You don't know me, but you don't like me.
—HOMER JOY, "Streets of Bakersfield"
How can the polls be neck and neck when I don't know one Bush supporter?
—ARTHUR MILLER
IN THE SPRING before the 2004 election, I heard from LaHonda Jo Morgan. Jo Morgan lived in Wauconda, Washington, a one-building town (combination grocery, cafe, and post office) about 150 miles northwest of Spokane. She was convinced that Wauconda remained on the map "simply because mapmakers don't like to leave a lot of empty space on their products." Jo Morgan was writing about segregation—political segregation. She had seen an article I had written about the tendency of places to become politically like-minded, either increasingly Republican or Democratic. She noticed that the article came from Austin, her hometown. So she recounted that through fifty years of marriage, she had lived in a number of places across the United States and elsewhere in the world. And then she described a change she had noticed taking place in Wauconda:
This is a predominantly conservative area with most residents tied to ranching, mining and apple orchards. A few years back I began to feel somewhat disconnected in my church community, but I chalked that up to the struggle between pre—and post—Vatican II concerns. Since the strife of the 2000 election, I became increasingly uncomfortable in conversations in a variety of situations. Perhaps I had more flexible views because of having been exposed to different cultures. In fact, I felt like a second-class citizen, not entitled to have opinions. I even wondered if I [was] becoming paranoid since being widowed.
Of course, now I understand. Increasing divisiveness arising from political partisanship is giving rise to the same sort of treatment I observed growing up in racially segregated Texas, only now it is directed at people who think differently from the majority population of an area. Sort of scary, isn't it?
Jo Morgan was right about Wauconda changing. In 1976, Okanogan County in Washington had split fifty-fifty in the nearly fifty-fifty race between Jimmy Carter and Gerald Ford. That made sense. Americans in 1976 were more likely to live close to somebody who voted differently from themselves than at any time since the end of World War II. And then, like the rest of the country, Jo Morgan's community changed. Okanogan County went for Clinton in 1992 and then veered Republican, strongly so, in the next three elections. In 2000, 68 percent of Okanogan County voted for George W. Bush. No wonder Jo Morgan felt lonely.
But "scary"? I kept a file of the more outrageous examples of political anger in 2004. They ranged from the psychotic to the merely sad. There was the Sarasota, Florida, man who swerved his Cadillac toward Representative Katherine Harris as she campaigned on a street corner. (Harris had been the Republican secretary of state in Florida during the presidential vote recount in 2000.) "I was exercising my political expression," Barry Seltzer told police.1 The South Florida Sun-Sentinel reported just a week before the election that "when an 18-year-old couldn't convince his girlfriend that George W. Bush was the right choice for president, he became enraged, put a screwdriver to her throat and threatened to kill her." The man told her that if she didn't change her vote, she wouldn't "live to see the next election."2 Two old friends arguing about the war in Iraq at an Eastern Kentucky flea market both pulled their guns when they got tired of talking. Douglas Moore, age sixty-five, killed Harold Wayne Smith because, a witness said, "Doug was just quicker."3
The destruction of campaign yard signs and the vandalism of campaign headquarters was epidemic in 2004. The Lafayette, Louisiana, Democratic Party headquarters was struck twice; in the second assault, miscreants wrote "4 + GWB" on the building's front windows in a mixture of motor oil and ashes collected from burned John Kerry signs.4 The most pathetic display of partisan havoc started at the Owens Crossroads United Methodist Church near Huntsville, Alabama. The youth minister at the church sent children on a "scavenger hunt" shortly before the election. On the list of items to be retrieved were John Kerry campaign signs. Once the kids toted the placards back to the church, the minister piled them in the parking lot and set the signs on fire.5 The scavengers did the best they could, but in Republican Huntsville they found only eight signs, barely enough for kindling. Had the same hunt taken place in, say, Seattle, the kids could have rounded up enough fuel to signal the space shuttle.
Living as a political minority is often uncomfortable and at times frightening. In 2000, more than eight out of ten voters in the Texas Hill Country's Gillespie County cast ballots for Bush. Two years later, Democrats prepared a float for the Fourth of July parade in the county seat of Fredericksburg. "We got it all decorated," county party chairman George Keller recalled, "but nobody wanted to ride." Nobody wanted to risk the stigma of being identified as a Democrat in an overwhelmingly Republican area. "Thank goodness we got rained out," Keller said of the orphaned float.
