The first half of the Fourth Amendment bans “unreasonable searches and seizures.” The second half, known as the Warrant Clause, states a set of basic requirements for search and arrest warrants—that they must be supported by an affidavit that establishes probable cause, and that they must describe both the location and objects of the search or the person to be seized.

On its face, the Warrant Clause would appear to be one of the most clearly written clauses in the Constitution. It requires that warrants be supported by probable cause, that the police officer seeking the warrant swear to the truth of the facts used to support the application for the warrant, and that, once issued, the warrant describe who is to be arrested, where the search is to take place and what the officer is allowed to look for. All this is plain from the text. Perhaps because they are so plain, the rules have been relatively easy for courts to apply, although close questions can arise on how specific a search warrant must be in describing the place to search and the items the police are looking for.

There are, though, two important questions the text does not answer, or at least does not answer clearly. Those questions have been the subject of a great deal of litigation and commentary: First, what does “probable cause” mean? Second, a trickier question, are officers ever required to obtain warrants in order to carry out a search or make an arrest, and if so, in what circumstances? The text leaves the question open, but it implies that the answer is that warrants are not required: the phrasing of the Warrant Clause limits warrants but does not mandate their use.

The first of these questions can be quickly answered. In Brinegar v. United States (1949), the Supreme Court defined “probable cause” as information that would lead “a man of reasonable caution” to believe “that an offense has been or is being committed.” In Illinois v. Gates (1983), the Court put it more succinctly, describing probable cause as “a fair probability” that evidence will be found in the place searched or that the person arrested committed a crime. Those definitions may sound too vague to be useful, but in practice the standard seems clear enough. In most cases “probable cause” means what the ordinary definition of “probable” would suggest: more likely than not. That “51 percent” standard does not always apply: in practice, courts seem to give the police a little more leeway when the crime being investigated is especially serious, and a little less when the crime seems minor. The Supreme Court itself has refused to quantify the degree of certainty needed to establish probable cause.

As with any vague standard, the phrase “probable cause” has occasioned a great deal of litigation and commentary, but the contested territory is small. All sides agree that the phrase means more than just a possibility, and less than a near-certainty. A clearer definition than that may be impossible.

The second question, whether warrants are ever required, is more complex. At first blush the question seems nonsensical. Of course, warrants are sometimes required; otherwise, why would the Fourth Amendment mention them? When the Fourth Amendment was written, the sole remedy for an illegal search or seizure was a lawsuit for money damages. Government officials used warrants as a defense against such lawsuits. Today a warrant seems the police officer’s foe—one more hoop to jump through—but at the time of the Founding, it was the constable’s friend, a legal defense against any subsequent claim. Thus it was perfectly reasonable to specify limits on warrants (probable cause, particular description of the places to be searched and the things to be seized) but never to require their use.

That is probably (though not clearly—some historians disagree) how the clause was understood when it was written. Like the state constitutional provisions on which it was modeled, the Fourth Amendment arose as a response to three famous cases decided in the 1760s. In each of those cases, agents of the Crown conducted very broad searches; in each, the agents had warrants authorizing the searches; finally, in none of the three searches did those warrants meet the requirements that were later spelled out in the Fourth Amendment. The point of the text was to forbid the kind of behavior seen in the three cases—not to require warrants, but to prevent the government from using them to justify overly broad searches.

The first of the three cases was Wilkes v. Wood (1763). John Wilkes was a London pamphleteer critical of the king’s ministers; he was also a Member of Parliament and perhaps the most popular man in England. One of the King’s secretaries issued a sweeping warrant, ordering the arrest of Wilkes and those associated with a pamphlet he had authored, as well as the seizure of all Wilkes’s books and papers. Wilkes sued, and won the then-staggering total of five thousand pounds. Wilkes v. Wood was a famous and celebrated case in the colonies, so much so that several towns were named after John Wilkes (as was Abraham Lincoln’s assassin).

The second case, Entick v. Carrington (1765), was similar. Like Wilkes, John Entick wrote pamphlets criticizing the government. As with Wilkes, one of the King’s underlings issued a warrant commanding officers to seize Entick and all his papers. As with Wilkes, the warrant extended to all Entick’s papers, not merely to those that might offer evidence of a crime. Entick also sued and won; the case was likewise famous in the colonies, prompting local officials to name several towns after the judge in Entick’s case—Lord Camden.

The third case is the famous Writs of Assistance Case (1761) in Boston. The warrant in that case authorized the search of any place in which the Crown’s agents thought smuggled goods might be hidden. The things to be seized were described, but the places to be searched were not. A number of Boston merchants challenged these “writs of assistance.” James Otis, representing the merchants, argued that the common law banned such “general warrants.” Otis lost his case, but his argument struck a chord in the increasingly rebellious colonies.

Historians generally agree that the Warrant Clause was written to adopt the decisions in Wilkes and Entick and the losing argument in the Writs of Assistance Case. General warrants, meaning both warrants not supported by probable cause and warrants that failed to describe the places or objects of the search, were banned. But the police (at that time, constables) were probably free to not use warrants at all. The reason that this last point is not entirely clear is that no one seems to have thought much about the question. When the Fourth Amendment was adopted, police forces did not yet exist (they arose in America beginning in the 1830s). A good deal of criminal investigation was conducted by private parties, with evidence turned over to the local constable or magistrate after the suspect was charged. Constables became involved only when it was time to make an arrest (and sometimes not even then), at which time they typically searched the arrestee’s person and home. It is clear that those actions did not require a warrant in 1791.

