Democracy posits the radical idea that political power and legitimacy should ultimately be found in all of the people, rather than a small group of experts or for that matter arbitrarily-chosen hereditary dynasties. Nevertheless, a good case can be made that the bottom-up and experimental nature of democracy actually makes for better problem-solving in the political arena than other systems. Political theorist Henry Farrell (in collaboration with statistician Cosma Shalizi) has made exactly that case. We discuss the general idea of solving social problems, and compare different kinds of macro-institutions — markets, hierarchies, and democracies — to ask whether democracies aren’t merely politically just, but also an efficient way of generating good ideas.
Support Mindscape on Patreon.
Henry Farrell received his Ph.D. in Government from Georgetown University. He is currently the Stavros Niarchos Foundation Agora Institute Professor of International Affairs at the Johns Hopkins School of Advanced International Studies. He was the 2019 recipient of the Friedrich Schiedel Prize for Politics & Technology. He is a member of the Council on Foreign Relations and co-leader of the Moral Economy of Technology initiative at Stanford University. He is a co-founder of Crooked Timber blog, as well as the Monkey Cage blog at the Washington Post.
- Web site
- Johns Hopkins web page
- “Pursuing Cognitive Democracy,” Farrell & Shalizi
- Crooked Timber posts
- Washington Post columns
- Wikipedia
[accordion clicktoclose=”true”][accordion-item tag=”p” state=closed title=”Click to Show Episode Transcript”]Click above to close.
0:00:00.2 Sean Carroll: Hello, everybody, welcome to the Mindscape Podcast. I’m your host, Sean Carroll. And on the podcast, we’ve talked about democracy quite a bit in different contexts. We’re gonna keep talking about it. I think it’s a very important, as well as fascinating topic. But usually, when you think about democracy, when you have a discussion about it, there’s different angles you can take. You can talk about it from a political point of view, obviously, historical point of view as we’ve done, philosophical point of view. The justifications for democracy are generally that it’s the right thing to do, it gives voice to the people who are in a country. But what about the question of: Is democracy any good at achieving whatever goals the people in the country actually have? So this is almost a scientific point of view of democracy or at least, sort of instrumental, computational point of view. Is democracy good at problem-solving? That’s the question being asked by today’s guest, Henry Farrell.
0:00:55.0 SC: Now, Henry is a well-known political theorist. He writes a lot about, not just democracy, but international relations and so forth. But with his collaborator, Cosma Shalizi, he’s been thinking about democracy as a problem-solving mechanism, the cognitive aspects of democracy. And basically, what you can do is think about democracy as one kind of emergence. When we talk about countries, we often anthropomorphism them. We say this country has these desires or these goals or something like that, a little bit sketchy because a country is not a person. But in some sense, countries act as if they are agents that emerge out of all the people that make them up, and there’s different ways that the viewpoint of a country can be instantiated, given what its people want. You could have an autocracy where there’s basically the people have some way of giving all the power to one person or some very small number of people. You could have markets, in some sense. So you have different sets of people who are bidding against each other. And of course, you can have democracy where, in some sense, everyone is voting, expressing a preference, and then you sort of either take a majority or you have some mechanism for turning those individual votes into preferences for the wider country.
0:02:10.0 SC: And so what Henry is gonna argue is that democracy is actually really good at solving problems, sort of surprisingly good, in some sense, not perfect, but these other mechanisms can get stuck. These other mechanisms of taking the views of many, many people and converting them into a single action can get stuck in ruts, can miss all these solutions because it’s not as creative as a democracy can be. And like many Mindscape discussions, even though Henry’s a political theorist, he takes a scientific point of view on this, so we have analogies to not only physics, but also, evolutionary biology, and things like that. So it’s an interesting angle on an old question, democracy how it works. Let’s be science-y about it, let’s think about how this actually happens in the real world, so let’s go.
[music]
0:03:11.0 SC: Henry Farrell, welcome to the Mindscape Podcast.
0:03:13.2 Henry Farrell: Wonderful to be here talking to you.
0:03:15.2 SC: So let me start by asking you a question so that I can better understand the landscape of political science/government academia, which I don’t know that much about. It seems reading your stuff, obviously, you’re perfectly respectable political scientist. You write about international relations and institutions, and things like that, but it seems to me that you are more sympathetic than the average political scientist who’s thinking about these institutions like a physicist would think about them or like a scientist would think about them, as dynamic, complex systems, interacting in different ways with equations and the whole bit. Is that an accurate gauge, or is there a whole discourse going on out there that I just don’t know about?
0:03:58.4 HF: No, that is right. There isn’t nearly as much discussion of this stuff in political science, as you might expect. So we are seeing people who are beginning to start to use machine learning techniques of one sort or another, random forests and similar types of things for extracting meaning from data. But in terms of actually thinking about political processes in these ways, which is something that I think is really important, we really don’t have nearly as much as we ought to have. There is one wonderful semi-forgotten political scientist from the 1960s, Charles Lindblom.
0:04:29.4 SC: Oh, yeah.
0:04:29.6 HF: He thought a lot about the world in these kinds of ways. And of course, political scientists are aware of the work of Herbert Simon, even if they tend to focus not on the areas of Simon’s work that really talks these questions, and much more on his ideas about satisficing, more generally. But we don’t have that kind of a conversation that is really yet to get going. And I think the reason why I am interested in this is because of an experience we both share of having been around in the dawn ages of the blogosphere when there were so few people actually writing and thinking as academics about this stuff, that you begin to have these different social networks. And I think it was through talking to people like you, talking to people like Cosma Shalizi, who has become a regular partner in crime at Carnegie Mellon, that I started to get interested in these questions, and then getting exposed to a lot of people in Santa Fe and other places that are also thinking about the world in these kinds of ways.
0:05:25.7 SC: That’s excellent. When people ask me, “Did the blogosphere, in its heyday, ever lead to anything good?” My best example is that I found my wife through the blogosphere since Jennifer and I were both bloggers. But you have a much more academic intellectual reason to give you credit, which is good.
0:05:43.0 HF: And I also found my job. My spouse was living in DC at the time, and was very enthusiastic for me to get a job in DC. And then one day, I got an email from somebody who later became a colleague of mine, inquiring gingerly, I was at University of Toronto, whether I wanted to apply for a job that they had going, and I jumped on that. And my spouse promised she would never complain about me blogging again, and she has mostly managed to keep that promise over the intervening decades.
0:06:11.0 SC: This probably has lessons for the spread of information over networks and diverse opinions, and things like that, which we’ll get to, good. So even though we’re chit-chatting now, it actually does have to do with substance of the conversation. So let’s start with this idea of democracy and other kinds of institutions in a more or less theoretical sense. Probably the thing that is gonna guide the conversation the most is the paper that you wrote with Cosma Shalizi on cognitive democracy. And I take it that the idea is to think about democracy as a way of making decisions and to compare it with other institutional ways of making decisions. Is that fair?
0:06:48.8 HF: That’s right. So we can think about markets, we can think about democracy, we can think about hierarchy, we can think about all of these different modes of problem-solving. And what I mean by problem-solving here is something like the following: So imagine that there are a number of us and we have some shared problem in common that we would like to solve. But either A, it is a complex problem that the solution is not immediately and readily apparent; or B, we have strong disagreements about how to solve it. I think that is where we can start to begin to get some traction on which modes of governance, whether these be markets, whether this be democracy, whether this be hierarchy, are better or worse at solving individual problems.
0:07:30.9 HF: And so I think that the argument that Cosma and I have made there, and I think we would probably modify it some. I think we’re a little bit too enthusiastic about democracy as against its competitors in that piece, but I think that the argument that we make is that the ability of democracy to solve problems is underestimated because people don’t pay attention to the ways of which democracy forces people, ideally, with very different perspectives, very different wants to work together. And sometimes, from the very diversity of our goals and our wants, information emerges and the democratic process works best when it is actually able to harness that process, and to turn that information into something useful and something actionable to help us solve problems.
0:08:14.0 SC: And so just to put the audience in the right mindset here, when we think about democracy, most of us are gonna be thinking about the government, the society in which we live in, it’s a system of governance. But we can think about it more abstractly. I take it that your view of democracy works when it comes to governments, but also, just other decision-making processes, more broadly.
