A look at News, Memes, and Wireless Tech from More than 100 Years Ago (Heidi Tworek calls from Germany)

You are listening to Policy Prompt from the Centre for International Governance Innovation. I'm Vass Bednar.

And I'm Paul Samson.

Our in-depth interviews find nuances in the conversation, with leading thinkers that work at the intersection of technology, society, and public policy.

Listen now wherever you find your podcasts.

Hey, Paul. What's happening?

Hey, not much.

Okay. I want to read you the text of a cartoon. You could say it's a meme, but it's from back in 1917. Is that cool with you?

Yeah, we're doing a podcast, though. People can't see what you're showing. So what is it?

I know, I know. So here's the caption, okay? "The lie is the law of the world. Reuters Cable Network teaches that." I don't know, I wondered if that language sounds familiar to you today.

Yeah, it does. It speaks to the fake news and distrust being a big issue a hundred years ago, and it is totally today for sure. And so, it's not new in that way. It may be easier to do now in a digital age, and you certainly don't have to look very far to find it.

Right. You definitely don't. I mean right now in the US, two major social networks are owned by people who will either lead or serve in or influence the next presidential administration. President Trump owns a 58.9% stake in Trump Media and Technology Group, which is the ultimate owner of Trump Social, his self-styled free speech social network. Of course, we've seen very close ties between him and Elon Musk, who controls Twitter or X. That's fascinating.

Yeah. It really came up in the very run up there to the election. A couple of newspapers chose not to endorse a presidential candidate this cycle, Washington Post and the LA Times and a couple of others. That was unusual and it raised a lot of questions about the roles of newspapers in politics, how they should endorse or not candidates and when and how. All these questions are circulating out there.

Right, and also how people view these institutions. Where and how people get their news is really, really changing, and that was also exemplified in that last election. More people getting news from podcasts, trusted influencers who may not be trained as journalists. So not only is where and how people get their news changing, but also the very idea of what news even is is really contested right now.

That risk of having powerful political or corporate interest influence the news is not a new worry. We've been here before. So maybe we can learn a little bit more about how to best meet this moment and understand it from a historian.

Yeah, and we will unpack further what actually happened. There was no major deep fake moment that people seemed to think was going to happen. There were a few small ones. Let's see if there was something that was hidden that comes out as time goes on.

So today we have Heidi Tworek, who's an associate professor of history and public policy at the University of British Columbia. She is a non-resident fellow at both the German Marshall Fund of the United States and the Canadian Global Affairs Institute and is project coordinator of the United Nations History Project.

Welcome, Heidi. How are you?

Heidi Tworek:

I'm great. How are you?

We're doing pretty well. We're both here in Waterloo, at the CIGI campus. Where did we find you today?

Heidi Tworek:

You found me in Bielefeld, Germany, at the Center for Interdisciplinary Research, where I am on a fellowship for a few months.

Wonderful. Thank you so much for joining us today.

It's great to see you, Heidi. We've never met in person, so this is a great start virtually here today.

Heidi, you describe the grasp that news agencies had over national and international news in the first half of the 20th century as being more powerful than the monopolies that Facebook and Google ... Sorry, Meta. I have so much trouble calling that company Meta, more powerful than the monopolies that Meta and Google hold over information provision now. What made that possible?

Heidi Tworek:

I love that we're diving straight into news agencies. It's delightfully nerdy and fantastic way to begin. So some of the reasons are that there are so few of them, and they are supplying basically all of the newspapers. So we don't exist in a multimedia environment like we have today where, yes, of course you have Meta and you have Google or Alphabet, but you still have TV, podcasts, radio, newspapers. There are many different ways that people inform themselves. So while those social media companies and platforms are clearly very important, they're not really a sole mechanism of news. For much of the first half of the 20th century, news agencies are really the main suppliers of news. They're the bottleneck behind newspapers and behind most radio news as well.

There were lots of things I learned in this book, and one of the things that really struck me was this idea of a technology race that was happening around telegraph systems, and particularly the line-based telegraphy undersea cables.

Speaker 4:

The cable is laid out by the bow. Through the anchor port holes, we see the cable slipping down into the water. Four miles out ...

Lead large largely by the British, and the idea that the Germans were advancing very quickly on wireless telegraphy. The question is, can you say a little bit more about that, but did the Germans ever have a chance of winning this technology race either before or during World War I?

Heidi Tworek:

Yeah. So I think what's really interesting about this longer span of history is that when we first get submarine cables, so there's one between Britain and France in the early 1850s, and then finally successful transatlantic submarine cable in 1866. So when that first happened, there isn't a lot of contestation about it. It's mostly laid around the world by British-American owned companies and other nations like the French, the Germans, et cetera, all using this system.

There's then an international telegraph union which starts to regulate these cables. But you don't see ... If you're reading things in the 1870s, 1880s, you don't see a lot of German officials feeling particularly worried about this system. It really starts to be in the 1890s onwards, as Germany starts to think of itself as a colonial power, wants to be a global power that officials and others start to think, "Hang on a moment, these cables are not neutral infrastructure actually. They can be used to censor content. They can be part of great power rivalry."

And so, Germans start to think about how can we maybe lay cables that are along exactly the same lines? Not because we need more capacity, but because this is actually part of politics. So these are not neutral cables in any way. At the same time, they also start to invest in this new technology of wireless, thinking that that's actually going to be a way to ... If we want to think about it as a competition or a race, that's going to be a way to win. This is not just to duplicate cables but actually find a way to bypass this system altogether.

Part of the reason they want to do that is because of war planning, a fear that if war does break out, one of the first things that the British will do is cut the cables. It turns out they're right. I mean usually in our stories of the outbreak of World War I, this part doesn't feature, but it's actually a huge part of what happens in the first bits of the war and actually throughout World War I, is lots of cable-cutting on both sides to try to disrupt communications.

