Searching for Terrorists: Why Public Safety Is Not a Special Need

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

Duke Law Journal

VOLUME 59 FEBRUARY 2010 NUMBER 5

SEARCHING FOR TERRORISTS:

WHY PUBLIC SAFETY IS NOT

A SPECIAL NEED

RIC SIMMONS†

BSTRACT

In the wake of the terrorist attacks of September 11, 2001, local

police across the country instituted blanket searches without

individualized suspicion at various venues—including political

protests, sporting events, subway platforms, and public ferries—all in

an attempt to prevent further terrorist attacks. When evaluating these

searches, courts rely upon the special needs doctrine, which allows the

government to conduct a suspicionless search as long as the search

serves a special need distinct from the goals of law enforcement. Over

the past eight years, courts have struggled to determine whether and

how the special needs doctrine applies to these searches, and these

struggles have produced inconsistent results.

This Article first reviews the history of antiterrorism searches,

which can be roughly divided into three different time periods. In the

early 1970s, in response to an epidemic of hijackings and bombings of

public buildings, the government instituted a regime of suspicionless

† Associate Professor of Law, Moritz College of Law, The Ohio State University. I would

like to thank Joshua Dressler, Michael J.Z. Mannheimer, Angela Lloyd, Jeanna Volp, and

James Marra for their feedback on earlier drafts of this Article. I would also like to thank the

faculty of the Moritz College of Law for their comments about this Article during faculty

workshops.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

844 DUKE LAW JOURNAL [Vol. 59:843

searches at airports and public buildings—searches which continue to

this day. During the second period, the imminent danger of these

terrorist actions abated, but courts continued to reject challenges to

the searches, and suspicionless searches spread to other contexts far

removed from the terrorist threat. Finally, in the third era, which

began in 2001 and continues to the present day, the government has

aggressively expanded its use of antiterrorism searches, and courts

face a new set of challenges in evaluating their constitutionality.

This Article then explains why antiterrorism searches cannot be

justified under the special needs doctrine, and indeed why—in their

current form—these searches cannot be justified under any Fourth

Amendment doctrine. It then proposes a solution: suspicionless

searches to prevent terrorism should be permitted, but only if the

fruits of the search cannot be used in a subsequent criminal

prosecution. Although the solution at first seems controversial, it

represents a reasonable balance between the need to protect the

country from terrorist attacks and the need to draw a principled

distinction between special needs searches and general searches.

TABLE OF CONTENTS

Introduction .............................................................................................845

I. The First Era: 1968–1976 ...................................................................850

A. The First Wave of Terror: Hijackings and Bombings.......850

B. The Birth of the Administrative Search Doctrine and

the End of Individualized Suspicion ...................................855

II. The Second Era: 1977–2000 .............................................................859

A. Order Is Restored .................................................................859

B. Broadening the Scope of Permissible Suspicionless

Searches..................................................................................862

1. The Birth of Special Needs: Searches and Drug

Tests in Schools................................................................863

2. Searches and Drug Tests of Public Employees..............866

3. Searches of Probationers and Parolees...........................868

4. Vehicular Checkpoints .....................................................871

III. The Third Era: 2001–2010...............................................................873

A. The War on Terror................................................................873

1. Reservoirs ..........................................................................875

2. Political Protests................................................................876

3. Sports Arenas ....................................................................879

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 845

4. Public Transportation .......................................................881

5. The Current Status of Antiterrorism Searches..............883

B. Special Needs in Other Contexts.........................................884

IV. Can Suspicionless Antiterrorism Searches Be Justified by

the Special Needs Doctrine?.......................................................887

A. Categorizing Permissible Suspicionless Searches..............887

B. Placing Antiterrorism Searches into the Proper

Category .................................................................................890

V. Other Possible Justifications for Suspicionless Antiterrorist

Searches.........................................................................................894

A. Evaluation Under a Generalized Reasonableness Test ...894

B. Legislatures Should Decide What Is Reasonable .............899

C. Consent...................................................................................903

D. Antiterrorism Searches Are Sui Generis; Courts

Should Create a New Category of Suspicionless

Searches..................................................................................907

VI. What If Suspicionless Antiterrorism Searches Are

Unconstitutional? .........................................................................911

A. Abolishing Antiterrorism Searches.....................................912

B. Allowing the Searches, but Precluding the Evidence .......915

C. Responding to Potential Criticisms.....................................921

Conclusion................................................................................................926

INTRODUCTION

Consider a hypothetical. The United States is the victim of

multiple terrorist attacks,

and traditional law enforcement techniques

have consistently been proven ineffective. In response, federal law

enforcement agents institute a regime of blanket searches, conducted

without any degree of individualized suspicion. These suspicionless

searches work, and in a very short period of time the terrorist attacks

abate. Meanwhile, the search regimes are inevitably subjected to a

series of challenges in court. The judges, seeing the efficacy of these

programs, unanimously approve of the searches by exploiting—and

expanding—a relatively obscure loophole in Fourth Amendment

jurisprudence. Stability returns to the nation. The question: what

happens to the suspicionless searches once the terrorist attacks have

been eliminated?

1. Throughout this Article, the word “terrorism” will be used to denote the use or threat

of violence with the intent to intimidate, usually for ideological or political purposes. See

MERICAN HERITAGE DICTIONARY 1854 (3d ed. 1992).

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

846 DUKE LAW JOURNAL [Vol. 59:843

The answer: they continue indefinitely. And why not? They have

proven themselves effective—indeed, indispensable—in preventing

widespread acts of violence. They have been sanctioned by the courts.

And most importantly, every year that they continue, the citizens who

are consistently subjected to these searches become more and more

accustomed to them, until they no longer seem to be intrusive.

Instead, if the searches are noticed at all, they are thought to be

merely a minor annoyance, an inevitable facet of modern life.

This is not a hypothetical from our potential future; it is a lesson

from our recent past. In the late 1960s and early 1970s, international

hijackers and domestic terrorists caused widespread violence and

social unrest in this country. These attacks stopped only after the

government ordered searches of every airplane passenger and every

visitor to federal courthouses. Courts upheld these suspicionless

searches by applying a legal theory called the administrative search

doctrine, which had been recently developed to allow health

inspectors access to homes without needing to demonstrate

individualized suspicion. Now, nearly forty years later, entire

generations have grown up assuming the only way to board an

airplane or enter a courthouse is to be subjected to a search by law

enforcement personnel.

But if we are to learn a lesson from this history, it is not clear

what the lesson should be. We have effectively traded liberty for

security, but even now it is difficult to measure how much liberty we

have lost and how much security we have gained. Are suspicionless

searches at airports and public courthouses merely a minor

annoyance, or a severe and unjustified privacy intrusion into our

privacy made all the more Orwellian by the fact that we only perceive

them to be a minor annoyance? And how much more secure are we

with these searches—or rather, how much security would we lose if

we abolished these searches and required police to play by the normal

rules to prevent these crimes?

Following the terrorist attacks of 2001, these are no longer

academic questions. History is repeating itself. In the wake of the

September 11 attacks, local police across the country instituted

suspicionless search regimes at political protests, sporting events,

subway platforms, and public ferries, all in an attempt to prevent

further terrorist attacks. Courts now struggle to determine whether

the same doctrine that justified suspicionless searches at airports and

public courthouses can be used to support this second generation of

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 847

searches. And many academics—usually ready to denounce and

criticize new restrictions on Fourth Amendment rights—have

scrambled to find ways to justify these searches under existing Fourth

Amendment jurisprudence,

or have argued that the old Fourth

Amendment rules simply do not or should not apply in the face of a

terrorist threat.

This Article evaluates the constitutionality of today’s

antiterrorism suspicionless searches. It argues that one cannot

effectively evaluate the validity of these searches without first

understanding and critiquing the initial wave of antiterrorism

suspicionless searches, all of which were unanimously approved by

the courts in the early 1970s. Thus, this Article breaks up the

development of antiterrorism searches into three time periods: the

early 1970s, when courts first instituted and upheld airport and

courthouse searches; the 1980s and 1990s, when terrorist attacks

abated and the searches became ingrained in American culture; and

the modern era, from 2001 until the present, in which the government

seeks to extend the scope of the searches and courts are once again

being called upon to evaluate their constitutionality.

Of course, suspicionless searches are not confined to the

antiterrorism context. In fact, just before the first antiterrorism

searches were being reviewed by the appellate courts in the early

1970s, the Supreme Court created an entirely new doctrine which

2. See, e.g., Ricardo J. Bascuas, Fourth Amendment Lessons from the Highway and the

Subway: A Principled Approach to Suspicionless Searches, 38 R

UTGERS L.J. 719, 722 (2007)

(stating that the New York City subway searches are “uncontroversial” and that the case for

such searches is “relatively easy to make”); Richard C. Worf, The Case for Rational Basis

Review of General Suspicionless Searches and Seizures, 23 T

OURO L. REV. 93, 131–37 (2007)

(arguing that suspicionless searches should be seen as reasonable and thus constitutional if they

have been approved by a representative legislative body, because the legislative process will

correct any overreaching by law enforcement). But see Anthony C. Coveny, When the

Immovable Object Meets the Unstoppable Force: Search and Seizure in the Age of Terrorism, 31

M. J. TRIAL ADVOC. 329, 384 (2007) (criticizing the holdings in antiterrorism search cases and

noting that “whenever a bright line rule is replaced by a balancing test, civil liberties are likely

to lose”).

3. See, e.g., Ronald M. Gould & Simon Stern, Catastrophic Threats and the Fourth

Amendment, 77 S.

CAL. L. REV. 777, 777 (2004) (asserting that “traditional Fourth Amendment

search-and-seizure doctrine was fine for an age of flintlocks,” but that “large-scale searches

undertaken to prevent horrific potential harms may be constitutionally sound”). But see John T.

Parry, Terrorism and the New Criminal Process, 15 W

M. & MARY BILL RTS. J. 765, 834–35

(2007) (concluding that “the war on terror has generated extraordinary criminal processes

applicable to people suspected of terrorism” and that the costs of this change are significant,

including that “state power over all of us—over our bodies, our mobility, our words and actions,

and of course our lives—continues to increase”).

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

848 DUKE LAW JOURNAL [Vol. 59:843

allowed the government to conduct a search or seizure without any

showing of individualized suspicion. It is this doctrine, known as the

administrative search doctrine, which the early courts ultimately used

to justify the antiterrorism searches. Although the Court has

described these cases as a “closely guarded category of

constitutionally permissible suspicionless searches,”

they in fact

cover a broad range of situations. They include searches to enforce

regulatory violations, mandatory drug testing of schoolchildren and

public employees, inventory searches, immigration and drunk driving

checkpoints, and searches of probationers and parolees. The doctrine

underlying these cases is murky at best, and judges and commentators

disagree as to whether these cases all fall into the same category or

should be considered doctrinally distinct.

For now, I will refer to

them all as “permissible suspicionless searches,” meaning searches

that are permissible even if the government does not have any

amount of individualized suspicion of the subject being searched.

The first three Parts of this Article detail the changes to

antiterrorism searches and the permissible suspicionless search

doctrine for each of these three time periods. Part I covers the first

era, approximately 1968–1976, and examines both the government’s

reaction to the first wave of terrorism, as well as the birth and early

application of the permissible suspicionless search doctrine. Part II

describes the second era, roughly 1977–2000, which was a time of

relative calm between the two waves of terrorism. Despite the relative

calm during this period, the permissible suspicionless searches at

airports and courthouses became more firmly entrenched in the law,

and among the general public they became an unquestioned—and

perhaps even comforting—aspect of traveling by air or entering a

federal building. Meanwhile, outside the terrorism context, the

permissible suspicionless search doctrine was evolving and expanding

4. Chandler v. Miller, 520 U.S. 305, 309 (1997) (declaring unconstitutional a Georgia law

that conditioned one’s candidacy for state office on passing a drug test).

5. The Ninth Circuit, for example, breaks these cases down into three categories that are

“not necessarily mutually-exclusive”: searches at “exempted areas” (such as international

borders, prisons, airports, and entrances to government buildings), “administrative” searches,

and “special needs” searches. United States v. Kincade, 379 F.3d 813, 822–23 (9th Cir. 2004).

Professor Schulhofer, meanwhile, distinguishes the “administrative” searches and the “internal

governance” searches (such as those that occur in schools and public workplaces) from the

mandatory drug testing laws that are actually intended to deter and punish the use of illegal

drugs. See Stephen J. Schulhofer, On the Fourth Amendment Rights of the Law-Abiding Public,

1989 S

UP. CT. REV. 87, 162–63.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 849

to cover a variety of different contexts, acquiring along the way a new

name: special needs searches. Finally, Part III discusses the modern

era, from 2001 to the present, when terrorism again became a

palpable threat to this country, and law enforcement once again

responded with suspicionless searches to prevent future attacks.

During the present era, however, courts have become more critical of

suspicionless searches, both in the antiterrorism context and in other

areas, ultimately leading to a confusing legal test and an inconsistent

application of that test when it is applied to antiterrorism searches.

After the Article traces the development of both antiterrorism

searches and the permissible suspicionless search doctrine, Part IV

critically analyzes the special needs doctrine. This Part breaks the

special needs cases up into three different categories: searches made

outside the criminal context, searches made in the context of criminal

activity but whose results are not turned over to law enforcement, and

searches made in the context of a criminal case whose results are used

by law enforcement. This Part concludes that suspicionless

antiterrorism searches fall into the third category—a category that

cannot be justified by the special needs doctrine. Part V examines

other possible justifications for suspicionless antiterrorism searches,

including analyzing these cases under a generalized reasonableness

test and relying on implied consent. This Part concludes that no other

justifications are valid, given the very real danger of a slippery slope

and the need for a principled distinction between searches that

require individualized suspicion and those that do not. This creates a

significant problem: if suspicionless antiterrorism searches are

unconstitutional under the principles of the Fourth Amendment, how

can law enforcement protect us from the very real threat of

terrorism?

Part VI will propose a solution to this problem: suspicionless

antiterrorism searches would be constitutionally justified under the

special needs doctrine as long as any contraband that is recovered

cannot be used in a criminal prosecution. By giving law enforcement

the opportunity to choose between a suspicionless search that is truly

aimed only at preventing terrorism and a more traditional search that

could result in criminal prosecution, this proposal gives the police

more power to search when there is a true terrorist threat, yet

encourages them to develop more sophisticated and less intrusive

methods of investigation when they are seeking to apprehend and

prosecute wrongdoers.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

850 DUKE LAW JOURNAL [Vol. 59:843

THE FIRST ERA: 1968–1976

The government’s use of antiterrorism suspicionless searches

began in the late 1960s, in response to a series of hijackings and

bombings of government buildings across the country. At about the

same time, the Supreme Court created the administrative search

doctrine, which permitted suspicionless searches as long as they were

not “aimed at the discovery of evidence of a crime.”

When it came

time for courts to review the antiterrorism suspicionless searches,

many of them turned to the administrative search doctrine to evaluate

and ultimately approve of this method of detecting and apprehending

terrorists.

A. The First Wave of Terror: Hijackings and Bombings

The late 1960s was a time of severe social unrest in this country.

The civil rights movement, opposition to the Vietnam War, protests

against the regime of Fidel Castro, and Palestinian claims of

sovereignty all motivated radical elements of the population to

engage in acts of domestic terrorism. Bombings of public buildings

and other high-profile targets increased dramatically. Although the

country had experienced forms of domestic terrorism before, the

unrest had previously been limited to political fringe groups or

specific segments of the population.

During the late 1960s, police

departments, courthouses, foreign missions, university offices,

prisons, the Pentagon, and the United States capitol building were all

targets of bomb attacks.

According to a report by the Federal Bureau

6. Camara v. Mun. Court, 387 U.S. 523, 537 (1967) (asserting the reasonableness of code-

enforcement inspections).

7. For example, in 1857 a group of Mormons in Utah declared themselves to be in open

rebellion against the United States and slaughtered 120 settlers on their way to California. See

Eugene E. Campbell, Governmental Beginnings, in U

TAH’S HISTORY 153, 165–71 (Richard D.

Poll et al. eds., 1989). The late-nineteenth century saw a number of violent acts committed

during labor disputes or by anarchists attempting to topple the United States government. See,

e.g., Nicholas von Hoffman, To Kill Everyman; A New Chapter in American Terrorism, W

ASH.

POST, Oct. 15, 1995, at B3 (describing the changing targets of terrorist attacks). And racial

violence perpetrated by groups such as the Ku Klux Klan caused many deaths in the early-to-

mid-twentieth century. However horrible these terrorists’ actions were, in sheer number they

did not compare to the widespread bombings of public buildings in the late 1960s. See infra

notes 8–9 and accompanying text.

8. See Thomas R. Brooks, Editorial, The Radical Underground Surfaces with a Bang, N.Y.

TIMES, Mar. 15, 1970, (Magazine), at 171; see also Michael Taylor, ‘70s in the Bay Area—Era of

Radical Violence, S.F.

CHRON., Jan. 24, 2007, at A1.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 851

of Investigation, there were 3,000 bombings and 50,000 bomb threats

in 1970 alone.

These attacks, alongside a series of high-profile

political assassinations, including Robert Kennedy and Martin Luther

King, Jr., heightened the perception of a tidal wave of violent social

unrest sweeping the country.

Around the same time, the country experienced an epidemic of

airplane hijackings. In the decade preceding 1968, hijackings had

averaged only one per year, but in 1968 there were eighteen, and in

1969 there were thirty-three.

In one particularly notorious incident

in 1970, a radical Palestinian group hijacked four planes bound for

New York City and flew them to Jordan and Egypt, where the

terrorists ultimately blew up all four planes on the landing field.

The government’s response to this rise in both domestic and

international terrorism included broad new search procedures at the

entrances to public buildings and at airports. In the fall of 1970, the

federal General Services Administration issued an order requiring

searches of all bags and packages at entrances to every federal

building.

Meanwhile, on September 11, 1970, President Richard

Nixon issued a directive for the Department of Transportation,

requiring its agents to work with airlines to institute surveillance

programs at all domestic airports.

This directive was followed by a

series of rules promulgated by the Federal Aviation Administration

in 1972 requiring all airline passengers and their luggage to be

9. See FBI, History of the FBI: Vietnam War Era: 1960s–1970s, http://www.fbi.gov/

libref/historic/history/vietnam.htm (last visited Dec. 22, 2009) (describing violence in opposition

to the Vietnam War).

10. See generally The American Century, 1960–1969, W

ASH. TIMES, Aug. 30, 1999, at A10

(detailing the major political and social events of the 1960s).

11. United States v. Davis, 482 F.2d 893, 898 (9th Cir. 1973), abrogated by United States v.

Aukai, 497 F.3d 955 (9th Cir. 2007) (en banc). The high number of hijackings continued into the

next decade: twenty-five in 1970, twenty-five in 1971, and twenty-six in 1972. The numbers

finally dropped to single digits for most of the rest of the 1970s, and dropped to zero by the

1990s. See O

FFICE OF CIVIL AVIATION SEC., FED. AVIATION ADMIN., U.S. DEP’T OF TRANSP.,

CRIMINAL ACTS AGAINST CIVIL AVIATION 75 (2000).

12. See Eric Pace, Disclosure Is Made by Amman Radio, N.Y.

TIMES, Sept. 27, 1970, at A1.

The four planes were all destroyed on September 12, 1970. Id.; see also Cynthia R. Fagan, Iraq’s

Oil for Terror; $72 Million to Palestinians, N.Y.

POST, Oct. 17, 2004, at 13.

13. Downing v. Kunzig, 454 F.2d 1230, 1231 (6th Cir. 1972). The Government Services

Administration (GSA) order stated: “[B]ecause of the recent outburst of bombings and other

acts of violence, effective at once, at all entrances to federal property under the charge and

control of GSA, where there are guards on duty, all packages shall be inspected for bombs or

other potentially harmful devices. Admittance should be denied to anyone who refuses to

voluntarily submit packages for examination.” Id.

14. Davis, 482 F.2d at 899–900.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

852 DUKE LAW JOURNAL [Vol. 59:843

screened prior to boarding an airplane.

These requirements have

remained in place ever since,

and security screenings have become a

familiar fixture at every federal building and commercial airport in

the country.

