Chapter 5 — Civil Liberties: The Bill of Rights and the Ongoing Argument Over Individual Freedom

"The very purpose of a Bill of Rights was to withdraw certain subjects from the vicissitudes of political controversy, to place them beyond the reach of majorities and officials and to establish them as legal principles to be applied by the courts. One's right to life, liberty, and property, to free speech, a free press, freedom of worship and assembly, and other fundamental rights may not be submitted to vote; they depend on the outcome of no elections." — West Virginia State Board of Education v. Barnette, 319 U.S. 624 (1943) (Jackson, J.)

Why this chapter exists

If the Constitution is the operating system, the Bill of Rights is the firewall. The first ten amendments do something that ordinary statutes cannot: they tell the government what it may not do, even if a majority of voters wants it to. That makes civil liberties the most counter-majoritarian feature of American government — and the most important one, because everyone, eventually, ends up in the minority on something.

Three points up front, because they govern the rest of the chapter.

First, civil liberties and civil rights are not the same thing. Civil liberties are limits on government power over the individual: the government cannot censor your speech, search your home without a warrant, jail you without a trial, or compel you to incriminate yourself. Civil rights are guarantees of equal treatment by the government and, in some contexts, by private actors covered by federal statute: the government cannot deny you the vote, the schoolhouse, or the lunch counter on the basis of race, sex, or other protected characteristics. The two often overlap — the right to marry the person you choose involves both — but they answer different questions. Civil liberties asks what the government may not do at all. Civil rights asks how the government must treat people equally when it does act. Chapter 6 takes up civil rights. This chapter is about the limits.

Second, civil liberties cut across partisan lines. The same First Amendment that protects a Klan rally also protects a Black Lives Matter rally. The same Fourth Amendment that protects a suspected drug dealer from a warrantless search also protects a small-business owner from a warrantless IRS audit. The same Fifth Amendment that protects an organized-crime boss from compelled self-incrimination also protects a whistleblower hauled before a hostile congressional committee. A textbook that defends civil liberties only when they protect speakers and defendants the author likes is not a textbook on civil liberties. It is a brief for one side. This book defends the doctrinal protections regardless of who benefits from them in any given case.

Third, the law in this area is alive and changing. The Roberts Court has reshaped major doctrines — Second Amendment, religion, abortion, administrative search — in ways that may settle into stable doctrine or may be overturned in the next decade. We will note where the doctrine has shifted and why, and present the strongest versions of the arguments for both the new doctrine and the old.

The Bill of Rights and the long road to incorporation

The first Congress, in 1789, sent twelve amendments to the states. Ten were ratified by 1791. They are the Bill of Rights — and for the first 75 years of American history, they did almost nothing.

The reason is that they were written to bind the federal government, not the states. Barron v. Baltimore (1833) made this explicit: a Baltimore wharf-owner whose business had been ruined by city construction sued under the Fifth Amendment's Takings Clause, and Chief Justice John Marshall held that the Bill of Rights "contains no expression indicating an intention to apply [its provisions] to the State governments." If you wanted protection from your state government, your remedy was your state constitution, not the federal one.

That changed, slowly and incompletely, after the Civil War. The Fourteenth Amendment (1868) said in plain text that "No State shall make or enforce any law which shall abridge the privileges or immunities of citizens of the United States; nor shall any State deprive any person of life, liberty, or property, without due process of law." The Reconstruction Congress wrote those words to make the states answerable to the federal Constitution — including, the Reconstruction-era authors believed, the Bill of Rights.

The Supreme Court, however, took its time. The Slaughter-House Cases (1873) effectively read the Privileges or Immunities Clause out of the Constitution. Hurtado v. California (1884) held that the Fifth Amendment's grand-jury indictment requirement did not apply to the states through the Fourteenth Amendment — using a textual argument that, taken seriously, would have meant no part of the Bill of Rights applied. For two generations, the Bill of Rights remained almost as inert against the states as it had been before the Civil War.

The reanimation began in the 1920s. Gitlow v. New York (1925), upholding a conviction for socialist pamphleteering, included almost in passing the proposition that "freedom of speech and of the press . . . are among the fundamental personal rights and 'liberties' protected by the due process clause of the Fourteenth Amendment from impairment by the States." With that one sentence the Court began the doctrine of selective incorporation: the idea that the Fourteenth Amendment's Due Process Clause "incorporates" against the states those rights in the Bill of Rights that are "fundamental" or "deeply rooted in the Nation's history and tradition."

Over the next century, the Court incorporated almost the whole Bill of Rights: - First Amendment — speech (Gitlow 1925), press (Near v. Minnesota 1931), assembly (De Jonge v. Oregon 1937), religion (Cantwell v. Connecticut 1940; Everson v. Board of Education 1947). - Fourth Amendment — search and seizure (Mapp v. Ohio 1961, with the exclusionary rule). - Fifth Amendment — self-incrimination (Malloy v. Hogan 1964) and double jeopardy (Benton v. Maryland 1969); takings (Chicago, B. & Q. R. Co. v. Chicago 1897, very early). - Sixth Amendment — counsel in felony cases (Gideon v. Wainwright 1963), speedy trial (Klopfer v. North Carolina 1967), confrontation (Pointer v. Texas 1965), jury trial (Duncan v. Louisiana 1968). - Eighth Amendment — cruel and unusual punishment (Robinson v. California 1962). - Second Amendment — McDonald v. Chicago (2010), the most recent major incorporation.

A few provisions remain unincorporated. The Third Amendment (no quartering of soldiers) has never been litigated to the Supreme Court for incorporation, because it has almost never come up. The Fifth Amendment's grand-jury indictment requirement is still not binding on the states (a state may charge a felony by prosecutor's information, and many do). The Seventh Amendment's civil-jury right is also unincorporated — a state may try a civil case to a judge alone, or with a six-person rather than twelve-person jury. The Eighth Amendment's prohibition on excessive fines was not formally incorporated until Timbs v. Indiana (2019).

The plain-English version: most of the Bill of Rights binds the states today, but it does so by a doctrinal route — the Fourteenth Amendment's Due Process Clause — that took a century to build, and a few corners of it still bind only the federal government. When we say "the First Amendment protects you from your state's speech regulations," we are reciting the result of the long story of incorporation, not the original 1791 design.

