Read an Excerpt
NO MIDDLE GROUND
How Informal Party Organizations Control Nominations and Polarize Legislatures
By Seth E. Masket THE UNIVERSITY OF MICHIGAN PRESS
Copyright © 2009 University of Michigan
All right reserved. ISBN: 978-0-472-11689-8
Chapter One
The Modern Political Party
For decades after its 1974 publication, David Mayhew's Congress: The Electoral Connection was considered the premier text for anyone wishing to understand how members of Congress behaved. The description was not a pretty one for anyone who believed in the value of party government. According to Mayhew, it was easiest to think of members of Congress as "single-minded seekers of reelection." Candidates, he found, didn't have much use for parties other than to help mobilize voters. Attention to party agendas didn't change much once politicians got into Congress, either. "To a remarkable degree," Mayhew noted, "members can successfully engage in electorally useful activities without denying other members the opportunity to successfully engage in them" (82). Why didn't members of Congress press a party agenda? Apparently, it was a matter of choice, and they chose not to. "American congressmen," Mayhew wrote, "could immediately and permanently array themselves in disciplined legions for the purpose of programmatic combat. They do not" (98). Any party battles that occurred were, for the most part, theater;members of Congress maintained strong friendships across party lines and hewed closely to the median voters in their districts. Reelection was far more important than the advancement of any ideological agenda.
Mayhew's descriptions were merely the most eloquent of an era of scholarship extending from the early 1950s to the early 1990s that depicted American politicians as nonideological poll-watchers, sticking closely to the median voter in their districts and avoiding any stances that would offend the general electorate. The beginnings of this period are marked by the American Political Science Association's report "Toward a More Responsible Two-Party System," which decried the lack of responsibility in the weak party system. "Alternatives between the parties are defined so badly," the authors claimed, "that it is often difficult to determine what the election has decided even in the broadest terms" (APSA 1950: 3-4). One of the last gasps of such depictions was Hill Rat, an entertaining tell-all from a former Capitol Hill staffer. As the author described congressional committee work, "You sit around a table and divide up the money. Anything that gets in the way of that process-philosophy, conscience, and so on-gets checked at the door" (Jackley 1992: 103).
The literature of this period generally describes a Congress full of ambitious, nonideological politicians. Interestingly, it closely charts an unusual period in congressional elections during which candidates actually tended, on average, to represent the median voter. An innovative study by Ansolabehere, Snyder, and Stewart (2001) demonstrated that, at least since the mid-nineteenth century, candidates for Congress have deviated strongly from their districts' median voters, stubbornly defying Downs's (1957) prediction of convergence. The exception seems to be a period from the late 1940s to the early 1980s.
Despite the relative oddity of this weak party period, its view of nonideological politicians is the one that is most ingrained in American popular culture. Notably, one of the most common epithets in modern political discourse is "flip-flopper," directed at those who appear to have no core beliefs. The essential plotline of 1990s political movies like Bullworth, Dave, and The American President is that a politician has lost his way, abandoning the issues and agenda that drove him to run for office in the first place, and is now "so busy keeping my job I forgot to do my job." Only when the politician has some sort of catharsis (a near-death experience, falling in love, etc.) does he remember to "fight the fights that need fighting," rather than the fights he can win (Reiner 1995).
This view of politics is hardly confined to Washington. Many state governments went through a similar period of weak partisanship, avoiding the most controversial issues in order to protect their political class. "I've been on a tour of state legislatures," humorist Mark Russell joked in 1991. "Mostly they are a bunch of fat white guys pretending to hurt each other" (Richardson 1996: 360).
The pretense, for the most part, is gone. The literature from that weak party era is difficult to reconcile with the modern Congress and state legislatures. Today, by virtual consensus, the parties in Congress and in many state legislatures have diverged (Poole and Rosenthal 1997). Candidates no longer converge on the median voter. They have returned to their old historical pattern of representing the ideologically extreme elements within their parties, despite the electoral risk that this strategy carries (Canes-Wrone, Brady, and Cogan 2002; Wright and Berkman 1986). While it was once common for members of different parties to call each other friends, today that almost never happens, and insults and even physical threats across party lines are becoming more common in Congress (Jamieson and Falk 2000).