The suspicionless searches were challenged almost immediately,

and were universally upheld by the courts, which went out of their

way to emphasize the grave risk posed by the threat of terrorism. The

Sixth Circuit conceded that “[o]rdinarily . . . a person should not have

his person or property subjected to a search in the absence of a

warrant or probable cause to believe that a crime is being

committed,”

but also argued that, given the recent wave of violence,

“the dangers to federal property and personnel were imminent.”

The court concluded that “in times of emergency[,] government may

take reasonable steps to assure that its property and personnel are

protected against damage, injury or destruction.”

The Ninth Circuit

similarly upheld suspicionless searches at a San Francisco

courthouse.

In reaching its conclusion, the court relied on the fact

that there were “specific instances of bomb threats and bomb attacks

directed at San Francisco police stations, Oakland police stations, and

the Los Angeles federal building.”

Moreover, the court noted that

terrorists had recently kidnapped four individuals and killed a judge

in a nearby county.

Given this background of violence, the court

determined that “a serious threat of violence existed at the Hall of

Justice,” and therefore courthouse searches were a reasonable

15. See Davis, 482 F.2d at 900–02 (noting a February 1 rule requiring air carriers to

implement a screening system, an August 1 directive requiring airlines to search the baggage

and screen or search the person of passengers meeting a particular profile, and a December 1

order requiring searches of all carry-on items and screening of all passengers).

16. In 2002, federal law enforcement agents replaced private airline employees as

screeners. Federal Screeners Just One Component of Air Safety Net, USA

TODAY, Nov. 19, 2002,

at 20A.

17. Downing, 454 F.2d at 1232–33.

18. Id. at 1232.

19. Id. at 1233. The court also held that the searches were not very intrusive, stating that

inspection of bags and packages was a “very minimal type of interference with personal

freedom,” which was reasonable given the government’s need to protect itself against the

“ruthless forces bent upon its destruction.” Id.

20. See McMorris v. Alioto, 567 F.2d 897, 901 (9th Cir. 1978).

21. Id. at 900.

22. Id.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 853

precaution to prevent further violence against the courthouse and its

personnel.

Judges reacted similarly to suspicionless searches in airports.

approving the new security measures, these courts emphasized the

extraordinary harm that was caused by terrorist attacks. The Ninth

23. Id. Legal commentators at the time expressed some concern that the government was

overreacting, even given this background of violence. See, e.g., Kenneth L. Jesmore, The

Courthouse Search, 21 UCLA

L. REV. 797, 825 (1974) (“[Courthouse searches] are seldom

founded upon an adequate correlation between the scope of the search procedures and the

necessity for their implementation. . . . An ongoing emergency—a current, serious threat of

violence—may provide the justification necessary . . . . [but t]he threat, of course, can be neither

stale nor insignificant, and the intrusion must be limited according to the severity of the

emergency.”).

The Supreme Court did not rule on any case regarding courthouse bombings or

hijackings. However, in United States v. United States District Court, 407 U.S. 297 (1972), the

Court acknowledged the heightened danger of terrorism in a case in which the government

wiretapped a suspected terrorist. The Court noted that “threats and acts of sabotage against the

Government exist in sufficient number to justify investigative powers with respect to them. The

covertness and complexity of potential unlawful conduct against the Government and the

necessary dependency of many conspirators upon the telephone make electronic surveillance an

effective investigatory instrument in certain circumstances.” Id. at 311–12 (footnote omitted).

However, the Court ultimately ruled against the prosecutor, concluding that “the Government’s

concerns do not justify departure in this case from the customary Fourth Amendment

requirement of judicial approval prior to initiation of a search or surveillance.” Id. at 321.

24. See, e.g., United States v. Edwards, 498 F.2d 496, 499–501 (2d Cir. 1974) (holding that

the warrantless preboarding search of an airline passenger’s beach bag was reasonable when the

passenger was given sufficient notice that she was free to avoid the search by leaving the line);

United States v. Albarado, 495 F.2d 799, 806 (2d Cir. 1974) (“[T]he use of a magnetometer is a

reasonable search despite the small number of weapons detected in the course of a large

number of searches. The absolutely minimal invasion in all respects of a passenger’s privacy

weighed against the great threat to hundreds of persons if a hijacker is able to proceed to the

plane undetected is determinative of the reasonableness of the search.”); United States v.

Cyzewski, 484 F.2d 509, 512 (5th Cir. 1973) (“[C]ourts have consistently held airport security

measures constitutionally justified as a limited and relatively insignificant intrusion of privacy

balanced against the need to protect aircraft and its passengers.”); United States v. Davis, 482

F.2d 893, 910–11 (9th Cir. 1973) (holding that preboarding screening of passengers and carry-on

items for weapons or explosives is reasonable so long as the passenger can choose to avoid the

search by not boarding the aircraft), abrogated by United States v. Aukai, 497 F.3d 955 (9th Cir.

2007) (en banc); United States v. Skipwith, 482 F.2d 1272, 1276 (5th Cir. 1973)

(“[R]easonableness does not require that officers search only those passengers who meet a

profile or . . . who otherwise appear suspicious.”); United States v. Slocum, 464 F.2d 1180, 1182

(3d Cir. 1972) (concluding that, because of the potential serious dangers of aircraft hijackings,

the use of a magnetometer to screen airline passengers is justified by a reasonable governmental

interest); United States v. Bell, 464 F.2d 667, 673 (2d Cir. 1972) (“In view of the magnitude of

the crime sought to be prevented[ and] the exigencies of time which clearly precluded the

obtaining of a warrant, the use of the magnetometer is . . . a reasonable precaution.”); United

States v. Epperson, 454 F.2d 769, 771 (4th Cir. 1972) (asserting that magnetometer searches are

justified by the minimal invasion of personal privacy and overwhelming governmental interest

in preventing air piracy).

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

854 DUKE LAW JOURNAL [Vol. 59:843

Circuit noted that “[t]he need to prevent airline hijacking is

unquestionably grave and urgent. The potential damage to person

and property from such acts is enormous. The disruption of air traffic

is severe. There is serious risk of complications in our foreign

relations.”

Suspicionless searches were seen as the only way to

prevent these terrorist acts because “[l]ittle can be done to balk the

malefactor after such material is successfully smuggled aboard, and as

yet there is no foolproof method of confining the search to the few

who are potential hijackers.”

In upholding suspicionless airport

searches, the Fifth Circuit implied that they would be temporary:

“Certainly all citizens look forward to the day when skyjackings and

their sequels, airport search and security measures, cease. When the

threat of air piracy disappears the standards of reasonableness which

we here recognize will go with it.”

The language of these opinions demonstrates that these

suspicionless searches were seen as an extraordinary solution to an

extraordinary problem. As one judge noted in 1976: “In the wake of

unprecedented airport bombings, aircraft piracy and courtroom

violence, the courts have approved precautionary security measures

which cannot be reconciled with previously conceived notions of the

citizen’s protection under the Fourth Amendment against warrantless

intrusions without probable cause.”

Although courts generally agreed that these searches were

necessary, they initially struggled to find a doctrinal justification for

their holdings. Some courts simply applied the “unreasonable

searches” language of the Fourth Amendment and concluded that the

searches were reasonable given the nature of the threat and the

relatively low level of intrusion.

Others analogized suspicionless

searches in airports to searches at the border, and applied a three-part

balancing test, including the nature of the threat, the level of the

intrusion, and the efficacy of the search in actually preventing the

25. Davis, 482 F.2d at 910 (footnote omitted). The Ninth Circuit also held that these

searches have an element of implied consent: “[A]irport screening searches are valid only if they

recognize the right of a person to avoid search by electing not to board the aircraft.” Id. at 910–

11. The consent justification for these searches has recently been questioned by some courts. See

infra notes 316–20 and accompanying text.

26. Davis, 482 F.2d at 910.

27. Skipwith, 482 F.2d at 1279.

28. Collier v. Miller, 414 F. Supp. 1357, 1362 (S.D. Tex. 1976).

29. See, e.g., Downing v. Kunzig, 454 F.2d 1230, 1233 (6th Cir. 1972).

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 855

harm.

Still others held that the individual being searched had given a

sort of implied consent because he or she could always avoid the

search by simply choosing not to board the plane or enter the

government building.

But most courts placed these searches into a

recently developed category of suspicionless searches known as

administrative searches—and this category is where they have

remained for the subsequent four decades.

B. The Birth of the Administrative Search Doctrine and the End of

Individualized Suspicion

The Fourth Amendment prohibits “unreasonable searches and

seizures,” and states that “no Warrants shall issue, but upon probable

cause.”

This pair of restrictions has led to differing interpretations as

to how the Amendment should be applied in practice,

which in turn

has led to a thoroughly confusing body of jurisprudence. Courts have

generally held that a search is unconstitutional unless it is supported

by a warrant or probable cause,

but they have also used the

“reasonableness” language to support exceptions to this general rule.

For example, courts allow brief pat-down searches for weapons on a

showing of less than probable cause to protect an officer’s safety.

30. See, e.g., Skipwith, 482 F.2d at 1275.

31. See, e.g., Singleton v. Comm’r, 606 F.2d 50, 52 (3d Cir. 1979) (“By electing to proceed

and board the aircraft, with advance notice of the search requirement, [the passenger] impliedly

consented to the search.”); Davis, 482 F.2d at 910–11. The implied-consent justification was

recently rejected by the Ninth Circuit. See United States v. Aukai, 497 F.3d 955, 960–61 (9th Cir.

2007) (en banc). For a critique of the implied-consent justification, see infra Part V.C.

32. U.S.

CONST. amend. IV.

33. See Bascuas, supra note 2, at 723–25 (explaining that the “warrant preference”

interpretation, with certain exceptions, requires that searches and seizures be made under a

warrant and with probable cause, whereas the “general reasonableness” interpretation requires

only that searches and seizures be reasonable in the context of the particular case). Most

scholars today focus on the “unreasonable” language, arguing for a broad balancing test in

determining whether or not a given search is constitutional. See, e.g., Akhil Reed Amar, Terry

and Fourth Amendment First Principles, 72 S

T. JOHN’S L. REV. 1097, 1118, 1120–25 (1998)

(discussing different factors that may help to determine whether or not a search is reasonable,

including: the scope and intrusiveness of the search, the weight of the governmental interest at

issue, and the identity of the subject being searched).

34. See, e.g., Nat’l Treasury Employees Union v. Von Raab, 489 U.S. 656, 667 (1989)

(“Even where it is reasonable to dispense with the warrant requirement in the particular

circumstances, a search ordinarily must be based on probable cause.”).

35. See Terry v. Ohio, 392 U.S. 1, 23–24, 28–31 (1968) (holding that a police officer’s

limited search of three men for weapons was reasonable because the men’s unusual conduct

gave the officer reasonable ground to believe, in light of his experience, that the search was

necessary to prevent harm to himself and others).

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

856 DUKE LAW JOURNAL [Vol. 59:843

And if there is probable cause to make an arrest, the Supreme Court

has held that a search of the suspect is reasonable to protect officer

safety and preserve evidence for later use at trial.

In each of these cases, however, courts require some form of

individualized suspicion (albeit short of probable cause) before the

search can take place.

In fact, prior to 1967, the only context in

which the Supreme Court allowed nonconsensual suspicionless

searches

was at the international border, pursuant to longstanding

and relatively noncontroversial case law.

But all that changed in the

case of Camara v. Municipal Court.

In Camara, the Court held that a health inspector was not

required to demonstrate probable cause—nor indeed any form of

individualized suspicion—to obtain a warrant to investigate a home.

According to the Court, there was no need to require individualized

suspicion because the searches were not “aimed at the discovery of

evidence of crime.”

The Court also stated that a probable cause

requirement would be impracticable because the hidden nature of

36. See United States v. Robinson, 414 U.S. 218, 234 (1973); Weeks v. United States, 232

U.S. 383, 395–96 (1914) (affirming the principle that evidence found incidental to the execution

of a legal search warrant is admissible at trial when material and properly offered in evidence

because this evidence was not the product of an unreasonable search and seizure).

37. For searches incident to arrest, no further amount of individualized suspicion is

required after the arrest is made—but of course the arrest itself is not legal unless the police

have some amount of individualized suspicion against the suspect, and if they do not, the

subsequent search incident to the arrest is invalid. See Chimel v. California, 395 U.S. 752, 762–63

(1969).

38. Of course, the most common form of search used by law enforcement is the consent

search, which does not require any amount of individualized suspicion—but does require the

suspect to agree to be searched. See United States v. Miller, 20 F.3d 926, 930 (8th Cir. 1994)

(“[P]olice officers may search an area, even without probable cause or a warrant, if someone

with adequate authority has consented to the search . . . .”); United States v. Morris, 910 F.

Supp. 1428, 1446 (N.D. Iowa 1995) (explaining that for a consensual search to be valid, the

consent must be voluntary, must not have been “tainted by any other Fourth Amendment

violation,” and must not exceed the “reasonable scope” of the consent).

39. See Carroll v. United States, 267 U.S. 132, 154 (1925) (“Travelers may be so stopped in

crossing an international boundary because of national self protection reasonably requiring one

entering the country to identify himself as entitled to come in, and his belongings as effects

which may be lawfully brought in.”); see also infra notes 135–38 and accompanying text.

40. Camara v. Mun. Court, 387 U.S. 523 (1967).

41. Id. at 528 (overruling Frank v. Maryland, 359 U.S. 360 (1959), which held that a health

inspector could enter a home even without first obtaining a warrant).

42. Id. at 537.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 857

dangerous conditions in homes would prevent an inspector from

being able to demonstrate probable cause.

Over the next fifteen years, the Court applied the administrative

search doctrine in numerous regulatory contexts, approving warrants

issued in the absence of individualized suspicion for Occupational

Safety and Health Administration workplace inspections,

and

permitting warrantless inspections of certain industries that were

closely regulated, such as liquor stores,

sellers of firearms,

and

mines.

In these cases, the Court added another rationale for

abandoning the individualized suspicion requirement: the individuals

and businesses engaged in these industries were fully aware of the

pervasive regulation in their field and therefore had a reduced

expectation of privacy.

Courts also applied the administrative search doctrine to a

category of cases known as inventory searches. Police officers

frequently take custody of personal property, either pursuant to

criminal activity (for example, when they arrest an intoxicated driver

and impound the car),

or pursuant to other duties (for example,

when they tow a car in violation of parking regulations, or find an

abandoned bag in a public place).

When police officers take custody

of such property, they routinely conduct an “inventory search.” As

the Supreme Court noted in 1976, the goals of such searches are not

the discovery or investigation of a crime, but rather to (1) protect the

owner’s property, (2) avoid false claims of lost or stolen property by

the owner, and (3) protect the police from potential danger.

Thus,

the Court concluded that inventory searches were part of the

43. Id. The Court also noted that these types of suspicionless inspections had been

traditionally accepted by the courts and by the general public. Id.

44. Marshall v. Barlow’s, Inc., 436 U.S. 307, 320 (1978).

45. Collonade Catering Corp. v. United States, 397 U.S. 72, 77 (1970).

46. United States v. Biswell, 406 U.S. 311, 317 (1972).

47. Donovan v. Dewey, 452 U.S. 594, 606 (1981).

48. See, e.g., Biswell, 406 U.S. at 315–16.

49. See, e.g., People v. Trusty, 516 P.2d 423, 424 (Colo. 1973).

50. See, e.g., People v. Sullivan, 272 N.E.2d 464, 466 (N.Y. 1971).

51. South Dakota v. Opperman, 428 U.S. 364, 378 (1976). Opperman merely confirmed the

overwhelming conclusion of the lower courts; prior to the case, almost every state and circuit

court that considered the question had held that inventory searches were permissible. Some

lower courts had held that an inventory search was not technically a search under the Fourth

Amendment, whereas some had concluded—as the Supreme Court ultimately did—that

inventory searches were reasonable under the Fourth Amendment. See id. at 369–71 (listing

over twenty-five state and circuit court opinions).

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

858 DUKE LAW JOURNAL [Vol. 59:843

“routine, administrative caretaking” functions of the police,

rather

than part of its law enforcement functions, and approved inventory

searches even in the absence of individualized suspicion.

In one sense, it is not surprising that some appellate courts

turned to the administrative search doctrine when seeking to justify

suspicionless antiterrorism searches in the early 1970s because at the

time there was no other doctrine that specifically permitted

suspicionless searches.

And given the very real and immediate

danger posed by the terrorists who were hijacking airplanes and

blowing up public buildings in the late 1960s, the government had a

reasonable argument that these searches served a purpose other than

crime control: to wit, they were designed to secure the safety of

airline passengers and government personnel.

As the Ninth Circuit

explained, the searches were “part of a general regulatory scheme in

furtherance of an administrative purpose, rather than as part of a

criminal investigation to secure evidence of crime.”

This rationale was not surprising—and perhaps even justifiable,

given the quasi emergency the country faced at the time. But even

during this early stage, the use of the administrative search doctrine

to justify antiterrorism searches was legally suspect. When juxtaposed

against inspections for building code safety violations and routine

cataloging of the contents of impounded vehicles, widespread

suspicionless searches at airports and at courthouses stand out rather

dramatically. In the first place, they are much broader in scope,

affecting a much larger portion of the population—today, it is fair to

say that the vast majority of the country has been subjected to a

52. Id. at 370 n.5.

53. Id. at 375–76; accord Colorado v. Bertine, 479 U.S. 367, 376 (1987).

54. As noted above, some courts did not apply any doctrine at all; they merely conducted a

reasonableness balancing test. Some courts analogized these searches to border searches,

whereas other courts applied the doctrine of implied consent. See supra notes 29–31 and

accompanying text.

55. See, e.g., United States v. Davis, 482 F.2d 893, 910 (9th Cir. 1973) (“In this and other

relevant respects, the airport search program is indistinguishable, for Fourth Amendment

purposes, from the warrantless screening inspection of air passengers and their luggage for plant

pests and disease . . . .”), abrogated by United States v. Aukai, 497 F.3d 955 (9th Cir. 2007) (en

banc). Although the Supreme Court has never directly reviewed the constitutionality of

suspicionless searches at public buildings or in airports, it has cited the circuit court cases with

approval in dicta, implying that upholding these suspicionless searches is a proper application of

the administrative search doctrine. See Nat’l Treasury Employees Union v. Von Raab, 489 U.S.

656, 675 n.3 (1989).

56. Davis, 482 F.2d at 910.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 859

search at an airport or a courthouse (or both). And in the second

place, the public safety purpose, although no doubt legitimate, does

not seem far removed from general crime control—in fact, if one

looks at the searches from a slightly different perspective, the purpose

seems to be indistinguishable from crime control.

As it turns out, the problem of distinguishing between public

safety and crime control became more severe over the next thirty

years, as two changes occurred. First, the administrative search

doctrine itself evolved to allow different types of searches, many of

which are de facto crime-control searches masquerading as

administrative searches. Second, as terrorism once again became a

real threat to the country, law enforcement officers sought to expand

the use of suspicionless antiterrorism searches to other contexts, and

the courts—unlike their predecessors in the early 1970s—began to

push back.

II.

THE SECOND ERA: 1977–2000

Although the danger from terrorism receded from 1977 to 2000,

suspicionless antiterrorism searches continued—if anything, they

became more entrenched in the law and in the public consciousness.

At the same time, law enforcement began to use—and courts

approved of—suspicionless searches in a number of other contexts: in

schools, in the workplace, on probationers and parolees, and on the

public roads.

A. Order Is Restored

Over the following two-and-a-half decades, the threat of

terrorism—perceived and actual—receded from the nation’s

concerns. The major outbreaks of violence conducted by such groups

as the Weathermen, the Black Panthers, and Cuban exile groups

faded into history,

whereas hijackings were virtually eliminated by

the end of the 1970s. But the suspicionless searches at airports and

most public buildings remained in place, finding a new practical (if

not doctrinal) justification in the minds of the populace and in the

courts. Given the relative period of calm, courts could no longer

57. There were some exceptions. The Fuerzas Armadas de Liberación Nacional (FALN), a

Puerto Rican Marxist terrorist group, carried out a number of bombings between 1974 and 1985.