Gerald Daugherty used to live in the hip and shady section of Austin known as Clarksville. When he became active in a campaign against a proposal to build a light rail system in town, Daugherty put NO LIGHT RAIL bumper stickers on his car and on his wife's Mercedes. That apparently didn't go over too well in Democratic and pro-rail Clarksville. Somebody "keyed" the Mercedes at the local grocery and for good measure punched out the car's turn signal lights. Was Daugherty sure the damage had been politically motivated? Not really. But then one morning he found his car coated with eggs. "There must have been two dozen eggs all over my car," he remembered. "Splattered. And then deliberately rubbed on the 'No Rail' bumper stickers. You knew where that was coming from." So Daugherty sold his house in a precinct that gave George W. Bush only 20 percent of the vote against Al Gore. He bought a place in a precinct where two out of three people voted Republican in the same election. Two years later, Daugherty became the only Republican elected to the county governing body. His move out of Clarksville, he admits, was a political exodus. He left a place where he "stuck out like a sore thumb" and moved to a neighborhood that was more ideologically congenial. He reasoned, "You really do recognize when you aren't in step with the community you live in."6
People don't check voting records before deciding where to live. Why would anyone bother? In a time of political segregation, it's simple enough to tell a place's politics just by looking. Before the 2006 midterm elections, marketing firms held focus groups and fielded polls, scouring the countryside to find the giveaway to a person's political inclination. Using the most sophisticated techniques of market profiling, these firms compiled a rather unsurprising list of attributes.
Democrats want to live by their own rules. They hang out with friends at parks or other public places. They think that religion and politics shouldn't mix. Democrats watch Sunday morning news shows and late-night television. They listen to morning radio, read weekly newsmagazines, watch network television, read music and lifestyle publications, and are inclined to belong to a DVD rental service. Democrats are more likely than Republicans to own cats.
Republicans go to church. They spend more time with family, get their news from Fox News or the radio, and own guns. Republicans read sports and home magazines, attend Bible study, frequently visit relatives, and talk about politics with people at church. They believe that people should take more responsibility for their lives, and they think that overwhelming force is the best way to defeat terrorists. Republicans are more likely than Democrats to own dogs.
None of this is particularly shocking. We've all learned by now that Republicans watch Fox News and Democrats are less likely to attend church. Okay, the DVD rental clue is a surprise, and Democrats in my part of town own plenty of dogs, but basically we all know these differences. What is new is that some of us appear to be acting on this knowledge. An Episcopal priest told me he had moved from the reliably Republican Louisville, Kentucky, suburbs to an older city neighborhood so that he could be within walking distance of produce stands, restaurants, and coffee shops—and to be among other Democrats. A journalism professor at the University of North Carolina told me that when he retired, he moved to a more urban part of Chapel Hill to escape Republican neighbors. A new resident of a Dallas exurb told a New York Times reporter that she stayed away from liberal Austin when considering a move from Wisconsin, choosing the Dallas suburb of Frisco instead. "Politically, I feel a lot more at home here," she explained.7 People don't need to check voting records to know the political flavor of a community. They can smell it.
To explain how people choose which political party to join, Donald Green, a Yale political scientist, described two social events. Imagine that you are walking down a hall, Green said. Through one door is a cocktail party filled with Democrats. Through another is a party of Republicans. You look in at both, and then you ask yourself some questions: "Which one is filled with people that you most closely identify with? Not necessarily the people who would agree were you to talk policy with them. Which group most closely reflects your own sense of group self-conception? Which ones would you like to have your sons and daughters marry?"8 You don't compare party platforms. You size up the groups, and you get a vibe. And then you pick a door and join a party. Party attachments are uniquely strong in the United States. People rarely change their affiliation once they decide they are Democrats or Republicans. No wonder. Parties represent ways of life. How do you know which party to join? Well, Green says, it feels right. The party is filled with your kind of people.*
How do you know which neighborhood to live in? The same way: because it feels right. It looks like the kind of place with boys and girls you'd like your children to marry. You just know when a place is filled with your kind. That's where you mentally draw a little smiley face of approval, just as my wife did as we moved from Kentucky to Austin in 1999.
Texas voted in 2005 on whether to make marriage between people of the same sex unconstitutional. Statewide, the anti—gay marriage amendment passed with ease. More than seven out of ten Texans voted for it. In my section of South Austin, however, the precincts voted more than nine to one against the measure. The difference between my neighborhood and Texas as a whole amounted to more than 60 percentage points. It's not coincidence that in our narrow slice of Austin, a metropolitan area of more than 1.4 million people filling five counties, the liberal writer Molly Ivins lived just five blocks from the liberal writer Jim Hightower—and at one time we lived five blocks from both of them.
