 Thank you all for being here. I'm David Sarota. I'm chairing this panel. And it's called, Fake News and Fake Experts. Or should the experts in the media find new citizens? This is a topic near and dear to my heart as a journalist. And defining what fake news is and all of that is extremely important to those of us who are trying to not do fake news. I'll just introduce the panel here. On the panel we have Rob Farris, who's the research director at the Berkman Client Center. Philip Morowski, who's a professor of economics and history of science at the University of Notre Dame. Sheila Dow, who is an emeritus professor of economics at the University of Stirling here in Scotland. And an adjunct professor of economics at the University of Victoria in Canada. Tom Ferguson, who is INET's research director. And as we always do, we'll start with presentations and then we'll do a discussion afterwards. And I think Rob, you are first. Great. Wonderful. Can folks hear me? I would stand so you can see me better, but I think I need to be tethered to this mic. Does it lift right off? We'll give it a go, see if this works. All right, here we go. So it's a pleasure to be here. My talk today is going to be basically about media and democracy. I'm also going to talk about expertise, authority, and the flow of information and disinformation in media ecosystems. In a prior generation of communication scholarship, there was a very simple model that explained agenda setting. So political elites would shape the media agenda. This agenda would then be passed on to the electorate through large media organizations. Now that we live in this age in which consumers can choose from a vast array of media sources and can choose their own media diet, what we see is we see the emergence of unique media ecosystems that emerge around a given topic at a given time for specific communities. And I'm going to show you how we've gone about studying this. The example we're using is from the US presidential election from 2016. I think the challenges and the questions that come out of this apply to other issues and other places. This is all done using a platform that we've developed at the Berkman Klein Center with our friends at the Center for Civic Media, MIT. It's called Media Cloud. If you're interested in media analysis, go check it out. Give it a whirl. It's open and free source. I'm going to show you three perspectives on media ecosystems. Each one of them by themselves is very telling, but inadequate. When we put them together, we have a more complete vision of what the media ecosystem looks like. So this is based upon the collection of approximately 2 million stories over an 18-month period produced by 70,000 media sources. And this first view of the media ecosystem is based on interlinking between media sources. We often call it the link economy. So this is a media-centric view of the media ecosystem. Each one of the nodes is a media source. The size of the dot is a function of the number of other media sources that link to it. Proximity in space is based upon common end links. So for the New York Times and the Washington Post, there are many other media sources that link to them both. Huffington Post and Breitbart have relatively fewer media sources that link to both of them. We also created a measure of media partisanship for this study that's based on the behavior of Twitter users. And so what we have is the colors represent five quintiles from the left to the right. And the way we created this was looking at the retweeting behavior of those people who retweeted Donald Trump and those that retweeted Hillary Clinton. On the left in the dark blue, those are media sources that are more frequently retweeted by Hillary Clinton retweeters. I'm sorry, these are media sources that are shared more frequently. They have URLs embedded in their tweets by those who retweet Hillary Clinton, by at least a margin of four to one. On the other side, the red ones are the far right. The light blue is the center left. The green is the center. And it doesn't show up very well in this resolution, but there's pink ones which are almost non-existent in this. So this, again, is a media-centric look at the media ecosystem. This is how media sources look at one another. And what we can take away from this is that the center left is really the most important, the most influential in this part. We also see that the center and the center left and the left are pretty well integrated and the right is off by itself. The center of the right is Fox News and Breitbart News. Here's another look. This is based upon Twitter, similar logic. This time the size of the nodes and the measure of influence is the number of times the media source is shared on Twitter. Proximity is based on users sharing both media sources. So in this example, there are many users that shared stories from Fox and Breitbart, bringing them close together. What we see is a similar shape of the media ecosystem here, except that Breitbart is more prominent than it was in the media ecosystem. We also see other media sources that become more prominent. The Gateway Pundit pops up here. Hill does very well. The basic left center core of the left side is holding place. The right starts to look a little more different as we go in this way. So this is the last map I'm gonna show you. And this is based upon Facebook rather than Twitter. And we have similar trends here, except that Breitbart is even more prominent than it was in the other views. And we see the emergence of many other sources on the right that we would not see in the link economy unless so in Twitter. Daily caller is one. Truth feed is one. Conservative Tribune, what we see is a lot of the less reputable on news sources emerging on Facebook. We used to term for them political clickbait. They tend to be sensational headlines. News sources that do very little original reporting. They're grabbing information from other places. And there's a tendency for highly biased, if not outright false reporting. You'll see Gateway Pundit appears there. Infowars appears there for those of you who know them. This is basically a story of two very separate media ecosystems. These are screenshots from July 18th when the Republican efforts to repeal Obama's healthcare plan failed. The New York Times and the Washington Post are talking about that in the headline. Meanwhile, Breitbart is talking about a festival in Germany that resulted in violence. And Fox News is talking about the Clintons, although there is a little headline down there about the failure of the healthcare bill. There are many, many such examples. This is a particularly telling one. This is telling a similar story in a different way. This is the distribution of the top 250 media sources by these three different perspectives. So the link economy is on the top, Twitter on the left, and Facebook on the right. And what we see is what we call asymmetric polarization. One of the themes that came out of this is that the left is very, very different from the right. The right is more skewed towards the far right. The left center is suppressed. And the center of gravity from the left is within the left center and the left across those places. When you move to Twitter, this pattern becomes more pronounced. The partisanship of the right is even more distinct. The prominence of the center and center right is diminished. When you move to Facebook, again, this pattern is even more pronounced. So these are three different views of this media ecosystem. They're all legitimate in their own ways. They all have their own traction in their own ways, but they look different. Disinformation propaganda is not new, of course. We see the manifestation of old forms of that. We also see new forms of propaganda occurring in a network phenomenon. So a lot of the things are kind of old school propaganda, the mixing of truth and fiction, insinuation, leaps of logic. What we see in the network public sphere is we see the emergence of many, many different media sources that copy each other, tell the very same storylines, and gain legitimacy by the kind of the numbers that they put out there. Here's a few examples I couldn't resist. So Info Wars and Gateway Pundit, they're infamous for peddling conspiracies. So Info Wars is talking about the