None of these incidents, however, provoked a national shift in law enforcement tactics. See

Oscar Avila, Ex-Puerto Rican Radicals Work to Keep Cause Alive,

L.A. TIMES, Sept. 16, 2009, at

A17.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

860 DUKE LAW JOURNAL [Vol. 59:843

defend these suspicionless searches as extraordinary responses to an

“emergency” situation in which public buildings and airplanes faced

“imminent” danger from the “ruthless forces” bent on harming the

United States. Instead, the Supreme Court offered precisely the

opposite argument when it gave suspicionless airport searches its

stamp of approval (albeit in dicta) in 1989:

In the 15 years the [suspicionless search] program has been in

effect, more than 9.5 billion persons have been screened, and over

10 billion pieces of luggage have been inspected. By far the

overwhelming majority of those persons who have been

searched . . . have proved entirely innocent—only 42,000 firearms

have been detected during the same period. When the

Government’s interest lies in deterring highly hazardous conduct, a

low incidence of such conduct, far from impugning the validity of the

scheme for implementing this interest, is more logically viewed as a

hallmark of success.

There is little dispute that the hijacking epidemic ended in large

part because of the new security procedures at airports; thus, the

Court’s emphasis on the efficacy of such a program is entirely

sensible. No judge, commentator, or (probably) passenger would

have considered eliminating the searches even in the year 2000, after

a decade without any domestic passenger airplane hijackings.

The

thought of eliminating them in the wake of the terrorist attacks in

2001 is practically unthinkable. Nevertheless, both the Fifth Circuit’s

sentiment that Americans all “look forward to the day when

skyjackings and their sequels, airport search and security measures,

cease”

and the promise that the relaxed standards that allow

suspicionless airport searches will disappear “when the threat of air

piracy disappears”

now seem naïve at best. It is now clear that the

“threat of air piracy” will never disappear, and thus, the security

measures will never disappear either. But if suspicionless searches at

airports are now a permanent fixture in American society, it is even

more imperative that courts find a principled justification for these

58. Von Raab, 489 U.S. at 675 n.3 (citations omitted).

59. See O

FFICE OF CIVIL AVIATION SEC., supra note 11, at 75. There was an attempted

hijacking of a nonpassenger airplane in 1994, when a Federal Express employee attempted to

hijack one of the company’s planes. See J

EFFREY PRICE, PRACTICAL AVIATION SECURITY:

PREDICTING AND PREVENTING FUTURE THREATS 67 (2009).

60. United States v. Skipwith, 482 F.2d 1272, 1279 (5th Cir. 1973).

61. Id.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 861

searches—and as we will see,

the need to justify these searches has

been one of the factors that has turned the permissible suspicionless

search doctrine into a confusing tangle of inconsistent jurisprudence.

The policy of suspicionless searches at public buildings has also

long outlasted the violent epidemic of bombings and shootings that

created it. Like searches at airports, these security procedures have

become a familiar and unquestioned part of daily life; one simply

accepts that there is no way to enter many public buildings—to meet

with a public official, or to attend a session of open court—without

submitting to a search by a law enforcement official.

But searches at

public buildings are different from those at airports in three

fundamental ways. First, although almost everyone traces the

decrease in hijackings to the suspicionless search policy that was

implemented to prevent them, the decrease in bombings and

shootings inside public buildings can probably be traced to a number

of different factors.

Second, courts continue to point out that

suspicionless searches in airports are the only effective way to prevent

hijacking,

whereas there may be other ways of keeping public

buildings and courtrooms safe.

Finally, as the events of September

11, 2001 have shown, an airplane’s unique combination of

vulnerability and destructive power means that a hijacking could

potentially lead to catastrophic harm in terms of loss of life and

property. In short, although the two types of suspicionless searches

came into existence at the same time and were justified under similar

theories, there is, at least from a policy perspective, a much stronger

62. See infra Part III.

63. The unquestioned acceptance of these searches by the general population is perhaps

the most troubling aspect of permissible suspicionless searches. Not only does it mean that there

is no political will to change—or even critically examine—the policy of suspicionless searches at

airports and courthouses, but it could also ultimately mean that legally these procedures are no

longer even considered to be searches at all. Under the test put forward in Katz v. United States,

389 U.S. 347 (1967), a government surveillance or detection procedure is not a search unless it

violates the suspect’s “reasonable expectation of privacy.” Id. at 360 (Harlan, J., concurring). If

suspicionless searches at airports and courthouses are generally accepted by society, it could be

argued that they no longer violate a “reasonable expectation of privacy” because it would be

unreasonable to believe one had the right to board an airplane without being subjected to a

search. See Ric Simmons, From Katz to Kyllo: A Blueprint for Adapting the Fourth Amendment

to Twenty-First Century Technologies, 53 H

ASTINGS L.J. 1303, 1331–35 (2002).

64. Most importantly, the decades since the Vietnam War have not seen anything like the

level of social unrest that was produced by the racial tensions and antiwar protests of that era.

See supra note 9 and accompanying text.

65. United States v. Marquez, 410 F.3d 612, 616 (9th Cir. 2005).

66. See infra note 355 and accompanying text.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

862 DUKE LAW JOURNAL [Vol. 59:843

reason for continuing one kind of search than there is for the other.

Thus, a doctrine that permits suspicionless searches at airports need

not also cover suspicionless searches at public buildings.

B. Broadening the Scope of Permissible Suspicionless Searches

During this second era, the permissible suspicionless search

doctrine underwent an important evolution outside the context of

antiterrorism searches. Courts expanded the doctrine into a number

of different areas, including drug testing in schools,

drug testing of

public employees,

searches of probationers,

and drunk-driving

checkpoints.

But the most dramatic expansion in the law of

permissible suspicionless searches was a change in what it meant for a

search to have a noncriminal purpose—and this change occurred

subtly, perhaps accidentally, without any express acknowledgement

from the courts. As Part III demonstrates, this change had dramatic

implications for later courts who applied the permissible suspicionless

search doctrine in the terrorism context.

One example of this change in the suspicionless search doctrine

came in the administrative search context. In the 1987 case of New

York v. Burger,

the Supreme Court upheld the suspicionless search

of an automobile junkyard.

Under then-existing New York law,

every automobile junkyard had to keep records of all the auto parts

that came onto the premises, and submit both the records and the

junkyard itself to inspections whenever requested by an appropriate

state official.

Pursuant to this law, police officers from the Auto

Crimes Division of the New York Police Department searched

Burger’s junkyard, copied down the vehicle identification numbers of

some auto parts there, and later determined that the parts had come

from stolen vehicles.

Burger was subsequently charged and

convicted for possession of stolen property, and he appealed the

67. See infra Part II.B.1.

68. See infra Part II.B.2.

69. See infra Part II.B.3.

70. See infra Part II.B.4.

71. New York v. Burger, 482 U.S. 691 (1987).

72. Id. at 718.

73. See N.Y.

VEH. & TRAF. LAW § 415-a5 (McKinney 1986). The law allowed inspection by

an agent of the Department of Motor Vehicles or by a police officer. Id.

74. Burger, 482 U.S. at 694–95.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 863

conviction, claiming that the lack of individualized suspicion rendered

the search unconstitutional.

The Supreme Court applied the administrative search doctrine

and ruled that the search was not “aimed at the discovery of a

crime”—even though it was conducted by police officers and resulted

in a criminal prosecution.

The Court explained that “an

administrative scheme may have the same ultimate purpose as penal

laws, even if its regulatory goals are narrower;” and that this

particular administrative scheme “serves the regulatory goals of

seeking to ensure that vehicle dismantlers are legitimate

businesspersons and that stolen vehicles and vehicle parts passing

through automobile junkyards can be identified.”

During this same period, the Supreme Court began to approve of

suspicionless searches in other contexts, and in each context the

Court moved toward a weakening of the “noncriminal purpose”

requirement. Specifically, the Court approved of four new categories

of suspicionless searches: drug tests in public schools, drug tests of

public employees, searches of probationers and parolees, and

vehicular checkpoints. In broadening the scope of suspicionless

searches, the Court refined its terminology, moving from

“administrative search” to the broader term of “special needs

search.”

1. The Birth of Special Needs: Searches and Drug Tests in

Schools. The term special needs was first coined in New Jersey v.

T.L.O.,

in which a high school assistant principal searched the purse

of a student whom he suspected of smoking in school.

Although the

principal had neither a warrant nor probable cause, the Court upheld

the search for two reasons: first, the search was conducted for the

purpose of “maintaining discipline in the classroom and on school

grounds,”

not law enforcement purposes; and second, high school

students have a “lesser expectation of privacy” than the general

75. In applying the administrative search doctrine, the Supreme Court found that the

defendant had a reduced expectation of privacy because he was in a closely regulated industry,

id. at 703–07, and because suspicionless searches were necessary to further a substantial interest

of the state, id. at 708–10. The Court also noted that the statute provided notice, which was a

valid substitute for a warrant, and that it placed legitimate time, place, and scope restrictions on

the searches. Id. at 711–12.

76. Id. at 713–14 (emphasis added).

77. New Jersey v. T.L.O., 469 U.S. 325 (1985).

78. Id. at 336.

79. Id. at 339.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

864 DUKE LAW JOURNAL [Vol. 59:843

population.

In his concurrence, Justice Blackmun set out a test that

would be used in future cases: a lower standard for searches was

permitted “only in those exceptional circumstances in which special

needs, beyond the normal need for law enforcement, make the

warrant and probable cause requirement impracticable.”

In contrast with the administrative search cases, the special needs

test does not involve any explicit finding that the government

possessed no other viable means of meeting its objective. Blackmun’s

test, however, incorporates (and dilutes) this standard by stating that

the traditional requirements of the Fourth Amendment can only be

loosened if the special needs make adherence to those requirements

“impracticable.”

In the context of a search conducted by school

officials, the majority in T.L.O. stated that “[t]he warrant

requirement . . . is unsuited to the school environment” because it

“would unduly interfere with the maintenance of the swift and

informal disciplinary procedures needed in the schools.”

This first special needs case was not a suspicionless search

because the assistant principal had some reason to believe the

defendant was carrying contraband in her purse. Thus, the special

needs doctrine initially only meant that a court would relax the strict

warrant and probable cause requirements if there were special needs

beyond the normal needs of law enforcement.

In T.L.O., for

example, the Court held that the school officials must have

“reasonable grounds” to believe the student had broken a law or a

school rule.

But this requirement of individualized suspicion for searches of

public school students was abandoned ten years later in Vernonia v.

Acton,

in which the Court upheld mandatory, suspicionless drug

80. Id. at 348 (Powell, J., concurring).

81. Id. at 351 (Blackmun, J., concurring).

82. See id. at 340 (majority opinion) (“The warrant requirement, in particular, is unsuited

to the school environment: requiring a teacher to obtain a warrant before searching a child

suspected of an infraction of school rules (or of the criminal law) would unduly interfere with

the maintenance of the swift and informal disciplinary procedures needed in the schools.”).

83. Id.

84. See, e.g., Vernonia Sch. Dist. 47J v. Acton, 515 U.S. 646, 653 (1995); Griffin v.

Wisconsin, 483 U.S. 868, 873 (1987).

85. T.L.O., 469 U.S. at 341–42.

86. Vernonia Sch. Dist. 47J v. Acton, 515 U.S. 646 (1995).

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 865

testing of school athletes.

As in T.L.O., the Court in Acton

emphasized that schoolchildren had a “lesser expectation of

privacy.”

And the Court found other factors—not present in

T.L.O.—which would logically lead to a further loosening of the

Fourth Amendment standards: athletes in particular have a lower

expectation of privacy than other students;

a urinalysis is in some

ways more limited and less intrusive than the search of a purse;

the

results of the tests were revealed only to school administrators and

parents, not to law enforcement;

and alternate methods of deterring

drug use were likely to be less effective.

Unfortunately, the Court

did not explain which of these factors justified the critical doctrinal

shift from T.L.O.’s requirement of “reasonable grounds” to Acton’s

abandonment of individualized suspicion.

At first glance, abandoning the requirement of individualized

suspicion was the most significant part of the Acton opinion. But as

this Article demonstrates, suspicionless searches had already been

allowed in the context of administrative searches, so this was not an

unprecedented step.

In truth, the most significant aspect of Acton

was the specific special need that the Court used to justify deviation

from the strict Fourth Amendment requirements in the first place.

Unlike the special need in T.L.O. (maintaining discipline in the

classroom and on school grounds), Acton’s special need was barely

distinguishable from a standard law enforcement purpose:

“[d]eterring drug use by our Nation’s schoolchildren.”

In the Acton case, however, the Court could still legitimately

claim that the test was not designed to serve the ordinary needs of law

enforcement because the results of the test were not turned over to

law enforcement personnel.

Since Acton, this expansive definition of

87. Id. at 663–64.

88. Id. at 655–57.

89. See id. at 657.

90. See id. at 658.

91. Id. at 651, 658.

92. See id. at 663–64.

93. One can conclude by inference that the “reduced expectation of privacy” of athletes

was not one of the dispositive factors, because seven years later the Court approved

suspicionless drug testing of all students involved in extracurricular activities. See Bd. of Educ. v.

Earls, 536 U.S. 822, 825 (2002).

94. Suspicionless searches had also already been permitted in the context of automotive

checkpoints. See infra notes 139–46 and accompanying text.

95. Acton, 515 U.S. at 661.

96. Id. at 658.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

866 DUKE LAW JOURNAL [Vol. 59:843

special needs has broadened to other cases in which the results of the

search are turned over to law enforcement,

thus effectively removing

the original justification for departing from the Fourth Amendment’s

traditional requirement of individualized suspicion.

2. Searches and Drug Tests of Public Employees. Only one year

after T.L.O. was decided, a plurality of the Court applied the special

needs test in O’Connor v. Ortega,

in effect upholding a search of a

public hospital employee’s desk.

As in T.L.O., the reason underlying

the intrusion was not the normal need of law enforcement, but rather

“the government’s need for supervision, control, and the efficient

operation of the workplace;”

100

thus, the search was not impermissible

even though there was no warrant and no probable cause.

Ortega’s analysis was similar to T.L.O.’s in many ways. The

Court stated that a warrant or probable cause requirement would be

impracticable in the context of the public workplace because the

supervisors who conduct these searches—like school administrators—

“are hardly in the business of investigating the violation of criminal

laws,” and therefore could not be expected to understand or comply

with “unwieldy warrant procedures”

101

or “the subtleties of the

probable cause standard.”

102

Also, just like the schoolchildren in

T.L.O. and the heavily regulated businesses in Camara, the Court

noted that office workers have a diminished expectation of privacy in

97. See infra notes 147–50 and accompanying text.

98. O’Connor v. Ortega, 480 U.S. 709 (1987).

99. Cf. id. at 722 (“In our view, requiring an employer to obtain a warrant whenever the

employer wished to enter an employee’s office, desk, or file cabinets for a work-related purpose

would seriously disrupt the routine conduct of business and would be unduly burdensome.

Imposing unwieldy warrant procedures in such cases upon supervisors, who would otherwise

have no reason to be familiar with such procedures, is simply unreasonable.”).

100. Id. at 720; see also id. at 723 (“Government agencies provide myriad services to the

public, and the work of these agencies would suffer if employers were required to have probable

cause before they entered an employee’s desk for the purpose of finding a file or piece of office

correspondence. Indeed, it is difficult to give the concept of probable cause, rooted as it is in the

criminal investigatory context, much meaning when the purpose of a search is to retrieve a file

for work-related reasons. Similarly, the concept of probable cause has little meaning for a

routine inventory conducted by public employers for the purpose of securing state property. To

ensure the efficient and proper operation of the agency, therefore, public employers must be

given wide latitude to enter employee offices for work-related, noninvestigatory reasons.”

(citations omitted)).

101. Id. at 722.

102. Id. at 724–25.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 867

their workplace.

103

And unsurprisingly, the Court ordered the district

court to apply the same standard to evaluate the search: whether the

government had “reasonable grounds” to suspect that the search

would turn up evidence of wrongdoing.

104

And like T.L.O., Ortega laid the groundwork for future cases in

which the Court abandoned the individualized suspicion requirement

for public employees. In a pair of cases in 1989, the Court upheld

suspicionless mandatory drug tests for railroad employees who had

been involved in accidents

105

and customs agents who used firearms or

engaged in drug interdiction.

106

In Skinner v. Railway Labor

Executives’ Ass’n,

107

the Court determined that the purpose of drug

tests for railroad employees was “ensuring the safety of the traveling

public and of the employees themselves,”

108

noting that the tests were

administered “not to assist in the prosecution of employees, but

rather to prevent accidents and casualties in railroad operations that

result from impairment of employees by alcohol or drugs.”

109

National Treasury Employees Union v. Von Raab,

110

the Court held

that the purpose of drug tests for United States Customs agents was

“to deter drug use among those eligible for promotion to sensitive

positions within the [Customs] Service and to prevent the promotion

of drug users to those positions.”

111

Specifically, the Court noted that

the Customs Service sought to prevent employees with guns from

injuring themselves or others, and to ensure that those protecting the

nation’s borders maintained good judgment and were not vulnerable

to blackmail.

112

In both Skinner and Von Raab, the Court cited the familiar

justifications for allowing suspicionless searches: a limited intrusion

103. Id. at 717 (“An office is seldom a private enclave free from entry by supervisors, other

employees, and business and personal invitees. Instead, in many cases offices are continually

entered by fellow employees and other visitors during the workday for conferences,

consultations, and other work-related visits. Simply put, it is the nature of government offices

that others—such as fellow employees, supervisors, consensual visitors, and the general public—

may have frequent access to an individual’s office.”).

104. Id. at 726.

105. Skinner v. Ry. Labor Executives’ Ass’n, 489 U.S. 602, 606 (1989).

106. Nat’l Treasury Employees Union v. Von Raab, 489 U.S. 656, 660–61 (1989).

107. Skinner v. Ry. Labor Executives’ Ass’n, 489 U.S. 602 (1989).

108. Id. at 621.

109. Id. at 620–21 (internal quotation marks omitted).

110. Nat’l Treasury Employees Union v. Von Raab, 489 U.S. 656 (1989).

111. Id. at 666.

112. See id. at 670–71.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

868 DUKE LAW JOURNAL [Vol. 59:843

on privacy,

113

reduced expectation of privacy on the part of the

subjects,

114

and the fact that imposing a warrant requirement would be

impractical and would frustrate the government purpose.

115

But most

significantly, Skinner and Von Raab paralleled the evolution of

special needs that would later be seen in Acton: in both Skinner and

Von Raab, the purpose of the drug test was essentially to deter illegal

activity (drug use), with the dubious argument that deterring illegal

activity went beyond the standard goals of law enforcement because

of the highly dangerous or sensitive positions that the employees

occupied.

116

As with Acton, the Court in Skinner and Von Raab could

credibly claim that the test was not meant to serve the ordinary goals

of law enforcement because the results of the drug tests were not

turned over to the police

117

—but as in Acton, the Court did not

expressly state that this restraint on the part of the state was a

necessary element in determining that the search was conducted for a

special need.

3. Searches of Probationers and Parolees. In the same year that

Ortega was decided, the Court in Griffin v. Wisconsin

118

found a

special need justifying the search of a probationer’s home.

119

Again,

the search was conducted without a warrant or probable cause, and

again, the Court upheld the search for two reasons: first, because of

the lower level of Fourth Amendment protection due to

113. See Skinner, 489 U.S. at 624; Von Raab, 489 U.S. at 672 n.2.

114. See Skinner, 489 U.S. at 627 (finding a diminished expectation of privacy due to the

heavy regulation of the industry); Von Raab, 489 U.S. at 672 (ruling that customs agents should

expect inquiries into their fitness to perform their job).

115. See Skinner, 489 U.S. at 623–24, 628 (noting that drug users sometimes show no

outward signs giving rise to probable cause; further noting that private railway employers are

not experts in the warrant requirements or the subtleties of the law on probable cause); Von

Raab, 489 U.S. at 666–67 (holding that a warrant requirement would “divert valuable agency

resources from the Service’s primary mission”).

116. The greater danger posed regarding drug use by schoolchildren, railway operators, and

customs agents is an important factor: a few years later, the Court found there was no special

need to justify a Georgia law that required drug testing of all candidates for elected office

because the officials did not “perform high risk, safety-sensitive tasks.” Chandler v. Miller, 520

U.S. 305, 321–22 (1997).

117. See Skinner, 489 U.S. at 621 n.5 (noting that although the results of the test could be

turned over to law enforcement, there was no evidence that the public employer had ever done

so, nor that it ever intended to do so); Von Raab, 489 U.S. at 666 (“Test results may not be used

in a criminal prosecution of the employee without the employee’s consent.”).