During the same years that Americans were slowly sorting themselves into more ideologically homogeneous communities, elected officials polarized nationally. To measure partisan polarization among members of Congress, political scientists Howard Rosenthal, Nolan McCarty, and Keith Poole track votes of individual members, who are then placed on an ideological scale from liberal to conservative. In the 1970s, the scatter plot of the 435 members of the House of Representatives was decidedly mixed. Democrats tended toward the left and Republicans drifted right, but there was a lot of mingling. Members from the two parties overlapped on many issues. When the scholars fast-forward through the 1970s, 1980s, and 1990s, however, the votes of the 435 representatives begin to split left and right and then coalesce. The scatter plot forms two swarms on either side of the graph's moderate middle. By 2002, Democratic members of Congress were buzzing together on the left, quite apart from a tight hive of Republicans on the right.9 In the mid-1970s, moderates filled 37 percent of the seats in the House of Representatives. By 2005, only 8 percent of the House could be found in the moderate middle.10
Members from the two parties used to mingle, trade votes, and swap confidences and allegiances. (In 1965, half the Republicans in the Senate voted for President Lyndon Johnson's Medicare bill.) That kind of congressional compromise and cross-pollination is now rare. More common is discord. The Washington Post's Dana Milbank and David Broder reported in early 2004 that "partisans on both sides say the tone of political discourse is as bad as ever—if not worse."11 Former Oklahoma congressman Mickey Edwards said that on a visit to Washington, D.C., he stopped at the barbershop in the Rayburn House Office Building. "And the barber told me, he said, 'It's so different, it's so different. People don't like each other; they don't talk to each other,'" Edwards recalled. "Now, when the barber in the Rayburn Building sees this, it's very, very real."
Some very smart people have questioned whether the American public is polarized to begin with, whether there really are vast and defining differences among Americans. Some argued that, viewed over the centuries, the increase in geographic segregation since the mid-1970s has been minor, a subtle fluctuation—and compared to the Civil War period, that is certainly the case.12 At the same time, Stanford University political scientist Morris Fiorina proposed in the mid-2000s that Americans were not particularly polarized in their politics: "Americans are closely divided, but we are not deeply divided, and we are closely divided because many of us are ambivalent and uncertain, and consequently reluctant to make firm commitments to parties, politicians, or policies. We divide evenly in elections or sit them out entirely because we instinctively seek the center while the parties and candidates hang out on the extremes."13
Fiorina argued that the fractious politics Americans were experiencing were wholly a result of polarized political leadership and extreme issue activists. Elected officials might be polarized, the professor wrote, but people were not. Journalists miss what's really happening in the country, he contended, because "few of the journalists who cover national politics spend much of their time hanging out at big box stores, supermarket chains, or auto parts stores talking to normal people ... When they do leave the politicized salons of Washington, New York and Los Angeles, they do so mainly to cover important political events which are largely attended by members of the political class ... The political class that journalists talk to and observe is polarized, but the people who comprise it are not typical."14
Fiorina announced that his book was needed to debunk what he described as the "new consensus" that Americans were deeply divided.15 In the meantime, however, Fiorina's view became the new truism. Jonathan Rauch wrote in the Atlantic that when scholars went to look for the red and blue division, "they couldn't find it."16 Joe Klein in Time blamed the "Anger-Industrial Complex" for ginning up a division that didn't exist in real life.17 Columnist Robert Kuttner scolded a "lazy press corps" for overplaying the red and blue division when "the reality is quite different."18 Fiorina's argument was even picked up in 2005 by the yellow pages of conventional wisdom, Reader's Digest.19
The abortion question was a favorite of those who contended that the middle was wide and the fringe narrow. Both Klein and Kuttner used abortion as such an example. Likewise, E. J. Dionne wrote in the Atlantic that "60 to 70 percent of us fall at some middle point" on most issues. Dionne wrote that only 37 percent of the people interviewed in a 2004 Election Day exit poll said that abortion should be "always" legal or "always" illegal.20 Indeed, if we accepted the notion that a person who believed that abortion should be legal for victims of rape but illegal for victims of incest qualified as a moderate, then we would find nearly two-thirds of the population in the "middle" on this issue.* But a late 2005 poll from Cook/RT Strategies posed the abortion question in a slightly different way. Instead of asking if abortions should "always" be illegal or legal, Cook asked if people were "strongly pro-life" or "strongly pro-choice." In response to that question, the "middle"—those who were only "somewhat" committed to a position—shriveled to 25 percent. Those who felt "strongly" about this issue totaled 70 percent of the population, split just about evenly between the two poles.
