It is a truth so universally acknowledged that it is a cliché: Education is essential to democracy. Thomas Jefferson viewed an informed citizenry as the indispensable bulwark against tyranny, writing that no civilized nation could be both ignorant and free. Franklin Roosevelt called education “the real safeguard of democracy,” and Ronald Reagan said education was a “mainspring” for both democracy and freedom. The idea that education is the fuel and the binding agent for democratic citizenship animates high school civics classes and the concept of public education itself.
But what if this logic isn’t airtight? What if, in a polarized country where political, cultural, and social divisions make conflict in the public discourse raw and visceral, people use the skills they get from education in ways that reinforce those divisions and harm democracy?
In fact, scholars have found that highly educated Americans are central to the political polarization that is fracturing our country. They are less likely than the average American to communicate with people who don’t share their views and more likely to view their political adversaries with hostility. Their views are often inaccurate and their political reasoning is often poor — precisely the characteristics that education is supposed to counteract.
What is going on? And what can we do about it?
Sorting ourselves into silos
Thanks to increased mobility and the revolution wrought by electronic communication, Americans have increasingly sorted ourselves into geographic and online communities of people who share our lifestyles, values, and political beliefs. The highly educated have done so with particular vigor, clustering tightly around just a few urban centers, like Boston and San Francisco. One result is that highly educated Americans are less likely than others to communicate with people who disagree with them.
In addition, scholars have long known that ideological thinking increases with education: The highly educated are more likely than others to hold consistently liberal or consistently conservative views across a wide range of issues. Even in the middle of the 20th century, when few Americans held an ideologically consistent set of political beliefs, the highly educated often did.
However, there were fewer of them then. Less than 10 percent of Americans held a college degree in the early 1960s. More than a third do today. So it is not surprising that there are more ideological Americans than there were a half a century ago. What is surprising is that the college educated have become more ideological over time. Between the 1980s and the early 2000s, ideological thinking barely rose among Americans without a college degree. But it soared among the college educated. Whereas just over a third of college-educated Americans held consistently liberal or conservative views on multiple issues in the 1980s, almost half did by 2004. The trend has not abated. A 2016 study by the Pew Research Center found that ideological polarization was driven heavily by Americans with bachelor’s degrees or higher.
Measuring political beliefs is one way to measure polarization, but it’s not the only way. Individuals can disagree about politics and still like each other personally. But even when we look at feelings rather than issues, the highly educated still stand out, and not in a good way. Social psychologists P.J. Henry and Jaime Napier found that liberals and conservatives were more likely to dislike each other at every step up the educational ladder. Liberals with a high school diploma were less prejudiced against conservatives than those with some college, who were, in turn, less prejudiced than those with a college degree. The pattern was precisely the same for conservatives: Education increased political animosity.
Those with more education are also more likely to misperceive their political opponents and to engage in what psychologists call “motivated reasoning” — the tendency to use our reason in defense of a strongly held opinion, discounting evidence or logic that may contradict it. High levels of education do not make a person more prone to motivated reasoning, but they give that person skills and tools — the ability to marshal evidence, construct an argument, parry counterarguments — to do it effectively.
Let’s start with misperceptions. In today’s fragmented media environment and Internet-enabled information echo chambers, Democrats and Republicans at every level of education seriously misperceive each other. Political scientists Douglas Ahler and Gaurav Sood found, for instance, that a majority of Democrats believed that 44 percent of Republicans made more than $250,000 per year; only 2 percent do. Similarly, a majority of Republicans thought that 43 percent of Democrats were union members; 10 percent are. The examples went on and on.
Then Ahler and Sood found something fascinating: Misperceptions were greatest among respondents with the most political knowledge. Perhaps “political knowledge” is not synonymous with “education level.” After all, people with low levels of education can be politically knowledgeable, and the highly educated might tune out politics. But in most cases, the more educated a person, the more likely she is to be politically knowledgeable and engaged.
Then there is political reasoning. Whereas Ahler and Sood looked at what people believe they know about their political opponents, Yale law professor Dan Kahan and colleagues examined how people use their cognitive and academic skills when they think about political issues. They started with the mundane finding that people with better math skills were more likely than others to solve a tricky math problem. But when they gave the same subjects a math problem with a political element — it involved the question of whether gun control laws affect urban crime rates — conservatives who were good at math repeatedly got the answer wrong if the right answer showed that gun laws helped reduce crime, while liberals with strong math skills got it wrong if the correct answer showed that gun laws do not reduce crime.
Kahan’s findings might be new, but the phenomenon they capture is not. Asked in 1988 whether inflation had gone up or down during Ronald Reagan’s presidency, the most politically informed Democrats were the most likely to answer, erroneously, that it had gone up. At the end of the 1990s, the most politically engaged Republicans were more likely than any other group to believe that the budget deficit had risen under Bill Clinton, when in fact it had fallen dramatically.
The power of human connection
Democrats and Republicans had their differences in the middle of the 20th century, but it was not uncommon for them to live in the same neighborhood or watch the same TV shows or attend the same religious services. These are all much less likely now.
Today, Democrats and Republicans, liberals and conservatives, disagree on many political issues, but the bigger problem is that they have less and less in common as humans and fellow citizens. Political conflict in 21st century America is not just about politics. It is about group membership and identity.
As humans, we are hardwired by evolution to attach ourselves to social groups. For millennia, ostracization from the group meant near-certain death. If an issue threatens our group identity or membership — if a debate about gun laws and crime rates isn’t just a policy debate but a proxy battle between my people and those people — then we engage in motivated reasoning and talk past each other, if we talk to each other at all.
Social science offers insights into how we can break or transcend hostility between groups, which is the underlying cause of motivated reasoning, so it is here that some solutions may be found. First, increase interaction between members of opposing political tribes. Even incidental engagement, such as having a friend of a friend in an opposing group, raises empathy. So does hearing a fair portrayal of the other side in the media or exposing people to differing viewpoints in a nonconfrontational setting.
And if the goal is opening minds, why not start early, the better to keep them open? What if, in our digitally connected age, teachers in classrooms in various parts of the country collaborated on lessons and projects designed to connect young people from all walks of American society? Rust Belt kids with city kids with farmers’ kids. They would work toward a shared goal — connection — and they would all be on equal footing in a setting where norms and rules demand civility and constructive interaction.
Continuing with the notion of starting early, what about national service as a powerful antidote to polarization? Mix young people from different social and cultural groups so that college-bound high schoolers from the golden suburbs of Boston and San Francisco and rural and non-college-bound young people serve alongside each other.
I’d even suggest compulsory military service. Perhaps it is politically unimaginable, but as a hypothetical model, consider its advantages to breaking down polarization: It has all the salutary characteristics of joint classroom activities and the added benefit that it imposes a shared identity on participants. Furthermore, it is intensive and long-lasting and involves shared sacrifice, which is among the most powerful of all human bonding agents.
The solutions are not easy, but they have a clear theme: more mixing, less sorting. Piercing the boundaries between opposing groups means humanizing both sides, which frees us to use the knowledge and skills we get from education not as artillery in an ideological battle of us versus them, but as map and light in the search for truth and mutual understanding.
Todd Washburn, a former assistant provost for international affairs at Harvard, teaches “The Polarization of American Politics” at the Harvard Extension School.