Judicial Scrutiny: How Courts Protect Your Constitutional Rights

Last updated 4 weeks ago. Our resources are updated regularly but please keep in mind that links, programs, policies, and contact information do change.

Murgia was a uniformed state police officer, and four months before he was forced to retire he passed his physical: the demanding annual exam officers over 40 had to take, with an electrocardiogram and tests for gastro-intestinal bleeding. By the time his case reached the Supreme Court, nobody disputed that his excellent physical and mental health still rendered him capable of performing the duties of a uniformed officer. He was retired at 50 anyway, and he lost.

He lost because of a decision the Court made before weighing anything about him: age is not a suspect classification, so the retirement age only had to be rationally related to the state’s objective. Once his claim was sorted into that box, his fitness stopped mattering. The state never had to show that 50 was the right number, or that any number was.

That sorting is the whole game, and it is worth understanding before you ever speak to a lawyer. A court asked to strike down a law does not weigh whether the law is wise or cruel. It sorts your claim into one of three tests, and each test fixes in advance who has to prove what, so the sorting usually decides the case before anyone argues the facts.

What the three tests actually demand

The constitutional text behind most of this is one clause. Section 1 of the Fourteenth Amendment says that no state shall deny to any person within its jurisdiction the equal protection of the laws. That is the entire instruction. The three tests are nowhere in it, which is why the argument over whether to keep them never quite dies.

The Congressional Research Service (CRS), the nonpartisan agency that writes legal briefings for members of Congress, sets them out plainly. Most classifications get rational basis review and survive as long as they are reasonably related to a legitimate government purpose.

Race and national origin are suspect classifications, and they, along with certain distinctions touching fundamental rights like the right to vote, draw strict scrutiny, under which a classification must be narrowly tailored to achieve a compelling government interest. In between sit classifications based on sex or illegitimacy, which get intermediate scrutiny.

Read those three formulas side by side and they sound like three settings on one dial. They are not. The difference that decides cases is not the adjective in front of the word scrutiny. It is which side walks into the courtroom already owing an explanation.

Rational basis: the state can win on a reason it never had

Oklahoma once made it unlawful for anyone but a licensed optometrist or ophthalmologist to fit, adjust, adapt, or to in any manner apply lenses, frames or prisms to a person’s face, or to put lenses back into frames, without the written prescriptive authority of a licensed Oklahoma eye doctor. In practice, an optician could not move your existing lenses into a new frame after you sat on the old one. You had to go get a prescription first.

The Supreme Court upheld the law, and the sentence doing the work is worth reading slowly: the legislature might have concluded that a prescription was needed often enough to require one in every case. Might have. The Court was not describing a reason Oklahoma gave. It was supplying one, and a supplied reason is enough.

The Court later said out loud what that implies. A legislative choice is not subject to courtroom factfinding and may be based on rational speculation unsupported by evidence or empirical data. So the state need not put on a witness, produce a study, or show that the problem it worried about is real.

Your side of the ledger is heavier. The law arrives in court bearing a strong presumption of validity, and you have to knock down every conceivable basis that might support it. Take that literally, because courts do: not the reasons the government actually had, not even the reasons it argues in court, but every reason anyone could think up.

Whether that is deference or abdication is the oldest fight in this corner of the law, and it has real casualties. After Hurricane Katrina destroyed the timber income of St. Joseph Abbey, thirty-eight Benedictine monks started selling two models of handmade cypress casket, at $1,500 and $2,000.

Rules issued by the Louisiana Board of Funeral Directors granted funeral homes an exclusive right to sell caskets, and the board moved against the monks. The monks sued; the district court blocked the rules, finding that they deny equal protection and due process of law; and the Fifth Circuit affirmed that judgment.

Clark M. Neily III of the Institute for Justice, which brings cases like that one, calls rational basis “judicial abdication,” “make-believe judging,” the rubber-stamp style of judging, and an empty charade. He argues instead for judicial engagement: real judging in every case, with courts examining the justification the government actually had rather than one the judge invents on its behalf. On that view, a test that lets a court manufacture the state’s reasoning is not really a test.