This to me is I think a really important lesson for today for a couple of reasons, if I can already bring us a little bit into the connection to today. One is that this is not a history that's in the path. This is really the beginning of the story that leads us up to fiber optic cables, because the way in which we repair fiber optic cables is pretty much the same as we did in 1866. Many of the places where fiber optic submarine cables are laid are exactly the same places where we laid submarine cables in the 1860s onwards, because we laid telephone lines there and then we laid fiber optic cables there.

I think lots of the ways in which nation states think about these cables, them being part of war, surveying them, these are all things that we see today. Then the final thing is that they often get forgotten and the flash points emerge at moments of great tension. So we've seen that more recently. Suddenly we've remembered that this infrastructure exists because it's a moment of great tension between a whole host of nations. So it's both the path dependency that's, I think, really important and the way that these kind of conflicts keep recurring.

Then, finally, to answer your actual question, Paul, did the Germans think they were going to win this or could they have won it? I guess they didn't always think about it as winning before World War I. They were just thinking about needing some sort of alternative at all. In that, at least with wireless, they succeed just before World War I. They do set up a world wireless network, but it only exists for a few months before World War I breaks out.

Then I guess they win in the sense that the British and others take their system very seriously. They really believe this is a key part of World War I. Again, take a conventional course on World War I, you're not going to hear about this, but it's actually a massive part of how the British and others run this war, particularly outside of Europe. They're looking to find the islands where these wireless towers are. So that tells me that, at least in the context of the war, people were taking communications very seriously in a way that historians have forgotten in the hundred years since.

Heidi, out the gate, you shared that you were a little bit excited that we were talking about news agencies right away. I love you just went deep, literally deep in the water too, with cables and talking about the physical infrastructure.

I do want to point out that when you mention wireless, it's probably not wireless in the way that we're used to, thinking about maybe connecting with Wi-Fi that you're thinking. You are presenting it to us in terms of radio and some associated freedoms in terms of sending information that way.

What I wanted to just ask you is we started talking about news agencies without chatting about what those actually are. What is a news agency?

Heidi Tworek:

So news agencies emerge basically around the same time as telegraphy. So they're a new business form, with the idea that this is going to be a new way to collect news and send it out to a whole bunch of newspapers. So the easiest way to describe them is to think about them like a news wholesaler. So a couple of news agencies, they start to emerge, they station correspondents around the world to gather news, and then compile it and send it to newspapers who are their subscribers, who are like their retail customers. So that's really the easiest way to think about it.

From the gate, there are very few news agencies because there's really high fixed and sunk costs in station correspondents all around the world when you don't exactly know where news is going to happen. Initially, sending telegrams is very, very expensive. So what you end up with then is a very small number of news agencies. The first one is founded in Paris just after the 1848 revolutions. Then you get the first one in Britain, early 1850s, Reuters that still exist today, one founded in Berlin. The three founders know each other and they decide to even strengthen this system by creating a cartel.

So they say, "Listen, there's no point in all three of us all covering the world. We should get together. We should divide the world between us. We'll station our correspondents around the world. Then we'll swap the news because we all have different retail customers." So that's really the easiest way to think about it.

Wholesalers who then provide to newspapers, to their retail customers. The final part of this is that, of course, most people who are reading newspapers aren't necessarily aware of this system, and that's part of its power and that's why states get really interested in it, because it's a way of influencing news by influencing one firm and by the users not really knowing that you're doing it.

So you're describing a time when it was getting harder and harder to understand the source of the news because the world was divided up in a way. Now with so many additional sources and so many technologies, it feels like a lot of the sources are buried, sometimes intentionally. Is this getting worse in the current world, particularly with generative AI potentially thrown in there and synthetic data thrown into the mix that muddies the waters here? Do you think things are getting worse now in that space of what is the source of the actual news?

Heidi Tworek:

I guess I don't think about it in terms of worse, more that this is a problem that has existed for at least 150 years, that one of the ways in which states or even companies or powerful individuals want to influence news is by influencing at a point when people don't know what the source is. And so, I see the generative AI, synthetic AI part of this.

This is part of a much longer debate as to where you intervene and how you influence news, because you see various moments where, for example, rich people say, "I'll buy a newspaper, and then I'll dictate exactly what the article is," and then it backfires because everybody knows that you've exactly dictated it. So then they turn to what are other hidden ways that we can influence news? So I think what we see is the latest technological evolution in that.

Now the question is how's that going to fit into the overall media ecosystem and how is that going to influence questions of what we do and don't find reliable? So I think one fairly convincing argument is part of the reason that this is more concerning is that it's going to be harder to discern what is reliable and what isn't.

But I guess I'm just arguing this isn't at all a new problem and that I think that just helps us look at it in a slightly different way. It helps us not immediately fall into a huge panic about it, but rather think about what are some real tactics and strategies that we can employ to give a sense of what is reliable information? Maybe there are new techniques, for example, of transparency that we can implement today that weren't possible at the time that I was writing my book about.

So we've mentioned straight up fake news, kind of false elements, and then you're pointing to this spectrum perhaps of reliability, or the work that the reader does to try to discern reliability if they're able to take the time to do that. There were really great examples through your research in your book of what might be called fake news today, like the kaiser's abdication in 1918, which had widespread attention and impact.

What's different about today's fake news, but also maybe what's a little bit of the same? As you're saying to us, these aren't new problems. They're maybe presenting just in different ways.

Heidi Tworek:

Yeah. So I mean I think that the example of the kaiser's abdication, just to give people a little bit of background on it. So in the last days of World War I, it's clear that Germany's going to lose, but it's not clear exactly how it's going to happen. There's a new chancellor in Berlin, and he is really worried that there's going to be a communist uprising. There's going to be blood on the street. And so, he basically decides that the way he's going to forestall that is to say and send out through the main news agency that the kaiser has abdicated.

