The Fourth Amendment

Probable cause is the legal standard required before police can obtain a warrant or, in many cases, conduct a warrantless search or arrest. A court considers whether the known facts and circumstances would lead a reasonably prudent person to believe that a crime has been committed or that evidence of a crime will be found in the place to be searched. It is more than a hunch or suspicion but less than certainty or proof beyond a reasonable doubt.

The requirement is written into the amendment itself: no warrant may issue except on probable cause, supported by oath or affirmation, and particularly describing the place to be searched and the persons or things to be seized. The particularity rule answered the colonial general warrants described on the pre-Constitution history page.

Warrants must come from a neutral and detached magistrate. In Coolidge v. New Hampshire (1971), the Court threw out a warrant signed by the state attorney general, who was acting as a justice of the peace while also directing the investigation.

Arrests follow related rules. Police may arrest someone in public on probable cause without a warrant, but in Payton v. New York (1980) the Court held they generally need a warrant to enter a home to make a routine arrest. A person arrested without a warrant is entitled to a prompt judicial check of probable cause, usually within 48 hours, in the court system.

In Illinois v. Gates (1983), the Supreme Court replaced an older two-part test with a totality of the circumstances standard. Judges are to consider all available information, including tips from informants, police observations, and circumstantial evidence, as a whole, rather than evaluating each piece in isolation. This standard gives courts more flexibility and has generally been viewed as favorable to law enforcement.

The older approach, known as the Aguilar-Spinelli test after cases from 1964 and 1969, required magistrates to evaluate an informant's tip on two separate grounds: how the informant knew the information and why the informant should be believed. A weakness on either ground could defeat probable cause.

Gates began with an anonymous letter to police in Bloomingdale, Illinois, in 1978, describing how a couple drove drugs back from Florida. Officers confirmed the travel details before seeking a warrant. Justice William Rehnquist wrote for the 6-3 majority that the corroborated tip was enough.

The Court has applied the standard in both directions. Florida v. J.L. (2000) held that an uncorroborated anonymous tip that a person carries a gun does not justify a stop, while Florida v. Harris (2013) unanimously held that a trained drug-detection dog's alert can establish probable cause. Both are part of the Supreme Court's criminal procedure caselaw.

Reasonable suspicion is a lesser standard than probable cause. Established in Terry v. Ohio (1968), it allows police to briefly stop and pat down a person if they have specific, articulable facts that suggest criminal activity may be afoot. Reasonable suspicion does not require probable cause and does not require a warrant. It must be based on something more than a mere hunch, but it can be based on an officer's experience and training in interpreting behavior that might seem innocuous to a layperson.

Later cases show how little it can take. In Illinois v. Wardlow (2000), the Court held 5-4 that a person's unprovoked flight on seeing police in a high-crime area can supply reasonable suspicion, and United States v. Sokolow (1989) allowed a stop based on a combination of travel patterns that matched a drug courier profile.

Traffic stops need only reasonable suspicion of a traffic violation. In Whren v. United States (1996), a unanimous Court held that the officer's real motive does not matter if a violation occurred, and Heien v. North Carolina (2014) upheld a stop based on a reasonable misreading of a brake-light law.

There are limits on how long a stop can last. In Rodriguez v. United States (2015), the Court held 6-3 that police may not extend a completed traffic stop to wait for a drug-sniffing dog without separate reasonable suspicion. Technology-driven policing raises related questions covered on the privacy and surveillance page.

If a person voluntarily consents to a search, no warrant is required. The consent must be voluntary, meaning not the product of coercion, but police are not required to inform you that you have the right to refuse. Courts have held that the presence of uniformed officers does not automatically make consent involuntary, a finding criticized by civil liberties advocates who argue that most people feel they cannot practically refuse.

The leading case is Schneckloth v. Bustamonte (1973), which held that voluntariness depends on the totality of the circumstances, and that knowing one could refuse is only one factor. In Florida v. Bostick (1991), the Court applied a similar test to police sweeps of buses, asking whether a reasonable person would feel free to decline.

Consent can come from someone other than the suspect. A roommate or spouse with authority over shared space may consent, and police may rely on someone who reasonably appears to have that authority. But in Georgia v. Randolph (2006), the Court held that a physically present occupant's refusal overrides a co-occupant's consent.

Consent also has a scope. Florida v. Jimeno (1991) held that consent to search a car for drugs reasonably included a closed paper bag on the floor, while a person may limit or withdraw consent before the search is complete. Rights at the station house, including the right to remain silent, are covered on the Fifth Amendment page.

