We need political parties. But their rabid partisanship could destroy American democracy.

It’s December 2020, and President Donald Trump has still refused to concede that he lost the tumultuous presidential election.

A month earlier, Sen. Kamala Harris narrowly defeated Trump. Even though the incumbent held onto the Rust Belt states he had gained in 2016, record-high minority turnout gave Harris narrow wins in Arizona, Florida, Georgia, and North Carolina. Democrats also narrowly took back both the House and the Senate, with razor-thin pickups in North Carolina and Georgia. The urban/rural partisan divide continued to widen. And for the first time in American history, a majority of one party’s voters (the Democrats) were nonwhite.

The Trump campaign demanded recounts in the four states Harris narrowly won, blaming “illegal voting” and Chinese hacking. Recounts confirmed the original totals. Trump dismissed the recounts. With the balance of power in both the Senate and House potentially implicated as well, House Speaker Kevin McCarthy and Senate Majority Leader John Thune quietly backed the Trump administration.

Sen. Harris, daughter of Indian and Jamaican immigrants, gained momentum early with a strong pro-immigrant, pro-civil rights campaign that electrified the Democratic base. A former prosecutor, she promised to intervene with an even hand in the increasingly contentious battles over policing and to prosecute Trump once and for all for collaborating with Russia. Though special prosecutor Robert Mueller indicted and convicted a few Trump campaign affiliates in early 2018, Trump kept up the drumbeat about a “witch hunt,” containing the damage.

Trump ran a brutal campaign, accusing Harris of harboring a secret plan to implement Sharia Law in the US, and promising to open our borders and grant amnesty to millions of “illegals.” He repeatedly said that if Harris stole the presidency (he never acknowledged she could win legitimately), America may as well surrender to radical Islam.

After the recounts, Trump held increasingly large rallies in Arizona, Florida, Georgia, and North Carolina, saying a Democratic victory would mark the “end of America.” He openly encouraged violence if Harris failed to concede. Nobody knew what would happen if Trump had to be forcibly removed from the White House.

Could something like this really happen here? The prospect may be remote. But it feels a lot more likely now than at any other time in the past 150 years.

American democracy, like all political systems, rests on norms. Rules can only save us if they are agreed upon and respected.

For a long time, we collectively assumed that respect for elections and peaceful transfer of power were so sacred to the stability of our political system that nobody would ever challenge them. But in 2016, Trump, a major party candidate, promised to keep the country in suspense over whether he would concede if he lost. (Since he won, we’ll never know what would have happened had he pressed the issue.)

For a long time, we assumed that while we might have strong political disagreements with each other, there were certain neutral arbiters in society whose authority we would all respect and abide by. There were enough generally agreed-upon facts that our disputes wouldn’t threaten the foundations of our political system.

But for years now, we’ve been retreating into our separate tribal epistemologies, each with their own increasingly incompatible set of facts and first premises. We’re entering a politics where the perceived stakes are higher and higher (“the fate of our nation lies in the balance”) that they justify increasingly extreme means. When it is a war of good versus evil, “norms” and “fair play” seem like quaint anachronisms.

Ralph Freso/Getty Images

We often talk about this in terms of record-high polarization. This is quantitatively true: Both elites and voters are now highly separated into partisan camps.

But qualitatively, this is something more. It’s not just how much we are divided, but more fundamentally how we are divided. The core problem is that the fundamental disagreement in our politics is now over what it means to be an American — it’s over what our nation’s core values are. And that has historically spelled trouble.

Recent events in Charlottesville bring these divisions into sharp relief: Can “very fine people” march alongside Neo-Nazis? Do counterprotesters on the alleged “alt-left” deserve just as much (or maybe even more) of the blame for any violence? Do Confederate generals deserve commemoration in our public squares? Answers to these questions reveal very different visions of both the past and the future of our country. And they break overwhelmingly along partisan lines.

To the political left, Donald Trump is un-American: His xenophobic, racist rhetoric stands in opposition to the true American vision of tolerance. It’s an affront to our nation of immigrants, a country in which equality is written into our founding documents. Any Republican who supports or voted for him is guilty by association.