0:08:38.6 HF: I think that’s right, and this is work that I’ve been doing more recently together with two colleagues of mine, Hugo Mercier, whose work is mentioned in the Confidence with Democracy piece, and Melissa Schwartzberg, who is a wonderful political theorist at New York University. And part of what we’ve been trying to do is come up with this research program that we call No Bullshit Democracy. And the idea behind it is as follows, that there is a lot of language being used by democracy, and so democracy is awesome, according to its defenders. Democracy is terrible according to a variety of very influential Libertarian critics who popped up over the last decade or 15 years. And what we want to do is to try and to change the conversation a little bit to say that there are all of these different forms of democracy, these different forms in which democratic decision-making can happen. Some of them work well. Some of them work terribly. Let’s start to try and figure out a more scientific way of understanding how to do this.
0:09:33.1 HF: And because Hugo is a cognitive psychologist and because we are really working from that micro level, a lot of the argument we make in the No Bullshit Democracy paper is really saying we need to build up from these insights that cognitive psychologists have developed about when group problem-solving works and when it doesn’t, to start trying to figure out: What are the parameters under which democracy work well or badly? But equally, one could imagine that a very similar research programme and a complementary research programme working downwards from some of the work that statistical physicists and others have been doing on contagion over large-scale networks, and trying to use that as ways of modelling, have these much bigger social processes, which are not small groups, but in fact, are taking place at a much higher level of social articulation, how those can have both good and bad facets, too.
0:10:21.6 HF: And again, to try and push this kind of insight into a feedback loop between, on the one hand, expiration of the theoretical possibilities; and on the other hand, try to figure out what works practically and what doesn’t because of course, you can theorise all you want, but there are going to be major facets of actual real-life democratic politics that you aren’t going to get until you see what works in practice and what doesn’t. And this is the kind of debate that at least, I think, Hugo and Melissa and I would like to see happening, and of course, Cosma and I, as you say, have talked about this in a slightly different space earlier. And there is a lot of work that needs to be done.
0:11:00.7 SC: Yeah it’s very much in line with the spherical cow philosophy that physicists have. Let’s do the simple thing first, then we can add in the complications later and hopefully, that doesn’t change everything about the problem that we’re working on. But you said, before we get back to that train of thought, you said one little provocative thing about there are Libertarian critiques of democracy. And maybe you could share with us what those are ’cause I think that at the very naive level, we might think of Libertarians as being maximally democratic, but are they worried that the majority rule is gonna interfere with their rights? Is that the basic worry?
0:11:35.2 HF: So I think that that probably is their basic worry, but they proceed from a different direction. So here, I’m thinking about three books which have gotten public attention over the last 10 or 15 years; one by Jason Brennan called Against Democracy, which is a little bit of a, if… He’s a Business School Professor and trained as a philosopher at Georgetown; and is a bit of a shock jock book. He is very clearly trying to be maximally annoying to the people whose opinions he dislikes. And there are also books by Bryan Caplan, who I think is doing this from a less provocative, but nonetheless, extremely strongly-worded perspective; and Ilya Somin, who is a George Mason University law professor. All of these are arguing, more or less, that democracy is doomed to failure. And democracy is doomed to failure for two reasons. First of all, because individual citizens are ignorant, that is that they do not have the kinds of political knowledge that they would need to have in order to take good decisions. And secondly, that they are hopelessly cognitively-biased, that my side bias, and a variety of other biases mean that they simply aren’t able to think straight about politics.
0:12:44.2 HF: And so we have these authors proposing different alternatives to democracy. So Brennan proposes what he calls epistocracy, which is a kind of an aristocracy of those who know better, and he has different flavours of this. Some involving a pretty strong rule of an elite; others involving elites having greater voting weight or whatever. And we have Caplan arguing that we need to rely much more on markets. And Somin saying both markets and the possibility of exit, that is the possibility if you don’t like a political arrangement, to be easily able to leave it and go for another, as being different alternatives to democracy. Somin says that we should limit democracy, and the two others, I think, are looking for much more radical pushback against democracy. So these are the people who are, in a sense, disagreeing with this perspective and disagreeing from a point of view which says that democracy, as a problem-solving system, is doomed because effectively, the citizens don’t have the knowledge or the capability to judge that they would need to have in order for democracy to work well.
0:13:50.7 SC: That point of view has seeped into the practical discourse a little bit just over the past couple weeks. I’ve seen claims in the space of voting rights, certain… Let’s say one political party in our current system is trying to expand voting rights, and another one is fine to shrink it a little bit. And there’s been little tentative efforts to say, “Well, we shouldn’t count votes equally among people who are equally educated or equally informed. This is not how our democracy actually works right now,” but I’ve at least seen some people on one side of the ledger bring up that argument. I don’t know if it’s directly related to the academic discussion you’re talking about.
0:14:30.8 HF: Certainly, there are linkages and connections. It’s very, very hard to say that somebody is saying this because they’ve read this, that, and the other. And of course, there is a very, very long history in US politics of voting restrictions that are nominally supposed to be based upon knowledge of one sort or another. So that when you look at the pre-Civil Rights cells, you very often had tests being administered to voters, which were very deliberately specifically intended to exclude African-American voters and to make sure that they had no chance of having a say. So there is this long-lasting independent tradition of looking for spurious justifications from stopping people from voting if you worry that they are going to vote in ways that you don’t want them to vote.
0:15:14.1 HF: And this very often tends to be on the conservative side of the political spectrum because if you’re conservative, you’re somebody who believes in property. And one of the great worries of the Conservative movement, as Daniel Ziblatt and others have discussed; Daniel Ziblatt is a very, very interesting political scientist, is that Conservatives have always worried that as the franchise expands, people will use the franchise to vote the rich out of their property. And so equally, if you are a rich person, you are in the market, perhaps, for ideological justifications for why it is that we ought to restrict the franchise. So there has always been that tradition there, but I think that what these people do, in particular, Brennan’s book, I think, has lent a patent of respectability to these arguments and really, has given a sense that there is some intellectual weight behind them. I frankly think that this is an extremely bad book in a variety of ways, I think it’s poorly argued. It hopelessly misrepresents much of the cognitive science literature that it purports to draw upon. And of course, this is a topic that Hugo, Melissa and I are interested in talking more about.
0:16:23.4 SC: Very good, yeah, it’s good. Sometimes, bad books can lead to the good things, the response to them is productive. Not always true, but okay, look, it’s very clear that we could have a six-hour long conversation ’cause I’m fascinated by all this stuff that you’re saying, but let’s retrench a little bit and get back to this question of using democracy as a decision-making tool. Before we get into comparing democracy with markets and hierarchies and things, let me think about what that means. Democracy is both a way to not just make a decision, but also to decide what it is we value. Typically, a decision-making process, if I’m a computer scientist and writing an algorithm, I know what my fitness function is or whatever that I want to maximise, and I just figure out a way to do it, but democracy mixes that up. I’m deciding what I value and deciding how to get it at the same time, so it’s a little bit different, maybe.
0:17:18.0 HF: There are a whole lot of interesting differences there. And I think there’s a very interesting… I was listening to a podcast a couple of weeks ago. This was the Quanta Podcast with Steven Strogatz. And he was interviewing Moon Duchin, who is a mathematician at Tufts. And she said, and I think this is absolutely right, that democracy is not an optimisation problem. And this is, I think, a temptation that a lot of people with engineering or mathematical backgrounds have, is to think in a certain sense, “Well, why is it that people disagree? If we could just come up with some sweet engineering solution, everybody would realise, and we would find some kind of optimal solution”. There are a number of reasons why this is not true.
0:17:58.9 HF: One of those is what you’ve already touched upon, which is that we often figure out what our goals are in the process of actually seeking after them. And this is an argument that political theorists such as John Dewey have made at length. So I think that the best articulation of this is a wonderful, albeit, I think also flawed book that Dewey wrote back in the 1920s called The Public and Its Problems, where he conceives of democracy as effectively a means of trying to figure out these broader problems that we have and to… But in the process of discovery, which both involves ordinary members of the public and experts. Ordinary members of the public understand how the problem affects them in their lives, and experts perhaps understand the more subtle causal chains that mean that…
0:18:47.1 HF: Public has a shared problem. So first of all, a public identifies itself, and then the public tries to create the means through which it can actually address the problem. But the implication of all of this, and Dewey’s a pragmatist, is that this is going to be a never-ending process of discovery where, in trying to solve the problems, the public will, of course, figure out that some of its goals are appropriate and some of its goals widely misconceived the nature of the problem and so, there’s going to be an endless process of revision, and that is something that is crucial and important.
