Democracy for Realists
How Democracies Die
How Democracy Ends
Danny Oppenheimer and Mike Edwards. Democracy Despite Itself: Why a System That Shouldn't Work at All Works So Well. MIT Press, January 2012.
The "paradox of democracy" is that democratic societies work pretty well even though voters are biased and irrational, meaning they make "inconsistent or imperfect decisions." Some reasons why democracy works out:
- "mass participation in contested elections creates psychological pressures for individuals to be better citizens and for politicians to be better leaders"
- "regularly alternating power between different factions helps to avoid political instability"
- "individuals and crowds can overcome their ignorance and make informed decisions"
- "people will occasionally punish politicians, thereby helping to curb the worst abuses of the public trust"
But here's what goes wrong:
Voters are often ignorant of relevant facts, the question at stake, and where the candidates stand. (Even experts are often ignorant, such as when Alan Greenspan admitted after the 2008 financial crisis that he hadn't understood important things.) Thus they "often are ignorant and foolish pawns of a system they don’t understand." (Note: Being ignorant of relevant facts is generally bad, but being ignorant of irrelevant facts may help narrow their focus.)
All kinds of heuristics are at play:
Confirmation bias, in which people "seek out information that supports what we already believe, and actively avoid information that contradicts our current opinions." One study showed that after being provided with information on two sides of a debate, people became more entrenched in their preexisting opinions. [Update: Not necessarily in the book, but discussed in many other places including this 2018 article, confirmation bias is an important driver of Internet algorithms, since they tend to replicate a user's past behavior.]
They tend to forget the details of why they formed an opinion about a candidate and just remember the positive or negative impression. They may go with whatever brand they recognize, even if it's lower quality. They may abdicate responsibility for forming their own opinion and follow the opinion of a trusted expert.
People follow the crowd and do things just because everyone else is doing them, like drink too much at parties ("pluralistic ignorance"). Or they behave in a "principal-agent" dynamic in which the principal (a car owner, for example) hires an agent (a car mechanic) to do something and must trust them because it is difficult to check their work, opening the door to "corruption and inefficiency."
Media bias occurs because journalists are humans writing for humans. They write about what already interests people (plane crashes, not car crashes). This magnifies distorted beliefs in the public imagination about, for example, how people are really more likely to die (car crashes, not plane crashes). There's also no such thing as perfect balance in media representation especially given that people hold false beliefs or opinions that do not depend on facts. Should someone who is demonstrably wrong be given any air time at all? Does it matter if they have a large following? "There is simply no way to give balanced coverage of an imbalanced world."
Hearing about death makes people more conservative. Seeing pictures of schools biases them in favor of education. Physically leaning to the left or right or using their left or right hand biases them toward the metaphorical "left" or "right" of the political spectrum. They'll consider shoplifting as a major or minor crime depending on whether they consider it alongside lesser or greater crimes. When given a spectrum of possible options, they'll choose something middle-of-the-road on the given scale. When asked to grade politicians, they think the lower end of a -5 to 5 spectrum is worse than the lower end of a 0 to 10 spectrum. They prefer initiatives phrased as a two-out-of-three chance of a good outcome rather than a one-out-of-three chance of a bad outcome. They prefer simple, easily understood statements that are repeated, and believe those to be the smartest. They agree to initiatives phrased as A because B even when B is not a justification for A and especially when B is it is good for children and families. They form opinions on initiatives depending on how they are named ("estate tax" but not "death tax"). They value consistency so much that they will consent to strange requests if it maintains an identity or relationship that they've started to form, but they can be inconsistent about their strongly held principles when a particular example catches their imagination.
Some irrationality has to do with ingrained moral behavior. People will jettison their rational self-interest in favor of an arrangement that seems more fair or principled, as in preferring to pay higher taxes to their own leader rather than lower taxes to a foreign leader. They'll voluntarily use less water during a drought when they believe water prices are fair. They'll support or oppose foreign reporting depending on whether they're first asked if their own nation's reporters should be allowed to report from other countries or if foreign reporters should be allowed into their own country. If they do not feel that all votes are counted equally, the nation tends toward "instability, unrest, and chaos."
(Here, I personally would quote Chuck Klosterman in his 2013 book I Wear the Black Hat: Grappling With Villains (Real and Imagined), paraphrasing Thomas Frank in What’s the Matter with Kansas?: “Voters don’t know who they are. They don’t view themselves objectively, because no person can. Instead, they see themselves as a self-generated projection of who they could be, striving for whatever best-case scenario they consider plausible. The contradiction between what they think and how they feel does not pose a problem; as far as they can tell, no contradiction exists.”)
They also favor candidates for unprincipled reasons, such as whether the candidate's name is pronounceable. (Even when the candidate's name is easy, like "John Kerry," a voter may write a misspelled version of "John Edwards" on their ballot.)
There are different ways to run an election. None is perfect. "Closed primaries and intraparty elections can cause a political system to reject popular, moderate choices. Open primaries open the door to manipulation by the other side." You can pick a single ballot at random and declare it the winner, which has the benefit of giving a greater chance to minority groups who would otherwise never win the popular vote, but also of generating resentment at the perception of unfairness. You could also entirely do away with elections and draft random citizens to serve in the government, which eliminates the threat of challenges to power transitions, but it will result in some very unfit politicians indeed.
There are problems specific to voting. In the United States, candidates are given credibility by winning early opinion polls that aren't necessarily representative of the electorate, and losers drop out even before the primary election. In the primary, people sometimes vote strategically (based on who they think is ultimately electable, or who they think will most harm the other party) rather than idealistically, and they also vote strategically in the general election for the "lesser of two evils." (Sometimes the two evils are both very bad indeed, even in democracies.) Polls are often phrased poorly (such as who are you going to vote for rather than who most closely matches your ideals), and "there is no such thing as neutral wording." The most liked candidate can lose.