Thus the original understanding of the Warrant Clause was in one sense clear, and in one sense not. It was clear what the conditions were for a valid warrant—those conditions are spelled out in the Fourth Amendment’s text. It was not clear whether warrants were ever required (though they probably were not), because the issue had not arisen with any regularity.

Today’s Warrant Clause doctrine differs from the historical understanding in some important respects. That doctrine can be divided into two parts. The first deals with the conditions of a valid warrant. The second deals with when warrants are required.

The conditions of a valid warrant are relatively straightforward, because Warrant Clause doctrine continues to track the Fourth Amendment’s text. Probable cause and particular description are required, and the application for a warrant must be made under oath, as the text says. Another requirement is not mentioned in the text: early on, American courts decided that warrants should be issued only by judicial officers (usually, that means magistrates) and not by anyone in the prosecutor’s office or the executive branch of government more generally.

One important qualification to the Fourth Amendment language concerns probable cause. The Supreme Court has approved warrants not based on probable cause in some regulatory settings. Thus, in Camara v. Municipal Court (1967), housing inspectors were allowed to use what the Court called “administrative warrants”—orders authorizing the random selection of some buildings for code inspection. Such administrative warrants are sometimes used, as in Camara, to enforce building and fire codes, but not for much else.

The police are not allowed to use administrative warrants to investigate and enforce the criminal law, although information uncovered when executing these warrants may be used in a criminal prosecution. The justification of this state of affairs is that police officers investigating crime tend to have more power than other government officials: the police can break down doors, use force to subdue suspects, and, in some cases, they may destroy suspects’ property if that is a necessary consequence of the search for evidence. Other government officials tend not to have those powers. Consequently ordinary citizens tend not to find a building code inspection as frightening as a police search or arrest. The distinct legal requirements reflect those differences in official power and in the fear that such power inspires.

The second issue, when are warrants required, is more complicated. In summary, warrants are required when the police search a home or an office, unless the search must happen immediately, and there is no opportunity to obtain a warrant. Containers such as luggage that are not found in cars or at airport security check-points may be seized without a warrant, but normally cannot be searched without one. Warrants are also required for wiretaps and some computer searches—a special category covered by federal statute. And while the scope of the rule is not entirely clear, the Court has said that before the police can attach a GPS unit to a suspect’s car that will allow them to continuously track the car’s movement, a warrant is needed. Outside those categories warrants are almost never required.

There is a slightly more elaborate way to put the point. Until recently the Supreme Court said that warrants were required for all searches and seizures, save those that fell within some exception to that requirement. The classic statement of this rule, and the classic defense of a broad warrant requirement, was penned by Justice Robert H. Jackson in Johnson v. United States (1948). Today, the Court uses different language, emphasizing not the second half of the Fourth Amendment’s text, but the first (the ban on “unreasonable searches and seizures”). See Indianapolis v. Edmond (2000). Notwithstanding this change in legal rhetoric, the old categories, a warrant requirement with a list of exceptions, still exist. The scope of the requirement is defined by the many exceptions to it. The major ones are these:

1. Exigent circumstances. The police need not get a warrant when doing so is practically impossible, because of, for example, the risk that the evidence will be destroyed or moved, the risk that the suspect will flee, or if there is some danger created if the police do not act immediately.
2. Arrests outside the home. The police must have probable cause to justify the arrest, but they need not have a warrant.
3. Searches incident to arrest. This means a search of the arrestee’s person and any baggage he or she may be carrying; if the person is arrested in his car, the police may search the car if they have reason to believe that evidence related to the crime of arrest may be found there.
4. Inventory searches. The police may seize any belongings the arrestee has in his possession at the time of arrest (including his car), bring those items back to the police station, and make a record of them and their contents. So long as the police conduct the inventory according to standard procedures, they do not need to have any suspicion that the inventoried items contain evidence, but may use any such evidence that they find.
5. Automobiles. Cars, including their trunks, may be searched without warrants, as long as the searching officers have probable cause to believe that the car contains evidence of a crime or contraband.
6. Street stops and frisks. Officers are allowed to detain a suspect for a brief period, and to frisk him for weapons, given “reasonable suspicion” of criminal activity, a standard that is easier for the police to satisfy than probable cause.

In addition to these exceptions, there are several categories of searches that involve government officials other than police officers (e.g., searches of lockers by school principals, and government employers searching employees’ desks), or government interests separate from the interest in criminal law enforcement (e.g., searches of vehicles at the nation’s borders, searches of persons and baggage at airports). Such searches generally do not require warrants, and often do not require probable cause.

That list of exceptions and special categories aside, other searches and seizures do require warrants. Notice, however, that the major categories of searches and seizures that do not appear on the above list are searches of homes, arrests within homes, searches of private offices or other privately owned buildings (other than for fire inspection and the like), and wiretaps. The overwhelming majority of search and arrest warrants are issued in such cases.

A generation ago those propositions were widely contested; the scope of the warrant requirement was the subject of a great deal of litigation, including a number of Supreme Court decisions. That is no longer the case. Today Fourth Amendment litigation focuses on warrantless searches and seizures. The Searches and Seizures Clause—the first half of the Fourth Amendment’s text—is now the primary source of Fourth Amendment litigation and commentary.