The Kaiser is not in Berlin. He's in a place called Spa that's quite close, but he gets news of this and he, of course, is outraged. He hasn't abdicated. He tries to figure out, "How can I overturn this?:" He's sending telegrams, he's mad as hell, but he doesn't manage to. Then he has to flee to the Netherlands. A few weeks later, there's a tiny little item in the newspaper saying that Kaiser has actually officially abdicated.

But you read all the histories about the last days of World War I, and they all say the kaiser abdicated on the night in November, et cetera, et cetera, and don't have this story. So it just is, I think, one good reminder of how powerful this kind of news can be.

So that's an example. I guess you can call it fake news if you want. You can think of similar examples that we've had today as a reminder that disinformation, if you want to call it that, instead of fake news, also can come from the top, and it can be employed for very specific reasons. So I think that's one good and important point that comes out of that.

Another is, though, the question of how much did that kind of news really change? I mean this is the political scientist's question, is like, well, how much does it really matter? I mean would the kaiser have had to abdicate? Anyway, does it really matter how this unfolds in the news is a deeper question, I think.

One would certainly say he was going to have to abdicate anyway, if we look at the broader political situation, but it actually really does matter how things unfold on the day, whether you would've had, for example, a slightly different government within what becomes Democratic Germany. Would you have had different compromises between different parties? There's all sorts of potential ways in which this could have unfolded somewhat differently. So that's another question that I always try and ask myself, but sometimes it doesn't really matter.

So another example is one of the US news agencies gets wrong, when World War I ends. They actually say it ended a couple days before it actually ends. People go on the street, they start celebrating, and then they all have to be told, "No, no, it's not over yet." A couple of days later, it's actually over.

That doesn't really change the course of history. And so, we have to, I think, be careful about some of these. Maybe they really matter, but in other cases it's a fun story. When we're in the moment, I think, of the last eight years, probably at least since 2016, we're not often asking ourselves that question, when is something that is incorrect, that's been planted, or is by mistake, when does it really matter and when is it something that perhaps we don't need to panic about as much?

Yeah, it's super interesting to look at these historic examples, both because, as you said, did they matter or not? Did they change the course of what happened? But also just the way the public reacted. In some cases, they tended to believe the announcement quite readily and maybe they were ready to hear it. Then in other cases in history, they start to ignore what they're hearing.

You think of the ... During the Soviet Union, when the secretary general of the Communist Party, the politburo, was missing for a few days, there would be news releases saying it's all good, when in fact a death had occurred or something and they couldn't deal with the transition. But people didn't believe it really because they got habituated to the misleading news. And so, there's a really interesting analysis about how populations are reacting. Then I guess it's very time-dependent.

Heidi Tworek:

Yeah, I'm really glad that you brought up that example as well because I think it gets us towards another question that I try to address in the book, which is sometimes it's about what news people believe, but equally importantly is what do elites think people believe? Sometimes we see this very simplistic equation of, oh, well, if something is in a newspaper, people must believe it.

But of course our belief systems as individuals are much, much more sophisticated and there are great degrees of skepticism that can develop and so on and so forth. The Soviet Union is one good example of that.

I think that's a really personal lesson for today as well, because we do occasionally see policymakers or others falling into the trap of thinking, well, if there is some sort of story that's floating around on social media, this, therefore, influences people's belief systems.

There's two important things to tease out there. One is that maybe people are more skeptical because of their general experience, as you point out. Two is that even if we mistakenly fall for a piece of fake news, disinformation, whatever you call it, maybe it's just that one item. It's not that it will change the way that we vote, or our belief system in some fundamental fashion. So I think distinguishing those two things is also really important, and it's not always clear that policymakers or others do that.

But it does contribute to mistrust. I mean I'm thinking it's a pretty big head fake to think the war is over, even if it's a couple of days before. So it's fascinating to think that something like that could be viewed as a benign mistake versus some of this more intentional skewing of our views of the world. When we zoom forward, I suppose, to today and the role of media in shaping international relations. I, of course, think of these debates related to the banning of TikTok.

Speaker 5:

Now to breaking news in our nation's capital. The US House of Representatives has passed a bill that would ban TikTok in the US unless its Chinese parent company sells the social media platform. KTLA's Trevor Shirley live in Washington.

Speaker 6:

[inaudible 00:20:13].

And whether government officials should have TikTok on their phones and if there's a threat, but also that it is a source of news for many people, for better or for worse. How can we apply some of what we learned or should have learned from the Germans and their use of technology during the war to our discussions today across different social media platforms?

Heidi Tworek:

It's a great question. Maybe I'll start by saying some of the stuff that's a little bit different. So obviously something that's dramatically different is just the huge amount of individualized data that can be sucked up from our phones, and that's completely different than the situation I write about in the book where often those who are producing the news have comparatively little understanding of how people are reading or absorbing it.

So there are different ways they try to measure it. It's quite blunt, like how many words does X newspaper in Chile print of our News From Germany? It's not as deeply individualized or as sophisticated as what can now be gathered from phones. So that to me is quite a big difference between these two areas.

But I think there's still plenty that we can learn. One thing is that there are multiple different potential reactions to concerns about a foreign country bringing news or information into your space. There's ones where you can go down the more espionage route, which is what we see in the US, including after the war.

What the British try and do instead is they try to invest more in the BBC. So their idea is what we need to do is send out more news from Britain. This is how we're going to combat a lot of the news that is coming out of Germany. So that to me is one interesting lesson. There's actually multiple different ways that you can react.

We're seeing that today as well. The US reaction is we want TikTok to no longer be Chinese-owned, whereas the EU's reaction through the Digital Services Act is actually to really try and implement the Digital Services Act very firmly, and TikTok has actually already reacted by withdrawing a product that they were going to introduce, but it's not by banning TikTok altogether.

