Democracy for Realists assails the romantic folk-theory at the heart of contemporary thinking about democratic politics and government, and offers a provocative alternative view grounded in the actual human nature of democratic citizens.
Christopher Achen and Larry Bartels deploy a wealth of social-scientific evidence, including ingenious original analyses of topics ranging from abortion politics and budget deficits to the Great Depression and shark attacks, to show that the familiar ideal of thoughtful citizens steering the ship of state from the voting booth is fundamentally misguided. They demonstrate that voters even those who are well informed and politically engaged mostly choose parties and candidates on the basis of social identities and partisan loyalties, not political issues. They also show that voters adjust their policy views and even their perceptions of basic matters of fact to match those loyalties. When parties are roughly evenly matched, elections often turn on irrelevant or misleading considerations such as economic spurts or downturns beyond the incumbents' control; the outcomes are essentially random. Thus, voters do not control the course of public policy, even indirectly.
Achen and Bartels argue that democratic theory needs to be founded on identity groups and political parties, not on the preferences of individual voters. "Democracy for Realists" provides a powerful challenge to conventional thinking, pointing the way toward a fundamentally different understanding of the realities and potential of democratic government."
On the tube, book open, hardback, head down, three quarters in. Man interrupts: Sorry I couldn't help noticing what you are reading. It sounds timely! Is it any good? Me: It's ok, not as Machiavellian as it may sound. The main... Man interrupts: What's the argument? You must be into it I see you put yellow post-it notes throughout. Me: The main idea is that democracy as practiced has nothing to do with sovereignty of the p... Man interrupts: Would you recommend it? Me: Not the hardback version ... Man: Why? How much does it cost? Me: No, I meant ... it's not about the price. It's that it does not provide any solution to our urgent pro... Man interrupts: Are you Greek? Me: ... (Huge sigh) Sorry I've gotta get on. (Head back down into the book).
"Democracy for Realists" is a rich and sobering assessment of the state of democracy. The book is at once a literature review, an empirical contribution, and an agenda proposal for the future study of democracy.
It's structured around critiques of two leading theories of democracy: the "folk" or "populist" theory, and the theory of retrospective accountability. Populists hold that democracy involves political equality and popular control of policy, in which elections translate voters' preferences into policy outcomes. Retrospective accountability is a less ambitious theory of democracy, which holds that periodic elections allow parties to experiment with different policies and voters to assess these experiments. Achen and Bartels summarize an extensive theoretical and empirical literature criticizing both theories.
In brief, the populist or folk theory of democracy fails because voters lack well-formed preferences and cannot translate these preferences into meaningful choices. Beginning with Philip Converse's studies of voter knowledge in the 1960s, political scientists have shown lack the political knowledge to connect their political preferences to electoral and policy choices. Voters also have contradictory policy preferences such as for lower taxes and more extensive public services. Retrospective accountability fails because voters are not very good at attributing responsibility for policy outcomes or judging the real effects of policy. In a particularly humorous example, Achen and Bartels show how a spate of shark attacks on the Jersey Shore before the presidential election in 1916 led voters to turn against Woodrow Wilson. Voters punish incumbents randomly and ineffectively, contrary to the requirements of retrospective accountability.
Achen and Bartels propose an alternative theory of elections in which group identity rather than policy preferences or retrospective accountability is the primary cause of voter behavior. Voters' sense of social distance to candidates and group affiliation with candidates determines their choices.
"Democracy for Realists" may seem like a depressing book. However, I found it to be quite a hopeful read. If we can put to rest the bad arguments for democracy, we can identify more clearly and confidently why democracy has value. That's a reason to be hopeful.
THis was a tough read since I'm not a political scientist and I don't understand a lot of the jargon. The overall point was interesting and thought provoking; I think that the intro/conclusion would be worth reading carefully for the non-academics out there and then maybe skim the rest? I think that the authors outline the problem - too much democracy is not the solution to the problems that ail our democracy - but, as they themselves say, they don't really have solutions. But they do an excellent job debunking the theories that they, and other political scientists over the course of the last century have proposed for how democracy works.
Identity ('group-based') politics is older than you think, was once recognized as a dominant paradigm of voter behavior in the early 20th c. - and tons of empirical evidence from the last century backs that assertion up and shows the concrete impact of identity ('group-identification') on political behavior and the formation of ideology - and it's here to stay.
The authors also examine, deconstruct, and refute the 'folk theory of democracy' (and the famous spatial, i.e. left-right view, of voting), analyze the impact of random events on voter preferences, and demonstrate that other theories of democracy and voting taken from economics (e.g. ones assuming rational choice theory) fail in practice, namely retrospective voting, where the electorate is so myopic as to only count real income growth in the 2 quarters before an election (out of 16) in deciding whether to keep or replace the incumbent. Nevertheless, too great a neoliberal emphasis is placed on economics as driving voting and economic growth as the be-all, end-all for a hypothetical 'rational voter'.
They have a rather new analysis of the New Deal realignment which argues that the victories of FDR weren't actually approval of his policies, but were accounted for by real income growth in the months leading up to his reelections, delaying 'blind retrospection' - when the depression hit, countries with right wing governments threw them out in favor of leftists; states with left wing governments threw them out in favor of rightists.
