When discussing a proposal to place all nuclear weapons in the hands of an international body, Thomas Schelling accused its proponents of wishing for "government without politics". Whether or not he was correct in that instance, it's a line that's stuck with me ever since I first read it and, well, it's a line I found myself thinking about quite a bit while reading this.
You cannot simply wish away humans' propensity to form groups. It's been a feature of virtually every political system humans have come up with, and probably every one we can come up with. I do think that lottocracy is an interesting idea that is worth trying in some capacity but, well, if you put 100 randomly-selected people in a room together, guess what, they will eventually sort themselves into groups.
Similarly, Pigouvian taxes are probably a genuinely good idea, you just have to sort out the tiny problem of how you're quantifying negative externalities. And it turns out that different people will think certain externalities should be weighted differently, whether due to simple differences of opinion, naked self interest, or something in between. And some of those people might group together to advocate for certain externalities being weighted more or less heavily and whoops, we're back in the messy realm of politics.
It doesn't help that you seem to conflate instances of genuinely irrational behaviour from groups (where most of the members of a group want one thing, but group dynamics push them to do another) with instances of groups simply accurately reflecting the views and interests of their members. i.e., there isn't some secret giant group of people who individually support nuclear power, but whom group dynamics push to oppose it instead; the groups that oppose nuclear power oppose it becauae most of their members oppose it, and they oppose it because they (mistakenly) believe it to be dangerous. If you randomly selected 100 people from the population, they wouldn't suddenly become more pro-nuclear.
Thanks, Ben. Part of my answer to your comment would be repeating my answer to Patrick's comment, so I'd recommend checking out my response to Patrick's comment. The only thing I'd add to my response there is that jurors generally avoid forming tribes, and they generally make pretty good decisions about people's guilt or innocence. They're not perfect, of course, but they tend to be better-informed and more impartial than voters are. So it is possible for people to make important decisions about other people's lives without being tribal, while being relatively well-informed about the details, and while being relatively impartial. Yet this is not how democracy works by any stretch of the imagination, and that is a very bad thing. I think democracy would be better if voters and politicians were more like jurors (rather than mean, dumb tribes), and if the media was more like the prosecution and the defense of a case (rather than a machine for delivering one-sided tribal propaganda). Given that we already have a trial-by-jury system, and it already makes important decisions, and it has not yet devolved into a patchwork of mean, dumb tribes, it suggests that the sort of tribalism you think is inevitable is by no means inevitable. I think if you randomly selected 100 people from the population, and subjected them to the voir dire process to make sure they're impartial (just like we do for jurors), and presented them with the pro-nuclear and anti-nuclear side of the debate, I would bet they'd decide on something more pro-nuclear than the average voter. In general, I think they would make better policy decisions than the average voter on pretty much any policy issue. Yes, we cannot "wish away" the propensity to form groups, but we can try to reduce it and mitigate it. And we can avoid glorifying it and pretending it's virtuous.
It’s probably one of the most traumatizing things one can do. At least the inner city ones that had experience with. The idea that it doesn’t devolve into mean dumb tribes…have you been on a jury?
Anyway, if I was on a jury that determined tax policy I think that I would determine that jury members should pay no taxes and everyone else should pay lots of taxes to jury members (I think we are back to your authoritarian example).
This was entertaining to read and certainly contains some truth, but nonetheless I think it is overly cynical about group behavior, and then simultaneously not cynical enough about "crowds".
In defense of group behavior, it provides the conditions under which the good kinds of "crowds" can emerge (or be regulated into existence). You gesture at this with the traffic light example, but I don't think fully examine its implications. We need rigid rules to coordinate around in order to have a predictable, orderly society where positive sum exchanges can routinely occur. A hypothetical pure crowd with no group-based order around it incentivizes violence, fraud, and abuse of unequal power dynamics.
Compare labor conditions in the early industrial revolution to the post-New Deal era. What broke those were group-based labor unions and political parties advocating for blunt policies like the 40-hour workweek, OSHA, bans on child labor, and the minimum wage. I don't see any other realistic way for the downsides of industrial capitalism to have been curbed. It wasn't going to organically emerge via some "crowd" deliberating on it, and the market itself did not deliver it.
You often need blunt instruments to shove a crowd out of an undesirable equilibrium. Sure, you can imagine a more efficient and intellectually satisfying system of Pigouvian Taxes, but those don't actually solve any of the political "group" problems you mention, because we'd still have to decide what harms count and how to measure them. It's very easy to imagine them not working in practice.
I think you aren't cynical enough about crowds. Even setting aside coercion and fraud, crowds often converge on stupid policies and bad values. For example, you cite NIMBYism as "group" behavior, but actually it seems to me that a lot of NIMBYism emerges through crowd-like behavior. The more local you go, and the more voice individuals have, the more NIMBY the politics are. In fact, YIMBYs have been trying to make the issue of housing more group-oriented (and the YIMBY movement is a much more coherent "group" than NIMBYs, who seem more dispersed and localized). It turns out that blunt, group-oriented state politics are more conducive to good policy outcomes on housing. Why? Because focusing on the blunt goal of "affordable housing and economic growth good" is better than some nuanced deliberative process of local residents who, usually, converge on the idea that they don't want change and like their city the way it is thank you very much. Most people are NIMBYs when the develop is literally in their backyard.