So I think the historical example shows us there are multiple different potential reactions. The British one was more monitoring, let's invest in news that we disseminate around the world. The US one was more the J. Edgar Hoover paranoia, put Germans on trial ... This pre-World War II. Put Germans on trial, call them spies, do all sorts of J. Edgar Hoover-like things like send people surreptitiously into offices in Latin America to take photostatic copies of stuff, draw up big bizarre network diagrams, et cetera, et cetera. Those are both reactions, and we can debate about which one of those we think would be more effective.

One of the things that you're seeing now is the complexity of company structures. Back in the day, there were more state-owned enterprises. There were more things that were perhaps clear to identify this is country X behind it, or they're trying to do something to us, or they're controlling something. Now it is quite complex. In the case of TikTok, we're seeing that to unravel the TikTok structure is complex and it can be overly simplified in terms of who's really controlling what. And so, the environment is much more complex than it was, I think, on companies.

Heidi Tworek:

Although maybe if I can pick up on that, actually a really good example of how that's not true is the right-wing news agency that I write about, that is owned by the industrious Alfred Hugenberg and a bunch of his other friends after World War I. It's actually really difficult to pass out all of the stuff that they owned.

They're really the first to create a very vertically integrated structure of media, which the news agency is only one part of it. Even people at the times were writing about the hidden power of Hugenberg. It's really difficult to discern exactly how they created the ownership structures, what were their agreements, and so on. Partly it's difficult because of the things that get destroyed, but even at the time, people find it really, really hard to disentangle.

So I think that there are some parallels actually, just in terms of how company structures are used to obfuscate, who's really owning what and exactly the relationship between companies. That's at least one example where that's definitely true.

Maybe we could pick up on that and touch on news dependency on the state for financial support. So during economic downturns, this looks to be a constant, but also it's a different source of trust or mistrust for people who receive news from those conduits. Have there been exceptions? Is this an argument for the need for state-backed news that operates as more of a protected institution going forward?

Heidi Tworek:

So one of the things that I do look at in the book is one of the main German news agencies, which is founded even before Germany exists. It was found in the mid-19th century. But it enters into a semi-dependent relationship with the Prussian and then the German state, where they have a contract where they basically agree to show political news to the German state for approval and, in return, they get priority over the telegraph.

So it gives them a very strong position in terms of a firm in the market, helps them really sort consolidate a monopoly all the way until the 1920s, 1930s. This is something that becomes really important during things like times of war actually. So it's both during economic downturns, but also during war.

So it's counterintuitive because usually you would think when a war is happening, it's a boon for newspapers. Everybody wants news, so everybody buys. That's true for newspapers, but it's not true for news agencies because they saw the same number of customers, because they're just selling to newspapers. They're not new newspapers that tend to get founded during war.

So, for example, during World War I, it's actually a terrible time for this German news agency because they've got so much more news to report on, but they're operating under the older contract. We also see that in places like Britain. Reuters comes under much more pressure during World War I to cooperate with the British government. I think that's just a dynamic for us to generally be aware of, that there's both economic downturns open this space for news influence, but so too does war.

National security would be another example, which at least for many people, I think, is part of what's created some distrust initially in social media platforms is things like national security and Edward Snowden. So if we want to think about social media platforms as a news agency of the day, then that war/national security footing is another opening there.

In terms of then what we can think of as potential solutions, so there are, I think, multiple different ways in which you can have state funding, but create some clear content divisions by being much more transparent about it, and that the BBC is one example of trying to do that, though there are different ways in which the state funding is very directed towards at least sending these particular places in the first years of the BBC. It's very much dictated by the vision of its initial founder, John Reith, who wants it to be a place that will elevate the working classes through lots of classical music and high entertainment and things like that.

But it at least points to a potential model where you have state funding, but you do have quite clear, transparent division in terms of who's actually creating the content, that there's not what I described with that German news agency where you have a contract, where you're actually showing political news for approval in advance.

So I think those subtle nuances are ones that at least some scholars argue are some ways of trying to get back some trust in news is by having some state funding, but making sure that it's independent in terms of content at least, which is a little bit different than something like NPR in the US, where you'll have lots of consistent funding drives. Then there are other kinds of discontents about a model like that.

Then, finally, I think we've got some really interesting ideas about what you can do that go beyond state backing, sort of ideas like nonprofits saying even something like the BBC, there's questions around influences and how you reported it around the time of something like Brexit. But if you have maybe nonprofit models, these could perhaps be an even better way to go. So you're not worried about any kind of state funding influence.

So the state media, state broadcasters are in the news a lot themselves. You mentioned the BBC and NPR. This Canadian broadcaster, CBC, also is heavily in the news, as you know, about is it too expensive? Is it aligned to one particular viewpoint too much? A lot of criticisms, particularly from right of center voices. It's not a financial question per se. There's a financial question in the background of the cost, but it seems to be an ideological question now that's driving some of the attacks on state-run broadcasters in democracies. I would underline that we're talking about democracies here. Elsewhere, I think it's a totally different story.

But are they going to survive this, or are they just going to be downsized and they'll wax and wane over time as they do, or is there some kind of existential threat against national broadcasters in democracies right now?

Heidi Tworek:

Yeah, it's a great question. It's actually a good question to ask me, I guess, because the period of this book is the one in which things like the CBC emerged. So I guess the first thing is that the CBC is much younger than Canada. So there was a Canada before there was a CBC.

What?

Heidi Tworek:

So there could be a Canada-

Are you sure?

Heidi Tworek:

I'm sure. I'm sure.

The archive.

I guess the historian knows.

Heidi Tworek:

There could potentially be a Canada after a CBC, as it emerges to be a broadcasting corporation with radio and then with TV. And so, its role has really transformed with something like the internet. I think that's the big ... That remains the big debate, and the question is what is the role of the CBCs and the BBCs in this internet world?

The BBC's news is read all around the world, but then its production is mainly paid for by British taxpayers. So finding ways to deal with those kinds of questions for an internet age. I think it's not necessarily existential, but it just requires potential rethinking, just as the very founding of the CBC was a rethinking of the role of the state in information provision.