This kind of ideological allegiance has grown over time, as successful politicians know. Bill Bellamy has been an Oregon state representative and was a Jefferson County commissioner in the small town of Madras when we talked in 2005. Madras is on the dusty side of Mount Hood, where the Cascades flatten into fields that circle around irrigation rigs. In Bellamy's real estate office parking lot, a cowboy pulled in with a blue heeler barking and twirling on the toolbox just behind the back window of his pickup. In Portland, trailer hitches are bright chrome and virginal. Here a trailer hitch ball has seen some action. "In 1976, when I first ran and they would ask me my position on abortion, out of one hundred people, it was really important to only ten of them," Bellamy said. "By 1988, when I ran for the [state] senate, out of that one hundred people, for probably sixty of them it was very important."
Emory University political scientist Alan Abramowitz argued that Morris Fiorina "systematically understates the significance" of divisions over abortion, gay marriage, and other cultural markers. Abramowitz collected national polling data to show that differences among Americans were deep and growing deeper, increasing between 1972 and 2004, just the period when the country was segregating geographically. People who identified themselves as Democrats thought differently about issues than those who considered themselves Republicans. And those differences—on issues such as abortion, living standards, and health insurance—were growing larger. People's evaluation of George W. Bush in 2004 were more divided along party lines than at any time since the National Election Studies started asking questions about presidential approval in 1972.21
The sharp divisions among Americans appeared again in the results of the 2006 midterm elections. Voters split most dramatically on the war in Iraq: 85 percent of Democratic House voters said the invasion had been a mistake, compared to only 18 percent of Republican voters. But those divisions extended to most other issues. Sixty-nine percent of Democrats were strongly pro-choice, compared to 21 percent of Republicans. Only 16 percent of Democrats supported a constitutional ban on gay marriage, a position favored by 80 percent of Republicans. Nine out of ten Democrats, but less than three out of ten Republicans, felt in November 2006 that government should take some action to reduce global warming. Plotted on a graph of how they felt about the issues of the day in November 2006, American voters didn't form a nice, high-peaked bell, with most people clustered toward the happy ideological center. Instead, there was a deep, sharp V, with voters pushed hard left and right. How many voters wavered between the two parties as true independents in 2006? About 10 percent.22
Typically, two reasons are given to explain our polarized politics. The most popular is gerrymandering: through years of redistricting, politicians have packed their districts to produce overwhelming majorities, creating such partisan uniformity that there is no reason or call to compromise. We elect extremists, especially for Congress, the argument goes, because politicians have drawn their districts to be extreme. And when legislators come out of these partisan districts—districts where the two parties don't compete—they push the entire country into a choice between the far left and the far right. Voters polarize not because everyday Republicans are all that different from everyday Democrats, but because political leaders are ideologues.
The second explanation—one favored by Democrats—holds that conservative activists built an interlocking structure of propaganda and money that moved the Republican Party, and the nation, to the right. The aim of the New Right after Goldwater's defeat in 1964 was to exacerbate divisions in the country and then exploit them.
Gerrymandering is a convenient—and popular*—explanation because it does conform to an objective reality. Every ten years, legislators do, in fact, redraw districts, and an ever-increasing number of those districts are becoming more ideologically lopsided. Gerrymandering also has science behind it. Legislators use "powerful computers," which make the process nefariously exact. In addition, the gerrymandering thesis has "bad guys"—better than bad guys, really; it has politicians. Elected officials, not moderate-loving voters, have caused the problem and deserve the blame.
It's certainly true that congressional districts have grown largely uncontested. Even in the middle of an unpopular war, 90 percent of incumbent members of Congress were reelected in 2006, and although the number of competitive races increased, only 66 out of 435 House races were at all close.23 And it's true that House districts, on average, have grown overwhelmingly either Democratic or Republican since the 1970s. By 2004, nearly half the members of Congress came from districts that had unassailable majorities. The question, however, is whether the increase in ideologically pure districts was caused by redistricting.