Thanks, Marcus. This is some good pushback. I will grant that NIMBYism vs. YIMBYism is an interesting exception to my thesis. YIMBYs appear to be more thoughtful than the average political group (I’m inclined to say they are unusually crowdy but I also realize that is a bit too convenient for me). I would deny, though, that local politics is that much more crowdy than national politics. I think it’s groups all the way down. I think the point about unions being needed for better treatment of workers is debatable and has been challenged by some scholars, who think these measures were already underway by the time unions were pushing for them (I don’t have an opinion on this, just noting the uncertainty about cause and effect). I would question whether violence and “unequal power dynamics” aren’t themselves group dynamics. I think they are: you need well-coordinated groups to gang up on and dominate people. The problems you’re talking about (I assume) involve groups ganging up on each other—not lone psychopaths preying on people. Most lone individuals are not psychopaths. Most groups tend to act like psychopaths toward enemies and outgroups. That’s why I (like George Carlin) generally prefer individuals to groups, where the odds of psychopathic behavior are lower. Not zero—lower. I would also grant that some crowds have better incentive structures than other crowds, and a crowd that cannot detect and punish psychopaths is one example of such a poorly incentivized crowd. Perhaps there is another continuum, from chaotic collections of individuals on one end, to well-incentivized and wise crowds on the other. Perhaps we need groups to nudge us toward that well-incentivized end, but perhaps not. Either way, I don’t think we can count on groups to do that for us, because they’re too dumb and mean.
The pessimistic frame is that all of these problems with democracy are unsolvable, which may be true. I don’t think the answer lies in controlling voters access to the ballot box but instead on the societal narrative of what makes the good life. Pluralism makes this a challenge but the stakes are high. The alternative is nihilism which is doom. Based on all of these problems, why has the US succeeded so far? Caplan himself finds it puzzling that democracy in the US works as well as it has. I find that as a reason to be optimistic and promote optimistic values. The downside case is overstated (definitely in the media chattering class) and putting out positive vibes is the answer in my frame. It’s costless and directly benefits the recipients as well as yourself. The enemy is not the other tribe but nihilism itself. I read your blog as the prosecutor’s case in combatting nihilism. I reject the thesis that democracy generally produces bad results. Bad compared to what? Life in Western democracies is far better today than in was 200 years ago. You can list a bunch of problems that are yet to be solved as evidence of being bad. Without recognizing the benefits and tradeoffs, it’s a meaningless statement. As an analogy, you could say that life is bad because you die. Agree with premise that death is bad but does that make life itself bad? If dying is the cost of living, is that a worthwhile tradeoff?
Thanks, Patrick. Maybe these problems are solvable; maybe they aren’t. The only way to find out is to admit that they are problems and take them seriously. To use your analogy, yes, death is bad. Maybe we’ll never get rid of death. But it would still be good to delay and prevent deaths. Saving lives is good. Increasing life expectancy is good. Halting or slowing the aging process, if that is possible, is worth looking into. We should not glorify death and pretend it’s awesome in the way we pretend democracy is awesome. As another analogy, maybe war will always be with us. But we can still say that war is bad and we should do whatever we can to mitigate it and reduce its likelihood. Likewise, maybe some form of tribeocracy will always be with us. But we can still say that tribeocracy is bad and we should do whatever we can to mitigate it and reduce its likelihood. In fact, the analogy may be deeper than it appears, insofar as war is just politics by other means. As for the “compared to what?” question, I address that in the “Coda” section.
I am not sure you should revealing this much truth. I don't know how many people can handle it.
Bryan Caplan's explanation for voting (that it gives us a warm glow) is not incorrect; it is only incomplete because it cites a proximate cause without mentioning a distal, evolutionary cause. Many explanations in the behavioral sciences boil down to people seeking what makes them feel good and avoiding what makes them feel bad. Evolutionary psychology explains why things make us feel good or bad, providing a more complete picture. Thank you for offering a plausible evolutionary explanation for voting.
Groups do suck for many reasons, and you mention some of the important reasons. I especially appreciated the observation about the way thinking in terms of groups leads to black-and-white thinking, a topic I have also written about, https://www.psychologytoday.com/us/blog/cui-bono/201209/the-unsavory-psychology-two-party-politics . One thing that you did not emphasize about groups is just how inefficient and ineffective they are for helping individuals acquire what they want. Just think about how helpful typical faculty meetings are for helping you achieve your personal goals. The "strength in numbers" mentality might have worked for the limited, basic goals of small hominid bands, but I've found that operating as an individual with a limited number of other individuals-as-individuals has been enough for me to lead a pretty satisfying life.
In my view, quasi-rationalistic schemes such as Pigouvian taxes, utilitarianism, and effective altruism are always going to fall short because the number crunching is too difficult for individuals. How in the world are we supposed to accurately predict through computations the future negative externalities of goods and services? Or which acts will bring the most happiness to the most number of people? Or which of the issues addressed by effective charities will be the most important for overall flourishing in the world? (Effective altruists do have data on the efficiency of specific charities, but who knows which of the issues addressed by the various charities will truly matter for the long-run wellbeing of everyone--including nonhuman animals and plants--on the planet?)
Markets, on the other hand do not require us to crunch numbers as individuals. Support for free markets led me to the Libertarian Party for a period of time. Until I saw that the LP was a supergroup party that behaved just as badly as Republicans and Democrats.
When it comes to voting, I am going to keep voting for whoever seems to be the least offensive candidate, even though I realize my vote makes no difference. The advantage I see to voting is that when some asshole gets elected, people can't criticize me for not voting for the person who didn't get elected. No warm glow for me. Just freedom from nagging people.