Democracy has some good outcomes. Democracies don't go to war with other democracies. As Churchill put it, "Democracy is the worst form of government, except for everything else." And democracy itself has been a kind of self-fulfilling prophecy in which people "created the outcome that they expected to happen." The more we think about the importance of equality, the more work we do to make our society more equal. With a large enough population, random anomalies are statistically eliminated and thus the more moderate and informed voters will win the election. (But that assumes that most voters are indeed informed, and the book mostly argued against that assumption. Another problem is that not all extremist views are random; there might be a large number of extremists and they might appear clustered at one end of the spectrum.)
Also, democracy forces politicians to at least try to maintain the appearance of legitimacy (by obeying the courts, for example), to avoid accepting money or endorsements from unsavory organizations, to avoid offending even minority populations (who can change the outcome of an election), and also to consider what the majority of voters want even when it's hard to gauge popular opinion. (As on The Price Is Right gameshow, "One audience member will shout 'two hundred dollars,' another will shout 'one hundred fifty dollars,' a third will shout 'five hundred dollars,' and some prankster in the first row will shout 'thirty thousand dollars.' Of course what the contestant actually hears is 'two fifty hundred five thirty dollars.'") It stabilizes a society when "the most powerful people in the country regularly trade power back and forth between them." And, although starting a war temporarily boosts a leader's popularity, this effect doesn't last long. Politicians, however, have to fear for their employment at the hands of an electorate that is in large part irrational.
We can observe how public opinion correlates with policy, but this is not the same as a fuller understanding of "why public opinion changes" and why, in that situation, the public may choose either to tolerate the politician or vote the politician out.
"The brilliance of democracy does not lie in the people’s ability to pick superior leaders. It lies in the many ways that it subtly encourages the flawed people and their flawed leaders to continually work toward building a better society. ... Voters are driven more by snap judgments and gut feelings than by a rational consideration of the merits of the arguments. We are fed misleading information by the media and the polls, and then told to use that information to go and vote in elections that are subtly manipulated by the very people whose names appear on our ballots. Then we expect our politicians to understand what we want and come up with coherent policies to faithfully represent the popular will." Yet it works because people believe it is "fair," that it employs "processes that spread power among the many" which somehow compensates for the fact that the electorate is ignorant. "Policies do matter, but they don’t matter nearly as much as the simple fact that we are all working together to create them." (Incidentally, the authors think that eliminating the Electoral College would improve the perception of the fairness of U.S. democracy since it only "undermine[s] the fairness of presidential elections" by "rubber-stamping the popular vote total" except for two instances, as of the book's publication date, when it declared a winner who had not won the popular vote.)
Christopher H. Achen and Larry M. Bartels. Democracy for Realists: Why Elections Do Not Produce Responsive Government. Princeton University Press, August 2017.
Democracy is very popular. Achen and Bartels write that "most people almost everywhere accept the proposition that their own political system is (somehow) democratic — and even more accept the proposition that democracy is (somehow) a good thing." Democracy is often believed to be essential to human dignity. Yet: "What makes a country democratic and why that is a good thing have generated much less agreement." Indeed, existing concepts of democracy require "willful denial of a great deal of credible evidence." People live our this denial in part by holding contradictory beliefs" "For example, a substantial majority of Americans say that democratic government is a very important factor in the nation’s success; but most also believe that 'the government is pretty much run by a few big interests looking out for themselves.'" We do not seem to have a way to more accurately analyze and interpret what happens in our societies so that we can resolve these contradicting beliefs and approach the ideal. Furthermore, if democracy is to be rehabilitated in practice, then "it must at least be the case that the ideals are not too unrealistic." The authors want to abandon "simplistic faith." They aim to "document the gap between democratic ideals and realities." They endorse what they call the "group model" and they say they want to reduce the influence of money in politics insofar as it sways voters away from principled decisions.
The book was submitted for publication over a year before Trump was elected. The formulas in their book would have predicted that Clinton would win the popular vote by just over 2 percentage points, which she did, but the authors admit not having predicted the Electoral College outcome in favor of Trump.
The populist or "folk" ideal of democracy, "derived from Enlightenment concepts of human nature and the political views of 19th-century British liberalism," assigns a policy-guiding role to citizens. The idea of popular sovereignty gives "legitimacy and stability to political systems whose actual workings are manifestly — and inevitably — rather less than divine." A famous essay by Philip Converse in 1964 argued that most ordinary citizens “do not have meaningful beliefs, even on issues that have formed the basis for intense political controversy among elites for substantial periods of time” and therefore, as Achen and Bartels explained it, that "the political 'belief systems' of ordinary citizens are generally thin, disorganized, and ideologically incoherent" compared to what the folk ideal requires of them.
In 1982, William Riker feared this ideal was a "totalitarian sleight-of-hand...used to justify coercion in the name of temporary or spurious majorities". The authors furthermore find that this model "obscures or ignores" how individuals' identities affect their voting and party loyalty.
Whereas kings used to explain away errors as the result of poor advice from humans so that the error did not challenge their divine status, democracy (believing its own ideas to be nearly divine and incorruptible) blames the influence of "special interests" for "mistranslation or temporary deflection of the people's will" rather than admitting the people themselves might be unable to make good decisions. The authors want to dispense with "cheerful illusions and wish fulfillment" and instead begin "grappling seriously with all the evidence undermining the standard versions of democratic theory."