To the political right, it’s the Democrats who are un-American. They denigrate our founding as a Christian nation and want to secularize everything. They want to sacrifice our sovereignty to globalist institutions under the guise of invented problems like global warming and to undermine our exceptional heritage by opening our borders to anybody, even those who want to blow us up. There is only one “real America” and it doesn’t include the coasts or cities where many Democrats live.

We now have two political parties with very different and increasingly irreconcilable ideas about what it means to be American, and, perhaps more saliently, what it is to be un-American.

Political scientists have documented how the spirited disagreements that used to characterize our political system have turned to rancor and disdain. Democrats and Republicans alike are far more likely today than they were only a few decades ago to say their rivals are not just wrong but stupid, selfish, and close-minded

This statistic is particularly telling: In 1960, only 5 percent of Republicans and 4 percent of Democrats said they would be displeased if their son or daughter married somebody with the opposing party affiliation. In 2010, 49 percent of Republicans said they would be unhappy if their son or daughter married a Democrat, and 33 percent of Democrats said they’d be unhappy if their son or daughter married a Republican. My strong hunch is that if this survey were conducted in 2017, those percentages would be even higher.

The paradox of partisanship: essential for politics yet potentially toxic

This partisan divide is obviously deeply problematic. But in thinking clearly about our partisan divisions, we first need to recognize that partisan conflict is a healthy and necessary aspect of democracy. In many ways, it’s the lifeblood.

As political scientist E.E. Schattschneider famously observed in his 1942 book, Party Government, “Modern democracy is unthinkable save in terms of parties.” It is unthinkable, because without competing parties, voters lack meaningful choices. Partisan conflict is necessary for democracy, because one-party politics is not democracy. It’s totalitarianism.

Competition gives parties incentives to respond to voters. And losing parties keep winning parties accountable by threatening to take away their supporters.

Parties mobilize and engage citizens to win elections, in the process bringing many otherwise apathetic citizens into politics. They bind disparate citizens together in a common purpose, providing a shared sense of collective energy necessary for a functioning democracy. Absent parties to structure and organize politics, democracy would crumble under chaos or apathy.

But the good things that parties accomplish come with side effects. To unite people, parties must also divide, by offering a common enemy to everyone on their side. As psychologists have long known, in-group loyalty and out-group hostility are two sides of the same coin. And under certain circumstances, particularly ones of high stress and high threat, and usually with active goading from above, out-group hostility can easily take on very dark and destructive forces.

Here’s the paradox: We can’t have democracy without partisanship. But when partisanship overwhelms everything, it becomes increasingly difficult for democracy to function.

Political sociologist Seymour Martin Lipset puts it this way in his 1959 classic Political Man: "A stable democracy requires the manifestation of conflict or cleavage so that there will be struggle over ruling positions, challenges to parties in power, and shifts of parties in office.” But he added that the system must permit “the peaceful 'play' of power,” and “the adherence by the 'outs' to the decisions made by the ‘ins.’”

If the “ins” fail to recognize the rights of the “outs,” Lipset concluded, “there can be no democracy."

In 1967, Lipset, along with Stein Rokkan, edited a volume entitled Party Systems and Voter Alignments: Cross-National Perspectives, in which they noted that political conflicts come in many varieties. They conceptualized them along a spectrum from most tractable — that is to say, workable and manageable — to least tractable.

At the tractable end of the spectrum, they placed a politics of pure economic materialism: They include conflicts over allocation of resources — disputes between producers and buyers, workers and employers, tenants and owners, and so on. These “can be solved through rational bargaining and the establishment of universalistic rules of allocation.”

At the intractable end are what Lipset and Rokkan call “Ideological oppositions.” These are all-consuming, 24-hour disputes “incompatible with other ties within the community.” In this kind of conflict, each side strives to “protect the movement against impurities and the seeds of compromise."

This is the kind of politics that leads to democratic breakdown and violence, and it’s where we appear to be headed today.