This is a terrific piece. IMHO the recognition of the overwhelming importance of groups/tribes in social affairs is the beginning of wisdom
I have one point I would like to add.
We can divide society across different access and thereafter there to the one side it the other as a group/tribe (hereinafter tribe)
The resultant tribes have confident loyalties and worldviews and such trinal belief systems near heavily upon social action
IMHO a there is one critically important difference defines what is likely the most important social divide in human history - that between people whose world view is importantly structured epistemic by the inheritance of The Enlightenment and people whose world view is pre Enlightenment. Between people who believe in the importance of evidence and argument and people who do not put whose grasp it's so weak as to be non effective.
I am not suggesting that post enlightenment minds are perfectly rational or indeed that they are particularly rational at all times. I am however suggesting that whilst their commitment to rationality may be more or less intense and their epistemic competence may be more or less thorough they're is a real line here.
The consequences of this divide that splits humanity into two groups are enormous.
Thanks, Pete. You may be right, but I worry that there is something dangerous about this way of thinking--the idea that there are the rational, post-enlightenment folks on one side and the dumb, tribal, benighted folks on the other. This way of thinking is precisely the sort of ingroup supremacy narrative that groups/tribes tend to push on their members. So even if there is some truth to it, I would worry that pushing too hard on it would create precisely the sort of group/tribal boundary that the so-called post-enlightenment folks think they have transcended. I think to avoid the transformation from individual to group, we should try to see the people around us as individuals who are our equals, vulnerable to the same biases as we are, no matter how irrational they might seem. As soon as we start to see a large collection of individuals as inferior in any way (intellectually or otherwise), we start to go down the group/tribe path. What's so insidious about this path is that people cannot see that they are on it and actively deny being on it. No one who's in a tribe sees themselves as being purely tribal. They simply see themselves as individuals fighting for truth and justice, who agree with the wise, virtuous people around them on what is true and just. Avoiding the slide down the group/tribe path is really hard, and I don't even know if I've succeeded, but the only way to even possibly succeed is to be aware of how alluring it is and notice the warning signs.
This is an extremely important point and it deserves serious attention. This is likely not the best forum for it. (You might want to look at the small number of pieces I have posted on here about it. If you can face it then start with ‘Two Tribes’)
However…
The implications of not understanding this distinction, unappealing though it may be, are catastrophic. Let me give you a simple example.
When Hilary C debated T the very style of their speech betrayed the audiences they were reaching out to. One participant knew the audience the other confused it.
Forget the negligible intellectual content of T’s contributions. The content isn’t the point. The style however was crucial. It screamed ‘I am one of your tribe.’ The fact that it was a ridiculous lie didn’t matter. His speech was an effective tribal rallying call and that was its only purpose. Such calls overwhelm policy statements.
Hilary’s post Enlightenment rationally presented policy outlines just went right over most people’s heads. All they knew was they she wasn’t one of them.
Democrats make this mistake constantly. They implicitly believe that the ‘others’ are fundamentally like them - they’re just not as good at it..not as articulate…not as well educated…. etc etc etc. But the break from pre to post Enlightenment thinking is not a continuum, there is a very real break. You either consider sound evidence and argument to be epistemically critical or you don’t.
If you don’t understand this your messaging will fail and your election campaign will fail.
Look at some recent campaigns:
Obama ran on hope - not policies
The UK Brexit campaigns could not have been more starkly different:
Leavers - we want our sovereignty back
Remainers - remain. :). possibly the worst campaign messaging ever
==
S. Stevens explores this in his book on the Republican Party elections playbook.
The stakes on not understanding that most of your electorate aren’t especially receptive to well crafted and substantiated arguments are extremely high.
Republicans have a much better grasp on this reality. All their messaging is based on it.
1) David Butterfield is interesting to read/listen to regarding Athenian democracy, which was a lot more direct and with an incentive structure giving much more significant accountability
2) James Madison in the Federalist papers has very sharp critiques on groups, or what he calls factions, with notable similarities to your own. Many of the checks and balances set in place in the US Constitution were constructed to add friction to a potential stampede of destructive policies brought on my motivated factions.
3) Joseph Schumpeter predicted in his famous book in the 1940s that Democracy and Capitalism will bring prosperity which will result in people voting in an ever expanding bureaucratic state pursuing various symbolic status games. In this sense although, as you note in your other articles, democracy, free markets, and meritocratic scientific institutions were the magical cocktail that brought the modern world's prosperity, their synergy started to crack with the expansion of the managerial state. Another important point is what Moses of Random Walk regularly points out: all of the current productivity growth comes from AI, which employs a very small portion of the population, while all the employment is growing in nonproductive (economically) health care and elder care services. This is the acceleration of Baumol's Cost Disease as far as the the sharp growth in services and decline of manufacturing. That means the people generating economic productivity don't have much voting influence, but the people drawing money from the state (directly or indirectly, that is by employer funded health care) are a growing bloc. This is concerning as far as the future of this synergy.