"The most systematic and sophisticated instantiation of the populist ideal is the 'spatial model' of voting and elections," the authors write. This holds that individuals will vote for the party that is ideologically closest to them on a spectrum and that parties adopt platforms that will get them elected. Of course, this requires the voters to know what they want, a premise that is dubious at best. If the model is correct, the implication is that "both parties will adopt identical platforms" based on the way voters are going to vote. (This is called the "median voter theorem.") Anthony Downs argued for the spatial model in 1957 and the theory seemed to be upheld "when more extreme candidates emerged on the political scene — Barry Goldwater in 1964, George McGovern in 1972 — [as] they were trounced at the polls, just as the theory suggested they should be." But subsequently there has been more political polarization. Furthermore, research "has suggested that Goldwater and McGovern’s losses had less to do with their issue positions than with election-year economic conditions; ideological 'extremism' probably cost them just a few percentage points of the popular vote". The authors also note a partisan shift in voter's allegiances based on racial identity that had begun to occur by this time, where southern blacks gravitated more toward the Democratic Party and southern whites toward the Republican Party.
While voting on the issues "is widely regarded as a hallmark of good citizenship," in the real world, individuals vote for reasons other than the candidates' issues. Another problem with this model of democracy is that, in the real world, politicians, once elected, do not modify their behavior to act as the "median voter" would want them to act.
Even when this theory first emerged in the 1950s, most people "had distinct — indeed, virtually uncorrelated — views about economic, social, and foreign policies." This model works best when the political spectrum has only one dimension, which does not reflect the real world. The real world presents us with multiple options and there is rarely a single clearly favored option that has majority support above all possible alternatives. This reveals the populist ideal as "logically incoherent " and means "there may be no policy platform with a logical claim to represent 'the will of the majority,' much less 'the will of the people.'" Charles Plott wrote in 1976 that the notion of "what society wants or should get" is basically "a thing that cannot, in principle, exist".
Spatial models also tend to assume that voters know a lot. One popular model assumes that even "uninformed" voters know where candidates stand on a left-right spectrum and they also know the opinions of these candidates held by "informed" voters, with which information they "infer the candidates' positions." The authors wryly observe that "most voters do not know what political 'left' and 'right' mean, much less know what informed voters think."
If voters are even slightly more likely than chance to make a good decision, then it stands to reason (as the Marquis de Condorcet showed) that a large group of voters is very likely to reach a good decision. Unfortunately, we can't rely on that smidgen of intelligence from each individual because voters leverage a herd mentality. Millions of people "misconstrue the same relevant fact or are swayed by the same vivid campaign ad," and they make decisions based on limited founts of information, some of which may even be intentionally "censor[ed]" or "distort[ed]" by the government or may be the result of unintentional errors that have been highly publicized. As a result, "the range of policies that can be sold to a majority of the voters may be very broad indeed". This may be how people prefer to behave; in any case, "[m]ost ordinary citizens do not want politics to be more like a philosophy seminar" and many are "intensely averse to political disagreement".
Since the electorate, as E. E. Schattschneider put it in 1942, is "a sovereign whose vocabulary is limited to two words, 'Yes' and 'No,'" the way political questions are framed becomes very important to accomplishing any given mission. In the 1970s, only a quarter of Americans would "forbid" a Communist to give a speech, but twice that number would "not allow" the same action. In the 1980s, less than a quarter of Americans said the government should increase "welfare," but nearly two-thirds said it should increase "assistance to the poor." Before the 1991 Gulf War, less than a third of Americans wanted to "go to war," but twice as many wanted to "use military force." We know, then, how to frame these questions based on how we want people to answer. While such manipulation, however, may convince people to let you do something, it does not reveal what they really want. Which of these options represents the majority will? "We can suggest no sensible way to answer that question," the authors say. Thus, crucially: If the majority does not have a preference, how can democracy claim to advance the will of the people?
Sometimes, even when the difficult question is better explained, the electorate continues to be attracted to an answer that contradicts democratic principles. A survey in 2014 found that a third of Americans said that significant corruption would justify a military coup.
Voters like the idea of term limits for politicians even though this just moves power from established politicians to their up-and-coming aides ("a different elite coalition") and does not directly give power to the people. Moreover, "it is not deliberate policy change".
California has introduced an "open primary" election in which candidates from all parties are on the same ballot, and the top two winners regardless of party will face off in the general election. The authors argue that this system requires voters to have even more "fine-grained knowledge" of the candidates' positions and that it has not led to more political centrism.
They question whether it is necessary to have primary elections at all. Cannot party elites choose the candidates they want to run? After all, "well-informed commitment to democratic values" is found in only "a sliver of the population," and it is this elite that comprises the "public conscience," so why shouldn't an elite group be trusted with political decisions?
The election-oriented model means that people do not actually rule themselves but can choose their rulers. As John Dunn put it in 1999, it is necessary to be ruled, but demanding accountability "softens its intrinsic humiliations". The "retrospective" theory of voting is today's most influential model. In this view, voters focus on whether to keep or reject the incumbent leaders on the basis of their performance, not on the basis of ideals. This explanation is attractive to many because "it puts much less pressure on the voters to have elaborate, well-informed policy views." The authors nevertheless point out its flaws.
In U.S. presidential elections, "voters do indeed reward or punish incumbents for real income growth," but pretty much only that "in the months just before each election." They have "short memories and little apparent regard for ideology or policy" even during large-scale crises like the 1930s Depression.
For the retrospective model to function as intended, voters need to be able to do the following things:
- Recognize when they are in a good or bad situation. (Is your neighborhood "safe"? How do you know?)
- Know whether there is, in principle, a political solution to it. (Your whole city might be unsafe because it was built at the base of an active volcano, but there's nothing the mayor can do. Citizens may need to take individual responsibility to decide whether to continue living there or to leave.)
- Be willing, on election day, to disregard concerns that have no cause or solution within the scope of the political office in question. (U.S. voters in oil-producing states should not use global oil prices to determine whether to reelect their governors. A 2002 study showed that they do.)