118. Griffin v. Wisconsin, 483 U.S. 868 (1987).

119. Id. at 872–74.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 869

probationers;

120

and second, because the goal of the search was not

law enforcement, but instead ensuring “that the probation serves as a

period of genuine rehabilitation and that the community is not

harmed by the probationer’s being at large.”

121

And what of the third factor cited in previous special needs

cases—that requiring individualized suspicion was infeasible? In past

special needs cases such as T.L.O. and Ortega, the Court relied on

this factor, noting that it was impracticable for teachers, school

administrators, or public agency supervisors to learn and understand

the complexities of the warrant requirement and probable cause.

122

But in Griffin, this factor was whittled down more or less into

oblivion because there is no real argument that probation officers

cannot understand and apply the warrant or probable cause

standards. The best the Court could do was to explain that the

government interest might be “unduly disrupted by a requirement of

probable cause,”

123

because “so long as [the probationer’s] illegal (and

perhaps socially dangerous) activities were sufficiently concealed as

to give rise to no more than reasonable suspicion, they would go

undetected and uncorrected.”

124

This statement is equally true in the

traditional law enforcement context; thus, it provides no extra

ammunition for the argument that a lesser Fourth Amendment

standard should apply.

125

The Griffin Court required the government to show “reasonable

grounds” before searching a probationer’s home, but as with public

schoolchildren and public employees, the Court eventually began

allowing these searches without any showing of individualized

suspicion at all.

126

As this shift occurred, however, searches of

120. See id. at 874 (noting that probationers and parolees “do not enjoy ‘the absolute liberty

to which every citizen is entitled, but only . . . conditional liberty properly dependent on

observance of special [probation] restrictions’” (alterations in original) (quoting Morrissey v.

Brewer, 408 U.S. 471, 480 (1972))).

121. Id. at 875.

122. See O’Connor v. Ortega, 480 U.S. 709, 720, 722 (1987); New Jersey v. T.L.O., 469 U.S.

325, 352–53 (1985).

123. Griffin, 483 U.S. at 878.

124. Id.

125. The Griffin Court also added another consideration, unique to the probationer context:

that the ongoing, supervisory relationship between the probation officer and the probationer

would provide extra information to the probation officer in deciding whether to conduct a

search. Id. at 879.

126. The latest product of this evolution is Samson v. California, 547 U.S. 843 (2006),

discussed infra at note 133.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

870 DUKE LAW JOURNAL [Vol. 59:843

probationers and parolees slowly left the orbit of the special needs

category. A series of circuit court cases in the early 2000s focused on

mandatory DNA extraction from prison inmates, parolees, and

probationers, and these cases made it clear that the primary

consideration in analyzing these searches was the subjects’

dramatically reduced expectation of privacy rather than the purpose

of the search. Numerous circuits struggled to find a way to

characterize the purpose of taking DNA samples as a “special need

beyond the normal need for law enforcement,”

127

which turned out to

be a logically impossible task because the DNA database being

assembled was used for traditional law enforcement purposes. Other

circuits abandoned the special needs test altogether and simply

applied a “totality of the circumstances” test, allowing the DNA

extraction because of the subjects’ substantially diminished

expectation of privacy.

128

Finally, in 2001 the Supreme Court itself abandoned the special

needs requirement for searches of probationers in United States v.

Knights.

129

At issue was a probationary condition that allowed the

authorities to search the probationer’s home at any time.

130

The Court

acknowledged that the breadth of the probationary condition meant

that a search could be conducted for any purpose, including a law

enforcement purpose,

131

and so the special needs doctrine could not

justify the condition. The Court cited Griffin, however, for the

proposition that a probationer’s expectation of privacy was very

127. Nicholas v. Goord, 430 F.3d 652, 671 (2d Cir. 2005); see also id. at 669 (upholding DNA

collection from certain convicted felons because “[a]lthough the DNA samples may eventually

help law enforcement identify the perpetrator of a crime,” they are not being used for evidence

or investigations “at the time of collection”), overruled in part by Samson, 547 U.S. 843, as stated

in United States v. Amerson, 483 F.3d 73, 79 n.5 (2007); Green v. Berge, 354 F.3d 675, 677, 679

(7th Cir. 2004) (upholding DNA collection from prisoners because “the government has a

special need in obtaining identity DNA samples”).

128. See, e.g., United States v. Sczubelek, 402 F.3d 175, 184 (3d Cir. 2005) (upholding DNA

sample of probationers as a reasonable search under the totality of the circumstances); Padgett

v. Donald, 401 F.3d 1273, 1280 (11th Cir. 2005) (same for prisoners); Groceman v. Dep’t of

Justice, 354 F.3d 411, 413 (5th Cir. 2004) (per curiam) (probationers); United States v. Kincade,

379 F.3d 813, 839 (9th Cir. 2004) (en banc) (prisoners); Jones v. Murray, 962 F.2d 302, 307 (4th

Cir. 1992) (same).

129. United States v. Knights, 534 U.S. 112 (2001).

130. Id. at 114.

131. Id. at 116.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 871

low,

132

and thus the search was reasonable, given the strong

government interest in monitoring its probationers.

133

4. Vehicular Checkpoints. Suspicionless stops at vehicular

checkpoints have a long history. Even before the special needs

doctrine was articulated in T.L.O., the Court had extended the

administrative search doctrine of Camara

134

into the automotive

checkpoint context.

The first automotive checkpoint cases involved immigration

stops. As noted above, the legal authority to conduct these stops at

the border has been well established for nearly one hundred years.

135

Similarly, international mail is subject to suspicionless searches by

customs officials,

136

and even files on laptop computers can be opened

and read during customs searches without individualized suspicion.

137

As a circuit court explained, border searches are conducted “for the

purposes of collecting duties and intercepting contraband destined for

the interior of the United States.”

138

After Camara in 1967, however, the Court began to use the

administrative search doctrine to expand the scope of border searches

to include seizures that were nowhere near the border. In 1975, the

Court stated that the warrant and probable cause requirement did not

apply to roving patrols inside the border.

139

One year later, in United

States v. Martinez-Fuerte,

140

the Court upheld the constitutionality of a

suspicionless immigration checkpoint that took place nearly sixty

miles inside the border.

141

The Court relied on Camara and its

progeny for the principle that, although “individualized suspicion is

usually a prerequisite to a constitutional search or seizure . . . . the

132. Id. at 119.

133. Id. at 120–21. In a case five years later, the Court similarly ignored the special needs

test in upholding a California law that allowed parolees to be subject to suspicionless searches.

Samson v. California, 547 U.S. 843, 846, 855 n.4 (2006).

134. See supra notes 40–43 and accompanying text.

135. See supra note 39 and accompanying text.

136. See, e.g., United States v. Ramsey, 431 U.S. 606, 622 (1977); United States v. Seljan, 547

F.3d 993, 1008 (9th Cir. 2008) (en banc).

137. United States v. Arnold, 533 F.3d 1003, 1008 (9th Cir. 2008).

138. United States v. Robles, 45 F.3d 1, 5 (1st Cir. 1995).

139. United States v. Brignoni-Ponce, 422 U.S. 873, 883 (1975) (stating in dictum that law

enforcement officers in border areas could pull over cars based on less than probable cause, but

stating that there must be some individualized reasonable suspicion).

140. United States v. Martinez-Fuerte, 428 U.S. 543 (1976).

141. Id. at 566–67.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

872 DUKE LAW JOURNAL [Vol. 59:843

Fourth Amendment imposes no irreducible requirement of such

suspicion.”

142

The Court justified the lower Fourth Amendment standard on

numerous grounds. One was the low level of intrusion involved in the

stop—a brief detention, a few questions, the production of

documents, and a visual inspection of the car.

143

Another justification

was the diminished expectation of privacy possessed by individuals in

automobiles.

144

But these rationales alone were not sufficient to justify

suspicionless stops, so in an attempt to tie the automotive checkpoint

cases to the administrative search doctrine, the Court also

emphasized two other aspects of the case: the necessity of these

techniques to prevent the influx of illegal aliens (a flow that “cannot

be controlled effectively at the border”);

145

and a non–law

enforcement purpose, which it described as the public interest in

denying illegal aliens “a quick and safe route into the interior.”

146

The Court later upheld vehicular checkpoints for the purposes of

detecting and apprehending drunk drivers. In that case, Michigan

Department of State Police v. Sitz,

147

the Court rejected the special

needs test and instead applied a balancing test from other seizure

cases.

148

Because the danger posed to public safety by drunk drivers

was so great, and the intrusion on Fourth Amendment rights by the

stops was so slight, the Court concluded that the checkpoints were

reasonable.

149

As with the suspicionless searches in Burger, Skinner, Von Raab,

and Acton, there is a fine line between vehicular checkpoints

designed for general crime control and those designed to meet a

special need or a distinct public interest. By the year 2000, the

suspicionless search doctrine had evolved from covering a narrowly

defined group of cases involving truly routine and administrative

procedures to covering a wide range of searches that revealed

142. Id. at 560–61.

143. Id. at 558.

144. Id. at 561; see also United States v. Chadwick, 433 U.S. 1, 12 (1977) (claiming a

“diminished expectation of privacy” with regard to vehicles).

145. Martinez-Fuente, 428 U.S. at 556.

146. Id. at 557.

147. Mich. Dep’t of State Police v. Sitz, 496 U.S. 444 (1990).

148. Id. at 455. The Court relied most prominently on Brown v. Texas, 443 U.S. 47 (1979).

See Sitz, 496 U.S. at 448–50, 453–54.

149. Sitz, 496 U.S. at 449–52.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 873

evidence of criminal activity—though the doctrine’s approval rested

on whether the purpose of the search was something other than

detecting criminal activity.

150

As a result, by the end of the second era,

antiterrorism suspicionless searches at airports and courthouses no

longer seemed out of place as an administrative or special needs

search—the doctrine had effectively caught up with the reality of

antiterrorism searches.

But in this final period, during the second wave of terrorist

attacks, the government began overstepping its bounds, and courts

began to resist applying the special needs doctrine to suspicionless

antiterrorism searches. At the same time, the Supreme Court began

to limit the application of the special needs doctrine in other contexts,

further complicating the doctrinal justification for antiterrorist

searches.

III.

THE THIRD ERA: 2001–2010

The fight against terrorism has been one of the defining

characteristics of this decade. In response to the devastating attacks of

September 11 and subsequent attacks in Europe, the United States

government instituted a number of aggressive law enforcement

initiatives to prevent future terrorist attacks. Among these initiatives

was a sweeping expansion of antiterrorism suspicionless searches,

which were instituted near potential terrorist targets such as

reservoirs, political protests, sports arenas, and public transportation.

At the same time, however, courts finally began to limit and restrict

the use of suspicionless searches, both inside and outside the context

of terrorism.

A. The War on Terror

The attacks of September 11, 2001 killed nearly 3,000 Americans

in a single day.

151

They were followed by similar attacks on civilians

throughout Europe, including the 2004 bombing of four commuter

trains in Madrid, which killed 191 people and injured nearly 2,000

150. In fact, the antiterrorism searches provided some authority for this shift because a

number of cases that expanded the scope of suspicionless searches cited the earlier courts’

unanimous approval of airport and courthouse searches as evidence that public safety was a

legitimate special need, distinct from general crime control. See, e.g., Nat’l Treasury Employees

Union v. Von Raab, 489 U.S. 656, 675 n.3 (1989).

151. Sean Alfano, War Casualties Pass 9/11 Death Toll, CBS

NEWS, Sept. 22, 2006,

http://www.cbsnews.com/stories/2006/09/22/terror/main2035427.shtml.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

874 DUKE LAW JOURNAL [Vol. 59:843

more,

152

and the 2005 bombing of three subway trains in London,

which killed 52 people and injured over 700.

153

The threat of terrorist attacks, and the resulting “war on terror,”

have been two of the defining characteristics of this decade.

154

Unlike

the late 1960s, when the country faced hundreds of small-scale acts of

violence as a result of widespread social unrest, the war on terror in

the 2000s is primarily defined by the devastating attacks on a single

day in 2001.

155

But the popular sentiment and the government reaction

has otherwise been similar, and just like in the early 1970s, the

government has responded to the terrorist threat with aggressive law

enforcement techniques, many of which have restricted civil

liberties.

156

The most dramatic responses—the USA PATRIOT Act,

indefinite detention at Guantánamo Bay, warrantless wiretapping,

and enhanced interrogation techniques—have been subject to

significant public criticism,

157

and most have subsequently been

repudiated by the Supreme Court or reversed by the Obama

administration.

158

152. Sean Clarke, Major Terrorist Attacks Since 9/11, GUARDIAN, Jul. 7, 2005, http://www.

guardian.co.uk/world/2005/jul/07/terrorism.uk1.

153. Attack on London, G

UARDIAN, July 7, 2005, http://www.guardian.co.uk/flash/0,,15388

19,00.html.

154. Terrorism did not even register as an election issue in 2000, but in 2002 it was seen as

the second-most important challenge for the government to address. See Economy Now the

Most Important Issue for Americans by a Wide Margin, H

ARRIS INTERACTIVE, Sept. 24, 2008,

http://www.harrisinteractive.com/harris_poll/index.asp?PID=951 (showing that, of people

surveyed in 2002, 17 percent offered unprompted replies of “terrorism” as an election issue,

making it the second most common unprompted answer).

155. There have been a series of threats and even fatalities since then, such as an individual

who attempted to board a passenger airplane with a bomb in his shoe, see Shoe Bomb Suspect to

Remain in Custody, CNN.

COM, Dec. 25, 2001, http://archives.cnn.com/2001/US/12/24/

investigation.plane/index.html, and an unidentified person who sent Anthrax through the mail,

see FBI, Amerithrax Investigation, http://www.fbi.gov/anthrax/amerithraxlinks.htm

(last visited

Jan. 3, 2010).

156. For an overview and analysis of many of these responses, see Parry, supra note 3, at

770–82.

157. See, e.g., Cate Doty, Gore Criticizes Expanded Terrorism Law, N.Y.

TIMES, Nov. 10,

2003, at A1 (criticizing expanded security measures).

158. See, e.g., Hamdan v. Rumsfeld, 548 U.S. 557, 635 (2006) (ruling that the Bush

administration’s tribunals trying enemy combatants held at Guantánamo Bay were

unconstitutional); Scott Shane, Mark Mazzetti & Helene Cooper, Obama Reverses Key Bush

Security Policies, N.Y.

TIMES, Jan. 22, 2009, at A1 (“Mr. Obama signed executive orders closing

the detention camp at Guantánamo Bay, Cuba, within a year; ending the Central Intelligence

Agency’s secret prisons; and requiring all interrogations to follow the noncoercive methods of

the Army Field Manual.”).

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 875

One of the less publicized reactions to this second wave of

terrorism has been local law enforcement’s increasing use of

suspicionless searches in new contexts. Local police have conducted

suspicionless searches at subway entrances,

159

on ferries,

160

near the

Republican and Democratic conventions,

161

near reservoirs,

162

protest rallies,

163

at hockey arenas,

164

and at football stadiums.

165

Courts

have analogized these cases to the suspicionless antiterrorism

searches from the past era, applying the administrative search

doctrine to determine whether the search was permissible. But unlike

the searches in airports and public buildings that comprised the first

wave of antiterrorism suspicionless searches, these searches have not

been met with unanimous approval by the courts.

1. Reservoirs. The opening act in this second stage of

antiterrorism searches occurred one month after the September 11

attacks, in the middle of the night on a small country road near the

Cobble Mountain Reservoir in Blandville, Massachusetts.

166

The

Cobble Mountain Reservoir supplies water to a number of towns and

cities in Massachusetts, and in the wake of the September 11 attacks,

the state police were concerned that a terrorist might try to

contaminate the water supply or destroy the dam itself.

167

Troopers

were stationed on the Cobble Mountain Road, which runs along the

length of the reservoir, with orders to stop all vehicles that travelled

along the road and speak to the drivers.

168

The officers stopped David Carkhuff and immediately noticed

that he was intoxicated, and then arrested him for driving under the

influence.

169

Carkhuff challenged the constitutionality of the stop,

159. See infra Part III.A.4.

160. See infra Part III.A.4.

161. See infra Part III.A.2.

162. Commonwealth v. Carkhuff, 804 N.E.2d 317, 318 (Mass. 2004); see also infra Part

III.A.1.

163. Bourgeois v. Peters, 387 F.3d 1303, 1306 (11th Cir. 2004); see also infra Part III.A.2.

164. State v. Seglen, 700 N.W.2d 702, 705 (N.D. 2005); see also infra Part III.A.3.

165. Johnston v. Tampa Sports Auth., 442 F. Supp. 2d 1257, 1259–60 (M.D. Fla. 2006), rev’d

per curiam, 490 F.3d 820 (11th Cir. 2007), vacated and superseded on reh’g, 530 F.3d 1320 (11th

Cir. 2008) (per curiam); see also infra Part III.A.3.

166. Carkhuff, 804 N.E.2d at 318.

167. Id. at 318–19.

168. Id. The troopers also had orders to conduct a search of every truck that drove down the

road. Id.

169. Id. at 319.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

876 DUKE LAW JOURNAL [Vol. 59:843

noting that the troopers had no reason to suspect him of having

committed any crime at the time he was ordered to stop.

170

The state

urged the court to apply the special needs doctrine, arguing that the

troopers’ policy of suspicionless stops was analogous to the well-

established searches at airports and public buildings.

171

As the state

noted, “the search and seizure protocols at those facilities,” like the

seizure protocol near the reservoir, are not designed for a law

enforcement purpose, but rather “to assure that persons entering do

not have the means to destroy those facilities or to disrupt their

operation.”

172

But the Massachusetts Supreme Court disagreed. The court held

that the purpose of the search crossed the line from protecting public

safety to straightforward law enforcement, and thus the state

troopers’ actions did not meet the requirements of an administrative

search.

173

The court based this conclusion on the lack of notice given

to the defendant, noting that in the case of searches at airports and

public buildings, the individual receives prior notice that the search

will occur.

174

This allows the individual to avoid the search altogether

by “electing not to board the aircraft (or enter the court house).”

175

Giving prior notice reinforces the argument that the purpose of the

search is to protect the vulnerable facility rather than to detect and

apprehend criminals because “the objective of preventing dangerous

persons from gaining access is still accomplished” if a terrorist simply

avoids the facility altogether.

176

2. Political Protests. One year after Carkhuff was illegally

searched near the Cobble Mountain Resevoir, another antiterrorism

suspicionless search took place outside a military base in Columbus,

Georgia. A protest group was planning its annual demonstration

outside of Fort Benning,

177

and the city of Columbus instituted a new

170. Id. at 318.

171. Id. at 320.

172. Id. The state also used analogies to cases upholding searches at the entrances to

military bases. See, e.g., United States v. Miles, 480 F.2d 1217, 1219 (9th Cir. 1973) (per curiam).

173. Carkhuff, 804 N.E.2d at 320.

174. Id. at 322–23.

175. Id. at 322.

176. Id. at 323.

177. Bourgeois v. Peters, 387 F.3d 1303, 1306 (11th Cir. 2004). Fort Benning was the host of

the Western Hemisphere Institute for Security Cooperation, otherwise known as the “School of

the Americas,” which trains military leaders from other Western Hemisphere countries in

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 877

policy requiring every participant in the protest to submit to a search

prior to proceeding to the protest site.

178

The city argued in its brief

that the September 11 attacks had—or at least should have—

fundamentally changed the rules for antiterrorism suspicionless

searches, stating that “[l]ocal governments need an opinion that,

without question, allows [suspicionless] non-discriminatory, low-level

magnetometer searches at large gatherings.”

179

The Eleventh Circuit refused to give the city such an opinion. In

Bourgeois v. Peters,

180

the court found the city’s argument “troubling”

because “the threat of terrorism . . . cannot [be] use[d] . . . as the basis

for restricting the scope of the Fourth Amendment’s protections in

any large gathering of people.”

181

If a generalized threat of terrorism

were sufficient to justify mass suspicionless searches, law enforcement

could institute these searches for any large event, including “a high

school graduation, a church picnic, a public concert in the park, an art

festival, a Fourth of July parade, sporting events such as a marathon,

and fund-raising events such as the annual breast cancer walk.”

182

Like the state of Massachusetts, the city of Columbus invoked

the special needs doctrine, arguing that the purpose of searching the

protesters was not to detect crime or enforce the criminal law, but

instead to “ensure the safety of participants, spectators, and law

enforcement.”

