小蓝视频

0:07 聽聽聽聽 Kay Summers: From the School of International Service at 小蓝视频 in Washington, this is Big World, where we talk about something in the world that truly matters. What's going on right now? How do you know? Did you see it on social media? Did it have a link that looked legit so you didn't bother clicking on it and just took the post at face value? Were you angry or frustrated by what you read? What did you use the information for? Did it help you decide who to vote for or whether to get vaccinated or who should go to prison for activities that are clearly corrupt? Just read the post.

0:40 聽聽聽聽 KS: And then, suppose you found out it wasn't true. Did you believe that or the original stuff you read? How are you supposed to know what to trust? Today, we're talking about disinformation. I'm Kay Summers and I'm joined by Sumitra Badrinathan. Sumitra is a professor here at the School of International Service, and she focuses on political communication in South Asia with an emphasis on new platforms like WhatsApp and their effects on political disinformation, media trust, and the quality of democracy. Sumitra, thanks for joining Big World.

1:15 聽聽聽聽 Sumitra Badrinathan: Thank you so much for having me, Kay.

1:17 聽聽聽聽 KS: So, just to get us all level set of, for people listening, Sumitra, what is your definition of disinformation?

1:25 聽聽聽聽 SB: It's a good question to start. So, I am actually going to start by telling you what the literature defines as disinformation, because mine derives from there. So, the literature normally identifies three components of what makes a story a false story. And the first one is low levels of veracity. This actually has to be factually untrue. The second is it should be presented in a journalistic format. And then, if there is the intention to deceive, it becomes disinformation as opposed to misinformation.

1:58 聽聽聽聽 SB: In your introduction, you very nicely pointed out that I work n South Asia. And in South Asia, I'm sure we'll talk about this a little more later, information is largely spread via WhatsApp, and much of it is in the form of text messages with the content copied and pasted into the body of the message and such. So, it really cannot mimic legitimate standalone news websites because it isn't presented in a journalistic format.

2:25 聽聽聽聽 SB: And then, while the creation of lies in this context can actually stem from organized attempts, sometimes you can just receive one of these messages as a forward from your uncle who isn't trying to deceive you. So in that case, is it really disinformation? It's a bit of a tricky question to answer.

2:42 聽聽聽聽 SB: So, I use the term misinformation for things that are just factually incorrect. They're rarely presented in journalistic formats. So, as long as they're factually not true, whoever they're coming from, because it's hard to pinpoint intent, I prefer using the term misinformation for the context I work in.

3:01 聽聽聽聽 KS: Okay. All right. Good to know. So, looking at disinformation broadly, with the intent behind it, there are different kinds of disinformation including health information, as seen during the COVID-19 pandemic. And political disinformation, as seen in elections around the world. The only thing different varieties of disinformation seem to have in common is a goal to use digital media to broadly promote things that aren't true. But how do these different kinds of disinformation pose threats to society?

3:38 聽聽聽聽 SB: Absolutely. So, health is the big one right now, correct? Because we're still kind of going through the ramifications of the pandemic. Medical misinformation is actually pretty big in India where I work, where most of my research is based on. This can take the form of promoting alternative medicines instead of science. And inadvertently leads to low trust in authority sources and could lead to less vaccination across the globe. And that's something we've been seeing for a while.

4:09 聽聽聽聽 SB: Another type of misinformation is conspiracy theories. These, in my context, can exist in the form of sometimes scapegoating minorities. For example, during COVID, actually, there was a good intersection between conspiracy theories and medical/health misinformation. One example is that in some villages in India, Muslim fruit and vegetable vendors were being accused of spitting on their fruits and vegetables in order to spread COVID intentionally to the majority Hindu population.

4:44 聽聽聽聽 SB: It made an already complicated communal situation even more polarized and fraught. And many people got injured because of it. And so, conspiracy theories like this could sometimes even affect violence and worsen outcomes that already exist. In many cases in India, this kind of misinformation has actually directly been linked to mob violence and even small-scale riots. So that's, I think, for me, another big category of misinformation.