"Stuff will keep getting worse because we won’t know how to fix it, and our stupid attempts to fix it will make things worse. We will make fewer and fewer babies, dwindle in number to the brink of extinction, become increasingly old and decrepit, run out of bold new ideas, over-regulate the few bold ideas we have, technologically stagnate, economically stagnate, lose trust in science and markets, destroy liberal democracy, forget to do anything about climate change, bungle the ensuing refugee crisis, wreck the economy, violently compete over the remaining limited resources, and slowly succumb to global institutional rot and sclerosis, perpetually collapsing and rebuilding until we slowly peter out or evolve into a more violent, less innovative kind of animal (perhaps something more religious). We’ll never conquer the galaxy. The center will not hold. The world won’t end with a bang, but with a whimper. "
So it may be the end of the line for modern civilization. But if you look at the repeating patterns of the Holocene, the last 10k years, it's not the end of the line for humanity. More likely after 3-500 years of Dark Ages people will come up with new institutions that improves structurally upon the current ones, and then economically and then militarily dominate the rest of the world. For hypersocial species selection works at the civilizational level as well.
Reading your essay made me feel even more confident that the roughly 3 decades the US had with the New Deal coalition was the best form of democracy we ever had. Among other things, it guarded against the vibe voting that is so easily weaponized by populist leaders while still giving working-class people real political representation. And even though the coalition existed only within the Democratic Party, I think it also effectively disciplined Republican Party elites. But once that coalition was uprooted and the state primaries were given greater totalizing leverage within the Democratic Party, voters no longer had any defense against political elites exploiting the vast information asymmetries between themselves and voters.
I’m also pretty sure that populist energy is an inherent property of democracies. So, as you point out, it is a bad idea to structure a voting system that incentivizes and selects for party elites who are better at engaging in populist “good versus evil” rhetoric than at dealing with the real complexity of implementing effective policy in a large-scale, multidimensional society. It means both parties are always vulnerable to capture by populist leaders like Trump.
Great essay as always. Funnily enough, despite naming your publication “Everything is Bullshit”, I usually come away from your essays feeling less pessimistic about the world.
David, amazing analysis. I'd like to know your thoughts on people who vote about elections in their home country AND are immigrants in another country and have been living there for a long time. How would you define this? getting an even "warmer" glow? reinforcing a belief they still belong to a group they are technically no longer a part of? since they've been away of their homeland for so long. I ask, because I don't vote as you may notice and I've always said it's pointless till certain extent, but after reading this, seeing many people still fighting over politics in a place they don't live anymore, it's just nonsense to me. Thx.
Beautiful writing … thanks to my dear friend Will Lindsey for bringing this to my attention. The writer certainly demonstrates many ‘human truths’, including his early statement that he is more cynical than most anybody in the world … and, the ‘truth’ Will Lindsey taught me years ago — there is nearly always an element of TRUTH in all cynicism …. Write on!!
Great post. I agree with your final point about sortition being underrated. I would love to replace the US senate with 500 randomly selected US citizens.
One thing you mentioned is that democracy works well because it is multitribal. IMO a country with only two tribes is in serious trouble. I also hypothesize a country with a million tribes is ungovernable. Any thoughts about what is the best/most practical way to delineate the various tribes? By this I mean: what is the ideal number of tribes, should the tribes be based on geography, policy positions, age, values, relation to the means of production? Should each person be a part of one tribe, or multiple tribes? I have some intuitive notions but I wonder if there's a sound approach to the question.
Thanks, Will, it's a great question. I think there is a big difference between one tribe and two tribes. There is a much smaller difference between two tribes and several tribes, but the latter is still mildly preferable. And a thousand tribes would be even more preferable, insofar as each one has less power and can do less damage. Ideally, the number of tribes would be equal to the number of individuals, which is just another way of saying that the ideal number of tribes is zero. So in general, I think it is good to move in the direction of breaking up and weakening tribes. Tribes are weaker when they overlap with one another, because their members will feel more conflicted and less coordinated, so more tribal overlap is probably better too.
I wonder if anyone’s tested this empirically. Something like the correlation between voter turnout and party affinities or alignment of issues. Eg what would happen in a world where there weren’t parties and issues couldn’t easily be bundled together? I imagine people would still vote, but how much less if at all.
Yes, you’re right that this is one testable implication of my thesis here. I know there’s some empirical support for it in the Diana Mutz book I linked to. I recall there’s some other data out there but I haven’t done a deep dive into the voter turnout literature in some time. I think if there were no parties or group identities or intergroup conflicts related to public policy, voting would almost entirely disappear. But maybe you’d still get some residual voting as virtue signaling or simple statistical illiteracy. Not sure. (Then again, if voting almost entirely disappeared then votes would actually start to matter at the individual level, so it’s kind of a weird thought experiment).
Yeah and to some degree there are still other levels of group identity like “voters” and “citizens” that likely motivate people. These groups probably have weaker ties than from party or policy but enough to still drive that Caplan warm glow; same thing that drives someone to wear an I Voted sticker even if they don’t share what they voted for.
Interestingly, in the birthplace of democracy, ancient Athens, it was quite common for many offices to be assigned via lottery. They had a mixed lottery/election system. Definitely an underrated idea.
Hey have you checked out Robert Burton's work on being certain. It would likely make one even more sceptical of democracy. David Hume was right norms, convention and tradition matter more and whatever people find legitimate.
In an ideal world whole of pol science and adjacent subjects should talk about legitimacy all the time.
The notion of a Lottocracy is a SERIOUSLY underrated idea. There's no perfect system but we can do better than democracy.
Lottocracy with time off for good behavior?
When discussing a proposal to place all nuclear weapons in the hands of an international body, Thomas Schelling accused its proponents of wishing for "government without politics". Whether or not he was correct in that instance, it's a line that's stuck with me ever since I first read it and, well, it's a line I found myself thinking about quite a bit while reading this.
