If Roe Wasn't Safe After 50 Years, What Is?

Justice Thomas’s concurrence in Dobbs did more than celebrate the fall of Roe: it pointed the Court’s spotlight at the rest of the substantive-due-process canon—contraception (Griswold), same-sex intimacy (Lawrence), and marriage equality (Obergefell).

The deeper point is methodological. Dobbs didn't just overrule an abortion precedent. It legitimized a new permission structure for overruling: declare an older case "egregiously wrong," discount reliance as "intangible," and tether rights to a selectively applied "history and tradition" baseline.

The Scale of the Reversal

Since 2018, the Court has formally or effectively overruled major precedents spanning roughly 200 years of combined legal stability:

Case	Year	Overruled	Duration	Subject
Janus v. AFSCME	2018	Abood	41 years	Union agency fees
Dobbs v. Jackson	2022	Roe, Casey	49 years	Abortion rights
Kennedy v. Bremerton	2022	Lemon test	51 years	Church-state separation
Students for Fair Admissions	2023	Grutter	45 years	Affirmative action
Loper Bright v. Raimondo	2024	Chevron	40 years	Agency deference (cited 17,000+ times)

From 1810 through 2020, the Supreme Court overruled precedent in constitutional cases approximately 145 times—barely 0.5% of its 25,544 opinions. The pattern since 2018 marks a dramatic acceleration.

Public confidence has collapsed to historic lows. The General Social Survey found only 18% of Americans had a "great deal" of confidence in the Court—the lowest point in the survey's 50-year history. The partisan gap has widened to an unprecedented 64 points: 75% of Republicans approve while only 11% of Democrats do.

The Thomas Concurrence

The Dobbs majority tried to build a firewall—abortion is “unique,” the holding is limited, don’t worry about other precedents.

Thomas didn’t play along. He argued that substantive due process itself is “demonstrably erroneous,” and urged reconsideration of the Court’s other SDP precedents.

The conspicuous omission—Loving v. Virginia—only reinforces the point: the doctrine isn’t being applied with a neutral limiting principle.

Dobbs as a "Precedent About Precedent"

One reason Dobbs reverberates beyond abortion is that it collides with Casey on the very question of stare decisis. As the Harvard Law Review observed, Dobbs "overruled not only the substantive holding of Casey but also its approach to stare decisis"—abandoning Casey as "a precedent about precedent."

Casey treated reliance broadly—women and families organizing intimate and economic lives around a stable constitutional rule. It explicitly recognized that "the ability of women to participate equally in the economic and social life of the Nation has been facilitated by their ability to control their reproductive lives."

Dobbs reframed that as "generalized" or "intangible" and therefore unworthy of protection. The majority distinguished between "concrete" reliance (property and contract interests) that merits protection and "intangible" reliance (organizing one's reproductive life) that does not. This redefinition effectively eliminates reliance as a barrier to overturning decisions related to privacy, speech, or equal protection.

Chief Justice Roberts concurred only in upholding Mississippi's 15-week ban, refusing to join the full overruling. He advocated a "measured course"—discarding only the viability line while preserving a "reasonable opportunity" to obtain an abortion. Roberts warned that "the Court's decision to overrule Roe and Casey is a serious jolt to the legal system" and criticized the majority for "ripping out walls all the way down to the studs" when something narrower would suffice.

That shift turns a doctrine meant to make overruling hard into one that can be satisfied by the majority's moral certainty.

Why the Firewall Won’t Hold

If the legal test is “deeply rooted in the Nation’s history and tradition” as of 1868, then the rights Thomas singled out are structurally exposed.

That’s not speculation—it’s the logic of the method. Under this framework, the question is not “Are these rights popular now?” but “Were these liberties recognized at the chosen historical baseline?”

And there's a second vulnerability: the Court applies history inconsistently. The "tradition" filter can be widened or narrowed depending on which result the majority wants. The same method can justify either expansion or contraction with equal confidence.

The Method is Applied Selectively

The contrast between Dobbs and New York State Rifle & Pistol Association v. Bruen (2022)—decided within days of each other, both invoking "history and tradition"—reveals the methodology's malleability.