0:19:21.1 HF: The other part of this which is something that Dewey, I think, is bad at, is understanding how it is that people, in a democratic society, have very different goals. We are tossed together in American society, and I think this is part of the issue that we have at the moment with polarisation with people who have very, very different goals, very different understandings of what the shared outcome ought to be, what our goals ought to be, what kinds of values ought to guide those goals. And so democracy then becomes a process of group warfare, to some degree, when it works badly, or group accommodation where you work together and you sometimes figure out messy and painful and agonising solutions, which don’t please ever anybody, but which, nonetheless, sort of allow you to continue on together in relative peace while addressing some of the problems that everybody can agree are problems or whatever.
0:20:15.9 HF: And so this is a really messy process, but it also is not an optimisation process; this is much more like a… So if you think about it in slightly more abstract terms, you’re searching across a rugged solution space where you do not know what the solutions are. You know that some solutions are going to be better than others, but you want to make sure that you don’t get trapped at some local peak, which is much lower than the peaks that you might be able to achieve otherwise. And so then, you have all of the really messy and complicated processes of actually figuring out what kinds of modes of decision-making might or might not be better able to guide you to the better local optima.
0:20:57.5 SC: And maybe it’d be useful just to back up even further because this notion, this metaphor, ’cause that’s what it is, of a fitness landscape is, I guess, borrowed from evolutionary biology. Physicists used the landscape metaphor, but we just reverse the sign of the function because we think that balls and water flow downhill and you end up at the bottom of the fitness landscape. Whereas, in evolution, we think that animals like to climb mountains, for some reason, so you go to the top of the fitness landscape. But pretend I’ve never heard of what a fitness landscape is; tell me how you think about what that metaphor actually does for us.
0:21:32.0 HF: Well, like all metaphors can be misleading. And I think one of the problems when you try to apply these evolutionary metaphors too easily to political processes is that trying to figure out what the actual fitness function is so that you’re trying to achieve. And this is, in itself, now an obvious task. This is not something, it was especially given to people who are alike and have strong disagreements about the goals that they are supposed to reach. And I think it’s a useful metaphor, but I think it’s one that one shouldn’t push too far. That said, what… So Cosma and Danielle Allen, who is a political purist at Harvard and is also currently a candidate for… She is running for governorship of Massachusetts.
0:22:18.5 SC: A political practitioner as well, yeah. [chuckle]
0:22:19.0 HF: So yeah. And is an all-together wonderful person. I have got to say I’m extremely lucky in my collaborators, so all of whom are terrifyingly far more brilliant than I am. But what we have tried to do there is try and use this evolutionary, this notion of evolutionary fitness states to try and think a little bit more clearly about democracy and more or less, an undirected process of problem-solving. So the idea here is something as follows, that if you think about… And here, I’m simplifying radically.
0:22:55.2 SC: Yeah.
0:22:55.3 HF: That if you think about how democracy works versus, for example, systems such as autocracy, so you can think about them as drawing upon possible solutions to problems that a society faces. And here, again, I’m abstracting away all of these difficult and complex problems about, “Do we agree what a solution is? Do we even agree on what a problem is?” etcetera, etcetera. But if you start from that kind of metaphor, then you can reasonably see that some solutions, which are likely to pass muster in a democratic space because they are to the benefit of the collective majority, are probably going to get locked in an autocratic situation or in a democracy where there are extreme power disparities because some of these solutions, even if they are overall beneficial for the society, are going to be uncongenial to the powerful minority the elite was in that society.
0:23:50.0 HF: And so what that suggests then is that under reasonable circumstances, we can say that democracy is likely to have a dynamic advantage, vis-à-vis totalitarian systems, and its ability to search this landscape for possible solutions for problems that pop up. And we think about this primarily in terms of problems of institutional change, finding new rules because there’s a literature there that we want to talk to. And one could also say the more equal democracies, ipso facto, and holding all things equal, are probably going to be better at doing this, at searching for good solutions than less equal democracies precisely because it is less likely that elites are going to be able to block these solutions from being adopted in a more general way.
0:24:38.6 SC: So maybe this is a useful place to distinguish how physicists think about this from how biologists do because like I said, if physicists, it’s thinking about a ball rolling down the hill, it’s one thing in a landscape and it will find the minimum, or at least a local minimum and then sit there. But biologists had this, well, I don’t know if it’s an advantage or disadvantage, but it’s a difference that you have a population, that there’s a whole species with slightly different genomes, slightly different traits, some of us are taller than others or shorter, whatever, and constantly shuffling around in a non-foresighted way, trying different things. And so the whole population climbs up the fitness landscape by trying different things. And if there’s another peak that is not right nearby, maybe you could find it just because evolution is randomly trying out different things far away. And so is that quite analogous, or just a little bit analogous to the idea that in a democracy, you have a bunch of people with different ideas? Is the people with different ideas and opinions analogous to the different genomes in a population?
0:25:42.0 HF: So this is, so what we do, the way in which we try to solve this in this paper, and of course, there are many ways in which you could solve this, is by thinking about the population question being what we call a population of institutional beliefs. And so here, the idea is something that like as follows: If we think about our lives, our lives are governed by a variety of social and political institutions. So if you walk out of the door and you start driving home, it matters whether you are in Ireland or in the United States, as to whether or not you’re going to drive on the left side or the right side of the road. And if you drive on the wrong side of the road, that’s going to result in problems. So this is one very simple and one very arbitrary institution. And it’s also an institution which is largely self-enforcing because those people who don’t abide by the institution tend to get selected out of a broader population of human beings very, very quickly.
0:26:36.0 HF: But we can imagine that when there are more complicated institutions, then there’s going to be a significant degree of variation among us in how we actually understand those institutions. So one thing that has come to my mind over the last year, obviously, is looking at mask-wearing in my local neighborhood. And clearly, there are informal rules about you wear masks under this circumstance, you don’t wear masks under those. And there also is a great amount of churn in the institution; the institution is changing, perhaps, in response to individuals’ assessment of the underlying risks, perhaps also in response to purely stochastic sort of a chance and sort of meaning that somebody makes a slightly different decision, and then there’s some sort of rebounds throughout the entire population.
0:27:23.3 HF: So then, if we think about institutions as being these populations of beliefs, so everybody has a slightly different belief about what the institutional rule is in people’s heads, then we could also assume that the, and that’s the second thing we do, that the ways in which we interpret the world, our underlying heuristics for interpreting the world are likely to lead to greater variation in the beliefs that pop up. And sometimes, whether or not it is because we mistake an institution, whether or not it is because we are trying to resist an institution, deliberately trying to subvert it, or perhaps we’re just coming together and we’re talking and we figure out a better way because our two different ways of thinking about the world collide in a way that allows us to discover something new, new variant beliefs can spring up. And then the question of which beliefs, which institutional beliefs spread across the population of human beings to then become a new institution, you can then begin to model that according to the relative power of the elite versus the non-elite by treating this as a process of network contagion. So that’s more or less the insight behind that.
0:28:31.2 HF: There obviously is a lot more complicated mathematics. But here, I am frankly and completely freewriting on my cause or Cosma who is… I’m able to appreciate and to use his insights, but I am not, under any circumstances, capable of replicating those insights in a reasonable way.
0:28:52.7 SC: Well, I have never freewritten on any of my colleagues’ superior math skills, so I have no idea what you’re talking about. But let me pick out one little persnickety science thing here, which probably is not important, but maybe it is just so I’ll check. When biology, when natural selection operates, one of its key features is that it does have zero foresight, that the mutations in the genome or the sexual selection or whatever is not trying to solve a problem. And Darwin’s great insight is that it nevertheless solves a problem. And so we, human beings, hopefully, have this extra ability to imagine different possibilities, to think about a possible solution without yet trying it out and seeing if it works. Does that change the math or the dynamics in some important way? Or is it just sort of go along with the ride?