If polarization were simply a matter of parties negotiating on behalf of competing economic interest groups and allocating federal dollars, it follows that there are deals to be made (and plenty of earmarks!). Under such a politics, political leaders of both parties can trade roads and bridges over whisky cocktails at after-hours parties. Different sides might offer different perspectives, creating contrasts for voters. But at the end of the day, everyone understands that there are no permanent winners or losers — just temporary electoral swings. This is normal “interest-group politics,” in the jargon of political scientists.

When division involves purity and impurity, when it devolves into a pure contest between “us” and “them” — then there is no bargaining, because there are no negotiable principles, just team loyalties. “We” are good and pure, while “they” are evil and corrupt. And, of course, you cannot compromise with evil and corrupt. The preferred cocktails of such a politics are of the Molotov variety, and the roads and bridges are not to be traded, but to be burned.

This is doom-loop partisanship, because it contains many reinforcing dynamics that can quickly spiral out of control.

American politics has been transitioning from interest-group politics to doom-loop politics for decades, and we are now deep into a crisis.

In decrying contemporary hyperpartisanship, we must not over-romanticize the old interest-group politics, in which party leaders worked out bargains among competing interests behind closed doors.

Old-school transactional politics did grease the wheels of functioning government. But it also had plenty of problems, which its contemporary commentators frequently pointed out. For one, when the parties were loose overlapping coalitions of interests with minimal differences, it was hard for voters to send clear signals to elected leaders and hold them accountable.

In the 1950s, for example, the party system was like a faucet that produced only varying kinds of warm water. It also effectively stymied progress on civil rights, because civil rights groups were effectively cut out of the insider deal-making. The smoke-filled rooms of yore were great if you were inside them, but unless you were a white male, and frequently a Southerner, it was very difficult to get an invite.

There is also a very pragmatic reason not to over-romanticize the past. The earlier era existed on a foundation of cultural, demographic, economic, and technological conditions that are dramatically different today. America is a very different country than it was in the 1950s and 1960s.

More significantly, over the past half-century or so, partisan identities have become much more closely aligned with other social identities. Partisan divides now overlay religious divides, cultural divides, geographical divides, and racial divides. In the past, these identities used to cross-cut each other more often. Thirty years ago, you could be a culturally conservative Democrat, or culturally liberal Republican. These overlaps made the parties less distinct. They also made it easier to find common ground with opposing partisans based on other shared identities.

But as social sorting took place, we lost those potentially bridging ties. Moreover, our collective sense of cultural, regional, and ethnic status become more and more linked to the status of our two political parties, which came to represent these different identities. This made politics more emotional because it felt like even more was at stake with each election. It was not just the parties fighting each other, but also competing ways of life they represented.

As political scientist Lilliana Mason convincingly argues, “The more sorted we become, the more emotionally we react to normal political events.” And when emotions are heightened, everything becomes a threat to status. Politics becomes more about anger. And, here’s the warning from Mason that should give you goose bumps: “The angrier the electorate, the less capable we are of finding common ground on policies, or even of treating our opponents like human beings."

This is what doom-loop partisanship looks like. There’s no possibility for rational debate or middle-ground compromise. Just two sorted teams, with no overlap, no cross-cutting identities, and with everyone’s personal sense of status constantly on the line.

George Washington predicted that political partisanship would lead to democratic instability

The founders feared doom-loop partisanship from the beginning; it was why they were hostile to political parties. In making the case against parties, George Washington prophesied in his farewell address: “The alternate domination of one faction over another, sharpened by the spirit of revenge, natural to party dissension, which in different ages and countries has perpetrated the most horrid enormities.” Washington agonized precisely about the arms race of incivility and nastiness that has overwhelmed national politics over the past few decades.

Washington feared that instability would “gradually incline the minds of men to seek security and repose in the absolute power of an individual.”

George Washington’s vision of no parties, just men of good character, was obviously unworkable. American politics quickly organized around the competing Hamiltonian Federalist and Jeffersonian Democratic-Republican parties. Schattschneider was right: Democracy does require parties to structure and organize conflict.

But Washington did have a point in fearing that “the founding of [parties] on geographical discriminations,” could spell trouble. Citizens of each region would be surrounded only by fellow partisans, those reinforcing their shared grievances and making it far easier to demonize the “other.”

