The Limits of Precedent

John Yoo & Robert Delahunty

Winter 2024

In the last two years, the Supreme Court achieved two of the goals for which the conservative movement had fought for half a century. In June 2023, the Court handed down Students for Fair Admissions v. Harvard, which rejected racial preferences in college admissions. The preceding summer, in Dobbs v. Jackson Women's Health Organization, the Court reversed Roe v. Wade and Planned Parenthood v. Casey, returning control over abortion policy to the states.

Critics allege that in doing so, activists on the Court thrust aside settled case law and destabilized the legal system. President Joe Biden accused the Harvard Court of having "once again walked away from decades of precedent." Justice Sonia Sotomayor's dissent in Harvard, which Biden proceeded to quote, declared the majority had rolled back "decades of precedent and momentous progress." Yet an investigation of the Court's most important decisions in recent years reveals not an outright rejection of stare decisis — the doctrine that calls on courts to adhere to precedent — but well-crafted adjustments in existing law designed to observe established legal and constitutional principles.

Still, in the conservative movement's campaign to revive the original meaning of the Constitution, the question of whether to obey stare decisis has become a critical one. After all, there are times when precedent directly conflicts with the original meaning of the Constitution. In those cases, does an originalist jurist respect stare decisis? Or does he jettison precedent in favor of returning to the Constitution's original meaning?

If the Court wishes to adhere to the rule of law, it must honor the meaning of the Constitution above its own past decisions. The judicial role the Constitution establishes within our republican government demands no less.


Under Chief Justice John Roberts, the Supreme Court has deployed at least three strategies in cases involving constitutional precedents. The first, which can be called "recovery," describes Roberts's majority opinion in Harvard. The second, "reaffirmation," appears in his majority opinion in the student-loan case, Biden v. Nebraska. The third, "rejection," features in Justice Samuel Alito's majority opinion in Dobbs.

President Biden and Justice Sotomayor claimed that the majority in Harvard overturned decades of precedent. Yet it did no such thing; not a single precedent permitting universities to engage in racial preferences in admissions, from Regents of the University of California v. Bakke through Grutter v. Bollinger to Fisher v. University of Texas at Austin, was formally disturbed. Instead, the Court restored the standard of judicial review, known as "strict scrutiny," that those decisions had purported to apply, and infused the standard with its characteristic force and severity.

The 14th Amendment to the Constitution commits state governments to the ideal of civil equality of the races before the law. But for about a century, the promise of a color-blind legal system was more honored in the breach than in the observance. In 1896, to its lasting shame, the Court upheld Jim Crow racial segregation in Plessy v. Ferguson, with only one justice — John Harlan of Kentucky, a former slave owner who had fought for the Union in the Civil War — dissenting. Harlan alone grasped the true meaning of the 14th Amendment: "Our Constitution is color-blind," he declared, "and neither knows nor tolerates classes among citizens."

Nearly 50 years later, the Court began to develop a strict-scrutiny standard for race-based policies. For any racial classification to survive, the government would have to demonstrate: a) that the classification serves a "compelling" purpose, and b) that the use of race is "narrowly tailored" to that purpose (i.e., that no feasible race-neutral alternative exists). So inimical is this standard to racial classifications that, for many years, it was "strict in theory but fatal in fact" to any law or practice to which it was applied.

Strict scrutiny played a crucial role during the civil-rights movement as the Court dismantled the structure of legalized segregation. Puzzlingly, in Brown v. Board of Education — the judicial centerpiece of the civil-rights era — the Court did not invoke strict scrutiny. But in other cases, the justices applied the rule to attack Southern resistance to Brown. By the end of the 20th century, the Court had gone beyond the civil-rights movement to uproot the use of race by government in employment, contracting, law enforcement, and virtually all other operations.

Yet despite its near universal application of the color-blindness principle to strike down race-based classifications, the Court permitted a solitary exception for post-secondary institutions. In Regents of the University of California v. Bakke, a fractured Court held that the 14th Amendment permitted college and university admissions committees to take race into consideration when reviewing student applications. Justice Lewis Powell, who wrote the opinion that was later upheld in Grutter v. Bollinger, defended the Court's departure from past practice by finding that the state's interest in promoting diversity in college and university classrooms is sufficiently compelling to survive strict scrutiny.

For the following 45 years, the use of racial preferences in college admissions stood as the lone exception to "strict in theory but fatal in fact." The Harvard decision, written by Chief Justice Roberts, eliminated that exception.