In Dobbs, Justice Alito applied a narrow historical inquiry, finding no tradition supporting abortion rights and treating territorial abortion restrictions as significant evidence. In Bruen, Justice Thomas applied a broader historical-analogical approach to gun rights, dismissing territorial gun regulations because of low population density—the same characteristic of territories whose abortion restrictions Dobbs credited.

The Columbia Law Review documented how the Court "struck past laws and practices out of this nation's historical tradition by applying a shadow strict scrutiny review" in Bruen while "incorporat[ing] past laws and practices into the historical tradition by applying a shadow rational basis review" in Dobbs. In Bruen, Justice Thomas excluded racially discriminatory gun laws from the historical tradition because they were unconstitutional. In Dobbs, Justice Alito incorporated abortion restrictions despite historians' arguments that they were motivated by sex-role stereotyping.

As Justice Sotomayor observed: "History matters to this Court only when it is convenient."

The "Major Questions Doctrine" operates similarly. Articulated in West Virginia v. EPA (2022), it holds that agencies must have "clear congressional authorization" before making decisions of "vast economic and political significance." But the criteria for what constitutes a "major question" remain undefined—allowing courts to second-guess any regulation they disfavor while claiming neutral methodology.

Reliance Doesn't Protect You Anymore

Reliance was once the stabilizer. If people have built real lives and institutions around a rule, the Court should hesitate before pulling the floor out.

Dobbs narrowed what counts. If reliance must look like property or contract reliance, then most personal liberties become easy to revoke—because their reliance is human and social, not commercial.

That’s why the question “but people relied on Obergefell” no longer functions as a brake. Once reliance is narrowed this way, rights-based precedents become easier to discard.

Stealth Erosion: Rights Can Die Before They're Overruled

Overturning isn't the only mechanism. A right can be hollowed out by allowing procedural end-runs to stand—especially through the "shadow docket."

The statistics are stark. During the Bush and Obama administrations combined (16 years), the Justice Department filed only 8 emergency applications to the Supreme Court. During Trump's first term alone (4 years), DOJ filed 41 applications, with 28 granted—more than 20 times the historical rate.

Texas's SB8 exemplifies how procedural manipulation can nullify constitutional rights. The law banned abortions after approximately six weeks but was deliberately structured to evade judicial review: no state officials enforce it, only private citizens who can sue providers for $10,000 or more. When providers sought emergency relief, the majority issued a one-paragraph unsigned order at 11:58 PM on September 1, 2021, allowing the law to take effect while citing "complex and novel antecedent procedural questions" it refused to resolve.

Justice Sotomayor's dissent was devastating: "Without full briefing or argument, and after less than 72 hours' thought, this Court greenlights the operation of Texas's patently unconstitutional law... The Court thus rewards Texas's scheme to insulate its law from judicial review by deputizing private parties to carry out unconstitutional restrictions on the State's behalf."

Just months earlier, the same 5-4 majority had issued emergency injunctions protecting religious liberty rights while ignoring procedural obstacles. In the abortion context, it invoked open procedural questions as reasons not to act.

Justice Jackson has characterized the Court's emergency docket approach as "Calvinball"—the game from Calvin and Hobbes whose only rule is that there are no fixed rules. In a 2025 dissent, she wrote: "We seem to have two [rules]: that one, and this Administration always wins."

That's a template: you don't always need a merits opinion to change constitutional reality.

The Super-Precedent Illusion

"Super-precedent" turned out to be more branding than protection.

The concept gained prominence during John Roberts's 2005 confirmation hearing, when Senator Arlen Specter asked whether Roe had achieved "super stare decisis." Constitutional scholar Michael Gerhardt defined super-precedents as "those constitutional decisions in which public institutions have heavily invested, repeatedly relied, and consistently supported over a significant period of time."

But if the test is continued political contestation, Roe never qualified—people continued pushing for its reversal. As Justice Barrett observed at her confirmation hearing, "descriptively it does mean that it's not a case that everyone has accepted."