183

The Bourgeois court rejected this argument, ruling

that the goals of public safety and law enforcement were “inextricably

intertwined” in this context.

184

The difference is no more than

semantic, according to the court; under the city’s argument, “a search

intended to enforce a given law would be permissible so long as the

government officially maintained that its purpose was to secure the

counterinsurgency tactics. Id. The protest group, known as the School of the Americas Watch

(SAW), had conducted a peaceful protest of approximately 15,000 people every year for the

thirteen years prior to the events in this case. Id.

178. The search was essentially identical to an airport screening—every protester was

required to walk through a metal detector, and if the detector indicated the presence of metal,

the police would conduct a more thorough search of the protestor’s person and possessions. Id.

at 1307.

179. Id. at 1311 (quoting Brief of Appellees at 13, Bourgeois, 387 F.3d 1303 (No. 02-16886),

2003 WL 23960109).

180. Bourgeois v. Peters, 387 F.3d 1303 (11th Cir. 2004).

181. Id. at 1311.

182. Id. (quoting Reply Brief of Appellants at 4, Bourgeois, 387 F.3d 1303 (No. 02-16886),

2003 WL 23960108).

183. Id. at 1312 (quoting Brief of Appellees, supra note 179, at 11).

184. Id. at 1312–13.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

878 DUKE LAW JOURNAL [Vol. 59:843

objectives that motivated the law’s enactment in the first place (e.g.,

public safety) rather than simply to enforce the law for its own

sake.”

185

But the Bourgeois court did not explain why this argument

should apply to searches at political protests and not to searches at

airports and public buildings, which are also searches intended to

detect criminal activity that are justified on the grounds that they

protect public safety. Indeed, the Bourgeois court—somewhat

inexplicably—does not mention the airport and public buildings cases

at all. Thus, the court does not provide any principled way to

distinguish between antiterrorism suspicionless searches at airports

and public buildings (which courts have consistently held to be

permissible) and antiterrorism suspicionless searches at political

protests (which the court held were impermissible).

186

Two years later, in preparation for the Republican National

Convention of 2004, New York City attempted to institute its own

suspicionless search policy for political protesters. To enhance

security during the convention, the New York City Police

Department (N.Y.P.D.) planned to search the bags of every protestor

before allowing them to proceed to the demonstration site.

187

Stauber v. City of New York,

188

the protestors challenged this policy

and sought a preliminary injunction to prevent the police from

conducting these searches.

189

Once again, the defendant invoked the

special needs doctrine, citing the cases upholding suspicionless

searches at airports.

190

And once again, the court rejected this

argument.

191

But unlike the Bourgeois court, the judge in Stauber addressed

the airport cases directly, and distinguished them from searches at

185. Id. at 1313.

186. The Bourgeois court did note in the facts that the demonstrations had been peaceful for

thirteen years, with no weapons ever found and no arrests for violence, id. at 1306, but it did not

refer back to these facts during its legal analysis.

187. Stauber v. City of New York, Nos. 03 Civ. 9162, 03 Civ. 9163, 03 Civ. 9164, 2004 U.S.

Dist. LEXIS 13350, at *3 (S.D.N.Y. July 19, 2004). The N.Y.P.D. also announced a number of

other security measures, such as requiring demonstrators to assemble within “pens” that were

made up of metal barricades and restricting access to the sites of the demonstration. Id. at *2–3.

188. Stauber v. City of New York, Nos. 03 Civ. 9162, 03 Civ. 9163, 03 Civ. 9164, 2004 U.S.

Dist. LEXIS 13350 (S.D.N.Y. July 19, 2004).

189. Id. at *1.

190. Id. at *84–87.

191. Id. at *84–90.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 879

protest sites. According to the court, the primary difference between

the two cases was the nature of the security threat: for the New York

City protestors, the threat was “overly vague”: the defendant had

only shown that the United States government “considers the

Convention to be a potential terrorism target,”

192

and a “general

invocation of terrorist threats” was not sufficient to justify the search

program.

193

In contrast, searches at airports were “implemented in

response to specific information about the threats faced by

officials;”

194

thus, the threat was real and concrete, and law

enforcement was allowed to respond accordingly.

195

3. Sports Arenas. A year earlier, the issue of suspicionless

antiterrorism searches arose in yet another context: a college hockey

game between the University of North Dakota and the University of

Minnesota in Grand Forks, North Dakota.

196

For security reasons,

University police officers at the entrances to the game subjected

every student to a pat-down search before entering the arena.

197

When

the police searched Scott Seglen, an underage student, they detected

and recovered a can of Coors Light in his pocket.

198

Seglen challenged

the constitutionality of the search, and the state analogized the search

to suspicionless searches at airports and public buildings, arguing that

such searches were even more reasonable in the wake of the attacks

of September 11.

199

The North Dakota Supreme Court also rejected this analogy,

citing Bourgeois, and held that the suspicionless searches were

unconstitutional.

200

Unlike the Eleventh Circuit in Bourgeois,

192. Id. at *88.

193. Id. at *89.

194. Id. at *88–89.

195. The court also noted other significant differences between airport searches and

protestor searches: (1) the searches at airports involve (as a first step) only metal detectors,

which are less intrusive than bag searches; (2) airport searches do not affect a person’s

constitutional right to expression; and (3) the police department gives no advance notice as to

whether they will be searching at a particular demonstration. Id. at *84–88.

196. State v. Seglen, 700 N.W.2d 702 (N.D. 2005).

197. Id. at 705. One of the original justifications for the search was to prevent students from

carrying animal carcasses into the arena and then throwing them on the ice during the game. Id.

at 705. During the court case, however, the state argued for broader “security needs” that are

“similar to airports and courthouses, especially in recent years.” Id. at 707.

198. Id. at 705.

199. Id. at 707–08.

200. Id. at 708.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

880 DUKE LAW JOURNAL [Vol. 59:843

however, the court in State v. Seglen

201

did provide one hint as to why

searches at hockey arenas should be treated differently than searches

at airports and public buildings—and its reasoning was similar to that

of the judge in Stauber. The court noted that airport and public

building searches were held to be constitutional only after

“unprecedented airport bombings, aircraft piracy, and courtroom

violence;”

202

in contrast, there was “no history of injury or violence

presented” in the Seglen case.

203

Whether a history of injury and

violence at sports arenas would actually have changed the outcome of

the case—and if so, what level of history of injury and violence would

be required—was left unexplained.

The next court to hear a suspicionless search case at a sporting

event was even more explicit in its reasoning. In Johnston v. Tampa

Sports Authority,

204

a federal district court judge ruled that

suspicionless searches of fans entering a professional football stadium

were unconstitutional. The court applied a special needs analysis, and

(unlike the Eleventh Circuit in Bourgeois), did not dispute the fact

that ensuring public safety could be a special need distinct from law

enforcement purposes.

205

But the court then followed the reasoning of

the Stauber court, explicitly stating that for an antiterrorism

suspicionless search to be constitutional, the danger of a terrorist

attack must be a “concrete danger”—that is, a “substantial and real”

danger, not merely a generalized risk.

206

In the case of Tampa Bay

stadium, the Johnston court concluded that “there was no testimony

or evidence of a particularized threat to NFL games or to the

[Tampa] stadium,”

207

merely a “general threat that terrorists might

attack any venue where a large number of Americans gather.”

208

201. State v. Seglen, 700 N.W.2d 702 (N.D. 2005).

202. Id. at 707.

203. Id. at 708.

204. Johnston v. Tampa Sports Auth., 442 F. Supp. 2d 1257 (M.D. Fla. 2006), rev’d per

curiam, 490 F.3d 820 (11th Cir. 2007), vacated and superseded on reh’g, 530 F.3d 1320 (11th Cir.

2008) (per curiam).

205. Id. at 1265–66. The judge in Johnston was worried about the slippery slope that would

be created if no such principled distinction were drawn. Echoing Bourgeois, the court noted that

allowing special needs to justify any antiterrorism suspicionless search would lead to searches at

“virtually all professional sporting events, high school graduations, indoor and outdoor concerts,

and parades.” Id. at 1269.

206. Id. at 1266.

207. Id.

208. Id. at 1268.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 881

4. Public Transportation. The final skirmish in the second wave

of suspicionless antiterrorism searches involved public transportation.

The first case, American-Arab Anti-Discrimination Committee v.

Massachusetts Bay Transportation Authority,

209

arose from searches

that took place on the buses and subway trains of Boston in July of

2004. As part of enhanced security during the Democratic National

Convention, the Massachusetts Bay Transportation Authority

(MBTA) conducted suspicionless searches of all individuals riding

buses or trains that passed near the convention site.

210

When some bus

riders sought an injunction against the searches, the MBTA

responded with the now-familiar argument: these were merely

“administrative” searches “similar to the security inspections of

personal belongings conducted at airports and the entryways to

certain kinds of property, such as courthouses.”

211

But this time the

argument won the day. Pointing to the then-recent subway bombings

in Madrid and Moscow, the trial judge noted that “it is not without

foundation to worry that a terrorist event might be aimed

simultaneously at the convention and the transit system.”

212

The judge

was not bothered by the lack of a specific threat against the

Democratic Convention, noting that “there is also no reason to

believe that specific information is necessarily, or even frequently,

available before a terrorist attack,”

213

and that airport searches take

place in the “absence of specific threat information about a particular

flight or even a particular airport.”

214

The court also noted that the

visual inspection of bags, though intrusive, is “very similar to the

intrusions imposed under other, increasingly common, administrative

security search regimes.”

215

During the same month, approximately 230 miles to the

northwest, the Lake Champlain Transportation Company (LCT)

began conducting suspicionless searches of passengers on the ferry

209. Am.-Arab Anti-Discrimination Comm. v. Mass. Bay Transp. Auth., No. 04-11652-

GAO, 2004 WL 1682859 (D. Mass. July 28, 2004).

210. Id. at *1. The searches consisted of a “visual search of the hand-carried items of all

passengers.” Id.

211. Id.

212. Id. at *2.

213. Id.

214. Id.

215. Id. at *4.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

882 DUKE LAW JOURNAL [Vol. 59:843

between Grand Isle, Vermont and Plattsburgh, New York.

216

Cassidy v. Chertoff,

217

two of the passengers on the ferry claimed the

searches violated their Fourth Amendment rights.

218

Once again, the

defendants responded by claiming that the searches served a special

need beyond those of ordinary law enforcement,

219

and once again,

this argument prevailed.

220

Even though there was no record of ferries

being targeted in the United States or anywhere else in the world, the

Second Circuit held that “the airline cases make it clear that the

government, in its attempt to counteract the threat of terrorism, need

not show that every airport or every ferry terminal is threatened by

terrorism;”

221

the search is reasonable as long as the government has

determined that the ferries “are at a high risk of being involved in a

transportation security incident.”

222

One year later, the N.Y.P.D. began a program of suspicionless

searches at various subway stations. Unlike the Boston subway

searches in American-Arab, which targeted every passenger on a

certain train or bus line, the New York searches were based on

seemingly random checkpoints set up at different stations at different

times,

223

and officers at the checkpoints only searched one out of

every series of passengers.

224

The Second Circuit reviewed the

constitutionality of these searches in MacWade v. Kelly,

225

and the

government again invoked the special needs doctrine, leading the

court to determine whether the search “serve[d] as [its] immediate

purpose an objective distinct from the ordinary evidence gathering

associated with crime investigation.”

226

As in the other two public

216. Cassidy v. Chertoff, 471 F.3d 67, 72 (2d Cir. 2006). These searches consisted of visual

inspection of carry-on bags and of the trunks and interior of automobiles on the ferry. Id. at 73.

217. Cassidy v. Chertoff, 471 F.3d 67 (2d Cir. 2006).

218. Id. at 73.

219. Id. at 74–75.

220. Id. at 86–87.

221. Id. at 83.

222. Id. at 83–84 (quoting Implementation of National Maritime Security Initiatives, 68 Fed.

Reg. 39,246 (July 1, 2003) (codified at 33 C.F.R. pt. 101)).

223. MacWade v. Kelly, 460 F.3d 260, 264 (2d Cir. 2006). Although the times and locations

of the checkpoints are meant to “appear random, undefined, and unpredictable,” they are

actually based on a “sophisticated host of criteria, such as fluctuations in passenger volume and

threat level, overlapping coverage provided by [the N.Y.P.D.’s] other counter-terrorism

initiatives, and available manpower.” Id.

224. Id. at 265.

225. MacWade v. Kelly, 460 F.3d 260 (2d Cir. 2006).

226. Id. at 268 (quoting Nicholas v. Goord, 430 F.3d 652, 663 (2d Cir. 2005)).

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 883

transportation cases, the court held that the searches did indeed serve

a special need, in this case “preventing a terrorist attack on the

subway.”

227

As in all the previous cases, there had been no specific,

concrete threat to the New York subway system, but the Second

Circuit still found the threat to be “sufficiently immediate.”

228

Unlike

the Eleventh Circuit in Bourgeois, which refused to allow searches of

protestors that were merely based on a generalized “threat of

terrorism,”

229

the Second Circuit held that “[a]ll that is required is that

the ‘risk to public safety [be] substantial and real’ instead of merely

‘symbolic.’”

230

5. The Current Status of Antiterrorism Searches. The first wave of

terrorism left us with routine suspicionless antiterrorism searches at

all airports and most public buildings. These searches have been

unanimously upheld by courts under the administrative search

doctrine, based on the theory that the searches further an

administrative purpose and were not primarily designed to detect or

investigate criminal activity. And perhaps just as significantly, these

searches have now become accepted by nearly every citizen as a

necessary part of modern life.

The second wave of terrorism has left behind the possibility of

suspicionless antiterrorism searches at subways and ferries, though

such searches are by no means widespread. As of now, these searches

have generated very little political opposition. Meanwhile, courts

have rejected suspicionless searches at sporting events, protest rallies,

and other potential terrorist targets.

231

227. Id. at 263.

228. Id. at 272.

229. Bourgeois v. Peters, 387 F.3d 1303, 1311 (11th Cir. 2004).

230. MacWade, 460 F.3d at 272 (quoting Chandler v. Miller, 520 U.S. 305, 322–23 (1997)).

231. Throughout this debate, the Supreme Court has never directly reviewed a single

antiterrorism suspicionless search case. As noted above, it has implicitly approved of the “first

wave” of suspicionless searches in dicta—as well as the application of the special needs doctrine

to those searches—in three separate cases. See City of Indianapolis v. Edmond, 531 U.S. 32, 47–

48 (2000) (“Our holding . . . does not affect the validity of border searches or searches at places

like airports and government buildings, where the need for such measures to ensure public

safety can be particularly acute.”); Chandler, 520 U.S. at 323 (“[W]here the risk to public safety

is substantial and real, blanket suspicionless searches calibrated to the risk may rank as

‘reasonable’—for example, searches now routine at airports and at entrances to courts and other

official buildings.”); Nat’l Treasury Employees Union v. Von Raab, 489 U.S. 656, 675 n.3 (1989)

(“[W]e would not suppose that . . . the government would be precluded from conducting

[searches at airports] absent a demonstration of danger as to any particular airport or airline.”).

And although these cases were all decided before September 11, 2001, the broad language used

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

884 DUKE LAW JOURNAL [Vol. 59:843

Taking a cue from the earlier airport search cases, courts analyze

this new wave of cases under the special needs doctrine, but this

second wave of suspicionless antiterrorism searches—and the judicial

response to it—both continue to evolve. Law enforcement agencies

across the country continue to experiment with where to conduct

these searches and how extensive they should be, and courts still

struggle to determine the point at which these suspicionless searches

become unconstitutional—if indeed such a point exists. At one end of

the spectrum, the Eleventh Circuit held in Bourgeois that

antiterrorism searches do not serve a special need, whereas the

Second Circuit in Cassidy and MacWade reached the opposite

conclusion. And many of the cases that do accept the Cassidy

rationale—such as Seglen and Johnston—require the threat to reach

some undefined threshold of specificity before the searches can be

justified.

In short, the special needs doctrine does not provide much in the

way of principled guidance to courts struggling to evaluate

antiterrorism searches. To make matters worse, outside the

antiterrorism context, the Supreme Court has been engaged in a

similar struggle to find a principled distinction between searches for

special needs and searches for a law enforcement purpose.

B. Special Needs in Other Contexts

Two cases in particular show the limits that the Court has tried to

create for suspicionless searches over the past decade: Indianapolis v.

Edmond,

232

a vehicular checkpoint case, and Ferguson v. City of

Charleston,

233

a suspicionless drug testing case.

In City of Indianapolis v. Edmond, the Court considered a

suspicionless vehicular checkpoint meant to detect cars carrying

drugs.

234

The city of Indianapolis argued that its checkpoint was

legally identical to the immigration checkpoint in Martinez-Fuerte and

points toward an approval of the second wave of antiterrorism searches: for example, the Von

Raab Court stated that even if there were no demonstration of risk or danger, “[i]t is sufficient

that the Government have a compelling interest in preventing an otherwise pervasive societal

problem from spreading to the particular context.” Von Raab, 489 U.S. at 675 n.3. The Second

Circuit relied upon this broad language in upholding ferry searches in Cassidy v. Chertoff, 471

F.3d 67, 83 (2d Cir. 2006), and subway searches in MacWade, 460 F.3d at 272.

232. City of Indianapolis v. Edmond, 531 U.S. 32 (2000).

233. Ferguson v. City of Charleston, 532 U.S. 67 (2001).

234. Edmond, 531 U.S. at 34.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 885

the sobriety checkpoint in Sitz because those checkpoints “had the

same ultimate purpose of arresting those suspected of committing

crimes.”

235

But the Court disagreed, stating that the checkpoint in

Martinez-Fuente was designed to maintain “the integrity of the

border”

236

and the checkpoint in Sitz was designed to “reduc[e] the

immediate hazard posed by the presence of drunk drivers on the

highways.”

237

In contrast, the primary purpose of the checkpoint in

Edmond was “to advance ‘the general interest in crime control.’”

238

The Court noted that it had “never approved a checkpoint program

whose primary purpose was to detect evidence of ordinary criminal

wrongdoing.”

239

The Edmond Court realized that the distinction it was making

was difficult to draw, conceding that “[s]ecuring the border and

apprehending drunk drivers are, of course, law enforcement activities,

and law enforcement officers employ arrests and criminal

prosecutions in pursuit of these goals.”

240

The Court, however, argued

that these valid checkpoints were closely related to the distinct public

interests of border security and ensuring safe roadways, whereas the

public interest in preventing illegal drug distribution was nothing

more than general crime control.

241

In Ferguson v. City of Charleston, a public hospital in Charleston,

South Carolina began mandatory drug testing of all pregnant mothers

who sought treatment.

242

The government argued that these

suspicionless searches served the “special need” of protecting the

health of the mother and the unborn child,

243

citing Skinner, Von

235. Id. at 42.

236. Id. at 38 (quoting United States v. Montoya de Hernandez, 473 U.S. 531, 538 (1985)).

237. Id. at 39.

238. Id. at 44 (quoting Delaware v. Prouse, 440 U.S. 648, 659 n.18 (1979)).

239. Id. at 41.

240. Id. at 42. The dissent also pointed out that the distinction made by the majority was

“not at all obvious.” Id. at 50 n.2 (Rehnquist, C.J., dissenting).

241. Id. at 42–44. The definition of “general crime control” was further refined in Illinois v.

Lidster, 540 U.S. 419 (2004), in which the Court upheld the use of a suspicionless vehicular

checkpoint to ask drivers for information about a recent hit and run that had occurred at the

same location. Id. at 427–28. Although the Court conceded that the Lidster checkpoint, like the

invalid Edmond checkpoint, was designed to meet the “general interest in crime control,” the

Court held there was a distinction between determining whether the motorist in the car had

committed a crime (as in Edmond) and asking the motorist for information about a crime that

was probably committed by someone else. Id. at 423–24.

242. Ferguson v. City of Charleston, 532 U.S. 67, 70–71 (2001).

243. Id. at 81.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

886 DUKE LAW JOURNAL [Vol. 59:843

Raab, and Acton.

244

But the Court rejected this argument, holding that

the primary purpose of the searches was crime control rather than

protecting the fetus’s health.

245

The Court distinguished Ferguson

from the other drug testing cases because the test results in Ferguson

were turned over to law enforcement and used in criminal

prosecutions.