0:29:44.2 HF: It’s something that’s really difficult to incorporate in the model. So I think that what our approach does, I think it does two things reasonably well. First of all, I think it’s much better capturing informal institutions that is these rules like the rules of our mass-wearing that arise in a spontaneous way from a broad population than it is at capturing formal institutions such as the question of whether or not the filibuster is or is not going to survive in the US Senate. I think you would not want to start with our kind of a framework to be coming up with useful insights as to what is going to happen there. And the second thing that it does is it shoehorns all of those forces that you’re talking about, human creativity, and it more or less makes them into a… It shovels them into the noise term into the equation.
0:30:30.0 HF: So these are… So the notion is more or less that we’re going to see more creativity happening when we have a population of diverse actors interacting with each other to some degree, and this is going to create more noise, which then generates new evolutionary variations, which appear. But very clearly, that actually cuts out a lot of what is really important and really interesting and really ingenious about the ways in which human beings craftly managed to manipulate the circumstances around them in ways that… So are to their advantage or to the advantage of their loved ones or whatever. So like all models, it helps, but you don’t want to take it entirely seriously as a complete picture of the world. What you’re trying to do is you’re trying to take one part of an incredibly complex set of causal relationships and make them tractable so that we can better understand this, while at the same time, being aware that there are going to be some pretty profound limitations to the model, and you wanna make sure that you’re careful in extrapolating too simply and too easily from the model to the real world.
0:31:42.9 SC: Well, one of my long-standing goals…
0:31:43.5 HF: So spherical cows, yeah.
0:31:44.9 SC: Yeah. One of my long-standing goals for the podcast is to inspire young graduate students that there’s plenty of problems out there not yet solved, or at least, plenty of techniques worth trying. So maybe this is something, some room for building on what you’ve done. But let’s get… Okay, good, let’s… I think we can maybe anchor this a little bit more by contrasting the ability of democracies to solve problems, to find the maximum fitness landscapes with the alternatives that you give. And so in the paper, you talk about markets and hierarchies as two alternatives. And first, that first struck me a little bit weird. I think that in the popular imagination, markets and democracies go along together, but you’re sort of opposing them as different ways of decision-making, which does make sense once you read the paper, but maybe you could explain it.
0:32:33.0 HF: Well, here, what we’re doing, in a sense, we’re disagreeing with people like Friedrich Hayek who has a pair of very, very interesting and indeed, revelatory papers where he’s talking about the price mechanism and how it is that price mechanism serves as a process of discovery.
0:32:48.0 HF: And more or less, the argument he’s making is something like the following, that you have diffused knowledge about the world. I have diffused knowledge about the world. And there isn’t any real way for us to be able to share this in a way that helps us to solve problems, but we can rely upon the price mechanism very often to do this. You may be an incredibly good grower of tomatoes. I may be an incredibly good maker of tomato sauce. And the ways in which we figure out here, so for example, so which kinds of tomatoes to grow, what kinds of sauces to make is by using the price mechanism as a summary statistic of all of these, so complex relationships of production and hence, in a sense, the price mechanism becomes this extraordinary means of capturing signal out of a set of latent processes that are very, very… That are fundamentally invisible to human beings or at least, inarticulable by human beings. It’s very, very hard to explain without showing or doing how it is to make tomatoes.
0:33:52.6 HF: So what we’re trying to do is we’re trying to argue in a sense that, Hayek, so it’s a real insight and there are incredible things that markets can do. And also, this is a point that Lindblom makes, is that you don’t ever get any democracies, there are no historical examples of democracies which have not had markets going together wisdom. So there is some plausible, complex causal relationship between the two. But that there are certain things that markets are going to be terrible at doing and these, very often, are going to be those instances where human interaction and where human verbal exchange and reasoning and giving of reasons can help us to understand how politics works.
0:34:32.8 HF: And this is where I would turn back to Mercier’s work, and in particular, a landmark book that he’s written with Dan Sperber called The Enigma of Reason, where they argue, first of all, I think it’s a really interesting argument, is that first of all, human reasoning evolves not as a way of figuring out the world, but as a way of justifying our positions to each other in a social context so that I am… I reason because this provides me with a way to justify why it is that I deserve all of the tomato sauce; whereas, you sort of basically deserve nothing. And you then come back at me with a set of vigorous counterarguments as to why it is that you deserve all the tomato sauce, and we then bicker in a way that I’m all too familiar with from being the parent to children. But that is a lot of how human society works.
0:35:24.5 HF: And so, but then, I think the interesting thing that Mercier and Sperber do is they say, “Well, these blindnesses that we have”, so that reason is basically a means of justifying self-interest, “Go together as a set of cognitive mechanisms so that we’re really good at understanding, at seeing the holes in other people’s bullshit, even if we are not good at seeing the holes in our own. And therefore, if you can construct appropriate group scenarios and if there’s at least a minimal degree of willingness on our part to self-correct, we can then begin to actually use these groups”, so to have you criticising and telling me why my arguments suck, and vice versa, and then we can converge upon something that actually helps us to solve this bigger problem of how to distribute the tomato sauce in a way that actually works for everybody.
0:36:14.8 HF: And so there’s a lot of work that cognitive psychologists have done, trying to figure out what are the circumstances under which you are able to do this, under which group processes work in order to discover. And this is something that democracy purely can take advantage of, and this is getting back to the point about the relationship to markets, that democracy can do in a way that markets are probably pretty, pretty suited to do because they are really about pure buying and selling and the price mechanism. And hierarchy also is pretty hard to do because it is really hard, for example, for you to offer good, compelling reasons for why your boss, who can fire you at a moment’s notice, shouldn’t have all the tomato sauce and hence, a lot of that discussion tends to get short-circuited by the reality of power relations.
0:37:00.4 SC: Is it, I’m sure it is an oversimplification. How much of an oversimplification is it to cast that argument against markets as good decision-makers or compared to democracy, anyway, simply as the fact that the price is too little information. It’s a very… The price of a good is a very nice quantitative thing that is set by some equilibrium condition that we all understand, but it’s only one number. And maybe the point is that there’s a whole bunch of other numbers and more subtleties that actual real conversation between people can involve in a way that simply setting a price cannot.
0:37:40.1 HF: That’s not to be wry in. This is something that Joe Stiglitz, who is a Nobel Prize winning economist, his Economics of Information makes this point at great lengths, that the price signal, it does wonderful things, but that you simply cannot expect the price signal to bear sufficient information to produce many of the nice results that economists would like it to produce in terms of having markets work well. So in a sense, what you can see, I think that the intellectual trick that Hayek did there, and what he’s doing, he’s arguing against people who argue that, who claim that state planning can effectively serve as a substitute for markets.
0:38:19.8 HF: And so he takes a somewhat extreme position, which is that he suggested, really, the knowledge that is most important to us is precisely the knowledge that is tacit, the knowledge that cannot be readily verbally communicated, that cannot be systematised. And this allows him to make a very, very strong case against state socialists and others who want state planning as an alternative to markets, and in many ways, obviously, a correct argument, it turns out that those systems work terribly badly. But it also means that he more or less systematically precludes the possibility that people, through discussing and through arguing and through talking with each other, can now, often, actually solve problems in ways that cannot be reduced to the price mechanism.
0:39:08.9 HF: And the final point, which is I think an interesting point, is that very often, markets can be a place where this goes on. So this is one of the arguments that John Stuart Mill makes back in the 19th century and where he says that the reason why markets are so beneficial to us isn’t necessarily because of the buying and the selling, but because when you have world markets, people are suddenly introduced to a wide variety of different ways of knowing, of different societies, of different approaches to problems, and he sees this as being the important dynamic benefit of markets. And obviously, this is not a Hayekian benefit, and equally, the more the markets approach spot market type relationships in which there is much less communication, much less banter, much more sort of somebody anonymous on eBay buying or selling something else anonymous on eBay, the more that we approach towards that, the less that markets are capable of carrying these other signals in the price signal, which help us to coordinate and help us to learn.