- Be able to identify, if there is indeed a political solution, exactly which politicians should be responsible. ("In politics, thousands of cobblers are at work, and pinches [from ill-fitting shoes] come and go, many with nobody at fault. Ordinary citizens with limited free hours in their day are very unlikely to learn who deserves credit or blame for the pains they suffer." Voting out a politician who couldn't possibly have helped anyway is like "kicking the dog after a hard day at work.") (Update: Not in the book, but in a 2018 study of climate data, it seems that Roman emperors were more likely to be assassinated during droughts. There may have been less food and higher anxiety during a drought, but there was probably little the emperor could have done to remediate the problem. Hence, where weather is concerned, killing the emperor seems like kicking the dog.)
- Be willing to reward good incumbents with reelection as well as to punish bad incumbents with loss of power, and do this in a roughly fifty/fifty split so that the half of politicians who are above-average get elected. (If politicians think it's too easy or too hard to get reelected, they won't be properly motivated. Besides, when voters punish a politician or party, it "can neither relieve their distress nor produce more competent or caring political leaders.")
- Gauge what politicians have "above-average" performance. (Tricky, since each politician faces their own set of specific challenges in their own age and it's hard to imagine how a perfectly "average" politician would have reacted.)
- Consider the incumbent's performance for the entire duration of his or her term, not just in the months leading up to the election. (In the real world, voters have very short memories. This reduces the incentive for politicians to perform well every day.)
- Don't forget the policy lessons learned immediately after the election.
"Blind retrospection" is when voters "punish incumbents for conditions beyond their control." Commonly, in the twentieth-century United States, voters have punished incumbents for recent damaging climate or natural disasters. If voters learn from natural disasters that they want "more government intervention" to help cope with the crisis, such realizations "ought to push electorates to the ideological left. What they actually do, however, is reduce support for incumbents regardless of their ideological commitments." And it isn't reasonable to justify this by saying that all of these incumbents gave "substantially worse than average" disaster responses. Some of them had to have been above average. But they, too, were punished. Sadly, incumbents then have no election-day incentive to prepare well for disasters.
Voters tend to reward incumbents for income growth in the six months leading up to election day. Louis Uchitelle in 2004 likened this to what happens when the music stops during Musical Chairs. Politicians know this, because, accordingly, "the average election-year income growth rate exceeds the corresponding GDP growth rate by 40% (2.9 versus 2.0 percentage points)" and in the late twentieth century the average income growth rate was over 1 percent higher "in presidential election years than in other years."
Thus, while retrospection may seem easier than ideology, it is "not easy enough" and "it simply does not work" to improve government.
The authors endorse the "group" theory which more realistically reflects "social identities and group attachments figuring crucially in their political loyalties and behavior." They appreciate the critics of the individualism of the "French Enlightenment...British liberalism...[and] American Progressivism" for seeing that "human life is group life." "For most people," they write, "partisanship is not a carrier of ideology but a reflection of judgments about where 'people like me' belong."
Columbia University studies in the mid-twentieth century showed that "showed that group memberships — being a Protestant rather than a Catholic, a union member rather than not, or a white person rather than an African American or other minority — powerfully shaped vote choices. People adapted their ideas to those of the presidential candidate they favored, or if they were less informed, they simply assumed (sometimes incorrectly) that his ideas matched their own."
What does it mean to be well-informed?
"Even on purely factual questions with clear right answers," the authors write, "citizens are sometimes willing to believe the opposite if it makes them feel better about their partisanship and vote choices." They give the example of President Bill Clinton halving the federal budget deficit during his first term. No one, not even Democrats and independents, were clear on the fact of this decrease, and a majority of Republicans believed the deficit had increased. This suggests a partisan inclination to view the Democratic president negatively. ("Democrats behaved in much the same way on other issues," the authors write.) A lack of facts underlies much apparent irrationality. Of course, voters may not be highly motivated to answer mere surveys correctly, but why, then, the authors ask, should we imagine that voters are more highly motivated to educate themselves before taking real political action? Ilya Somin asked in 2013 "whether the modern electorate even comes close to meeting the requirements of democratic theory."
James Bryce noted in 1894 "how little solidity and substance there is in the political or social beliefs of nineteen persons out of every twenty. These beliefs, when examined, mostly resolve themselves into two or three prejudices and aversions, two or three prepossessions for a particular leader or party or section of a party, two or three phrases or catchwords suggesting or embodying arguments which the man who repeats them has not analyzed." The statistic "nineteen persons out of every twenty" was validated by Converse who classified only 3 percent of voters as "ideologues," while most voters and especially non-voters appeared to think "in terms of group interests or the 'nature of the times,' or in ways that conveyed 'no shred of policy significance whatever,'" as Achen and Bartels put it.
Hence, H. L. Mencken in 1916 called democracy "the theory that the common people know what they want, and deserve to get it good and hard."
Being "moderately well-informed" sometimes worsens the problem. In the real world, people learn first about what their partisan loyalty expects them to believe, and only after that what actually is true. So when people begin to educate themselves, they learn how to "spin" before they learn how to think independently and objectively. And what of the minority who curates accurate, relevant information? The authors believe that "even the most informed voters" tend to vote their identity rather than their ideology. Because an election "reflects and reinforces voters' social loyalties, it is a mistake to suppose that elections result in popular control of public policy." Elections are therefore "mostly just erratic reflections of the current balance of partisan loyalties in a given political system. In a two-party system with competitive elections, that means that the choice between the candidates is essentially a coin toss."
When one's political allegiance begins to reveal conflict between one's identity and one's belief, identity usually wins out. This means that, if a political party challenges the previously held policy opinions of its constituents, if the constituents identify strongly with that political party then they will likely just change their opinions about policy to match whatever the party is currently promoting. But if the party seems to be squeezing out people from a certain identity group (apart from the political party's identity itself), those people will tend to defect to another party. The authors give the example of the abortion rights debate in the US in the late twentieth century. When the Republican Party challenged abortion rights, men did not feel that their identity was threatened and thus Republican voters who were men simply adjusted their political opinions on this issue to match the Republican Party's, but women felt rejected in their identity as women and thus many Republican voters who were women defected to the Democratic Party which more closely reflected their existing beliefs. Both results were predictable.