ties of Jeb Bush, I'm sorry, to Nazis. There's many people that believe that the Clintons are responsible for knocking off people who get in the way of their political aspirations. So here's Gateway Pundit reporting another mysterious death. That's Breitbart there. Breitbart is masterful at taking a core bit of truth and reporting it and twisting it into a nefarious liberal conspiracy to undermine the white Christian core of the United States. The Daily Caller is another one that gets the facts wrong frequently. Here's an example of how information bounces around within the right-wing media ecosystems. This is a story around Hillary Clinton and ISIS. It started with a story published on a Jordanian news source that was picked up by a source called Zero Hedge that reported a Saudi prince claiming to have funded 20% of Hillary Clinton's campaign, which is illegal and a little bit absurd in the face of it. The story was subsequently taken down. The news source reported that it was hacked. Nevertheless, it made it to infowars. It made it to Fox News. And then this infamous fake news source ending the Fed somehow took this and conflated it into a story about Hillary Clinton funding ISIS. So we see the memes kind of traveling through these different ecosystems in interesting ways, sometimes amplifying things that come from the fringe and vice versa. Another thing that we see is of course using emotionally loaded messaging in putting things together. We using media cloud were able to track the topical coverage of media over the course of the campaign. This doesn't include any of the scandals. This is just topical coverage. The things that got the broadest coverage are immigration and things related to Muslims in Islam. These far surpassed the other substantive topics of the election and within that, Breitbart spent a much larger proportion of their coverage on the top of immigration. When we map immigration by itself, Breitbart is far and away the center of the discourse around this topic. And when they're talking about immigration, they're talking about disease, they're talking about terrorism, they're talking about burdened on government services. The racism is not explicit, but it's unmistakable. The last piece of propaganda I wanna talk about is the way that the right wing was able to influence the agenda of the center-left media. Actually, I'm going here first, which is the attacks on other sources of authority. Breitbart played a very prominent role in going after Fox News early in the election, which was quite interesting. Fox News during the primary session was kind of caught within their conservative audiences. And wasn't willing to place their bets on anyone early on in the campaign, unlike Breitbart. So Breitbart used this interim period to go after Fox very heavily. And it took a toll on Fox when we look at a network map in January. Fox was kind of out in the middle of media ecosystems there and of lesser prominence. After the convention when Trump won the nod from the Republican Party, Fox pulled in line behind the Trump campaign and the media ecosystems went back to where they should be. As I started to say before, the right wing media was quite successful in influencing the agenda of the center-left. This is looking at topics across the 18 months of the campaign that we tracked. And the one issue that got more attention than any other was Hillary's emails. Trump's coverage was led on immigration and jobs and trade more than his scandals. Clinton's coverage was focused more on her scandals than on the substantive issues, which she would have preferred the topics to be covered. A lot of this was planted early on in the campaign. It was a very prominent book done by one of Steve Bannon who runs Breitbart. One of his partners wrote a book called Clinton Cash, The New York Times and The Washington Post. Both got an advanced copy of this. And The New York Times wrote this article here. Cash flowed to Clinton Foundation of a Russian uranium deal. This was one of the most cited articles in The New York Times across the election. And what it did is it provided cover for Breitbart to say, yeah, we were telling the truth, see? So Breitbart, this is one of many examples of this, but you can see the headline, which is 11 Explosive Clinton Cash Facts Mainstream Media Confirm or Accurate. So what they did is they planted these stories to the best of their ability and then cited mainstream media in furthering these narratives about Hillary Clinton. So I'm gonna wrap it up there. So what we see is we see a very nimble digital media ecosystems that are trafficking a lot of half truths without outright fantasy. They are very resistant to correction. There's a great deal of manipulation going on there. The jury, I think, is still out on that to what extent the bots and Russian meddling in Facebook had what the impact of that was. It's unclear and hopefully we will learn more about that in the future. Fake news, I think is fake news in the sense of purely fabricated news was a very small part of the conversation. We found examples of these stories that were just made out of whole cloth, but they were by far the exception and very few of them gained prominence within the top stories that we tracked. The more common is the propaganda and disinformation, this mix of truth and innuendo and leaps of logic. The very asymmetry of these things we see to us suggests that anyone who wants to point the finger at technology is that's not a great description of what went wrong here. If so, you would expect to see the same kind of dysfunction on the left that you did on the right. There was some of that on the left, but by far it was much more prevalent on the right than on the left. So I think the problems are undoubtedly embedded in political and social phenomena. Technologies certainly didn't help. I think it helped to further this, but it worked very, very differently on the right and on the left. And so I think I will leave it at that. Thank you so much. This idea goes back at least 100 years. The board made themselves to the market. As a validator of truth, this is all through, goes back at least 100 years. And I think that at minimum there's an opportunity to go look and do that. Now, every single... What that does is that gets rid of the role of education in a citizenry too. So citizens are not being educated to be citizens anymore. They're just getting human capital. Only entrepreneurs live fully realized lives for a neoliberal. But of course you can't tell people this directly. And for that reason, and I will say this twice, truth becomes unmoored from argumentation. That is, as part of a political doctrine, truth becomes unmoored from argumentation. I hope you see the relevance of this shortly, okay? Now, what I'm gonna do here is to use this to begin to analyze fake news, okay? Now, what I'm surprised is how many people on the left keep claiming that fake news even exists. And I've got some quotes there. I'm not gonna go into them. What's important for me is to be very precise about its definition. Maybe I might differ with you a little, Rob, because of this. Fake news, by the way, if anyone has checked out Snopes. Snopes is one of those websites that are supposed to be kind of truth checkers, okay? So this guy, this is his whole life. And his quote is, fake news was a term specifically about people who purposely fabricated stories for clicks and revenue. Now, it includes bad reporting, slanted journalism, outright propaganda. We are doing a disservice to lump all those things together. Amen. Fake news is a narrow phenomenon, not whatever you feel like it should be. Which, by the way, is Donald Trump's attitude, too, right? You don't wanna play that game. Okay, so, fake news is, in a sense, a neoliberal innovation. It is not like Orwell's 1984, and I feel sad for all these people who are picking up copies of it and think that it sheds light on the modern situation. Actually, it does not at all, okay? The difference here is that befuddlement became an active political strategy very different from the top-down broadcast model of early 20th century propaganda. Here I'm possibly differing from Rob because propaganda is intentional deception, and what I'm going to argue is fake news is not intentional deception, it's intentional