# Why Courts Sometimes Undermine Democracy *A political scientist set out to understand why democracies are in danger. The answer led him somewhere he didn’t expect — the courts.* October 29, 2026 ![Close-up of scales of justice with a broken chain](/sites/g/files/omnuum8171/files/styles/hwp_16_9__480x270/public/2026-10/justice%20scale-nellie-adamyan-unsplash.jpg?h=510e58a8&itok=UbtR2Lpr) *by Courtney Hayes* ## **Istanbul, 2016** On a warm summer night in 2016, Andrew O’Donohue was a twenty-something intern at the US Consulate in Istanbul, when tanks blockaded a nearby bridge. The official explanation — a bomb threat — didn’t make sense. ![Andrew O'Donohue](/sites/g/files/omnuum8171/files/styles/hwp_1_1__360x360_scale/public/2026-05/Andrew%20ODonohue%202026.jpg?itok=mf-lt-H5) *Andrew O'Donohue* is a Postdoctoral Fellow at Princeton University and a Fellow at the Carnegie Endowment for International Peace. As a graduate student affiliate at IQSS, he received IQSS funding for his doctoral research.Then his twin sister texted from New York. CNN was reporting a coup. As O’Donohue made his way to safety, citizens used public buses to block roads and resist the coup attempt. Gunshots cracked from the street. Late into the night, low-flying jets rattled the chandelier in the house where he took shelter. When the coup failed, he saw tens of thousands of Turks celebrating in the streets. But by the time he returned to work at the US Consulate, human rights organizations were already calling in reports of torture. In the weeks that followed, Turkey’s government abused emergency powers to imprison opponents, purge thousands of judges, and remove sitting members of the Constitutional Court. The elected government had won; democracy, in some sense, had survived, but was rapidly eroding. That summer set the course for the next decade of O’Donohue’s life. He returned to Harvard for a PhD with two questions driving him: why do democracies fail, and what can we do to protect them? What he would discover is that democracy's most trusted guardian — the courts — is sometimes the weapon used to destroy it. ## **From Coups to Courts** When O’Donohue returned to Turkey for his PhD fieldwork in 2022, he arrived carrying an assumption most Americans share: Judges are democracy’s referees, a check on whoever holds power. But the Turkish judges and lawyers he interviewed told him it was “crazy” to think of courts as defenders of democracy. In Turkey, one interviewee explained, the judiciary is “like a gun — the weapon used to control the media, opposition, and civil society.” American veneration of the courts runs deep. Alexander Hamilton called the judiciary “an excellent barrier to the encroachments and oppressions” of elected leaders — a line that appears often in civics textbooks. But what most textbooks leave out is the warning that follows: the judiciary is also “in continual jeopardy of being overpowered, awed, or influenced by its coordinate branches.” The slogan got passed down, but Hamilton’s warning got dropped. O'Donohue's research explores why courts [defend](https://www.journalofdemocracy.org/articles/how-courts-survive-democratic-backsliding/) democracy — or [*undermine*](https://www.journalofdemocracy.org/articles/how-courts-undermine-democracy/) it instead. Democracies today rarely die through military coups. They erode through gradual changes to constitutions, statutes, and courts that can subvert elections, strip rights, and enable expansions of executive power. More surprising still, explains O'Donohue, courts can undermine democracy even when fully independent of an elected government. But a court doesn't betray democracy on its own; someone takes hold of it. O'Donohue calls this “court capture.” And during his time in Turkey, Andrew began to realize, “there was an interesting story to tell about how courts can be captured by political parties, ethnic groups, militaries, business interests, and many other actors who can weaponize courts against democracy.” What makes a court vulnerable to capture, O'Donohue argues, comes down to three factors: - **Judicial power-sharing** — whether choosing judges forces rival parties to compromise, or lets one side pack the bench. - **The political establishment** — whether elites, parties, and institutions are strong enough to block an attack on the courts. - **Societal mobilization** — whether ordinary citizens will turn out to defend them. Power-sharing is the foundation — it determines how easy or difficult it is for political actors to capture the courts. The other two factors matter most when political leaders attempt to defy or