The confirmation hearing statements of justices who later voted to overturn Roe have generated fierce controversy:

Justice Kavanaugh called Roe "settled as a precedent of the Supreme Court" and emphasized that Casey was "precedent on precedent"—making it doubly established
Justice Gorsuch described precedent as "the anchor of law" and "the starting place for a judge"
Justice Barrett declined to "grade precedent" but stated she had "no agenda to try to overrule Casey"

None explicitly promised to preserve Roe, and legal experts note they maintained "escape hatches" through qualifying language. But Senators Susan Collins and Joe Manchin publicly stated they felt misled about the justices' commitment to precedent.

If a right can be reaffirmed, embedded for half a century, and still be erased once personnel change, then "settled" means only "until we have the votes."

What Can Protect Rights Now?

If constitutional protection is conditional on Court composition, then durable protection shifts outward:

Federal statutes can help—but can also be repealed.
State constitutional law can insulate some rights in some states.
Structural reforms (term limits, enforceable ethics rules) are often proposed as legitimacy repair tools, though reforms may not be enough once the perception of politicization hardens.

The point is not that nothing can be done—but that the old assumption ("the Court said it, so it's secure") no longer holds.

The Legitimacy Crisis

The justices themselves have publicly clashed over whether the institution retains legitimacy.

Justice Kagan has repeatedly warned that "if, over time, the court loses all connection with the public and public sentiment, that's a dangerous thing for democracy." She has stated that judges "create legitimacy problems for themselves... when they stray into places where it looks like they're an extension of the political process." When a new justice arrives and "all of a sudden everything is up for grabs," Kagan argues, "people have a right to say... that just seems as though people with one set of policy views are replacing another."

Justice Alito has responded defiantly. In a July 2023 Wall Street Journal interview, he declared: "I marvel at all the nonsense that has been written about me in the last year." He controversially compared critics of the current Court to segregationists who defied Brown v. Board of Education, stating: "If we're viewed as illegitimate, then disregard of our decisions becomes more acceptable and more popular." Most provocatively, Alito claimed: "Congress did not create the Supreme Court... No provision in the Constitution gives them the authority to regulate the Supreme Court—period"—a position legal scholars call "belied by 234 years of practice."

The legitimacy crisis is compounded by ethics controversies. ProPublica documented that Justice Thomas accepted luxury trips from Dallas billionaire Harlan Crow "virtually every year" for over two decades without disclosure—international cruises on Crow's superyacht, flights on his private jet, vacations at his Adirondacks resort. Crow purchased Thomas's mother's house and paid roughly $100,000 in private school tuition for a relative Thomas was raising. Justice Alito accepted a private jet flight to an Alaska fishing trip from hedge fund billionaire Paul Singer, who had business before the Court—a trip organized by Leonard Leo, the Federalist Society's former executive vice president.

The Court adopted a formal Code of Conduct in 2023—but with no external enforcement mechanism, leaving justices to police themselves.

What Remains

The honest answer is the one this era keeps returning to: we don't know which rights survive, because the limiting principle is no longer doctrinal—it's political and personnel-based.

The Dobbs dissent stated this explicitly: the decision exists "for one reason and one reason only: because the composition of this Court has changed." Nothing in law or fact shifted between Casey's reaffirmation in 1992 and Dobbs's reversal in 2022 except the identity of the justices. If that is sufficient to overturn 49 years of precedent, stare decisis provides no meaningful constraint—only a delay until the composition changes.

Legal scholars increasingly describe stare decisis as operating as a "one-way ratchet"—constraining liberal majorities from overturning conservative precedents while providing no limit on conservative majorities. The UK Supreme Court's 1966 Practice Statement, by contrast, permits departure from prior decisions "when it appears right to do so" but directs using this power "sparingly." Between 1966 and 2009, this power was explicitly invoked in only approximately 21 cases. The Canadian Supreme Court has overruled itself over 500 times—but consistently in the direction of expanding rights.

If Roe wasn't safe after 50 years, the rest of the substantive-due-process edifice is not "settled." It's contingent—on who sits on the Court, and how aggressively they are willing to use the tools Dobbs normalized. The constitutional order depends on shared premises that once transcended partisan divides—that precedent constrains, that the Court is not simply a super-legislature, that law is something more than power dressed in robes. When those premises collapse, what remains is not constitutional democracy but its simulation.

This is the ninth article in a series examining democratic decline. The next article examines what happens when the executive branch simply ignores court orders—reducing judicial rulings to suggestions.