You cannot simply wish away humans' propensity to form groups. It's been a feature of virtually every political system humans have come up with, and probably every one we can come up with. I do think that lottocracy is an interesting idea that is worth trying in some capacity but, well, if you put 100 randomly-selected people in a room together, guess what, they will eventually sort themselves into groups.
Similarly, Pigouvian taxes are probably a genuinely good idea, you just have to sort out the tiny problem of how you're quantifying negative externalities. And it turns out that different people will think certain externalities should be weighted differently, whether due to simple differences of opinion, naked self interest, or something in between. And some of those people might group together to advocate for certain externalities being weighted more or less heavily and whoops, we're back in the messy realm of politics.
It doesn't help that you seem to conflate instances of genuinely irrational behaviour from groups (where most of the members of a group want one thing, but group dynamics push them to do another) with instances of groups simply accurately reflecting the views and interests of their members. i.e., there isn't some secret giant group of people who individually support nuclear power, but whom group dynamics push to oppose it instead; the groups that oppose nuclear power oppose it becauae most of their members oppose it, and they oppose it because they (mistakenly) believe it to be dangerous. If you randomly selected 100 people from the population, they wouldn't suddenly become more pro-nuclear.
Thanks, Ben. Part of my answer to your comment would be repeating my answer to Patrick's comment, so I'd recommend checking out my response to Patrick's comment. The only thing I'd add to my response there is that jurors generally avoid forming tribes, and they generally make pretty good decisions about people's guilt or innocence. They're not perfect, of course, but they tend to be better-informed and more impartial than voters are. So it is possible for people to make important decisions about other people's lives without being tribal, while being relatively well-informed about the details, and while being relatively impartial. Yet this is not how democracy works by any stretch of the imagination, and that is a very bad thing. I think democracy would be better if voters and politicians were more like jurors (rather than mean, dumb tribes), and if the media was more like the prosecution and the defense of a case (rather than a machine for delivering one-sided tribal propaganda). Given that we already have a trial-by-jury system, and it already makes important decisions, and it has not yet devolved into a patchwork of mean, dumb tribes, it suggests that the sort of tribalism you think is inevitable is by no means inevitable. I think if you randomly selected 100 people from the population, and subjected them to the voir dire process to make sure they're impartial (just like we do for jurors), and presented them with the pro-nuclear and anti-nuclear side of the debate, I would bet they'd decide on something more pro-nuclear than the average voter. In general, I think they would make better policy decisions than the average voter on pretty much any policy issue. Yes, we cannot "wish away" the propensity to form groups, but we can try to reduce it and mitigate it. And we can avoid glorifying it and pretending it's virtuous.
Have you ever served on a jury?
It’s probably one of the most traumatizing things one can do. At least the inner city ones that had experience with. The idea that it doesn’t devolve into mean dumb tribes…have you been on a jury?
Anyway, if I was on a jury that determined tax policy I think that I would determine that jury members should pay no taxes and everyone else should pay lots of taxes to jury members (I think we are back to your authoritarian example).
This was entertaining to read and certainly contains some truth, but nonetheless I think it is overly cynical about group behavior, and then simultaneously not cynical enough about "crowds".
In defense of group behavior, it provides the conditions under which the good kinds of "crowds" can emerge (or be regulated into existence). You gesture at this with the traffic light example, but I don't think fully examine its implications. We need rigid rules to coordinate around in order to have a predictable, orderly society where positive sum exchanges can routinely occur. A hypothetical pure crowd with no group-based order around it incentivizes violence, fraud, and abuse of unequal power dynamics.
Compare labor conditions in the early industrial revolution to the post-New Deal era. What broke those were group-based labor unions and political parties advocating for blunt policies like the 40-hour workweek, OSHA, bans on child labor, and the minimum wage. I don't see any other realistic way for the downsides of industrial capitalism to have been curbed. It wasn't going to organically emerge via some "crowd" deliberating on it, and the market itself did not deliver it.
You often need blunt instruments to shove a crowd out of an undesirable equilibrium. Sure, you can imagine a more efficient and intellectually satisfying system of Pigouvian Taxes, but those don't actually solve any of the political "group" problems you mention, because we'd still have to decide what harms count and how to measure them. It's very easy to imagine them not working in practice.
I think you aren't cynical enough about crowds. Even setting aside coercion and fraud, crowds often converge on stupid policies and bad values. For example, you cite NIMBYism as "group" behavior, but actually it seems to me that a lot of NIMBYism emerges through crowd-like behavior. The more local you go, and the more voice individuals have, the more NIMBY the politics are. In fact, YIMBYs have been trying to make the issue of housing more group-oriented (and the YIMBY movement is a much more coherent "group" than NIMBYs, who seem more dispersed and localized). It turns out that blunt, group-oriented state politics are more conducive to good policy outcomes on housing. Why? Because focusing on the blunt goal of "affordable housing and economic growth good" is better than some nuanced deliberative process of local residents who, usually, converge on the idea that they don't want change and like their city the way it is thank you very much. Most people are NIMBYs when the develop is literally in their backyard.