The authors demonstrate that group identity and partisanship precede ideology and form ideology, instead of the other way around. They demonstrate thoroughgoing cognitive distortions and polarization according to group identification (identity) and ideology.
The authors examine the political salience of identity using examples like the Catholicism of JFK - identity issues which are no longer live in politics - while avoiding modern racial identity for the most part and giving very short shrift to nonpartisan ideology. (The authors' account argues that party identification gives rise to ideology instead of vice versa, which can't account for and leaves aside the entire group, albeit a minority, of people who have an ideology but are outside of the party system, like libertarians, integralists, Communists, ethnic nationalists, etc.)
The authors do not venture an explanation of why and how identity politics came about. Evolutionary theory does. Identity politics is the instantiation of ingroup altruism in representative democracy. Thus, for an evolutionary-theoretic account of how identity politics evolved, see Salter, 'On Genetic Interests'; as to examples of how it's practiced and why it's effective, see MacDonald, 'Separation and its Discontents'.
Might write a more detailed review. I have mixed feelings about this book and the authors skittered away from the implications of their data in a hurry. Many important parts are glossed over and minor parts given longwinded and repetitive exposition, but for raw data this book gets a partial pass on not following the argument and a higher rating for the data than for the analysis of it.
How do voters affect the political process? The folk theory says voters are knowledgable about issues & about candidates positions on the issues & they select the candidate who mostly closely reflects their preferences. FALSE! Voters know very little about the issues, are not willing or able to invest the time & study in becoming knowledgeable & besides optimizing a choice of candidate is impossible (Arrow). Voters retrospectively reward good performance & punish malfeasance. FALSE! Voters reward outcomes which are beyond the scope of politicians to affect; do not use appropriate evaluations on phenomena which are within the purview of politicians. Finally, voters align into parties according to group identity. Good analysis. A bit of a tedious read.
Great when it comes to documenting how the political behavior of most democratic citizens is a lot less informed and less rational than idealistic theories tend to assume. In particular, it shows how group loyalties are critical to people's political behavior, which is underappreciated and important. Unfortunately, while the authors show that group loyalties are extremely significant, they seem to have little understanding as to why that is the case, which renders the picture they draw painfully incomplete. In particular, the picture presented by the authors seems to have absolutely no grounding in evolutionary theory, which is, to me, a rather staggering omission. The work of evolutionary theorists such as Robert Kurzban and Peter DeScioli has a lot to say about the "why" question. So does The Elephant in the Brain. (In fact, Democracy for Realists and The Elephant in the Brain complement each other quite well, as they each cover each others' main shortcoming when it comes to politics: an evolutionary framework and a lot of empirical data on political behavior, respectively.)
In addition to that, I think the book does a poor job when discussing the implications of the presented findings, and when it comes to critiquing existing "theories of democracy" (ch. 11). To take just one example, they write (paraphrased slightly) that all the conventional defenses of democratic government are at odds with the facts they document in their book. But this is hardly the case. For example, consequentialist defenses of democratic government must be considered among the conventional defenses of democratic government, yet they do not rest on the claim that democratic citizens are any different from what the authors here describe. Such defenses merely require that democratic governments are better than alternative forms of (non-)government. The authors seem to make many such overly strong statements in the final chapter.
Read this for school so I’m not rating but there are some really interesting ideas in this book. I definitely didn’t agree with everything and I appreciated being able to discuss this with my political science professor because it is dense but I still think there’s value to think through the arguments in this book.
According to the “folk theory” of democracy—the naïve, rose-tinted understanding put forward by schoolteachers, politicians, and pundits—democracy is a mechanism for self-government. Individuals determine where they stand on the issues of the day, learn about the positions of candidates and parties, and vote for the option that most closely resembles their views. As voters are the ultimate source of legitimacy, and they bestow that legitimacy on those with whom their policy views align, politicians ignore the “will of the people” at their own peril. Since the democratic will represents the collective wisdom of the masses, it is inevitably greater than the understanding of any elite faction or cabal that would subvert it. Thus, democracy, while not perfect, is better than any other system of government, and when problems arise within democratic systems, this does not mean that democracy itself is inherently flawed, but merely that the current system is not democratic enough. There is nothing wrong with democracy that cannot be fixed with more democracy.
Despite the dogged persistence of this optimistic view in the popular consciousness, it has been thoroughly debunked by more than a century of social science. Democratic citizens are simply unwilling and unable to fulfill the role allotted to them by democratic theory. They do not follow politics closely, they rarely develop any kind of coherent political ideology, and they lack the time, means, and inclination to make the type of informed choices among a potentially infinite range of policy options that would make the “popular will” meaningful or even discernible. One study after another has found that voters routinely struggle to articulate their own political beliefs, let alone those espoused by the parties, and they often vote for candidates whose policy views are markedly different from their own. When voters do have a well-developed set of political views, this almost always constitutes an elaborate rationalization of their partisan identity, not a real basis for it. Voters mirror the policy views of their chosen party as they learn of them; they do not drive policy in any meaningful way. Party elites and interest groups hold the reins.
