Reversibility
Democracy, Loss, and the Promise of Return
This essay is the second in a series exploring the core conditions of democracy—those that make democratic loss possible. Read the introduction to this series here, and the introduction to this Substack here.
Democracy’s core feature, its innovation, is that people who lose an election willingly give up power. People will only do so—and democracy can only exist—if they believe that a loss today is reversible (you’ll have a chance to win next time) and survivable (it won’t destroy you or the things that are most important to you). In upcoming essays we’ll talk about survivability. Today we will discuss reversibility. What are our reversibility mechanisms? Where do they come from? How are they failing?
Surprisingly, our constitutional system has produced fewer reversibility mechanisms than survivability ones. Surprising, I say, because the first thing that we think of when we think about democracy is elections, the core of the reversibility mechanisms.
But the Framers spent most of their energy ensuring that factions could not destroy each other. As we will see, the Constitution is saturated with survivability protections. They were less systematic about ensuring that today’s losers could become tomorrow’s winners.
What does the Constitution actually give us on reversibility? Elections, obviously. That’s the big one. Regular, scheduled elections for the House every two years, the Senate every six, the President every four. This staggering is itself important. It means that no single election can sweep everything away, that change happens incrementally and gradually, that there are always opportunities to course-correct.
Beyond elections themselves, the Constitution offers the First Amendment freedoms: speech, press, assembly, and petition. These are reversibility mechanisms. They allow political minorities to argue their case, to organize, to build coalitions, to change minds. Without them, the winners could crush dissent; the losers of one election would have no tools to become winners of the next.
That’s mostly it, as far as the original constitutional design goes. The electoral franchise itself was left largely to the states, and initially extended primarily to propertied white men. Later amendments dramatically expanded the right to vote: the Fifteenth Amendment (race), the Nineteenth (sex), the Twenty-Fourth (eliminating poll taxes), the Twenty-Sixth (age eighteen). Each expansion was a fight, sometimes a bloody one. Each was a recognition that reversibility requires broad participation.
In addition, a handful of amendments added structural anti-entrenchment devices: the Seventeenth Amendment’s move to direct election of senators and the Twenty-Second Amendment’s presidential term limits. The original Constitution also provided regular census-based reapportionment and impeachment. But these tools are blunt, indirect, or rarely used.
Much of the actual work of reversibility has therefore fallen to subconstitutional mechanisms: statutes, norms, and institutions.
Our political system has produced some reversibility mechanisms through legislation. The Voting Rights Act of 1965 was perhaps the most significant, a recognition that the formal right to vote meant nothing if states could suppress it through literacy tests and outright intimidation. The Act’s preclearance provisions required jurisdictions with histories of discrimination to get federal approval before changing their voting rules.
Campaign finance reforms, particularly after Watergate, attempted to prevent money from overwhelming democratic competition. The Federal Election Campaign Act, the creation of the FEC, and various disclosure requirements were all efforts to keep elections clean and meaningfully competitive. These were bipartisan achievements. The Bipartisan Campaign Reform Act of 2002, known as McCain-Feingold (after the Republican and Democrat who sponsored the bill), was also a product of cooperation across party lines.
These mechanisms emerged from crises. The Voting Rights Act came from the blood of Selma. Campaign finance reform came from the corruption of Nixon.
We tend not to build reversibility protections until we have already seen the cost of entrenchment.
Here is an uncomfortable truth: for the past several decades, the Supreme Court has focused mostly on survivability questions while systematically dismantling reversibility mechanisms.
Consider what the Court has spent its energy on. Gun rights. Religious liberty. Abortion. Parental rights. Substantive due process. These are survivability questions. They concern what the government can and cannot do to you regardless of who holds power. Can it take your guns? Can it suppress your religion? Can it control your body, your family, your intimate choices? You can agree or disagree with how the Court has resolved these questions. But that has been the focus.
Meanwhile, on reversibility, the Court has been a demolition crew.
First, campaign finance: gutted. When people talk about this, they usually mention Citizens United v. FEC (2010), which is the most famous case. But the erosion began earlier and continued after. Buckley v. Valeo (1976) first established that spending money on elections is a form of speech protected by the First Amendment. First National Bank of Boston v. Bellotti (1978) extended First Amendment protections to corporate political speech. Citizens United itself struck down limits on corporate independent expenditures. McCutcheon v. FEC (2014) eliminated aggregate contribution limits. Across this line of cases, the Court progressively narrowed its understanding of corruption to effectively mean only explicit quid pro quo bribery, ignoring the subtler but more pervasive corruption of a political system systematically responsive to those who fund it.