Identity played significantly in the 2016 primary election between Clinton and Sanders. Sanders "did eleven points worse among women than among men, eighteen points worse among nonwhites than among whites, and a whopping twenty–eight points worse among those who identified as Democrats than among independents."
In 1914, Walter Lippmann said: "Once you touch the biographies of human beings, the notion that political beliefs are logically determined collapses like a pricked balloon."
Even "well-organized 'ideological' thinking often turns out to be just a rather mechanical reflection of what their favorite group and party leaders have instructed them to think," the authors write. For example, the American public did not have opinions on the privatization of Social Security until this became the subject of hundreds of ads just before Election Day 2000 in the Bush/Gore campaign. The result was the doubling of the statistical relationship between voters' opinions on this policy question and their preferred political candidate; however, this was "due not to changes in vote intentions, but to Bush and Gore supporters learning their preferred candidate’s position on the issue and then adopting it as their own." For a far more consequential example, the authors say that Germans liked Hitler because he presided over a dramatically improving national economy ("It is not easy to imagine a clearer example of blind retrospection, nor a more damaging blow to retrospection as a defense of democracy"), and many Germans who had not been anti-Semites to begin with adopted anti-Semitic views because Hitler was already their preferred politician.
To the extent that good ideological thinking ever exists, it tends to be undemocratic, left in the hands of a few experts and dilettantes by its very nature, where the majority of people have no interest in joining that cocktail party.
People do not always analyze complex issues to the end of their logical implications. They sometimes choose to eliminate essential safety services such as fire departments (as in one cited case in Illinois), not realizing that the tax savings might be eaten up by higher insurance premiums and disaster recovery. Similarly, in the mid-twentieth century, voters listened to "crackpots, rogue doctors, and extreme right-wing interest groups" that opposed adding fluoride to drinking water, and many localities rejected the proposal, in effect accepting likely higher dental bills in the future for a few pennies of tax savings in the present.
Philip Converse noted in 1962 that the disproportionate power of undecided voters, so-called "swing voters," to determine elections by their shift in opinion at the last minute basically points to them as being "the least informed members within the electorate who seem to hold the critical balance of power".
The authors write:
"Gifted in their own spheres, artists and intellectuals have no special expertise in politics. In our political judgments and actions, we all make mistakes, sometimes even morally indefensible errors. Thus, when we say that voters routinely err, we mean all voters. This is not a book about the political misjudgments of people with modest educations. It is a book about the conceptual limitations of human beings — including the authors of this book and its readers.
"As we have seen, effective democracy requires an appropriate balance between popular preferences and elite expertise. The point of reform should not simply be to maximize popular influence in the political process, but to facilitate more effective popular influence. We need to learn to let political parties and political leaders do their jobs, too. Simple-minded attempts to thwart or control political elites through initiatives, direct primaries, and term limits will often be counterproductive. Far from empowering the citizenry, the plebiscitary implications of the folk theory have often damaged people’s real interests."
Steven Levitsky and Daniel Ziblatt. How Democracies Die. Crown, January 2018.
Prompted by Donald Trump’s election and first year in office, How Democracies Die by Steven Levitsky and Daniel Ziblatt is a timely read. We are told that the Republican and Democratic parties are both at fault for the recent breakdown of democratic norms, but the authors blame Republicans more. Twentieth-century anecdotes illustrating antidemocratic tendencies within the United States, South America, and Europe give a wider historical context and a quick overview of some aspects of authoritarianism. The book is capped with recommended strategies to defend democracy. It is educational, useful, and enjoyable despite the weight of the topic.
A litmus test for authoritarians
When a leader hits any of these criteria of authoritarianism, it is reason to worry:
- Rejecting the rules
- Denying opponents’ legitimacy
- Tolerating violence
- Threatening to restrict civil liberties
By the authors' count, Trump hit all four before he even took office. Furthermore, during his first year, he revealed authoritarian instincts through the strategies of “capturing the referees” (which they define as dismissing nonpartisan officials and selectively enforcing laws), “sidelining the key players” (for example, by intimidating journalists), and “rewriting the rules” (changing the laws). Trump violates other unwritten norms regarding nepotism and divestment from personal business. "A country whose president attacks the press, threatens to lock up his rival, and declares that he might not accept election results," they warn, "cannot credibly defend democracy.”
Avoiding authoritarian politicians
In the past, large percentages of Americans have supported extremist appeals from men like Charles Coughlin, Henry Ford, Huey Long, Joseph McCarthy, and George Wallace. Amidst such popular support, the authors say, political parties should function as “democracy’s gatekeepers” to filter out those unfit to lead. This strategy is not foolproof, however. Establishment leaders do not always speak from conscience. They may reject riskier candidates in favor of more electable ones or, in a worse twist, make fateful alliances with up-and-coming demagogues in a bid for short-term gain. They may validate the demagogue either because he shares some of their agenda or because they believe they can control him. He may then seize power amidst “collective abdication” of responsibility.
This helps explain the rise of Benito Mussolini, Adolf Hitler, and Hugo Chávez, among others. “It’s hard to find any evidence of majority support for authoritarianism in 1920s Germany and Italy,” the authors write, a time when under 2 percent of the population claimed membership as Nazis or Fascists. Similarly, when Hugo Chávez was elected in Venezuela, only a quarter of Venezuelans believed that authoritarianism could ever be better than democracy, and “public support for democracy was higher there than in Chile — a country that was, and remains, stably democratic.” These dictators got their foot in the door not primarily because of popular support for their ideas, Levitsky and Ziblatt argue, but because establishment politicians helped usher them in.