befuddlement, which is not the same thing. Disinformation is predicated upon the creation of a frog of confusion and disillusion and less directly promoted by straightforward medium manipulation, okay? Now, there's also, as Rob mentioned, there's this idea that somehow it has something to do with technology. Now, my argument would be the neoliberals theorized this before the internet ever appeared on the scene. I don't have time to do it, but I've done some history on this, that they've theorized the problem and what to do about it before the internet arrived. They didn't know the internet was gonna be invented, but once it was privatized in the 1990s, it supercharged this neoliberal vision that it developed in discussions at Montpeler and elsewhere over the centrum. Basically, what it allows is to recast the market as an amplifier to recycle the vulgarity, the twaddle, the gibberish, and noise back into the public that generated it in the first place in a cybernetic feedback loop, to such an extent that the people involved, the populace, would have no clue what is actually going on in their world. That's not like propaganda. And indeed, this didn't have to be intentional because it's driven by two really neoliberal political programs. One, the transformation of this befuddlement into a lucrative source of profit. Do you see how this can happen? Even if it's not intended by, let's say, the people who start the platforms. But secondly, and neoliberals discuss this, it rendered the populace much more docile in the face of a neoliberal takeover of the government. By the way, I hope you guys understand, neoliberals don't want a small government. They want to run the government. They want a strong government, okay? And this is part of it, okay? Now, the way this happened is through four rough stages, which I can just briefly point to. The first, and I'm sure you all know this, by the way, when I wrote this paper back in March, no one was talking about this, and now it's in the newspaper every day, so I don't have to do it quite so much. That basically, the first effect of having news being conveyed in this way is the de-skilling and the casualization of journalist labor. No journalist can specialize in a beat anymore. They can't know something in depth, all right? So you're already sort of undermining the sources from which this knowledge is going to come from, and I won't go into that anymore, but I will go into the other three. The second thing that happens is that you replace people with algorithms to call, curate, and convey the news through platforms. I'll tell you a story about Facebook in a minute. This is a common thing that happens in the recent history. The third stage is then why stop there? You also need to automate the selling and placement of advertising, which funds the platforms, which perform function too. This is where the economists enter in, and then there's a fourth stage, which I probably won't have time to talk about. If you're gonna automate the news itself, its distribution, its payment, why don't you automate the audience as well? It's not a joke. And that's the role of political bots. All right? Let me tell you a story about phase two. You might say to me, wait a minute, where are the neoliberals in all this? It's mostly about Facebook and entrepreneurs and all this sort of stuff. Well, you guys, I hope you trust and know the story. You have to do it very quickly. Facebook only added this news feed in 2012, but for a long time, they had actual human beings to curate it. Now, by the way, this is different than people in India looking at 36 frames per minute and trying to decide if it violates some rule. These people actually decided, yeah, we'll put this in the news trending box, and we won't put stuff from here. In May 2016, somebody in a neoliberal think tank just claimed that the human editors who were curating the trending box on Facebook were biased and routinely suppressed so-called conservative websites, which you've just heard about in Robs. For example, Breitbart was blocked a lot, okay? In other words, a group of neoliberals were miffed that somebody would take into account the source of the information that was being conveyed on the platform. Do I need to say that again? You're not supposed to do that, according to these neoliberals, and why, let me just be clear about this, because the market is the validator of truth, not people. So what happens is the think tanks drew this up into a causal lab. Three months later, the human editors were fired and replaced by algorithms, and this has been documented by a number of people. Immediately, the volume of fake news on the trending box exploded on Facebook. That's the direct intervention of neoliberals, right? By the way, this is how the algorithms work now, because, and by the way, I don't blame people's psychology. I mean, this is a different way of getting people's attention and then validating it. Is it a headline that makes people click? Does it have pictures or videos? Put it on automated, real-time selling to advertisers. That's how Facebook still works. This is not the only way that automation fosters fake news. Advertisers cannot keep track of where their ads appear, and in order to be able to act in real time, in other words, why you're clicking through, they, with the help of market designers, have developed real-time markets that auction and place the ads while you're clicking through. Now, my point here is, because people ask me, how do you fix this? My point here is this is the route that needs to be cut. This means that anyone who both generates or just even passes along, a lot of this crap will make money. So maybe we disagree. That money is central to the way this works. Back before this happened, Facebook was not having these problems, or at least not to the extent that they're having them now. Moreover, I have a quote from, I love this, from Fox Networks, honestly, the long tail is to advertising what subprime is to mortgages. They have no idea what crap is out there. And it's economists that did that because they believe the market validates truth. And almost done. As I said, the final step is of course, not only to automate all those other parts of it, but to automate the receiving end as well. There's a lot of work done on political bots. I don't have time to talk about it. It's very interesting. It's role in recent elections, why it didn't have a lot of role in Germany, but did have a lot in the United States. Very interesting question we could go into if you wanted to. But here's my point. This is also an attack on democracy explicitly. Suppose you believe in deliberative democracy. Say you're Juergen Habermas or somebody like that. But the trouble is, most of the people online think they're arguing with someone, but really they're arguing with a robot. How much longer are they gonna believe in deliberative democracy? This stuff is poison. And it needs to be eradicated. Sorry, my fault. So I'm going to pick up on some of the things that Phil's been talking about, but I'm not going to deal with social media. I'm going to look at this whole area from our perspective as experts and to discuss it in a different way from the way in which Phil has discussed it, how far we're complicit in fake news and fake expertise. And what I want to do is develop a non-binary or non-dualistic approach to addressing this question. And I'll explain what that means as I go along. But the purpose is to try to envisage a framework for economics and economists which allows us to bridge the gap between us and public discourse. Clearly there's been a disconnect there which needs to be addressed. So my starting point is not a US focus, but a more British focus. And it reflects the way in which economists have been talked about here, both in the wake of the crisis and in the wake of the Brexit referendum. I should say as an aside, this does not apply to the Scottish referendum during which economists were actively engaged in the debate on both sides and seemed to command a reasonable amount of respect. And there was a meaningful debate across society. But this differs from the Brexit debate where there was a very clear buildup of mistrust in experts and their opinions, verbalised by a government minister who said the people have had enough