When police lawfully arrest someone, they may search the person and the area within that person's immediate reach without a warrant. The rationale is officer safety and preventing destruction of evidence. However, as the Supreme Court held in Riley v. California (2014), this exception does not extend to the digital contents of a cell phone, because those contents pose no safety threat and cannot be destroyed by a quick reach.

Chimel v. California (1969) set the modern boundary. Before it, courts had allowed police to search an entire house after an arrest inside; Chimel limited the search to the arrested person and the area within his immediate control, sometimes called the grab area.

For vehicles, Arizona v. Gant (2009) held 5-4 that police may search a car's passenger compartment after an arrest only if the arrestee is unsecured and within reach of it, or if it is reasonable to believe the car holds evidence of the crime of arrest.

Riley was unanimous and was decided together with a Boston case involving an older flip phone. Two years later, Birchfield v. North Dakota (2016) drew a similar line for drunk-driving arrests: police may require a breath test without a warrant but need one for a blood test. Digital privacy is also covered on the privacy and surveillance page.

Because vehicles are mobile and can be moved before a warrant is obtained, and because people have a reduced expectation of privacy in their cars, police may search a vehicle without a warrant if they have probable cause to believe it contains contraband or evidence of a crime. This exception, established in Carroll v. United States (1925), has been extended broadly and allows search of the entire vehicle, including containers, if probable cause exists.

Carroll was a Prohibition case. In 1921, federal agents stopped George Carroll and John Kiro on a Michigan highway and found 68 bottles of liquor hidden behind the car's upholstery. Chief Justice William Howard Taft wrote the opinion, now one of the landmark Supreme Court cases on police searches.

The exception has grown over time. California v. Carney (1985) applied it to a motor home parked in a lot, California v. Acevedo (1991) allowed police with probable cause to open containers found in the car, and Wyoming v. Houghton (1999) extended it to passengers' belongings.

It stops at the home. In Collins v. Virginia (2018), the Court held 8-1 that the exception did not let an officer walk up a private driveway beside a house to lift a tarp covering a stolen motorcycle, because that area is part of the home's protected curtilage. Related privacy protections are discussed on the privacy and surveillance page.

If police are lawfully present somewhere and they observe evidence of a crime or contraband in plain view, they may seize it without a warrant. The officer must have a lawful right to be in the location, the incriminating nature of the item must be immediately apparent, and the officer must not have moved the object to observe it. This doctrine has generated significant litigation over what counts as immediately apparent.

The doctrine took its modern form in Coolidge v. New Hampshire (1971). A plurality there required that the discovery be inadvertent, but in Horton v. California (1990) the Court dropped that requirement, holding that an officer's expectation of finding the item does not invalidate a plain-view seizure.

Moving an object can turn plain view into a search. In Arizona v. Hicks (1987), officers lawfully inside an apartment after a shooting moved stereo equipment to read its serial numbers, and the Court held 6-3, in an opinion by Justice Scalia, that this was a separate search requiring probable cause.

The Court has recognized a parallel plain-feel rule. In Minnesota v. Dickerson (1993), it held that an officer frisking for weapons may seize contraband whose identity is immediately apparent by touch, but suppressed the drugs in that case because the officer had squeezed and manipulated the lump to identify it. See other rulings on the landmark Supreme Court cases page.

When an emergency makes getting a warrant impractical, police may act without one. Recognized exigencies include hot pursuit of a fleeing suspect, imminent destruction of evidence, and situations where someone inside is in danger. Courts look at whether the urgency was real and whether the police themselves created the exigency through their own conduct.

Hot pursuit has limits. In Lange v. California (2021), the Court held, with no dissent, that chasing someone suspected only of a misdemeanor does not automatically justify entering a home without a warrant; officers must assess whether a genuine emergency exists in each case.

The police-created exigency rule is narrow. In Kentucky v. King (2011), officers knocked on an apartment door, smelled marijuana, and heard what sounded like evidence being destroyed. The Court held 8-1 that the entry was lawful because the officers had not threatened or engaged in any unlawful conduct before the exigency arose.

Emergencies involving people are treated differently from routine policing. Brigham City v. Stuart (2006) allowed officers to enter a home to stop a fight in progress, but Caniglia v. Strom (2021) unanimously rejected a broader community caretaking exception that would let police enter homes to seize guns without a warrant or emergency. In Case v. Montana (January 14, 2026), a unanimous Court reaffirmed that officers may enter a home without a warrant when they have an objectively reasonable basis to believe someone inside needs emergency help, and it declined to require probable cause. All three are Supreme Court decisions.

