APSA Presidential Task Force on Political Parties, More Than Red and Blue: Political Parties and American Democracy
Given the fact that it is now almost commonplace to shove a significant portion of blame for the pernicious hyper-partisanship now evident in the U.S. on the 1950 American Political Science Association study on political parties [See, e.g. The Polarizers or Breaking the Two-Party Doom Loop and my reviews of those two books here at 3:16 AM] one can’t deny the courage it must have taken for that august society of social scientists to give the same issue another public look. It’s not only the numerous mentions of the Schattschneider-led post-WWII report that make clear that the authors often had the earlier work at the back of their minds, but the much more cautious tone exemplified by the new report.
There was a supreme confidence exhibited in the 1950 advocacy for a system consisting precisely of two policy-coherent, “responsible” parties to replace the somewhat mushy links and fences then holding together Republicans and Democrats and separating them from each other. In contrast, the new study is hesitant and replete with caveats.
John Ishiyama’s Foreword is explicit: “The focus of the task force is not on recommending grand solutions to the current dilemma facing the country (such as recommending the introduction of a system of responsible parties that was part of the first APSA report on parties so many years ago)” Mark Brewer’s Preface leaves little doubt that the 1950 report was the impetus for what is said to be the sixth U.S. party “system.” The suggestion there is that we may now be entering a seventh, though we are not told exactly what a “system” is. I’m embarrassed to admit to flashing on an R. Crumb “Dis is a system?” drawing after seeing the term used repeatedly without explanation, but my ignorance was remedied to some extent by David Lublin and Lilliana Mason’s meaty Introduction. (There is indeed a Foreword, a Preface, and an Introduction here. An Executive Summary, Conclusion, and Afterward too!)
The Introduction notes that while the 1950 report “contended that the two major parties were not sufficiently differentiated to provide voters with meaningful choices, [t]oday American politics is in a very different place….[T]he contemporary Democratic and Republican parties are so polarized and incompatible that their differences prevent legislative action, fuel zero-sum thinking, and even provoke violence.” Later, there is an even clearer display of earlier-study-wariness: “[W]e do not intend to repeat the mistakes of the 1950 report–widely criticized for, among other things, ‘normative slovenliness and empirical inaccuracy [and for] contributing to some of the very problems, particularly polarization, that we now find our democracy facing.” The diffidence is so prevalent that while an opening chapter, by Susan Scarrow, is titled “How Much Should We Trust Political Parties?” it may be natural for readers to wonder how much we ought to trust APSA. But however one may answer that question, I certainly have not come to bury E. E. Schattschneider, who I believe to have been generally correct about most of the matters he discussed and who I think was wronged by the 1971 Kirpatrick “slovenliness” attack mentioned above. In fact, it’s my view that some of his most important insights are still not sufficiently appreciated–particularly his “Yes/No” limitation regarding what voters should be allowed/expected to express.
My point is rather to agree with the authors of the new study that APSA publications of this kind can have dramatic effects on the world, though, of course, this is less likely when they don’t speak with a single voice on behalf of any specific reform(s). In spite of the obvious difficulty in maintaining an impartial stance throughout the study, the authors have tried their best to display no favoritism toward either of the two major U.S. parties, without disguising the fact that most of the anti-democratic activity has been taking place among Republicans. After all, there were few (if any) Democrats involved in the attempt to prevent Congress from counting and certifying electoral votes on January 6, 2020 (though, if I recall correctly, there were a few days when “antifa” was accused by Republicans of committing most of the mischief).
So, it is natural that a pro-Democratic-Party bias sometimes evades the intended filters. In addition, a couple of the included papers indicate that a stronger labor movement would provide democratic safeguards, and one specifically mentions (Republican supported) “right to work” laws as having been damaging to democracy-enhancing unions. Obviously, it can be difficult for those who support democracy or lean left to maintain an appearance of impartiality; but a sincere attempt has clearly been made here. In her piece, Susan Scarrow reports that recent opinion polls indicate that in both the U.S. and Europe, political parties are trusted less than any other public institution, including police, courts, and legislatures. But she notes too that since every party is in an ongoing battle with one or more other parties–presumably unlike police, courts, or legislatures in that regard–it’s not terribly surprising that few people trust parties-in-general. After all, maybe they DO trust their own party and are only suspicious of the other guys: certainly, Americans tend to love their own Congressional representatives while loathing Congress-in-general.