of experts. They're consistently getting things wrong. And this fed into mistrust which had been growing in any case in the wake of the crisis. There was a widespread perception that economists had not predicted the crisis, were finding difficulty explaining it in retrospect, but that then economists were complicit in a policy agenda which supported austerity in fiscal policy which had dire distributional consequences which were made worse by the fact that the financial sector seemed to be relatively unscathed. And of course, Donald Trump, I mean, if one can believe, I realised I used a news feed for this quote that's in the paper, maybe I shouldn't believe what I read. Trump was quoted as saying that experts are terrible, they've made a real mess of things. And indeed that had real consequences because the number of experts, civil servants in Washington was drastically cut. So there was a very real effect of that viewpoint. But within society as a whole, there was a growing sense that experts and in particular economic experts were not to be trusted. The fact that economists were seen to support austerity policies implied that economists had bought into a particular sectional interest, that economists were not concerned about the distributional consequences of austerity policies. So what should our response be? I mean, our business is expertise. This is what we do. And therefore, automatically, any suggestion that expertise should be rejected is something that we would want to dispute. And similarly, any suggestion that we convey fake news is something automatically that we would react against. But this is made difficult by the fact that there is this disconnect between the public discourse, which sees what we do very differently. And it all fits into the whole issue of a power struggle between sectional interests in society. And this is something that's very much at the heart of the growth of populism on the left and on the right. Both involve suspicion of establishment. And part of the establishment is perceived as the economic expertise. This follows from the way economists have input into the policy debate. Now, in this political environment, if we can't trust expertise as a separate phenomenon, then it becomes perfectly legitimate to pick the expertise that supports your own point of view. And this is unashamedly what many politicians have been seen to be doing. And I see this as an extreme form of post-modernism. An extreme post-modern position is that any viewpoint is legitimate and cannot be challenged from an alternative viewpoint. And therefore, it's not a question of an alternative viewpoint. And therefore, it's perfectly legitimate to seek support for your arguments from those experts you choose, those facts you choose, anything goes. Now, how have economists responded to this? Well, the mainstream reaction takes the form of what I would characterize as a modernist approach. This is the other side of the dual. On the one hand, you have the post-modern anything-goes argument. On the other hand, you have the modernist response from mainstream economists, which holds to a view of expertise as being unified by the principles of science, being legitimate, being expert, being something that is separate from society, picking up on part of Phil's argument. This is the way in which mainstream economics developed, that it's a technical subject which doesn't touch on values. Values are something that come in later on in the policy process. David Collenders has an article where he talks about the art of economic policymaking, all the political, institutional, empirical issues that come into actual policymaking. But he argues that this should draw on economic theory, which is something set apart. So there's this fiction, as I would regard it, of economics as being something technical, which is not complicit in any support of any sectional interest, but is something set apart. And the rules of science which it follows involve an unquestioning acceptance of the rules of classical logic to derive propositions which are then tested against independent facts. OK, predictions were a bit awry when it came to the crisis. If you look at the response to the Queen's questioning of why economics failed to predict the crisis, OK, we didn't get it right. So let's try harder. We just need to do better. If you look at a whole array of articles, Cavalier was one that was mentioned earlier, there is an admission that things haven't gone quite right, but they would just need to do it better. So my non-dualistic suggestion is that economics should be done differently. It's not that we should throw it out or just pick up any idea that suits our agenda, nor that there is one form of truth that we should seek to uncover, but that we should accept that epistemology as applied to social science is far more complicated than something that can be dealt with by classical logic. There are other forms of logic, and in particular, I would draw attention to all the literature on Keynes's human logic, which is built on the premise of uncertainty. Classical logic requires certainty as far as assumptions are concerned, and then deductive logic applies. If you start with uncertainty, then that's no use. You have to develop a pluralistic kind of logic. But pluralism also applies to our understanding of the facts, and we know this because of this disconnect that's emerged between public discourse and the discourse of economists and indeed many establishment politicians. Politicians have been asserting, in the British case, British economy, strong employment levels have been rising, and that's the root out of poverty, but at the same time in-work poverty has been rising. And a lot of the reaction following the crisis, which is continuing now, is that people's real experience is other than what the experts are seeing. And this needs to be taken seriously. If indeed people's real experience is of an economy not functioning successfully and it is the goal of the government to make the economy work successfully, then there has to be some way of dealing with this question. And in the process of providing advice to governments from this kind of perspective, inevitably for all we are seeking truth as pluralist economists, we know that we're never going to reach there. The best we can do is to, from our own perspective, produce the best kind of advice we can, but be modest about it, accept that no one approach is going to cover all of a situation, particularly as the context evolves, and indeed that other disciplines can provide useful input. And secondly, there needs to be, if it is the case that economics embodies value judgments, as I would argue it does, then there needs to be much more honesty about the kind of value judgments that are embedded in any one approach. So now I want to leap back over the history of thought that we've already heard about and go back to Adam Smith, who we've heard a fair bit about already in the conference, but he has really useful things to say on this subject. Clearly the context was different in the 18th century, but in that situation where a moral philosopher was operating within a small society, economics was evolving as applications of moral philosophy, and there was direct engagement with people in other disciplines and with people in industry and less so in government because the government had moved south by that stage. But there was engagement with an audience, and Smith's first job actually was to provide a series of lectures on rhetoric in Edinburgh, publicly available, where he talked about how one communicates. Because if you can't demonstrate the truth of an argument, if demonstrable logic is not enough within an evolving open system, then we're in the business of persuasion. What a scientist or philosopher needs to do is to persuade an audience that his or her argument is a good one and no enough of an alternative argument to argue that it's better than the others. And Adam Smith was fierce in his arguments, but the starting point is it has to be persuasion rather than demonstration. So he was concerned about how people gather knowledge and how people communicate knowledge. And expertise was his first example of the division of labour before he applied it to the economy. People specialise in being