A word about Citizens United specifically, because it deserves special criticism. The plaintiffs had not asked the Court to strike down corporate expenditure limits. They asked for a narrow ruling that their film about Hillary Clinton did not fall within the restrictions of the Bipartisan Campaign Reform Act. The Court could have ruled in their favor on that narrow question. Normally, that’s what the Court would do. It doesn’t usually offer arguments that the parties didn’t make and provide relief that the parties didn’t seek. Nevertheless, it ordered re-argument and expanded the case to ask whether the Court should overrule precedents that had upheld limits on corporate political spending. In other words, the Court reached out for a question it was not asked to decide and used it to dismantle decades of bipartisan campaign finance law. This is judicial activism in the truest sense.
I should be fair. The Court’s campaign finance decisions rest on understandable First Amendment principles. The freedom to spend money to disseminate political speech is connected to the freedom of speech itself; restrictions on spending do burden political expression. But what the Court got wrong, catastrophically wrong, is the failure to recognize that the political concerns reflected in the First Amendment must be balanced against the democratic concerns reflected in campaign finance laws. The First Amendment exists to protect democratic self-governance. Campaign finance laws also exist to protect democratic self-governance. When the Court treats one as absolute and the other as suspect, it undermines the very democracy that both are meant to serve.
The result? In 2024, 100 billionaire families poured $2.6 billion into federal elections, one of every six dollars spent. That is two-and-a-half times what billionaire donors spent in 2020. It is a 160-fold increase since Citizens United. Elon Musk alone spent nearly $300 million, almost all of it supporting Donald Trump and his allies. He then sat in the government he helped purchase.
Second, the Voting Rights Act: hollowed out. In Shelby County v. Holder (2013), the Court struck down the preclearance formula. The preclearance process required some jurisdictions, typically those with an extensive history of racial discrimination, to seek pre-approval of any new election practice. The purpose was to make certain that the new practice would not reinstate racism in the voting system. It was one of the most effective tools to fight entrenched racism, and it won overwhelming bipartisan in Congress every time it was up for renewal. Nonetheless, the Court decided that things had changed in the South and the old coverage formula was outdated. Justice Ginsburg’s dissent was prophetic: “Throwing out preclearance when it has worked and is continuing to work to stop discriminatory changes is like throwing away your umbrella in a rainstorm because you are not getting wet.” Texas immediately announced it would implement a voter ID law that had previously been blocked. Other states followed. In Brnovich v. Democratic National Committee (2021), the Court made it even harder to challenge discriminatory voting laws under what remained of the Act.
Third, partisan gerrymandering: untouchable. In Rucho v. Common Cause (2019), the Court held that federal courts cannot review partisan gerrymandering claims at all. They present “political questions beyond the reach of the federal courts.” Chief Justice Roberts acknowledged that extreme partisan gerrymandering is “incompatible with democratic principles.” He just concluded that courts could do nothing about it. Think about that. Extreme partisan gerrymandering is incompatible with democracy; but too bad. The result is that elected officials choose their voters instead of voters choosing their elected officials. Justice Kagan’s dissent: “Of all times to abandon the Court’s duty to declare the law, this was not the one. The practices challenged in these cases imperil our system of government. Part of the Court’s role in that system is to defend its foundations. None is more important than free and fair elections.” She recognized and valorized one of democracy’s first principles; he recognized it but dismissed it.
These pressures are not in the past. They are live. In December 2025, the Supreme Court heard argument in National Republican Senatorial Committee v. FEC, a challenge to federal limits on coordinated party expenditures. Observers widely noted skepticism from several Justices toward those limits. If they fall, one of the few remaining statutory barriers to donor-driven entrenchment will weaken further.
Voting-rights doctrine remains unsettled as well. The Court has again taken up questions that go to the VRA’s core in Louisiana v. Callais. It may well gut what remains of the VRA this term.
Election-administration rules are also in flux. The Court has agreed to hear a case addressing whether federal Election Day statutes preempt state laws that count ballots received after Election Day but cast on time. States are already adjusting their rules in anticipation. And mid-decade redistricting has become an increasingly open strategic tool, with the Court allowing Texas to use a newly enacted congressional map—explicitly designed to win more seats for Republicans—for the 2026 elections.
The pattern is consistent. When the question is whether the government can take your guns or burden your religion, the Court finds judicially manageable standards. When the question is whether politicians can entrench themselves in power by manipulating electoral rules, the Court throws up its hands.
The Court’s abdication is indefensible, and I want to explain why.
When Marbury v. Madison established judicial review in 1803, Chief Justice Marshall did not point to any specific constitutional text granting the Court that power. The Constitution does not say “the Supreme Court shall have the power to strike down unconstitutional laws.” Marshall recognized instead that the structure of the Constitution required someone to resolve disputes about constitutional meaning, and that the Court was the only institution positioned to do it. Judicial review was an act of interpretation in service of a deeper structural principle about how a democratic system must work.