Scarrow also points out that at least since the 1920s, powerful parties have been destructive of democracy in various countries around the world, a fact responsible for the growth of Karl Loewenstein’s “militant democracy” theory, according to which strict limits should be put on the sort of mischief that parties can get up to. Indeed, if militant democracy principles had been in place, there could never have been blackshirts in Spain or brownshirts in Germany. For it was Loewenstein’s view that states must be given a monopoly on allowable force. Scarrow notes, though, that a number of militant democracy tenets are difficult to square with political rights as understood in many contemporary jurisdictions, especially those with constitutions anything like the U.S. version. As she says, “militant democracy responses to perceived threats to the democratic order…resemble the very malady they seek to treat.”
When she turns from the question of how much we DO trust parties to how much we SHOULD, Scarrow tells us that this depends on how they are organized. If they are too weakly institutionalized, NO; if they are over-institutionalized, ALSO NO. Her conclusion, roughly, seems to be that the only parties we ought to trust are the ones that are trustworthy. Although it is not so much about parties specifically as about the attitudes of today’s voters generally, there are certainly no pulled punches in Zoltan Hajnal’s fine contribution, “Understanding the Demographic Sources of America’s Party Divisions.” Hajnal is persuasive in making his case that when determining whom voters will support, race far outpaces class, gender, education, age, religion, or anything else that has been considered by political scientists.
While Trump won only eight percent of the Black vote in 2016, he picked up 58 percent of the White vote, a whopping 50-point differential. This wasn’t quite matched by the 39-point difference for Asian Americans and Latinos, and it swamps explanations involving rich v. poor, male v. female, young v. old, or college v. no college. God-fearing v. non-religious was next biggest, but still not as important. It hasn’t just been a matter of Trump either. As Hajnol points out, “About 90 percent of the vote that McCain won in 2008, that Romney won in 2012, and that Trump garnered in 2016 came from White Americans. There was a slight shift in 2020, but 82 percent of Trump’s support still came from White America. The Republican Party is for almost all intents and purposes a White party.” Not only have racial fears and concerns now been clearly shown to be the dominant predictive factor for our partisan choices, but the world seems to be getting worse on this front. At any rate, the predominance of race considerations in our voting choices has been rising since the 1960s. And while that correlation has strengthened since those days, “there has been no clear increase in the importance of class in American politics over time.” In fact, Hajnol tells us that over that period, the “very rich have evolved from slightly Republican to slightly Democratic.”
In his interesting contribution on primaries, Seth Masket argues for the surprising thesis that “neither the existence of primaries nor their rules plays a particularly large role in the parties’ ideological polarization.” Masket nevertheless holds that “[p]arty nominations…are in need of significant reform–not so much to advantage the nomination of moderate candidates, but to advantage the nomination of candidates who will protect and advocate for democracy.” The argument here, so far as I understand it is that (1) there is no evidence that a different electorate is involved in primaries than the group that votes in general elections, and (2) the openness of primaries to independents and those of other parties is actually negatively correlated with the polarization of the legislators in the states where those primaries occur. I may be missing something here, but those two premises seem to me insufficient to show that primaries do not play a central role in the polarization of both candidates and the electorate. It seems to me quite plausible, at any rate, that the use of, e.g., Approval Voting over a list containing all the candidates (or even something more like the old system in which candidates could be selected by party bosses in smoked filled rooms) would produce more moderate victors than does the existing system, even if it is agree that, at present, the same voters are supposed to be involved in primary and general elections and the closedness of primaries is uncorrelated with victor polarization. (With respect to the latter point, I wouldn’t expect it to be uncommon for Democrat-leaning voters to support the most extreme Republican candidates in primaries, and for Republican voters to do the same thing in Democratic primaries: that’s just strategy. Also, I doubt many believe that primary elections are ENTIRELY responsible for polarization–only that they likely exacerbate it.
Of course, it must be difficult to provide hard evidence one way or the other on this matter since primaries are currently used nearly everywhere in the U.S. While Masket is unconvinced that primaries affect polarization, he is concerned that they bring other dangers. In particular, he takes the loss of effectiveness of party leaders to mean that voters may not be able to distinguish loyal partisan candidates from pretenders and that “populist movements and factionalism continue to undermine the choices of party leaders.” This is claimed to be bad because a number of candidates with little experience in politics are now winning elections and are “prioritizing posturing over legislating.’ And those proclivities, in turn, have meant that the sort of coalition building necessary for good government has declined rapidly in recent years. The electorate is seen to need strong parties to warn them of authoritarians today, just as they did back in the days of Lindbergh and Long. Among the reforms Masket suggests to address the (in his view non-polarizing) problems engendered by primaries, is a requirement that party officials approve of candidates before the latter can carry their banners. It is further suggested that parties could raise debate participation thresholds or even impose more hurdles to be cleared in order to vote in primaries at all. Those both seem like anti-democratic moves to me, but perhaps my views are heretical–or simply confused–on this matter: see the final paragraph below.