The founders did not have to wait long to see evidence of the dangers of partisan conflict. In the 1800 presidential campaign, Democratic-Republican Thomas Jefferson’s winning campaign accused Federalist John Adams of having a “hideous hermaphroditical character, which has neither the force and firmness of a man, nor the gentleness and sensibility of a woman.” Adams’s campaign in turn called Jefferson "a mean-spirited, low-lived fellow, the son of a half-breed Indian squaw, sired by a Virginia mulatto father." Adams dominated New England, but Jefferson won the entire South.

But the divisions then healed, and American politics briefly enjoyed a relatively tranquil period of one-party dominance referred to as the “Era of Good Feelings.” The 1824 election brought divisiveness back to American politics, when Andrew Jackson won the plurality of the popular vote but the US House gave the presidency to John Quincy Adams, angering Jackson and his supporters. They organized the Democratic Party and won in 1828. In response, a coalition of Jackson’s opponents organized the Whig Party.

For two decades, both parties were national parties, and national political debates tended to center on economic materialism, issues like tariffs, internal improvements, and banking. Classic interest group politics had returned.

For the system to hold, though, both parties had to suppress the one big issue that would divide the nation by region: slavery. As long as both parties had Northern and Southern wings, they had strong incentives to work out internal compromises on the issue (and they did). But in the 1850s, as westward expansion forced the slavery question to a head, passions erupted. The Whigs split over the issue first, with Republicans emerging out of the ruins of the Northern Whigs, a splinter group of Northern Democrats who also opposed slavery, and a few other third parties

In 1860 election, a regional candidate named Abraham Lincoln swept the North, got no support in the South, and was elected with only 39.8 percent of the national popular vote in a four-party election. Separate Southern and Northern Democrat candidates ran after Democrats couldn’t agree on a candidate at their convention because of the slavery question.

After Lincoln won, Southern states responded by sending the strongest “Not my President” message possible: They seceded from the Union. An estimated 750,000 men lost their lives in the four-year war that followed.

The lesson of this admittedly potted history is that when the central political division shifts from economic materialism to disputes over fundamental values and questions of national identity, democracy threatens to become unstable.

How close are we to a new breaking point?

Are we at such a cataclysmic moment today? There’s more and more evidence that we are.

Our politics is now both regionalized and racialized in ways that we haven’t seen in a long time. As we separate into our separate, all-encompassing tribal loyalties, we’re falling into three very dangerous and related self-reinforcing cycles: 1) the disappearing trust doom loop; 2) the disappearing electoral legitimacy doom loop; and 3) the growing inequality doom loop.

The disappearing trust doom loop

“Political trust,” note Marc Hetherington and Thomas Rudolph note in their recent book Why Washington Won’t Work, “is critical because it helps create consensus in the mass public by providing a bridge between the governing party’s policy ideas and the opinions of those who usually support the other party.” Without some trust from the other side, it is almost impossible to govern in a 50-50 nation that requires supermajorities to pass legislation.

Years of bad faith and negative partisanship have convinced both elites and voters that the other side cannot possibly represent them, and that therefore negotiation is impossible. These animosities are nurtured and honed in conflicting media narratives, with each side consuming only the information diet that puts them in the right, nodding at the commentators who say the other side is acting in an “un-American” fashion.

Political gridlock follows. Institutions don’t function. Trust declines. Anger grows. Somebody needs to be blamed. That somebody is always the other side. They cannot be trusted. They must be crushed.

The disappearing electoral legitimacy doom loop

Growing distrust feeds into another doom loop: the disappearing legitimacy loop.

Again, if the other side is bad and untrustworthy, it follows that extraordinary means should be taken to secure electoral victory. And when everybody we surround ourselves with agrees we are on the side of good, it is much easier to explain away defeat as the product of cheating and illegitimacy.

When Barack Obama was president, many Republicans treated Obama as illegitimate. “Where was the birth certificate?” they repeatedly asked. “He’s not an American. He’s not a Christian. He’s not my president.”

Now that Trump is president, many Democrats have adopted their own delegitimizing narrative. “Clinton won the popular vote,” they say. Or “Trump is only president because of Comey.” Or: “If the Russians hadn’t hacked the election, Clinton would have won.”