Thanks, Marcus. This is some good pushback. I will grant that NIMBYism vs. YIMBYism is an interesting exception to my thesis. YIMBYs appear to be more thoughtful than the average political group (I’m inclined to say they are unusually crowdy but I also realize that is a bit too convenient for me). I would deny, though, that local politics is that much more crowdy than national politics. I think it’s groups all the way down. I think the point about unions being needed for better treatment of workers is debatable and has been challenged by some scholars, who think these measures were already underway by the time unions were pushing for them (I don’t have an opinion on this, just noting the uncertainty about cause and effect). I would question whether violence and “unequal power dynamics” aren’t themselves group dynamics. I think they are: you need well-coordinated groups to gang up on and dominate people. The problems you’re talking about (I assume) involve groups ganging up on each other—not lone psychopaths preying on people. Most lone individuals are not psychopaths. Most groups tend to act like psychopaths toward enemies and outgroups. That’s why I (like George Carlin) generally prefer individuals to groups, where the odds of psychopathic behavior are lower. Not zero—lower. I would also grant that some crowds have better incentive structures than other crowds, and a crowd that cannot detect and punish psychopaths is one example of such a poorly incentivized crowd. Perhaps there is another continuum, from chaotic collections of individuals on one end, to well-incentivized and wise crowds on the other. Perhaps we need groups to nudge us toward that well-incentivized end, but perhaps not. Either way, I don’t think we can count on groups to do that for us, because they’re too dumb and mean.
This is really good writing. You have a talent for putting these ideas into engaging and readable words.
The pessimistic frame is that all of these problems with democracy are unsolvable, which may be true. I don’t think the answer lies in controlling voters access to the ballot box but instead on the societal narrative of what makes the good life. Pluralism makes this a challenge but the stakes are high. The alternative is nihilism which is doom. Based on all of these problems, why has the US succeeded so far? Caplan himself finds it puzzling that democracy in the US works as well as it has. I find that as a reason to be optimistic and promote optimistic values. The downside case is overstated (definitely in the media chattering class) and putting out positive vibes is the answer in my frame. It’s costless and directly benefits the recipients as well as yourself. The enemy is not the other tribe but nihilism itself. I read your blog as the prosecutor’s case in combatting nihilism. I reject the thesis that democracy generally produces bad results. Bad compared to what? Life in Western democracies is far better today than in was 200 years ago. You can list a bunch of problems that are yet to be solved as evidence of being bad. Without recognizing the benefits and tradeoffs, it’s a meaningless statement. As an analogy, you could say that life is bad because you die. Agree with premise that death is bad but does that make life itself bad? If dying is the cost of living, is that a worthwhile tradeoff?
Thanks, Patrick. Maybe these problems are solvable; maybe they aren’t. The only way to find out is to admit that they are problems and take them seriously. To use your analogy, yes, death is bad. Maybe we’ll never get rid of death. But it would still be good to delay and prevent deaths. Saving lives is good. Increasing life expectancy is good. Halting or slowing the aging process, if that is possible, is worth looking into. We should not glorify death and pretend it’s awesome in the way we pretend democracy is awesome. As another analogy, maybe war will always be with us. But we can still say that war is bad and we should do whatever we can to mitigate it and reduce its likelihood. Likewise, maybe some form of tribeocracy will always be with us. But we can still say that tribeocracy is bad and we should do whatever we can to mitigate it and reduce its likelihood. In fact, the analogy may be deeper than it appears, insofar as war is just politics by other means. As for the “compared to what?” question, I address that in the “Coda” section.
Agreed - truth is not my enemy but nihilism is. I’m optimistic that we have that in common.
I am not sure you should revealing this much truth. I don't know how many people can handle it.
Bryan Caplan's explanation for voting (that it gives us a warm glow) is not incorrect; it is only incomplete because it cites a proximate cause without mentioning a distal, evolutionary cause. Many explanations in the behavioral sciences boil down to people seeking what makes them feel good and avoiding what makes them feel bad. Evolutionary psychology explains why things make us feel good or bad, providing a more complete picture. Thank you for offering a plausible evolutionary explanation for voting.
Groups do suck for many reasons, and you mention some of the important reasons. I especially appreciated the observation about the way thinking in terms of groups leads to black-and-white thinking, a topic I have also written about, https://www.psychologytoday.com/us/blog/cui-bono/201209/the-unsavory-psychology-two-party-politics . One thing that you did not emphasize about groups is just how inefficient and ineffective they are for helping individuals acquire what they want. Just think about how helpful typical faculty meetings are for helping you achieve your personal goals. The "strength in numbers" mentality might have worked for the limited, basic goals of small hominid bands, but I've found that operating as an individual with a limited number of other individuals-as-individuals has been enough for me to lead a pretty satisfying life.
In my view, quasi-rationalistic schemes such as Pigouvian taxes, utilitarianism, and effective altruism are always going to fall short because the number crunching is too difficult for individuals. How in the world are we supposed to accurately predict through computations the future negative externalities of goods and services? Or which acts will bring the most happiness to the most number of people? Or which of the issues addressed by effective charities will be the most important for overall flourishing in the world? (Effective altruists do have data on the efficiency of specific charities, but who knows which of the issues addressed by the various charities will truly matter for the long-run wellbeing of everyone--including nonhuman animals and plants--on the planet?)
Markets, on the other hand do not require us to crunch numbers as individuals. Support for free markets led me to the Libertarian Party for a period of time. Until I saw that the LP was a supergroup party that behaved just as badly as Republicans and Democrats.
When it comes to voting, I am going to keep voting for whoever seems to be the least offensive candidate, even though I realize my vote makes no difference. The advantage I see to voting is that when some asshole gets elected, people can't criticize me for not voting for the person who didn't get elected. No warm glow for me. Just freedom from nagging people.
This is a terrific piece. IMHO the recognition of the overwhelming importance of groups/tribes in social affairs is the beginning of wisdom
I have one point I would like to add.