disempower the judiciary. Turkey was the first place O'Donohue saw the foundation give way. ## **A Court Falls** ![Turkish President Recep Tayyip Erdoğan](/sites/g/files/omnuum8171/files/styles/hwp_1_1__360x360_scale/public/2026-10/Turkish_President_Recep_Tayyip_Erdogan.jpg?itok=W4CxvpJU) *Recep Tayyip Erdoğan*In Turkey, court capture came gradually — not in a single seizure but over years of erosion. And when a religious conservative named Recep Tayyip Erdoğan rose to power as Prime Minister in 2003, he ran headlong into a judiciary built to stop him. The bench was controlled by the secular elite, and Erdoğan needed a way to turn that weapon against them. His opportunity arrived in the form of a long-standing cultural flashpoint: a ban on Muslim women wearing the Islamic headscarf at universities. Erdoğan proposed a constitutional amendment lifting the ban — engineered to pick a fight with Turkey’s staunchly secular judiciary. Erdoğan’s policy was wildly popular: Three-quarters of Turkish citizens supported repealing the ban, and 1 in 5 women said they’d rather leave university than remove it. But in a 9–2 decision, the Constitutional Court struck down the amendment, ruling it violated bedrock secular principles. The ruling was the opening Erdoğan needed. He cast himself as the true democrat and the court as the enemy of the people's will. It’s a tactic O'Donohue calls “[court-baiting](https://www.theatlantic.com/ideas/archive/2025/05/trump-court-autocrat-popular-issues/682797/):” a leader provokes the courts into striking down a popular policy, then uses the ruling as political ammunition. “That decision gave me a personal appreciation of how Erdoğan skillfully mobilized public opinion against the judiciary,” O'Donohue explains. Then Erdoğan pressed his advantage. In a 2010 referendum, Erdoğan pushed through changes that expanded the Constitutional Court and gave his government control over its membership. The secular establishment that had ruled the bench was swept aside. With the courts no longer able to check him, Erdoğan jailed journalists and opposition politicians, took control of the universities, and stripped the central bank of its independence. "That is how Erdoğan, in the name of democracy, crushed the courts," O'Donohue explains. ## **Inside the Numbers** To test his observations in Turkey, O’Donohue had to measure something slippery: the slow capture of a country’s institutions. Courts, he realized, were the place to look. Judges cast votes and publish decisions, case after case, year after year. And buried in that record is a measurable fingerprint of capture even before Erdoğan took control of the bench in 2010. O’Donohue web-scraped every constitutional decision issued by the Turkish Constitutional Court since its founding in 1962 — 5,294 cases in total. That original dataset enabled him to test his theory that the absence of judicial power-sharing enabled court capture. In Turkey until 2010, the nonelected president wielded vast power over selecting judges, whereas the elected parliament and prime minister had no say at all. That absence of judicial power-sharing was associated with a clear pattern of judicial bias. From 1997 to 2007, when Turkey’s president used the courts to challenge the elected government’s legislation, a startling pattern emerged: the president won 83 percent of the time. When opposition parties brought their own challenges, they won just 59 percent. The same court, but with very different outcomes, depending on who was asking. ## **Why Israel Held** ![Israeli Supreme Court](/sites/g/files/omnuum8171/files/styles/hwp_1_1__360x360_scale/public/2026-10/500px-Israeli_supreme_court_building_nightshot.jpeg?itok=dTXBzYaB) Supreme Court of IsraelTurkey showed how a court falls. To understand why others hold, O'Donohue's research turned to a comparative case. Israel's longest-serving prime minister, Benjamin Netanyahu, had spent years attempting to weaken a Supreme Court that kept ruling against him. The atmosphere was ripe for capture — and yet the court held. The difference was O'Donohue’s three factors, running the other way. Israel's judicial selection committee forces compromise; no single actor can install loyalists, because the committee required an unusually high super-majority to appoint new Supreme Court justices. That power-sharing institution helped protect judicial independence because it was complemented by an unwritten norm of compromise. O'Donohue calls this informal institution the “Noah's Ark” strategy: when the opposing sides couldn't agree on a judge, they appointed them two by two, one from each camp. "In Israel, the