This matters for more than antiquarian reasons. Some scholars and judges — Justice Thomas most prominently — argue that the Court has incorporated the Bill of Rights through the wrong textual provision, and that the Privileges or Immunities Clause (rather than the Due Process Clause) is the correct vehicle. McDonald v. Chicago (2010) addressed this directly: Justice Alito's plurality opinion incorporated the Second Amendment through the Due Process Clause along the established route, while Justice Thomas concurred in the judgment but argued at length that Privileges or Immunities was the historically correct path. The doctrinal disagreement has practical implications. If the Court were to revive the Privileges or Immunities Clause as a vehicle for incorporated rights, it could potentially reach a wider category of substantive protections — economic liberties, perhaps even rights not currently treated as fundamental — than the Due Process Clause has been read to support. Reviving the Privileges or Immunities Clause is an unenumerated agenda item of some originalist scholars; it has no current majority on the Court but is one of the doctrinal possibilities a serious reader should be aware of.

The First Amendment: speech

"Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the Government for a redress of grievances."

That single sentence does six different things. Take them one at a time.

The doctrinal categories of unprotected and lesser-protected speech

The First Amendment does not literally protect "speech" against any conceivable government action. It protects most speech against most kinds of government regulation, while permitting regulation in specific categories the Court has carved out over a century.

Incitement. Until 1969, the Court applied a "clear and present danger" test that, in practice, allowed the government to punish speech advocating illegal conduct. Brandenburg v. Ohio (1969) replaced that test with a much stricter one: the government may punish advocacy of unlawful action only when the advocacy is "directed to inciting or producing imminent lawless action" and "is likely to incite or produce such action." Both prongs must be satisfied. Speech that advocates violence at some indefinite future time, no matter how repulsive, is protected. (Case Study 1 walks through this in depth, including how it applies to recent contested speech.)

True threats. A statement that a reasonable person would understand as a serious expression of intent to do harm to another is unprotected (Virginia v. Black 2003). Counterman v. Colorado (2023) added that the speaker must have at least recklessly known the statement would be received as threatening — pure negligence is not enough.

Defamation. False statements of fact that damage another person's reputation are not constitutionally protected when made about private figures. But for public officials and public figures, New York Times v. Sullivan (1964) requires proof of "actual malice" — that the speaker knew the statement was false or recklessly disregarded the truth. The point of the doctrine is that "debate on public issues should be uninhibited, robust, and wide-open," even if some of that debate is wrong. The doctrine is now under fresh political pressure (Justices Thomas and Gorsuch have written separately questioning Sullivan's scope), and a serious student of civil liberties should track the question as it develops.

Commercial speech. Truthful, non-misleading advertising of lawful products or services receives intermediate First Amendment protection (Central Hudson 1980). The government can require warnings (cigarettes, prescription drugs), prohibit deceptive ads, and regulate professional advertising — but blanket bans on truthful information about lawful products generally fail.

Obscenity. Miller v. California (1973) defines obscenity by a three-part test: (1) the average person, applying contemporary community standards, would find the work, taken as a whole, appeals to the prurient interest; (2) the work depicts or describes sexual conduct in a patently offensive way; and (3) the work, taken as a whole, lacks serious literary, artistic, political, or scientific value. The third prong is the one that matters in practice — almost no commercial pornography fails it, which is why the Miller test functions less as a ban on adult content than as a sorting mechanism between protected and unprotected at the extreme margin (child pornography is treated separately and is categorically unprotected, New York v. Ferber 1982).

Fighting words. Face-to-face words "which by their very utterance inflict injury or tend to incite an immediate breach of the peace" are unprotected (Chaplinsky v. New Hampshire 1942). The category exists in the case law but the Court has not affirmed a fighting-words conviction since 1942. Functionally, it is a narrow exception kept alive in doctrine but rarely applied.

Content-based vs. content-neutral regulation

Whatever else the First Amendment does, it draws a sharp line between regulations that target speech because of what is said and regulations that target speech because of where, when, or how it is said.

Content-based regulations — laws that restrict speech because of its message, subject matter, or viewpoint — receive strict scrutiny. The government must show that the regulation is narrowly tailored to serve a compelling state interest. Almost no content-based regulation survives that test.

Content-neutral regulations — reasonable time, place, and manner restrictions that apply regardless of the speaker's message — receive intermediate scrutiny. A city can require parade permits, prohibit amplified sound near hospitals at night, and restrict door-to-door solicitation, as long as the restriction serves a substantial government interest, is narrowly tailored, and leaves open ample alternative channels of communication.

The line between the two can be subtle. Reed v. Town of Gilbert (2015) held that a municipal sign code that treated "ideological," "political," and "directional" signs differently was content-based on its face — even though the town swore it had no censorial intent — and applied strict scrutiny. After Reed, courts have been more willing to find facially neutral regulations content-based when the categories the regulator drew turn on what the sign or speech says.

Government-employee speech

When the government acts as employer rather than as sovereign, the First Amendment analysis shifts. Pickering v. Board of Education (1968) held that a public-school teacher could not be fired for writing a letter to the editor criticizing the school board on a matter of public concern. The Court balanced the employee's interest in commenting on public matters against the government's interest in efficient operation of its workplace.

But Garcetti v. Ceballos (2006) added a critical limitation: when public employees speak pursuant to their official duties, the First Amendment does not protect them from discipline. A district attorney who wrote a memo as part of his job arguing that a search warrant was based on misrepresentations had no First Amendment claim when his superiors retaliated. The line between job-duty speech and citizen speech matters enormously to teachers, police officers, and bureaucratic whistleblowers.

Public-forum doctrine

The First Amendment regulates government regulation of speech in spaces the government controls. The doctrinal taxonomy:

• Traditional public forums (streets, parks, sidewalks) — content-based regulations get strict scrutiny; content-neutral time, place, and manner regulations are permitted.
• Designated public forums (a public auditorium opened to private speakers) — same rules as traditional public forums while the forum is open.
• Limited public forums (a public school after hours opened to community groups) — the government may limit subject matter and speaker categories, but cannot discriminate among viewpoints within whatever it has opened.
• Nonpublic forums (military bases, prisons, internal government offices) — only reasonableness and viewpoint neutrality are required.