And if the outcome is illegitimate, it follows that the procedures somehow need to be changed, or the results need to be challenged, or both.

We saw this in 2016. Candidate Trump tossed off reckless allegations of likely illegal voting, and refused to preemptively accept the election results (“I will keep you in suspense”).

Even after he won, he held onto the claim that Clinton only won the popular vote because 3 to 5 million people voted illegally, and directed his Justice Department to investigate. Certainly, some of this is paranoia — but it’s also a collective narrative that right-wing media has been pushing for a long time. And it’s a great way to prepare to challenge any future electoral results the right doesn’t like.

This logic helps to explain what Richard Hasen calls The Voting Wars — the escalating attempts to shape voting rules and procedures to gain partisan advantage, the most controversial aspect of which has been the Republican-led introduction of voter ID laws.

The movement began slowly in the early 2000s, as states began to ask voters to bring some kind of identification to the polls. In 2005, both Indiana and Georgia passed strict rules requiring such documentation (they were somewhat delayed by court challenges). By 2016, 11 states had “strict” voter ID laws, according to the National Conference of State Legislatures.

Republicans have made repeated attempts to justify these laws on grounds of illegal voting, claims which are almost always racialized. This fuels the perception of illegitimacy on both sides. If Republicans win in close elections, Democrats say it’s only because they cheated by making it harder for Democratic constituencies to vote; if Democrats win in close elections, Republicans say it’s only because they voted illegally.

Evidence is nonexistent for Republicans’ case. But in our hyperpartisan information world, perception fuels reality.

Hence my worry about a close election in 2020 and the constitutional crisis it could bring.

The growing inequality doom loop

Finally, it’s hard not to notice that inequality and polarization have grown almost in tandem over the past four decades. As American society has become more unequal, it has become more polarized, and vice versa.

In their 2006 book, Polarized America: The Dance of Ideology and Unequal Riches, political scientists Nolan McCarty, Keith Poole, and Howard Rosenthal posit that polarization contributes to inequality because it increases gridlock, which makes it harder to adjust policy in redistributive ways.

That’s convincing as a partial explanation, but something else is probably happening here as well. Given that most people find it unthinkable to vote for the other party because of regional and social identities, that makes it easier to for politicians to take the continued support of their rank and file for granted. They are then freed, Democrats and Republicans alike, to take cues from their donors in the top 1 percent.

But voters are not dumb: They can tell that they’re mostly getting short shrift while a few people at the top are doing very, very well. And they resent it — even if they don’t see a way out of the predicament. This makes them angry at the political system. Not surprisingly, the more unequal the society, the less likely its citizens are think that the country is governed democratically, as recent research by Roberto Foa and Yascha Mounk shows.

Roberto Stefan Foa and Yascha Mounk, Journal of Democracy web exchange

What America is and isn’t

To understand the implications of these intertwined trends, let’s dig into the ur-factors that are powering this distrust and division: the two competing visions of what America is and isn’t, and the ethnic/cultural-educational/geographic/partisan split driving these two visions.

This may be the most important schism of all: White Christian America, once dominant, realizes it is dominant no more. Now it competes, almost equally, with the multiracial secular America.

It’s this moment of slippage that creates a the most pressing threat. During the 2016 election, Donald Trump played directly to fears of diminishing white prestige: “If I don’t win I think this will be the last election that the Republicans have a chance … because you're going to have people flowing across the border, you're going to have illegal immigrants coming in and they’re going to be legalized and … vote and once that all happens you can forget it.” Immigration, here, became a stand-in for the fear of non-whiteness — and that is what connects anti-immigrant policies with birtherism, two of the pillars of Trump’s campaign

Or, in even more extreme language, here’s what one Trump supporter told a CNN reporter about Muslim immigration: "It's basically a form of genocide … They're trying to breed out the white population.” “Genocide,” though, is not an issue where productive debate can lead to a moderate compromise.

When whites are reminded that America is on the verge of becoming a majority-minority nation, they tend to become more conservative, political science tells us, and that may explain the rise of right-wing populism.

