Skip to main content
Find a Lawyer
Michael C. Dorf

The New Arizona Immigration Law Raises an Old Question: What is "Reasonable Suspicion"?


Monday, May 3, 2010

Recently, the state of Arizona catapulted immigration issues back onto the national agenda with the passage of a new law mandating that tough measures be taken against undocumented immigrants by state and local law enforcement officers. The law raises a number of important legal and policy issues, such as these: Can a person who is not lawfully present in the United States be found guilty of the further state-law crime of being present anywhere in the state–as authorized by a provision of the Arizona law? Is it wise for a state to adopt an overwhelmingly hostile stance to illegal immigration, thereby providing undocumented immigrants with strong disincentives to cooperate with law enforcement? Last, and most broadly, to what extent do federal statutes and the Constitution permit states to "over-enforce" federal immigration laws, as Arizona has attempted?

One feature of the Arizona law has received special attention. It provides:

[F]or any lawful contact made by a law enforcement official or agency of this state or a county, city, town or other political subdivision of this state where reasonable suspicion exists that the person is an alien who is unlawfully present in the United States, a reasonable attempt shall be made, when practicable, to determine the immigration status of the person. The person's immigration status shall be verified with the federal government pursuant to 8 United States Code Section 1373(c).

Critics have argued that, in most cases, "reasonable suspicion" to believe any given person is an undocumented immigrant will merely be the product of the person's ethnicity. How, the critics ask, is a police officer or other government official supposed to distinguish between, on one hand, undocumented immigrants and, on the other hand, U.S. citizens and others who are lawfully present in the country? Won't they likely use racial profiling to focus on Latinos? After all, a person's immigration status is not easily detectable.

In this column, I shall unpack the "reasonable suspicion" provision of the Arizona law. In doing so, I will question (and then ultimately defend) a basic precept of the constitutional law limiting police investigations: the prohibition on policing by "hunch."

What Is Reasonable Suspicion?

The Arizona legislature did not invent the term "reasonable suspicion." It is a term of art in American constitutional law, traceable to the U.S. Supreme Court's 1968 landmark decision in Terry v. Ohio.

The Fourth Amendment protects a right "against unreasonable searches and seizures." It further provides that "no Warrants shall issue, but upon probable cause." Taken together, these limits have long been understood to require that before the police subject someone to search or arrest–since an arrest counts as a "seizure" of the person–they must first obtain a warrant from a judge or magistrate that is based upon probable cause that the search will yield evidence of crime, or that the person to be arrested is guilty of a crime. A search or seizure without a warrant that is backed by probable cause is presumptively an "unreasonable" search or seizure.

However, the language of the Fourth Amendment does not literally require warrants, or even probable cause, for every search and seizure. Accordingly, various lines of cases uphold some warrantless searches and seizures, as well as some searches and seizures not backed by probable cause.

Consider a routine example: A police officer walking the beat observes someone acting suspiciously–perhaps walking back and forth in front of a bank furtively placing his hand in his pocket. That person could be fingering a gun as he works up the courage to rob the bank, but he could instead be absentmindedly touching his wallet as he works up the courage to apply for a loan. The officer does not have probable cause to believe the person is engaged in or about to commit a crime, and even if she did, she would not have sufficient time to obtain a warrant to search him from a magistrate or judge. Does this mean that the officer has no power to intervene?

The Supreme Court addressed that question in Terry. In a decision by Chief Justice Earl Warren, the Court ruled that a police officer may "stop and frisk" a suspect if the officer observes "articulable facts" that make it "reasonable to assume" that the suspect is armed and dangerous. This standard is less demanding than the "probable cause" that is needed for a full arrest because the nature of the intrusion–a brief stop and a weapons pat-down–is less serious than with an arrest. Although the majority in Terry did not specifically use the term "reasonable suspicion," subsequent cases have made clear that Terry was applying what has come to be called the reasonable suspicion test.

What Counts as Reasonable Suspicion?

Courts have resisted calls to quantify the "reasonable suspicion" standard. However, it appears to be a relatively undemanding test.

Consider, for example, the sorts of factors that law enforcement authorities have successfully invoked in court as part of the so-called "drug courier profile" to meet the reasonable suspicion test. In his 1999 book, No Equal Justice, Georgetown law professor David Cole listed factors that courts had approvingly cited as justifying the conclusion of reasonable suspicion for a stop. Every factor on Cole's list was matched by one or more contrary factors.

Thus, "made eye contact with officer" and "avoided making eye contact with officer" were suspicious. So were "arrived late at night," "arrived early in the morning," and "arrived in afternoon." "Traveled alone" was suspicious, but so was "traveled with a companion." And so on.

Cole uses these and other examples of non-factor factors to argue that when officers purport to employ a drug-courier profile they are often engaging in racial profiling. That may well be right, but there is another possible (or additional) explanation: The problem may not be the factors police articulate for their suspicions, but the requirement that they articulate the basis for their suspicions at all.

Is It Sensible to Require "Articulable" Suspicion?