The Internet has scrambled this taxonomy. Are social-media platforms public forums? They are not government property. But they function as the dominant places where public discourse happens. The Court took up the question in NetChoice v. Paxton and Moody v. NetChoice (2024), reviewing Texas and Florida laws that restricted platforms' ability to moderate user content. The Court vacated and remanded for fuller First Amendment analysis, holding that platforms' content-moderation choices implicate the platforms' own First Amendment rights (because curation is itself an editorial act). Lower courts and legislatures are still working through the implications. The contested doctrinal question is whether the major platforms should be treated as private speakers exercising editorial discretion (the position the platforms and most of the NetChoice majority took) or as something closer to common carriers obligated to carry all lawful speech (the position the Texas and Florida statutes embodied). Each side has serious arguments — about the scope of the First Amendment, about the structure of the modern public square, and about the economic power of the platforms. This is one of the most active doctrinal frontiers in American First Amendment law.

The First Amendment: religion

"Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof . . ."

The Religion Clauses do two distinct things, and they are in genuine tension with each other.

The Establishment Clause

The Establishment Clause forbids the government from establishing a religion. What that means in practice has been the subject of a century of doctrinal evolution.

The mid-twentieth-century answer was Lemon v. Kurtzman (1971): a law violates the Establishment Clause unless (1) it has a secular purpose, (2) its principal effect neither advances nor inhibits religion, and (3) it does not foster excessive entanglement between government and religion. The Lemon test was used to invalidate school-organized prayer, evolution-disclaimer requirements, holiday displays without secular context, and dozens of other arrangements.

Critics of Lemon — including Justice Scalia, who memorably described it as "some ghoul in a late-night horror movie that repeatedly sits up in its grave and shuffles abroad" — argued it was unmoored from the historical meaning of the Establishment Clause and produced erratic results.

Kennedy v. Bremerton School District (2022) explicitly abandoned Lemon and replaced it with a "history and tradition" test: courts now ask whether the challenged practice is consistent with the way the Establishment Clause was understood at the founding and in early American history. The case involved a public-school football coach who knelt to pray on the 50-yard line after games. The Court, 6-3, held that the coach's prayer was protected private religious expression and that the school district's claim of an Establishment Clause obligation could not override the coach's free-exercise and free-speech rights.

Defenders of Bremerton argue it restores the original meaning of the Establishment Clause and ends the doctrinal incoherence of Lemon. Critics argue that "history and tradition" is malleable in a country whose founding generation tolerated arrangements (state-supported churches in some early states; explicit religious tests for office in others) that no one today would defend, and that the new test will produce more, not less, doctrinal uncertainty. Both sides have a point. The doctrine is unsettled in real time.

The Free Exercise Clause

The Free Exercise Clause protects religious practice from government interference. The doctrinal question — sharper than the Establishment Clause's question — is what to do when a religiously motivated practice conflicts with a generally applicable, secular law.

The mid-twentieth-century answer was Sherbert v. Verner (1963) and Wisconsin v. Yoder (1972): if a generally applicable law substantially burdened religious practice, the government had to justify the burden under strict scrutiny. Yoder exempted Amish parents from compulsory-school-attendance laws past eighth grade.

Employment Division v. Smith (1990) — written by Justice Scalia — substantially narrowed that doctrine. Two members of the Native American Church were denied unemployment benefits after being fired for sacramental peyote use. The Court held that a neutral, generally applicable law does not violate the Free Exercise Clause merely because it incidentally burdens a religious practice. "The right of free exercise does not relieve an individual of the obligation to comply with a 'valid and neutral law of general applicability.'"

Smith triggered a strong political backlash from a coalition that crossed every ordinary partisan line. Congress passed the Religious Freedom Restoration Act (RFRA) in 1993 with overwhelming bipartisan support — voice vote in the House, 97-3 in the Senate, signed by President Clinton. RFRA restored the Sherbert/Yoder strict-scrutiny test by statute. The Supreme Court in City of Boerne v. Flores (1997) held that RFRA exceeded Congress's powers as applied to the states but remained valid as applied to the federal government. About half the states have enacted state-RFRAs that cover their own laws.

RFRA produced Burwell v. Hobby Lobby Stores (2014): a closely held for-profit corporation was held to be a "person" within RFRA's protection, and the Affordable Care Act's contraception-coverage mandate substantially burdened the religious exercise of the corporation's owners. The case sharpened a question that has dominated Free Exercise litigation since: when religious exercise conflicts with antidiscrimination law or with healthcare access, who wins?

Masterpiece Cakeshop v. Colorado Civil Rights Commission (2018) and 303 Creative LLC v. Elenis (2023) addressed related questions in the speech register. Masterpiece held narrowly that the Colorado Civil Rights Commission had shown anti-religious hostility in adjudicating the case of a baker who declined to make a cake for a same-sex wedding. 303 Creative held more broadly that a website designer could not be compelled by a public-accommodations law to design wedding websites celebrating same-sex marriages, because compelled creation of expressive content violated the designer's First Amendment speech rights. Critics argue the cases license discrimination under the cover of religious or expressive exemption. Defenders argue they vindicate the principle that the state may not conscript individuals into producing expressive content that violates their conscience. The doctrinal question — what kinds of commercial activity are sufficiently "expressive" to claim First Amendment exemption from antidiscrimination law? — remains unresolved.

The genuine tension

Notice the structural problem. The Establishment Clause restrains government accommodation of religion. The Free Exercise Clause requires government accommodation of religion in some cases. A school district that lets a coach pray after a game might be either complying with the Free Exercise Clause (the Bremerton result) or violating the Establishment Clause (the pre-Bremerton result). The two clauses are not redundant — they pull in different directions, and the Court is engaged in working out where one ends and the other begins. A serious civil-libertarian recognizes the tension instead of pretending only one clause matters.

The First Amendment: press, assembly, petition

The Free Press Clause has historically been read to protect the press's institutional role — gathering and publishing news — but the Court has been clear that the press has no constitutional rights beyond those of ordinary speakers. The doctrinal cornerstone here is Near v. Minnesota (1931), which struck down a state law authorizing prior restraints on "malicious, scandalous and defamatory" publications. The presumption against prior restraint — government censorship of material before publication — has been sustained even in cases of plausible national-security claims. New York Times Co. v. United States (1971), the Pentagon Papers case, denied the federal government an injunction against publication of classified material on the Vietnam War, with each justice writing separately to articulate a different theory of why prior restraint was impermissible there. The doctrine permits post-publication punishment in narrow cases (the Espionage Act remains in force) but treats pre-publication censorship as constitutionally suspect in nearly every conceivable situation.