0:40:07.1 SC: So to make it very, very practical, it was the kind of scenarios that we’re contrasting here, a truly Hayekian one, which is very laissez-faire capitalist, and everyone is just buying and selling according to their individual self-interest, and he tries to make an argument that in fact, this is optimising various good things. But the counterargument is that there’s more that we should be taking into consideration, and a democracy would allow us to, for example, just be very down to earth about it, put regulations on what a corporation can do, which the market would not necessarily do, but the society decides that’s a better solution to its collective problems.
0:40:49.6 HF: That’s right, and I think… Now, in fairness to Hayek, he does not, he is not an optimizer, he is not one of the people who argues that you can come up with optimal solutions. Instead, he is trying to think in a roughly evolutionary way, and he’s making evolutionary arguments for the benefits of markets. But he also, and this is a point that I think comes out really well in a book that a historian, Quinn Slobodian wrote, which came out about two or three years ago called the Globalists. Hayek and his followers, on the one hand, they’re committed to this kind of global constitution, you might say, of WTO-type roles as a kind of straitjacket on the ability of states, whether they’re democratic or non-democratic, to do stuff that might, for example, involve heavy regulations that might limit property rights and whatever. But also, Hayek’s disciples are very, very strongly opposed to the gathering of useful knowledge, statistical knowledge, and all of these other forms of knowledge, which they see as being a kind of perversion, which might give policy makers the false belief that they can actually impose these regulations and impose these rules and have anything go on.
0:42:00.2 HF: And I think that there’s some other very interesting twists. He has one question that I think is a huge question that we need to think about, is: What happens when economic activities become sufficiently complex that it is impossible for regulators to keep up with them? You can think about this, for example, with regard to financial regulation. It’s a huge problem that financial regulators face, is that there is an incredible asymmetry of comprehension that it is really hard to understand what many of the more complex financial products are actually doing, what the market arrangements are. And so then, this poses a very, very considerable problem for how democracy or how democratic decision-making works. What do we do when it is impossible, effectively, even for expert policy makers to actually affect the regulating given field of activity? And a whole bunch of other questions spring from that, which are probably too far afield for us to discuss in a time that…
0:42:58.8 SC: There is a time limit, yeah. But maybe this is the time, just take a little bit of an aside and talk about the word “neoliberalism” because you haven’t quite said it, but you’ve been thinking about it as you’ve been speaking, I’m guessing. And I think that this word is used a lot and no one agrees on what it means. And specifically, there are people who think that neoliberalism is the worst thing in the world, and they have a definition for it. And people who think it’s the best thing in the world, and they have a definition for it. And they’re almost non-overlapping. So do you think this is a useful term? And how do you define it?
0:43:29.5 HF: It’s not a term that I use very often myself, but it is a term, put it this way: There are many people who use the term “neoliberalism”, and I would include Quinn Slobodian, who I just mentioned, among these people, who use it in a specific and precise way. Of course, their specific and precise definition will disagree with somebody else’s specific and precise definition, so that creates some confusion. But nonetheless, I think that it can be useful. For me, I prefer to look at smaller segments of the bigger problem, so to… That’s just where I think my skills lie. I am not somebody who is a grand social or economic thinker. I’m more of a tinkerer, in some ways, than a thinker. And I think that what I like to do is to try and pull together collaborators or people from other disciplines and learn from them and to try to jam ideas together to see what happens. But when it comes to… And sorry, that’s a bit of an aside.
0:44:25.4 HF: When it comes to neoliberalism, I think Slobodian, so Slobodian’s definition of it is, as I say, that he thinks about this as being this almost global institutional straitjacket, that we have all of these people who burrowed into these institutions such as the GATT, the originator and progenitor of the world’s trade organisation, into the International Chamber of Commerce back in the 1930s, 1940s, 1950s, and managed to create these global rules such as rules on investment, which then constrained the things that both democratic and non-democratic countries can do. So he thinks about it as being a market backed up by a constitution.
0:45:06.0 HF: And so here, again, he’s building on Hayek, and Hayek has a lot of arguments about constitution and the way that constitutions help to provide liberty; as against that you get many neoliberals who think about… Or not many neoliberals; many critics of neoliberalism who would think about this as being pure markets running rampant, that these are… That this is not about global rules. This is instead about markets and the ways in which markets are effectively replacing our more traditional ways of getting on with each other. And a lot of these people are followers of Polanyi, of Karl Polanyi, who’s a great thinker of the early 20th century, and who, coincidentally, is the brother of Michael Polanyi, who’s the guy who gives Hayek all of his ideas about tacit and diffuse knowledge. So there’s a very complex intellectual set of debates happening there. So I think that each of these definitions of neoliberalism can be useful, but I guess I don’t like to talk about neoliberalism as such, precisely because it has these many definitions, although I do think that you can learn from people who define it in a reasonably precise way.
0:46:22.3 SC: Okay, very good.
0:46:22.7 HF: Sorry, that’s a very messy answer, I know.
0:46:25.0 SC: No, I think it’s a necessarily messy answer because people do use it in slightly different ways, but it was an aside. Let me get back to draw a connection between what you were previously saying about markets versus democracy, where democracy has, in some sense, a richer informational content in what it’s allowed to use to make decisions, versus what we were saying before about evolution and natural selection. Because you said that Hayek sort of uses evolution as a metaphor. And maybe that’s right, maybe evolution really is like the free market, but evolution is really just a bunch of individuals with their own self-interests in mind. They’re interacting with their nearest neighbours and is sort of giving up on this other thing that human beings can do, which is to think more broadly and sort of conceptualise collectively in a way that evolution doesn’t allow individual organisms to take advantage of.
0:47:21.2 HF: Well, I think there are two things here. First of all, I think that one of the arguments that Charles Lindblom makes about markets, I think, is a… It’s a very astute one, which is to say that markets, in a sense, they remove problems from the sphere of political arguments so that rather than arguing… And this is a different take in the Hayekian thing; rather than arguing or fighting over whether or not I should have this amount of tomatoes and you should have this amount of tomatoes, it depends on what amount I can buy of tomatoes. And of course, there is a lot of inequality baked into that, but also, there is a sense that there is a solved political problem so that you have turned this political problem of distribution into one in which we are agreeing to disagree, but within a market system, which effectively serves as the means of allocation of these things.
0:48:08.9 HF: But the second and more fundamental point that I think you’re making here is really, I think, a Herbert Simon kind of argument because Simon, who I mentioned at the beginning of this, is I think somebody who… He is an economist, he is a, God knows what he is. He is one of these people whose insights travel across half a dozen different fields. But what he thinks about is he thinks about markets and democratic institutions and bureaucracies as being structures of information, in a sense, and of collective information. This is, I think, the key difference between Simon’s understanding of the economy and, say your standard in sort of Econ 101 textbook understanding, which is, if you look at Econ 101, if you look at the ways in which they think about individuals, if you look at game theory, all of these, they argue an effect that individuals have pretty well unlimited ability to understand the environment that they are in and to figure out what is the optimal strategy given that environment.
0:49:14.2 HF: Obviously, in game theory, you could have sort of complications. There may be a game of imperfect information in which you do not know which particular branch of the game tree you’re on, but nonetheless, you have… You and all of the other actors have common knowledge regarding the overall structure and the strategies that are possible, and all of these things. And Simon’s argument is: In fact, human beings are not like that. Human beings, our ability to process information is extremely limited.
0:49:43.2 HF: And so what this means is that in reality, we tend to toss out a lot of our information processing tasks to broader structures. And those can be markets. Those can be political institutions. Those can be democracies. Those can be bureaucracies. There are a wide variety of different things that we can do here, but the crucial point then is that we have to pay attention to these collective modes of problem-solving. And in a sense, if we think about markets from a Simon perspective, we’d have to think about markets, too, as being a mode of collective problem-solving, rather than a new aggregate of the individually self-centered and fully-rational and fully-understood actions of different people. So coming together to form this market, we have to look at how the market, in a sense, becomes a set of institutions that have ineluctably collective properties, and if we don’t understand that, we’re not going to be able to understand how actual real-life markets actually work.
0:50:44.9 SC: It’s emergence, yeah, I like it as a…
0:50:46.9 HF: Yeah.