Terry v. Ohio (1968) established that police may briefly detain a person and conduct a pat-down for weapons based on reasonable suspicion alone, without probable cause and without a warrant. The scope is limited: the detention must be brief, and any frisk is limited to the outer clothing for weapons only. Terry stops have been the subject of extensive civil rights debate, particularly regarding racial disparities in how they are applied.

In Terry, Cleveland detective Martin McFadden, a 39-year veteran, watched the men pace past a store window about two dozen times. Chief Justice Earl Warren wrote for the 8-1 majority, and Justice William O. Douglas dissented, warning that the ruling gave police more power to seize a person than a judge had to authorize a search.

Later cases broadened the practice. Hiibel v. Sixth Judicial District Court (2004) held that a state may require a person lawfully stopped to give his name, and in Utah v. Strieff (2016), the Court held 5-3 that evidence found after an unlawful stop was admissible because the officer discovered an outstanding arrest warrant during the stop.

The practice has been challenged on equal protection grounds. In Floyd v. City of New York (2013), a federal judge found that New York City's stop-and-frisk program, which peaked at about 685,000 stops in 2011, violated both the Fourth Amendment and the Fourteenth Amendment because of how it targeted Black and Latino residents. The civil rights page covers the broader history.

At international borders and their functional equivalents such as airports, the government has broad authority to search travelers and their belongings without a warrant or probable cause. Courts have historically recognized that the government's interest in controlling what enters the country justifies reduced Fourth Amendment protections at the border. Whether this exception extends to full forensic searches of digital devices at the border is an active and unresolved legal question.

The Court has repeatedly upheld suspicionless searches at the border. United States v. Ramsey (1977) allowed customs officers to open international mail, and United States v. Flores-Montano (2004) allowed agents to remove and disassemble a car's gas tank without any suspicion. More intrusive detentions, such as holding a suspected drug smuggler for many hours, require reasonable suspicion.

The power extends inland in limited ways. Border Patrol may operate fixed immigration checkpoints within a reasonable distance of the border, which the Court upheld in United States v. Martinez-Fuerte (1976), but roving patrols need reasonable suspicion to stop a car. The immigration page explains how border enforcement is organized.

Customs and Border Protection, part of the Department of Homeland Security, searches tens of thousands of electronic devices at the border each year. Federal appeals courts disagree on the rules: the Ninth Circuit requires reasonable suspicion for a forensic search of a phone or laptop, while the Eleventh Circuit requires no suspicion at all.

The exclusionary rule bars evidence obtained through an unconstitutional search or seizure from being used in a criminal prosecution. It was established by the Supreme Court in Weeks v. United States (1914) for federal courts and extended to state courts in Mapp v. Ohio (1961). The rule's purpose is not to benefit the defendant but to deter police misconduct by removing the incentive to violate the Fourth Amendment.

Between Weeks and Mapp, the Court took a middle position. In Wolf v. Colorado (1949), it held that the Fourth Amendment's core protection applied to the states through the Fourteenth Amendment, but did not require states to exclude illegally obtained evidence. Mapp overruled that part of Wolf.

The rule does not apply everywhere. The Court has declined to extend it to grand jury proceedings, civil tax cases, and deportation hearings, and prosecutors may use illegally obtained evidence to challenge a defendant's credibility if the defendant testifies falsely at trial.

The Court has also narrowed when violations lead to exclusion. Hudson v. Michigan (2006) held 5-4 that violating the knock-and-announce rule does not require suppression, and Herring v. United States (2009) held that evidence found because of a negligent police recordkeeping error need not be excluded. Mapp and Weeks appear among the landmark Supreme Court cases.

The fruit of the poisonous tree doctrine extends the exclusionary rule to evidence derived from an unconstitutional search, not just the primary evidence itself. If an illegal search leads police to a witness who then provides a confession, the confession may also be suppressed as fruit of the poisonous tree. The doctrine was established in Silverthorne Lumber Co. v. United States (1920).

In Silverthorne, federal agents illegally seized a lumber company's books, returned the originals after a court order, and then tried to use copies they had made to force production of the same records. Justice Oliver Wendell Holmes wrote that knowledge gained through the government's own wrong cannot be used by it. The phrase fruit of the poisonous tree came later, from Justice Felix Frankfurter in Nardone v. United States (1939).

The doctrine has three main exceptions. Evidence is admissible if police also obtained it from an independent lawful source, if they inevitably would have discovered it lawfully, or if the connection to the illegal search is so weakened, or attenuated, that the taint has dissipated. In Nix v. Williams (1984), a murder victim's body was admitted because search teams would have found it anyway.

The doctrine also applies to confessions obtained after illegal arrests, which intersects with the protections against self-incrimination on the Fifth Amendment page. In Brown v. Illinois (1975), the Court held that Miranda warnings alone do not automatically break the chain between an illegal arrest and a later confession.