Branzburg v. Hayes (1972) held that journalists have no First Amendment privilege to refuse to testify before grand juries. Most states have enacted shield laws protecting journalists from compelled disclosure of confidential sources; there is no federal shield law as of 2026, though one (the PRESS Act) has been proposed in successive Congresses. The status of online publishers and individual writers under shield-law protections is itself a contested doctrinal question — does a journalist who has never worked for a traditional outlet receive the same protection as a Times reporter? — and state laws answer the question differently.

The Free Assembly Clause protects the right to gather peacefully — at protests, rallies, religious services, political meetings. It has been read together with the Free Speech Clause to protect "expressive association" (NAACP v. Alabama 1958, protecting the NAACP from compelled disclosure of its membership lists in segregation-era Alabama). Boy Scouts of America v. Dale (2000) extended the doctrine to permit a private organization to exclude members whose presence would interfere with the organization's expressive purposes — a doctrine that cuts in different directions for different organizations. The right to peaceably assemble does important work: it draws a constitutional line between protected protest (even loud, disruptive, traffic-blocking protest) and unprotected violence or property destruction. The line is sometimes hard to administer in practice — a single window broken in a 10,000-person march is not the act of the march — but it is the line the doctrine draws.

The right to assemble overlaps with the right to not be compelled to associate. Janus v. AFSCME (2018) held that public-sector unions cannot compel non-members to pay agency fees as a condition of employment, because such fees implicate compelled political speech. The decision overruled Abood v. Detroit Board of Education (1977), which had distinguished collective-bargaining fees from political-advocacy fees. Janus is one of the cases in which the Court's commitment to robust First Amendment protection cut against a position favored by labor and most contemporary progressive policy preferences — a pattern we will see again in Case Study 1.

The Petition Clause protects the right to address the government with grievances. It is often the forgotten clause, but it underlies things ranging from lobbying (Chapter 22) to litigation against the government to public-comment processes in agency rule-making (Chapter 10). The Court has held that the Petition Clause does not give petitioners a right to a response — the government can ignore your letter — but it protects you from retaliation for sending it. The clause also has a long historical pedigree: the right to petition the Crown predated the Bill of Rights by centuries in English constitutional history, and the Founders' inclusion of it reflected their concern that any government, however well-designed, could become unresponsive without protected channels for citizens' complaints.

The Second Amendment: the right to bear arms

"A well regulated Militia, being necessary to the security of a free State, the right of the people to keep and bear Arms, shall not be infringed."

For most of American history, the Supreme Court said almost nothing about what those twenty-seven words mean. The leading case for over a half-century was United States v. Miller (1939), which upheld a federal ban on sawed-off shotguns on the ground that the weapon had no "reasonable relationship to the preservation or efficiency of a well regulated militia." Lower federal courts read Miller to mean that the Second Amendment protected a collective right — a right of the people, organized into militias, to bear arms — rather than an individual right of private citizens to own firearms unconnected to militia service.

District of Columbia v. Heller (2008) ended that reading. The Court, 5-4 (Scalia for the majority, Stevens dissenting), held that the Second Amendment protects an individual right to keep and bear arms unconnected to militia service, "for traditionally lawful purposes, such as self-defense within the home." The majority based the holding on close textual analysis of the prefatory and operative clauses, on Founding-era usage of the relevant terms, and on the structure of state constitutions and early state cases. The dissent contested all of those readings and argued that the prefatory clause limited the operative clause to militia-related arms-bearing.

McDonald v. City of Chicago (2010) incorporated the Second Amendment against the states through the Fourteenth Amendment. After McDonald, Heller's individual-right reading bound state and local gun regulations as well.

For the next twelve years, lower federal courts mostly used a "two-step" test: first, ask whether the regulated conduct fell within the Second Amendment's scope; if so, apply some form of intermediate or strict scrutiny. Most regulations survived. Many gun-rights advocates argued the test was being applied too deferentially.

New York State Rifle & Pistol Association v. Bruen (2022) replaced the two-step test with a "history and tradition" test analogous to the test from Bremerton. Now, when the Second Amendment's plain text covers a regulated practice, the government must justify the regulation by demonstrating that it is "consistent with the Nation's historical tradition of firearm regulation." (Case Study 2 walks through Bruen in depth, including the empirical data on gun violence and the contested question of whether the new test is workable.)

The empirical landscape

The Second Amendment is the area where it is easiest to confuse empirical and normative claims, and most important to keep them separate. Here is what the data shows.

In 2022 (the most recent year for finalized federal data as of this writing), the Centers for Disease Control reported approximately 48,000 firearm-related deaths in the United States. About 58% (~27,000) were suicides; about 38% (~18,000) were homicides. Accidents, legal interventions, and undetermined causes accounted for the remainder. The United States has substantially more firearm deaths per capita than any other developed democracy — by a factor of roughly 5-10 depending on the comparison country and the year.

Within the United States, the geography is uneven. Firearm-suicide rates are highest in rural states with high firearm-ownership rates (Montana, Wyoming, Alaska). Firearm-homicide rates are highest in concentrated urban-poverty areas in cities including Chicago, St. Louis, Baltimore, and Memphis. Mass-shooting events — defined varyingly, but typically as four or more victims in a single incident — receive the most media coverage but account for under 1% of total firearm deaths.

Public-opinion data is more fragmented than the loud political debate suggests. Pew and Gallup polling consistently show majority support for some regulations (universal background checks, red-flag laws, restrictions on high-capacity magazines and bump stocks) and majority opposition to others (a national handgun ban, mandatory firearm licensing comparable to driver's licenses). The same polls show large majorities affirming the individual-rights reading of the Second Amendment.