0:50:47.2 SC: I need to read Herbert Simon. I’ve actually never read anything by him, so that I keep hearing people talk about him and so, I’m gonna have to follow that up. But the other thing, but I also wanna give due to the idea of hierarchies because you’ve contrasted in the paper markets and democracies and hierarchies. And I think that you, there’s some aside in which you say there’s actually not a lot of literature trying to justify hierarchical decision-making as much maybe because in certain cases, it’s just obviously true. Like in a business, the boss is gonna decide, but in other cases like in a democracy, we think it’s obviously, or we pretend, anyways, obviously, false, that the decision should bubble up from the bottom. But so what is the principle case for a hierarchical decision-making as the right way to do this sort of social optimisation problem we’re thinking of?
0:51:36.7 HF: Well, hierarchies can get it done, is I think one way of thinking about it, that if you are… If you want to get big, complex problems solved, at some stage or another, you probably want to have some degree of hierarchy in there. You want to have some decision, which is taken; some means of communicating that decision to a wide variety of actors who are the people who are concerned with implementing the solution, and some means to make sure that they are actually implementing the solution in ways that reflect the grand design, albeit it was necessary fudging around the edges in order to be able to withstand the unexpected local conditions. And that their way of solving this, that this all adds up, that they are not undercutting each other in some way or another. So this is, I think, the standard notion of hierarchy, is that it is a way of compartmentalising and organising and understanding the world and answering them once you have a problem of delivering solutions to it.
0:52:40.0 HF: So I think that one of the classic works on hierarchy, which we don’t talk about in this book, is Seeing Like a State, which is by James Scott, which is a wonderful, beautifully-written book, which I can’t say about many other political science books. We are not known for our writing style as a discipline. But Scott writes beautifully. And so he argues, more or less, what he’s doing, he’s looking at what he calls high modernism, which is this set of hierarchical institutions which come into being in the… Really come into play in the 18th and 19th century.
0:53:16.9 HF: And a lot of what they’re trying to do is to make the world legible, that is that if you are, for example, a Haussmann trying to redesign Paris, which is this incredibly complex maze of rookeries and all sorts of unpleasant places with some pleasant people doing unpleasant things from your perspective, you start to build these grand boulevards and some sort of this very, very straightforward layout, and you’re doing this in order to make the people of Paris visible to the state so that they can be organised, so that they can be schooled, so that the problems of riots and of crime become less deep than they were because it is hard for people just to disappear to this complex morass where the state can’t follow them. And so Scott’s argument is that this involves huge losses of information, that there is a lot of tacit acknowledge and here, he is, in some ways, semi-Hayekian. But you can also look at this from the point of view of there are a lot of things that hierarchy is capable of doing that you really want to do.
0:54:19.9 HF: And you can think about this in the United States at the moment, with respect to coronavirus, that there’s one thing that we desperately want above all else, which is a effective means for distributing the vaccines that we have as efficiently as possible in ways that minimise the impact of the disease. And we are learning that one of the… Over the last 20, 30, 40 years, that a lot of the state instrumentalities, which you might have expected to be able to do this, have withered away and have disappeared. And so this is one of those circumstances where hierarchy comes to the fore. So the hit against hierarchy is that it is not particularly flexible, it is top-down, it tends also, according to Scott, to resegment the world in ways that fit its own concept, that fit its own ways of categorising the world, and this involves the loss of information. The benefits are sewage, so dealing with diseases, all of these provisions of crucial services which we couldn’t live without, and which we need some kind of central organised way of doing in order to get that stuff done.
0:55:31.0 HF: And I have a whole other spiel I can give you. I basically work with Marion Fourcade about how machine learning has a lot more in common with Scottian versions of hierarchy than it does with many of the… With the market notions that people very often apply, but that’s a red herring, which you can follow our lead to one side if you…
0:55:51.5 SC: No, no, no, I wanna hear it. I wanna hear about the relationship between machine learning and hierarchy. But just to sum it up, the good thing about hierarchy, like you said, since decision-making is concentrated, the decisions get made quicker and things get done, but they might not be the right decisions ’cause there’s sort of less wisdom being collectively integrated in there. And if there’s some relationship there to machine learning, I definitely wanna hear more about it.
0:56:16.0 HF: Okay. So I think that the argument that this… Again, as I say, I try to work with co-authors who are brighter than me, and Marian very definitely. This is applied versions of her ideas without a huge amount of originality on my part. But more or less, we have this notion that we call “high-tech modernism”, that the idea behind it is that if you think about how machine learning works, it really it’s a process… It’s a process of categorisation, a process of classification, whether it’s supervised or primarily supervised but also unsupervised machine learning. What you’re trying to do is you’re trying to take a dataset and you’re trying to sort it out into some kind of a classification scheme, which can then allow you to figure out stuff to identify relationships that you might not have done otherwise. And so if you think about it from that perspective, you immediately see that there is a very, very strong analogy to bureaucracy, because that is exactly what traditional bureaucracies have done.
0:57:16.9 HF: Traditional bureaucracies have always looked to classify and to figure out if you are a, I don’t know, a deviant Irishman, or a upstanding loyal British citizen, or what have you. And they have done this immensely and efficiently and whatever with paper files, but there’s something pretty strongly the case… So that, there’s something very, very similar. And the way many of the fights that we see happening around machine learning and about machine learning and equity machine learning, racial bias are in many ways their new versions of the fights that we have been having since forever about the way, for example, in which redlining was used by financial institutions to effectively make it impossible for people in entire neighbourhoods to get loans. So we see, not the same debate because of course as the technologies change, things change about how the technologies are applied and their social impact, but there’s a strong relationship there.
0:58:19.7 HF: The key difference, I think, with machine learning to the kinds of stuff that James Scott was talking about, is that when you think about James Scott and the way that he talks about these cities being created to map, map the bureaucratic categories onto the world around them, the process with machine learning, it’s much more subtle because we are all carrying around with us these little devices in our pockets which are endlessly feeding information to the mothership, allowing the mothership to classify us and categorise us in different ways, cookies, whatever new thing Google is coming up with to replace cookies. All of these are intended to categorise and classify us, and they have these very substantial consequences, they guide you into being the kind of person who gets a sweet mortgage offers versus the kind of person who gets the kinds of predatory loans that we see advertised on the side of subway cars, but these processes of guiding tend to be much more invisible than the more direct forms of repression that we’ve seen in previous eras.
0:59:22.7 HF: And so then the question… This gets back to what you’re saying about the ways in which hierarchy can go wrong, the question is, what kinds of feedback loops are there through which you can correct some of these problems? If you think about hierarchy, say for example, if you think about authoritarian regimes, authoritarian regimes are notoriously terrible at taking feedback. They try to monitor what their citizens are doing and thinking because they want to stay in power, but when the feedback tell some things that they don’t really want to know, it’s very hard to penetrate through the system. And one of the interesting questions with machine learning is, if you have these feedback loops which are largely automated and which provide very little room for actual people to insert themselves into the process and say, “Hey, this isn’t actually what we want”. How do you change that? How do you get into a world in which machine learning is democratised, and that’s a huge problem.
1:00:17.6 SC: Well, this is good because I think it leads us in, even if I buy, which I do, the idea that democracy is a good way of making decisions, in comparison on a fair level playing ground with markets or hierarchies, democracy has its issues. When we go from the ideal theory to the real world, we have a whole bunch of people as the critics would let you know that are not very well-informed or they have incoherent beliefs themselves, so it’s not even just a majority vote would get you a simple decision-making process. And increasingly, people in the world are of the opinion that the real power is too far away from them, even if they live in a democracy, it’s a representative democracy, it’s a republic, right, as we’re told, not a democracy.
1:00:36.5 HF: Yeah.
1:01:07.0 SC: And they don’t see this direct influence of their voice on the decisions being made. So, is the modern world making it harder for democracies to function in this ideal way?