One of the most popular academic models for the folk theory is the spatial model, according to which the policy views of voters can be represented as ideal points on a single left-right spectrum, and parties try to win electoral majorities by staking out positions in the center of the spectrum, where a preponderance of voters are represented. This would mean, in theory, that the major political parties do in fact represent the popular consensus, and it could explain why, in a two-party system like that of the United States, the parties often espouse very similar issue-preferences.
Unfortunately, the very premises of the spatial model do not account for the real behavior of voters. The model portrays “issue proximity” as the primary determinant of voter behavior, when in fact tribal loyalties—partisan, ethnic, racial, occupational, religious—are far more operative. It is also highly problematic to attempt to reduce the vast complexity of the realm of political choice to a single spectrum, or even a number of single-issue spectrums. There are simply too many possible ways to view the political world to meaningfully map out the contours of the political beliefs of voters. There are never just two options, and that makes the notion of gleaning any sort of policy consensus from the binary choices of the electorate almost paradoxical.
An alternative, if somewhat more attenuated, defense of the populist model of democracy is the retrospective theory of voting: the idea that even if citizens do not proactively create policy out of whole cloth, they may still regulate the behavior of their elected representatives by rendering a retrospective verdict on their tenure of office. Voters may not have sophisticated policy views, but they can at least be trusted to determine whether their own lives and the wellbeing of their communities have improved or deteriorated during a politician’s term, and to reward or punish him accordingly.
But as with the spatial model and with the folk theory more generally, the retrospective theory runs aground at the epistemological level. It is not at all clear that voters can reliably discern changes in their own quality of life or in the wellbeing of the surrounding community; and even if they could, they would have to surmount the additional hurdle of coherently connecting these changes to the performance of political incumbents in order to rationally ascertain their culpability. Voters do tend to “punish” elected officials during difficult times, but the bases for these punishments are often myopic, arbitrary, and even downright nonsensical. Woodrow Wilson lost as much as ten percent of his support in the coastal counties of his native New Jersey after a spate of deadly shark attacks in the summer of 1916. Incumbent parties suffer at the polls during periods of drought.
Even on as momentous an occasion as the presidential election of 1936, when Franklin Roosevelt won a landslide reelection victory in what most have interpreted as a popular ratification of the New Deal, the retrospective judgement of voters was far more limited in scope than is usually assumed. The election was not so much an endorsement of the New Deal as it was a reaction to the positive economic conditions that prevailed in the few months immediately preceding the election. Achen and Bartels conjecture that if the recession of 1938 had occurred two years earlier, Roosevelt would most likely not have won reelection, and the entire New Deal realignment of the American party system would have been stopped in its tracks.
The overwhelming majority of voters are not ideological; they are identitarian. They place themselves within an in-group, vote for the politicians who seem to represent “people like them”, and adjust their stated policy positions to match those of their political tribe rather than the other way around. They view political parties not as slates of policy ideas, but as collections of social groups. Realignments are not driven by changes in policy positions on the part of voters or parties, but rather by reevaluations of identitarian belonging.
In one of the book’s most fascinating sections, Achen and Bartels refer to a collection of studies on the dramatic realignment of white southerners from the Democratic Party to the GOP after the 1960s. While the common understanding is that the defection of southern whites from the Democratic Party was motivated by opposition to the Civil Rights Act and subsequent efforts to improve the lot of African-Americans in the south, such as affirmative action, these studies demonstrate that the issues of civil rights and affirmative action were not nearly as large a contributing factor to the southern realignment as changes in the perceived place of belonging for white southern identity. White southerners who supported and opposed government aid for African-Americans and racial integration both defected from the Democratic Party in the latter half of the twentieth century. The partisan gap between southern whites who supported race-based affirmative action and those who opposed it did not become significant until well into the 1980s, more than two decades after the southern realignment had begun to take shape. This suggests that southern whites first moved to the GOP due to changes in tribal affiliation, and only afterwards began to adopt the policy views of their new party.
Achen and Bartels don’t put forward any new systematic theory of democracy, but they have successfully demonstrated the necessity of a radical rethinking of its role and function.
This book was really a 2.5. The authors show various areas where direct democracy breaks down. If you aren't familiar with Arrow's Impossibility Theorem or survey research on partisan bias this book could be interesting for you. If on the other hand you are familiar with those sorts of results then this book doesn't really offer anything new.
Many of the statistics presented in the book are represented has have much more statistical power than they actually do so take the statistical results with a hefty grain of salt. Overall, the reason I rounded down to two stars rather than up to three is because the authors fail to give an explanatory account of identity politics that is causally distinct from voters simply being uninformed. Their inability to parse these concepts dramatically weakens their argument even if you think it could have merit.
For a more rigorous account of political dynamics, I'd recommend "The Logic of Political Survival" by Bueno de Mesquita et al.
A lot of interesting ideas and information - presented in a stilted boring tedious academic style of writing. Too bad. I got about a hundred pages in before I finally gave up. Not too technical or hard to understand. Just painful to read. Sorry to say that, I’ve never written a book myself, and if I did I’m sure it would be far worse...
I opened this book as a skeptic of democracy, and it didn't do much to change my views, but it did give me some new reasons to be skeptical.