What to do about the data is contested in a way the data itself is not. Gun-rights advocates argue that proposed restrictions either fail constitutional scrutiny under Heller/Bruen or fail to address the empirical drivers of firearm violence (which are concentrated in illegally trafficked firearms used in concentrated geographic areas, not in newly purchased firearms used by typical owners). Gun-control advocates argue that the United States's outlier status in firearm deaths cannot be explained without reference to the United States's outlier status in firearm availability, and that incremental regulations modeled on those of peer democracies could meaningfully reduce harm at acceptable cost. Both sides invoke serious empirical and constitutional arguments. This book treats the empirical question (the per-capita gap exists) and the normative question (what to do about it) as distinct, and does not pretend either side has a monopoly on rigor.

The Fourth Amendment: search and seizure

"The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no Warrants shall issue, but upon probable cause, supported by Oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized."

The Fourth Amendment has two clauses: the reasonableness clause ("no unreasonable searches and seizures") and the warrant clause (warrants require probable cause and particularity). The relationship between them — does the reasonableness clause require warrants, or is "reasonableness" the master concept? — is the source of much of the doctrinal complexity.

The general rule: the government must have a warrant, supported by probable cause, to search a person, a home, or an effect. The exceptions: many. Consent searches, searches incident to arrest, plain-view seizures, automobile searches (lower expectation of privacy), exigent-circumstances searches (hot pursuit, imminent destruction of evidence), border searches (no warrant or probable cause required, with limits at the digital border), administrative searches (inspections of regulated industries), and Terry stops (a brief investigatory stop on reasonable suspicion, less than probable cause).

Terry v. Ohio (1968) held that a police officer with reasonable suspicion that a person is engaged in criminal activity may briefly detain that person for investigation, and may pat down the person for weapons if there is reason to believe the person is armed. The doctrine has expanded over the decades to authorize the entire field of "stop and frisk" policing, and is one of the most contested points where Fourth Amendment doctrine intersects with race, urban policing, and concentrated-poverty enforcement patterns.

The exclusionary rule and good faith

Until 1914 in federal court (Weeks v. United States) and 1961 in state court (Mapp v. Ohio), evidence obtained in violation of the Fourth Amendment could still be used at trial. The exclusionary rule changed that: evidence obtained through an unconstitutional search or seizure must be suppressed at trial.

The rule has been controversial since adoption. Critics argue it lets factually guilty defendants go free because of police misconduct that the defendants themselves did not suffer. Defenders argue it is the only effective remedy for Fourth Amendment violations — civil suits against police are difficult to win, internal discipline is uneven, and without exclusion, the constitutional protection is functionally unenforceable.

The Supreme Court has steadily narrowed the rule. United States v. Leon (1984) created the good-faith exception: if police conducted a search in objectively reasonable reliance on a warrant later found defective, the evidence is admissible. Herring v. United States (2009) extended good faith to police database errors. Each contraction has reduced the rule's deterrent force.

Carpenter and the digital Fourth Amendment

The Fourth Amendment has had to evolve to accommodate technologies that did not exist when it was ratified. Olmstead v. United States (1928) had once held that wiretapping was not a Fourth Amendment "search" because there was no physical trespass. Katz v. United States (1967) replaced Olmstead with the "reasonable expectation of privacy" test: a Fourth Amendment search occurs when the government violates a reasonable expectation of privacy that society is prepared to recognize as legitimate.

For decades, the third-party doctrine (Smith v. Maryland 1979) held that there is no reasonable expectation of privacy in information voluntarily turned over to a third party — bank records, telephone numbers dialed, and so on. That doctrine, applied to the modern digital world, would mean that almost everything you do online is unprotected by the Fourth Amendment, because almost everything you do is mediated by third parties.

Carpenter v. United States (2018) declined to extend the third-party doctrine to cell-site location information (CSLI). The government had obtained, without a warrant, 127 days of cell-site records that placed Timothy Carpenter at the scene of multiple robberies. The Court, 5-4 (Roberts for the majority), held that obtaining seven days or more of CSLI is a Fourth Amendment search, requiring a warrant. The decision did not overrule Smith v. Maryland but pointedly distinguished it: cell-site records are detailed, comprehensive, and held by third parties only as a technological necessity, unlike the limited bank records or dialed-number lists in earlier cases.

Carpenter opened the door to a broader rethinking of the third-party doctrine in the digital age, but the Court has not yet walked through it consistently. Lower courts are working out how Carpenter applies to email metadata, smart-home device data, web-browsing histories, and the like. Border searches of digital devices are particularly contested — the Court has not yet decided whether the broad border-search exception survives Riley v. California (2014), which required a warrant for cell-phone searches incident to arrest.

The administrative-search and special-needs doctrines

A significant body of Fourth Amendment doctrine carves out administrative searches: inspections of regulated industries (mines, firearms dealers, liquor stores), routine workplace searches by public employers, school searches of students by educators, drug testing of certain government employees and student athletes, and various airport, subway, and border checkpoint screenings. The Court has typically applied a balancing test rather than a warrant requirement, weighing the government's regulatory interest against the individual privacy interest, and finding the regulatory interest sufficient when the search is part of a general scheme rather than an investigation of specific wrongdoing.

The administrative-search doctrine sits uneasily with the warrant-clause text but reflects a practical reality: thousands of routine government inspections happen every day under regulatory regimes designed in advance, and treating each as a discrete Fourth Amendment search requiring individualized probable cause would render entire fields of regulation impossible. Critics argue the doctrine has expanded far beyond its initial limits — that mass screening, dragnet surveillance, and "consent" searches at points where consent is functionally unavailable (the airport gate) have eroded the Fourth Amendment's core protections. Defenders argue the doctrine simply recognizes that the warrant clause was designed for criminal investigation, not regulatory inspection, and that requiring individualized probable cause everywhere would be both impractical and constitutionally unsupported. The doctrine is a place where text, history, and practical necessity all pull in different directions.

Stop-and-frisk in practice

Terry gave police a tool that has been put to enormous use. New York City's stop-and-frisk program at its 2011 peak conducted roughly 685,000 stops in a single year, in a city of about 8 million residents, with about 87% of those stopped being Black or Hispanic. Floyd v. City of New York (2013) found in federal district court that the program as implemented violated the Fourth and Fourteenth Amendments. The program was substantially curtailed; reported crime statistics did not increase materially in the years following. Critics of the curtailment argue the data is harder to read than that summary suggests; defenders point to it as evidence that aggressive Terry policing was not, in fact, the load-bearing element of public safety its proponents claimed. The empirical question — what does proactive policing actually do, and at what cost in liberty and racial equity? — is contested in social-science literature and is one of the points where Fourth Amendment doctrine, civil-rights doctrine, and policy questions converge.