1:01:18.9 HF: Yes. Well, this is one of the arguments that I think people like Piketty have made at length, which is that as you see wealth inequality going up, this begins to feed into the political process in some important way. You see political scientists like Chris Achen and Larry Bartels, you see also Paul Pierson and Jacob Hacker who have written books that try to describe this in a broader public arena, so you see these feedback loops developing between wealth and political power, so that many of the forms of democratic representation which are supposed to work in order to give people some power really don’t work anymore. You also see, I think, other more subtle forms of differentiation, which I think are more uncomfortable for people, the liberal left deal with, which is that the lives that you and I live as college professors who have pretty comfortable sweet deals, are more and more divergent from the lives of a lot of people who don’t see very much in the way of economic opportunities.
1:02:21.4 HF: And there is a degree of invisibility there, that I think we have begun to realise that there are many, many problems that have been simmering away beneath the surface of American democracy, that we got entirely too comfortable with and that we don’t… We simply haven’t confronted, and I think in some way still are not confronting because they are too difficult for us to confront. And I think the final question is, to what extent democracy itself is a stable system? And so here… And again this is yet another co-author, I think, in terms of co-authors, I think… Try to think, in terms of my work with co-authors. Another co-author of mine is Bruce Schneier, who is a computer scientist and a cryptographer, we have been thinking together about, what is democracy as a dynamic system, and what kinds of feedback loops democracy needs to work and what kinds of feedback loops can go astray?
1:03:24.8 HF: And roughly, our argument is that many of the things that have gone bad over the last number of years, our feedback loops which under other circumstances could be reinforcing for democracy; that is, if we think, for example, about politicians’ ambitions, if you want politicians to be selfishly ambitious, to want to win power and whatever in the democratic system, but you want them to do this by appealing to voters, that is when democracy works, is when politicians are vying with each other to try to appeal to voters and try to figure out ways to attract voters to them. And of course, when that ambition begins to get disconnected so that they begin to refuse election results, that becomes much more problematic. You see similar things with distrust in election systems, you probably do want to have citizens not being entirely trusting of election officials or of election systems because things can indeed go very badly wrong. Again, that is a kind of a suspicion which has metastasized in some importantly problematic ways, and finally want democracy to be a relatively open system to new ideas, this is where it gets its benefits, is from the fact that new groups can articulate their interests and organise, and we can think about good which in which that has happened, such as Black Lives Matters.
1:04:40.3 HF: We can also think of something that probably would not have been possible in the more closed-off democratic system of the 1950s, but we also can think about a negative aspect such as QAnon, and the various forms of conspiratorial communities which have formed on the internet. And so then this doesn’t necessarily help us to solve the problem, but it helps us to see how it is that some of the feedback loops that under other circumstances can reinforce democracy, or tearing American democracy apart to some degree, and to start to try to think more systematically about how to redirect those feedback loops in ways that are more positive and more beneficial. Because it’s hard to argue a unqualified case, in particular for US democracy at the moment because US democracy sucks, there are huge, huge problems.
[chuckle]
1:05:32.7 SC: Well, I recently talked on the podcast with Zeynep Tufekci, and I think she said something important, that I think she gave credit to Herbert Simon for saying about the importance not of necessarily of information but of attention in the modern world, because we’ve crossed some threshold where there’s just too much information out there. And now the question is, what do you pay attention to? One way that bad actors can combat or try to get their point across is just to flood the zone with bullshit, right? And…
1:06:07.4 HF: Yeah.
1:06:07.8 SC: That does seem to raise a new kind of challenge for democracy. People were lying and misleading all the time, but just the volume of it that we can get, and now we’re able to immerse ourselves in pools of falsehoods in ways that we couldn’t before. Is there a mechanism internal to democracy that will help us get out of that kind of dilemma?
1:06:32.7 HF: Well, the ways… My version of Zeynep’s argument to the quip, you may have heard a saying among many First Amendment people, that the best solution to bad speech is more speech. And so, I think what authoritarian rulers discovered a number of years ago is that the best solution to more speech is bad speech, that is that if you flood the zone with bullshit, it becomes incredibly difficult for people to figure out what is happening anymore. And if you are the rulers, and if you’re able to prevent people from, in the opposition, from communicating with each other, from organising, from knowing what the reality of politics is, what people actually want them to do and so on, then you are already starting with a substantial advantage. I think one of the key research questions that we have to figure out at the moment is how we can try to mitigate. And this is something Bruce and I wrote a paper, making these arguments about the vulnerabilities of democracies to information flooding, more as an argument that we had… We know a lot about how autocracies are vulnerable to certain kinds of information flows, we need to start thinking about how democracies too have these kinds of vulnerabilities.
1:07:47.6 HF: And I think where we need to start experimenting is trying to figure out how best to mitigate because we do not want to move into a world which is a world of strong gatekeepers again, because that world is a world in which much of the benefit of democracy disappears. But we want to figure out ways in which we can at least tap down disinformation flows, and some of this will of course involve as everybody says, working with social media and figuring out the ways in which both algorithms and self-selection by people who are looking for crazy stuff at the start, how those things work and how to minimise and to mitigate their negative consequences, while preserving benefits, the benefits of free organisations that they provide. Some of it also will involve us thinking in a different sense about media regulation of traditional media, and I think this is work that in particular we see coming from people associated with the Berkman Klein Center at Harvard, there has been a lot of work which suggested if you really want to see where the bad stuff happened in recent elections, it happened via Fox News, so that there is effectively a conveyor belt that you see where both the left and the right have a lot of nonsense floating around the fringes. So that is just a necessary fact of how it is that people are.
1:09:08.6 HF: But that on the conservative side of the spectrum, we saw a lot of this nonsense being effectively hoovered up by Fox News through a variety of means, and then injected into the mainstream of public debate. So I think, if we want to think about reform of media, I think that going after platform companies is one good and important thing, but figuring out ways that we can force more traditional media to also be more responsible while preserving their ability to exercise independent voice is something that we need to think about very carefully as well. One of the few moments of hope in the last couple of months of misery have been seeing Fox News suddenly begin to realise that it can’t make these crazy statements about election machine manufacturers, without having some legal liability. So, one of the things that Bruce and I have toyed around with a little bit is the possibility of whether or not… Are there ways in which that can be built upon to try and create some kind of a moderating mechanism towards making it more in the… Incentivising these places to tell the truth, even if they still provide a slanted version of the truth, a slanted version of the truth which is not entirely disconnected from reality.
1:10:26.9 SC: Yeah, no, maybe you could even elaborate on that a little bit because I think this is the really crucial question. We want there to be a diversity of opinions and voices and values out there in the media, but we should also be able to say, “We would like the factual information to be true, as much as it can be”. And right now the incentive structure is not necessarily aligned in that way, and you brought up the example of the voting machine companies, and so here you had some big companies that had both the chutzpah and the resources to bring lawsuits, and say, “Don’t lie about us”, and that seemed to work. How scalable is that mechanism to stop major media corporations from just spreading false facts out there into the world?
1:11:09.0 HF: Not particularly, but one idea… This is… I am not a policy person, and Bruce is in different ways, but this is not his area of policy. But one thing that… One way that you might be able to think about doing this is by obliging as a licensing requirement, TV stations to have, and other broadcasters have a significant fact-checking, truth-checking apparatus as part of their make-up. And the idea behind this is as follows, that then this makes it far harder for them to argue and course that they… That they have not been committing malicious slander or liable, if they have a place… If they have parts of their apparatus which are specifically devoted to figuring out this stuff, it could be taken to imply a greater level of assumed responsibility for them actually getting stuff. Right? Whether that actually works or not depends upon a lot of specific and detailed regulatory knowledge and legal knowledge that I don’t have, and that this is not Bruce’s specific area either, but those are the kinds of things that I think that… The kinds of places that I would start to look in. I would also look to the act that the Democrats are trying to get through, which is trying to minimise your various forms of gerrymandering among other forms of abuses of the electoral system.
1:12:41.5 HF: And finally, I think that there is a lot of argument to be made for different approaches to voting, transferable voting of one sort or another. Lee Drutman has a book where he recently makes this argument. And you can think about this also in terms of benefits for information, because one of the problems we have at the moment is that on the right side of the spectrum, the extreme crazies in the Republican party and the more moderate people are bundled in together, and it becomes a [1:13:10.9] ____ that you more or less have to subscribe to certain kinds of beliefs or else not be considered to be a Republican in good standing. If you had a different voting system which might make it easier for different parties on the right to emerge, then it would become… That might help you to help provide people who don’t subscribe to the full, “Donald Trump was illegitimately denied the election” mantra to form their own party, and hence might make it easier… To isolate the crazy people and the crazy beliefs from those who probably don’t believe all or much of the stuff, but who feel that they have to go along with it because of pressures toward social conformity.