246

Once again the Court reaffirmed that “[i]n none of

our previous special needs cases have we upheld the collection of

evidence for criminal law enforcement purposes.”

247

As in Edmond, the Court worked hard to draw a principled

distinction between the Ferguson holding and other cases in which the

Court had upheld suspicionless searches even when the incriminating

evidence was used in a criminal prosecution. The Court noted that the

drug tests in Ferguson served the “immediate” purpose of law

enforcement, whereas the broader social goal of protecting the health

of the mother and fetus was an “ultimate” goal.

248

In the valid special

needs cases, public safety was the “direct and primary” purpose of the

search—and using the results in a subsequent prosecution was

presumably nothing more than a beneficial side effect of the search—

whereas in Ferguson the drug test used law enforcement as a “means

to an end.”

249

“This distinction is critical,” said the Court, “[b]ecause

law enforcement involvement always serves some broader social

purpose or objective.”

250

Without this distinction, “virtually any

nonconsensual suspicionless search could be immunized under the

special needs doctrine by defining the search solely in terms of its

ultimate, rather than immediate, purpose.”

251

244. Id. at 77.

245. See id. at 85–86.

246. Id. (“The stark and unique fact that characterizes this case is that [this program] was

designed to obtain evidence of criminal conduct by the tested patients that would be turned

over to the police and that could be admissible in subsequent criminal prosecutions.”).

247. Id. at 83 n.20.

248. Id. at 82–83.

249. Id. at 83–84.

250. Id. at 84.

251. Id. Most recently, the Court’s newfound reluctance to allow suspicionless searches

could be implied from Arizona v. Gant, 129 S. Ct. 1710 (2009). Although Gant is not a

suspicionless search case because it covers searches of a suspect’s car after he or she has been

arrested, it includes language that supports the need for probable cause or a warrant. See, e.g.,

id. at 1721 (“Construing Belton broadly to allow vehicle searches incident to any arrest would

serve no purpose except to provide a police entitlement, and it is anathema to the Fourth

Amendment to permit a warrantless search on that basis. For these reasons, we are

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 887

IV.

CAN SUSPICIONLESS ANTITERRORISM SEARCHES BE JUSTIFIED

BY THE SPECIAL NEEDS DOCTRINE?

Suspicionless searches can be generally grouped into three

different categories: true administrative searches, which do not

uncover evidence of criminal activity; searches that uncover evidence

of criminal activity, but whose fruits cannot be used in a subsequent

criminal prosecution; and searches whose fruits are in fact used in a

subsequent criminal prosecution. As this Article demonstrates, the

special needs doctrine cannot justify this third category of searches,

and it is in this category that antiterrorism searches belong.

A. Categorizing Permissible Suspicionless Searches

The law surrounding permissible suspicionless searches is in a

state of disarray,

252

but it is possible to distill some general principles

from the various cases. First, there is only one context in which the

Supreme Court has openly admitted that a suspicionless search is

permissible for criminal law purposes: searches of probationers and

parolees.

253

In this category of cases, the Court has held that the

defendant has such a diminished expectation of privacy

254

that the

state has the ability to conduct a suspicionless search and use the

results to criminally incriminate the defendant.

255

For every other type of suspicionless search, the Court has

claimed that the search is invalid unless it serves a purpose other than

traditional law enforcement. The Court uses different language in

different contexts—the search must not be “aimed at the discovery of

unpersuaded by the State’s arguments that a broad reading of Belton would meaningfully

further law enforcement interests and justify a substantial intrusion on individuals’ privacy.”).

252. See generally Edwin J. Butterfoss, A Suspicionless Search and Seizure Quagmire: The

Supreme Court Revives the Pretext Doctrine and Creates Another Fine Fourth Amendment Mess,

40 C

REIGHTON L. REV. 419 (2007) (“[T]he ‘category’ of suspicionless searches [has become] a

jurisprudential mess, with the only consistent theme being that suspicionless search schemes

[have been] regularly upheld as lawful.”).

253. See United States v. Knights, 534 U.S. 112, 116 (2001).

254. The Supreme Court has held that government surveillance does not constitute a

“search” within the meaning of the Fourth Amendment unless the surveillance infringes upon

an individual’s “reasonable expectation of privacy.” Katz v. United States, 389 U.S. 347, 360

(1967) (Harlan, J., concurring). Thus, if an individual does not have a reasonable expectation of

privacy in a certain context—for example, if he is conducting an activity in a public place, or if

he is in prison—then the government can watch his actions without violating his reasonable

expectation of privacy and thus not violate the Fourth Amendment.

255. Knights, 534 U.S. at 119.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

888 DUKE LAW JOURNAL [Vol. 59:843

evidence of crime;”

256

it must serve “special needs[] beyond the

normal need for law enforcement;”

257

or it must fulfill a purpose other

than “general crime control”

258

—but the requirement itself remains

constant. This requirement—assuming it is actually followed in

practice—implies that the Fourth Amendment’s primary purpose is to

prevent the state from invading the rights of citizens during a criminal

investigation, and so there is no reason why it should limit the

government’s actions outside the criminal context.

259

This is a

controversial proposition, but it is an inevitable conclusion when

examining the special needs doctrine. One benefit of this rule—if it

were actually followed—is that it would result in a bright-line rule

providing clear guidance to law enforcement and protecting

individuals from random, suspicionless searches during criminal

investigations.

But as this review of the case law has demonstrated, the non–law

enforcement purpose requirement has not been followed in practice.

In many of these suspicionless searches, the government conducts

searches that—if it detects what it is designed to detect—will produce

evidence of criminal activity: possession of firearms or explosives,

possession of stolen vehicles, illegal presence in the country,

possession or use of illegal substances, or driving while intoxicated.

And to state that the purpose of the search is distinct from the

“normal need for law enforcement” is merely a semantic game

because the special need that justifies the weaker search standard is

usually nothing more than the policy justification for the original

criminalization of the conduct. Hijacking is illegal because of the

dangers it poses to airline passengers; immigration laws exist to

secure the country’s borders; drugs are illegal because those who use

them suffer health effects and an impairment of their abilities; drunk

driving is prohibited because those who drink and drive create a

danger on the public roadways.

In this sense, the suspicionless search jurisprudence is little more

than an exercise in redefining the nature of criminal activity and

256. Camara v. Mun. Court, 387 U.S. 523, 537 (1967).

257. New Jersey v. T.L.O., 469 U.S. 325, 351 (1985) (Blackmun, J., concurring).

258. City of Indianapolis v. Edmond, 531 U.S. 32, 43 (2000).

259. See, e.g., United States v. Montoya de Hernandez, 473 U.S. 531, 563 (1985) (Brennan,

J., dissenting) (“[The Fourth Amendment safeguards] should govern border searches when

carried out for purposes of criminal investigation.”).

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 889

thereby redefining the permissible methods of detecting that activity.

A police officer could enforce the law against drunk driving, a law

that exists to keep the highways safe, or she could set up a checkpoint

to detect drunk drivers to keep the highways safe, and in so doing

detect individuals who are breaking the law. In the former case, she

would need to have probable cause before pulling over a car, whereas

in the latter case, she would not need any level of individualized

suspicion because keeping highways safe is a special need.

To be fair, this objection does not fit all of the permissible

suspicionless search cases equally. For some of them, the lack of a law

enforcement purpose is more than a semantic difference; it is an

important distinction that alters the nature and extent of the search in

significant ways. To understand this point, it is helpful to divide the

special needs cases into three different categories.

The first category consists of the original administrative searches,

in which non–law enforcement officials conduct searches to enforce

health and safety violations. These include building code inspections,

Occupational Safety and Health Administration workplace

inspections, mine safety inspections, and the like. Such searches do

not seek to enforce criminal laws, and so the searches represent a

“less hostile intrusion than the typical policeman’s search for the

fruits and instrumentalities of crime.”

260

This category does not

include all administrative search cases because courts have applied

the “administrative search” term to a number of cases in which the

search has a clear crime-control purpose. For example, the search in

New York v. Burger—in which the police searched a junkyard for

evidence of stolen auto parts and ultimately arrested the junkyard

owner on criminal charges—was termed an administrative search

even though it was designed to uncover evidence of criminal

activity.

261

The second category is comprised of searches that are designed

to uncover evidence of unambiguously criminal activity, but that can

legitimately be classified as special needs searches because the results

are not used for a law enforcement purpose. In these cases, the search

serves a purpose “other than that of traditional law enforcement,” not

because of the type of information being sought, but rather because

260. Camara, 387 U.S. at 530.

261. See New York v. Burger, 482 U.S. 691, 708–10 (1987). Burger belongs in the third

category of suspicionless search cases because the search was upheld based on the defendant’s

reduced expectation of privacy. Id. at 703–07.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

890 DUKE LAW JOURNAL [Vol. 59:843

of what is done with the information after it is obtained. This category

includes the drug tests of school children and public employees, which

are designed to find evidence of criminal activity, but whose results

are not used in a subsequent criminal prosecution.

The third category involves searches that have a clear law

enforcement purpose, and whose results are used by law enforcement

in criminal prosecutions. This category includes antiterrorism

searches, vehicular checkpoints, and some of the later administrative

searches in cases such as Burger. For some of these searches, the

Supreme Court has stressed that the nature of the intrusion is very

slight and that the subject has a reduced expectation of privacy—

briefly stopping a vehicle or checking records and inventory for a

highly regulated business—though these factors are not present for

antiterrorism searches. But assuming individuals in these situations

have some expectation of privacy (unlike probationers and parolees),

courts must still engage in the semantic game of redefining a crime-

control search as a special needs search.

B. Placing Antiterrorism Searches into the Proper Category

Once permissible suspicionless search cases are organized in this

manner, it becomes easy to see that suspicionless antiterrorism cases

fall into the third category of special needs cases. These searches are

conducted by law enforcement personnel—police, TSA officers, court

officers—and any incriminating search results may be used against

the suspect in a criminal trial. And what is more, these searches are

even less defensible than the other special needs cases in that

category because the other cases involve situations in which the

suspect has a reduced expectation of privacy. In contrast, courts have

held that subjects of antiterrorism searches—unlike parolees or

occupants of cars—“possess[] an undiminished expectation of

privacy” for items carried on their bodies or in their bags.

262

Furthermore, the ultrafine distinction that courts make between

special needs and law enforcement needs is especially strained in the

antiterrorism context. As the Eleventh Circuit pointed out in

Bourgeois, the alleged special need of “public safety” is “inextricably

intertwined” with the goal of crime control in the antiterrorism

262. See, e.g., MacWade v. Kelly, 460 F.3d 260, 273 (2d Cir. 2006).

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 891

context.

263

Antiterrorism searches are designed to prevent individuals

from intentionally committing acts of violence, and to apprehend

those who attempt to commit these acts; to claim that this is a public

safety purpose rather than a law-enforcement or crime-control

purpose is simply disingenuous.

264

As noted above,

265

the early cases that justified antiterrorism

searches as administrative searches were reacting to an extraordinary

situation in which there appeared to be no other way to prevent an

epidemic of hijackings and bombings of public buildings. The only

doctrine available at the time to justify suspicionless searches was the

administrative search rationale from Camara. These early courts

therefore borrowed the terms “administrative” and “regulatory” from

Camara and used them in a way that had nothing to do with the

terms’ actual meaning. The Ninth Circuit, for example, noted that

searches at airports were “conducted as part of a general regulatory

scheme in furtherance of an administrative purpose, namely, to

prevent the carrying of weapons or explosives aboard aircraft, and

thereby to prevent hijackings.”

266

To say that blanket, suspicionless

searches of everyone who boards a plane is a “regulatory scheme” is

simply an exercise in making a term mean whatever one wants it to

mean; likewise, the prevention of violent crimes is an “administrative

263. Bourgeois v. Peters, 387 F.3d 1303, 1312–13 (11th Cir. 2004); see also supra notes 180–

85 and accompanying text.

264. There is, perhaps, a distinction between trying to prevent crime and trying to detect

crime. If a government activity is only meant to prevent crime, the individualized suspicion

requirements of the Fourth Amendment should not apply. For example, installing a window

made of bulletproof glass at all post office counters is a method of preventing crimes—there is

surely no need to demonstrate individualized suspicion for every customer who approaches the

window. It could be argued that suspicionless antiterrorism searches should be analogized to

these bulletproof glass windows—if a government agent searched everyone who entered the

post office, for example, the search would only be intended to prevent crimes, and so there

would be no need to demonstrate any individualized suspicion. The problem with making this

distinction is that in practice the suspicionless antiterrorist searches are used both to prevent

and detect crime—and ultimately to punish the criminal as well. If the search recovers a

weapon, the subject of the search is arrested and the weapon is used as evidence in a subsequent

prosecution. The only way to ensure that the suspicionless searches are truly only being used to

prevent a crime would be to preclude the government from using any fruits of the search in a

subsequent prosecution. See infra Part VI.B.

265. See supra notes 18–28 and accompanying text.

266. United States v. Davis, 482 F.2d 893, 908 (9th Cir. 1973), abrogated by United States v.

Aukai, 497 F.3d 955 (9th Cir. 2007) (en banc).

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

892 DUKE LAW JOURNAL [Vol. 59:843

purpose” only if one invents a completely new definition for the term

“administrative.”

267

This contrivance has been encouraged by the confused way in

which the special needs doctrine has evolved over the past few

decades, with different justifications becoming significant in different

contexts. If the same doctrine is applied to widely different factual

situations (from conducting building code inspections to drug testing

schoolchildren to detecting terrorists on the subway), it creates

confusion when one court interprets the language of another court’s

precedent. Consider Von Raab, the case in which the Supreme Court

upheld suspicionless drug testing of United States customs

employees. The search in this case falls into the second category of

permissible suspicionless searches—because the results of the test are

not used for criminal prosecution, the Court can legitimately claim

that the search serves a special need unrelated to law enforcement.

268

But the court also attempts to place this search into the first

category—claiming that it serves a purely “administrative” or

“regulatory” purpose by analogizing the use of cocaine by customs

employees to the inventory searches: “[T]he traditional probable-

cause standard may be unhelpful in analyzing the reasonableness of

routine administrative functions, especially when the Government

seeks to prevent the development of hazardous conditions or to

detect violations that rarely generate articulable grounds for

searching any particular place or person.”

269

In making this argument,

the Court cites a number of legitimate administrative search cases,

including an inventory search case,

270

a border search case,

271

and a

building code inspection case.

272

It is already a bit of a stretch to say that cocaine use by customs

officials is a “hazardous condition” equivalent to building code

267. Once this terminology had become enshrined in precedent, it was perpetuated by later

generations of cases. The Ninth Circuit recently upheld an airport search as an administrative

search, quoting this same language from the Davis case. United States v. Aukai, 497 F.3d 955,

960 (9th Cir. 2007) (en banc).

268. Nat’l Treasury Employees Union v. Von Raab, 489 U.S. 656, 666 (1989). The Court

also notes that customs employees have a diminished expectation of privacy, id. at 672, but this

was not the primary reason that the searches were found constitutional.

269. Id. at 668 (emphasis omitted) (citation omitted).

270. Colorado v. Bertine, 479 U.S. 367, 369–71 (1987).

271. United States v. Martinez-Fuerte, 428 U.S. 543, 556–57 (1976).

272. Camara v. Mun. Court, 387 U.S. 523, 534–36 (1967).

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 893

violations, and it is difficult to equate suspicionless drug testing with

the “routine administrative function” of searching an automobile

before it is impounded. But the Court persisted in its analogy, stating

that drug use by customs agents was a “latent or hidden condition[],”

and “discover[ing] . . . or . . . prevent[ing] the[] development” of such

conditions justified a suspicionless search.

273

The lower courts that applied the special needs doctrine to

suspicionless antiterrorism searches merely carried this analogy one

step further. For example, this is how the Second Circuit in MacWade

upheld suspicionless searches on the subway:

As a legal matter, courts traditionally have considered special the

government’s need to “prevent” and “discover . . . latent or hidden”

hazards, Von Raab, 489 U.S. at 668, in order to ensure the safety of

mass transportation mediums, such as trains, airplanes, and

highways. We have no doubt that concealed explosives are a hidden

hazard . . . . Accordingly, preventing a terrorist from bombing the

subways constitutes a special need that is distinct from ordinary post

hoc criminal investigation. Further, the fact that an officer

incidentally may discover a different kind of contraband and arrest

its possessor does not alter the [p]rogram’s intended purpose.

274

A terrorist attempting to carry a bomb onto a subway (or an

airplane) is not a “latent or hidden hazard.” He is an individual

attempting to commit an extremely serious crime. Any attempt to

deter or detect his actions is purely a law enforcement function, and

should be treated as such. And any attempt by courts to claim that a

search designed to detect and prevent this crime is an administrative

or regulatory search cannot ultimately be sustained.

Thus, even the contorted jurisprudence of the special needs

doctrine cannot justify these searches. But this conclusion leads to

another question: if suspicionless antiterrorist searches cannot be

justified by the special needs doctrine, is there any other way to

justify them? We will address this question in Part V.

273. Von Raab, 489 U.S. at 668.

274. MacWade v. Kelly, 460 F.3d 260, 270–71 (2d Cir. 2006) (first alteration in original)

(citations omitted).

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

894 DUKE LAW JOURNAL [Vol. 59:843

OTHER POSSIBLE JUSTIFICATIONS FOR SUSPICIONLESS

NTITERRORIST SEARCHES

At first, the search for a legitimate justification for suspicionless

antiterrorist searches may appear to be a waste of time, or at best a

purely academic pursuit. In reality, antiterrorism searches in airports

and courthouses are ubiquitous, widely accepted, and seemingly

indispensable. But it is important to find a legitimate doctrinal

underpinning for these searches for two reasons. First, courts’

responses to the second phase of antiterrorism searches have been

wildly inconsistent, both in terms of reasoning and results. And

second, continued application of the special needs doctrine to these

searches will create confusing precedent that could be applied to

other areas of the law, leading to even greater damage to the

individualized suspicion requirement. Thus, it is worth seeking

another justification for suspicionless antiterrorism searches—or, if

one cannot be found, to find another way to reconcile these searches

with the Fourth Amendment.

275

Other commentators have suggested

numerous justifications for suspicionless antiterrorism searches—

using a generalized “reasonableness” test, deferring to the legislature,

applying the doctrine of implied consent, and creating an entirely new

exception to the Fourth Amendment. But as this Article

demonstrates, none of these proposals can serve to justify these

searches.

A. Evaluation Under a Generalized Reasonableness Test

Even if antiterrorism searches cannot be justified as a special

need, most courts and commentators (and indeed lay people) would

still find them constitutional under the general test of reasonableness.

As the Supreme Court has repeated dozens of times, reasonableness

is the “touchstone of the constitutionality of a governmental

search,”

276

and whether or not a search is reasonable “depends on all

of the circumstances surrounding the search or seizure.”

277

When the

circuit courts first began approving the suspicionless antiterrorist

searches in the early 1970s, many of them relied on a generalized

275. For a discussion of how to reconcile these searches with the Fourth Amendment, see

infra Part VI.

276. See, e.g., Bd. of Educ. v. Earls, 536 U.S. 822, 828 (2002).

277. Skinner v. Ry. Labor Executives’ Ass’n, 489 U.S. 602, 619 (1989) (quoting United

States v. Montoya de Hernandez, 473 U.S. 531, 537 (1985)).

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 895

reasonableness standard, “balancing the need for a search against the

offensiveness of the intrusion.”

278

Indeed, all of the current

suspicionless antiterrorism cases—after applying the special needs

test to bypass the general requirement of a warrant and probable

cause—essentially apply a reasonableness analysis. The Second

Circuit defines the suspicionless antiterrorism test as follows:

First, as a threshold matter, the search must “serve as [its]

immediate purpose an objective distinct from the ordinary evidence

gathering associated with crime investigation.” Second, once the

government satisfies that threshold requirement, the court

determines whether the search is reasonable by balancing several

competing considerations. These balancing factors include (1) the

weight and immediacy of the government interest; (2) “the nature of

the privacy interest allegedly compromised by” the search; (3) “the

character of the intrusion imposed by” the search; and (4) the

efficacy of the search in advancing the government interest.

279

So why not abolish the somewhat artificial threshold test of

special needs and simply skip to the second step and conduct a

reasonableness analysis? The extremely violent and destructive

nature of terrorist acts surely affects the appropriateness of a search

meant to detect or deter such acts, and this factor would be reflected

in a general reasonableness test. For example, if police have twenty-

four hours to find an atomic bomb that they know is hidden in one of

a hundred houses, it would be reasonable for them to conduct

suspicionless searches of each of those houses.