The authors marshal a lot of support for their first thesis -- that swing voters mostly care about what happened to their wallets in the past year when they vote, and that not much else makes a difference. This didn't surprise me, but I hadn't thought much about the perverse incentives it offered to non-dictatorial governments (e.g. Richard Nixon manipulating the economy to create a false economic boom just before reelection, with the downsides only arriving once the votes were in). The second thesis is splashier (we care about identities we create for ourselves far more than we care about policy), but comes with fairly light empirical evidence (though I can't think of many reasons to doubt it).
The book's occasional missteps (happily quoting some sketchy priming studies, leaving out economic theory almost entirely) stop it from getting a fifth star. More complaints: the prose can be repetitive, and the graphs are muted and sometimes under-labeled. There will be a market somewhere for a popular-science book that quotes the same ideas.
On the other hand, that popular-science book may leave out some of the best parts of the book: The authors' description of the benefits of democracy (despite everything), their political-science-insider discussion of how folk democratic theory developed almost independently of any evidence, and a solid collection of international examples that help them dodge American myopia. They also avoid talking down to "the people", freely admitting that education is no panacea, and that irrationality dominates the voting patterns of even the most knowledgeable partisans. (If they are partisans themselves, they never mention any personal ideological affiliations.)
I think most pro-democracy, let-the-people-rule types will get more out of this than I did. If any of you are reading this review, pro-democracy people: Give the book a try! You'll find the authors friendlier and less elitist than you suppose, and you can skim without missing too much.
There's a pretty widespread idea that "democracy isn't working." But what does that mean? The authors, two leaders in political science, go through a series of naive theories of democracy and debunk each.
Ultimately, they settle on (an admittedly unsatisfying answer to them) story about democracy as a feedback loop from parties, interest groups, and elites on the one hand, and the voters on the other. This is an *account* of how democracy acts.
As someone working in the space, this is really helpful.
They also close with a question about what it would be to be more democratic grounded in the theory that they articulate and what it would look like to know. If parties, interest groups, and elites are so central to the functioning of democracy, why is there so little research on how those operate?
They lay out a research plan based on that. They do note, as have others, that there is an inequity in who is in the interest groups and parties and have some suggestions about opening those up. But those are VERY preliminary and quite different from the ideas that most reformers engage in.
Numerous books have been devoted to arguing for one side of the political spectrum to sway the reader's political alignment. However, very few books take a step back and analyse the political process without any party loyalties. And within those who do, none go to depths and statistical precision that is undertaken by ‘Democracy for Realists’.
The book is divided into 3 sections. It begins with dismantling the traditional ‘folk theorem of democracy’, that voters are well-informed rational citizens and elections produce the aggregate preferences of the populace. Next, they attack models of retrospective voting and the belief that voters are rationally able to measure the positive impact of a politician and vote accordingly. Finally, they present their own idea of group identities and how social groups trump any rhetoric, debate or evidence when it comes to partisan loyalties.
The book covers each section with enormous depth and various statistical tests. In a world full of books with empty rhetoric it is a fresh take on the political process. As someone who has followed politics in India for a long time, a lot of the results help explain phenomenons that are frequently observed. Indeed, the book does what every great book aims to do. Creates a worldview that changes your perspective forever.
The few criticisms are that the authors fail to provide a complete theory or set of policy reforms for their own version of the democratic process. However, the authors are aware that what they have taken is the first step in developing a new understanding of democracy. Moreover, the authors seem to be narrowly focused on the USA. Studies could have been easily repeated over multiple countries to produce similar results.
At last, I believe this is one of the books which will stand the test of time and become the bedrock of a new form of thinking. While the book may have been praised for explaining the 2016 US elections, the core of their argument applies to all election cycles, 100 years ago and 100 years in the future.
