In the momentous 1803 case Marbury v. Madison, Chief Justice Marshall observed that the “very essence of civil liberty certainly consists in the right of every individual to claim the protection of the laws, whenever he receives an injury” and warned that a government cannot be called a “government of laws, and not of men . . . . if the laws furnish no remedy for the violation of a vested legal right.”

When the government itself violates individuals’ rights, it is especially important for courts to furnish a remedy. To be sure, providing remedies to the victims of unconstitutional conduct after the fact is often at best an imperfect solution. While money plausibly provides full compensation to, say, a government worker denied income while suspended for engaging in First Amendment–protected activity, it may be far less effective in a case involving an unconstitutional strip search: Can money really restore the sense of security that the victim has lost? If not, perhaps it can at least enable her to begin rebuilding her life.

Ideally, backward-looking remedies can deter future violations. If government officials or agencies know that they will be held to account, they will be less likely to commit violations in the first place. And injunctive remedies—judicial orders either to stop engaging in a particular practice (for example, racial profiling of motorists) or to start doing something the government has so far failed to do (for example, issuing marriage licenses to same-sex couples)—have been critical to enforcing civil rights and civil liberties.

More broadly, judicial remedies perform an important expressive function: they drive home to the public that the law takes constitutional violations seriously. This can galvanize popular movements to vindicate constitutional values even more fully. For example, the Supreme Court’s decisions in Brown v. Board of Education (1954–55), held that purposeful racial segregation of public schools violated the Fourteenth Amendment’s equal protection clause and ordered that school boards dismantle their dual school systems. The Brown decisions by themselves achieved very little actual integration, but the Court’s condemnation of segregation provided critical support to a mass movement that culminated in statutes such as the Civil Rights Act of 1964, which crafted more effective tools for dismantling Jim Crow in schools, public accommodation, employment, and housing.

So a lot depends on the availability of a remedy: when the courts refuse to provide one, rights can be reduced to mere lines on paper.

Indeed, recent years have shown a troubling trend as the Supreme Court has retrenched on protecting civil rights and civil liberties in two important ways. Sometimes the Court has straightforwardly redefined an underlying right in more limited terms. For example, in Montejo v. Louisiana (2009) the Court narrowed Sixth Amendment protections against police interrogation of criminal defendants without their lawyers being present. And in Gonzales v. Carhart (2007) the Court upheld restrictions on abortion rights almost identical to those it had earlier struck down. (I represented the California Medical Association as an amicus curiae in the case.)

The second approach is more insidious. Here the Court leaves the formal right in place but constricts the remedial machinery. At best, this will dilute the value of the right, since some violations will go unremedied. At worst, it may signal potential wrongdoers that they can infringe the right with impunity.

Remedial abridgment is a pervasive tool of the contemporary Supreme Court. One area in which it’s often used is criminal procedure. For decades the so-called exclusionary rule forbade prosecutors from using evidence obtained through unconstitutional searches or impermissible interrogation techniques. But while the Burger, Rehnquist, and Roberts Courts have left in place most of the Warren Court’s restrictions on police activity, they have developed a series of what criminal procedure scholar Carol Steiker calls “inclusionary” rules: exceptions to the exclusionary rule that result in many defendants being convicted on the basis of illegally obtained evidence. These inclusionary rules weaken the deterrent force of the exclusionary rule by signaling to police that illegally obtained evidence won’t be worthless.

The Supreme Court may be signaling potential wrongdoers that they can infringe rights with impunity.

In its cases cutting back on the exclusionary rule, the Court often points to the availability of alternative mechanisms for enforcing the Fourth Amendment’s prohibition of unconstitutional searches and seizures. The Court has repeatedly highlighted the option to seek civil damages under a Reconstruction-era statute (42 U.S.C. § 1983), which authorizes suits against state and local officials and governments that deprive individuals of their constitutional rights. Yet the Court has substantially weakened the section 1983 remedy at precisely the same time that it has weakened the exclusionary rule, engaging in a sort of shell game in which the presence of each remedy serves to justify weakening the protections of the other.

For example, section 1983 suits are subject to an increasingly government-friendly doctrine of “qualified immunity,” which shields officials from liability unless they violate constitutional rules that were “clearly established” at the time of the offense. To get a sense of how protective that rule is, consider Wilson v. Layne (1999). First the Supreme Court held unanimously that, according to the Fourth Amendment’s prohibition of unreasonable searches, law enforcement officials could not bring the media uninvited into people’s homes (think Cops). Then the Court held, 8-1, that until it issued its unanimous opinion, it would not have been clear that such intrusions were impermissible.

Although qualified immunity provides no barrier to suits against governments themselves, the Court has developed other doctrines to shield them. A particularly callous example of the Court’s willingness to leave victims of even atrocious constitutional violations uncompensated came in last term’s Connick v. Thompson. There, the Court tossed out the $14 million jury verdict John Thompson obtained against the Orleans Parish District Attorney’s Office after he spent fourteen years in solitary confinement on death row for a murder he did not commit. Prosecutors had deliberately withheld favorable evidence from his lawyers. The Court did not contest that the prosecutors’ behavior violated clear constitutional standards. But it held that the government could not be held liable because it was not sufficiently obvious that the failure to train its prosecutors about their constitutional responsibilities would cause such violations. (I represented a group of former high-level officials in the Department of Justice, from Republican and Democratic administrations alike, who filed an amicus brief stressing the necessity of such training.)

The current term is only half over, and already the Court has further narrowed the availability of remedies for constitutional violations and has given hints at oral argument of additional remedial abridgements to come. In Minneci v. Pollard, for example, the Court ruled that federal prisoners held in private facilities (as opposed to ones run by the Bureau of Prisons) cannot seek damages in federal court for violations of the Eighth Amendment’s prohibition of cruel and unusual punishment. Sure, the Eighth Amendment applies: privately run prisons, like government-run ones, cannot deny inmates adequate medical care or nutrition, or physically mistreat them. But when they do, the inmates’ only resort is to go into state court under a state-law theory, such as medical malpractice. They cannot vindicate their constitutional claims directly.

The Court also is poised to expand qualified immunity even further and to limit Congress’s ability to authorize a variety of private lawsuits. Decisions such as Connick v. Thompson or the recent series of cases ejecting workers and consumers from the courts and forcing them into arbitration—often preventing victims banding together as a class—don’t explicitly change what the law tells governments or businesses they can or cannot do. But these decisions can have that effect, by making it harder for victims of wrongful conduct to exercise their rights. They thus betray the Supreme Court’s long-ago promise that it would ensure a government of laws.