Strict scrutiny: a compelling reason, and a tight fit

The top tier arrived in an opinion now treated as a byword for judicial failure.

Upholding that conviction, the Court wrote the rule later used against laws like the one it was approving: restrictions curtailing the civil rights of a single racial group are immediately suspect, courts must subject them to the most rigid scrutiny, and while pressing public necessity may sometimes justify such restrictions, racial antagonism never can.

Strict scrutiny asks two things of the government, and the second does most of the work. It must demonstrate a compelling interest, and it must show its action is narrowly tailored to further that interest. Where the interest claimed is remedying discrimination, precedent requires a strong basis in evidence that the remedial action is necessary, which is a long way from rational speculation.

A commendable goal is not automatically a compelling one. In Students for Fair Admissions the Court held that the interests the universities offered, training future leaders, acquiring new knowledge based on diverse outlooks, promoting a robust marketplace of ideas and preparing engaged and productive citizens, were not sufficiently coherent for purposes of strict scrutiny, because no court could measure them or tell when they had been reached.

Even so, the top tier is not an automatic death sentence, and the Court has said as much. In Adarand it moved to dispel the notion that strict scrutiny is strict in theory but fatal in fact, adding that the lingering effects of racial discrimination are an unfortunate reality and that government is not disqualified from acting in response.

Who brings a claim at the top tier has changed, though. Barbara Grutter was a white Michigan resident who applied to the Law School at the University of Michigan in 1996 with a 3.8 grade point average and a 161 score on the LSAT, the law school admissions test; the school put her on a waiting list, then rejected her, and she sued. Because the top tier covers every racial classification whoever it benefits, the plaintiff invoking it is often not the person the doctrine was written for.

That reversal is the heart of the objection from the lawyers who lost Students for Fair Admissions. The NAACP Legal Defense and Educational Fund argued in its friend-of-the-court brief that the version of colorblindness being demanded of judges would require decisionmakers to willfully ignore ongoing racial inequality, and that Brown v. Board of Education was an indictment of racial apartheid rather than a rule against noticing race. On their account, a doctrine written to catch laws that harmed Black Americans now most often catches programs meant to help them.

Why sex stopped one rung short of race

Sharron Frontiero was a lieutenant in the United States Air Force. She applied for the increased quarters allowance and the housing and medical benefits that would automatically have been granted with respect to the wife of a serviceman, this time for her husband Joseph, and was refused because she could not show he depended on her for more than half his support. A man in her uniform had to show nothing at all.

She won, and four Justices, with Brennan writing, would have gone further and held sex an inherently suspect classification. That was one vote short. Powell, joined by the Chief Justice and Blackmun, agreed she had been wronged but thought it inappropriate to decide at this time whether sex is a suspect classification, in part because the Equal Rights Amendment had been submitted to the states for ratification.

So the constitutional rank of sex discrimination turned partly on a pending amendment, and the Court chose to wait for a democratic answer rather than supply a judicial one. By the time the wait was over, the middle tier had been built instead.

Ruth Bader Ginsburg built much of it at the American Civil Liberties Union (ACLU), one case at a time, and, as the organization’s own history notes, a number of the plaintiffs in Ruth’s litigation were men. That was strategy. A line a court can see hurting men as well as women looks like a problem with the line itself, not with the sympathies of the person complaining.

Which is how the standard came to be set in a beer case. Appellant Craig, a male then between 18 and 21 years old, and appellant Whitener, a licensed vendor of 3.2 percent beer, challenged an Oklahoma scheme that sold nonintoxicating 3.2 percent beer to women at 18 but made men wait until 21. The Court held that classifications by gender must serve important governmental objectives and must be substantially related to achievement of those objectives. That sentence is the middle tier.

