Concept
Reasonable Suspicion
Terry stop and frisk
Reasonable Suspicion
Once a seizure exists, the officer needs a justification for it, and for a brief investigative stop that justification is reasonable suspicion. In Terry v. Ohio the Court held that police may stop and briefly detain a person for investigative purposes if the officer has a reasonable suspicion, supported by articulable facts, that criminal activity may be afoot, even without probable cause.1 The suspicion may concern a crime that is underway or one already completed, since the standard reaches both criminal activity and involvement in a crime already committed. For a crime that is over, the Court has confined stops to felonies and crimes that threaten public safety, where prompt solution of the crime and detention of the suspect serve a strong public interest.2
The standard sits between two boundaries. The lower boundary is that the officer must be able to articulate something more than an "inchoate and unparticularized suspicion or 'hunch,'" because the Fourth Amendment requires some minimal level of objective justification for the stop.3 The upper boundary is that the level of suspicion required is considerably less than proof of wrongdoing by a preponderance of the evidence, and it is less demanding than probable cause, which asks for a fair probability that contraband or evidence of a crime will be found.4 The gap runs in two directions. Reasonable suspicion can rest on information that differs in quantity or content from what probable cause demands, and it can also rest on information that is less reliable, so the two dimensions are weighed together and a less reliable source calls for more supporting information.5
Each boundary has its own reason. The upper one exists because a stop is brief and its purpose is to clarify an ambiguous situation, so an officer who had to establish probable cause before asking a single question would be barred from the very inquiry that might produce it.6 The lower one exists because a bare hunch gives a reviewing judge nothing to examine. If a feeling sufficed, any person on the street could be stopped at an officer's whim and no court could test the decision afterward, whereas a demand for objective facts lets the judge ask what the officer actually saw. Neither boundary can be drawn as a formula, because human behavior produces probabilities rather than certainties, and the Court has declined to reduce reasonable suspicion to a neat set of legal rules.7 Courts instead look at the whole picture, in which officers, like jurors, may draw common-sense conclusions from what they observe.8
The difference between the two levels of proof shows up when a description fits several people. Stopping a person near the scene of a recent robbery because the person matches the victim's general description is proper, even though an arrest would be improper, because the same description would fit others in the area.9
On an exam, the tempting error is to test a stop against the probable-cause standard that governs arrests. Because a stop needs only reasonable suspicion, a description that would fit several people can justify a stop though it could not justify an arrest, and the same facts fail only when the officer cannot point to anything objective at all. The subsections below take up how that premise is applied: what an officer must be able to point to, how the facts are weighed together, and what to make of conduct that has an innocent explanation.
Articulable Facts, Not Hunches
The first demand the standard makes is that the officer be able to point to facts. Reasonable suspicion must rest on articulable facts, meaning observations the officer can put into words and a court can examine, and a feeling that something is wrong does not count however sincerely the officer holds it.11 The requirement follows from the way stops are reviewed. A stop happens on the street without a warrant, so the only check on it comes later, when a judge asks whether the officer had enough to justify the intrusion. A judge can test a statement that two men walked past a store window a dozen times, because it describes something that either happened or did not. A judge cannot test a statement that the men looked like trouble, and a rule that accepted such a statement would leave every stop to the officer's unreviewable discretion.
In Terry v. Ohio the facts were of the describable kind. Detective Martin McFadden, a 39-year police veteran who had worked downtown Cleveland for thirty years, watched from 300 to 400 feet away as John Terry and Richard Chilton walked back and forth along Huron Road, paused repeatedly to look into the same store window, and conferred with each other, a routine they repeated roughly a dozen times over ten to twelve minutes.12 A third man briefly joined them and left, and the two then followed the path he had taken.13 McFadden concluded that the pair were casing the store for a robbery, and that conclusion sat on top of a sequence of observations he could recite in order. Each observation had a place, a count, and a duration, which is what lets a court see how the officer got from conduct to suspicion.
Those facts were articulable even though none of them was unlawful. Pacing, looking into a window, and talking are ordinary behavior, and the Court has since observed that the conduct justifying the stop in Terry was ambiguous and open to an innocent explanation, yet Terry allowed the officers to detain the men to resolve the ambiguity.14 What the officer needs, then, is a set of specific facts from which crime is a reasonable inference, and no single fact has to be incriminating on its own. How far innocent conduct can go in building suspicion is the subject of a later subsection. The point here is narrower: the officer must have observed something, specifically and often enough, to reason from.