Who, then, should decide a candidate’s political future: the people or the party elites? Too much party control is inherently undemocratic. Yet if the people choose an authoritarian, their democracy may need paternalistic protection bestowed by someone with more foresight. “There is no escape from this tension,” Levitsky and Ziblatt say.
Polarization: A threat to democracy
American democracy has many checks and balances, “hard guardrails,” enshrined in law. These include tools like the president's executive orders, pardons, and “court packing” as well as Congress’s filibuster, power of advice and consent, and impeachment. Levitsky and Ziblatt focus instead on unwritten norms, what political scientist Donald Matthews calls “folkways,” the “soft guardrails” that bolster democratic culture. History has taught us that institutions rely heavily on "shared understandings of appropriate behavior.” Since the founding of the United States it has been necessary to create such workable norms.
Polarization based on identity or ideology tends to degrade norms. In the three decades leading up to the Civil War, against a background of disagreements over slavery, there were over a hundred violent incidents in Congress. Such a culture could only have contributed to the lead-up to war. Afterward, political stability “came at the price of racial exclusion and authoritarian single-party rule in the South.” The more complete democratization fulfilled through the Civil Rights movement in 1965 simultaneously kicked off another round of increasing polarization which in turn has given us the escalating violation of democratic norms we see today. In other words, a democratic nation must have inclusive voting rights, but the push for inclusion can itself set off power struggles that endanger democracy. Building a multiracial society is necessary in principle but is not always a safe, stable process.
Polarization, at least ideally, can be dialed down. Levitsky and Ziblatt identify an essential Golden Rule. Their terms are “mutual toleration and institutional forbearance,” but those aren’t very catchy. What we might prefer to call the Golden Rule goes roughly as follows:
Don’t treat your political rival as an existential enemy in thought or in action, and don’t assume they will treat you as such. Don’t have a fundamentally hostile attitude toward them. Don’t pull out all the stops to use all the legal weapons at your disposal; avoid the “weaponization” of checks and balances. View your rival as legitimate. Expect some disagreements and disappointments but also expect good faith and a willingness to stay in the game (“mutual toleration”). Fight your battles with the restraint that has previously served the culture well (“institutional forbearance”).
Commitment to the Golden Rule facilitates business as usual. If an opponent begins fighting dirty, one should avoid responding in kind and escalating the conflict. The authors say that “scorched-earth tactics” in the long-term keep the two sides fighting. The party in power rationalizes crackdowns, moderates flee the scene, and this “plays directly into the hands of authoritarians.” For this reason, the authors fear an impeachment of Trump if it were to be perceived as having a partisan, obstructionist agenda.
Planning for the future
What will happen next in the United States? Will democratic norms rapidly restore themselves or will they implode under the pressure of white nationalism? The authors think the more likely near-term outcome is a gradual descent into increased polarization and “democracy without solid guardrails.”
The authors reject the proposal that Democrats should strategically deemphasize multiracial voices and interests and attempt to regain the support of more white working-class people. Pandering to white supremacist assumptions and agendas, implicitly or explicitly, for the sake of votes is “the wrong way to reduce polarization” and would “repeat some of our country’s most shameful mistakes.”
They offer general guidelines for political parties to plan their futures. Parties should keep apparent authoritarians out of their grassroots supporters and definitely keep them off the ballot. They shouldn’t normalize authoritarian behavior and shouldn’t ally with parties that do. The most effective defense of democracy will be, the authors further recommend, a coalition forged between people who are otherwise ideological adversaries but who come together to defeat antidemocratic extremists.
How Democracies Die is a creative, in-depth, and — if not exactly optimistic — invigorating exploration of one of the major questions of our time.
In September 2019, New York Times opinion columnist Paul Krugman cited the book How Democracies Die, saying that current events show that the slow decline of democracy "can happen right here in America." He said: "Authoritarianism is on the march across much of the world, but its advance tends to be relatively quiet and gradual, so that it’s hard to point to a single moment and say, this is the day democracy ended. You just wake up one morning and realize that it’s gone."
David Runciman. How Democracy Ends. Basic Books, June 2018.In advance of the release of his book, How Democracy Ends, David Runciman wrote in The Guardian (1 May 2018) about "epistocracy: the rule of the knowers."
"It is directly opposed to democracy, because it argues that the right to participate in political decision-making depends on whether or not you know what you are doing. The basic premise of democracy has always been that it doesn’t matter how much you know: you get a say because you have to live with the consequences of what you do. In ancient Athens, this principle was reflected in the practice of choosing office-holders by lottery. Anyone could do it because everyone – well, everyone who wasn’t a woman, a foreigner, a pauper, a slave or a child – counted as a member of the state. With the exception of jury service in some countries, we don’t choose people at random for important roles any more. But we do uphold the underlying idea by letting citizens vote without checking their suitability for the task."
Democracy mainly is concerned that "the voters should be around long enough to suffer for their own mistakes." Epistocracy wants us to "prevent them in the first place – then it wouldn’t matter who has to take responsibility."
"Democracies do seem to be doing some fairly stupid things at present. Perhaps no one will be able to live with their mistakes. In the age of Trump, climate change and nuclear weapons, epistocracy has teeth again."
"Technocrats are the people who understand what’s best for the machine. But keeping the machine running might be the worst thing we could do. Technocrats won’t help with that question. ... technocracy is not really an alternative to democracy. Like populism, it is more of an add-on. What makes epistocracy different is that it prioritises the 'right' decision over the technically correct decision. It tries to work out where we should be going. A technocrat can only tell us how we should get there."