5:13 聽聽聽聽 SB: And I think the last one is political. We talk about ideological misinformation a lot in every context. My party or leader is superior because of X, Y, Z reasons. But it could also be I, as a politician, built a bridge that's five villages away that you can't see, but here is a photo that's actually fake, but you believe it and it leads to you voting for me. So, that's the other type of political misinformation that I think is quite specific to the Indian context. Because votes are oftentimes based off of tangible goods and service provision, especially in rural areas.

5:50 聽聽聽聽 SB: So in theory, you're right, all of this could worsen democratic outcomes. I think the challenge for researchers is to sometimes be able to causally identify whether it's the information that's actually causing these outcomes like violence or polarization or whether it's other factors. And for a bunch of reasons, this is really hard to do. So, definitive evidence on what misinformation actually worsens is quite hard to pin down. That said, we all know and that's why we're here, it's a normatively very bad problem.

6:20 聽聽聽聽 KS: Right. And you mentioned India and, Sumitra, you have looked most closely at the spread of misinformation in India, which is now the world's most populous country designation just a few months ago. How does India's size impact the spread of misinformation?

6:37 聽聽聽聽 SB: Yeah. I'll also say, actually, that we're not just the world's most popular country. For the first time in the almost 30 years of my existence, we are not labeled a democracy anymore. I wonder if that has anything to do with the spread of misinformation. But to answer your question, I'll say less about the size and more about, I think, some of the population features that make India quite unique.

7:00 聽聽聽聽 SB: And I think the first of those is the share of formal education and literacy in the country is quite low compared to developed countries that we normally see misinformation research on. Now, that's not to say that less literacy leads to more vulnerability to misinformation. In fact, there's actually quite a lot of evidence to the contrary, that more educated people are more entrenched in their views. But what this could lead to is that there's a lot of people in India who are actually not connected to the internet even now.

7:34 聽聽聽聽 SB: That's changing very quickly. But you have this cohort of people that have just been digitally connected to the grid. And so, they're unfamiliar with the medium of the internet. They also have less digital literacy, separate from actual educational literacy, or familiarity with what the internet and the social media apps entail. And also, the ways in which disinformation could spread and even whether it exists.

8:01 聽聽聽聽 SB: So, in interviews I've talked to people who sometimes conflate the internet with, say, Facebook, because that's the one app they have on their phone. So, there are a lot of studies actually showing that low digital literacy definitely leads to more vulnerability to misinformation. So, it's just a different population in terms of the uptake of some of these tools and technologies that we're seeing in the rest of the world.

8:25 聽聽聽聽 SB: So, as time goes by, I think my hunch would be that India in these ways would start to resemble many developed countries. But for now, that's not the case. And so, there are these unique factors that might make people more vulnerable and could impact the spread of disinformation just because so many people are just new to the internet and some are still not connected.

8:47 聽聽聽聽 KS: And because people's experience of the internet has become so personalized, I think it's easy to forget that not everyone sees what I see on my feeds, and not everyone uses the same platforms I use. You mentioned WhatsApp earlier. What are the major social media platforms people in India use, WhatsApp and others perhaps, and how do they factor into this spread of disinformation?

9:14 聽聽聽聽 SB: It's a great question. A good place to talk about WhatsApp, which has been both the boon and the bane of my existence a good part of the past six years. Okay. So, WhatsApp is the main app in India. India is its largest market in the world. And if you ever used it, it's actually super easy to use. It pulls a very small quantity of data to send photos and videos in large files. So, even in remote places with bad internet connections, it's actually the one app that works seamlessly. So, it's not a surprise that the uptake of WhatsApp is really big in India.

9:49 聽聽聽聽 SB: But one of the things about WhatsApp is that it's an encrypted messaging app, so it's not a public social media platform like Facebook or Twitter. So, what this means is that it's sort of like email where only the sender and the receiver of a text can see what's in the text, right? No one outside of those two people can see, read or analyze what's on it. This really limits what you can actually do to counter misinformation.