We can divide society across different access and thereafter there to the one side it the other as a group/tribe (hereinafter tribe)
The resultant tribes have confident loyalties and worldviews and such trinal belief systems near heavily upon social action
IMHO a there is one critically important difference defines what is likely the most important social divide in human history - that between people whose world view is importantly structured epistemic by the inheritance of The Enlightenment and people whose world view is pre Enlightenment. Between people who believe in the importance of evidence and argument and people who do not put whose grasp it's so weak as to be non effective.
I am not suggesting that post enlightenment minds are perfectly rational or indeed that they are particularly rational at all times. I am however suggesting that whilst their commitment to rationality may be more or less intense and their epistemic competence may be more or less thorough they're is a real line here.
The consequences of this divide that splits humanity into two groups are enormous.
Thanks, Pete. You may be right, but I worry that there is something dangerous about this way of thinking--the idea that there are the rational, post-enlightenment folks on one side and the dumb, tribal, benighted folks on the other. This way of thinking is precisely the sort of ingroup supremacy narrative that groups/tribes tend to push on their members. So even if there is some truth to it, I would worry that pushing too hard on it would create precisely the sort of group/tribal boundary that the so-called post-enlightenment folks think they have transcended. I think to avoid the transformation from individual to group, we should try to see the people around us as individuals who are our equals, vulnerable to the same biases as we are, no matter how irrational they might seem. As soon as we start to see a large collection of individuals as inferior in any way (intellectually or otherwise), we start to go down the group/tribe path. What's so insidious about this path is that people cannot see that they are on it and actively deny being on it. No one who's in a tribe sees themselves as being purely tribal. They simply see themselves as individuals fighting for truth and justice, who agree with the wise, virtuous people around them on what is true and just. Avoiding the slide down the group/tribe path is really hard, and I don't even know if I've succeeded, but the only way to even possibly succeed is to be aware of how alluring it is and notice the warning signs.
I respect your concern but IMHO the rangers if ignoring this social reality outright the risks
There is a lot to be said here and it is IMHO critically important. One's perspective on this point influences for example how one conducts politics.
This is an extremely important point and it deserves serious attention. This is likely not the best forum for it. (You might want to look at the small number of pieces I have posted on here about it. If you can face it then start with ‘Two Tribes’)
However…
The implications of not understanding this distinction, unappealing though it may be, are catastrophic. Let me give you a simple example.
When Hilary C debated T the very style of their speech betrayed the audiences they were reaching out to. One participant knew the audience the other confused it.
Forget the negligible intellectual content of T’s contributions. The content isn’t the point. The style however was crucial. It screamed ‘I am one of your tribe.’ The fact that it was a ridiculous lie didn’t matter. His speech was an effective tribal rallying call and that was its only purpose. Such calls overwhelm policy statements.
Hilary’s post Enlightenment rationally presented policy outlines just went right over most people’s heads. All they knew was they she wasn’t one of them.
Democrats make this mistake constantly. They implicitly believe that the ‘others’ are fundamentally like them - they’re just not as good at it..not as articulate…not as well educated…. etc etc etc. But the break from pre to post Enlightenment thinking is not a continuum, there is a very real break. You either consider sound evidence and argument to be epistemically critical or you don’t.
If you don’t understand this your messaging will fail and your election campaign will fail.
Look at some recent campaigns:
Obama ran on hope - not policies
The UK Brexit campaigns could not have been more starkly different:
Leavers - we want our sovereignty back
Remainers - remain. :). possibly the worst campaign messaging ever
==
S. Stevens explores this in his book on the Republican Party elections playbook.
The stakes on not understanding that most of your electorate aren’t especially receptive to well crafted and substantiated arguments are extremely high.
Republicans have a much better grasp on this reality. All their messaging is based on it.
There’s more…
Great deep dive.
1) David Butterfield is interesting to read/listen to regarding Athenian democracy, which was a lot more direct and with an incentive structure giving much more significant accountability
2) James Madison in the Federalist papers has very sharp critiques on groups, or what he calls factions, with notable similarities to your own. Many of the checks and balances set in place in the US Constitution were constructed to add friction to a potential stampede of destructive policies brought on my motivated factions.
3) Joseph Schumpeter predicted in his famous book in the 1940s that Democracy and Capitalism will bring prosperity which will result in people voting in an ever expanding bureaucratic state pursuing various symbolic status games. In this sense although, as you note in your other articles, democracy, free markets, and meritocratic scientific institutions were the magical cocktail that brought the modern world's prosperity, their synergy started to crack with the expansion of the managerial state. Another important point is what Moses of Random Walk regularly points out: all of the current productivity growth comes from AI, which employs a very small portion of the population, while all the employment is growing in nonproductive (economically) health care and elder care services. This is the acceleration of Baumol's Cost Disease as far as the the sharp growth in services and decline of manufacturing. That means the people generating economic productivity don't have much voting influence, but the people drawing money from the state (directly or indirectly, that is by employer funded health care) are a growing bloc. This is concerning as far as the future of this synergy.