This book is not recommended for the general or a casual political reader, as it comprises a highly detailed, scholarly statistical analysis of other in depth political works from the mid-1800's to the present, attempting to explain how and why American voters vote in the manner in which they do. The book mostly tries to answer the fundamental question: Do elections result in serving voters needs, or are they simply an exercise of a random process where the input of the votes originate from a largely uneducated, ignorant, and myopic voting public? The compiled data within the book is huddled around two competing political theories of processes to which an average voter employs in selecting a candidate. Both theories apply to incumbents and challengers. The first theory the author's present is termed ‘Populist Democracy" or "folk democracy." Within the theory lies the assumption that voters implicitly employ rational analysis when making voting decisions. These are based on awareness of associated policy connotations of various candidates, as well as political and psychological affiliations with party elites. The second theory of electoral mechanics identified is "Retrospective Accountability." Simply put, it is the idea that voters scrutinize past performances of politicians in office, as a method for forecasting future performance. In general terms, voters who experience pain, be it direct or perceived, economic or otherwise, tend to "throw the bums out" by voting for challengers of another party. The fallacy here is that the voting decisions based on this theory are myopic to the degree that politics and policy are not congruent. When one considers that global citizens are massively ignorant about particular policy machinations, and thus vote incorrectly, resulting in holding incumbents responsible for actions and circumstances which they are not liable. The authors used a laughable example highlighting the 1919 re-election of Woodrow Wilson. Apparently, there had been a series of shark attacks along the Jersey shore which resulted in two fatalities, the beaches that summer were closed, and the beach merchants hostility to the lack any "government action" actually dealing with the attacks, translated into votes cast for Wilson's challenger Theodore Roosevelt. Roosevelt subsequently carried the Jersey shore districts by a vast majority. Another significant influence upon which the authors identify as having considerable influence in producing inept election results is inherent, ingrained ideologies and pre-determined group controls. Numerous studies have demonstrated that most citizens of democratic countries have little interest in politics and do not follow the news of public affairs beyond browsing headlines. As stated before, they do not know details of even salient policy debates, nor a firm understanding of even what political parties stand for. As a result, they often vote for parties whose long-standing issue positions are at odds with these voters. Mostly these voters identify with ethnic, racial, occupational, religious, and other sorts of groups, and often- whether thru group ties or hereditary loyalties- with a political party. Even the more attentive citizens mostly adopt the policy positions of the parties as their own: they are mirrors of the parties, not masters. For most citizens most of the time, party and group loyalties are the primary drivers of voting choice. Hence we can see the disconnect when sovereignty rests with the people. A government derives its just powers, not merely from the "consent of the governed" but from their political judgments. Within that framework of understanding, we can see how overwhelmingly exists the likelihood for voters to vote incorrectly. Thus producing, as in our case, POTUS 45, or in the most extreme, the most disastrous democratic election result of the 20th Century; the election of Adolf Hitler. In the end, Democracy for Realists offers no solutions to the dilemma, but does clearly document the conditions upon which reside our political future.
To be honest I skipped a bunch of the middle chapters. I was just looking for the answers! Alas, this book doesn't really have many, besides "make society more equal" and "reduce the effect of money on politics".
The book was still immensely helpful, because it shook me of the naive view that we should be giving more power to individual citizens to have a direct impact on the political process.
My initial fears about the book, that two white dudes in Western academia would be constrained in their vision of what governing could be like, were not unfounded. the last chapters especially, that were focused on ways forward, lacked any knowledge from outside the western world. If the US and Europe have such silly views of how democracy works, why not look a little farther for some good ideas?
Also, the book was about how responsive a democracy is to it's citizens, but I didn't get a great explanation for what that metric looks like and of the merits of other metrics.
it's also hard for me to get an intuition on what ruling by group power feels like. in the ideal, does that mean that as an individual I shouldn't have say in the government? instead I meet with my group and they decide what to support? like what if I gave my "vote" proportionally to the different groups I wanted to represent me and then those groups got to vote? why weren't ideas like this discussed in the book?
also they dismiss deliberative democracy because it's too small scale, but what about ways for scaling up deliberative democracy with technology?
As the internet will tell you, 2016 was a disappointing year for democracy. Brexit, Trump, democratic decline in Poland, Hungary, Turkey, Israel, Venezuela ... Surely something has gone terribly wrong with the way we do democracy? According to the authors of this book, no, it's more like electoral democracy never really worked very well.
If you don't have the time to read the whole book, just read the final chapter (31 pages). In it, the authors summarize decades of research showing that the way most people think electoral democracy works is a hopelessly unrealistic ideal. Contrary to popular belief, requiring that governments be elected by their subjects does not, even in theory, ensure that they will rule according to "the will of the people" in any meaningful sense. Increasing the role of the voter through e.g. referendums doesn't fix this problem - it just leads to the most vocal and well-organized interest groups taking control. And in any case, voters are unqualified and uninterested in taking on the responsibilities of governance.
The authors have no real solutions to the problems they identify, but it does seem like accepting the flaws in both the current theories and the practice of electoral democracy is an important first step towards improving them in the future.
I can sum this up with "people can't be trusted to take the time to educate themselves and override natural biases and illogical thinking to be a part of an effective democracy".
This book became a bit of a chore to read. There were great points made to be sure, but the amount of research cited and examples given made the reading tedious.
I ended up reading the first few paragraphs and the "conclusion" sections of each chapter and didn't feel like I missed much.
It's unfortunate that this book is more along the lines of a research paper, as the issues and topics discussed should be more widely read. The very last few paragraphs offered some especially interesting ideas. The style will deter many from reading and absorbing, however.
I'd recommend if you are seriously interested in the topic of democratic theory and have the endurance to read through the entire book.
This is a very good book to read [it is a scholarly tone and style, though], but with a bunch of annoying (for me) statements that if toned down a bit would make the book even better. But they are trying to put extra attention on their subject, so it at least makes sense to play up their theories' strengths.
First, the good: this book will make abundantly clear that the idea of voters carefully and rationally looking at issues and choosing parties and candidates accordingly is completely wrong. This isn't entirely surprising, but the authors do a good job of showing how much of democratic thought is based on this unrealistic ideal. [They also claim Arrow's theorem proves there is no good way to accumulate voters' interest. This is false. Cardinal voting systems (give a candidate a rating out of 10 for example) are not constrained by Arrow's theorem. The Gibbard–Satterthwaite theorem does apply which basically means if people vote strategically you could get a second best candidate winning. That the authors don't recognize Arrow's theorem's limitations here is a bit disappointing.]