There are several arguments against the gerrymandering thesis. The first is that political parties aren't in the business of building supermajorities for incumbents. Parties exist to maximize their number of representatives. This imperative causes parties to spread votes around, creating more districts with, say, 10- to 15-point majorities and fewer with lopsided constituencies. Studies of redistricting have found that, indeed, "partisan redistricting often has the effect of reducing the safety of incumbents."*24 The results of the 2006 midterm elections provided some evidence that Republicans lost races not because they had been making seats safer, but because they had spread their majorities a wee bit thin. In Pennsylvania, Democrats targeted districts where Republican margins had been shaved through redistricting and narrowly picked up three seats. "If Republicans had been a little less aggressive (in redistricting), they could have won several of those seats," Nathaniel Persily, a redistricting specialist at the University of Pennsylvania told the Wall Street Journal. "If they gave the Democrats one more seat, they could have shored up by several percentage points the other seats."25
It doesn't appear that redistricting caused much, if any, of the increase in homogeneous districts. After all, if gerrymandering created landslide districts, you'd expect to see an increase in noncompetitive districts immediately after redistricting. Legislatures would draw new districts after the census and, bing-bang, there would be fewer competitive districts. That didn't happen. After each of the last three redistricting cycles (1980, 1990, and 2000), there were no immediate jumps in lopsided districts. When Emory University political scientist Alan Abramowitz examined the effects of redistricting in 2000, he found that the number of supersafe House seats (those with presidential vote margins of more than 20 percent) had increased by two, from 201 to 203. That's hardly a sign of much horseplay. Abramowitz found similar small effects after redistricting in the 1980s and 1990s. (After redistricting in 1980, in fact, the number of noncompetitive districts slightly decreased.) If legislative gerrymandering had caused the lopsided House, its effects certainly had been subtle, or perhaps one should say "prescient." For the districts hadn't grown more partisan at the time of redistricting, Abramowitz found. They had grown more partisan later, in the years between redistricting, when the districts' boundaries remained unchanged. From the first post-redistricting election in 1992 until 2000, the number of ideologically lopsided districts jumped from 156 to 201, but not a single district changed shape in those years.*26
Vanderbilt University's Bruce Oppenheimer looked at this phenomenon in another way. There are seven states with only one member of Congress. Five are red (Alaska, Montana, the Dakotas, and Wyoming), and two are blue (Delaware and Vermont). But none have had their legislative boundaries gerrymandered. Oppenheimer cobbled these seven districts into a single, hypothetical "state." He compared this seven-district "state-of-states" with twenty-six actual states with a similar number of districts in three very close presidential elections: 1960, 1976, and 2000. Oppenheimer checked to see which had become more lopsided, the made-up state-of-states with the static borders or the real states where politicians and their infernal computers had gerrymandered to their hearts' content.
Between 1960 and 2000, no real-life state saw partisan vote margins in its congressional districts increase more than in Oppenheimer's hypothetical state-of-states. Manipulative politicians in the twenty-six states had four chances to make their congressional districts less competitive, but even so the districts didn't match the lopsidedness that appeared naturally in the state-of-states.* "These data raise doubts about the ability of redistricting schemes to explain the decline in the underlying party competitiveness of congressional districts," Oppenheimer wrote.†27
If not gerrymandering, then how about conspiracy? Democrats have argued that the elections of 2000 and 2004—and the concurrent polarization of the nation's politics—were the culmination of a forty-year effort by Republicans. The story goes like this: In the wake of the Barry Goldwater defeat in 1964, Republicans devised a grand scheme. They built a tightly wound, highly coordinated movement from the top down. Corporations and foundations paid for think tanks and advocacy groups, which supplied the movement with ideas and leaders. The right created its own media—talk radio, Christian television networks, and conservative-minded college newspapers—in this centrally managed, ma-chinelike plot to split the country ideologically and then establish a permanent majority. The result of this multigenerational effort lay in the Republicans' congressional victory in 1994 and the election of George W. Bush in 2000 and 2004.