Indeed as you note here
https://www.everythingisbullshit.blog/p/mediocrity-as-an-existential-risk
"Stuff will keep getting worse because we won’t know how to fix it, and our stupid attempts to fix it will make things worse. We will make fewer and fewer babies, dwindle in number to the brink of extinction, become increasingly old and decrepit, run out of bold new ideas, over-regulate the few bold ideas we have, technologically stagnate, economically stagnate, lose trust in science and markets, destroy liberal democracy, forget to do anything about climate change, bungle the ensuing refugee crisis, wreck the economy, violently compete over the remaining limited resources, and slowly succumb to global institutional rot and sclerosis, perpetually collapsing and rebuilding until we slowly peter out or evolve into a more violent, less innovative kind of animal (perhaps something more religious). We’ll never conquer the galaxy. The center will not hold. The world won’t end with a bang, but with a whimper. "
So it may be the end of the line for modern civilization. But if you look at the repeating patterns of the Holocene, the last 10k years, it's not the end of the line for humanity. More likely after 3-500 years of Dark Ages people will come up with new institutions that improves structurally upon the current ones, and then economically and then militarily dominate the rest of the world. For hypersocial species selection works at the civilizational level as well.
Reading your essay made me feel even more confident that the roughly 3 decades the US had with the New Deal coalition was the best form of democracy we ever had. Among other things, it guarded against the vibe voting that is so easily weaponized by populist leaders while still giving working-class people real political representation. And even though the coalition existed only within the Democratic Party, I think it also effectively disciplined Republican Party elites. But once that coalition was uprooted and the state primaries were given greater totalizing leverage within the Democratic Party, voters no longer had any defense against political elites exploiting the vast information asymmetries between themselves and voters.
I’m also pretty sure that populist energy is an inherent property of democracies. So, as you point out, it is a bad idea to structure a voting system that incentivizes and selects for party elites who are better at engaging in populist “good versus evil” rhetoric than at dealing with the real complexity of implementing effective policy in a large-scale, multidimensional society. It means both parties are always vulnerable to capture by populist leaders like Trump.
Great essay as always. Funnily enough, despite naming your publication “Everything is Bullshit”, I usually come away from your essays feeling less pessimistic about the world.
This essay by Dan Williams dovetails well with DiB: https://persuasion1.substack.com/p/how-to-understand-factual-polarization?r=6p0s35&utm_medium=ios
David, amazing analysis. I'd like to know your thoughts on people who vote about elections in their home country AND are immigrants in another country and have been living there for a long time. How would you define this? getting an even "warmer" glow? reinforcing a belief they still belong to a group they are technically no longer a part of? since they've been away of their homeland for so long. I ask, because I don't vote as you may notice and I've always said it's pointless till certain extent, but after reading this, seeing many people still fighting over politics in a place they don't live anymore, it's just nonsense to me. Thx.
Thanks, Mauro. Not sure, but I suspect you’re right that it’s about maintaining ties to their old group.
Beautiful writing … thanks to my dear friend Will Lindsey for bringing this to my attention. The writer certainly demonstrates many ‘human truths’, including his early statement that he is more cynical than most anybody in the world … and, the ‘truth’ Will Lindsey taught me years ago — there is nearly always an element of TRUTH in all cynicism …. Write on!!
Great post. I agree with your final point about sortition being underrated. I would love to replace the US senate with 500 randomly selected US citizens.
One thing you mentioned is that democracy works well because it is multitribal. IMO a country with only two tribes is in serious trouble. I also hypothesize a country with a million tribes is ungovernable. Any thoughts about what is the best/most practical way to delineate the various tribes? By this I mean: what is the ideal number of tribes, should the tribes be based on geography, policy positions, age, values, relation to the means of production? Should each person be a part of one tribe, or multiple tribes? I have some intuitive notions but I wonder if there's a sound approach to the question.
Thanks, Will, it's a great question. I think there is a big difference between one tribe and two tribes. There is a much smaller difference between two tribes and several tribes, but the latter is still mildly preferable. And a thousand tribes would be even more preferable, insofar as each one has less power and can do less damage. Ideally, the number of tribes would be equal to the number of individuals, which is just another way of saying that the ideal number of tribes is zero. So in general, I think it is good to move in the direction of breaking up and weakening tribes. Tribes are weaker when they overlap with one another, because their members will feel more conflicted and less coordinated, so more tribal overlap is probably better too.
I wonder if anyone’s tested this empirically. Something like the correlation between voter turnout and party affinities or alignment of issues. Eg what would happen in a world where there weren’t parties and issues couldn’t easily be bundled together? I imagine people would still vote, but how much less if at all.
Yes, you’re right that this is one testable implication of my thesis here. I know there’s some empirical support for it in the Diana Mutz book I linked to. I recall there’s some other data out there but I haven’t done a deep dive into the voter turnout literature in some time. I think if there were no parties or group identities or intergroup conflicts related to public policy, voting would almost entirely disappear. But maybe you’d still get some residual voting as virtue signaling or simple statistical illiteracy. Not sure. (Then again, if voting almost entirely disappeared then votes would actually start to matter at the individual level, so it’s kind of a weird thought experiment).
Yeah and to some degree there are still other levels of group identity like “voters” and “citizens” that likely motivate people. These groups probably have weaker ties than from party or policy but enough to still drive that Caplan warm glow; same thing that drives someone to wear an I Voted sticker even if they don’t share what they voted for.
Interestingly, in the birthplace of democracy, ancient Athens, it was quite common for many offices to be assigned via lottery. They had a mixed lottery/election system. Definitely an underrated idea.
Hey have you checked out Robert Burton's work on being certain. It would likely make one even more sceptical of democracy. David Hume was right norms, convention and tradition matter more and whatever people find legitimate.
In an ideal world whole of pol science and adjacent subjects should talk about legitimacy all the time.
Thanks, haven’t read it but it will check it out.