philosophers. It's not that they're different from other people, and he talked about the philosopher and the porter, we heard about the janitor earlier, but he talked about the philosopher and the porter as being essentially the same, except for habit, custom and education. In other words, there's the potential for communication there. It's just that some people specialise in applying more reason. They're curious about explaining puzzles. So in trying to convey ideas to an audience, Smith talked about how a philosopher thinks in terms of imaginary machines. That a theory is like an imaginary machine. You try to paint a picture of a set of causal mechanisms and try to persuade an audience that this is a good way of understanding the real world. But people will understand that theory in terms of their own experience. So you have to communicate to an audience in terms of their experience. This is the kind of lesson that we as economists can learn, that we need to think about how we communicate to an audience. It's not that Smith said we shouldn't be experts far from it. Experts by definition have specialist knowledge. But in order to be useful in society and in order to take members of society along with us in the application of the particular values that are being input into a theory, then there has to be successful communication and for a theory to succeed, there has to be successful persuasion. Now he said something very interesting about that by contrasting natural philosophy, physical sciences, with moral philosophy, social sciences. He said that it's quite normal for theories in the physical sciences to survive for a long time, even though there's no element of truth in them. And that's because there isn't a direct connection with people's everyday experience. In the social sciences in contrast, because they're so directly connected with people's personal experience, theories can't depart from that experience for very long because they'd be challenged. So what's happened in economics? How come we got to this situation where theory has moved off in a direction which produced this disconnect between public discourse and economic expertise? Well, economics suffered from physics envy, from mathematics envy and pushed in a direction that... I mean, quite explicitly, aiming to become more like the physical sciences. In other words, more divorced from the real experience of the people most affected by economic policy, which is the citizenry. Now, one point I should just mention before I come to this final slide. Just going back to Cain's character... Sorry. Freudian slip, they thought very differently. Smith's characterisation of the difference between the philosopher and the porter was custom, habit and education. Smith had a lot to say about education. And he talked about education as being a way to protect people from superstition and running away with foolish ideas. In other words, a protection against fake news. I mean, he saw education more generally as something that required particular attention as automation became more prevalent with the industrial revolution in order to provide people with a more meaningful existence. But it was also a matter of arming the citizenry with the equipment to deal with experts. So let me just summarise where we've got to and what I mean by doing economics differently. The first point I want to reiterate is that expertise is not something to be confused with something that's exclusive, something that's done behind closed doors and only the output pops out to be used by governments and the populace. Nor is it something to be presented as having produced results with certainty. Experts need to be, and particularly in the social sciences, need to be modest about the capacity of anyone's theory to explain things. Secondly, we need to put a lot of attention into earning the trust of society, something that's been lost. I mean, this is something that's very difficult to reverse. I mean, we see this in the financial sector. But this involves taking seriously the need to have a much better discourse with the rest of the population. I mean, I commend to you the book Echinocracy, which goes into this in great detail and makes some very useful suggestions about how we might pursue this in a practical sense. On the one hand, helping provide education and economics to the general population, but also in taking the views of the citizenry seriously. And finally, we're in the business of persuasion. Not only should we be engaging in debates, but this in itself is a good way of promoting a successful form of educating and economics. Taking as a starting point the fact that no one approach is going to in itself be sufficient. Nevertheless, each of us takes an approach. We prefer it for one reason or another, but we should be able to articulate those reasons and be persuasive about their merits. And this applies to analysis, and it applies to the values that we employ, but it also applies to facts. Facts themselves are clear-cut if we're talking about a certain narrow range of things, but once we get to just interpreting data, never mind starting to impute causal connections to data, this becomes a whole area of debate in itself. So again, this is a non-binary approach to fake facts as well as fake expertise. It's not a matter of there are definite facts or there are no facts. Facts are things which themselves are to be debated. And ultimately, in terms of how we proceed, it seems to me that education is at the core of how we proceed. Thank you. I think I'd better stand up, at least because you couldn't see me, I suspect. I'm in the somewhat difficult position of agreeing with all the papers and actually liking the authors a lot. So I don't have too much critical to say. And then I have this other problem, which is saying much about anything is hard because the topics are hugely complex and we don't have that kind of time. So I think what I will do is make a few notes on various topics. I begin with my sort of ironic reflection which just oppresses me more and more every year. We live in an information age. It's obvious that we live in a misinformation age and it's confusing to me that, I mean, whence this confidence that we are actually profiting, I don't mean that in a Google or Facebook sense, from all of the abundant bits of information around us. But in any case, I want to sort of address the question of misinformation in several directions because of the fake news discussion. I mean, there are times, I think, I actually considered making an argument that economics invented the fake news back in the 1980s and I'm still susceptible to that but I'm not going to bother you with this question. I think if you look at, the one right question to pose is the historical one, is when do you slide into situations where people openly promote lies and don't care? I remind you that it was not, you know, somebody on the far right but somebody on the alleged center right. I mean, it was a spokesperson for George Bush's administration, that's W, there, who made the famous remark to some Democrats, you Democrats are still believing in facts when we've given that up and that was not the alleged extreme political party that did that. And I just actually stepped out of a panel on the Weimar Republic. Well, I was quite interested once in sort of where did all these, we know where the stab in the back legend originated but all this discussion of Jews going on on the right wing side of the Germans. So I was reading, I thought a place that might be illuminating and it was, it was the minutes of the various meetings of the All-Deutsche Verein. That was a front run by, as I recall, Alfred Hugenberg and others at the time they were in the steel industry, though Hugenberg may have already exited at the point I'm talking about that didn't mean he stopped being a right wing publicist that many bought newspapers and became enormously influential. Anyway, the, and this guy, you can hear, they're all arguing about what to say and somebody says it was the Jews and then some guy says no it wasn't and he says it doesn't matter, say it anyway. Just like that. And the point is, is that if you get people highly roused, if you like, passions are running high I do think that as passions rise whether we like it or not, the willingness to