1:14:01.5 SC: Let me start winding things up with getting one more big idea on the table here, because we’ve talked a lot about using democracy, putting it to work to make decisions, and I think it’s a persuasive case that a diversity of voices connected in a very functional way is a good way of solving complicated problems. But then there’s the flip side of that, which is the evolution of democracies themselves, something that you’ve also written about, the literal evolution, the evolution over time, and maybe there’s a biological metaphor lurking there too. But you talk about the examples of Athens and Sparta as two different ways that purported democracies can evolve, and I was amused to read in your paper that these are two of the best studied political systems in history, they were quite a long time ago, but why don’t you tell us the story very quickly of Athens and Sparta and what they teach us about the evolution of these institutions?
1:14:58.2 HF: So, what they are, they’re very well-studied, and they’re canonical cases, for better or for worse, that is that a lot of our ideas about politics came through the Romans, who in turn were profoundly indebted to the Greeks for their culture, for their understanding of how politics work. So there’s a lot of path dependence in the examples. And also they have a nice blend of, on the one hand they have enough factual information that you can figure out a lot of what happened, but there also are enough open spaces that you can… You can then begin to play around, and use them as examples, that make it easier to connect, that then might be the case if you had a much more detailed and boring set of… Some historian once said that fire and flood are historian’s two best friends, and I think that there is a little bit of that going on here as well. But nonetheless, I think what they do suggest, they do provide, all joking aside, they do provide instances of two very different trajectories.
1:16:10.6 HF: And if you think about Sparta as being a pretty clear example of autocracy, and a very strong form of autocracy where you have a relatively small self-reproducing ruling class, you have a somewhat bigger class of people who don’t have citizenship rights, but have some rights, Perioeci they’re called, they are around the limits of the community, of the household. And then you have a massive population of serfs or slaves, the Helots, and Spartan society is all organised in the same way that the 19th century [1:16:45.3] ____ was all organised about keeping the Helot stand, keeping them from getting power, keeping them from getting weapons, keeping them from getting organised, keeping them from getting leverage. So this is a very… And in some ways you could say it’s a successful society for very limited and unpleasant forms of success, and that it manages to survive for several centuries, but it managed to survive in an extremely static way, they don’t allow the introduction of money, they are extremely suspicious of innovations, it is a particularly unpleasant society, even by the standards of ancient societies. And even if you are a member of the elite, you probably have a pretty miserable time in Sparta.
1:17:32.2 HF: Athens in contrast is at best a flawed democracy from our perspective. It, again, is a slave society. There are a lot of people who are enslaved, and that is an important basis of the economy, slaves and the silver mines, as well as in a variety of other capacities. There, women don’t have anything even vaguely approaching full political rights, and there is a population of metics who don’t have citizen rights, but do have certain other rights as well. It’s not perfect, by any stretch of the imagination, but it’s a hell of a lot better than Sparta. And you see a lot of the benefits and disadvantages of democracy, its dynamic ability to adapt to bring in innovations from other countries when those innovations look to be useful or helpful innovations, it’s a place which is affirmant.
1:18:23.2 HF: And the reason, of course, why we remember Athens is because during that period when Socrates, Plato, Euripides, Aristophanes, in a very different way. These are people whose work flowered, and observe that… And there is a sense of a common culture there, which fertilisers, which explodes, and which really observe, even apart from the the post-hoc ways in which we think of this as being awesome. I think you can reasonably say that it is objectively, there is a lot of wonderful cultural material which is produced, and this is not divorced from the political and economic context that Athens was in. It has a lot of other problems, that’s an imperial power, and a lot of its wealth is developed, frankly, by exploitation of other Greek states. But again, it provides a way to see how it is that you can get a dynamic society where the power relations are much more moderate than they are in Sparta, even if they are pretty extreme by modern circumstances, and hence, this is why we use these examples as ways of thinking through this.
1:19:35.8 SC: And did they teach us something about the relative merits of stability versus dynamism? You want both a little bit, right, you wanna be a little bit stable, otherwise you’re in trouble, but you also wanna be able to adapt to change, and getting that balance right is always a tricky thing.
1:19:50.9 HF: It’s always a very tricky thing, and I think that these strengths of any political system… And this is a huge, huge problem. This gets back to the beginning of our argument for political scientists. Political scientists by and large don’t tend to think in dynamical terms, and to the extent that we do think in dynamical terms, we often find ourselves drifting back without even realising it, back into static frameworks and configurations. There is this approach in political science called “historical institutionalism”, which is all supposed to be a batch of processes of institutional change over time, but as Kathleen Thelen, who’s one of the chief progenitors of this approach has complained, very often people who are in historical institutions, they say they’re about change, but really when you look at it, it’s all about the same old, same old, same old, that they managed to get pulled back into it. If the strengths of any of these systems are dynamical, this means a number of things. First of all, it means that we can come up with accounts of the dynamical advantages of one thing over another across a large spectrum of possible parameters, but we also, we have to preserve in the back of our heads the possibility that…
1:21:01.9 HF: So even if we think that one particular system is better overall, that there may be specific shocks, specific problems that it is going to encounter, which is going to do worse in dealing with than other systems. We have to think about how institutions work in different ways, we have to think about them as flows. And it’s hard to come up with really, really hard and fast, this is universally better than that across all possible states of the world, but I think it’s still possible to come up with a reasonable, across most reasonable circumstances kinds of generalisations where you can… And generalisations that, even if they are more limited than one would have in a more static account of the world, generalisations that probably are at least a little bit more true to reality, and that we know that we live in this dynamic chaotic world and we need to understand better how these systems work. And two other people who I’d recommend here, Scott Page and Jenna Bednar, both of whom have done wonderful work in different ways, thinking about these problems for institutional change, and thinking in particular about the ways in which diversity plays an important role in all of this.
1:22:16.7 SC: Well, if it makes you feel any better, even within physics, the study of statistical mechanics has historically almost always been in the equilibrium context, and it’s only in the last 15 or 20 years that we’ve really been taking non-equilibrium statistical mechanics very seriously. So, I wonder if there isn’t room for importing some of the fluctuation theorems and studies of deviations from equilibrium, from physics, into these studies of societies, that sounds like an ambitious thing, but maybe worth looking into.
1:22:47.2 HF: Well, there’s gold in them thar hills, I would say…
1:22:48.8 SC: Exactly.
1:22:48.9 HF: For those of your listeners who are interested, the one thing I would say is that there’s a long-standing history of physicists coming in and saying, “Okay, here is a problem, here is my… ” Ising or whatever… How do you pronounce his name? I’ve never…
1:23:07.7 SC: Ising.
1:23:07.9 HF: Easing… Ising. Yeah, yeah. “Here is my workhorse model, I’m going to show you how my workhorse model generates something that kind of looks like what you’re talking about if you squint at it in the right kind of way, and there is my publication and you peons can then go and figure out how to apply it”. That’s not going to work. I think that if you want to generate real understanding, there is a huge amount of open space for you to do so, but you should do so through engagement with social scientists who are already working there, and through engagement with the actual messiness of real life, which is of course what we’re all trying to capture here.
1:23:47.7 SC: And I cannot think of a better final thought than that one. Henry Farrell, thanks very much for being on the Mindscape Podcast.
1:23:54.2 HF: Thank you so much.[/accordion-item][/accordion]
David Graeber’s “Debt: the first five thousand years” would be a useful contributor of context for this discussion. The rise of states and markets together, during the Axial Age empires, created the definition of “self- interest” that is common among economists ( and hence many others) to this day. “Human economies” (many Native Americans among them) early mostly, were both non-state and none-market, and were often arguably more democratic than Athens or Sparta, or the U.S. today. Great talk!
Pingback: Sean Carroll's Mindscape Podcast: Henry Farrell on Democracy as a Problem-Solving Mechanism | 3 Quarks Daily