They then take on the retrospective voting. They show that retrospective voting is only based on short term economic conditions before an election. This can lead to bad outcomes since it essentially makes elections more random. I think their argument is well-received here, but they often try to make it seem like this is a very bad thing. It's not actually clear it is. If determining how good a candidate is doing is so difficult, then democracy being random may not be so bad since no other system could do much better and there are benefits to democracy like legitimacy of outcome from the people. Still, it isn't great to learn that retrospective voting will lead to bad outcomes sometimes. There is also the case of shark attacks hurting Wilson, but the Spanish Influenza not hurting Wilson. This seems like it is incompatible with their theory, unless voters randomly decide which events should be seen as something the government should be punished for.
Also, they try to say that if voters always punish politicians for droughts, then this is a perverse incentive. They say that voters should reward better than average response half-the-time. I think it might be that voters basically always punish but may reward good responses by punishing the incumbent less. This would also destroy the perverse incentive, since a good response still helps the politician. I doubt they completely missed this, but they don't seem to address this argument head on well in the text.
In general, they have lots of analyses of data which usually does a good job of reinforcing their point. There are some caveats in that they don't always have a lot of data [like for presidential elections], but I think they still get suggestive results. Note that their model on Southern Identity however (p. 256) has an adjusted R² of 0.09 [note that an adjusted R² penalizes adding too many explanatory variables to your model and that you want it close to 1 for an accurate model. Low R² means that the data is not satisfying your model well, with <0 being a very poor model] which means the model doesn't explain very well, which I was disappointed they didn't comment on, giving the impression that this model is doing well.
The last part is on how group identities are far more important for determining people's political choices. This seems true to me, as far as it goes, but it gives a lot of freedom to positing a group membership to explain any choice away. This is a minor concern, but given the many groups people belong, it could be quite significant. The question then becomes, how do people decide which group is most salient for any issue, and this seems like it must ultimately come down to issues/policies of the parties. Like I said, I think exploring group membership is important, but I think it poses a lot of problems for interpretation.
I also wish they had information for other systems [rather than democracy] to see how they stack up against democracy. This is beyond the bounds of the book they set out to write, though. So I can't blame them for not taking this enormous task on.
Finally, the authors mention the rise of Hitler as a cautionary tale of democracy, but I strongly dislike how they presented the story. Hitler came to power essentially in a coup. See http://www.huppi.com/kangaroo/L-hitle... for a better rendering of the tale.
Still, with all of these caveats, I think the authors do a great service by telling us to come up with real reasons to support democracy. We can do better than rely on untrue "folk democratic" reasons. They come up with some reasons in the final chapter, as they are supporters of democracy as well. I still recommend this book, as it brings the justification for democracy to the fore. We should always try to improve, and be on the look out for ways to improve human flourishing. This includes improving democracy.
Generally interesting, although way technical for any reader who is not a social scientist or familiar with how they talk, or hasn't ever taken a statistics class... I kept going by skimming over stuff I didn't understand to get to their statements of the meaning of a given statistical analysis.
What they are up to is to demonstrate that what they call the "folk theory of democracy" not only is not valid on the ground but actively hinders the representative nature of our government. After demolishing it in every possible way and at great length, they get to the part I persevered for. If we can all agree with Winston Churchill that 'democracy is the worst form of government except all those others that have been tried from time to time' - and we can see that voters are ignorant and shortsighted and irrational and really can't be expected to have the time to fully understand public policy anyway - where to??
Through their examples they arrive at a conclusion that the thing that most predicts how voters will vote is group affiliation, which works in different ways at different times but can almost always shed light on how voters make decisions. So they say that yes, democracy is the best possible government form, but that it needs reform, and it needs to take into account how people work in and with groups.
They say there is a lot more work to be done, because no one has spent time studying the affect of group affiliation on politics in 100 years, but their main suggestions to achieve a better democracy are that (1) a society with less inequality would be more thoroughly democratic, and (2) the power of money in American politics is unfairly privileging some groups over others, skewing the democracy and institutionalizing the inequality. I do completely agree with that, and it seems to me that it could have been asserted without reading the book, but without the complete demolition of the idea that voters carefully consider policy issues and vote accordingly, it might not seem so urgent to a lot of people.
I have to say I raised my eyebrows at their repeated assertions that people generally take the political group affiliations of their parents. They are older than I am so they were alive in the 1970s - so I wanted to say, Really??
I didn't read the appendix because on the first page was an equation using the terms omega, theta, and delta.... So wander into a library that has this, turn to page 297, and spend an hour reading the summary of their arguments and their conclusions, if you're interested in why US democracy is so debased.
this is an extremely well researched with tons of very good data, and despite its academic tone, easy to read analysis of the fundamental issues with modern democracies in an information-heavy age. however, i am not sure if one of their central premises - that party loyalty drives views, and that therefore people get entrenched in their party choices - still holds after the recent european elections in 2017 (eg france, uk). perhaps we can be more hopeful than the authors suggest, at least in europe?
Bartels and Achen challenge what they call the “folk theory of democracy.” The “folk theory” seems to have its roots in the idea of the “rational man” — an Enlightenment idea, certainly, but one that seems to have made its way into popular politics. The idea is that democracy works (when it does) via choices of representatives or directly of policies as informed by their interests and values. Representatives and policies then reflect those choices — the government embodies and enacts the will of the people.