Below that floor lies the hunch. In United States v. Sokolow the Court repeated that an officer must articulate more than an inchoate and unparticularized suspicion or hunch, and that the Fourth Amendment requires some minimal level of objective justification for making the stop.15 The word unparticularized matters, because the suspicion must attach to the individual who is stopped. As United States v. Cortez recounts, Chief Justice Warren wrote in Terry that this demand for specificity in the information on which police action rests is the central teaching of the Court's Fourth Amendment jurisprudence.16 A general sense that crime occurs nearby, or that people of a certain kind sometimes commit it, gives the officer nothing specific about the person standing in front of him.
Placed beside the Terry scene, this one involves the same kind of street and the same kind of encounter, and the difference lies in the conduct that can be described. McFadden had a repeated sequence of specific acts, and the officer in the box has one short conversation and an impression. Add a dozen passes along the same block and a pause at the same window each time, and the officer would stand where McFadden stood.
Labels create the same problem in modern practice. In Floyd v. City of New York, a federal district court reviewing the New York police department's stop records found that in at least six percent of documented stops between 2004 and 2009 the officers' own stated reasons were facially insufficient to show reasonable suspicion, including more than 62,000 stops justified only by "furtive movement" and more than four thousand justified only by "high crime area."17 That is a trial court's finding and not a Supreme Court holding, but it fits the Supreme Court's treatment of location: presence in an area of expected criminal activity, standing alone, does not support a particularized suspicion, though officers need not ignore the character of a location when they weigh the circumstances.18 A label becomes useful once it is attached to something specific. In Illinois v. Wardlow the location was an area of heavy narcotics trafficking, and what turned the encounter into reasonable suspicion was the suspect's unprovoked flight on seeing the police, an act the officer could describe.19
The facts also need not describe a crime in progress. The standard reaches a person who is involved in criminal activity or in a completed crime, so articulable facts about a past offense can support a stop just as facts about a present one can.20 The green-jacket description earlier in this section works that way: a matching description, found near the scene soon after the crime, is something the officer can state and a court can check, subject to the limits on stops for completed crimes already noted. A bare sense that a person must have committed some earlier offense fails for the same reason that a bare sense about a present one fails.
On an exam, the useful question is what the officer actually saw, heard, or knew, stated as a fact, and whether a reasonable inference of crime runs from it. An impression or a label is a conclusion, not a fact, so a stem that recites an officer's bad feeling or a neighborhood's reputation and nothing more describes a stop without reasonable suspicion. A stem that recites repeated, specific conduct points the other way, even when every act would be innocent standing alone.
Totality and Officer Experience
Once the officer has articulable facts to point to, the next question is how a court weighs them. The governing method is the totality of the circumstances: the court considers everything the officer knew and observed together and asks whether "the whole picture" gives the officer a particularized and objective basis for suspecting the person stopped of criminal activity.22 In United States v. Cortez the Court divided that idea into two elements, each of which must be present before a stop is permissible.23 The first is that the assessment rest on all of the circumstances, including objective observations, available police reports, and the usual patterns of operation of certain kinds of lawbreakers. The second is that the result be suspicion directed at the particular individual stopped, which is the specificity demand already discussed.24 This subsection concerns the first element and a corollary that comes with it: the person drawing the inferences is a trained officer, whose experience lets facts carry more meaning than they would to a bystander.
The reason for looking at the whole picture is that suspicion is built from probabilities, and probabilities add up. A court that examined each fact alone could supply an innocent explanation for every one, discard it, and then announce that nothing remained, even though the combination might be far more telling than any single piece. United States v. Sokolow rejected a version of that error. The court of appeals there had divided the evidence into "ongoing criminal behavior," which counted, and "probabilistic" evidence, which did not, and the Supreme Court answered that the two kinds of evidence differ only in degree of probative value.25 Traveling under an alias might be innocent, as when a patient conceals a trip for an operation, and an evasive path through an airport might be an attempt to avoid a creditor.26 Each possibility shows that the fact lacks ironclad significance, and a fact that is highly probative without being conclusive still belongs in the weighing.