Jake Grumbach’s “Protecting Democracy from State Level Threats in the Age of National Parties” warns that the U.S. version of federalism distributes more power to subdivisions than is usual around the world. And, he tells us, it is particularly unorthodox in providing subdivisions with the authority to devise democracy rules. According to Grumbach, some good comes from such decentralization, since it makes an authoritarian takeover of an entire nation more difficult. But a lot of bad is also produced via gerrymandering, state laws making voting more difficult, appointments of clearly partisan election officials, and the like. I have argued elsewhere ["Abizadeh on Majoritarianism"] that the parceling out of government authority via a federal structure of government must be in one sense undemocratic, because it is necessarily anti-majoritarian, This seems to me also to be the case for divvying out election-determining authority to political parties. However, the benefits provided by federal architectures and political parties may outweigh the harms they impose on majority rule. Those are gnarly questions. My own sense is that no holy grail of perfect power balances will ever be found. But wherever those lines are drawn, certain democratic principles must simply be considered foundational and unalterable.
It doesn’t whether they arise locally or federally, because they have to be in place on both levels or there will be no real democracy. In the end, Grumbach seems to agree with this assessment, for after mentioning two sorts of reform strategies he thinks may help: more effective punishment of opponents’ antidemocratic norm violations and rebuilding of the labor movement, his closing paragraph turns both from such tactical counsel and from the concerns previously expressed over the level of government that is properly be involved in rule-making. He looks instead to the matter of what the rules are that should be passed. He writes, “[V]oting rights policy matters. In a previous era of democratic divergence in the states, Congress played a pivotal role in defending democracy by passing the Civil Rights Act and Voting Rights Act of the 1960s.” While he suggests that “neutralizing new threats to democracy that are arising from the states will very likely also require updated federal legislation that guarantees baseline civil and voting rights across the entire country,” it is obvious that if it were the feds wreaking havoc with democracy, as could clearly happen with a Republican Congress in the near future, the “neutralization” would be required to take place in the several states.
In other words, it’s not so much who does the regulating that matters, but what the regulations are. In her “Mass Political Behavior and Party Incentives,” Lilliana Mason turns to group psychology in an attempt to understand the messages that parties are both receiving from and sending to their variously motivated supporters, and looks into the levels of success produced by the strategies being implemented. Mason notes that when voters become extremely partisan, winning may take precedence not only over government handling of problems but even over the quality of life enjoyed by the partisan supporters. “In this scenario,” she warns, “partisanship can become a driver of political action, intolerance, ethnic resentment, and even violence.” Because the Republican Party is in the minority on a number of key issues like gun control and reproductive rights, Mason plausibly accuses its leaders of knowingly pushing “identity-centric rhetoric—which can motivate voters without offering them economic or practical benefits…By leaning on threats to group status and grievance narratives, the Republican Party is uniquely positioned to benefit from partisan animosity and conflict. A straight governance debate does not benefit their candidates.”
Things may not be entirely hopeless, however. Mason points out that “MAGA” supporters (who believe the election was stolen, that strong leaders are more important than democracy, that armed citizens should patrol polling places, that there’s a conspiracy to “replace” white Christians, etc.) do not make up the entirety of Republican supporters. That being the case, she believes that party leaders may have reasons to model less aggressive stances and behaviors. Furthermore, this situation may afford Democratic Party leaders the opportunity to target democracy-supporting Republicans. The article cites a number of studies seeming to show that things may improve in such an environment. Nearly every chapter of this report begins with a litany of dangers facing democratic self-government in the U.S. (dangers that, to be honest, are likely to be familiar to anybody with access to newspapers, magazines, TV, radio, or the internet), and closes with a few proposals that, perhaps based on a prior study or two, might help shore things up.
But the most sweeping (and, at least to me, discouraging) assessment here may be Rachel Beatty Reidl’s “Factions, Moderation, and Democratic Responsibility.” Her comprehensive description pinpoints a number of reasons that crumbling democratic safeguards may be impossible to rebuild. For example, the decentralization of media and campaign funding has served to make parties and Federal rules less essential to candidates, and, indeed, less relevant to the electorate at large. It’s hard to imagine those conditions changing dramatically. So, even if party leaders DID want to moderate their members–and, of course, there may be significant incentives for them NOT to do this–it is doubtful how far they could get. Some party elites may have personal reasons to keep things from going completely off the rails: they can’t retain power if their party disintegrates; but “[i]nternal party divisions and factions–whether over identity or ideology–can disrupt this elite pact undergirding democratic regime stability. Political entrepreneurs seek to build their own personalist following, [leave the] party, and build an anti-system democracy ‘for the people’ rather than through institutionalized mechanisms of horizontal accountability (checks and balances or vertical accountability (free and fair elections that represent the will of the electorate)....[H]igh degrees of partisan polarization…folds all social cleavages into one partisan dividing line, hardens loyalties, and prevents such distancing of the extreme factions from the levers of party power.”