Certainly, the conservatives wanted to take over. Winning, after all, is one goal in politics. But a conspiracy? One piece of evidence used to support the existence of this far-sighted plan is a 1971 memo written by Lewis Powell, the soon-to-be-appointed Supreme Court justice. Powell, writing to a friend with the U.S. Chamber of Commerce, warned of an "attack" on the "free enterprise system." In the early years of the George W. Bush administration, liberals unearthed this obscure manuscript and gave it nearly mythic significance. Former Democratic senator Bill Bradley described Powell's note (in what surely is an oxymoron) as a "landmark memo." The right had used the memo, Bradley wrote, as a "blueprint" to construct a "pyramid" of foundations, think tanks, and advocacy groups, all designed to support an interchangeable Republican leader.28 Pick—or mix!—your metaphor of all-embracing power. Harper's Magazine editor Lewis Lapham described the "Republican propaganda mill" as "tentacles of rage." He transformed Powell's memo into a "manifesto" that held for the political right the "hope of their salvation." According to Lapham, Powell's "heavy word of warning fell upon the legions of reaction with the force of Holy Scripture."*29 Skipping several generations, the bloggers Jerome Armstrong and Markos Moulitsas Zuniga wrote in 2006 that the Powell memo "eventually helped fuel nascent efforts to create the most sophisticated, well-funded political propaganda machine in world history," Joseph Goebbels notwithstanding.30
The belief on the left is that the machine (or mill or pyramid or giant squid) built of foundations, radio programs, and organizations powered the Republican comeback. The right-wing mechanism paid for scholars' sharp pencils and book contracts. Young leaders were fledged through summer camps, internships, and jobs with Republican congressional representatives until they could become self-supporting members of the movement. The right established a shadow society that built, grew, and eventually took over in the name of religion and free enterprise. And the entire operation was funded by the businesses that had suffered at the hands of Democratic government.
Mark Schmitt, the former director of policy at the liberal Open Society Institute, called this phenomenon the "legend of the Powell memo."31 He found few historians of the conservative movement who even mention the memo. For example, John Micklethwait and Adrian Wooldridge's chronicle The Right Nation gives the Powell memo exactly three sentences.*32 Moreover, Schmitt wrote, Powell was "far out of touch" with what would become the New Right.33 The memo was given an iconic status by liberals searching for some explanation of their minority standing in national politics. (Conspiracy was a more appealing theory than a simple lack of popular support.) Best of all, this explanation was duplicable: the left could write its own Powell memo and create its own matrix of foundations, think tanks, and leadership programs. (James Piereson, executive director of the conservative, and now defunct, Olin Foundation, observed that the left had a "near-obsessive interest in conservative philanthropies."†34) Schmitt contended that the "reality of the right is that there was no plan, just a lot of people writing their own memos and starting their own organizations—some succeeding, some failing, false starts, mergers, lots of money well spent, and lots of money wasted."35
There is some truth to the conspiracy stories. Republicans schemed and conservatives talked of creating a "shadow society"; they set up alternative foundations, research groups, and media outlets. Of course, Democrats schemed, too, and the left had its own support in the foundation world. But conservatives better understood the changes taking place in the country, and that is why, for a time, Republicans were more successful politically. Republicans didn't create a movement. They recognized the cultural shifts taking place across the country—the Big Sort—and then channeled what was happening into politics, to their advantage.
What both gerrymandering and the forty-year conservative conspiracy arguments miss is that politics is a two-way street. It flows both from the top down and from the bottom up. Most explanations for our current partisanship—gerrymandering and conspiracy are two good examples—are top-down only. They assume that public opinion follows the lead of presidents, politicians, and Capitol Hill journalists. In this worldview, elites (be they elected officials, media barons, or a cabal of well-funded Republicans) use the power of money or position to push society in a particular direction. Voters are largely powerless in this process. They just choose one of the alternatives that legislative manipulation, media bias, and party propaganda provide.
But politics is bottom-up as well. Society changes and politicians follow. The Big Sort is the story of real differences in the way people think, in what they value, in how they worship, and finally in where they live. The divisions in Congress aren't simply the consequence of manipulations by left-wing interest groups or the outcome of plots hatched in a bunker deep under the Heritage Foundation. The divisions are the reflection of how—and where—people have come to reside.
A less conspiratorial explanation for why national politics has grown more partisan over the past thirty years can be found in the studies of congressional redistricting. Alan Abramowitz and Bruce Oppenheimer looked at the evidence of increasing geographic polarization we first presented in the Austin American-Statesman in 2002 and 2004, and they came to the same conclusion: people have been sorting. Abramowitz: "Americans are increasingly living in communities and neighborhoods whose residents share their values and they are increasingly voting for candidates who reflect those values."36 Oppenheimer: "A final theory that I offer to explain the decline in partisan competitiveness at the congressional district level rests on the increased mobility of Americans and the corresponding growth in the freedom to select where they will reside."37
The overwhelming attention given to political celebrity—and political conspiracy—in our time has obscured the politics of place. If people simply respond to the faults, successes, and foibles of political elites, then it really doesn't matter that people are taking up residence in increasingly homogeneous neighborhoods. But politically like-minded regions practice a different kind of politics than do places with a greater mix of allegiances. Our politics are affected by our neighbors. Following is one example.