Then the burden. Deciding the Virginia Military Institute case, the Court said the justification must be genuine, not hypothesized or invented post hoc in response to litigation. Set that against rational basis and the contrast is stark: at the bottom, the court may invent the government’s reason for it; in the middle, the government may not even invent its own.

The case began with one student’s complaint. In 1990 a female high school student who wanted to attend VMI complained to the Attorney General, and the United States sued the Commonwealth of Virginia and the Virginia Military Institute (VMI) over the all-male admission policy. Trial took six days and an array of expert witnesses on each side. In the two years before the suit, the district court noted, VMI had received inquiries from 347 women and had responded to none of them.

Virginia lost. Measuring the record against that standard, the Supreme Court concluded that Virginia had shown no “exceedingly persuasive justification” for excluding all women from the citizen-soldier training VMI offered, and held that the separate women’s program Virginia proposed at Mary Baldwin did not cure the violation, because it did not provide equal opportunity.

The middle tier is not a formality the government always fails. CRS points to a federal citizenship statute upheld in Nguyen v. INS, where the Court found the sex-based line rested not on a stereotype but on an undeniable difference in the circumstance of the parents at the time a child is born.

Two doors up, and the wall most claims hit

There are two ways out of rational basis, and working out which one you are trying is the first practical question in any case.

The first door is the suspect class. Weighing whether a group qualifies, the Court has asked whether it has experienced a history of purposeful unequal treatment or been subjected to unique disabilities on the basis of stereotyped characteristics not truly indicative of its abilities.

The second door is the fundamental right, and it is narrower than it sounds, because importance is not the criterion. The question is whether the right is explicitly or implicitly guaranteed by the Constitution, not whether the thing at stake matters more than housing or subsistence. Plenty of things a family cannot live without are absent from the document.

When a claim does come through that second door, the facts on the other side of it can be very small and very human. James Obergefell met John Arthur more than two decades before the case that carries his name, and in 2011 Arthur was diagnosed with amyotrophic lateral sclerosis, or ALS, a progressive disease with no known cure.

Resolving to marry before Arthur died, the two traveled from Ohio to Maryland, where same-sex marriage was legal, and were wed inside a medical transport plane as it sat on the tarmac in Baltimore. Three months later Arthur died, and Ohio law would not let Obergefell be listed as the surviving spouse on the death certificate.

Behind both doors sits a wall. A law that lands hardest on one group but never mentions it usually stays at the bottom, because disproportionate impact is not the sole touchstone of an invidious racial discrimination forbidden by the Constitution, and standing alone it does not trigger strict scrutiny. You need purpose, not just effect, and purpose is hard to prove.

Left at the bottom: age, disability, and not having money

In July 1980 Jan Hannah bought a building at 201 Featherston Street in the city of Cleburne, Texas, to lease to Cleburne Living Center for a group home. Thirteen men and women with intellectual disability were to live there under constant staff supervision, in a house with four bedrooms and two baths; the city required a special use permit, and the council refused it, 3 to 1.

The Supreme Court struck down the permit requirement and, in the same opinion, refused to make disability a quasi-suspect class, meaning a category that would have drawn the middle tier.

Its reason for refusing was that legislatures had already acted on behalf of people with intellectual disability, which negates any claim that the mentally retarded are politically powerless in the sense of being unable to attract the attention of lawmakers. Justice Marshall called the method second order rational-basis review. He meant the Court was applying a stiffer test than it would admit to.

The Disability Rights Education and Defense Fund (DREDF), which filed its own friend-of-the-court brief in that case, records both halves of the result: the Court did not give disability “suspect class” status, but did hold there was no rational basis for keeping the group home out of a family neighborhood. Winning the building and losing the tier is not a small trade. A tier is a rule for every future case; a permit is one house.

The same door closed on poverty. Demetrio Rodriguez was a veteran and sheet-metal worker at Kelly Air Force Base in San Antonio. After a 1968 student walkout at Edgewood over insufficient supplies and the lack of qualified teachers, parents formed the Edgewood District Concerned Parents Association, and Rodriguez, who had worked with the American G.I. Forum, the League of United Latin American Citizens and the Mexican-American Betterment Organization, was one of them. He and seven other Edgewood parents filed suit on July 10, 1968.

