SPEAKER_00: I'm Jesse Single.
1:12:05
SPEAKER_00: And remember, it's not misinformation if I agree with it.
1:12:15
Thank you.