The Fifth Amendment: self-incrimination, double jeopardy, due process, takings

The Fifth Amendment is doing four things at once.

Self-incrimination. "No person . . . shall be compelled in any criminal case to be a witness against himself." Miranda v. Arizona (1966) translated that protection into the famous warnings: a suspect in custodial interrogation must be informed of the right to remain silent, that anything said can be used against him in court, the right to counsel, and the right to appointed counsel if indigent. Dickerson v. United States (2000) reaffirmed Miranda as constitutional rather than merely supervisory, and Vega v. Tekoh (2022) limited civil remedies for Miranda violations while leaving the suppression remedy intact.

Double jeopardy. "Nor shall any person be subject for the same offence to be twice put in jeopardy of life or limb." A defendant cannot be tried twice for the same offense after acquittal, conviction, or (with limited exceptions) a mistrial declared without manifest necessity. The dual-sovereignty doctrine permits successive prosecutions by federal and state governments for the same conduct (Gamble v. United States 2019), which has been criticized as a workaround of the protection but reaffirmed.

Due process. "Nor be deprived of life, liberty, or property, without due process of law." This clause does double duty: procedural due process requires fair process — notice and an opportunity to be heard — before government deprivation of protected interests; substantive due process is the much more contested doctrine that some "fundamental" rights are protected from government infringement regardless of process. The Ninth Amendment section below treats substantive due process at length.

Takings. "Nor shall private property be taken for public use, without just compensation." The clause permits the government to take private property by eminent domain, but only for public use and only with payment of fair market value. Kelo v. City of New London (2005) held, 5-4, that "public use" includes economic-development takings — a city could condemn private homes and transfer them to a private developer if the city judged the resulting economic activity to be in the public interest. The decision was constitutionally significant but politically explosive: 45 states subsequently passed laws restricting eminent-domain authority beyond what Kelo permitted, in one of the more remarkable bipartisan responses to a Supreme Court decision in modern history. The Kelo aftermath illustrates how the political process can supplement (or supersede) constitutional doctrine: the Court permitted the practice; the people, through their states, mostly forbade it.

Civil asset forfeiture

The Fifth Amendment's takings and due-process protections are also implicated in civil asset forfeiture — the practice by which law enforcement seizes property believed to have been used in or derived from criminal activity, and forfeits it through civil proceedings against the property itself. Because the proceeding is technically against the property and not the owner, conventional criminal-procedure protections do not fully apply. The owner often must prove the property's innocence to recover it, sometimes years later, sometimes after spending more in legal fees than the property is worth.

Civil-asset-forfeiture reform is one of the genuinely cross-ideological issues in current civil-liberties law. The American Civil Liberties Union and the Institute for Justice — organizations that disagree about almost everything — have collaborated on litigation challenging the practice. Timbs v. Indiana (2019) held that the Eighth Amendment's Excessive Fines Clause is incorporated against the states and applies to civil asset forfeitures. The case involved a $42,000 Land Rover seized after a $400 drug conviction. About half the states have enacted reforms requiring criminal conviction before forfeiture; federal reform has been repeatedly proposed but not enacted. This is one of those areas where the standard partisan map breaks down — a fact the chapter notes precisely because it cuts against the assumption that civil liberties are a partisan football.

The Sixth Amendment: trial rights

"In all criminal prosecutions, the accused shall enjoy the right to a speedy and public trial, by an impartial jury of the State and district wherein the crime shall have been committed . . . to be informed of the nature and cause of the accusation; to be confronted with the witnesses against him; to have compulsory process for obtaining witnesses in his favor, and to have the Assistance of Counsel for his defence."

Six rights packed into one paragraph.

Counsel. Powell v. Alabama (1932) required state-appointed counsel in capital cases. Gideon v. Wainwright (1963) extended the right to all felony cases. Argersinger v. Hamlin (1972) extended it to any case resulting in actual imprisonment. The right is real, but the public-defender systems that implement it are chronically underfunded — caseloads of 200, 300, or more felony cases per attorney per year are common, far above the standards established by the American Bar Association. The right to counsel is one of the doctrinally robust civil liberties whose practical implementation is often weakest.

Speedy and public trial. Klopfer v. North Carolina (1967) incorporated the speedy-trial right; the Speedy Trial Act of 1974 implements it in federal criminal cases. The public-trial right is generally thought to apply to the press and public as well as to the defendant.

Impartial jury. Duncan v. Louisiana (1968) incorporated the jury-trial right for serious offenses (those punishable by more than six months' imprisonment). Batson v. Kentucky (1986) prohibited racially discriminatory peremptory strikes; the doctrine has been repeatedly applied (and debated for under-enforcement). Ramos v. Louisiana (2020) required jury unanimity in state criminal cases, ending a long-tolerated Louisiana and Oregon practice.

Confrontation. Crawford v. Washington (2004) restored a robust reading of the Confrontation Clause: testimonial hearsay (statements made under circumstances that would lead an objective witness to reasonably believe the statement would be available for use later in trial) is inadmissible unless the witness is unavailable and the defendant had a prior opportunity to cross-examine. Crawford unsettled decades of more relaxed practice and has produced a complicated body of case law working out what counts as "testimonial."

The plea-bargaining reality

The Sixth Amendment is structured around a trial. About 95% of federal criminal convictions and around 94% of state criminal convictions are obtained by guilty plea, not by trial. The trial right is real but, in practice, rarely exercised. Plea bargaining is the dominant mode of American criminal adjudication.

The Court has held that the Sixth Amendment's right to counsel includes effective assistance during plea negotiations (Lafler v. Cooper 2012; Missouri v. Frye 2012). But the Court has not held that the dramatic gap between trial-conviction sentences and plea-conviction sentences — sometimes 4x or more — itself violates due process or the right to trial. The "trial penalty," as defense lawyers and reform advocates call it, is one of the genuine tensions between the constitutional design (which assumes most cases go to trial) and the operational reality (which assumes most cases plead).