280

If police pull over

every car that drives by a reservoir because the dam is a “potential

terrorist target,” the search would likely be unreasonable.

But these are the extreme cases—and therefore the easy cases.

As our review of the second wave of antiterrorism case law shows, it

is indeed the hard cases that make bad, confusing, and inconsistent

law. Thus, finding a reasonable, principled justification for these

searches is critical—and it is more important now than ever. If there

is no principled way to evaluate antiterrorism searches, then there is

no way of knowing the limits of these searches—if they have limits at

all.

281

Under the current law, the only thing resembling a principled

limitation to the searches is the factor explicitly stated in Johnston

278. See, e.g., United States v. Edwards, 498 F.2d 496, 500 (2d Cir. 1974).

279. MacWade, 460 F.3d at 268–69 (citations omitted).

280. For a discussion of this hypothetical, see Gould & Stern, supra note 3, at 779–80.

281. See Bourgeois v. Peters, 387 F.3d 1303, 1311–12 (11th Cir. 2004).

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

896 DUKE LAW JOURNAL [Vol. 59:843

and Stauber (and hinted at in Seglen and Bourgeois): that

suspicionless antiterrorism searches are unconstitutional unless there

is a “concrete danger” that is “substantial and real.”

282

In rejecting the

suspicionless searches of individuals at sporting arenas, Johnston and

Seglen noted that there had been no history of terrorist attacks or

threats against sporting arenas

283

—in contrast, Johnston argues that

searches of public buildings were allowed after “an outburst of acts of

violence, bombings of federal buildings and hundreds of bomb

threats, resulting in massive evacuations of federal property and

direct financial loss to the Government,”

284

and searches in the New

York City subway were allowed because “NYC’s mass transit system,

including the subway system, had been targeted by terrorists in the

past, and within the last year terrorists had bombed commuter trains

in Madrid, the subway system in Moscow and had attempted to bomb

the London subway system.”

285

Thus, under the current reasonableness analysis, the only

limiting factor for suspicionless antiterrorist searches is whether a

threat is “substantial and real” or merely the background threat of

terrorism that presumably has become part of everyday existence.

286

But this is rather thin protection; there is little doubt that if a terrorist

attack occurred at a sporting arena or a political protest, the holdings

in Johnston and Seglen or Stauber and Bourgeois would be overruled

by the next court to consider searches in these contexts. In this sense,

it is the terrorists who control the extent of our Fourth Amendment

rights: by hijacking planes and blowing up buildings, they persuade

courts to allow suspicionless searches at airports and public buildings;

282. Johnston v. Tampa Sports Auth., 442 F. Supp. 2d 1257, 1266 (M.D. Fla. 2006), rev’d per

curiam, 490 F.3d 820 (11th Cir. 2007), vacated and superseded on reh’g, 530 F.3d 1320 (11th Cir.

2008) (per curiam); Stauber v. City of New York, Nos. 03 Civ. 9162, 03 Civ. 9163, 03 Civ. 9164,

2004 U.S. Dist. LEXIS 13350, at *93 (S.D.N.Y. July 19, 2004); see also Bourgeois, 387 F.3d at

1311; State v. Seglen, 700 N.W.2d 702, 707–08 (N.D. 2005).

283. Johnston, 442 F. Supp. 2d at 1266–69; Seglen, 700 N.W.2d at 708.

284. Johnston, 442 F. Supp. 2d at 1269 (quoting Downing v. Kunzig, 454 F.2d 1230, 1231–32

(6th Cir. 1972)).

285. Id. (citing MacWade v. Kelly, 460 F.3d 260, 272 (2d Cir. 2006)).

286. This rule means that a judge must somehow determine at what point the ever-present

“background threat” of terrorism becomes a “substantial and real” threat. If a terrorist group

issues a statement that it will attack a sports arena sometime in the next year, does that make

the threat to all sports arenas substantial and real? What if it named a city but no date? Would it

matter if the terrorist group had been known to have successfully committed terrorist acts in the

past? In short, how could a judge intelligently determine the probability of such a threat, and

what is the probability threshold for “substantial and real”?

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 897

by bombing mass transit in London, Madrid, and Moscow, they

convince courts to allow suspicionless searches on ferries and

subways; if they successfully target a sports arena or a political protest

rally, courts will (under current doctrine) begin to allow suspicionless

searches in those contexts as well. And as this Article’s discussion of

the first wave of suspicionless antiterrorist searches demonstrates, the

search procedures (and their acceptance by the courts and by the

general population) does not end when the threat subsides.

Throughout the 1990s there was not a single domestic passenger

airplane hijacking, and the epidemic of shootings and bombings of

courthouses disappeared—yet the suspicionless searches continued,

and probably will continue forever. As noted above, the Supreme

Court cited the infinitesimally low number of airplane hijackers

caught through the suspicionless search program as evidence that the

programs were “successful.”

287

Yet when this “successful”

suspicionless search program failed spectacularly—resulting in the

death of 3,000 Americans on a single day—the government reaction

was to make the suspicionless searches even more rigorous and

intrusive, and the reaction of the courts was to reaffirm (in even

stronger language) the necessity and constitutionality of those

searches.

288

More generally, antiterrorist searches are particularly ill suited to

a generalized balancing test or a reasonableness analysis, for the

simple reason that the gravity of the potential harm is so great that it

overpowers any other variable that could be placed into the balancing

equation. Every first-year criminal law student is presented with some

variant of this classic hypothetical, intended to highlight the

difference between utilitarianism and retributivism: would the police

be justified in torturing an innocent person if it could provide

information that would prevent a nuclear bomb from detonating and

destroying a city? The threat of terrorism turns the second half of this

hypothetical into grim reality: wouldn’t it be reasonable to allow

nearly any type of surveillance or search program if it could prevent

287. See supra note 58 and accompanying text. In Von Raab, the Court noted that in fifteen

years, 9.5 billion people and ten billion pieces of luggage had undergone a suspicionless search,

and only forty-two thousand firearms had been detected. Nat’l Treasury Employees Union v.

Von Raab, 489 U.S. 656, 675 n.3 (1989). Thus, every suspicionless search that occurred had a

.00042 percent chance of detecting a firearm.

288. See, e.g., United States v. Yang, 286 F.3d 940, 944 n.1 (7th Cir. 2002) (“[T]he events of

September 11, 2001, only emphasize the heightened need to conduct searches at this nation’s

international airports.”).

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

898 DUKE LAW JOURNAL [Vol. 59:843

catastrophic harm?

289

Should a court somehow give less weight to the

government interest in a search seeking to prevent a terrorist attack

that may only kill a few hundred people as opposed to many

thousands of people? Can a court really evaluate the likelihood of a

terrorist attack in any given context, other than by simply accepting

the risk assessment the government provides for them?

As the Eleventh Circuit noted in Bourgeois, “reasonableness” is

a term of art in Fourth Amendment jurisprudence;

290

courts must

interpret the term “in the light of established Fourth Amendment

principles,”

291

including the warrant requirement, prior judicial

scrutiny, probable cause, and individualized suspicion.

292

It is very rare

for a search that lacks any of these requirements to be deemed

“reasonable.” If the term were given its ordinary meaning, the

constitutionality of a search would be determined by “the judge’s

personal opinions about the governmental and privacy interests at

stake.”

293

289. In a recent article, Anthony Coveny described the current balancing test for

suspicionless searches as a mathematical formula, in which PC stands for public concern, E is a

percentage that represents the efficacy of the searches, and LI stands for the level of intrusion of

the search. If PC x E > LI, the search should be allowed, but if PC x E < LI, then the search is

unreasonable and hence unconstitutional. Coveny, supra note 2, at 379–80. When public

concern is preventing terrorism, PC is essentially infinite. Coveny quotes the Second Circuit

statement that “the government interest in preventing a terrorist attack on the subway was ‘of

the very highest order.’” Id. at 379 (quoting MacWade, 460 F.3d at 267). Other courts use similar

language when applying the balancing test. See, e.g., United States v. Hartwell, 436 F.3d 174, 179

(3d Cir. 2006) (“[T]here can be no doubt that preventing terrorist attacks on airplanes is of

paramount importance.”); United States v. Marquez, 410 F.3d 612, 618 (9th Cir. 2005) (“It is

hard to overestimate the need to search air travelers for weapons and explosives . . . .”).

Therefore, “as long as the effectiveness measure is anything but zero, the level of intrusiveness

is near immaterial.”

Coveny, supra note 2, at 379. And if PC is not set to infinity, how should it

be calculated in evaluating the reasonableness of most antiterrorist searches?

290. See generally Orin S. Kerr, Four Models of Fourth Amendment Protection, 60 S

TAN. L.

REV. 503 (2007) (discussing why courts have not adopted a test for reasonableness).

291. Bourgeois v. Peters, 387 F.3d 1303, 1313 (11th Cir. 2004) (quoting Chimel v. California,

395 U.S. 752, 765 (1969)).

292. Id. at 1313–16.

293. See id. at 1314. A number of commentators have also recognized this problem.

Professor Ricardo Bascuas notes that using a balancing test to determine what is “reasonable” is

not even “a legal inquiry or test” because

it is not a process of discerning general rules or principles and applying them

evenhandedly to specific disputes as they arise. Rather, balancing is for judges, as

Justice Scalia put it, “a regrettable concession of defeat—an acknowledgement that

we have passed the point where ‘law,’ properly speaking, has any further

application.” The vagueness of the term “reasonable” makes not only the outcome

but the very criteria of the “test” unpredictable.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 899

Therefore, reasonableness alone provides neither a principled

reason for allowing suspicionless antiterrorist searches, nor a sensible,

robust limitation on their use. A search for a more principled basis to

justify these searches will not only help to legitimate them, but will

also contain them so that they do not swallow up the entire Fourth

Amendment.

B. Legislatures Should Decide What Is Reasonable

Political process theory offers a more sophisticated defense of

the reasonableness analysis by arguing that the democratic process

will provide necessary limits on the government’s behavior. The

argument runs like this: because the Fourth Amendment’s

prohibition on unreasonable searches is indeterminate, the task falls

to either courts or legislatures to interpret the meaning and scope of

the term.

294

According to political process theory, the legislature

should conduct this task because the interpretation of democratically

elected legislators is more legitimate than the opinions of unelected

judges.

295

The courts should thus defer to the legislatures in these

situations, and only conduct a rational basis review of any legislative

interpretations of ambiguous constitutional terms.

Bascuas, supra note 2, at 748 (footnote omitted) (quoting Antonin Scalia, The Rule of Law as a

Law of Rules, 56 U.

CHI. L. REV. 1175, 1181 (1989)). Professor Thomas Clancy, writing in

1995—well before the second wave of antiterrorism searches began—warned against

suspicionless “checkpoints and detectors . . . at stadiums, in schools, and at other public

gatherings,” all justified by a “totally subjective reasonableness analysis.” Thomas K. Clancy,

The Role of Individualized Suspicion in Assessing the Reasonableness of Searches and Seizures,

25 U. M

EM. L. REV. 483, 624–25, 627 (1995) (footnotes omitted). Clancy argues that there must

be a “return to the central importance given to individualized suspicion by the framers,” though

he does not explain how this will apply to airport searches. Id. at 634. Bascuas proposes his own

test for antiterrorism searches, which this Article discusses infra at notes 335–38 and

accompanying text.

Indeed, the vague nature of the term “reasonableness” was evident in the Supreme

Court’s most recent school search case, in which a schoolgirl was strip-searched on suspicion

that she was carrying illegal prescription drugs. Safford Unified Sch. Dist. No. 1 v. Redding, 129

S. Ct. 2633, 2641–43 (2009). Whether it was reasonable for school officials to conduct such an

intrusive search to combat the possession and distribution of drugs in school comes down to

little more than the personal opinion of the judge or judges hearing the case. In Safford United

School District No. 1 v. Redding, 129 S. Ct. 2633 (2009), the trial judge found the search to be

reasonable, a three-judge appellate panel agreed, an en banc panel split sharply on the question,

and the Supreme Court finally ruled that the search was unreasonable. Id. at 2637–39.

294. Political process theory focuses on ambiguous terms in the Constitution, such as due

process, equal protection, or unreasonable searches and seizures. See Richard C. Worf, supra

note 2, at 100–01.

295. Id at 101.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

900 DUKE LAW JOURNAL [Vol. 59:843

Aside from the increased legitimacy from the democratic

process, legislative interpretations of reasonableness are superior to

judicial determinations for two reasons. First, the legislature is in a far

better position than the courts to determine what is reasonable—it

can run a cost-benefit analysis and has better access to the necessary

facts in making such a determination.

296

Second, the legislature is

more flexible. It can vary from state to state or even city to city,

depending on the preferences of the community, and if circumstances

change—if terrorists begin a successful campaign of violence, for

example—it can adapt much more quickly than the precedent-bound

judicial system.

297

Political process theory recognizes that the democratic process

sometimes breaks down, at which point courts must be more

aggressive in reviewing a given search. For example, if a search

procedure had been instituted by the executive branch, either without

legislative approval or pursuant to an overly broad grant of discretion

on the part of the legislature, courts would need to conduct a strict

scrutiny analysis.

298

Similarly, courts should intervene when searches

disadvantage a discrete and insular minority, who may not have

adequate representation in the legislature.

299

In other words, the

Fourth Amendment is only meant to be a check on majoritarian rule

when the political process cannot work effectively.

300

Permissible suspicionless searches are particularly easy to justify

under political process theory—provided they are properly

authorized by the legislature—because they are nondiscriminatory by

their very nature. They are not directed at a specific class of

underprivileged or disenfranchised individuals; they affect everyone

more or less equally. Thus, assuming the benefits and costs of the

search are spread equally throughout the voting community, the

search program cannot be seen as the result of a failure of the

democratic process.

Furthermore, deference to legislators seems even more sensible

in the context of antiterrorism searches. It is hard enough for judges

296. Id. at 120–26.

297. Id. at 129–30 (“If mass terror strikes, it might be disastrous to apply old precedents to

determine what constitutes a reasonable general search or seizure.”).

298. Id. at 138–52.

299. Id. at 152–58.

300. Id. at 189–90.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 901

to make reasoned, informed judgments about the necessity and

efficacy of other types of searches, such as roadblocks to prevent

drunk driving, or frisks to determine if suspects carry a weapon. But

this task becomes even more daunting for antiterrorism searches. The

Second Circuit in MacWade explained why it deferred to the

government’s experts who testified that the suspicionless subway

searches were an effective method of deterring a terrorist attack:

We will not peruse, parse, or extrapolate four months’ worth of

data . . . . Counter-terrorism experts and politically accountable

officials have undertaken the delicate and esoteric task of deciding

how best to marshal their available resources in light of the

conditions prevailing on any given day. We will not—and may not—

second-guess the minutiae of their considered decisions.

301

But MacWade is an outlier in this instance. For the most part,

courts have rejected applying political process theory to the Fourth

Amendment,

302

and with good reason. Given the central role of the

Fourth Amendment in criminal procedure jurisprudence, deferring to

the democratically elected legislature to determine what is reasonable

is deeply troubling. Some commentators, such as Professor Akhil

Amar, have argued persuasively that the Fourth Amendment should

not be thought of primarily as a rule of criminal procedure

303

—but

there is no denying that, rightly or wrongly, this is what the Fourth

Amendment has become. And as a rule of criminal procedure—

whose primary purpose is to protect the individual citizen from

overreaching by the government during criminal prosecutions—the

Fourth Amendment is dangerously unsuitable for political process

analysis.

It is no accident that political process analysis has not been

applied to the rest of the criminal procedure rules in the Bill of

Rights—there is no deference to the legislature to determine when

Miranda rights apply, for example, or when the right to counsel

attaches. Granted, the term “unreasonable” is ambiguous enough to

require interpretation, but giving a majority of voters the right to

301. MacWade v. Kelly, 460 F.3d 260, 274 (2d Cir. 2006).

302. Even the strongest proponents of political process theory concede that the Fourth

Amendment has “remained largely invulnerable to political process theory.” Worf, supra note

2, at 103–04 (noting Akhil Amar’s acknowledgment that political process theory has had little

influence on Fourth Amendment jurisprudence).

303. Akhil Reed Amar, Fourth Amendment First Principles, 107 H

ARV. L. REV. 757, 758–59

(1994).

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

902 DUKE LAW JOURNAL [Vol. 59:843

interpret it—and thereby set the parameters for how police interact

with ordinary citizens—could lead to extreme results, for two reasons.

First, a majority of voters (or their representatives who make the

laws) will frequently overreact in times of crisis, trading freedoms in

exchange for short-term security. The government responses to this

decade’s second wave of terrorism are a case in point: in the wake of

the September 11 attacks, the federal government took a number of

steps to curtail civil liberties to detect and prevent other terrorist

activities, including detaining citizens indefinitely without charge and

without providing counsel, eavesdropping on conversations between

attorneys and clients, and broadening the scope of wiretapping

programs.

304

The courts have been very active in reviewing and in

some cases rejecting these reactions.

305

Political process theory would approve of judicial review of some

of these rules—for example, rules allowing the government to listen

in on conversations between criminal law attorneys and clients only

burden accused criminals, who hardly have a proportionate say in the

legislature. Others—those that apply broadly to all citizens—would

be left untouched by courts if judges agreed to defer to the legislative

enactments. But these rules are not the result of a reflective

legislative body calmly weighing the costs and benefits—indeed, they

are little better than pandering to the political moment.

This short-term pandering might not be so dangerous were it not

for the second problem: once civil liberties are restricted, the

restrictions are essentially permanent. What the majority of voters

think is reasonable will inevitably reflect the rules that they have

grown accustomed to in their everyday lives. The reason that searches

at airports and public courthouses are now thought to be reasonable

by the vast majority of citizens has nothing to do with whether they

are in fact a sensible, least intrusive solution to a severe and ongoing

problem—though they may in fact be. Rather, they are thought to be

reasonable because they have been around for so long that we no

longer even notice them or perceive them to be intrusive. And regular

searches in some contexts also tend to desensitize individuals to

identical searches in other contexts—that is, now that suspicionless

304. See supra note 157 and accompanying text.

305. See supra note 158 and accompanying text.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 903

searches are ubiquitous at airports, people are less likely to find them

unreasonable at schools, rock concerts, sporting events, and so on.

306

Courts, on the other hand, are far more insulated from the

politics of the moment, and far more likely to uphold the principles of

the Fourth Amendment—even in times of crisis, which is when such a

defense is needed the most. This is not to say that the courts’ analyses

and holdings are immune to changes in the world, merely that they

are more likely to apply constant principles of law to the evolving

threats and dangers that arise.

307

Simply put, it is not very comforting to conclude that the Fourth

Amendment—a fundamental protection against government

overreaching—says whatever the government wants it to say. If the

Fourth Amendment is to have any meaning at all, courts must have

the power to review the constitutionality of searches against some

principled standard.

C. Consent

Even if there is no principled reason to allow suspicionless

antiterrorism searches, law enforcement could conceivably rely on the

implied or express consent given by the individual being searched. In

other words, if the individual does not wish to be searched, he or she

can simply choose not to board an airplane, ride the subway, or

attend the political protest. Conversely, if the individual does choose

to engage in the activity, he or she is consenting—at least implicitly—

to the search that goes along with the activity.

306. See, e.g., Sara Kornblatt, Note, Are Emerging Technologies in Airport Passenger

Screening Reasonable Under the Fourth Amendment?, 41 L

OY. L.A. L. REV. 385, 404–05 (2007)

(“There are metal detectors at courthouses, schools, and stadiums. While some may find

magnetometers at these types of locations annoying, society in general has allowed their use

without an uproar that Fourth Amendment rights are being violated.” (footnote omitted)).

307. There is even an argument that once a threat has receded, courts are more likely than

voters to revisit the restrictions on liberty that they put into place. Professor William Stuntz has

noted that the twentieth century saw a pattern in the way the Supreme Court interpreted the

Fourth Amendment: “Crime fell in the 1940s and 1950s; Fourth Amendment rights expanded in

the 1960s. Crime rose sharply in the 1960s; Fourth Amendment protection receded in the 1970s

and 1980s. Crime fell again in the 1990s, and by the end of that decade Fourth Amendment

rights were once again expanding.” William J. Stuntz, Local Policing After the Terror, 111 Y

ALE

L.J. 2137, 2155 (2002). Thus, it is not that courts are unresponsive to the changing world, but

that they are merely slower to respond—both for institutional reasons, and because they are

applying fixed principles of law, which legislatures do not do. This deliberateness is probably a

virtue when courts are considering alterations to the scope of Fourth Amendment protections.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

904 DUKE LAW JOURNAL [Vol. 59:843

Consent is a well-established exception to the Fourth

Amendment’s requirement of individualized suspicion.