When torch-wielding white supremacists chant “You will not replace us” in Charlottesville, this is the fear of demographic change that is driving them.

It can be illuminating — and terrifying — to put white ethnic resentment into the context of ethnic resentment in other nations. In a transnational study of ethnic violence, MIT political scientist Roger Petersen found that a major risk factor for ethnic violence was anger stemming from “the feeling of being politically dominated by a group that has no right to be in a superior position.” Typically, that occurs when an ethnic group that was formally subordinate achieves new status and power.

Considering the horrific ethnic violence that other nations have seen — in Yugoslavia, for example, just two decades ago — it is truly chilling to note that (mostly white) anti-government militias in the US grew from 42 in 2008 to 276 in 2016. Gun sales under Obama more than doubled.

It seems unlikely that America will literally break apart as it did in 1861, despite the growing bull market for such predictions, and the continued efforts for Californians to declare their independence from the rest of the country.

A more plausible vision involves the slow dissolution of the United States. The federal government grows ever more dysfunctional because of deep political divisions. A growing number of states descend into gridlock because of their own urban-rural splits. Meanwhile, a limited number of one-party blue and red states put in place ever-more radical political visions. The United States recedes as an international actor, while large states like California, New York, and Texas increasingly operate independently on the world stage, signing accords and treaties.

There are certainly ways in which this dissolution could turn out to be relatively peaceful; if I’m feeling optimistic, I can imagine it diffusing some of our current political animus. But the strength of Washington, DC, in our constitutional system creates hurdles for such quasi-federalism. And in those states where red and blue pockets co-exist, continued street violence like what we witnessed recently in Charlottesville is likely to continue. We can only hope it remains relatively low-level.

Elites manipulate and stoke the tribal passions generated by modern partisanship

Is there an alternative to the slow-motion descent and disintegration I’ve just described? If there is, we’ll have to confront that our divisions are real, and that there’s no “moving past identity politics” within a political system in which partisan conflict is fundamentally organized around identity — white identity most definitely included.

One way to understand the deep-seated nature of identity is through the lens of economic anxiety.

Andalou Agency/Getty Images

That phrase became something of a joke in 2016, with many people noticing that voters described as economically anxious tended to have particular anxiety about black and brown people. But the two kinds of anxiety are deeply intertwined.

Despite Trump’s supposed appeal of economic populism, repeated studies found no evidence in the data that individual-level economic immiseration drove voting patterns in the 2016 election. One explanation is that truly objective economic anxiety (as opposed to the subjective kind voters report any time when the other party is in power) was affecting both Democrats and Republicans. Both Democratic voters and Republican voters had economic troubles. But they had very different stories about who was responsible for their troubles. And those stories were caught up in ethnocultural and regional identities.

As political scientist Kathy Cramer notes in The Politics of Resentment, economic conditions are not raw objective facts. Instead, “they are perceptions of who is getting what and who deserves it, and these notions are affected by perceptions of cultural and lifestyle differences.”

To be fair, it is not necessarily irrational for voters to think that what’s good for their group is also good for them individually. But this also makes it very easy for political leaders to manipulate voters’ perception that allocation of resources is unfair. “My fear,” Cramer goes on to argue, “is that democracy will always tend toward a politics of resentment, in which savvy politicians figure out ways to amass coalitions by tapping into our deepest and most salient social divides: race, class, culture, and place.” It’s my fear, too.

For a project called the Democracy Fund Voter Study Group, I analyzed voters who cast a ballot for Obama in 2012 but for Trump in 2016 (the much-discussed “Obama-Trump” voters, who seem paradoxical on the surface). They were almost all economically liberal, but conservative on social-identity issues. They shifted to Trump primarily because he spoke to their cultural and ethnic concerns.

This is why Trump won, and this is how Trump is governing. He has pushed hard to keep identity politics at the top of the political agenda, stoking further cultural animosities. Indeed, he employs the same strategy Southern Bourbons used 120 years earlier when agrarian populists started getting uppity and talking about class interests: The Southern grandees made ethnocultural identity more salient, to distract from economics.