So that's not really a prediction of where we're going to go. It's just, I guess, saying that the historian would say change over time is something we have seen in broadcasters, and that this is, I think, a moment that really pushes us towards that. It particularly pushes us towards that because the internet is a replacement medium.

So unlike something like radio, which doesn't replace newspapers, the internet can, of course, provide you with radio and an equivalent of television, et cetera. So it poses a really different question for any kind of national broadcaster as to why people in Canada would go to that on the internet rather than going to something else.

Let's talk about the internet. I mean we've been speaking about the news obviously, and I think it's still in somewhat of a traditional way, maybe newspapers evolving and then receiving news or interpreting news through dominant social media platforms. Your book also has political cartoons, and making the connections to modern day memes maybe is natural for me.

I also think of a social platform that doesn't come up too often in traditional discussions of the news, but should. That's Reddit. So Reddit as a space where people both receive news, discuss the news, largely anonymously, sharing their pros, cons, interpretations.

But, increasingly, I'm noticing that maybe Reddit is a little bit of a news agency in and of itself. I'm being a jokester, but I am seeing Reddit cited. Stories come from Reddit. Journalists are reflecting back to us what is being debated, celebrated, reigned on on this other platform. How are sites like that on the Internet changing news, our receipt of news, but also maybe some connections to trust that you were speaking about earlier?

Heidi Tworek:

Oh, I'm so glad you asked me this question, because once upon a time a few years ago, I wrote something basically saying that Twitter, and now I think reading Reddit is like the vox pop of the internet, right, because-

Mm-hmm.

Heidi Tworek:

So vox pop, if you remember, you've got the TV news reporter. Maybe there's a mall that's being ... Or a new bike lane, which would be a classic for Vancouver. A new bike lane. So the TV reporter goes out, has to find one person who says, "I love the bike lane," and one person who says, "I hate the bike lane." It's just some random people walking on the pavement. That's vox pop.

It has all sorts of problems because it's not really a random sample, et cetera, et cetera. It's any people who agree to be on television, so on and so forth. Then if we look at how a lot of journalists are using Twitter or Reddit, it's kind of similar. I mean you're just looking for some stuff, and then you amplify some stuff. Then you find one funny thing that one Reddit person said and another funny thing another Reddit person said. Then we say, "Social media says X." There's no real systematicity to it.

If you look at 2016, that's actually the way that a lot of the Russian internet research agency people got into all of these very credible news sources, was just by putting these kind of tweets, and then almost every credible news organization you can name would just be pulling tweets as examples of Americans responding to things. Then we could show, in fact, they were not Americans.

There's understandable economic reasons why journalists are doing this, because you are under a lot of pressure to produce a lot of stories. Here it's just raining material, it's amazing, but you don't necessarily take the time to go and check, A, if it's a real human, B, who is this really representing, and, C, what can I actually contextualize and make it as a paper.

[inaudible 00:33:51].

Heidi Tworek:

I wrote 2017 was a calling for let's do a little bit more of that, because, otherwise, we're just reinventing the vox pop, which is understandable, but comes with a whole host of dangers because we're portraying things as widespread that may just be one random person on the internet, or it may be actually deliberately being seeded so that it skews our vision of what people are thinking.

It makes me think of ... I'm going to throw to Paul in a second. There's one line you have in the book, "The telegraph union did not just falsify, it also changed rumors to facts." Maybe that's an example of rumor or frumor, a fake rumor, becomes a fact. Sorry, Paul.

Just to pursue Reddit a little bit more. I mean it's fascinating that the insatiable appetite for data right now has driven Reddit to become a public company or a listed company recently there. They're starting to negotiate agreements with data scrapers for generative AI, et cetera. It's a hot commodity, as Vass says. But you're saying it's probably fleeting and superficial ultimately, because the credibility ...

Again, it goes back a little bit to that what is the source of this? It's entertainment value information really in a way. It sounds to me like your conclusion is that that is in the end not going to be a credible ongoing source. But some people feel strongly the other way, that this is the authentic voices that are out there.

Heidi Tworek:

Yeah. I think it can be ... Just because it's a vox pop stuff, it doesn't make it not useful. There's all sorts of amazing stuff you can find on Reddit, and especially certain groups of people. I know lots of my students find it incredibly helpful just to hear about what's going on on campus. And so, there's all sorts of ways in which it can be incredibly helpful. It's just a reminder that it's not a sum total representation of an entire population in some way. Occasionally, when things are pulled from it, quite frequently it's portrayed in that way.

But it's simple things like we should just look at who are the people who use Reddit, and then we know, all right, we're really talking about mostly X, Y, and Z demographics. It's not used by actually the majority of people who are going online, and those who do use it are super active.

So it's a really particular type of community, and that also raises a whole host of questions for generative AI that's using it as a source. What kind of skew is then going to come out in your generative AI, because your source ... Every source base has some bias in it-

[inaudible 00:36:15].

Heidi Tworek:

... but unless you have a really good understanding of that, you're going to wonder why things are turning out in the way they do or why we're missing perspectives and so on and so forth.

We're speaking with Dr. Heidi Tworek and gleaning lessons from her book News from Germany: The Competition to Control World Communications, 1900-1945. If you want to do a deeper dive into this fascinating book, we really encourage you to check it out, especially if you're a history buff. It's absolutely brilliant. Get a copy at your local independent bookstore. All right, back to our interview with Dr. Heidi Tworek, where we pick up the conversation on how generative AI will have an impact on the media.

How do you anticipate that a generative system could have the right lines or draw the right lines between news and information, propaganda, espionage, sensationalism, and opinion when it comes to even remembering source information in the first place?

Heidi Tworek:

It's a massive challenge, but we're also talking about massive challenges at the point when you ask some generative AI to create a picture of eight people, it creates a picture of seven. I can think there's still some more foundational problems still to deal with. But this is a huge issue. It's a huge issue also because we as humans disagree on this, too. Sometimes we disagree on whether something is an accurate interpretation. That's the reason why there are so many different new sources and different ones appeal to us and so on.