Then they undertake a quantitative study of the validity of that folk theory. They actually examine two theories of rational voter behavior.
The first is policy voting — that, in simple terms, voters vote for candidates with whom they share policy positions more than alternative candidates.
Policy voting fails in part because voters are unable or don’t take the time to discern the policy positions of candidates. And in fact, there is little actual policy alignment between voters and the candidates they vote for to support the hypothesis that such a thing is behind voters’ behavior. This point echoes what Bartels showed in his earlier book, Unequal Democracy — that the policy positions of members of the House of Representatives do not correlate well with those of their constituents (particularly their lower income constituents).
In fact, foreshadowing some later discussions on group identity and group influence, Achen and Bartels hint that where there is agreement between voters and candidates on policies, the arrow of fit and influence may go in the opposite direction. Voters may not choose candidates who reflect their policy positions so much as adopt policy positions held by the candidates they choose.
The second theory of rational voter behavior is retrospective rationality. Voters assess the performance of office holders and vote them in or out depending on performance, viewed in terms of the voters’ individual or collective welfare.
Retrospective rationality fails in part because voters do not separate factors influencing their welfare that are due to the office holders’ action from those that are not. Famously, Woodrow Wilson lost re-election votes in New Jersey’s shore area in 1916 due to shark attacks on swimmers. Wilson had nothing to do with the attacks of course, but statistical analysis shows that in fact he did suffer at the polls. Voters in the area felt things weren’t going well, and they blamed the office holder. Achen and Bartels of course cite other cases, including a century long correlation of drought or severe rainfall with voters’ behavior that punished incumbents.
The second half of the book tries to pick up the pieces.
Suppose the critique of “folk democracy” is correct. Voting behavior is not rational, in either sense of policy voting or retrospective assessment. What then? For some, the obvious response is to re-assert what has been disproven, but this time as a “should” rather than an “is”. What we need is a more educated, more “rational” voter. Some readers may even at this point simply congratulate themselves as bucking the evidence, because they view themselves as exceptions, well-informed, rational voters. In fact, though, as Achen and Bartels show, the politically more informed voters are more, not less, likely to fail tests of rational voting behavior. Raising the information level of voters won’t correct the problem.
By contrast, Achen and Bartels pursue a “realist” theory of democracy. “In our view, a realist theory of democracy must be founded on a realistic theory of political psychology. At present, nothing of that kind exists.” (p. 230). They don’t pretend to have such a theory of political psychology themselves. But they believe they can begin. No such theory, they believe, can ignore the role of group identity. Research shows too clearly that policy positions, the starting point in the folk theory, are not the starting point at all, but are rather themselves heavily influenced if not produced by group identity. We adopt the policy positions we adopt, in large part because of the social group(s) with which we identify.
Of course this is ideological anathema to individualists (themselves a group, of course, no matter how some would like to deny it). But Achen and Bartels stand on realist grounds, and they subject their hypothesis to case studies. These case studies are the partisan political realignments of the New Deal in the 1930s, Kennedy’s Catholicism as a point of contention in 1960, the collapse of the solid Democratic south following the Jim Crow era, and the emergence of abortion as a powerful issue in the 1980s and 1990s. In each instance, they find compellng evidence, in the data, of powerful group influence.
So what direction would all of this lead us in, if we maintain a democratic ideology? As the authors argue, we must pay much more attention to the roles of groups in generating political positions and policies. “Groups” will include everything from political parties to unions to PACs to lobbyists to more informal citizen, professional, and business groups.
How do these groups influence the thinking of the voting public, and how do they influence the policies of the political parties in power? In particular Achen and Bartels recommend applying scrutiny to the role of money (and other forms of inordinate power) in politics — some groups are advantaged in their ability to influence, obviously, by their ability to speak more often, more loudly, and with more skill than others. As a consequence they are in a position to advance their group interests more effectively. Doing anything about them is more easily said than done, of course, for the very reasons Achen and Bartels have cited.
Bartels and Achen believe in democracy, and they are trying to determine how to help make it work. They believe in democracy in the sense that they believe a government that responds to and represents the interests of its people can be a positive force in their lives. Their obvious chagrin is with the fact, as shown in studies of voter behavior and the responsiveness of elected government, we neither have a responsive government nor do we behave at the polls in a way that will give us one.
I think the critique is a needed one. There is a need for throwing a wrench into our popular political discourse. We toss around tired ideological claims and perceptions as self-proclaimed liberals, libertarians, conservatives or whatever like blunt and tired tools, even deluding ourselves into thinking our “side” of the debate to be enlightened. It is too easy to claim that the public, as a whole, is too uninformed or doesn’t have sufficient time to educate itself for its role in a democratic system. Achen and Bartels don’t deny that that is the case, but their point directs us away from such a fatiguing defeatism.
Their arguments regarding the role of groups are not as tight, I think, as their critique of rational voting behavior. The notion of a “group” is itself pretty slippery and pretty complicated. I belong to many groups, with many associated identities — everything from explicit political affiliations to geographic identities, professional associations, cultural identities, and on and on. “My group identity”, if we can speak of it in any unified way, might best be seen as some sort of complex vector space rather than a simple assignment to some uber-identity.