John Stuart Mill wanted certain professional types to get more votes than others. Lawyers, Runciman explained, "got their extra votes because what’s needed are people who have shown an aptitude for thinking about questions with no easy answers." However, the same author pointed out that Jason Brennan, author of Against Democracy, claims that Americans on the whole are "systematically mistaken" about the relevant facts for major political decisions, so he thinks that it would be additionally necessary to subject voters to an examination of their knowledge. "Brennan also has to face the fact that contemporary social science provides plenty of evidence that the educated are just as subject to groupthink as other people, sometimes even more so," the writer says, citing Bartels and Achen.
Runciman says that Brennan believes:
"Voting is bad for us. It doesn’t make people better informed. If anything, it makes them stupider, because it dignifies their prejudices and ignorance in the name of democracy. 'Political participation is not valuable for most people,' Brennan writes. 'On the contrary, it does most of us little good and instead tends to stultify and corrupt us. It turns us into civic enemies who have grounds to hate one another.' The trouble with democracy is that it gives us no reason to become better informed. It tells us we are fine as we are. And we’re not.
In the end, Brennan’s argument is more historical than philosophical. ...we took a chance on democracy, waiting to see how it would turn out. Why shouldn’t we take a chance on epistocracy, now we know how the other experiment went? Why do we assume that democracy is the only experiment we are ever allowed to run, even after it has run out of steam?"
It would have been a far easier sell to try epistocracy first, since, having had full suffrage, people will resist giving it up.
Runciman identifies dangers of epistocracy.
One is that "we set the bar too high in politics by insisting on looking for the best thing to do. Sometimes it is more important to avoid the worst. Even if democracy is often bad at coming up with the right answers, it is good at unpicking the wrong ones." He elaborates: "Epistocracy is flawed because of the second part of the word rather than the first — this is about power (kratos) as much as it is about knowledge (episteme). Fixing power to knowledge risks creating a monster that can’t be deflected from its course, even when it goes wrong...Ignorance and foolishness [as frequently displayed by the demos don’t oppress in the same way that knowledge and wisdom do, precisely because they are incompetent: the demos keeps changing its mind."
In other words:
"You have to ask yourself where you’d rather be when things go wrong. Maybe things will go wrong quicker and more often in a democracy, but that is a different issue. Rather than thinking of democracy as the least worst form of politics, we could think of it as the best when at its worst. It is the difference between Winston Churchill’s famous dictum and a similar one from Alexis de Tocqueville a hundred years earlier that is less well-known but more apposite. More fires get started in a democracy, de Tocqueville said, but more fires get put out, too."
He writes about Kimera Systems' 2017 AI called "Nigel" that suggests who you should vote for. "Nigel is not trying to work out what’s best for everyone, only what’s best for you. It accepts your version of reality. Yet Nigel understands that you are incapable of drawing the correct political inferences from your preferences." Is it a "parody" or an "enhancement" of democracy, he asks? On the side of enhancement:
"Democratic politicians don’t much care what it is that we actually want. They care what it is they can persuade us we want, so they can better appeal to it. Nigel puts the voter first. At the same time, by protecting us from our own confusion and inattention, Nigel strives to improve our self-understanding. Brennan’s version effectively gives up on Mill’s original idea that voting might be an educative experience. Shita hasn’t given up. Nigel is trying to nudge us along the path to self-knowledge. We might end up learning who we really are.
The fatal flaw with this approach, however, is that we risk learning only who it is we think we are, or who it is we would like to be. Worse, it is who we would like to be now, not who or what we might become in the future. Like focus groups, Nigel provides a snapshot of a set of attitudes at a moment in time. The danger of any system of machine learning is that it produces feedback loops. By restricting the dataset to our past behaviour, Nigel teaches us nothing about what other people think, or even about other ways of seeing the world. Nigel simply mines the archive of our attitudes for the most consistent expression of our identities. If we lean left, we will end up leaning further left. If we lean right, we will end up leaning further right. Social and political division would widen. Nigel is designed to close the circle in our minds."
The AI could force us to question our existing beliefs, but this leads to the second problem with epistocracy.
"...we won’t be the ones telling Nigel what to do. It will be the technicians who have built the system. They are the experts we rely on to rescue us from feedback loops. For this reason, it is hard to see how 21st-century epistocracy can avoid collapsing back into technocracy. When things go wrong, the knowers will be powerless to correct for them. Only the engineers who built the machines have that capacity, which means that it will be the engineers who have the power."
And furthermore: "The power of engineers never fully comes out into the open, because most people don’t understand what it is they do."
In his book, Runciman says that democracy is sure to end someday and yet its end may come by surprise. We are, he says, “barely two decades into the twenty-first century, and almost from nowhere the question is upon us: is this how democracy ends?” Trump’s election: “looked like the reductio ad absurdum of democratic politics: any process that produces such a ridiculous conclusion must have gone seriously wrong somewhere along the way. If Trump is the answer, we are no longer asking the right question.” Trump himself may not be the “end of democracy,” but then we must ask: “What would democratic failure in a country like the United States actually involve? What are the things that an established democracy could not survive?” If we don’t know what to look for, “we are likely to be surprised” or “[w]e may not even notice that it is happening”.
Although the violence of the World Wars was devastating, wartime “created the conditions through which rising inequality could be brought under control.” More specifically, widespread participation in war efforts was used to justify political equality. Today, the US is divided into “those places where military service is the norm and those where it is almost unheard of,” and peacetime is ideal for populism to take root because the lack of need for a military draft “disputes the idea that democracy is still a genuinely collective experience.”
In 5th century BCE Athens, it seemed: “If there was no alternative to democracy, there was no alternative to putting up with it.” But then there was a coup by the aristocrats. “The only way to join the group [of the Four Hundred] was to be rich enough to work for the government without pay, because salaries for public office were suspended….Democracy had given way to oligarchy. Power was in the hands of a privileged clique, backed up by violence.” The moderates restored power to a larger group, and “democracy in Athens survived for the best part of another century.”