Cortez shows the method working on facts that would mean little to an untrained observer. Border Patrol officers had gathered clues over two months, including footprints left by chevron-soled shoes, suggesting that a single guide, whom they nicknamed "Chevron," was leading groups of aliens through the desert to Highway 86, where a vehicle picked them up by prearrangement.27 From that pattern the officers reasoned that the guide was due, since two weeks had passed since his latest crossing, that he traveled on clear weekend nights, and that rain on Friday and Saturday made Sunday the only clear night of that weekend. They then worked out a window between 2 a.m. and 6 a.m., a likely rendezvous near milepost 122, and a group of eight to twenty people, which pointed them toward enclosed vehicles of that capacity.28 They staked out a spot east of milepost 122 and watched for a large vehicle that passed going west and returned going east about an hour and a half later. In four hours only one vehicle fit that description, and the validity of the stop depended on what the officers knew beforehand, not on what the stop later turned up.29
The Court described this as the kind of police work in which "fact on fact and clue on clue" led the officers to the vehicle, and it stressed that objective facts meaningless to the untrained can be combined with permissible deductions to form a legitimate basis for suspicion of a particular person.30 Its test was not whether the officers had probable cause to conclude that the vehicle held the guide and a group of aliens. The question was whether, based on the whole picture, experienced Border Patrol officers could reasonably surmise that the particular vehicle was engaged in criminal activity, and on that record they could.31 Taken separately, a footprint pattern, a weather report, and a two-week gap would not have justified stopping any vehicle, but taken together they singled out one.
The same passage explains why the officer's training matters. A trained officer draws inferences and makes deductions that might well elude an untrained person, and the evidence is to be weighed as understood by those versed in law enforcement rather than as analyzed in a library by scholars.32 The idea runs through the stop-and-frisk cases from the start, since Terry framed the question as whether an officer who observes unusual conduct could reasonably conclude "in light of his experience" that criminal activity may be afoot.33 Border-area cases make the factor explicit by counting facts that are meaningful to the officer "in light of his experience detecting illegal entry and smuggling."34 The reason is functional. Criminal conduct is usually recognizable in patterns rather than in single acts, and the people best placed to see a pattern are those who have watched it recur. Letting experience inform the inference allows officers to act on what practiced observers know, and requiring the whole picture prevents a reviewing court from discounting each fact separately until nothing is left.
Experience is one source of inference, and ordinary common sense is another. In Kansas v. Glover the Court held that an officer who ran a license plate and learned that the registered owner's license was revoked could stop the vehicle, because the inference that a car's driver is its registered owner is one ordinary people make every day and needs no specialized training.35 A dissenting Justice had argued that reasonable suspicion allows only the common sense of the reasonable officer, developed through law enforcement experience. The majority replied that such a rule would make police point to training materials for every infraction in a municipal code and would tie a stop's validity to the officer's length of service.36 Illinois v. Wardlow points the same way: courts have no empirical studies of the inferences to be drawn from suspicious behavior, so the determination rests on commonsense judgments about human behavior, and neither judges nor officers are asked for scientific certainty.37
Experience therefore works on facts and cannot stand in for them. In the decisions above, the officers could state what they had seen: footprints, dates, weather, a vehicle's round trip. Their training supplied the step from those observations to the suspicion that crime was afoot. An officer who offers experience with no observations has given the court the second link of the chain without the first, and whatever the length of service, the officer must still articulate something more than an unparticularized hunch.38 A stem that mentions a veteran officer invites the assumption that the stop must be sound, and the trap is forgetting that experience cannot replace stated observations.
The two versions of the stop differ only in whether the officer can state observations, which is the same difference that separates the Border Patrol officers in Cortez from an officer with a bare impression. Because the totality inquiry has two layers, observations first and inference second, it is best run in that order on an exam stem.
The first step decides most stems, because the facts the officer can state fix everything the later steps have to work with. The next subsection takes up the hardest version of the second step, in which every observation has an innocent explanation and the question is what the combination adds.