10:17 聽聽聽聽 SB: So, WhatsApp is owned by Meta, which owns Facebook. And sometimes, you'll see... You might've seen this on your feed too, Facebook will - content by adding a label. These are now either false or partially false below a post. You can't do that for WhatsApp. You can't lobby tech companies to take down problematic content, however problematic they are, because nobody, including these companies, can actually get into your chats and see what's there.

10:44 聽聽聽聽 SB: There's some evidence now that companies have Metadata about the number of texts you've sent and things like that and who you've sent them. But the actual content of messages is pretty much invisible except to the people who are in that chat. So, that's a problem for misinformation because everything that's on there is virtually unchecked and we're talking about the biggest market for WhatsApp in the world, nearing 600 million people. And this is a huge obstacle to counter the spread of falsehoods.

11:16 聽聽聽聽 SB: So, the most promising solutions that have existed in the W and that are being implemented in many countries are just not applicable in this context. So -, correcting, putting warning labels and tags and so on and so forth, we just can't do those. So I guess the only solutions, in this case, are to help people become better consumers of information ex-ante. So, by teaching them tools that might make them critically think better about information they encounter.

11:45 聽聽聽聽 SB: I'm currently involved in a project to teach a misinformation curriculum to about 15,000 schoolchildren in one of India's poorest states. Sounds really good, but this sort of initiative is really expensive, requires a lot of coordination between multiple government and non-governmental actors, and that also makes it not so easy to scale up.

12:08 聽聽聽聽 SB: So, coming back to your initial question, WhatsApp is the major social media platform. It has features that are so unique to it that they directly impact the spread of misinformation to the point where we actually don't know how much misinformation is even out there. We don't know the scale of the problem. And the only solutions we have to counter it are expensive and hard to implement.

12:31 聽聽聽聽 KS: e. We talked about trusted sources and trying to develop that type of digital literacy, but in a platform where things are coming from people that you know and theoretically trust or have some sort of relationship with, it's hard to tell someone, "Don't trust your mom."

12:47 聽聽聽聽 SB: Absolutely. Or, "Don't trust your doctor." When your doctor could be sending you things that are sometimes not necessarily fully true. And that also taps into one of your earlier questions about, what is it about India? Does the size of the country or things about it make it different? And I think one of the things about it, which I didn't mention earlier, is that it is a developing country so it's a state of low capacity, low state capacity.

13:13 聽聽聽聽 SB: Which means that sometimes we have to rely on informal sources to get news and information, even if we're fantastic consumers of politics and the news, because official sources just sometimes don't exist in remote areas or don't provide us what we need. So, we're forced to tap into our informal family and other networks. And even if those people mean really well, you can sometimes get information that's completely incorrect.

13:46 聽聽聽聽 KS: It's time to take five. This is when you, our guest, get to daydream out loud and reorder the world as you'd like it to be by single-handedly instituting five policies or practices that would change the world for the better. On the practices front, what are five ways that researchers and academics have worked to counter misinformation?

14:05 聽聽聽聽 SB: Excellent question because I get a chance to now highlight some of my favorite research in this field. So, the first would be there is a new recent paper that looks at the Ivory Coast with a quite unique study that finds that actually targeting social identity and polarization through techniques that, say, foster empathy about the other side, that works to counter misinformation as opposed to some of the usual techniques that we've talked about. It's pretty unique and really cool.

14:33 聽聽聽聽 SB: The second. South America and Brazil researchers and colleagues and friends of mine actually ran an experiment where they encouraged people to deactivate WhatsApp before their recent election and found that that significantly reduced their exposure to false news, which is also pretty cool. Then, there's also podcasts that speak about news literacy and media literacy that have been used to counter misinformation. I've seen two recent studies that send people either podcasts or news digests or audio recordings straight to their phone over a sustained period of time and find that that actually works to help people.