The new version of Homo politicus is the eager party warrior; legislators recognize the advantages of partisan action and willingly engage in it. As with any other form of legislative organization, this one exacts a price: sometimes legislators must take stances that differ from their personal opinions or those of their constituents. A moderate Republican may feel that his party's deficit spending is endangering the country's fiscal standing and wish to oppose a tax cut, but he knows that he must stand with his colleagues. Similarly, a moderate Democrat may stand with his copartisans in opposing Social Security reforms even if he believes they are important for the program's long-term stability. Both these politicians recognize that they will, in the long run, do better, both in terms of their policy goals and their reelection efforts, if they stick with their party coalition. Even if they are derided for hackery, this is a decision that rational politicians will make.
Indeed, in recent years, an entire literature has evolved to explain why rational legislators would participate in strong parties rather than eschew them (Cox and McCubbins 2005; Aldrich 1995; Cox 1987; Cox and McCubbins 1993; Kiewiet and McCubbins 1991; Schwartz 1989; Volden and Bergman 2006). Parties, in this view, are "institutional solutions to the instability of majority rule" (Aldrich 1995: 72). Deliberative chambers are inherently chaotic-to get a bill passed, one must cobble together a majority and hold it together over numerous votes, any one of which could doom the bill. The easiest outcome in such a chamber is for nothing to get done. Parties, however, provide a solution. As standing coalitions that don't have to be constantly reassembled, parties make it possible for individual legislators to pass bills while providing a path for career advancement. Incumbents submit to parties because they make it easier to get legislative work done.
But why is it important to get work done? Cox and McCubbins elaborate on this with their theory of parties as legislative cartels (1993, 2005). According to cartel theory, politicians recognize that their reelection prospects rise and fall with those of their fellow partisans, so they seek to create and enhance their party's brand name, "the commonly accepted summary of the past actions, beliefs, and outcomes with which it is associated" (1993: 110). They do so by ceding power to a party leadership that can manipulate the legislative agenda, which it does in a way that helps their collective party reputation. Thus members can run in the next general election on a collective party label that resonates with a sizable chunk-ideally a majority-of the voters.
A key question, however, remains: Why has the behavior of politicians changed? Why are legislators who, by their nature, prefer to avoid real partisanship increasingly behaving as partisan warriors? This question is all the more puzzling since, as surveys show, voters increasingly claim to dislike parties and partisanship. If neither legislators nor voters like strong parties, who does?
As suggested in the introduction, these questions proceed from the wrong assumption: that parties are a function of politicians' preferences. When Aldrich (1995: 4) says that "the major political party is the creature of the politicians, the ambitious office seeker and officeholder," he is saying something of a piece with Mayhew's (1974: 98) claim that members of Congress "could immediately and permanently array themselves in disciplined legions for the purpose of programmatic combat" but "do not." Both, that is, are saying that we should look at politicians if we want to understand parties.
An alternative to this approach comes from the traditional party organization literature, which tends to see parties as hierarchical, patronage-based groups whose power lies in the control of nominations (Mayhew 1986; Dahl 1961; Kent 1924; Wolfinger 1972). The key to Mayor Richard J. Daley's power in Cook County, Illinois, for example, was his control over party nominations. Daley routinely slated candidates for every office from the board of aldermen to county judgeships to the congressional seats within the county limits, and these candidates just as routinely won nomination in direct primary elections. According to one ward boss, "He moves us around like a bunch of chess pieces. He knows why he's doing it because he's like a Russian with a ten-year plan, but we never know" (Royko 1971: 82). Daley also exerted considerable control over the Democratic Party's choices for governor and senate during his tenure, and his reputation as a kingmaker in Democratic presidential nominations during the 1960s is legendary (Mayhew 1986: 74-75).
Studying parties as organizations or machines that try to elect people helps us understand why the same politicians that are soulless poll-watchers in one era can be fierce party warriors in another. But it doesn't explain the dilemma at hand. That is, by virtual consensus, the party machines in the model of Daley's Chicago or Tammany Hall's New York don't exist anymore. Why would partisanship among elected officials increase as the machines go extinct?
We gain some purchase on this question from a strain of literature known as conditional party government (CPG) theory, which has the virtue of focusing both on events within the legislature as well as on forces outside it. As this theory describes, legislative party leaders have an array of tools available to foster party discipline. They can enforce unit rule voting in the party caucus, they can employ party whips to cajole members, they can award office space, staff, or popular committee assignments to loyal party members, and they can deny such perks to mavericks. However, the theory continues, the leaders' ability to use such methods is conditional on electoral forces outside the legislature (Rohde 1991; Aldrich and Battista 2002; Aldrich and Rohde 2001; Rohde and Shepsle 1987). Specifically, if the electorate sends an ideologically coherent party contingent to Congress, that party can build on its coherence by enforcing discipline on roll call voting. An ideologically incoherent party, such as the post-World War II Democratic Party with its Northern liberals and Southern segregationists, will have little success in whipping its members into line.