tolerate open garbage on the other side just goes way up. And so the first thing I always tried to do when I'm trying to figure out how tolerant are things of fakes is let's take the social temperature either, you know, maybe it's generally in the period or locally in the place because I am sure that's a big, people just won't say, they're willing to swallow a lot of stuff if they're really, if you like, angry at the other side. Now beyond that, on the general question of expertise I think if you discover this, it was pretty easy to become a lawyer say in America in the 19th century all you had to do was go read the law for a while with maybe find somebody to take in. By the early 20th century all, just about every form of expertise there is in fact if you want to be an expert, you've got to be credentialed and the capital costs of that stuff rise a lot. And I fear that that tends to make, not only does that, they may think they're entitled to higher incomes maybe for good reason, but it also tends to reorient the whole profession toward money in an amazingly direct way. And I have to say that I think the root of every evil in modern expertise is pretty much money more than anything else. When I came out of Princeton University as a young PhD I'd been working with William Baumol, a genuinely great economist, perennial candidate for the Nobel Prize Winner. But I knew two things. One was that he'd done a lot of work for AT&T and I had a friend named Herman Schwartz who was with the Federal Trade Commission. And Schwartz went to debate Baumol on the merits of the AT&T breakup because Baumol was defending AT&T along with a whole string of other Princeton folks and some non-Princeton people. And so they had this long argument all night and Schwartz later told me, well, after the CONFAB Schwartz and Baumol are talking and Schwartz says, look, you can't really believe there are economies of scale in AT&T anymore. And Baumol says, I didn't say there were. I said, if there are economies of scale. I mean, people made lots of money on that kind of fudging. And frankly, this is rampant in... I mean, there's no reason to just pick on economics. I could say law. Balzac's at a lawyer was a conscience for hire. And I mean, people have always picked up on this. But if you credentialize guilds or whatever and then they don't have to pay attention to facts and you pay them that, you can expect stuff to go terribly wrong if only because people aren't paying attention to stuff. And thus, if a large chunk of a modern population is sitting there saying, facing huge, basically stagnant wages for 30 years, while everybody with Nobel Prizes is writing in the New York Times that, well, this is really great. 2016's a wonderful year because we've finally made up our... We're all back from where we were in 2008. If you agreed with that Nobel Prize winner who wrote that, you did, in fact, vote for Hillary Clinton with a very high probability. Unfortunately, not everybody did agree and they weren't all in that kind of a position. That type of mistake is made over and over before we even get to the kinds of financial consulting stuff where you've got, as we all know now, idiots telling you it's fine to use a normal curve to predict stock price problems if you're a bank or something like that, looking for how much capital should you hold. And the truth is the economics profession disgraced itself in 2008. And from where I'm standing, it hasn't done a whole lot to rectify it. It's just amazing to me. I mean, look, I'd say, well, take a real case. One gentleman, I guess, I don't know if I will not use the name now, but in 2008, he actually argued seriously that what destabilized the whole world was not the breakup of Lehman and the AIG. He tried to argue that saving AIG did nothing to stabilize the financial situation. I'm not making a cut. I think those people should all have been fired. My slogan has been for many years, rescue banks, not bankers. But this guy then argued that what destabilized the world was when Paulson and Bernanke went to Congress. I mean, he actually wrote this in an article. I picked up on it in print. And then later he met my co-author and he said, yeah, you convinced me that was stupid. When I checked his textbook the other year, that example was still in there. I mean, this kind of insanity and double-talking runs all over the place. When I read that Google wants to require you, if you're going to do political ads, to sort of put all the political ads in one place, I'm thinking what a great idea. Maybe INET should really just build a sort of little panel thing where every prediction by economists gets stuck up where everybody can look at it. But I'm just ingesting, of course. Now, I don't want to take a lot of time. I think we go till five, but it seems to me we hear from you, not from me. But just a few remarks on Rob's presentation. I think I liked it a lot. I read the report. I want to address the fake news question just very explicitly, though, on one point. First of all, what that great graph does is show you, I mean, you just look at that, you know, the big colored one with all the money, and you quickly realize, look, this stuff is essentially all homegrown. It's not a Russian import. I mean, you really got to take that point there. And whatever people are doing, I mean, I myself think it's obvious the U.S. does all kinds of things in other countries, and other countries do all kinds of things in other places they can do. And so I would be a little surprised if the Russian, let's say, cutouts government, there's a problem about responsibility here and who did what if they completely set out the American election. But what you're actually getting here is a discussion that first bypasses the key point, though it is made in your paper that this is not a, fundamentally not a foreign story at all. This is made in America in terms of the whole panoply of multicolored spots. Beyond that, though, if you look at the evidence that's now floating out, I accept that Russian cutouts were probably doing the things they're said, but I suspect you'll find a few more. I don't think you've seen the limits of that, but it's a drop in the bucket compared to the domestic stuff. And in particular, like this Oxford Martin study that's out just a couple of weeks where it, what it did is took a sample of Twitter in the last three weeks, two weeks of the campaign, and then said they defined some as Russian, though they knew perfectly well there weren't, I mean, there was a lot of other stuff in there that had no Russian origins. And then they run, and then they say, well, those tweets from those were disproportionately concentrated in battleground states. And the test they use for that is actually, it's basically a sort of difference in, it's not technically a difference in means test, where they look at how many of these battleground states were above average in tweeting targeting versus below average, and they find there were some more. Well, you know, if you put a stronger test on that, you can see how poor it is in the sense that if you're really concentrating on swinging the election, you need to concentrate on battleground states. Let's just accept their characterization of it. And if they do it, I think my calculation was they got seven out of 13 or seven out of 14 right. If you treated this a very special case of an exam in American Studies, they'd clearly failed. And it's clearly bad targeting. And then Jonathan Albright, whose work I like a lot, and he has made precisely the point I made a couple of minutes ago on fake news, that it's, this is a local origin story, basically. He had a piece out where he looked at, he used the thing that Facebook immediately eliminated, a way to go back into data that passed ads and things. And he showed you that, well, there were a bunch of ads in there that probably were from Russian cutouts, if not from Russia, or at least might be. The only problem with the argument as far as it went was it didn't go very far. Why? Because almost half of them came out after the election. And, you know, if you're, look, this is not the way serious people target very well. I mean, I frankly, you know, I, myself, living in Boston where it turned out later the FBI and American authorities had all kinds of