Innocent Conduct and Profiles
The preceding subsection ended on the hardest version of the weighing step: every fact the officer can state has an innocent explanation, and the question is what the combination adds. The answer starts from a practical observation about how suspicion works. Almost any single fact, such as paying cash, traveling on short notice, or looking nervous near police, describes some law-abiding person, so a rule requiring each fact to point to crime by itself would disqualify nearly every stop. Reasonable suspicion is a probabilistic judgment, and probabilities accumulate, which means that facts each consistent with innocence can together make criminal activity a reasonable inference.40 The stop itself exists to clarify an ambiguous situation, so some ambiguity in the officer's facts is the condition the doctrine expects.41
Terry had exactly that structure. The officer there watched two men pace back and forth in front of a store, peer into its window, and periodically confer. Each act was lawful by itself, yet together they suggested that the men were casing the store for a planned robbery, and Terry allowed the officer to detain them to resolve the ambiguity.42
The modern leading illustration is United States v. Sokolow, in which DEA agents stopped a traveler named Andrew Sokolow on his arrival at Honolulu International Airport. The Court identified six facts the agents knew when they approached him.43
Taken one at a time, these facts prove little. Cash is a lawful way to pay, a nervous manner describes many air travelers, and carrying only carry-on bags is common. The name mismatch also had an explanation the agents did not know: the telephone number belonged to a roommate of Sokolow's, a detail the agents had not learned.44 The Ninth Circuit held that these facts did not amount to reasonable suspicion, and the Supreme Court took the contrary view and reversed.45 The reason lies in the combination. A quick round trip to a source city, paid for in cash from a roll of small bills and made without checked luggage, forms a pattern that none of the facts forms alone.
The scope of the decision is narrow. The Court assumed without deciding that a stop occurred when the agents grabbed Sokolow by the arm and moved him back onto the sidewalk, so the decision answers only whether the agents had reasonable suspicion at that moment, and the seizure question treated earlier in this chapter was not in play.46
The Court also refused to sort the facts into favored and disfavored categories. The Ninth Circuit had divided the evidence into "ongoing criminal behavior" and "probabilistic" evidence, and the Supreme Court answered that the two kinds differ only in degree of probative value, since even an alias or an evasive path through an airport can have an innocent explanation.47 The consequence is that the cash payment, the nervousness, and the empty baggage claim, none of which proves much alone, stay in the weighing rather than dropping out as merely probabilistic. The same objection appeared in Cortez, where the court of appeals thought the circumstances admitted "far too many innocent inferences" to justify the stop.48 The Supreme Court stressed instead that objective facts meaningless to the untrained can be combined with permissible deductions to form a legitimate basis for suspicion of a particular person.49
Illinois v. Wardlow shows the same arithmetic on a city street. Presence in an area of heavy narcotics trafficking does not support a particularized suspicion standing alone, and flight is not necessarily indicative of wrongdoing, but unprovoked headlong flight in such an area was enough, because flight is certainly suggestive of wrongdoing.50 The Court accepted that flight has innocent explanations and held that this fact does not establish a Fourth Amendment violation.51
Combining facts has a limit, and the limit keeps the principle from swallowing the requirement of articulable facts. Adding facts helps only when each is a specific observation with some bearing on crime, because a pile of labels adds up to nothing. In Brown v. Texas officers saw two men walk away from each other in an alley in an area with a high incidence of drug traffic, and the Court found no grounds for a stop, since it was not unusual for people to be in the alley and the officers pointed to no facts supporting their conclusion that the situation looked suspicious.52 A single common fact with nothing to combine it with sits in the same position.
Sokolow also settles what to do with a drug courier profile, which is a list of characteristics that investigators have observed in past offenders. Matching a profile neither by itself establishes nor negates reasonable suspicion, but the profile's elements can be part of the totality.53 Each half of that rule has its own reason. A profile cannot establish suspicion alone because it is a generalization about a category of people, while a stop requires a particularized and objective basis for suspecting the person stopped, and a list that many innocent travelers satisfy cannot supply that basis by itself.54 A profile cannot negate suspicion either, because its elements are observations, such as the cash, the short trip, and the empty baggage claim, that carry the same probative weight whether or not someone has collected them on a list. Discarding facts because they appear on a profile would remove evidence from the weighing for a reason unrelated to its value. When a stem says the suspect fit a courier profile, the label is therefore only a summary, and the question remains which specific facts the officer observed and whether they add up.
On an exam, the usual mistake is to stop at the observation that each fact could be innocent. An innocent explanation for each fact does not end the analysis, because the question is whether the combination reasonably suggests criminal activity, and the combination must still consist of specific observed facts rather than a label or a hunch. A stem that lists several ordinary facts about a traveler points toward reasonable suspicion when the facts fit together into a pattern, and a stem that offers one common fact, or a bare profile label, points the other way.