It would seem that understanding these forces outside the chamber is at least as important as understanding partisan tools within the chamber if we truly want to comprehend what parties are and how they function. It is surprising, then, that more attention is not devoted to such extra-legislative forces, and all the more important that we try to understand them. So what are these forces that determine the partisan nature of our elected governments? From the literature, it would seem that there are two of them, voters and elites.
It was voters that Rohde was speaking of when he described "the exogenous influence of electoral change" (1991: 162) as a cause of polarization or depolarization in a legislature. That is, voters shift their preferences from time to time and vote accordingly. If a party's voters are relatively unified on a set of issues, they will tend to elect a coherent party contingent to the Congress, which will vote consistently as a bloc. If a party's voters can't agree on much, they will elect leaders who can't agree on much, either. As Rohde notes, a good deal of the variation in unity among the congressional Democratic caucus on civil rights votes can be explained by voters' shifting preferences on racial issues. Jacobson (2004) and Aldrich et al. (2007) also find that much (though not all) of the recent polarization of the congressional parties can be explained by the increasing homogeneity of congressional districts. Indeed, the whole literature on party realignments is predicated on the notion that voters' preferences will occasionally shift in a dramatic, often unanticipated way, forcing elected partisans to change their behavior or risk electoral defeat (Burnham 1965; Schattschneider 1942; Sundquist 1983; Rogin and Shover 1970). Thanks to shifting voter preferences, for example, officeholders in the mid-1850s could no longer equivocate on slavery, as the Whig Party had essentially done for decades. You were either with the slaveholders or you were against them. An officeholder who preferred Whig economic policies but supported the institution of slavery suddenly had to make a choice, and had to make it quickly.
This view that voters determine changes in party coherence and agendas has a hard time surviving Bartels's (2000) observation that shifts in voter partisanship tend to follow, rather than precede, shifts in congressional partisanship. It seems, that is, that voters follow the cues of party leaders rather than the other way around (Hetherington 2001). These objections strongly suggest the importance of the second category of extralegislative forces: outside elites. This term refers to activists, party bosses, interest groups, and other key political actors outside the government with a strong interest in affecting government behavior. These outsiders wield power over legislators through their control over party nominations and the resources needed to win them. If such outsiders serve as gatekeepers to holding public office, we would expect that shifts in external party coalitions will lead to important changes in legislative voting behavior, even if there has been no concomitant shift in mass voting behavior.
Historically, these outside actors have been able to alter politicians' behavior without changes in the preferences of the mass public. For example, some Democratic officeholders' embrace of civil rights in the late 1940s was not an adjustment to voter demands. Rather, the interest group Americans for Democratic Action essentially forced the issue of racial equality on President Harry Truman, whose nomination in 1948 was far from assured. Truman sided with the ADA and won the nomination, but this led to the Dixiecrats bolting the Democratic Convention in 1948 (Bawn et al. 2006). Similarly, recent literature has cast doubt on the designation of the 1896 election as a critical realignment of the electorate (Mayhew 2004; Bartels 1998; Stonecash and Silina 2005). There were few changes in electoral behavior around that period. Instead, it was a matter of political activists seizing power: Western silver-coinage advocates took control of Democratic nominations from eastern gold-standard supporters. Other examples of outsider-led (rather than electorate-induced) changes in party priorities include the creation of the Republican Party in the 1850s and the takeover of that party in 1964 by the conservative Goldwater wing. What is left of realignment theory is pretty well decimated by Mayhew's (2000) analysis, in which he concludes, "The claims of the realignments genre do not hold up well, and the genre's illuminative power has not proven to be great" (471). Specifically, valence issues (Stokes 1966) or even random coin-tosses will explain more of presidential election history than realignment theory will.
Note, then, where this leaves us. Party leaders in a legislature may desire strong party behavior, but ultimately they are at the mercy of forces outside the legislature. And while voters may wish to control their elected officials, only a few key elites outside the legislature have the power over party nominations to determine just what the legislative parties will look like and fight for. But in order for these key elites to actually exert such control, they need to coordinate their activities. They need to be organized. So what does such an organization look like? What follows are two examples of modern party organizations in California. (Continues...)
Excerpted from NO MIDDLE GROUND by Seth E. Masket Copyright © 2009 by University of Michigan . Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