My own sense, though perhaps I’m mistaken, is that Beatty Reidl’s faith in the likelihood that any proposed reforms could be effective given the current disposition of parties, legislatures, courts, media, funding possibilities, etc., is pretty feeble. Two papers pushing electoral reforms, “Encouraging Cooperation and Responsibility” by David Lublin and Benjamin Reilly, and “Toward a Different Kind of Party Government: Proportional Representation for Federal Elections” by Jack Santucci, Matthew Shugart, and Michael Latner, if not much more optimistic about the likelihood of seeing their proposals become reality than are other authors in this volume, at least seem more sanguine about the partially curative properties of the potions they prescribe. The Lublin/Reilly piece focuses on the current ideological distance between the most liberal Republicans and the most conservative Democrats (a distance likely lengthened by the APSA “responsible party” advocacy of the 1950s and 60s) and suggests–contrary to the Masket paper discussed above–that “[m]ost primary challenges to incumbents come from the Left for Democrats and the Right for Republicans, creating incentives for incumbents to shift toward ideological extremes to preempt primary challenges….Bluntly put, combined with demographic sorting, electoral reform making primaries the method of choosing most party nominees got the U.S. into this hole and helps maintain polarization.”
Lublin and Reilly reasonably suggest that several voting reforms might counteract this trend. Both ranked choice voting (“RCV”) and the institution of two-round elections could, they say, incentivize more moderate positions and limit the effect of vote-splitting. As they put it, each would place “a centripetal counterweight…to the centrifugal forces of sorting, divisive mobilization and negative partisanship.” (They note, however, that while Arkansas adopted a two-round primary in 1924 with the intent of getting anti-Klan voters a more moderate candidate in the second round, Mississippi put runoffs in place for the obvious purpose of preventing any Blacks from win via plurality backing only.) In support of the moderating effects of their proposal they cite a 2007 study according to which “extremists have almost no chance” under a two-round system. (Ah, for the return of 2007!)
RCV, of course, is a kind of “instant runoff” and should therefore share some of the same merits as its slower-moving cousin. The authors suggest that RCV has been beneficial in Australia and that the early indications from Maine and Alaska in the U.S. are promising. As RCV can be confusing and exhausting if voters are given fifty candidates to choose from, an alternative proposed reform made here is for “Final Four (or Five) Voting” which is thought to produce many of the same benefits. Here, partisan primaries are replaced by a non-partisan, first round nominating election, and the top four (or five) candidates go on to face off in an RCV final. This scheme is said to produce moderate victors, but, because both the early votes and finals may contain multiple candidates from each party, it may also weaken political parties generally. Certainly, no party leader will have much control over any of its favored candidates. As the first (“pick one”) round of Final Four has multiple winners and is therefore, in essence, an example of a Single Non-Transferable Vote (“SNTV”), and that system has earned a fairly poor reputation around the world, a number of observers are here given two reasons (party-weakening, and poor track record) to dislike this proposal. I happen to like SNTV myself, though I would administer it differently than it has ever been hitherto [see my book], and I remain ambivalent about how powerful I think parties and their leaders ought to be, so I am likely more sympathetic with this suggested reform than other commentators may be.
For what it may be worth, I am also convinced of the absolute essentiality of proportional representation (“PR”) to any authentic democracy, so I find that aspect of the Santucci/Shugart/Latner article quite congenial. If more parties– or different kinds of parties than we now have–are necessary for minority viewpoints to be given their appropriate voice in government, then it is my view that such changes absolutely must be made. The questions for me are whether strong parties, which seem clearly needed for policy coordination [see my piece on this] are (i) required to achieve PR, or (ii) whether they would yield other benefits that are required for authentic democracy. (Fortunately, the final article in this anthology is focused on question (ii).) The Santucci/Shugart/Latner piece covers a lot of ground and is generally more complicated than the other articles found here.
I imagine the authors struggled with the space constraints imposed on them. The objections they make to the 1950 APSA report are interesting and incisive: “[A] responsible party is almost impossible under a presidential form of government. The separation of executive and legislative electoral processes and the separate survival of the elected branches (via fixed terms) provide too many countervailing incentives for party-policy responsibility to be feasible.” Where that study mostly went wrong, in their opinion, was its acceptance of a specifically two-party structure. Besides ignoring the presidential aspect of the U.S. government, the earlier APSA work both downplayed any polarization potential of a two-party system and failed to consider the importance of minority representation. It’s here claimed that what is really needed is a multiparty presidential democracy with PR for the U.S. House and some minor reforms (like fusion voting) for the single seat elections of the President and Senate.
