The Big Sort

by Bill Bishop

  These four politicians looked tired, survivors whose talents for politics had become obsolete. They found themselves living in a political ice age. Benson recalled how he and a Democratic legislator had put together the rarest of laws, a compromise on abortion. "The bill would have reduced the number of abortions, but it would have also guaranteed rights," Benson said. "We thought naively that we had the formula. It was a wonderful experience, and we went through the whole process and we got two votes. His and mine. Because everybody else was pushed to their base." Penny, suddenly excited, said, "Now get this. Their proposal would have measurably decreased the number of abortions in Minnesota, and it would have not made abortion illegal. But they couldn't get support from either extreme." There was a bit of silence, and then Benson said, "What we have today is idea segregation."

  Dave Bishop said that he used to get much of his best work done at receptions, where he could talk with Democrats and form cross-party coalitions. But legislators don't mix much anymore, he said. Ethics laws have cut down on the number of social gatherings, and besides, the two sides don't have much to talk about. They don't like each other. Kiscaden agreed. "The legislative system was designed for deliberation, to take people away from their homes and put them in one place to talk to each other and work it out," she said. "But that only works as long as people really trust their representatives." And people don't trust anymore, said Bishop. Legislation in St. Paul is "just pounded through" by the majority, he said. "It's a perversion of the strength of the legislative process."

  Kiscaden and I walked across the street and into the Mayo Clinic. Massive swirling blue and gold glass constellations by artist Dale Chihuly dangled over one staircase. We roamed about the clinic, and Kiscaden told me about the "coffee and conversation" meetings she had organized in Rochester. More than one hundred people would turn out on a Saturday morning just to talk about their town and state. One January, she held a "Beyond Bickering" seminar for state legislators. About 75 out of 201 showed up.

  Kiscaden eventually joined the Democratic-Farmer-Labor Party and in early 2006 agreed to run for lieutenant governor on a ticket with a moderate Democrat, Kelly Doran. The campaign never gained traction; Doran and Kiscaden dropped out before the primary. Kiscaden said that she wouldn't run for her old seat in the state senate, and so, for the time being, she is out of politics—not because she wanted out, but because she really doesn't have a place. Her situation isn't that unusual.

  In 2004, a rancher and former school board president from Waco, Texas, entered the Republican primary for a newly redrawn U.S. House district in central Texas. Dot Snyder ran against Arlene Wohlgemuth, who billed herself as "one of the most conservative members of the [Texas] legislature."* The winner would take on Democratic incumbent Chet Edwards in a district that was 60 percent Republican. Edwards, a conservative, had been elected seven times, and everybody expected the general election to be a doozy. But the real fight ended up being the primary between Snyder and Wohlgemuth.

  Wohlgemuth attacked Snyder for what in a less polarized time would be considered good citizenship. Snyder had served on the board of Waco's chapter of Planned Parenthood. (She had resigned when Planned Parenthood began to offer abortions.) And while serving on the school board, she had voted for some tax increases. (The state legislature at the time was systematically reducing its share of funds for public schools, so every Texas school board was raising local tax rates.) The conservative Club for Growth labeled Snyder a "Republican-in-name-only" (RINO) and pushed $400,000 into the Republican primary to support her opponent. Wohlgemuth was a "taxpayer superstar," the Club for Growth said, and the little primary in the blackland prairie of central Texas became the club's "highest priority" House race in the spring of 2004.

  Snyder's ranch is near Crawford. Her husband was one of George W. Bush's fraternity brothers. Snyder was a volunteer worker on the Bush campaign the week before the 2000 and 2004 elections. When I spoke with her in 2004, she described herself as a "rock-solid, 100 percent Republican." But she wasn't conservative enough for the Club for Growth. Overwhelmed by the Club's money, she lost in a primary runoff,† "The process was an awakening for me," Snyder said. "I did not expect to be misrepresented. And I think that the process perhaps forces that to happen ... Voters in the primaries are certainly not representative of those who vote in November. They skew to the edges." Would she run again? "No," she answered, laughing. "That's easy."

  Incumbent Arlen Specter received only 52 percent of the vote in the Republican primary for his U.S. Senate in Pennsylvania in 2004. His opponent, Representative Patrick Toomey, a conservative, labeled Specter a "Ted Kennedy" liberal, and the Club for Growth spent $2.3 million to broadcast that message. Toomey didn't win, but the Club for Growth didn't give up. The conservative group turned its attention to the Pennsylvania statehouse, targeting a dozen moderate Republicans. When Pennsylvania conservatives in the 2006 primary defeated twelve Republican state legislators they considered insufficiently doctrinaire, the Club for Growth announced, "Toomey Beats Specter." (These folks certainly can hold a grudge.)

  St. Paul mayor Randy Kelly, a Democrat, endorsed Bush in 2004 and was defeated one year later as punishment. After a few Republicans on the Houston city council supported the Democratic majority's proposal that stalled cars be towed immediately off the city's notoriously clotted freeways, local Republican officials promised retribution.* "We're not looking for council members who are going to go along and get along," said Jared Woodfill, chairman of the Harris County Republican Party. "We're looking for council members who are going to stand up for conservative values."47 Surely, political ideology has teetered over some high cliff when towing can be described as a "value." What's next, a doctrine of potholes, the water pressure credo?