"Low-class whites are precisely the group that rejects class politics,” Robert Huckfeldt and Carol Kohfeld argued in Race and The Decline of Class in American Politics, a book primarily about the South. They are “the group most sensitive to the presence of blacks and most unwilling to cooperate with them in the same political coalition … race is the wedge issue that disrupts lower-class coalitions,” italics mine.

That means that long as Trump continues to maintain the right enemies, it may not matter how little he improves the material circumstances of his alleged “base.” Even if Trump voters may feel let down by the lack of resurgence in manufacturing jobs, or their continued wage stagnation, many have reached a point where voting for the Democrats would be akin to voting for an America that feels completely alien to them: As long as Democrats are the party of cultural diversity, they will always be the un-American party.

And Democrats also have an electorate that cares deeply about cultural and identity concerns — even more so now — given the Trump administration’s hostility to traditional civil rights (see: Attorney General Jeff Sessions). Moreover, almost half of Democrat voters are now people of color, and a solid majority are women. These are Democrats core constituent groups, and they are energized and empowered.

Witness what happened when the Democratic Congressional Campaign Committee (DCCC) announced a willingness to support pro-life candidates — a move that could be charitably interpreted as trying to shift the Democratic Party back toward the older model of a big-tent ideology. The group was hammered by the Democratic base.

In short, neither party can escape from identity politics because identity is more than ever the ideological glue that holds together both party coalitions. The corollary is that any argument about economic policy can never be just an argument about economic policy.

How then, do we unscramble this? First, we have to understand the features of our current politics that are making this situation worse. Three conditions stand out:

• Our winner-take-all system of elections
• The expanding powers of the presidency, and the federal government generally
• The outsized importance of private money in politics

The winner-take-all system

Let’s begin with the single-winner plurality system of elections. This is the big one.

Unlike in other advanced democracies (most of which use various methods of proportional voting), the American legislative elections are held through a series of separate elections, each held in single-member districts, in which the candidate with the most votes wins. Because there is only one possible winner, any vote for a third party is essentially a “wasted vote.” This is the reason there are no third parties in the United States.

The deficiencies of a two-party system become glaring when the parties become as diametrically opposed as they are now. “A two-party democracy cannot provide stable and effective government unless there is a large measure of ideological consensus among its citizens,” writes the economist Anthony Downs in his widely cited classic An Economic Theory of Democracy.

The problem, as Downs explained, is that the country will swing from extreme to extreme, with each side antagonizing the other until the middle drops out entirely. In contrast to the ‘50s, the electoral system is now a faucet where the slightest change produce scalding and near-freezing water.

“When the distribution has become so split that one extreme is imposing by force policies abhorred by the other extreme,” Downs wrote, “open warfare breaks out, and a clique of underdogs seizes power." Just as George Washington feared.

FairVote

To expand the party system to create space for possible centrist or other alternative parties would require a change in electoral rules create space for proportional voting. Earlier this year, Rep. Don Beyer (D-VA) introduced the Fair Representation Act, which would create larger House districts, and have the voters choose several representatives in each one. FairVote has a helpful explainer on how this would work. I wrote earlier this year about why this plan makes a ton of sense.

The system would give conservatives in New York City, for instance, a say in who gets chosen for the House: They would be likely to win a seat or two. Likewise, liberals in red Alabama districts might be able to elect a candidate. Ranked choice voting would mean citizens could vote their hearts — yet still express a preference for the “lesser evil” of the two major-party candidates.

The power of the presidency

The high-stakes winner-take-all dynamic is made worse by concentrating so much power in a single unilateral actor -- the president — who is asked to do something impossible: be the single tribune of very divided nation. (A difference of 80,000 votes in three key states, and the country swings wildly in a different direction: That is not a foundation for democratic stability.)

Some have observed that Trump is acting as if he only represents those who voted for him, rather than the whole country. Of course he is. When the country is divided, especially over fundamental questions of national identity, it’s impossible to govern from a nonexistent middle ground. Trump stays in power based on the support of a minority that also happens to be the majority of the majority.