Roberts's reasons for reversing the outcomes of Bakke, Grutter, and Fisher did not rest so much on disagreements with past courts as on growing distrust of what universities had told them. In those three cases, the Court had believed universities' claim that employing racial preferences in admissions was necessary to improve the quality of higher education. But when universities began rejecting large numbers of well-qualified candidates on racial grounds and selecting instead students of other races with significantly weaker qualifications, the justices demanded that the schools prove, in some empirical way, that racial diversity was producing educational benefits. No such evidence was forthcoming.

The Court had also relied on universities' promises that their use of race would be limited — that race would be a marginal "plus" factor in an applicant's file. But the universities in Harvard argued that prohibiting consideration of race would lead to a dramatic decline in the number of black and Latino students admitted. Their own data revealed an unconstitutional quota system at work. Analyzing these claims with the severity they were due under strict scrutiny, the 6-3 majority had no difficulty finding that the universities were relying on race far more extensively than they had admitted.

The Roberts Court, therefore, cannot fairly be charged with overturning "decades of precedent" in Harvard; it was the Court's decisions upholding racial preferences that had departed from precedent. The Harvard majority was simply recovering the proper application of strict scrutiny so that it was indeed "strict in theory but fatal in fact."


In vindicating Justice Harlan's vision of a color-blind Constitution, the Harvard Court did not overturn deeply entrenched precedents. Instead, it brought a handful of outlier cases into alignment with a central theme of its jurisprudence. Chief Justice Roberts applied a related strategy — one that might be called "reaffirmation" — in another landmark case decided last year.

In Biden v. Nebraska, the Supreme Court invalidated the Biden administration's plan to forgive $430 billion in student-loan debt owed to the United States. The administration had claimed it had the authority to do so under the 2003 HEROES Act, a statute enacted during the Second Gulf War. That statute allowed the Department of Education, in the event of a national emergency, to temporarily freeze federal student-loan repayments by soldiers and first responders to ensure that they would not suffer financially while deployed. Nearly two decades later, the Biden administration seized on the statute to justify debt forgiveness on a massive and unprecedented scale. It found the necessary "emergency" ready at hand: the waning Covid-19 pandemic (which, the following month, the president would declare over).

Predictably, the Court found that the HEROES Act did not allow the executive branch to unilaterally cancel almost half a trillion dollars in student-loan debt. The Court drew on its "major questions" doctrine — which preserves the Constitution's assignment of legislative power to Congress — to reach this decision. Writing for the majority, Roberts held that while Congress may delegate highly consequential policy choices to the agencies, the major-questions doctrine requires that Congress make a "clear statement" of its intent to allow agencies to decide major social or economic questions. The "economic and political significance" of the loan-forgiveness program, Roberts observed, was "staggering by any measure"; such momentous policy decisions must be made, or at least clearly authorized, by Congress.

Justice Elena Kagan's dissent took sharp issue with what she called the majority's "made-up" major-questions doctrine. But her claim that the Court's new limits on legislative delegation somehow depart from precedent is mistaken. As both the chief justice and Justice Amy Coney Barrett (in her concurrence) pointed out, the name of the doctrine may be new, but the substance is not; the doctrine's requirement of a "clear statement" of Congress's intent to delegate authority to the other branches stretches back at least 40 years. By naming and applying the rule, the Court was not creating a doctrine out of whole cloth; it was simply clarifying and reaffirming its own precedent.


As outlined above, the Roberts Court has not tossed out precedent in several instances; it has instead brought case law into better alignment with the Constitution's text and basic constitutional principles. Yet in certain cases, the Court's originalist jurists have undeniably discarded long-standing precedent on fundamental issues. The question remains as to which precedents warrant more nuanced treatment, as in the strategies of recovery or reaffirmation, and which require outright rejection.

Back in 2020, the Roberts Court addressed the question of whether to respect stare decisis in the context of the Sixth Amendment's unanimous-jury requirement. Writing for the Court, Justice Neil Gorsuch observed that judicial precedent

warrant[s] our deep respect as embodying the considered views of those who have come before. But stare decisis has never been treated as an inexorable command. And the doctrine is at its weakest when we interpret the Constitution because a mistaken judicial interpretation of that supreme law is often practically impossible to correct through other means (internal citations omitted).

Unpacking this statement, we find three things: 1) that precedent is to be deeply respected, but 2) that it is not an "inexorable command" — it only carries a presumption of being correct, and 3) that it is weakest in its constitutional form.