So expecting when we humans fundamentally disagree, having a generative AI that can sort that out is going to be a massive challenge. I think there's so many other challenges. For me, though, the continual question is always that question about what is the material that is being fed into these systems so that there's, of course, a whole host of questions about how it's then being used, interpreted in each of these models. But if you have garbage in or some sort of bias, that you're going to get a garbage out. Without having a better understanding of what's being fed in, it's so hard to answer any of those questions, and it raises real concerns.

Then that's even before we start talking about the intellectual property questions, which are obviously about to start making their way through the courts with a whole bunch of cases that have been filed.

Heidi, one of the big takeaways in your book, at least for me, and I think it was part of the conclusion as well for you, was that if a free press does not guarantee the continuation of a democracy on its own, is it still a necessary, if not sufficient part of sustainable democratic models?

Heidi Tworek:

Oh, god, if you don't get a hefty question, Paul, and it's one that I really wrestled with. I mean I think we can say it is certainly something that sustains many democratic ideals, but of course it alone cannot sustain democracy. So I think my point was that if the investment of policymakers, academics, and so on and so forth is solely in something like media literacy or sustaining the press without thinking about border economic conditions, social conditions, and so on, that's not going to be enough to sustain democracy.

The press and the media is a huge part of this deliberative part of democracy, but that's not the only part. If people are fundamentally dissatisfied, and justifiably so, with things like inequality and so on, having access to a great host of news sources is not going to change those fundamental piece of dissatisfaction. I think sometimes we have seen too much focus on the problem of fake news or disinformation as if if we could only "solve" that, democracy would be fine.

I think it takes a media historian or a person who works on the history of policy of communications to come along and say, no, that's not enough. It doesn't mean it's not important or I wouldn't spend all my time thinking about it, but in so doing, that's how I can tell you that it's not enough.

So seemingly necessary, definitely not sufficient. It's a whole ecosystem of institutions in fact that are, in a way, integrated with supporting a similar democratic ideal.

Paul mentioned advertising earlier, and, Heidi, you threw out the word labeling. I think that some of our mistrust with media can be ... And this is not blaming the media, but sometimes that marketplace of news is murky in the sense that there's branded or sponsored content or advertorial that's not labeled at all or called something different. Sometimes our newspapers or magazines or favorite food or fashion bloggers mask that this link that they're suggesting as an influencer is an affiliate link and that they're earning money in return.

Do you see this activity of labeling as a prospect to add better or more transparency in our news marketplaces, or is it just a lazy hope that if we add certain stickers to certain sources, we'll be able to clarify their trustworthiness?

Heidi Tworek:

I mean I think the literature that we have on there says it's useful, but it's not a panacea. What does the literature say on panaceas? I'm just kidding. Tell us more about-

[inaudible 00:41:41].

Heidi Tworek:

If only there were some.

Yeah.

Heidi Tworek:

But I think what it tells us that this points to the bigger and deeper question, which I guess my whole book really is about digging below the surface. A label wouldn't have fixed Germany's problems in 1928 either. It's actually much more about a broader systemic set of questions. So in the case of advertising, why do we have all this stuff? Will we have a media ecosystem that's got some very deep financial problems? It didn't just begin with the internet goes back to at least the 1980s, and those are much deeper questions.

That's the reason why we've got to have an average world, et cetera, because you've got to figure out a way to make this stuff pay. Part of it is an even deeper question, which is if you look at the history of news the last 400 years since we have newspapers in the 17th century, most of the time it's extremely hard to make money with news. It's very, very difficult. There is one exceptional period, and that is between the 1940s and the 1970s, which is just like a golden age for newspapers.

Speaker 7:

Important story of newspapers, the story of how to read them. Do you know why you should read more than one newspaper? Can you use newspapers to help-

Heidi Tworek:

Owning a newspaper is like a license to print money. You are making margins like you wouldn't believe. It's incredible. Then it goes away. But, still, our baseline expectations of news are based on that golden era, that exceptional 30, 40-year period. I've argued with John Maxwell Hamilton who's at Louisiana State University, we've got to change our perspective. Take the 400-year perspective. We see it's just really, really hard to make money with news, and we are in a normal period in a way where that's true. And so, we should look at that longer perspective rather than taking the baseline as that 1940s to 1970s newspapers are a license to print money.

And so, then we start looking at different kinds of solutions, I think. Once we think like that, we say, okay, well, if it's going to be hard to make money with it, like I said, there's nonprofits, but then there's maybe a whole host of other things that we could be doing as well. If advertising is a big part of this ecosystem, how are we going to ensure truth in advertising? What does that look like, et cetera, et cetera? It's a bit of, I think, a different set questions and a different mindset to approaching how we deal with this issue than assuming we're going to get back to 1972.

I think where I was going to go next on this is if we wanted to talk about how companies generate money in the digital age, how do you get a slice of that intangible economy? It's tending to be captured by a very small number of players. And so, that works against certainly small media, but perhaps even medium-sized media.

Heidi Tworek:

I mean maybe if I can just ... I felt my answer was already long enough, so I wouldn't kick up. But just to be like, okay, the best way to think about Facebook and Google is that they are advertising companies. So we call them platforms, but where does all their money come from? It's from advertising. That's, I think, just ... It places them in a different perspective when you think about them as advertising companies, or whether they should be more subject to all of the things we expect from advertising.

That, too, has a history. Like adverts in the 19th century in the US, you say whatever you wanted. There were all these medical ads that are promising quack cures.

Speaker 8:

Too tired too often? Try Rybutol. If you two need extra amounts of thiamine and riboflavin, Rybutol can help you gain [inaudible 00:45:00].