This is precisely the danger of winner-take-all systems. All Trump needs is 60 percent support among Republicans (a narrow majority of the narrow majority); his poll numbers can stay in the low 30s and he can in theory retain power. This is how he is governing, because this is how the system is set up for him to govern under a divided public. He’s not going to win over any Democrats, and he has no incentive to try. This is why the power of the presidency is so dangerous.

The Constitution deserves some blame for this situation, but a considerable amount of the expansion of executive power over the last several decades has happened because Congress has ceded authority and oversight.

Changing this dynamic requires Congress reassert its authority — but Congress, of course, is also shaped by the dynamics of two polarized parties.

Absent the structural revolution of a shift to multimember districts, a more modest change would be for both parties in Congress to decentralize internally. Under current organizational structures, party leaders dictate and control policy in ways that punish any rogue members who want to try something that wouldn’t help the party win the next election — like working across the aisle.

That would require courageous votes from those entrepreneurial congressmen who would like to do much more freelance coalition building. But if they are willing to defy their leaders and whips, it would reveal that parties have far more internal ideological diversity than partisan voting patterns show.

For members to act independently, however, Congress would have to invest in more internal staff with policy expertise, and members have resisted such investments for a long time.

A shift toward localism, or federalism, could help, too. If partisans on both sides felt more secure that they could live their values at home regardless of who was in power in Washington, this could lighten some of the zero-sum nastiness and dysfunction of national politics. For those on the left who have long seen federalism as code for mistreatment of minorities, the resistance of blue cities to Trump’s more draconian anti-immigration policies should offer some reassurance.

The outsized importance of private money in politics

Finally, there is the outsized role of private money in politics. Again, because one-party majority dominance is always narrowly in reach, both parties have been in a decades-long fundraising arms race that has consumed their ability to govern.

In the campaign finance chase, both parties have gained profitably from catering to extremely wealthy individuals. In theory, the disconnect between voters’ wishes and donors’ priorities should create a problem for both parties, given that the parties economic policies are at odds with what most of their voters would prefer. But herein lies the great advantage of the two-party system — for party leaders. By dialing up the cultural conflict, they can distract voters from the disconnect between elite preferences and the public good. To the extent campaign finance reform can get politicians to refocus their attention on the economic needs of voters, not donors, that will help greatly in controlling polarization.

To return to the nightmare scenario I began with: Would Trump really refuse to concede, if he lost a close race in 2020? If so, would Republicans really stand by him? And if so, then what? Would the Courts step in? Would the military? Would Trump be forcibly removed from office? If so, what would he and his supporters do in response? Would a forcibly installed President Kamala Harris have to crack down on civil liberties in order to hold the peace? Would some deep red states then secede?

These hypotheticals are meant mainly to clarify the stakes of our predicament. But given current trends, such a scenario is frighteningly in the realm of the plausible. The raw division and conflict and mutual demonization are there, and are getting worse and worse. The events in Charlottesville are just the latest manifestation. Where will things be by 2020? What happens if there’s a major recession, and even more anger?

I honestly don’t know whether we’ll get escape this mess without a constitutional crisis.

We need partisan conflict to organize politics. Without political parties, there is no meaningful democracy. But we are deep into a self-reinforcing cycle of doom-loop partisanship. We need to think hard about how to escape this trap, before it is too late.

Lee Drutman, a regular contributor to Polyarchy, a Vox blog, is a senior fellow in the political reform program at New America. He is the author of The Business of America is Lobbying: How Corporations Became Politicized and Politics Became Corporate (2015).

The Big Idea is Vox’s home for smart discussion of the most important issues and ideas in politics, science, and culture — typically by outside contributors. If you have an idea for a piece, pitch us at thebigidea@vox.com.

Will you support Vox today?

We believe that everyone deserves to understand the world that they live in. That kind of knowledge helps create better citizens, neighbors, friends, parents, and stewards of this planet. Producing deeply researched, explanatory journalism takes resources. You can support this mission by making a financial gift to Vox today. Will you join us?

One-Time Monthly Annual

$5/month

$10/month

$25/month

$50/month

Other

$

Yes, I'll give $5/month

Yes, I'll give $5/month

We accept credit card, Apple Pay, and Google Pay. You can also contribute via