15:12 聽聽聽聽 SB: I think we're at four. The fourth, I think, is digital and media literacy interventions to help people learn about misinformation and foster critical thinking skills. These can take the form of school-based lessons. They can be one-shot training sessions. There's a bunch of studies and a bunch of different countries that have tried their best to work at this.

15:32 聽聽聽聽 SB: And finally, this is boring but important, fact checking itself. Study after study, and country after country shows that fact checks help correct misinformation across context and across a variety of formats. Whether they're delivered by text on WhatsApp or whether they're delivered otherwise, they're just a tried and tested tool that we, as academics, civil society, practitioners, shouldn't give up on whatever else we try to combat misinformation.

16:03 聽聽聽聽 KS: Thank you. And I am going to take away from that that one of the things that's helping save the world are podcasts. Thank you.

16:13 聽聽聽聽 KS: Sumitra, despite having the world's sixth largest economy, as you mentioned, India is classified as a developing country. Is there a correlation between developing countries and a higher rate of misinformation spread?

16:25 聽聽聽聽 SB: So Kay, I think this is a fantastic question because it really got me thinking a lot about how to answer it. And I realized, in the process of thinking about it, that to answer this question, we actually need an accurate estimate of the rate of disinformation spread in every country, which turns out is really hard to do. And the main reason is one that we've already discussed, which is that actually most of the world is on apps where you cannot detect spread like WhatsApp. And then, to calculate spillover effects from social to non-social or offline is practically impossible.

17:01 聽聽聽聽 SB: So long story short, the data doesn't exist to answer this question. But if we did have the data and if I had to take a bet on it, I think, if anything, we might find a negative correlation. That is more developed countries, I think, would likely have populations that are more vulnerable to misinformation and not less. Let me see if I can coherently explain why. I think it owes to the main reason why people believe misinformation to begin with.

17:31 聽聽聽聽 SB: There's a bunch of reasons, but the primary one is that, as humans, scholarship has shown that we have a strong preference for information that confirms what we already know rather than the truth. So, we tend to seek out information that reinforces our preferences. A phenomenon that the literature calls motivated reasoning. And this matters because there's research findings on misinformation that repeatedly show that more politically sophisticated individuals, that's people who have higher levels of political knowledge and education, are actually more likely to resist corrections to misinformation.And these people are also the least amenable to updating their beliefs when misinformation already supports their existing worldviews.

18:18 聽聽聽聽 SB: So, we know that developed countries have higher levels of education. We have this finding in the literature that more education doesn't necessarily protect you from misinformation. It might make you worse off. On the contrary, I've actually now run a few studies in India where we provide corrective information to respondents and it appears as though they are consistently willing to change their prior beliefs when they receive fact checks or corrections. So, miracle data doesn't exist, but if it did, maybe this is the direction in which it goes. The more education you get, the more resistant you are. So, developed countries are actually worse off.

18:55 聽聽聽聽 KS: That is so interesting, and it makes n awkward kind of sense that the more education we receive, the smarter we think we are and the less we feel like we need anyone to tell us anything or more apt to say, "Well, I found this one study that supports what I think. And so therefore, it must be true." And I guess it just is a reminder to people in developed countries to try and remain open to arguments against what they think and to at least examine those closely rather than shutting it down.

19:30 聽聽聽聽 KS: And Sumitra, you and I are talking about India a lot today. But to be clear, disinformation, and misinformation is happening pretty much everywhere. It's certainly prevalent in the US. And in the US, much of the disinformation, particularly related to our elections and to the COVID pandemic, has been driven by far-right interests in the US. And I'm wondering, is there a similar spread of disinformation or misinformation among far-right political parties in India as there is in the US? And how do political parties in India utilize these types of campaigns?