The Eighth Amendment: cruel and unusual

"Excessive bail shall not be required, nor excessive fines imposed, nor cruel and unusual punishments inflicted."

The Cruel and Unusual Punishments Clause has been the doctrinal vehicle for most modern Eighth Amendment litigation.

Capital punishment. Furman v. Georgia (1972) struck down then-existing state death-penalty statutes for arbitrary and capricious application; Gregg v. Georgia (1976) reauthorized capital punishment under more carefully drawn statutes. Subsequent cases have narrowed the death penalty's permissible scope: not for crimes against individuals other than murder (Kennedy v. Louisiana 2008), not for offenders with intellectual disability (Atkins v. Virginia 2002), not for offenders under 18 at the time of the crime (Roper v. Simmons 2005). The constitutionality of capital punishment in principle has not been overturned, but its implementation in practice has been substantially constrained — and, separately, public support and state-level use have both declined since the 1990s.

Conditions of confinement. Estelle v. Gamble (1976) held that deliberate indifference to serious medical needs of prisoners violates the Eighth Amendment. Brown v. Plata (2011) upheld a court order requiring California to reduce its prison population because severe overcrowding had produced unconstitutional conditions. The Eighth Amendment is the source of most prisoners'-rights litigation.

Excessive fines. Timbs v. Indiana (2019) incorporated the Excessive Fines Clause against the states (mentioned above in the civil-asset-forfeiture discussion).

Method of execution. Most modern Eighth Amendment death-penalty litigation has focused not on capital punishment in principle but on its method. Baze v. Rees (2008) and Glossip v. Gross (2015) sustained current lethal-injection protocols against challenges grounded in claims about pain risk. The Court has consistently declined to treat methodological challenges as opportunities to revisit Gregg. Justices Breyer (before retirement), Sotomayor, and Ginsburg argued in dissent that thirty years of capital practice had revealed problems — wrongful convictions documented through DNA exoneration, racial disparities in sentencing, geographic concentration in a small subset of counties — that warranted reconsideration of capital punishment's constitutionality. The argument has not commanded a majority and as of 2026 is unlikely to.

The Eighth Amendment was one of the constitutional provisions discussed in Trump v. United States (2024), the presidential-immunity decision. The principal holding concerned executive-branch criminal liability, but the secondary discussion of how cruel-and-unusual-punishment doctrine applies (or does not apply) to certain categories of executive action will be worked out in subsequent litigation.

Excessive bail. The Eighth Amendment's bail clause has produced surprisingly little Supreme Court doctrine. United States v. Salerno (1987) upheld the Bail Reform Act of 1984's preventive-detention provisions, holding that bail need not be available where there is a serious danger to the community. The doctrinal status of cash bail — the practice of conditioning pretrial release on payment of a deposit — is contested in current litigation in lower federal courts and in state courts, with reform advocates arguing that wealth-based detention violates equal protection and substantive due process even if it does not violate the bail clause itself. Several states (New Jersey, Alaska, New Mexico, Illinois in part) have enacted bail-reform legislation eliminating or substantially restricting cash bail; the empirical evidence on the consequences for public safety and court appearance rates is genuinely mixed and contested. Note again the structural pattern: when the Constitution's protection is incomplete, reform happens (or doesn't) at the state legislative and political level.

The Ninth Amendment, unenumerated rights, and substantive due process

"The enumeration in the Constitution, of certain rights, shall not be construed to deny or disparage others retained by the people."

The Ninth Amendment is the textual basis (or at least one of the textual bases) for the proposition that the Constitution protects rights not specifically listed in it. The doctrine of substantive due process — that the Fourteenth Amendment's Due Process Clause protects "fundamental" rights against state infringement, not merely the listed Bill of Rights protections — has been the doctrinal vehicle for most of these cases.

Griswold v. Connecticut (1965) struck down a state law banning contraception as applied to married couples, finding a "right of privacy" in the "penumbras and emanations" of the First, Third, Fourth, Fifth, and Ninth Amendments. Eisenstadt v. Baird (1972) extended the right to unmarried couples. Roe v. Wade (1973) extended the doctrine to abortion under a trimester framework. Lawrence v. Texas (2003) struck down sodomy laws as a violation of the constitutional liberty interest. Obergefell v. Hodges (2015) extended marriage to same-sex couples on substantive due process and equal protection grounds.

Dobbs v. Jackson Women's Health Organization (2022) overruled Roe and Casey, holding that the Constitution does not confer a right to abortion because the right is not "deeply rooted in this Nation's history and tradition." The decision returned the regulation of abortion to the political process at the state level. Following Dobbs, states have adopted a wide range of approaches: total or near-total bans (Texas, Mississippi, others), 6- or 15-week limits, weeks-of-viability frameworks similar to pre-Dobbs doctrine, and constitutional amendments protecting abortion access (Michigan, Ohio, others, by ballot initiative).

The doctrinal status of the Dobbs decision is itself contested. Defenders argue that Roe lacked constitutional grounding from the day it was decided (a position held by some progressive scholars on stare decisis grounds, including Justice Ginsburg before her elevation to the Court), and that Dobbs simply restored the political process to its proper role. Critics argue that Dobbs misread the relevant historical record, undermined fifty years of reliance interests, and signaled vulnerability for other substantive-due-process precedents (Justice Thomas's concurrence said as much). Both readings are held by serious legal scholars and judges. This book notes the doctrinal change without declaring which interpretation is correct, because that determination is properly the province of the reader.

The vulnerability question is particularly sharp for Lawrence (sodomy laws), Obergefell (same-sex marriage), and possibly Loving v. Virginia (interracial marriage), all of which rest in part on substantive due process. The Dobbs majority opinion explicitly disclaimed any implication for those decisions, but the dissent argued the doctrinal logic of the decision reaches them. As of 2026, the Court has not granted certiorari to revisit any of those cases, and Congress passed the Respect for Marriage Act (2022) to provide some statutory protection for same-sex and interracial marriages, but the constitutional question remains open in a way it did not before Dobbs.