There thus exists a distinction between constitutional and statutory stare decisis: If the Court misconstrues a statute, Congress can always correct the error. But if the Court makes a mistake in interpreting the Constitution, the only political fix is passing a constitutional amendment. That can be done, of course: The 14th Amendment corrected Dred Scott v. Sandford's grievous errors. But because an amendment usually requires approval by two-thirds of Congress and three-quarters of the state legislatures, practical considerations compel us to look to the Court itself to correct most of its constitutional blunders.

In Dobbs, originalism and stare decisis clashed head on — and originalism won. The Dobbs Court rejected almost 50 years of precedent stemming from Roe — a precedent that, although substantially reworked in Casey, had survived repeated challenges. If the current Court has garnered a reputation for casually overturning precedent, it is chiefly because of Dobbs.

Stare decisis loomed large in Dobbs because respect for precedent was Roe's last line of defense. Roe's reasoning was notoriously weak; the Casey Court didn't even attempt to rehabilitate the precedent on substantive, doctrinal grounds. If anything, the Casey plurality implicitly admitted that Roe was wrong as an original matter, and instead affirmed core parts of the decision solely on the basis of stare decisis.

Given Roe's glaring lack of constitutional rationale, the question for the originalist justices on the Court was whether stare decisis required them to adhere to precedent for precedent's sake. Writing for the majority in Dobbs, Justice Alito applied the stare decisis test that the Casey plurality had used in 1992 to uphold Roe, but this time to strike it down.

In Casey, the plurality listed the usual grounds for overturning precedent, which include the quality of the decision's "reasoning," its "workability," "subsequent constitutional developments" that have occurred since the case was decided, and "reliance interests." Respect for stare decisis usually turns on the fourth factor — indeed, in affirming Roe's central holding, the Casey Court placed great weight on such interests.

When the Supreme Court decides a constitutional question, behavior may have to change — often dramatically. If the Court announces a rule of, say, criminal procedure, police officers, prosecutors, juries, and judges have to conform to it — they rely on the Court's ruling to guide their behavior, and thus have an interest in keeping the rule in place. If the Court later reverses that rule, widespread or long-standing patterns of behavior must change. The transition costs may well prove high. Rapid shifts may also undermine legal stability and predictability.

Reliance interests traditionally arise and deserve protection where advanced planning is necessary. They therefore cannot hinge on unplanned acts. The Casey Court conceded that the termination of an unplanned pregnancy was not a planned act; it further admitted that "reproductive planning could take virtually immediate account of any sudden restoration of state authority to ban abortions." In other words, if Roe were overturned and a state banned abortions, residents could immediately abstain from having unprotected sex if they wished to avoid becoming pregnant.

At this point, however, Casey diverged from the Court's traditional view of stare decisis by inventing what Dobbs characterized as "a more intangible form of reliance": women's ability "to participate equally in the economic and social life of the Nation," which it claimed had been "facilitated by their ability to control their reproductive lives." Overruling Roe, Casey's authors maintained, would prove too disruptive "for people who [had] ordered their thinking and living around" abortion being legal.

In Dobbs, Justice Alito countered this claim by noting that those were not the kind of reliance interests the pre-Casey Court had protected — which paradigmatically included property and contract rights. The new type of reliance interest that the Casey authors invented had clearly been customized to fit their desire to reaffirm Roe. By rejecting it and instead finding that overruling Roe would not upset "conventional, concrete reliance interests," Dobbs was able to bring the Court's abortion jurisprudence in line with its precedent on stare decisis.


The originalist justices of the Court agreed that either all or part of Roe had to be discarded. Yet among those same justices, no consensus obtains as to the elements of stare decisis or how much weight to give the doctrine.

In Dobbs, Justice Alito applied Casey's four-factor stare decisis test to conclude that Roe should be overturned. Similarly, Justice Brett Kavanaugh's partial concurrence in Ramos v. Louisiana (2020) distilled stare decisis down to three factors: whether the precedent was not merely wrong, but "egregiously wrong"; whether the precedent has caused "significant negative jurisprudential or real-world consequences"; and whether overruling the precedent would "unduly upset legitimate reliance interests."

These and similar factor-based formulations are unlikely to achieve greater consistency and neutrality in applying stare decisis, however. Any factor that a test takes into account will ultimately require judges to make qualitative, normative, or even subjective judgments. In terms of Kavanaugh's triad, no justice has ever explained how to tell when a precedent is not merely "wrong," but "egregiously" so. Nor has the Court defined when overruling precedent would "unduly" upset reliance interests. It is also unclear what amounts to sufficiently "significant" negative consequences.