The same logic applies with even greater force to reversibility mechanisms. When the political branches produce rules to protect electoral competition, and those rules are challenged, the Court is uniquely positioned to protect them precisely because the political branches cannot be trusted to police their own entrenchment. Congress (by definition made up of incumbents) cannot typically be expected to fix gerrymandering that benefits incumbents. State legislatures cannot usually be expected to limit money that flows to their campaigns. The fox cannot guard the henhouse. This is not a problem the political process can solve on its own. It is the structural argument for judicial engagement, not abdication.
For decades, scholars have argued that courts have a special responsibility to police the rules of the democratic game, a tradition often described as “representation reinforcement.” The insight is that courts should be most vigilant precisely where the political process cannot be trusted to correct itself: where incumbents manipulate the rules to entrench their own power, where majorities suppress the political participation of minorities, where the channels of political change are obstructed. That insight has been articulated repeatedly. Alas, it has gone largely unheeded.
The irony is especially acute. The bipartisan campaign finance reforms that the Court struck down were themselves products of the democratic process. Elected officials, recognizing the corrosive influence of money in politics and motivated by Nixon’s corruption and the public’s crisis of confidence in elections—in reversibility—enacted laws to limit money’s influence. These are the rare cases where incumbents have reduced their own advantages. The Court then struck those laws down in the name of protecting democracy from government overreach. But democracy was not overreaching. Democracy was trying to protect itself, to protect reversibility. And the Court stopped it.
To some degree, the asymmetry between survivability and reversibility is understandable. Substantive rights are more galvanizing and more legible to the public than procedural ones. People know immediately when they have lost access to abortion or when their religious practices are burdened. The harm from gerrymandering or dark money is diffuse, structural, hard to see and harder to feel.
Consider the public response to Dobbs. When the Court overturned Roe v. Wade in 2022, Americans responded with overwhelming disapproval. Democrats bucked the historical trend of presidential party losses in midterm elections. After Dobbs, abortion ballot measures repeatedly mobilized voters. In 2022, even Kansas rejected an anti-abortion constitutional amendment. In 2024, abortion-rights measures prevailed in seven states, though they fell short in Florida (because of the 60% threshold the state requires) and lost in Nebraska and South Dakota. Voters understood exactly what they had lost and mobilized accordingly.
Compare that to the response to Rucho or Shelby County. These decisions were covered in the news, criticized by voting rights advocates, lamented by editorial boards. Then everyone moved on, leaving only people like me to sound the alarm. No ballot measures. No electoral backlash. No galvanized public demanding reform.
Citizens United generated more heat. It is one of the few Supreme Court decisions that non-lawyers can name, and it provoked genuine public outrage. But that outrage never translated into effective action. There were no ballot measures to overturn it, no sustained electoral consequences for its supporters. The anger was real but directionless, because what Citizens United damaged was not a right people could feel in their daily lives but the structural conditions of fair political competition. That is the pattern: when the Court strips away a substantive right, the public knows what it lost and fights to get it back. When the Court degrades the mechanisms of democratic accountability themselves, the loss is harder to see, harder to name, and harder to mobilize against.
The erosion of reversibility is quiet. It does not feel like losing a right. It feels like politics as usual. But suddenly it isn’t.
The consequences are becoming impossible to ignore. We are witnessing the emergence of something that deserves the name oligarchy.
At Trump’s second inauguration, billionaire tech executives were given conspicuously prominent placement—Musk, Bezos, and Zuckerberg among them—turning what used to be whispered influence into visible pageantry. This is not subtle. This is not hidden. This is oligarchy announcing itself.
Money in politics is a both-sides problem. Democrats have their billionaire donors too. Hunter Biden profited from proximity to power in ways that were ethically ugly even if the precise legal implications remain disputed. Bill Clinton’s pardon of Marc Rich and his use of the Clinton Foundation became an enduring symbol of donor-access politics. I am not claiming that one party is pure and the other corrupt.
But the current moment is particularly grotesque. Over 80 percent of campaign spending by the hundred wealthiest Americans in 2024 went to Republicans. Musk did not merely donate to a campaign. He received in return unprecedented authority for a private citizen, the power to reshape the federal government in ways that directly benefit his business interests. This is capture. And, since coming to power, this administration’s corruption has been dazzling to behold, as it enriches itself from those purchasing influence.
All of this erodes trust in elections. Not because elections are stolen (they are not), but because they are bought. And here is a dark irony: when people lose faith that elections are fair, they become susceptible to lies about stolen elections. January 6 can be understood—not justified, but understood—in part, as a response to a system that many Americans correctly perceive as rigged, just not in the way the insurrectionists believed. The system is untrustworthy. Not because of voting machine conspiracies or suitcases of ballots. Because the rules have been systematically tilted to favor the powerful. The rioters were catastrophically wrong about the specifics. They were not entirely wrong about the underlying reality.