With that in mind, I think Achen and Bartels do their best job here as challenging us to think differently about political behavior. The simple model of individual political rationality doesn’t work. And its failure isn't a matter of our failing to behave properly as voters — it fails because it is false to human political behavior per se. We don’t fall into the influence of groups because we fail to behave adequately as rational individuals. The influence of groups is simply an aspect of human behavior (I would even say human rational behavior, although that would get us into a much bigger argument about what constitutes “rationality” and whether it can have inherently social aspects or must properly be conceived on the scale of the individual).
Nothing could be more critical in reading this book than including yourself as a subject. It would be too easy to say that Achen and Bartels are talking about other people, not me. That’s not true. In reading their book, I see that tendency — “Most people fail to live up to rational standards, but I know I do.” But I am just like everyone else — I certainly have group identities, they certainly influence my perceptions, my judgements, and my political behavior. Thinking otherwise would be unbearably smug. As the authors say, “It is a book about the conceptual limitations of human beings — including the authors of this book and its readers.”
Slightly confused book built on statistical overreach that has been at least partly debunked (see https://www.journals.uchicago.edu/doi...). There were a few really interesting snippets, though, like :
- Expressed political attitudes can be remarkably sensitive to seemingly innocuous variations in question wording or context. For example, 63% to 65% of Americans in the mid-1980s said that the federal government was spending too little on “assistance to the poor”; but only 20% to 25% said that it was spending too little on “welfare” (Rasinski 1989, 391). “Welfare” clearly had deeply negative connotations for many Americans, probably because it stimulated rather different mental images than “assistance to the poor” (Gilens 1999). (...) The distinction between alternative frames is even more tenuous. For example, in three separate experiments conducted in the mid-1970s, almost half of Americans said they would “not allow” a communist to give a speech, while only about one fourth said they would “forbid” him or her from doing so (Schuman and Presser 1981, 277). In the weeks leading up to the 1991 Gulf War, almost two- thirds of Americans were willing to “use military force,” but fewer than half were willing to “engage in combat,” and fewer than 30% were willing to “go to war” (Mueller 1994, 30).
- In the 1950s and 1960s, for example, many American towns and cities decided whether to add fluoride compounds to their drinking water. The scientific evidence that fluoride reduced tooth decay was compelling, and cities whose administrators or city councils made the decision without a referendum overwhelmingly adopted fluoridation. However, when the measure went to the voters, 60% of the time the electorate voted it down. Nor were the losses confined to less educated parts of the country. Cambridge, Massachusetts, home to Harvard and MIT, voted three times, defeating it, then passing it by the narrowest of margins, and then defeating it decisively on the third try (Crain, Katz, and Rosenthal 1969, 4, 48). Around the country, voters who rejected fluoridation saved themselves a few pennies in taxes per year in return for many unpleasant visits to the dentist and substantial dental bills. Thus, the “more democracy” they had, the more likely they were to harm their finances and their children. Why the self-defeating choices? The simple answer is that the voters were confused. Crackpots, rogue doctors, and extreme right-wing interest groups all fought fluoridation, and many voters, including a substantial fraction of those with college educations, could not sort out the self- appointed gurus from the competent experts. The same sort of popular confusion has arisen in recent years with respect to childhood immunizations. Parents getting their medical information from unreliable sources have sought to exempt their children from inoculation. In states that have responded to public pressure for easy exemptions, outbreaks of long- suppressed diseases like measles and pertussis have resulted (Centers for Disease Control and Prevention 2015).
Democracy for Realists is a timely book of empirically and theoretically rigorous political science. It's strong suit is criticism--Christopher Achen and Larry Bartels painstakingly amass evidence and arguments against what they describe as the two leading theories of democracy. The first is the "folk theory" that Democracy allows the will of the people to be expressed, something they dismantle with evidence on voters' lack of knowledge in both choosing their representatives and, especially, in referendums. They also throw in some Arrow Impossibility Theorem arguments around the indeterminacy of the "will of the people" even if everyone was well informed but had views that differed along multiple dimensions. The second theory they dismantle is the agency theory or "throw the bums out", where they show that much of the voting based on performance is actually about luck (e.g., natural disasters or shark attacks) not performance. I would note, however, that they may overstate their case--disasters shift votes by a few percentage points, which can tilt an election but only because the two sides were within a few percentage points of each other to begin with which, itself, merits an explanation--like the median voter theorem.
The book attempts to develop an alternative theory based on people's group identities and social psychology. They have some fascinating analysis--for example, the shift of white Southerners from the Democratic to Republican Party in the Civil Rights era was concentrated less among people who had specific issue preferences on racial issues and more among people who identified as Southern. They also find that Democratic and Republican men have views of abortion that line up with their political leanings, but that it was their views on abortion that changed to match their parties not the other way around. As interesting as all of this is, Achen and Bartels essentially admit is just an early attempt at an alternative normative and positive theory of democracy. And the policy recommendations in the book are limited to the last few pages, are relatively thin (e.g., less economic inequality and money in politics), and do not necessarily follow uniquely from the analysis itself.