Both doors stayed shut. The Court held that the Texas school finance system does not disadvantage any suspect class, since people who live in comparatively poor school districts are not a definable suspect class, and that education, though one of the most important services performed by the State, is not among the rights the Constitution guarantees. Age, disability and wealth therefore share an address: the bottom, where the state wins by having a conceivable reason.

When the bottom test bites

Jacinta Moreno was a 56-year-old diabetic living on $75 a month in public assistance, and she shared a household with Ermina Sanchez, who helped care for her, and Sanchez’s three children.

Between her share of the rent, her share of the gas and electricity, $10 a month for transportation to a hospital for regular visits and $5 a month for laundry, she had $10 a month for food and other necessities. The household was disqualified from food stamps because the people in it were not all related to each other.

She won, at the bottom tier. If equal protection means anything, the Court said, it must at the very least mean that a bare congressional desire to harm a politically unpopular group cannot constitute a legitimate governmental interest. That is the escape hatch, and it is narrow: a conceivable reason will do, but hostility toward the group is not a reason at all.

Colorado’s Amendment 2 fell the same way: it was so sweeping and so unrelated to any identifiable legitimate purpose that it raised, in the Court’s words, the inevitable inference that the disadvantage imposed is born of animosity toward the class of persons affected.

How often does the bottom tier bite? A count in the New York University Law Review lists eighteen cases, meaning every Supreme Court decision that found an equal-protection violation under rational-basis scrutiny between the 1971 and 2014 Terms. If your claim lands at the bottom, that is the company you are trying to join.

Where the three tiers came from

They came from a footnote about imitation milk. The Filled Milk Act banned filled milk from interstate commerce as a fraud on the public, and the government indicted the Carolene Products Company for shipping “Milnut,” a compound of condensed skimmed milk and coconut oil made in imitation or semblance of condensed milk or cream.

Footnote four appeared in that opinion, and it set aside three situations where the usual deference to a legislature might be narrower: laws on their face within a specific prohibition of the Constitution, laws restricting the political processes ordinarily relied on to repeal bad legislation, and prejudice against discrete and insular minorities, which may call for a correspondingly more searching judicial inquiry. Every tier above rational basis grew out of that paragraph.

The labels came later, and not from a judge. Writing the Harvard Law Review’s Foreword in 1972, Professor Gerald Gunther described the Warren Court’s top tier as strict in theory and fatal in fact, and named its opposite too: the phrase rational basis with bite comes from his observation that some cases found bite in the equal protection clause after explicitly voicing the traditionally toothless minimal scrutiny standard. Both labels came from a professor describing what he saw, not from a court announcing a rule.

The tiers are being replaced by history, in the areas with the highest stakes

Robert Nash lived in Averill Park and Brandon Koch in Troy, both in Rensselaer County, New York. In 2014 Nash applied for an unrestricted license to carry a handgun in public; he did not claim any unique danger to his personal safety, he simply wanted to carry a handgun for self-defense. The state granted him a restricted license for hunting and target shooting only.

Their case did not produce a tier. It produced a different question. When the Second Amendment’s plain text covers your conduct, the Court held, the Constitution presumptively protects it, and the government must then demonstrate that the regulation is consistent with the Nation’s historical tradition of firearm regulation. The means-end scrutiny lower courts had been adding after that step was, in the Court’s phrase, one step too many.

The reasoning deserves stating at full strength, because it is the serious case against tiers. The Court’s earlier Second Amendment cases, Heller and McDonald, expressly rejected any judge-empowering “interest-balancing inquiry” that asks whether a statute burdens a protected interest out of proportion to its salutary effects.

On that view, a judge who declares an interest compelling and a fit narrow is legislating with a straight face, and words like important and substantially related are personal preference in a robe. A historical question, whatever its difficulties, has an answer that does not depend on the judge’s politics.