20:08 聽聽聽聽 SB: Yeah. Great question. I think partisan sources are the original and main ground zero for the creation of falsehoods. So, a very pertinent question. I don't think that India's parties are actually aligned on the left/right spectrum that we typically use to describe party systems in the W. So, we don't quite have the far right and the far left.

20:33 聽聽聽聽 SB: That said, the internet has spread so rapidly in the past decade or so, and it has actually led to election campaigning to be conducted over the internet to an extent that we hadn't quite seen before 2014 in India. And WhatsApp, chat-based applications like WhatsApp became primary communication tools for parties. There is one party, though, that does this more efficiently than others, and that is the BJP, which is the current ruling party... Hindu nationalist ruling party in India.

21:07 聽聽聽聽 SB: During the 2019 elections, they made plans to put voters, small geographical units of voters, into WhatsApp chat groups based on their location. So, they would classify voters by preexisting socioeconomic characteristics like religion and put them into groups based off of those characteristics, and then send information, propaganda, campaigning tools to those groups based on those characteristics.

21:36 聽聽聽聽 SB: And so, a WhatsApp group, at the time, could contain 256 people. And so, when they made those groups for every one of India's polling boots, which is our smallest administrative unit, that communication strategy potentially reached seven to 800 million voters. And while such campaigning is used to share information, I've been a part of these groups for research purposes in the past, oftentimes it can contain propaganda and disinformation.

22:05 聽聽聽聽 SB: In fact, there was one recent study actually trying to quantify the amount of misinformation in these private groups, and confirmed that it was the BJP that used the strategy more often than not as a very efficient propaganda tool, both during elections and outside of them. NPR has done some fantastic reporting on the ecosystem of WhatsApp groups and talks about them being in concentric circles where the innermost, the smallest circle, is party members who decide what content to push that day. The middle tier is party workers, so people who may or may not necessarily be on party payrolls, but who are tasked with the job of pasting these messages into group chats.

22:50 聽聽聽聽 SB: But the outermost circle, and that's where it gets tricky, is people who have absolutely nothing to do with the parties officially, but who are just fans who just care about the cause. And then, when they get these messages, believe they're true and pass them on, it may or may not be disinformation any longer, but the rate at which it spreads beyond that initial circle is quite exponential and pretty dangerous because, as we discussed on WhatsApp, it just can be checked in any way.

23:17 聽聽聽聽 KS: And Sumitra, I know that not everything has an analogy to the US, but I'm wondering if there is any analogy to what happened in the US in our 2016 presidential election when we know that Russia absolutely interfered through disinformation on behalf of Donald Trump. Does the spread and use of malicious types of disinformation in India include foreign powers trying to influence elections? Has that happened?

23:46 聽聽聽聽 SB: From my understanding, I think the threat from false information in India is largely domestic. And you're right, in the US, the focus has been on foreign-backed misinformation campaigns. But in India, from my understanding, political disinformation appears to be largely domestically manufactured. I guess the way I look at it is that, as somebody who cares about democratic outcomes and my country, there's enough of a threat coming from within at this point that we just haven't figured out how to stop because it's a serious problem and it's really hard to solve.

24:17 聽聽聽聽 SB: So, maybe this changes. Maybe I come back in five years with a different answer to this question. But as of now, I think everything is internally and domestically manufactured.

24:28 聽聽聽聽 KS: The problem is at home. Okay,

24:29 聽聽聽聽 SB: At home, yes.

24:31 聽聽聽聽 KS: Sumitra Badrinathan, thank you for joining Big World to discuss disinformation and misinformation in India. It's been a pleasure to speak with you. I learned a lot from you.

24:41 聽聽聽聽 SB: Thank you so much for having me. It was wonderful to be on.

24:44 聽聽聽聽 KS: Big World is a production of the School of International Service at 小蓝视频. Our podcast is available on our website, on iTunes, Spotify, wherever else you listen to podcasts. If you leave us a good rating or a review, it'll be like a social media feed full of puppy videos. Our theme music is It Was Just Cold by Andrew Codeman. Until next time.