  The One-Way Street of Polarization

  Over the past thirty years, most U.S. House districts haven't flipped back and forth in their political allegiance, or even in the ideology of their representatives. There's little ebb and flow in results because the composition of the districts isn't ebbing and flowing. Congressional districts are, as a rule, growing more politically homogeneous. When older officials leave office, they are inexorably replaced by someone further to the left or the right. Nolan McCarty, Keith Poole, and Howard Rosenthal have charted the increasing polarization of Congress. From the end of World War II until the mid-1970s, the political scientists found, the mix of the two parties in the U.S. House was fairly stable. Republicans tended to be conservative, and Democrats were largely liberal. But politicians of different parties overlapped in their ideology. After the mid-1970s, the two parties began to move apart. Democrats got more liberal and Republicans got more conservative.48 In the 95th Congress (1977–1979), 40 percent of the 435 members were moderates on the McCarty/Poole/Rosenthal scale. By the 108th Congress (2003–2005), this moderate bloc had been whittled down to 10 percent. The U.S. Senate has seen a parallel decrease in overlap between the two parties (see Figure 10.1). The middle has dropped out of Congress altogether.49 Some of that polarization was due to conservative southern Democrats being weeded out of the party (by either conversion to Republicanism or retirement). But members of Congress from every region in the country had moved away from the political center. The South has its peculiarities—country ham, Bob Wills, and Dr Pepper—but polarization is the product of a nation.*50

  McCarty, Poole, and Rosenthal observed that polarization has consequences. Without a sizable group of moderates, Congress has a harder time passing major legislation. The parties engage in the horseplay Americans see daily in Congress—the endless efforts to blame the other side for failures and the mountainous debate over trivial legislation—while significant issues are sidelined. Moreover, these political scientists found that when Congress is gummed up, the president and the courts are encouraged to act unilaterally.†51 "It's calcifying our politics," Democratic pollster Paul Maslin told me. "Where is the common ground if I as a Republican legislator in the Central Valley of California never h
ave to worry one damn bit about ever losing an election to a Democrat?" The same is true for a Democrat in Los Angeles, of course, and so both steer to the extremes, conceded Maslin, who was Howard Dean's pollster in 2004 and joined Bill Richardson's campaign in 2007. "So now if you have legislative bodies that can't act anymore, they either go lurching from one side to another, which means we have no continuity, or they are immobilized. If they are immobilized, that breeds cynicism: 'I hate the politicians."' Maslin described a perpetual motion machine of polarization. "It means the middle starts opting out. Now you are creating a real snowball effect. The middle is opting out because they're turious. This whole thing is a mess."

  The most effective sessions of Congress were from 1963 to 1966, just before the midsixties social and civic collapse. From that point on, the House and Senate grew more ideologically balkanized and, at the same time, less productive. The percentage of important issues ending in deadlock has increased since.52 "The moderates have been washed out of both parties," lamented former Republican congressman Mickey Edwards of Oklahoma. "There's just a homogeneity there" in the districts and among politicians, Edwards continued. Is the polarization of Congress led by elites or by partisans back home? It's both, said Edwards. "It's who they are and who they represent."

  Figure 10.1 The Decline of the Political Center, 1949–2005

  Moderates have been disappearing

  from the U.S. House and Senate.

  Percent of Moderates in Congress

  Source: Keith T. Poole, University of California, San Diego.

  Thomas Jefferson circumvented the boarding house factions during his presidency by inviting legislators to the White House for dinner, no more than a dozen at a time. He sat them at a large, round table, both to nullify questions of status in seating and to preclude private conversations. The wine was imported and plentiful. Jefferson's French chef was "his best ally in conciliating political opponents." Servants weren't allowed in the room—the president served out of a dumbwaiter near his chair—so the legislators felt free to speak in confidence. Jefferson wouldn't mix Federalists and Republicans, but he would use the dinners to cross over the boarding house factions in both parties.53

  Congress has been most productive when both parties have been ideologically mixed and when the members have soothed political differences with social grace. "In the fifties and sixties, these guys hung out in the gym," said Gary Jacobson, a University of California, San Diego, political scientist. "They socialized at functions held by lobbyists, across parties. They hung out in Washington together, and they formed friendship on affinities that had nothing to do with politics. It reduced the level of conflict. Now they don't talk to anyone except for people in their own party ... It's a very different world."

  People living in like-minded communities don't regularly encounter political disagreement, observed Alan Rosenthal, a state legislative specialist at Rutgers University. Voters living in segregated neighborhoods and churches "are less likely to have any empathy with politics or legislatures, because fundamental to those politics and legislatures is disagreement. If you live in a homogeneous community, you think, 'Everyone agrees this is right, so why the hell don't they do it?' Well, they don't do it because there is another homogeneous community over there that thinks quite the opposite. And candidates who come up through that [community] may also have less appreciation for democratic politics."