What is to be done? Two things, neither radical.
First, the courts must start enforcing the rules of the game. I am not asking for judicial activism. I am asking for courts to treat reversibility mechanisms with the same seriousness they treat substantive rights. Again: if there are judicially manageable standards for determining when a gun regulation is unconstitutional—and the Supreme Court majority claims there are—then there are judicially manageable standards for determining when a gerrymander is unconstitutional. Some state courts, at least, have shown that such standards can be developed and applied. Federal courts can do the same if they choose to.
Courts should also take seriously the anti-corruption interest in campaign finance law. Corruption is not limited to explicit quid pro quo bribery. Corruption includes the subtler but more dangerous reality of a government systematically responsive to the interests of those who fund campaigns. When a billionaire spends $300 million to elect a president and then receives a government position from which he can reshape policy affecting his businesses, that is corruption in any meaningful sense of the word. The Court’s refusal to recognize this is willful blindness.
None of this is a call for courts to abandon substantive rights protection. The claim is narrower: if democracy depends on both survivability and reversibility, courts cannot consistently treat the rules of democratic competition as marginal while treating substantive rights as their central constitutional project.
Second, the political system must find bipartisan ways to strengthen anti-entrenchment rules. This sounds naive, I know. But there are reasons for hope. Independent redistricting commissions have been adopted in states across the political spectrum, not just blue California but also red Arizona and Ohio (though Ohio’s has faced challenges). Ranked-choice voting is spreading. Automatic voter registration has bipartisan appeal when framed correctly.
The key is to frame these reforms as rules of the game, not as advantages for one side or the other. Everyone should want elections that the loser accepts as legitimate. Everyone should want a system where today’s minority can become tomorrow’s majority. Everyone should want to know that the other side, if it wins, won fairly.
I want to be clear about where I am placing blame. The political system bears responsibility for failing to update our reversibility mechanisms, for allowing partisan advantage to override democratic principle, for not responding to crises until they become catastrophes. At least they have the excuse that the incentives built into the political system run in the other direction.
And the Framers, too, bear some blame for neglecting to build in additional reversibility mechanisms into the Constitution. At least they have the excuse of having been working essentially from scratch in designing a national democracy, of not being able to predict the future.
The current Court, on the other hand, has no excuses. It bears special blame. It is the institution that struck down campaign finance protections that the democratic process itself had created. It is the institution that gutted the Voting Rights Act. It is the institution that declared partisan gerrymandering beyond judicial review. At every turn, when asked to defend the foundations of electoral competition, the Court has declined. And in doing so, it has failed to fulfill the structural role that Marbury itself recognized: the Court as the guardian of the constitutional order, precisely because no one else can be.
This is not inevitable. The Court has made choices. It could make different ones.
The rules of the game matter. When people believe the game is rigged, they stop playing by the rules. Some of them storm the Capitol. Others simply disengage from politics altogether, ceding the field to those with money and power.
Neither response leads anywhere good. If we want people to accept electoral outcomes, to believe that this loss can be reversed by the next election, we need to give them reason to believe it. That means tending to the mechanisms of reversibility with the same care we have devoted to the mechanisms of survivability.
Enjoyed the article. Here’s some slight pushback.
The Court is correct to keep their hands off the political gerrymandering issue. You rightly point out that the public’s belief in the reversibility of elections is the critical thing, and judicial tampering in districting rules will always undermine that. Any change that a court makes to redistricting will benefit one party at the expense of the other, and thus will be necessarily controversial. And unlike race-based redistricting, where there is some broad agreement as to a non-partisan aim, (reducing racism in our rules), court admonishment of political gerrymandering can have no broad legitimacy in the eyes of the electorate. (Ex. Party 1 wins an election. Jubilation, exuberance abound, champagne, etc. etc. Party 2 brings a lawsuit. Court redistricts. Next election, no change in voter turnout, Party 2 wins by a landslide. How does Party 1 feel about their ability to reverse that outcome?)
Also, I don’t buy the idea that the political branches are unable to develop fair and consistent rules without interference from the judiciary. I hear the logic of incumbents being ill-disposed to pass laws which will disfavor incumbents, but it only goes so far. By that logic, the 19th amendment should have never been passed; why would men voluntarily pass a law which gave so much power to women? Or why would whites do the same for blacks? It’s true that naked self-interest is a primary motivator of our elected officials, but it is not their only motivator. (And anyway, even if it was, the public has the power to make an incumbent’s failure to surrender power more politically painful than surrender would be, provided the political will is there).
Looking forward to seeing more of this series.