Heidi Tworek:

You name it. Then by the time we get to the 1930s, the FDA is regulating that because people are dying after taking these quack cures. And so, then there's a regulation of medical advertising. [inaudible 00:45:11], that may still be, there's still an idea that somehow the state has a role in regulating advertising and making sure that it's not completely falsified.

And so, anyway, I just think sometimes, although I certainly write about Facebook as a platform, and I think there's a lot of use in thinking about platform governance, obviously did a whole series with CIGI about that, it's also just really helpful to understand that their baseline financial model of being advertising companies and what that means, that they have to do what advertising companies do, which is you've got to understand your customers. That's why you need so much tracking. You've also got to sell to advertisers.

And so, that's why the campaigns that work really well with those companies are ones where advertisers start pulling their money. So I just wanted to make sure that I said that somewhere because I think it ties the history stuff to the present in a really useful way.

I really appreciate touching on the competition elements that have caused a shift and what the introduction and shift in advertising means for those competition for dollars and for eyeballs. If we can keep it in the competition space, just to indulge me a little bit, when we were talking about deep sea cables a little bit earlier, Heidi, do you observe that we're seeing maybe versions of the race that you documented in telegraph technologies today with the races for fastest wireless speeds or going beyond 5G? Should we be applying some more lessons there?

Heidi Tworek:

Yeah, I think we do see a bunch of interesting parallels and, again, part dependencies. So the first thing we see is that actually companies like Google, et cetera, are investing in cables. They have been since around 2015, 2016 trying to lay cables to Africa. And so, they're trying to lay new routes so that they can acquire new customers. So that's one interesting element.

Then, of course, we see lots of states being very interested in providing communication infrastructure. That's obviously part of China's Belt and Road Initiative, and there's a big focus on providing things like 5G to Africa. So those things for me fit very much in this longer history of moments when states will look to communications and communications infrastructure as something that they want to support for geopolitical reasons. Then that there will be firms which maybe begin as more content firms that then also want to invest in the infrastructure as well. Reuters tries to do that in the late 19th century.

But one of the very interesting things that is really a very, very old way of thinking about these cables is that the way we repair them is basically the same as it was in 1866. There's a really great article in the Verge that looks at one particular ship of ... I think they're basically all men who, after the nuclear disaster in Fukushima, they have to go and repair a bunch of cables.

Ironically, there's basically no internet on these ships that are repairing cables, and there aren't enough of these ships. And so, this is a massive concern because in fact cables can be cut through fishing and human error, et cetera, et cetera. Of course, if there were any military conflict, you cut those cables, everybody is in extreme trouble because that's 99% of internet traffic, probably at least 95%.

The way that they are still repaired, though, is basically the same as it was in 1866, which I find totally fascinating. It's something that is rarely talked about, but is massively important because although it seems like such a stable infrastructure, it's incredibly fragile. It just takes a few hours for the entire system to be knocked out, or one natural/unnatural disaster. I think that's worth bearing in mind, and it's certainly something that I think states have become more and more aware of. But what we do about it that's different than 1910, 1911, 1912, that's up for debate.

So cable infrastructure remains hugely geopolitical and hugely important, as you say, especially since there's a military connection as well, like taken very, very seriously. But there's a new technology on the block, low Earth orbit satellites, Starlink, the whole question there. That has rattled a lot of cages, including on the military side in the big powers, but also on the commercial side. Starlink has had a quasi-monopoly on this because they've got the ability to deliver these.

But it can be a game-changer in certainly remote areas that don't have cables now. But there's a lot of debate about whether there will be a displacement of the importance of cables or with these low Earth orbit satellites. So I don't know if you're ... I imagine you're following that, but do you have any comment on it?

Heidi Tworek:

Only that we've had that debate before, because of course these aren't the first satellites to exist. In the '60s and '70s, there's a real question whether satellite is going to be the next technology that potentially replaces radio or telephone, and that obviously doesn't end up working out, although we do have satellite TV.

So I think we're still up for debate as to exactly where this is going to go. It just reminds us that just because there is a new technology on the block, it doesn't mean that it's going to become ubiquitous. So I think as a historian, I've learned that predicting the future, very, very difficult.

What we can certainly say is that it's not guaranteed to be a replacement, though it may be extremely useful in a particular situation. I'm thinking of rural areas where it's just way too expensive to lay cables. We've obviously seen in the case of the Russian invasion of Ukraine, that Stalin came in incredibly helpful. So we may see that it develops into something that's more supplementary rather than replacement, and history tells us either one is possible.

Yeah, totally. It feels supplemental right now. When you think of electric vehicles, they took a hundred years to challenge combustion engines. Now it's still ... They haven't taken over instantly. So things that technology plays out in very, very surprising and interesting ways.

On that topic of displacing or supplementing, you also present news as a form of entertainment and surveillance. We haven't really dug into the surveillance aspect. So we can, if you like. Why can we not take news at "face value" and is news ... I mean not is, but how is news still a form of power, or has the News been ceding power to these other forms such as even podcasts or YouTube channels, where we are broadcasting forms of news, again, in different, more fractured places?

Heidi Tworek:

So I guess I see news as a very broad category, and that's why something like entertainment can absolutely be used. For a lot of people, when they bought a newspaper prior to the advent of something like television, they did not buy it to read foreign policy or even things that were going on in capital cities. They bought it because they wanted to know about sports, and that was news.

That was one of the reasons why newspapers had such a big problem when the internet came along. Suddenly the people who bought the newspaper for sport were just going to a website for sport instead. That disintermediation had a huge effect.

So that's why I think it's just really important to remind us that the things that I think a lot of us on podcasts like this are really concerned about the foreign news, the hard-hitting stuff is only one category of everything that comprises news. So that's the entertainment portion.

And also that entertainment can be really legitimate news, if we think about satire is a huge way that many people actually get their news, in part maybe because they want joy from news, they want to laugh at the news, and other ways of consuming the news feel too depressing. So they'd rather at least have a sardonic laugh at what's going on. I don't think we need to condemn that. That can be a very legitimate way of getting news in some fashion.