warnings from the Russians about the guys who actually did the, that April bombing with the marathon. I kind of feel that I'm not surprised when government bureaucracies try to target and don't do very well. But I actually think the, the available evidence here says, hey, this is a very low-level deal. It looks to me more like experiments than anything else, though I, I don't know that and I don't claim to. But the, the, I do agree with some folks who suggest that we have a real problem, the cover logic of Phil's about Facebook trying to preserve its, if you like, market flexibility. This is getting very interesting, though, because, you know, neoliberals just basically serve in the end, you'll forgive my plain English, the interest of money, specifically big money. And there are a lot of folks walking around saying the deplorables have really been just too much. And I am pretty sure Facebook has been experimenting with algorithms to de-emphasize stuff they think is on both the left and the right. And I can't prove that. I have talked, Don Hazen at Alternet has a dramatic graph of how their traffic tailed off after a change in a Facebook algorithm. And I am hearing, though, I have not yet seen similar experiences on our right-wing websites. And the question, and you can see inside Facebook and also outside, I thought there was a piece just quoted the other day in some major story, where folks say, well, we don't really want to get in the business of choosing between forms of news. I actually think this is a very tricky issue. Yeah, I don't think I want Facebook trying to do that either. And I don't want it doing it particularly with the National Endowment for Democracy advice, which I think that seems also, at least in the past, have been going on. We are, unfortunately, living in really strange times. And I think that Sheila's advice for some careful modesty and understanding you could end up being wrong, and pluralism opinions is very sensible. Thank you. So we're going to, I guess we'll do some questions. And if I can, as running this panel, I want to throw out a couple of questions myself. Because I work in International Business Times and Newsweek, and I work in a newsroom. So I'm actually living some of the things that were discussed on a day-to-day basis. One point I would raise about the point about beats, about a lack of beat reporting that was mentioned, I would add to that, that on top of that issue, which is a real issue, that is a lack of journalists, or a diminished number of journalists who have the ability and autonomy and professional wherewithal to be focused on a set of issues, there's also a problem with a somewhat deprofessionalized journalism profession that adds to the problem of fake news in this way. Which is, you may not be an expert on X, Y, and Z issue, reporting on that issue. All of us who come up in journalism, when you start out, you have got to report on everything. But you have to follow a set of standards that really is one key giveaway to when you're, the difference I think between a real news story and a fake news story, is if, as a reader, you can see those standards in action or you don't see those standards in action. And we're talking about basic stuff here. You go to the other side for comment. You explore multiple sides of an issue, as opposed to only giving one side of an issue. So this is a way, I think, to detect fake news or to distinguish between news and fake news, and that it's not just beats, it's that a lot of times those standards are increasingly not being followed at all. The question I have about much of this is when we go from, in my mind, there's fake news, which is obviously made up lies that are simply not true about facts that are not debatable. There are a certain set of facts that are either true or not true. Where it gets dicey, I think, is in the areas of facts that are marshaled for a message, facts that are marshaled for a partisan point, facts that are marshaled for ideology, where the news itself is not fake, but where the way it is presented may be driving an opinion. And what I wonder about is, because that, to me, seems like at least as vast, if not an even more vast universe than just pure fakery. And the question I would ask to anyone on the panel here is, it seems like we're at a conference that's about, for instance, in the economics profession, is about, in part, how bad assumptions, maybe not fake assumptions, but bad assumptions are oftentimes baked into the center of the economics profession. And oftentimes I find when you read news stories about politics, about economics, about business, there are equally bad, arguably discredited assumptions baked into the news coverage of that from the outlets that are cited as the non-fake news outlets, that the reliable outlets. And I don't want to name names, but the big outlets that are driving much of the news coverage, oftentimes you'll read a story in there about a piece of business news, a piece of economic news, in which assumptions that are not predicated in reality are baked into the assumptions of the story. I mean, trade is a good example. Free trade is good, has been baked into 30 years of legacy media coverage about economics for 30 years. And so the question is, how do you distinguish between fake news and that? Is it fair to assume that what you're getting from legacy news outlets isn't fake news when it may not be fake news, maybe in that gray area? And isn't the rise of partisan or ideological media content, isn't it in part, for better or worse, a reaction to a public that senses that a lot of what they're getting may not be purely fake news, but at least may not be telling the whole side of the story? So anybody can weigh in on that. I'm happy to start with that. So I agree with you that the fake news, as in completely made-up stuff, is a very, very small part of the problem, and that the larger part of the problem is where things are twisted or the facts just aren't the facts and they're reported as such. We know kind of the old-school way to try to address that, which is journalism practice, having the institutions of journalism itself police one another and have people issue corrections where they're wrong and do that. That's not a perfect system. I think it's better than what we have right now. The problem is where news gets tangled up in partisanship. It becomes that much more difficult. And despite the flaws of the traditional media model and where they got things wrong, the current situation is where you have half of the news playing by one set of the rules, which is let's be objective, let's chase down the facts, where the other half is being just blatantly partisan and has no regard for that. And that seems to me a particularly unstable situation. I'm not sure what the answer to that is. Okay. I think we can do questions. I think the easiest thing, because I think we want to be on the mic, or I guess, I mean, if it doesn't... Small off, you could repeat the question. Yeah, I can repeat the question. So you just raise your hand and... Yes, I certainly agree. So I guess the question is it goes back to this question. Yeah, go ahead. Just one very quick comment. I mean, Phil is making me rethink my position, but until now my line has been fake news. Probably was invented if we had the evidence back in Egypt or Samaria, but for sure in London coffeehouses. But my serious point is the easy way to do fake news, the mainstream media way, is what beautifully summarized in that Edward Hall at CARSA, what is history from the late 30s, where he just said, look, the historian's selection of facts is really crucial. I would say that the New York Times selection of facts is really crucial. And that's not accusing anybody of faking anything. And if you go to them and say, you said there was X and it was, in fact, not X, they'll fix it. But you know what? You can make an awful lot of disastrous errors on stuff that they don't consider worth talking about. Yes. Anyway, go ahead. Look, I think basically you're right on that point that way that, however, emerges is pretty straightforward, which is a classic one of where we run out of time and the ability to sort of mount enough of a program. And I mean, we did preface this with a three-day young scholar's, not gonna call it, was a sort of