In the early 1960s, political scientist John Fenton wondered why working-class voters in Ohio supported Republicans, a political act that was against their economic interests. Fenton explained this phenomenon by looking at the shape of the state's neighborhoods. Upper-class voters lived in tightly knit, geographically compact communities. Physical proximity made it easier for them to maintain political cohesion, to move and vote in an ideological herd. In Ohio's large number of midsize cities, however, there was no corresponding critical mass of workers. Working-class voters were dispersed. "In Ohio you had a fairly even distribution of these working-class voters across the state," explained the University of Maryland's James Gimpel. "And because they lived among farmers and clerks and ditch diggers, they were not as inclined to vote so monolithically."38 In nearby Michigan, Gimpel said, working-class voters lived close to one another, and their geographic proximity powered their ideological and political intensity. In Ohio, however, workers were spread out, and the effect of this diffusion, Fenton wrote more than forty years ago, was "profound ... The postman did not talk the same language as his accountant neighbor, and the accountant was in a different world from the skilled workman at Timken Roller Bearing who lived across the street. Thus, conversation between them usually took the form of monosyllabic grunts about the weather ... The disunity of unions and the Democratic party in Ohio was a faithful reflection of the social disorganization of their members."39
Thomas Frank recently bemoaned the failure of Great Plains residents to vote in their economic interests and asked, "What's the Matter with Kansas?"40 Frank's answer was that manipulative Republicans who offered intelligent design rather than a living wage had duped working-class voters in his home state. In addition, thin-blooded liberals who had gotten above their populist raisings had abandoned Democratic principles. When John Fenton asked a similar question more than forty years ago—What's the matter with Ohio?—he arrived at an explanation that didn't depend on either gullibility or duplicity. Fenton found that the way people lived—and the communities they lived in—shaped their political lives.
Unlike Ohio of the early 1960s, political divisions today are as much a result of values and lifestyle as they are of income and occupation. And with those divisions has come a pervasive and growing separation. Americans segregate themselves into their own political worlds, blocking out discordant voices and surrounding themselves with reassuring news and companions. For example, it's not surprising that supporters are more likely to watch a president's speech, whereas opponents tend to change the channel. But the spread between viewers and channel changers has been expanding. The Gallup organization found that during the Clinton administration, the television audience for the yearly State of the Union address was on average 9 percentage points more Democratic than Republican. Under George W. Bush, however, the audience from 2001 to 2005 averaged 21 percentage points more Republican than Democratic. In 1995, the viewing audience for Clinton's State of the Union address was evenly split between Democrats, Republicans, and independents. By the time Bush addressed the nation in 2005, 52 percent of the audience was Republican, 25 percent was Democratic, and 22 percent was independent.41 More and more, Americans watch and read the news that fits their political proclivities and ignore the other side. And should the choice between Fox News (on the right) and National Public Radio (on the left) seem impersonal, discriminating liberals can bob about the Caribbean on a cruise with writers from the Nation, while conservatives can board a different ship for a trip hosted by William Kristol and Fred Barnes of the Weekly Standard.
Is the United States polarized? Maybe that's the wrong term. What's happening runs deeper than quantifiable differences in a grocery list of values. Despite the undeniable sameness of places across America—is a PetSmart in a Democratic county different from a PetSmart in a Republican county?—communities vary widely in how residents think, look, and live. And many of those differences are increasing. There are even increasing differences in the way we speak. * Over the past thirty years, communities have been busy creating new and different societies, almost in the way isolated islands foster distinct forms of life, but without a plan or an understanding of the consequences.
The first half of the twentieth century was an experiment in economic specialization, as craft production gave way to assembly lines; cabinetmakers became lathe operators or door assemblers. The second half of the century brought social specialization, the displacement of mass culture by media, organizations, and associations that were both more segmented and more homogeneous. We now worship in churches among like-minded parishioners, or we change churches, maybe even denominations, to find such persons. We join volunteer groups with like-minded companions. We read and watch news that confirms our existing opinions. Politics, markets, economies, culture, and religion have all moved along the same trajectory, from fragmentation in the nineteenth century to conglomeration in the twentieth century to segmentation today. Just as counties have grown more distant from one another politically, regional economies are also separating—some booming and vibrant, others weak and dissipating. Mainline religious denominations gained parishioners through the first half of the twentieth century, the age of mass markets, but lost members beginning in the mid-1960s to independent churches designed for homogeneous communities. Media, advertising, city economies—they've all segmented, specialized, and segregated.