Three possible types of PR are discussed: mixed member proportional, the single transferable vote (“STV”), and open-list proportional. One gets the sense that the goal here is to mirror—as far as is possible in a Presidential country—a multi-party parliamentary systems, something which would put “responsibility” for making government policy in the hands of coalitions of several parties. Among other benefits, the authors believe this would provide more opportunities for voters to make meaningful shifts to focused parties with which they are more comfortable. (I myself think a simpler way of achieving new party creation would be to require party registration for anyone wishing to vote in a general election. It seems clear to me that a country with so many “independents” unwilling to join either the Democratic or Republican Party as they currently exist, would quickly sprout new parties if its citizenry were forced to commit to some group or other in order to be given the opportunity to vote at all.) The authors claim that concerns about party fragmentation and coalition difficulties that have long been associated with PR are “mostly caricatures of real-world PR systems” and would in any case be nearly impossible in the U.S. context because of the single-winner Presidential and Senatorial elections. In addition, the relatively small number of House seats currently available are said to prohibit a huge amount of proportionality in any sort of PR system that could be set up here.
The authors don’t suggest that PR is a panacea for achieving the expression of minority viewpoints in government, citing several articles indicating that “even in proportional systems candidate-centered ballots permit voters to withhold support from minority candidates,” but I hope most readers will agree with them regarding the need to provide members of larger minority groups who fail constitute a majority in any particular geographical area a real voice in their government. The authors note that “large parties” can intentionally foul up mixed member PR by creating “decoy lists,” but their discussion of STV makes clear that they do not think dilution of party influence is ever a very good answer to electoral problems. They note in conclusion that “PR adoption in the United States far from guarantees a multiparty system,” but I suspect there may be many readers who will join me in thinking that the provision of minority representation has great importance on its own, i.e., whether or not it is guaranteed to bring new parties with it. Both PR and multi-party democracies have come and gone around the world and it is naturally quite hard to make any bold suggestions regarding what will “work” or “last.” Political scientists have the extremely difficult task of making predictions of that sort.
Fortunately, political philosophers can focus on the (still vexing, but perhaps somewhat less perilous) question of what sort of architecture authentic democracy requires. There seems to me little doubt that PR is among the essential characteristics of any approvable structure. With respect to the necessity of “strong, responsible parties,” as indicated above, my admittedly only half-educated sense has for several years been [largely based on the work of Bawn et al.: see the blog piece linked above] that coherent policy coordination makes something like a party system unavoidable. I was therefore glad that the excellent final article in this collection, “Prioritizing Parties,” by Seth Masket and Hans Noel, sets forth an additional array of reasons why modern democracy in the U.S. can be said to be basically unthinkable without parties. These considerations have made these authors less excited about reforms like RCV, which can weaken existing parties. They don’t deny or make any excuses for the abuses that riddle the modern Democratic and Republican Parties (again, mostly the Republican one), but warn not only that there can be no decent democracy without strong parties, but that “efforts to block parties generally fail to exclude them altogether, and the party-weakened alternatives that emerge are often worse than what existed before.”
One can almost feel lurking behind many of their arguments and admonitions one loud and insistent rumble: If the Republican Party had been stronger, it could have prevented an authoritarian demagogue like Donald Trump from being its candidate for President in 2016. After all, the party leadership had very little use for the guy! Deep discussions of parties can never avoid this crucial–but quite confounding–argument, so “Prioritizing Parties” was a great choice to put as the denouement of this study. But even after turning its final page, the big question is likely to continue to hammer at the reader…or at least it still drums away at me. Suppose the American people want some particular person to be their leader. Let’s say a significant majority likes her more than they do anybody else. And suppose, too, that no party wants this woman to display their banner. Maybe they think she’s a corrupt, stupid, lying, self-serving, anti-democratic autocrat who hates parties and might actually do away with them entirely if given the chance. Now, democracy seems clearly to require that citizens get the leadership they want. But political scientists confidently assure us that democracy has also been found to absolutely require the existence of parties—indeed, strong ones—not only to produce meaningful elections, but for any sort of coherent governance at all. What should one conclude given this dilemma? Scylla or Charybdis? Is there a middle way?
O Idee, du Idee, das mir fehlt!
About the Author
Walter Horn is a philosopher of politics and epistemology.
His 3:16 interview is here.