In United States v. Leon (1984), the Supreme Court created the good faith exception: evidence obtained through an unconstitutional search may still be used if officers acted in objectively reasonable reliance on a warrant that turned out to be defective. If police genuinely believed a warrant was valid when they executed it, excluding the evidence provides little deterrent benefit. Critics argue this exception has significantly narrowed the exclusionary rule's protective power.

Leon was decided 6-3 on July 5, 1984, in an opinion by Justice Byron White, together with a companion case from Massachusetts in which a judge had used the wrong warrant form. Justice William Brennan's dissent argued that the exception would encourage careless warrant applications.

The Court has extended the exception several times: to officers relying on a statute later struck down (1987), on court clerks' database errors (1995), and on binding appellate precedent that was later overruled (2011).

Leon itself set limits. The exception does not apply if the affidavit contained knowing or reckless falsehoods, if the magistrate abandoned a neutral role, if the warrant was so lacking in probable cause that reliance was unreasonable, or if it was facially deficient. Several states, including New Jersey and Pennsylvania, reject the good faith exception under their own state constitutions.

The doctrine emerged in United States v. Miller (1976), where the Court held that bank customers have no Fourth Amendment right in their financial records because they are voluntarily shared with the bank. Three years later, in Smith v. Maryland (1979), the Court held that the numbers you dial on a phone are not protected because you voluntarily share them with the phone company. The government could obtain both without a warrant.

Miller involved a Georgia man charged with running an illegal whiskey still, whose bank records the government obtained by subpoena; the decision was 7-2. In Smith, police asked the phone company to install a pen register recording the numbers dialed by a robbery suspect who had been making threatening calls to the victim, and the Court ruled 5-3.

Congress responded with laws that go further than the Constitution requires. The Right to Financial Privacy Act of 1978 limits access to bank records, and the Electronic Communications Privacy Act of 1986 requires court orders for pen registers. See how a bill becomes law.

Smith later became the legal foundation for the National Security Agency's bulk collection of Americans' telephone records, revealed by Edward Snowden in June 2013. Congress ended that bulk collection in the USA FREEDOM Act of 2015. The privacy and surveillance page covers the program in detail.

In Carpenter v. United States (2018), the Supreme Court ruled 5-4 that the government must obtain a warrant before accessing historical cell-site location information (CSLI), the data that shows which cell towers a phone connected to over time. Chief Justice Roberts wrote that this kind of data is different: it creates a near-perfect surveillance record of a person's movements over years. The majority declined to extend the third-party doctrine to this qualitatively different category of information and emphasized the decision was narrow.

The records covered 127 days and nearly 13,000 location points. Investigators had obtained them under the Stored Communications Act, which required only specific and articulable facts showing the records were relevant, a far lower standard than probable cause.

The case was decided June 22, 2018. Chief Justice Roberts was joined by the Court's four liberal justices, and all four dissenting justices wrote separately. Justice Neil Gorsuch's dissent proposed an entirely different approach based on property rights in one's own data.

Carpenter built on United States v. Jones (2012), in which the Court unanimously held that attaching a GPS tracker to a suspect's car and monitoring it for 28 days was a search. Several justices in Jones suggested that long-term tracking is a search even without physical contact. More on location tracking appears on the privacy and surveillance page.

The Carpenter majority explicitly declined to decide many related questions: whether the warrant requirement applies to real-time CSLI, to shorter-term historical CSLI, to other digital data held by third parties like email content, or to internet browsing records. Lower courts and state courts reached different conclusions on these questions, creating a patchwork of Fourth Amendment law in the digital realm. The Supreme Court answered one of them in Chatrie v. United States (June 29, 2026), holding that people keep a reasonable expectation of privacy in phone location records even for a short window, two hours in that case. The others still await the Court's guidance.

Geofence warrants, which ask a company for every device present in an area at a certain time, were a leading example. In 2024, the Fifth Circuit held that such warrants are unconstitutional general warrants, while other courts upheld their use. In Chatrie, the Supreme Court ruled 6-3 that obtaining location data this way is a search, but it did not decide whether such warrants can meet the Fourth Amendment's requirements. Google changed how it stores location history in 2023 and 2024, which sharply reduced its ability to answer such requests.

Email content has more settled protection. In United States v. Warshak (2010), the Sixth Circuit held that the government needs a warrant to obtain the contents of emails stored by a provider, and major email companies now require one. The Justice Department adopted a policy in 2015 requiring warrants for cell-site simulators, devices that mimic cell towers.