In the mid-1970s, when counties were becoming politically integrated, most other measures of public life showed low levels of political separatism. The differences that we take for granted today were muted. For instance, how often a person went to church didn't mark him or her as a Democrat or a Republican. Women voted slightly more Republican than Democratic. The Democratic vote was slightly more rural than the Republican. Less than half the population saw important differences between the parties. The proportion of people describing themselves as true independent voters reached post-World War II highs. Fewer than half of Republicans described themselves as conservative. People often split their vote between Republicans and Democrats. Votes in the U.S. Congress were more bipartisan than at any time since World War II.
Beginning in the mid-1970s, the movement toward political mixing slammed to a halt and headed in the opposite direction. Women became more allied with the Democratic Party.* Rural areas and frequent churchgoers became more Republican. The percentage of independents and ticket splitters declined. People grew more ideological. Democrats were increasingly liberal; Republicans were increasingly conservative. Voters saw greater differences between the parties. Congressional Quarterly reported that 2005 was the most partisan year in Congress in the half century that the venerable publication had been keeping count.
The tale we've been told and have come to tell ourselves is that society cracked in 1968 as a result of protests, assassinations, and the melee in the streets of Chicago. Informed by the Big Sort, we can now see 1968 more as a consequence of gradual change than as a cause of the changes that followed. Old political, social, religious, and cultural relationships had begun to crumble years earlier. American culture had slowly shifted as people simultaneously grew richer and lost faith in the old institutions that had helped create that wealth: the Democratic Party, the Elks, the daily newspaper, the federal government, the institution of marriage, the Presbyterian Church. Party membership, newspaper circulation, trust in government, and the number of people in the pews of mainline churches all declined at the same time.
The old systems of order—around land, family, class, tradition, and religious denomination—gave way. They were replaced over the next thirty years with a new order based on individual choice. Today we seek our own kind in like-minded churches, like-minded neighborhoods, and like-minded sources of news and entertainment. As we will see later in this book, like-minded, homogeneous groups squelch dissent, grow more extreme in their thinking, and ignore evidence that their positions are wrong. As a result, we now live in a giant feedback loop, hearing our own thoughts about what's right and wrong bounced back to us by the television shows we watch, the newspapers and books we read, the blogs we visit online, the sermons we hear, and the neighborhoods we live in.
Politicians and parties have exploited this social evolution, and in doing so, they have exacerbated partisanship and division. Elites have always been more partisan, more extreme, and more ideological than regular voters. But today moderates on all sides are rebuffed, and those who seek consensus or compromise are squeezed out. Paul Maslin, Democratic presidential hopeful Howard Dean's pollster in 2004, explained it this way:
If I had to say one true statement about the entire process you are describing, I think that at the national or state level, it's making life increasingly difficult for people who are trying to thread the needle, to find the swing voter. In a way Karl Rove and Howard Dean and [Dean campaign manager] Joe Trippi were all right here. It's probably one of the things that's driving our politics into a more polarized situation. While the swing vote and the classic vote in the middle still matter, you are much more willing to say now that you ignore at your peril your own base. Because as everything spreads apart, the base becomes more important because they are demographically more together. You don't have a whole bunch of 51–49 communities out there. You have more and more 60–40, 65–35, 70–30 places. Well, you better damn well be sure you maximize your 70–30 votes, whether it's inner-city African Americans or liberal, educated Democrats or whether it's suburban, conservative Republicans or small-town, main-street, or Evangelical Republicans. We have to maximize our base, and they have to maximize their base. Ergo, polarization.
The country may be more diverse than ever coast to coast. But look around: our own streets are filled with people who live alike, think alike, and vote alike. This social transformation didn't happen by accident. We have built a country where everyone can choose the neighborhood (and church and news shows) most compatible with his or her lifestyle and beliefs. And we are living with the consequences of this segregation by way of life, pockets of like-minded citizens that have become so ideologically inbred that we don't know, can't understand, and can barely conceive of "those people" who live just a few miles away.