Then the question of whether news is surveillance. I mean there are lots of different ways in which news can become a form of surveillance, and that's in part because sometimes journalists have acted as spies. Sometimes they've been very wrongly accused of being spies that've had terrible consequences for them.

But they have sometimes served dual function. Sometimes they've also provided news to the state in the past. These are all ways in which the profession of journalism has been very blurry. It's now I think sometimes much more clearly defined in some places, but not everywhere. As I say, you can see ways in which journalists still get accused of being spies for geopolitical reasons.

Speaker 9:

Now a Russian court has found US journalist Evan Gershkovich guilty of espionage, sentencing him to 16 years in prison. Russian investigators claimed the Wall Street Journal reporter was collecting data on a Russian tank manufacturer for [inaudible 00:53:54].

Heidi Tworek:

There are also ways in which us reading the news today, we do actually get surveyed in a quite detailed fashion by media companies, again to know what to serve us. I mean a newspaper knows exactly how far down you have scrolled on a news item. We can call that surveillance or we can call that something else, but it seems to me that it's a way of knowing about you as an individual that is incredibly detailed, not akin to how we were reading newspapers 50 years ago.

But I'll also end by saying, from my understanding, an astonishingly small number of people scroll down below the first paragraph of an op-ed or news article. So if you've got a point, you've got to make it quickly before people stop scrolling.

Ooh.

Yeah. I feel like sometimes people want a dense message more than the first paragraph, even though ... But we won't go there. I wanted to note one thing in your book. You outlined how Weimar officials of the day created structures to oversee and regulate news supply, and that it backfired that they were too hands on. They were too hard. Are there any lessons that you drew out of that or things that have repeated again in history that you could speak to?

Heidi Tworek:

Yeah, thanks, Paul. It's a great question. So basically spoken radio emerges in the 1920s, and German officials are trying to figure out how to regulate this new medium. The man in charge of it, a man called Hans Bredow is trying to think about how you can actually regulate this content in such a way that it doesn't inflame an already fairly revolutionary situation in Germany. His belief is that you then need state supervision of this content in some way.

As the Weimar Republic becomes more and more febrile, there are several radio reforms where there's even heavier handed state supervision of content. This is all done in the name of protecting democracy. But the sad irony is that then when the Nazis come to power in January of 1933, what is the one medium where they have state supervision? It's radio. And so, by August of 1933, Joseph Goebbels gives a speech where he says, "The reason we could come to power and hold it ... It's the airplane that helps them come to power." He claims it's the radio that helps them hold it in those first few early months where they're consolidating power.

Obviously there's lots of other reasons as well, but I think it's a really crucial lesson that we're actually very careful about good intentions not always having good outcomes. But I think really carefully about any kind of concentration of power of media, what is the balance between where the state can actually be really important in protecting things like freedom of expression alongside trying to protect democracy? That to me is a cautionary tale then of where you could want to protect democracy, but in so doing reach so far that ultimately an authoritarian regime can take advantage 10 years later.

Heidi, we've been so fascinated by this conversation. I'm actually falling into my computer trying to get closer and closer to you. So I hope that hasn't creeped you out. As we round out, we'd love to hear a little bit more about a project either you're working on right now or something that's on your mind that you're thinking about. You mentioned your fellowship out the gate.

Heidi Tworek:

Yeah. So I'm actually here in Bielefeld with a group of international relations scholars, sociologists, and communications infrastructures to understand transformations of world politics. So usually communications wouldn't necessarily play such a big role, but we're really trying to see, have we used communications infrastructures to understand transformations of world politics over the last couple of centuries?

Heidi, thank you so much. Do you want to say thank you, too? Now we're just riffing to get our thank yous-

Thank you.

... and get a bit of text before we let you get to the hail.

Thank you, Heidi. You're normally at UBC, which is my alma mater. I'm from the west coast. Maybe we will one day meet on the UBC campus.

You could interview Paul about that new bike lane. He'll give you a very unique representative perspective.

Heidi Tworek:

I will look forward to it. Thank you so much for having me.

So, Vass, what remains in your head about that great discussion with Heidi?

So much. I mean I really appreciate all of her work. But I still think we, not just the [inaudible 00:58:00] we, but just generally. I think we lack an appreciation of what it means to "control" social media right now. It's not just the ownership or the board of directors. As individuals, it's difficult, if not impossible to autonomously navigate an advertising ecosystem. Companies don't just control what we're seeing on a web page or on a mobile browser, but whether we see it at all.

So I just really appreciate how deep Heidi is able to go on a very difficult and complex history in order to inform, or better inform, our present day thinking on these big regulatory questions.

Yeah, Vass, we're on the same page. What really hit home for me was the history of Germany and just all the nuances and stories she brought out about that eternal battle of controlling information, controlling the narrative. When you think back in history, when writing first got mainstreamed, it was controlled by groups, whether it was church, Mandarins, other government agents, and it's always been that way where there's been a big struggle for control and overturning these regimes that are imposed.

I also thought that the analogy was very pertinent about the idea of undersea telegraph cables, which were just so important as World War II was shaping up. It makes me think a lot about the tensions we have now around chips and access to AI technology and everything else. It's a little bit back to the future.

Oh, absolutely. Broadband access, the digital infrastructures that underpin these systems are also a source of contested control.

Policy Prompt is produced by me, Vass Bednar, and Paul Samson. Tim Lewis and Mel Wiersma are our technical producers. Background research is contributed by Reanne Cayenne, marketing by Kahlan Thomson. Brand design by Abhilasha Dewan and creative direction by Som Tsoi. The original theme music is by Josh Snethlage, sound mixing by François Goudreault, and special thanks to creative consultant Ken Ogasawara.

Please subscribe and rate Policy Prompt wherever you listen to podcasts. Stay tuned for future episodes.