Kavanaugh's three considerations, like Casey's four-factor test, offer a flexible standard, not a bright-line rule, for evaluating claims of stare decisis. It is chimerical to think that these or any other factor-based standard, regardless of the factors listed therein, could eliminate outcome-driven uses of the doctrine.

Chief Justice Roberts favors an alternative strategy for dealing with unfavorable precedents: He prefers not to overrule them, but to limit them to the point of near extinction. In Dobbs for instance, Roberts argued against directly overturning the basic right to abortion. "I agree with the Court that the viability line established by Roe and Casey should be discarded under a straightforward stare decisis analysis," he wrote in his concurrence. But "there is a clear path to deciding this case correctly without overruling Roe all the way down to the studs." In Roberts's view, the Court should decide a case with a rule no broader than the facts before it, and leave the rest of the Court's precedent untouched.

Roberts defended his view on stare decisis in the 2010 case Citizens United v. Federal Election Commission. There, he maintained that the doctrine's "greatest purpose is to serve a constitutional ideal — the rule of law." But he failed to elaborate on why adherence to a decision without a foundation in the constitutional text, structure, or history advances the rule of law. If Dobbs had affirmed Roe purely on the basis of stare decisis, the Court would have been denying rather than upholding the law. Such a ruling would have given a veneer of legality to what was originally a lawless act — an exercise of raw power, not of reason and judgment in constitutional interpretation.

In a concurrence in the 2019 case Gamble v. United States, Justice Clarence Thomas offered a third, seemingly provocative view of stare decisis. He began by tracing the origin of the doctrine to English common law — a body of unwritten rules grounded in custom and the laws of nature. Since the common law did not exist in writing, a common-law judge could not look to a text to interpret and apply the law; instead, he had to "discover" the law given by custom or nature using reason, and to declare it as such. And the principal, most authoritative evidence common-law judges looked to when affirming and applying the law was judicial precedent.

Stare decisis thus played a vital role in a common-law system: As Thomas put it, common-law judges were "expected to adhere to precedents because they embodied the very law the judges were bound to apply" (emphasis added).

Unlike the English common-law constitution, America's Constitution is written. This means that the text of the document is the law. Precedent may furnish evidence of the meaning of the Constitution's text, but it is not necessarily conclusive, and certainly not the most authoritative, evidence: That honor belongs to the text itself.

Thus, when judicial precedent conflicts with the Constitution's text, Justice Thomas believes that federal courts have a duty to follow the text. Tracking Marbury v. Madison, he asserted that the "judicial Power" is the power to say what the law is, not what the law should say. Adhering to a "demonstrably incorrect" decision, he explained, "is tantamount to making law," which usurps Congress's lawmaking power and is thus unconstitutional.

Thomas also pointed out that adhering to flawed precedent disregards the Constitution's Supremacy Clause. That clause states: "This Constitution, and the Laws of the United States which shall be made in Pursuance thereof; and all Treaties...shall be the supreme Law of the Land." The Constitution thus imposes a duty on the Court to obey the Constitution, not its own precedents — after all, judges take an oath to support "this Constitution," and consistency with "this Constitution" is not the same as consistency with "this Court's precedents."


Stare decisis need not pose an insurmountable obstacle to a proper understanding of originalism. A strict adherence to stare decisis, which no justice or scholar advocates, would require otherwise. But as Justice Thomas reminds us, the Constitution gives courts no power to elevate their own past decisions over the text itself. Precedent may represent the justices' accumulated wisdom, but that knowledge can be wrong — as it was in Dred Scott and Plessy v. Ferguson, among others. If the Court majority believes that the original understanding of the Constitution conflicts with precedent, the confines of the judicial role require it to enforce the former and discard — however gently — the latter.

John Yoo is a non-resident senior fellow at the American Enterprise Institute; Emanuel S. Heller Professor of Law at the University of California, Berkeley; and a visiting fellow at the Hoover Institution.

Robert Delahunty is a Washington fellow at the Claremont Institute's Center for the American Way of Life. They are the co-authors of The Politically Incorrect Guide to the Supreme Court (Regnery Publishing, 2023).


from the


A weekly newsletter with free essays from past issues of National Affairs and The Public Interest that shed light on the week's pressing issues.


to your National Affairs subscriber account.

Already a subscriber? Activate your account.


Unlimited access to intelligent essays on the nation’s affairs.

Subscribe to National Affairs.