“Twenty-first century America is nothing like Weimar Germany. Its democratic institutions are much more battle-hardened. Its society is much more prosperous. Its population has many better things to do than take up arms against democracy.” He criticizes Timothy Snyder’s 2017 On Tyranny in its injunction to “Figure things out for yourself,” since, Runciman says, conspiracy theorists these days “see themselves as the last vanguard against totalitarianism” and “If Trump never turns into Hitler - and he won’t - then everyone can claim to have been proved right,” but we may well get some other “true horror.” “The central division of our time is not conspiracy theory v. conspiracy theory in the name of democracy. This is not the 1930s all over again. It is the 1890s, without the prospect of resolution.” He says: “One possibility is that things carry on as they are. Democracy does not collapse into violence. It simply continues its drift into cranky obsolescence.”
“So what are the factors that make the current crisis in democracy unlike those it has faced in the past, when it was younger? I believe there are three fundamental differences. First, political violence is not what it was for earlier generations, either in scale or in character. Western democracies are fundamentally peaceful societies, which means that our most destructive impulses manifest themselves in other ways. There is still violence, of course. But it stalks the fringes of our politics and the recesses of our imaginations, without ever arriving centre stage. It is the ghost in this story. [On violence, see, however, my review of Corey Robin's The Reactionary Mind. - T.L.] Second, the threat of catastrophe has changed. Where the prospect of disaster once had a galvanizing effect, now it tends to be stultifying. We freeze in the face of our fears. Third, the information technology revolution has completely altered the terms on which democracy must operate. We have become dependent on forms of communication and information-sharing that we neither control nor fully understand. All of these features of our democracy are consistent with its getting older.
I have organized this book around these three themes: coup [a conspiracy]; catastrophe [some kind of total disaster]; technological takeover [the possibility that “we abdicate democratic responsibility” to AI, Big Data, and the like].”
“The minimal definition of democracy says simply that the losers of an election accept that they have lost.” Or: “democracy is civil war without the fighting. Failure comes when proxy battles turn into real ones.” Or: “the people with guns don’t use them.” And yet, seemingly contradictory, violence can catalyze the act of democracy: “If and when the fictions of environmental disaster turn to fact, democratic politics can be expected to snap out of it. That is one of democracy’s strengths: chaos and violence bring out the best in it.”
The United States didn’t become fully democratic until slavery ended, women got the vote, and the 1960s civil rights movement succeeded, which means it's “not elderly [for a democracy]. It is not youthful either, given how many democracies have been snuffed out before they have even got going. It is middle-aged.” He continues: “To be middle-aged is to have survived long enough to recognise the signs [of mortality]. Dramatic collapse is possible — it’s happened to others. At the same time, treating every ache and pain as a signal that the end is at hand is laughable. Hypochondria is a malady in itself. Life is still to be lived and the best may be yet to come. That’s where American democracy is now.”
Continuing the metaphor of "middle-age" and regarding the role of Trump, he says:
“When a miserable middle-aged man buys a motorbike on impulse, it can be dangerous. If he is really unlucky it all ends in a fireball. But it is nothing like as dangerous as when a seventeen-year-old buys a motorbike. More often, it is simply embarrassing. The mid-life motorbike gets ridden a few times and ends up parked in the street. Maybe it gets sold. The crisis will need to be resolved in some other way, if it can be resolved at all. American democracy is in miserable middle age. Donald Trump is its motorbike. It could still end in a fireball. More likely, the crisis will continue and it will need to be resolved in some other way, if it can be resolved at all.”
“In most functioning democracies, the people are bystanders much of the time anyway. They watch on as political decisions are taken on their behalf by elected representatives who then ask for their assent at election time. If that’s what democracy has become, it provides excellent cover for the attempt to undermine democracy, because the two look remarkably similar.
Contemporary political science has devised a range of terms to describe this state of affairs: ‘audience democracy,’ ‘spectator democracy,’ ‘plebiscitary democracy’. These terms might be too mild: ‘zombie democracy’ might be better. The basic idea is that the people are simply watching a performance in which their role is to give or withhold their applause at the appropriate moments. Democratic politics has become an elaborate show, needing ever more characterful performers to hold the public’s attention. The increasing reliance on referendums in many democracies fits this pattern. A referendum looks democratic but it is not. The spectators get dragged on stage to say a simple yes or no to a proposition they have played no part in devising.”
“Anyone who has lost faith in the possibility of political change is likely to believe voting is not worth the bother; anyone who stops voting is likely to find that the system ignores them because their views don’t count.
This is potentially a vicious circle. ... The danger comes when the permanent losers outnumber the occasional winners...We may be in the middle of just such a shift.”
And when we do not participate as performers, but rather as spectators:
“We do not walk the tightrope. It is done for us, by functionaries who are motivated by their anxious desire not to fall. The noise of the crowd is not an integral part of the performance. It is another hazard to be faced in the attempt to keep upright and moving forwards. ... Trump is no joyful high-wire artist. He is a sleepwalker and a gambler, unconcerned by watching others fall. To wish to put him on the wire is to believe one of two things. Either there is a safety net. Or the whole performance is a sham.”
He thinks AI is likely to play a bigger role in the future:
“I have considered three alternatives to modern democracy: pragmatic authoritarianism, epistocracy, and liberated technology. The first two have things to recommend them, but in the end they don’t stack up against the democracy we have, even in its parlous current state. They remain temptations rather than alternatives. The third is different. It includes all sorts of potential futures: some wondrous, some terrible, and most wholly unknowable. It is a spectrum of possibility, as wide as any human experience has ever known.”
"Carol Anderson: Seven Books on Democracy and Its Challenges." Jane Ciabattari. LitHub. Sept. 4, 2018.
"A Guide to Ranked-Choice Voting." David Leonhardt. New York Times. June 16, 2021.