A note on doctrinal method. Substantive due process has been criticized from both directions. From the originalist right, the doctrine looks like a form of judicial freelancing — courts protecting rights they happen to favor by reading them into a clause that, on its face, is procedural ("without due process of law" is a procedural phrase). From a different angle on the left, scholars including Akhil Amar and Jack Balkin have argued that many of the rights protected as "substantive due process" would be better grounded in the Privileges or Immunities Clause, the Equal Protection Clause, or the Ninth Amendment — clauses with stronger textual claim to substantive content. Dobbs did not endorse either of these alternative groundings; it overruled the substantive-due-process precedent at issue without offering a replacement framework for the broader doctrinal field. That leaves the post-Dobbs doctrinal landscape unsettled in a way that will keep constitutional scholars busy for at least a generation.

The implicit right to privacy

The word "privacy" appears nowhere in the Constitution. The Court has constructed a constitutional right to privacy from the text and structure of the Fourth Amendment (informational privacy from government surveillance), the Fifth Amendment (privilege against compelled disclosure), the Ninth Amendment (unenumerated rights), and the Fourteenth Amendment (substantive due process). After Dobbs, the future of the constitutional right to privacy is more uncertain than it has been at any point since Griswold.

Outside the constitutional context, privacy is also a matter of statute and tort: the Privacy Act of 1974 (federal records), HIPAA (health records), GLBA (financial records), state common-law privacy torts, state biometric-information laws (Illinois's BIPA, with active litigation), and sectoral state laws (California's CCPA/CPRA, Virginia's CDPA, Colorado's CPA). The legal protection of privacy is currently a patchwork. Comprehensive federal privacy legislation has been proposed in successive Congresses without enactment.

National security, war, and the limits of liberty in emergency

The Bill of Rights does not, by its terms, suspend during war or emergency. But the practical history of American civil liberties shows that emergencies test the protections in ways ordinary times do not.

The Sedition Act of 1798, passed during quasi-war with France, criminalized criticism of the federal government. It was allowed to expire, and is now widely understood as an early constitutional failure. The Civil War saw President Lincoln's suspension of habeas corpus and military trials of civilians (Ex parte Milligan 1866 limited the practice). World War I produced the Espionage Act of 1917 and the Sedition Act of 1918, used to prosecute opponents of the war (the cases that gave rise to Justice Holmes's "clear and present danger" formulation). World War II saw the internment of approximately 120,000 Japanese Americans, upheld by the Supreme Court in Korematsu v. United States (1944) — a decision that the Court has since treated as a constitutional embarrassment. Trump v. Hawaii (2018) explicitly described Korematsu as "gravely wrong the day it was decided."

The Cold War produced loyalty oaths, blacklists, and the McCarthy hearings; the doctrinal correction came slowly, with cases like Yates v. United States (1957) and Brandenburg (1969) narrowing earlier sedition-era precedents.

The post-9/11 era has produced its own set of liberty-versus-security questions. The Foreign Intelligence Surveillance Act (FISA), enacted in 1978 and substantially amended after 9/11, governs national-security surveillance through a specialized court system. The Patriot Act (2001) expanded surveillance authorities, and several of its provisions sunset or were modified after the Snowden disclosures (2013) revealed the scope of bulk telephone-metadata collection under Section 215. The USA Freedom Act (2015) ended the bulk-collection program. Section 702 of FISA, which permits warrantless surveillance of non-U.S. persons reasonably believed to be located outside the United States, was reauthorized in 2024 with reforms but remains controversial because of "incidental" collection of U.S. persons' communications.

The pattern across two centuries is consistent: emergencies expand executive power and contract civil-liberties protections; the protections are restored, sometimes incompletely, after the emergency passes; and the lessons of past contractions are imperfectly applied to new emergencies. A serious civil-libertarian, of any political stripe, recognizes that the test of constitutional commitment is not how the protections fare in calm times but how they fare under pressure.

Bringing it together

Civil liberties, properly understood, are a discipline of self-restraint by majorities. The First Amendment does not protect the speech everyone agrees is fine — that speech needs no protection. It protects the speech that is hated, in the moment it is hated, by the people who would otherwise punish it. The Fourth Amendment does not restrain searches that are obviously reasonable — those searches happen anyway. It restrains searches that the government wants to do but cannot constitutionally justify. The Ninth Amendment does not protect rights that are politically popular — those rights are protected by ordinary legislation. It guards against the tendency of every majority to assume that whatever it has not already authorized must be forbidden.

That discipline is uncomfortable. It protects speech you find repugnant, defendants you suspect are guilty, religious practices you find strange, and choices you would not make for yourself. It is uncomfortable for the left when it protects a Klan rally; it is uncomfortable for the right when it protects a flag-burning; it is uncomfortable for everyone when it protects something they think obviously should be illegal. The discomfort is the doctrine working as designed.

The Roberts Court has reshaped several major civil-liberties doctrines in the past decade — Second Amendment, religion, abortion, administrative search. Some of these changes will settle into stable doctrine; others will be contested in the next decade. A reader who finishes this chapter should be able to read a Supreme Court opinion in any of these areas, identify the doctrinal test the Court is applying, identify the strongest argument on each side, and recognize when the doctrinal change is being defended on principled grounds versus when it is being defended on results-oriented grounds. Both kinds of defense exist, on every side, on every issue. The civil-libertarian habit of mind is to keep them separate.

The next chapter — civil rights — turns from the question what may the government not do to the question whom must the government treat equally. The two questions are distinct, but they share a common foundation: both are commitments by an American majority to limit the operation of majority power in the name of something the majority believes is more important than its own preference of the moment. That belief is the reason the Bill of Rights exists at all.

Your Democracy Audit (Chapter 5 task)

Identify any active civil-liberties litigation in the federal district court covering your congressional district. Useful starting points: the U.S. Courts' PACER docket system (registration required, fee per search), the ACLU state affiliate's docket page, the Institute for Justice's case docket, the Foundation for Individual Rights and Expression (FIRE) campus-speech docket, and the Becket Fund religious-liberty docket. For one active case, identify (a) the constitutional provision invoked, (b) the doctrinal test the case will turn on, (c) the strongest argument on each side, and (d) the precedents most likely to control. Save the case identifier and your notes; you will return to it in Chapter 14 when we consider how the federal courts work.

Continue with exercises.md, case-study-01.md, case-study-02.md, key-takeaways.md, further-reading.md, and quiz.md.