Foreign intelligence surveillance follows separate rules. Section 702 of the Foreign Intelligence Surveillance Act allows warrantless collection targeting foreigners abroad, which also sweeps in Americans' communications with them; Congress reauthorized it in April 2024 for two years and, after two short extensions in 2026, that authorization lapsed on June 12, 2026, though collection continues under court-approved certifications that run to March 2027. The privacy and surveillance page explains the debate.

Without the exclusionary rule, the Fourth Amendment would be a meaningless guarantee. If police can violate it without consequence, there is no incentive to comply. The rule deters unconstitutional searches by removing the benefit of evidence obtained illegally. No other remedy, whether civil suits against officers or departmental discipline, has proven as effective at changing police behavior.

Supporters point to the Court's own reasoning in Mapp, where Justice Tom Clark wrote that without exclusion the Fourth Amendment would be reduced to a mere form of words, a phrase borrowed from Justice Holmes. States that had tried other remedies before 1961 had seen few results.

The alternatives face practical obstacles. Lawsuits against officers are limited by qualified immunity, which shields them unless they violated clearly established law, and the Court has sharply narrowed lawsuits directly under the Constitution against federal agents, as in Egbert v. Boule (2022) and, most recently, Goldey v. Fields (2025).

Supporters also note that the rule's real effect is on police training. After Mapp, departments across the country began teaching officers warrant procedures and search law in detail. Suppression rulings are made by trial judges in the court system, where they serve as ongoing feedback to police.

Critics argue that the exclusionary rule imposes enormous costs, including releasing guilty defendants on technicalities, without reliably deterring police misconduct. Many offending officers are never disciplined, and juries and judges are often unaware that evidence was suppressed before trial. Alternative remedies such as civil rights lawsuits, internal discipline, independent oversight boards, and statutory exclusion rules could deter misconduct while keeping reliable evidence in cases. Several conservative and even some liberal legal scholars have called for the rule to be replaced or significantly limited.

The objection is old. In 1926, New York judge Benjamin Cardozo, later a Supreme Court justice, famously wrote that under the rule the criminal is to go free because the constable has blundered.

Chief Justice Warren Burger argued in a 1971 dissent that the rule should be replaced by a statutory remedy allowing people whose rights were violated to sue the government for damages, which he believed would compensate innocent victims of illegal searches who never face trial.

In Hudson v. Michigan (2006), Justice Antonin Scalia wrote that increased police professionalism and the availability of civil rights lawsuits reduced the need for exclusion in some situations. Critics of the rule hope the Supreme Court will continue in that direction.

Defenders of the traditional third-party doctrine argue that when people share data with companies, they voluntarily assume the risk that those companies will share it with the government. People understand, or should understand, that their phone location data, email headers, and browsing history are held by third parties. Extending Fourth Amendment protection to all such data would create an unworkable standard and significantly hamper legitimate law enforcement investigations. Carpenter should remain a narrow exception for comprehensive location tracking only.

Justice Anthony Kennedy's Carpenter dissent argued that the location records belonged to the wireless carrier, not the customer, and that the decision would unsettle long-established law. Justice Samuel Alito warned that it could call into question the government's use of ordinary subpoenas for business records.

Subpoenas for business records, which require no probable cause, are among the most common tools in fraud, drug trafficking, and public corruption investigations. Defenders of the doctrine argue that requiring warrants for all such records would slow or stop many investigations at their earliest stage.

They also argue that Congress, not the courts, is better placed to set privacy rules for new technology, as it did with the Stored Communications Act of 1986 and the CLOUD Act of 2018. See how Congress passes such laws in how a bill becomes law.

Critics argue that the third-party doctrine is incompatible with modern life. People cannot meaningfully opt out of sharing data with third parties when doing so would require giving up phones, banking, email, and internet access entirely. The doctrine was designed for specific disclosures like bank records and phone numbers. Applying it to comprehensive surveillance data that reveals intimate details of daily life goes far beyond its original purpose and effectively eliminates Fourth Amendment protection in the digital world.

Justice Sonia Sotomayor raised the concern in her concurrence in United States v. Jones (2012), writing that the doctrine is ill suited to the digital age, in which people reveal a great deal of information about themselves to third parties in the course of carrying out mundane tasks.

Justice Neil Gorsuch, dissenting in Carpenter from a different direction, argued that Smith and Miller should be reconsidered and that people may retain property rights in data they entrust to companies, much as someone who stores goods with a warehouse still owns them.

Some states have gone further than federal law. Utah passed a law in 2019 requiring a warrant for electronic data held by third parties, and Michigan voters amended their state constitution in 2020 to protect electronic data and communications from unreasonable searches, with nearly 89 percent approval.