festival. Festival, thank you. Yes, where there was a direct attempt, surely, for people to be heard. I agree it's a problem here. I'm not trying to sugarcoat the issue. Yes. The history of cybernetics is a big, long argument in and of itself. I've done some stuff on it. Other people have, too. I think that, I mean, here's the short version, that while there was an attempt to kind of have a socialist cybernetics, I think it kind of went awry sometime in the 70s for whatever, and it would be interesting to talk about why that is. I'm just observing at this stage. Now, instead, what would be counter about this? Well, one thing is I hope people on the left have learned that Twitter is not the way to organize a political movement. However you felt about Occupy, however you felt about, you know, the Egyptian revolt, so forth and so on, it didn't succeed. And I don't hear people staring at that very carefully. Although there are one or two people. I mean, I think of Zaynep Tafakshi, I think that's how it's pronounced. She's actually gotten interesting lately about how she first praised it, but now she is having second thoughts as well about how one would organize a political movement. So, I mean, it's not like no one is thinking about it, but I think the ground idea is somehow social media is going to allow us all to, you know, more easily politically organize. I think people should learn their lessons. That's not what it's about. Yes. Wow, that's a good question. I mean, now in a way is the best moment to try it, because there is, I mean, I would say the economics is still quite vulnerable in the wake of the crisis. I mean, the mainstream has evolved and has changed and, you know, has addressed to some extent the issues that have arisen. But I think that there is such a public challenge still to the way in which economics is normally portrayed that now is, if anything, a good time to try to... Even at the level of opening up public discourse and alternative approaches to economics, I mean, you were talking about the difficulty for journalists being presented with theories and all sorts of stuff being embedded in it. But there are all sorts of different theories and there's all sorts of scope within economics for debate about, you know, which assumptions are critical and should, you know, what are the relative merits of different approaches. And if that was more widely observed and that might be picked up by journalists and become much more known about in wider society. I mean, that clearly relates to the sort of push for a change in economics education and the iconocracy idea of democratising economic education. So, I mean, given the difficulties of, I mean, sectional interests, whose interests are served by being closed and non-transparent, at least there would be some more open challenge of that. So, it might be... now presents us with opportunities which weren't there before. Yes. That research on girls' topics for almost 30 years, social cybernetizens are the secretary of social cybernetizens and social associations. So, I want to be very optimistic in this case. There are responses and then new economic thinking. So, what would be the response? I'll just present some issues which should be given attention. First of all, a new interdisciplinary that could save you problems with information. Then social constructivism, but also including a deeper understanding of mathematics, not only mathematics. So, this interdisciplinary should be much broader. Then we have problems with that. What is information? What is knowledge? When it's clear, look at this, we have to look at the ignorance. So, I'm just doing research, all these I could just share in the later. Then what do we have right now in the era of information? What does the department of assigning them mean? Next point, knowledge is hierarchy. I know that, I know that, I know that, I know that, I know that. And sexual ignorance has only two levels. In economics, we have a beautiful example of Robert Allman, common knowledge and formalization of common knowledge. And then it's a problem of the meaning. At which level of knowledge I am operating, you know? If you don't know the dialogue, I don't know the dialogue that you don't know what I'm truly talking about. So, then in social life, we have both of them, we have simulacra. You know, simulacra, a simulated reality of different levels. What do they have? So, in this problem, when you look carefully at the hierarchical knowledge and common knowledge and simulacra, then I would have a very interesting question, what is truly the phrase because meaning is created in an inter-subjective creation. We negotiate meaning according to behavioral sciences. So, it was a short kind of scope of a very optimistic direction of research in the field of information knowledge and ignorance. I think we have time for one more. Yes, sir. The entire careers of people writing things that never get cited. And so, I'm wondering as a professional and particularly thinking in terms of university and academic economists, what can we be doing to think more creatively about what it means to be an academic economist? I mean, one thing that happened within the British system is that, I mean, there's been this research assessment exercise which is the kind of thing that drives all this sort of over-publication, you know, which leads to lots of unread research. But there was a change at one stage to incorporate an element of public usefulness. I'm a bit apart from the system now, so I can't really talk to it. But it was a sign that public bodies were perceiving that there was an issue about the way in which economists, sorry, academics in general were connecting with society and performing a useful function. I mean, you can imagine the metrics of that or I mean, it must be quite a nightmare to put together. But it is a positive sign. And I mean, governments put a lot of resources into university education. And in a way, they're possibly fed the view from academics themselves that things are useful, which are not, in fact, the most useful things for society. I mean, it becomes a matter of the separability of expertise again. But I'm just going around the problem rather than suggesting the solution. But I mean, the only solution I can see is to sort of push on the public discourse angle and just build on that. And that in itself, hopefully, would feed through into changing perceptions within academia and also within government. Could just one quick comment. In the case you're talking about, lots of papers written to have been written not to be read. The red, it is, you could change promotion criteria somewhat. I mean, a lot of that stuff, I mean, you will see, I better stand up because I'm just behind this thing. If you actually look at these sort of spurts of publication by people, you see they are heavily concentrated in their pre-tenure year, and then sometimes when they need some requirement later on, it doesn't take, it wouldn't take much to have committees, though you'd have to make them actually read the publications, which is what the entire citation system is now designed to block. I mean, it's to replace systems of reading. Actual reading of the publications. But you could insist they read them, and where they're obviously garbage, well, you wouldn't promote them. I mean, that doesn't seem, and beyond that, you actually have plenty of room inside existing universities for, if you like, step processes that don't necessarily reward sheer longevity. A problem is, I am quite a supporter of unions. I was three times the head of, not the head, I was on the board of the local union in Mass Boston, and I think we could not function without them given the level of the state administrations there. But, you know, where you insist that everybody be rewarded on a step basis, you know, whether you do anything or not, you're obviously going to produce a system of junk. I mean, you can surely find some via media between just letting everybody sit on the vine but actually getting them to do something. I mean, when they put in what was a deliberately meaningless system of post-10-year review, I watched people who hadn't written anything in 20 years write something. Thank you all for being here. I really appreciate it, and thank you to all the panelists.