Although the Terry case is frequently cited as establishing the reasonable suspicion standard, it may be more important for its requirement that a "police officer must be able to point to specific and articulable facts" that make it reasonable to stop a person. "Anything less," the Terry Court insisted, "would invite intrusions upon constitutionally guaranteed rights based on nothing more substantial than inarticulate hunches . . . ."

On first blush, the requirement of articulable suspicion makes good sense as a means of ensuring effective and equitable policing. Officers who act on "hunches" may simply be giving vent to prejudices or otherwise acting on irrational grounds.

Yet a considerable body of neuroscience and psychology research suggests that for complex multi-factor judgments, actions based on unarticulated hunches actually produce better results than decisions in which people are required to articulate their specific grounds. Popular books like Malcolm Gladwell's Blink and Jonah Lehrer's How We Decide summarize and illustrate this research.

For example, Lehrer recounts the story of how, during the first Gulf War, a Royal Navy radar officer had an uneasy sense that a blip on his screen was an incoming hostile missile, rather than a returning friendly aircraft. The officer could not at the time explain why he "felt" that the blip was a missile, but he nonetheless staked the safety of his comrades in arms on his gut feeling. He was proved right but the explanation of why he was right took considerable post-hoc analysis. Experts eventually determined that because of their altitude differences, a missile would have entered the screen at a slightly different interval than a returning airplane. The officer's unconscious mind–what Lehrer calls the "emotional brain"–worked that out, even though his conscious mind could not explain the basis for his hunch, either in real time or thereafter.

Lehrer uses this and many other examples to show why hunches are often more accurate than conscious decisions that are based on articulable factors. Indeed, a number of studies show that asking people to articulate factors for decisions that are based on emotional reactions will lead to worse decisions. And because the often split-second decision whether to stop a person on suspicion of crime typically involves numerous factors that cannot readily be quantified, it probably falls into the category of questions that the emotional brain does a better job addressing than the rational brain does. Thus, in requiring police officers to articulate their grounds for stopping a person, the Terry rule may actually lead the police to make less accurate predictions of criminal conduct.

It does not necessarily follow, however, that the Terry Court was wrong to require police to be able to articulate the basis for their suspicion before conducting a stop-and-frisk. Even if a conscientious officer's hunch is a better predictor of criminal conduct than the facts he can articulate, hunches can misfire. They can also be infected with racial prejudice, a problem that was just barely below the surface in the Supreme Court's Terry decision. The requirement of articulable suspicion imposes a somewhat artificial and even harmful constraint on policing, but it does so as the price for ensuring fairness.

Should Government Officials Be Permitted to Act on Hunches About Immigration Status?

The new Arizona law does not directly implicate Terry. It only requires Arizona state and local officials to investigate a person's immigration status in the course of a lawful encounter. Such an encounter might be initiated by the suspected undocumented immigrant herself, rather than the official. For example, a clerk in a municipal hospital might develop suspicions based on what a person writes on an insurance form.

More often, however, suspicion about immigration status will arise in the mind of a police officer after he has lawfully stopped a suspect. The Fourth Amendment as construed in Terry already requires that the police officer have an articulable reasonable suspicion of crime to make the initial stop. The new Arizona law adds an additional twist: If, after the initial stop, the officer develops a further reasonable suspicion that the detainee is an undocumented immigrant, the officer then must take steps to ascertain the detainee's immigration status.

Should an officer's hunch be relevant to that further inquiry? The short answer is no. Hunches are only as good as the base of experience on which they rest. A person who has had numerous opportunities to sort undocumented immigrants from persons lawfully present in the United States–such as a federal Immigration and Customs Enforcement (ICE) agent–might develop an intuitive ability or "sixth sense" for detecting the former. But even then we would legitimately worry that skin color and accent would often play a large part in triggering the ICE agent's hunch. Moreover, Arizona state and local officials who lack training and experience sorting out undocumented immigrants would likely lean even more heavily on such illicit factors.

To be sure, there will be some circumstances that objectively give rise to reasonable suspicion, even without considering race or ethnicity. Writing in the New York Times last week, one of the law's authors gave the example of a police officer stopping a speeding minivan on "a known alien-smuggling corridor" and discovering twelve "passengers crammed inside," all lacking identification. Surely that would give rise to reasonable suspicion of immigration violations, he says. Indeed, that extreme case would satisfy the more demanding requirements of probable cause. But the Arizona law applies in numerous other contexts as well.

The critics of the new Arizona law are right. The core problem with the obligation to investigate that it places on state and local officials is that, in most settings, there are very few if any outwardly visible signs of immigration status that could give rise to reasonable suspicion. That will be true regardless of whether the officials act on expressly-articulated grounds or on the basis of a hunch informed by unarticulated grounds. Thus, whatever one thinks of how "reasonable suspicion" has been defined in the federal Fourth Amendment context, its use in the Arizona law is problematic.

Michael C. Dorf is the Robert S. Stevens Professor of Law at Cornell University. The second edition of his book, Constitutional Law Stories, is now available. He blogs at

Was this helpful?

Copied to clipboard