prime minister wasn’t able to only appoint judges closely aligned with him," O'Donohue explains. "He also needed to appoint judges aligned with the other side." Elected governments could influence the court, but they couldn't capture it. Then came the stress test. In 2022, Netanyahu formed a more extreme coalition, and the second factor — the power of the political establishment — gave way. The political elites that had long blocked attacks on the courts lost their grip, and in January 2023, Netanyahu’s government announced a sweeping overhaul of the judiciary. The attack encountered widespread resistance, as the third factor answered: societal mobilization. Military reservists, tech leaders, trade unions, lawyers, and ordinary citizens sustained months of mass protests; even Netanyahu’s own defense minister publicly opposed the overhaul. The fight isn’t over, but Israel's court held where Turkey’s fell. And in the lead-up to Israel’s 2026 elections, O’Donohue [observes](https://www.journalofdemocracy.org/online-exclusive/israels-slide-into-competitive-authoritarianism/), the Supreme Court has repeatedly ruled against the Netanyahu government. ## **Beyond Two Countries** Two cases, however vivid, can’t prove a global pattern. To find out whether Turkey and Israel were exceptions or examples, O’Donohue needed something no one had built: a record of how the world’s democracies choose their judges. This is where support from the Institute for Quantitative Social Science proved pivotal. With a grant from IQSS, O’Donohue and a research assistant spent months compiling the formal and informal rules governing judicial selection in 139 countries from 1990 to 2025 — more than 4,000 years of institutional detail that no existing dataset had captured. The case studies had shown the three factors in action. The dataset let O'Donohue test the most important of them — power-sharing — at scale, and the results pointed to a clear pattern. Countries that require a supermajority to appoint high-court judges, forcing rival parties to agree, have measurably more independent courts, and those same rules act as a buffer against populist erosion. What Turkey and Israel suggested, the data confirmed: power-sharing was doing the protective work globally. Yet as O’Donohue’s research reveals, those power-dispersing institutions are also double-edged. In highly polarized societies, when informal institutions of compromise break down, those same super-majority requirements that aim to compel compromise could also enable obstruction and gridlock. ## **The American Question** ![Seal of the Supreme Court of the United States](/sites/g/files/omnuum8171/files/styles/hwp_1_1__360x360_scale/public/2026-10/pngegg.png?itok=PCDz8SnM) When O’Donohue applied the same metrics to his own country, the results were alarming. On an [independent expert measure](https://www.v-dem.net/) of high-court independence from the Varieties of Democracy Project, the United States now ranks 39th of 87 democracies, behind Colombia, Brazil, Kenya, and South Africa. By O'Donohue's own measure of judicial power-sharing, the US is an outlier among democracies globally because its institutions for selecting judges require little power-sharing. "Many Americans are used to thinking about our Supreme Court as a global model of judicial independence," O'Donohue says. "My research turns that thinking on its head." The American judiciary, he argues, has much to learn from countries that now outscore it. O'Donohue points to three institutions that limit partisan control over judges: a supermajority requirement to confirm them, a selection process shared among multiple actors, and term or age limits. Israel has all three. Turkey had one. The United States effectively has none. But understanding those vulnerabilities, O'Donohue argues, offers a road map for strengthening the U.S. Supreme Court: restore the supermajority for confirmations, add 18-year term limits, and limit each president to two appointments per term. And as O'Donohue takes his findings public — in [articles](https://www.theatlantic.com/author/andrew-odonohue/) for *The Atlantic*, [testimony](https://humanrightscommission.house.gov/events/hearings/can-turkey-find-its-way-back-freedom-authoritarian-consolidation-versus-defense) before the U.S. Congress, and his [speech](https://www.youtube.com/watch?v=28BGGyh8lCA) at Harvard’s 375th Commencement — he tends to end on a note that is less of a prediction than a challenge: "Other countries have shown us divergent paths for our democracy. Which path we take is up to us." Share on:- [ Facebook ](#) - [ Twitter ](#) - [ Linkedin ](#)