The same move has been made elsewhere, and not in one direction. For religious exercise, the Court held that regulations are not neutral and generally applicable, and therefore trigger strict scrutiny, whenever they treat any comparable secular activity more favorably than religious exercise, and that treating some secular businesses just as badly is no answer.

For abortion, the Court went the other way, replacing the undue burden standard and holding that a law regulating abortion is entitled to a “strong presumption of validity” and must be sustained if there is a rational basis on which the legislature could have thought it would serve legitimate state interests.

One area got a test the government usually loses; another got the test the government almost never loses. Both were justified by the same appeal to history and tradition. The method has no fixed political direction; it moves a different question to the center.

Whether it is workable is contested, and by judges. The Brennan Center for Justice reports that federal judges appointed by Presidents Reagan, Clinton, George W. Bush, Obama, Trump, and Biden have all questioned the firearms decision, warning that history is an unworkable basis for deciding constitutional questions and pushes courts toward unreliable, unreasonable and unjust conclusions.

The Court has since loosened the test. In a later firearms case it explained that the analysis involves asking whether the challenged regulation is consistent with the principles that underpin our regulatory tradition and whether the new law is relevantly similar to laws the tradition permits. A modern statute, in other words, does not need a historical twin.

How the fight over the test is actually run

Most of these suits travel on one sentence of federal law. Section 1983 of title 42 of the United States Code makes every person who, under color of any statute, ordinance, regulation, custom, or usage of a state subjects another to the deprivation of rights secured by the Constitution liable to the injured party in an action at law or a suit in equity.

In plain terms: when a state or local official enforces a law against you, that statute is how you get into court to argue the law is unconstitutional.

The fight over the test starts in the complaint, before a shred of evidence. The complaint filed against Tennessee’s ban on hormone therapy for transgender minors pleads it outright, alleging that the classification is subject to heightened scrutiny under the Equal Protection Clause and is therefore presumptively unconstitutional, then walking through the suspect-class factors one at a time: obvious, immutable and distinguishing characteristics unrelated to the ability to contribute to society, a history of discrimination, and a lack of political power.

Behind the pleading was a family. L.W. was their 16-year-old daughter, and she was receiving the hormone therapy her doctors had prescribed. Tennessee’s S.B. 1, which took effect in July 2023, barred providers from giving transgender minors puberty blockers and hormone therapies that other minors could still get.

In a statement issued by the ACLU as the case went to the Supreme Court, she said: “I want this law to be struck down so that I can continue to receive the care I need, with the support of my parents and my doctors.”

Tennessee’s answer was about who decides, not about whether the treatment helps. Its brief argued that heightened scrutiny here would transfer “trying policy choices” and “fraught line-drawing dilemmas” from the arena of public debate and legislative action to the unelected federal judiciary, and that the federal government’s disagreement with a state’s assessment of the risks does not suffice to invalidate a democratically enacted law.

Even under heightened review, the state added, courts defer to legislative factfinding and give states wide discretion to pass legislation in areas where there is medical and scientific uncertainty, and a district court weighing the risks and benefits itself, with no deference to the legislature’s findings, was not the judiciary’s role.

Tennessee won on the sorting. The Court read the statute as drawing two classifications, one based on age and one based on medical use, neither turning on sex, and held that classifications based on age or medical use are subject to only rational basis review. The medical evidence both sides had assembled was never weighed by the Justices, because the choice of test made weighing it unnecessary.

So the honest answer to whether a court will strike down the law standing in your way begins one step earlier than most people expect. Before the hearing, before the evidence, a judge decides which test your claim gets, and that decision hands the burden to you or to the government.

If your claim turns on race or national origin, or on a right the Constitution names, the state has to justify itself. If it turns on age, disability, income or an ordinary economic rule, you have to rule out every reason the state might have had, including reasons nobody in the state ever thought of.

And where the Court now asks about history instead of tiers, the sorting question has changed shape without going away: someone still carries the burden, and the first thing worth learning about your case is who.