  And so may voters.

  11. THE BIG SORT CAMPAIGN

  THE EARLY RETURNS from exit polls on Election Day 2004 predicted that John Kerry would win. Kerry didn't win, of course, and the contrary findings of the exit polls launched a thousand conspiracies of stolen elections and monkeyed-with voting machines. The botched survey results were embarrassing, so the group that operates the exit polls investigated and eventually issued a long and quite detailed report on how the system had failed. To understand the findings, you have to know a little about how the polls were conducted. Interviewers set up at 1,480 polling places across the nation with instructions to interview a sample of voters. At some precincts, they were to interview, say, every fourth voter; at others, every other voter. Polling is meant to be exact, but there are always unexpected circumstances that can make obtaining a good sample difficult—even impossible. At some polls, for example, researchers are required to stand far away from the voting stations, which allows voters to escape questioning. At others, if it is pouring down rain, voters are disinclined to talk.

  There were snafus throughout Election Day 2004, but a major problem with the exit polls turned out to be a human reaction: Republicans sometimes refused to answer the queries of polltakers. Bush supporters were particularly unwilling to answer questions posed by young interviewers with graduate degrees. (No, polltakers weren't wearing their degrees on a badge. Although there was no overt display of educational achievement, Bush voters avoided people with graduate degrees more than the run-of-the-mill grungy college student.) The more educated or the younger the researcher, the less likely Bush voters were to answer questions. The official report came to no conclusion about why Republicans skirted the polltakers, except to note that "in this election voters were less likely to complete questionnaires from younger interviewers."1 That factor alone caused an undercount of Bush votes. Kerry's supporters were inclined to talk with young, educated interviewers, and so Kerry's vote was systematically overstated.

  The pollsters also carried paperwork and notebooks bearing the logos of the major news outlets sponsoring the survey. Republican voters may have so mistrusted the major media that this association, too, led some to avoid the polltakers. It's not a far-fetched conclusion. According to a June 2004 survey by the Pew Research Center, "Republicans have become more distrustful of virtually all major media outlets over the past four years."2 A Fox News poll in late October 2006 found that Republicans were less likely to respond to exit poll questions asked on behalf of "television networks."3 The early, leaked exit poll results were wrong because they failed to account for this imbalance.

  The exit poll failure remains a source of debate, fascination, and, for some, paranoia.* But it appears that the poll was wrong not because of malfeasance or mechanical breakdown, but because of a malfunctioning culture. The poll was wrong because people avoided talking to those they thought had political leanings different from their own. In 2004, with the fractured, segmented, and segregated state of the nation thirty years in the making, Americans believed they could determine another's politics just by looking. And maybe they could.

  By the time of the Kerry versus Bush contest, Americans were allergic to difference of opinion and were blind to compromise. That's why churches were homogeneous units and markets segmented—so that worshipers and consumers could be sure to get exactly the sermon and the product they desired. Politicians had resisted the techniques of the megachurch and the target marketers through the 2000 election, perhaps because candidates are always slow to innovate, but also because politicians had traditionally understood their job to be changing the minds of those with opposing beliefs. But by 2004, those techniques had caught up with politics. This was the first Big Sort election.

  "We Don't See Them Anywhere"

  Although both parties claimed to be concentrating on turnout in the '04 campaign, the two sides rarely crossed paths. To Democrats, the Republican campaign was invisible. "We have people looking hard, and we don't see them anywhere," a leader of the liberal activist group America Coming Together (ACT) told reporter Jeff Mapes of the Portland Oregonian.4 Another ACT leader, in Ohio, said that a report lauding Republican organizing efforts was "a dream." The ACT leader told Matt Bai of the New York Times, "We've got the reality. They're wishing they had what we've got."5 It was as if the parties were campaigning in parallel worlds.

  Even after they'd lost, Democrats still didn't understand what had happened. The Bush campaign had been innovative in a number of ways, but it received the most attention for the sophisticated techniques it had used to identify voters.6 The cam
paign compiled 182 pieces of information (according to one account) about each voter—the kind of consumer data that has been used for a generation to target customers for candy bars or computer gizmos. Then the sharp-pencil boys at Bush headquarters cross-referenced this data with political polls. With the resulting calculations, Republican organizers could tell (with a reported 85 to 90 percent accuracy) whether a person—any person—was a Republican or a Democrat.7 "If you drive a Volvo and you do yoga, you are pretty much a Democrat," Bush campaign manager Ken Mehlman liked to say after the election. "If you drive a Lincoln or a BMW and you own a gun, you're voting for George Bush."*8

  Democrats were convinced they had lost because they hadn't had the latest bit of technical expertise—in this case, the somewhat gray-haired technique of micro-targeting. Since the 2004 election, Democrats have been busy conducting polls and doing their own analysis of lifestyle data so that their candidates can better target and then pluck out individuals inclined to vote Democratic. But Democrats missed a big part of what happened in 2004—because they never saw it.