308

Law

enforcement officers are permitted to search anyone, for any reason,

as long as the individual agrees to the search. In the early years, some

courts relied—at least in part—on an implied-consent argument to

justify suspicionless antiterrorism searches. In United States v.

Davis,

309

for example, the Ninth Circuit argued that airport screenings

are constitutional “only if they recognize the right of a person to

avoid search by electing not to board the aircraft.”

310

This was in part

because a consensual search was thought to be less intrusive than a

compulsory search, but also because this helped to ensure that the

purpose of the search was the special need of preventing hijackings

rather than the law enforcement need of general crime control:

[A] compelled search . . . would not contribute to barring weapons

and explosives from the plane, [and therefore] it could serve only

the purpose of apprehending violators of either the criminal

prohibition against attempting to board an aircraft while carrying a

concealed weapon . . . or some other criminal statute. Such searches

would be criminal investigations subject to the warrant and probable

cause requirements of the Fourth Amendment.

311

Even today, implied consent is frequently cited as one of the

factors that courts consider in their balancing test, though not a

dispositive one. The Third Circuit has argued that searches of airline

passengers are “less offensive” because “[a]ir passengers choose to

fly,” and they are on notice that they will be searched at the airport.

312

Other courts have rejected suspicionless antiterrorism searches in

part because of a lack of implied consent. For example, the district

court that struck down the searches in the Boston subway noted that

proper notice would have “provide[d] an opportunity for persons who

do not want to permit inspection to avoid traveling on the MBTA,”

313

308. See, e.g., United States v. Miller, 20 F.3d 926, 930 (1994) (“[W]e note that police officers

may search an area, even without probable cause or a warrant, if someone with adequate

authority has consented to the search . . . .”).

309. United States v. Davis, 482 F.2d 893 (9th Cir. 1973), abrogated by United States v.

Aukai, 497 F.3d 955 (9th Cir. 2007) (en banc).

310. Id. at 910–11.

311. Id. at 911–12 (citation omitted).

312. United States v. Hartwell, 436 F.3d 174, 180–81 (3d Cir. 2006).

313. Am.-Arab Anti-Discrimination Comm. v. Mass. Bay Transp. Auth., No. 04-11652-

GAO, 2004 WL 1682859, at *3 (D. Mass. July 28, 2004).

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 905

whereas the court that struck down the searches at Tampa’s stadium

noted that the plaintiff was never informed of the search policy

before purchasing his tickets and therefore could not have consented

to the search at the time of purchase.

314

Conversely, the Second

Circuit upheld suspicionless searches in the New York subway in part

because the individuals being searched have a choice to either submit

to the search or turn away and not ride the subway.

315

But there is a significant problem associated with relying on

implied consent to justify suspicionless antiterrorism searches: the

doctrine of unconstitutional conditions. This doctrine holds that it

could be unconstitutional if the government conditions participation

in a certain activity on the waiver of a constitutional right.

316

The

314. Johnston v. Tampa Sports Auth., 442 F. Supp. 2d 1257, 1271–72 (M.D. Fla. 2006), rev’d

per curiam, 490 F.3d 820 (11th Cir. 2007), vacated and superseded on reh’g, 530 F.3d 1320 (11th

Cir. 2008) (per curiam).

315. MacWade v. Kelly, 460 F.3d 260, 273, 275 (2d Cir. 2006). The Second Circuit never

used implied consent as a legal justification for the search; instead it applied the fact that

individuals could choose to walk away as evidence that the search was “minimally intrusive,” id.

at 273, which is one of the three factors of the reasonableness balancing test for the special

needs doctrine, id. at 269.

There is a recent Ninth Circuit case that appears to hold that the constitutionality of

these searches does not depend on consent, but in reality it only specifies the type of consent

that is required. See United States v. Aukai, 497 F.3d 955, 960 (9th Cir. 2007) (en banc). In

United States v. Aukai, 497 F.3d 955 (9th Cir. 2007) (en banc), the defendant voluntarily

subjected himself to a metal detector and a secondary screening at the airport, but when he was

directed to empty his pockets, he refused, telling the TSA officials that he no longer wished to

board the plane and asking to leave the airport. Id. at 957–58. This request was refused, and a

subsequent search of the defendant revealed several baggies of methamphetamine and a glass

pipe. Id. at 958.

The Ninth Circuit relied on the administrative search doctrine to uphold the search,

and went out of its way to note that the constitutionality of the search did not depend on the

defendant’s consent; thus, the defendant could not revoke his consent and tell the officers that

he no longer wished to fly. Id. at 960–61. But a few sentences later it noted that “all that is

required [for the search to be constitutional] is the passenger’s election to attempt entry into the

secured area of an airport.”

Id. at 961 (footnote omitted). Thus, for all its strong language about

consent not being a necessary element, it appears that the Aukai case simply specifies what kind

of consent is required, without abolishing a consent requirement altogether. Once individuals

choose to undertake an activity (and thus implicitly agree to a search), there comes a point at

which they cannot back out of the activity and withdraw their consent. In other words, this

holding is not inconsistent with a theory that would justify all suspicionless antiterrorism

searches using implied consent. For a recent discussion of these issues, see generally Bethany A.

Gulley, Case Note, United States v. Aukai, 497 F.3d 955 (9th Cir. 2007), 31 U.

ARK. LITTLE

ROCK L. REV. 515 (2009).

316. See, e.g., Adams v. James, 784 F.2d 1077, 1080 (11th Cir. 1986) (“The doctrine of

unconstitutional conditions prohibits terminating [prisoner] benefits, though not classified as

entitlements, if the termination is based on motivations that other constitutional provisions

proscribe.”); Bertrand v. United States, 467 F.2d 901, 902 (5th Cir. 1972) (per curiam) (ordering

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

906 DUKE LAW JOURNAL [Vol. 59:843

Bourgeois and Johnston courts rejected an implied-consent

justification on these grounds, noting that an individual’s right to

attend a protest or even a “property interest in his season tickets and

his right to attend [football] games and assemble with other

Buccaneers fans” could not be conditioned on a surrender of his

Fourth Amendment rights.

317

The reason the doctrine of unconstitutional conditions presents a

problem is that—like a generalized definition of reasonableness—this

doctrine also fails to provide a principled, bright-line test. The

Seventh Circuit notes that “conditions can lawfully be imposed on the

receipt of a benefit—conditions that may include the surrender of a

constitutional right, such as the right to be free from unreasonable

searches and seizures—provided the conditions are reasonable.”

318

Thus, attempting to use implied consent as a justification for

suspicionless searches leads to the same conclusion, and creates the

same problem—in the end it is up to the judge to decide, after

balancing all the factors, whether it is reasonable to condition the

right to travel by air (or enter a public courthouse or attend a football

game) on the individual’s waiver of their Fourth Amendment right.

And once again, the case law shows the inconsistency that inevitably

results from such a broad, discretionary test: courts have said that it is

unreasonable to condition entry to a football game or a rock concert

on a Fourth Amendment waiver;

319

yet it is not unreasonable to

require consent before entering a courthouse or boarding a plane.

320

resentencing because “[t]he effect of the trial judge’s questioning [of the defendant] was to

impose an unconstitutional condition on the petitioner’s Fifth Amendment rights: he could go

into the details of the other offense . . . that might constitute a confession or he could exercise

his right to be silent and receive a long sentence”); Boykins v. Fairfield Bd. of Educ., 399 F.2d

11, 13 (5th Cir. 1968) (“The plaintiffs [were] admitted to the school system, but had been denied

the opportunity to transfer from a Negro to a white school. Once the plaintiffs had been

admitted to the school system, they had a constitutional right to a desegregated education, and

have standing to enforce that right—free of any unconstitutional condition precedent.” (quoting

Bossier Parish Sch. Bd. v. Lemon, 370 F.2d 847, 851 (5th Cir. 1967))).

317. Johnston, 442 F. Supp. 2d at 1271; accord Bourgeois v. Peters, 387 F.3d 1303, 1324–25

(11th Cir. 2004).

318. Burgess v. Lowery, 201 F.3d 942, 947 (7th Cir. 2000).

319. Johnston, 442 F. Supp. 2d at 1271; Nakamoto v. Fasi, 64 Haw. 17, 22–26 (1981).

320. The Supreme Court has perhaps provided a preview of how they would interpret the

doctrine of unconstitutional conditions in its “reasonable expectation of privacy” jurisprudence.

In Smith v. Maryland, 442 U.S. 735 (1979), the Court held that an individual has no reasonable

expectation of privacy in the telephone numbers he calls, because he “voluntarily”

communicates these numbers to the telephone company when he dials them. Id. at 744.

According to Smith’s reasoning, the only way to maintain a reasonable expectation of privacy in

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 907

D. Antiterrorism Searches Are Sui Generis; Courts Should Create a

New Category of Suspicionless Searches

Another possibility is simply to concede that antiterrorism

searches are sui generis and create a new, narrowly tailored exception

to the individualized suspicion rule. This rule could be designed so

that it would apply only to searches designed to prevent terrorist

actions, thus responding to contemporary dangers without

unnecessarily broadening the scope of special needs searches in other

contexts.

321

As one commentator noted: “[T]he exception might

expand to cover a variety of different search methods employed in

new contexts as the type of weapons and the types of targets changed

over time.”

322

If the exception stood on its own doctrinal basis, none

of these changes would affect other types of suspicionless searches.

323

This is not a trivial benefit. As one commentator points out, this

country has conventionally recognized a “traditional criminal

process” and an emergency criminal process.

324

The former, defined as

“investigation with the goal of proving criminal charges in an ordinary

criminal court,”

325

provides substantial rights to defendants, including

the right to a speedy and public trial. The latter, which involves the

military justice system as well as special military tribunals that are

created during times of emergency, is a more flexible and efficient

process in which “constitutional and statutory rules apply more

leniently, [so] executive officials have discretion to craft strategies for

the specific needs of a particular investigation or other activity.”

326

Historically, these two types of criminal processes have remained

somewhat separate, but the government’s response to the 2001

terrorist attacks has changed that somewhat. Today, many of the

telephone numbers is to not use the telephone at all. A similar analysis applied to the doctrine

of unconstitutional conditions would invariably result in findings of implied consent for almost

every activity.

321. See Charles J. Keeley III, Note, Subway Searches: Which Exception to the Warrant and

Probable Cause Requirements Applies to Suspicionless Searches of Mass Transit Passengers to

Prevent Terrorism?, 74 F

ORDHAM L. REV. 3231, 3287–91 (2006) (arguing for a sui generis

approach to mass transit searches).

322. Id. at 3289.

323. Id.

324. See Parry, supra note 3, at 766 (“[T]he ‘war on terror’ has accelerated the development

of a new criminal process and . . . this new process has increasingly displaced traditional

methods of investigating, prosecuting, and punishing people who have engaged in conduct that

is subject to criminal penalties . . . .”).

325. Id. at 791–92.

326. Id. at 792.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

908 DUKE LAW JOURNAL [Vol. 59:843

extraordinary actions taken by the government in response to the

crisis posed by terrorist threats affect the traditional criminal process,

resulting in a hybrid process in which “legally authorized discretion is

increasingly valued as a way to respond to a steady stream of

perceived crises.”

327

In practical terms, the current war on terror has

resulted not only in detention without charge and trial by a special

tribunal for terrorist suspects in Guantánamo, but also in enhanced

search and seizure powers against all citizens, including warrantless

wiretapping of overseas telephone calls, data mining of financial

information using national security letters, and more frequent

detention of material witnesses.

328

Although suspicionless searches are

(for the most part) carried out by local law enforcement, they are part

of this same trend. If these heightened powers of surveillance can be

limited more explicitly to terrorism investigations, the incremental

expansion of these powers to everyday investigations can be

prevented.

Creating a new exception would also help to shore up the

integrity of the special needs doctrine—because preventing terrorism

is not really a special need distinct from law enforcement, courts

would no longer have to use terms like “administrative,”

“regulatory,” and “law enforcement purpose” in ways that have

nothing to do with their actual meaning. This could be a critical first

step to bringing logic and doctrinal consistency to what is now a very

confused jurisprudence.

The Supreme Court has even hinted that a new exception might

be appropriate for antiterrorism searches. Edmond, in which the

Court struck down a suspicionless roadblock designed to detect illegal

drug trafficking, commented in dicta that “the Fourth Amendment

would almost certainly permit an appropriately tailored roadblock set

up to thwart an imminent terrorist attack.”

329

Although the Supreme

Court used the word imminent, the Foreign Intelligence Surveillance

Act (FISA) court later relied in part on this dictum to approve

wiretap applications that were meant to detect terrorist activity.

330

Noting that the Supreme Court had implicitly approved of applying

327. Id. at 835.

328. Id. at 770–82.

329. City of Indianapolis v. Edmond, 531 U.S. 32, 44 (2000); see also Gould & Stern, supra

note 3, at 821–23 (analyzing the implications of this statement).

330. In re Sealed Case, 310 F.3d 717, 745–46 (FISA Ct. Rev. 2002).

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 909

the special needs doctrine in the case of “an imminent terrorist

attack,” the FISA court argued that “[t]he nature of the ‘emergency,’

which is simply another word for threat, takes [antiterrorism cases]

out of the realm of ordinary crime control.”

331

In a recent article, Judge Ronald Gould of the Ninth Circuit,

writing with his former law clerk Simon Stern, used this language to

support a new exception to the individualized suspicion

requirement.

332

The authors proposed broadening the special needs

doctrine to allow any search as long as (1) the search is meant to

“prevent catastrophic harm;” (2) the search “goes beyond routine

police functions;” (3) the search is “as effective as is

practical . . . [and] minimize[s] harm to the public;” (4) the officers’

discretion is constrained and the search is not discriminatory; and (5)

under the totality of the circumstances, the balance favors the

governmental interest when compared with the infringement on the

privacy of those searched.

333

One problem with creating this new exception is evident both in

the FISA court’s opinion and in Gould and Stern’s proposed test:

even with a rule designed specifically for antiterrorism searches, the

slippery slope problem still exists. This slippery slope is even apparent

in the transition from Edmond’s dictum to the FISA court opinion:

Edmond spoke of a search to meet an imminent threat in an

emergency situation, but the FISA court—inexplicably equating

“emergency” with “threat”—used this language to justify a wiretap

when there was no evidence of an emergency or imminent harm. And

the Gould and Stern test contains a final factor that simply asks courts

to balance the governmental interest against the privacy

infringement—thus providing no principled bright-line test to guide

courts and limit the types of search that may be permitted.

334

331. Id. at 746.

332. Gould & Stern, supra note 3, at 823.

333. Id.

334. The Gould & Stern test is only one of many proposed tests that ultimately rely on the

reasonableness of the search. For example, Professor Edwin Butterfoss proposes a test first

developed by Scott Sundby twenty years ago—divide searches into two categories: “initiatory

intrusions” and “responsive intrusions.” See Butterfoss, supra note 252, at 488 (quoting Scott E.

Sundby, A Return to Fourth Amendment Basics: Undoing the Mischief of Camara and Terry, 72

INN. L. REV. 383, 418–21 (1988)). Responsive intrusions would be covered by the full range of

Fourth Amendment protections whereas initiatory intrusions would be subject to a balancing

test based on the subject’s reasonable expectation of privacy, the level of intrusion, and the

magnitude of the government interest. Id. at 488–95.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

910 DUKE LAW JOURNAL [Vol. 59:843

Recently, Professor Ricardo J. Bascuas proposed his own test

that avoids the problematic “totality of the circumstances” language

found in many of the proposed antiterrorism exceptions. In an

attempt to set out concrete, principled guidelines for suspicionless

antiterrorism searches, Bascuas proposes a test that permits such

searches if: (1) they are “justified by credible, non-speculative

evidence of . . . danger;” (2) the “danger must entail imminent

physical injury;” and (3) “anything seized unrelated to the

danger . . . [is] suppressed.”

335

This test goes a long way toward

providing stable guidelines for these searches, as well as preventing

pretextual searches in which law enforcement officials conducting

ordinary crime control claim to be deterring terrorist activity to

bypass ordinary Fourth Amendment rules.

But even a well-crafted, narrowly tailored exception for

antiterrorism searches is problematic; depending on how it is

interpreted, it is either too narrow or too broad. Bascuas’s proposal,

for example, requires “credible, non-speculative evidence” and

“imminent physical injury.”

336

Bascuas uses these terms to distinguish

between the vague “alarmist policy arguments” that failed to justify

the search of protesters in Bourgeois with the concrete “sober

evidence” produced by the N.Y.P.D. to justify the subway searches in

MacWade.

337

But what exactly constitutes credible, non-speculative

evidence of an imminent attack? In MacWade, the government

showed that there had been two plots to bomb the subway over the

past eight years, and that terrorists had recently bombed the subway

systems of three European cities.

338

Although this surely demonstrates

that subway systems are a tempting target for terrorists, does it

335. Bascuas, supra note 2, at 781. The concept of precluding any evidence found that is not

related to the danger which justifies the search is an intriguing one. Justice Scalia suggested this

idea at one point in the context of Terry stops, saying that it is supported by “the theory that

half a constitutional guarantee is better than none.” Minnesota v. Dickerson, 508 U.S. 366, 382

(1993) (Scalia, J., concurring) (“If I were of the view that Terry was (insofar as the power to

‘frisk’ is concerned) incorrectly decided, I might—even if I felt bound to adhere to that case—

vote to exclude the evidence incidentally discovered . . . .”). But ultimately this “half a

constitutional guarantee” is insufficient in the context of antiterrorism searches because it still

allows the government to conduct suspicionless searches and use the weapons that are

recovered in a subsequent criminal prosecution, thus bypassing the individualized suspicion

requirements of the Fourth Amendment.

336. Bascuas, supra note 2, at 781.

337. Id. at 786.

338. MacWade v. Kelly, 460 F.3d 260, 264 (2d Cir. 2006).

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 911

represent credible evidence of an imminent attack? And how would a

court apply this test to the original antiterrorism searches at airports

and public buildings? Terrorists have targeted airplanes, and

airplanes are particularly vulnerable to massively destructive acts of

violence, but is there any evidence that an attack on airplanes is

imminent? And what evidence exists to support an imminent attack

on judges or courthouses?

This critique is not specific to Professor Bascuas’s test; it applies

to any attempt to create a new antiterrorist exception to the Fourth

Amendment. Either it will only apply for a short period of time after

a credible threat has been detected in a certain area or venue, in

which case the search will no longer be justified once the imminence

wears off, or it will remain in place in perpetuity, like the searches at

airports and public courthouses, long after the concrete and imminent

threat has subsided. The latter, of course, is far more likely.

339

And in

any case, it will still be the terrorists who control the scope of the

exception; successful attacks against a certain type of target—or even

foiled plots that are uncovered by law enforcement—will represent

credible evidence that an imminent (yet never-ending) threat exists

for that category of target nationwide.

* *

This Part has considered four possible justifications for

suspicionless antiterrorism searches: using a generalized

reasonableness standard, relying on the legislature and the

democratic process to set the limits of what is reasonable, applying

the doctrine of implied consent, and creating an entirely new category

of suspicionless searches tailored for antiterrorism cases.

Unfortunately, none of these potential justifications are viable, which

leads—perhaps reluctantly—to the conclusion that these searches

should not be permissible at all.

VI.

WHAT IF SUSPICIONLESS ANTITERRORISM SEARCHES ARE

NCONSTITUTIONAL?

We have seen that suspicionless antiterrorism searches do not fit

into the administrative or special needs categories to which they have

been assigned. Because they are designed to prevent and detect

criminal behavior, and because law enforcement uses the results of

339. See supra notes 57–66 and accompanying text.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

912 DUKE LAW JOURNAL [Vol. 59:843

these searches, it is a legal fiction to say that they serve a purpose

other than law enforcement. Thus, only three options remain: abolish

the special needs threshold and simply analyze these searches under a

generalized reasonableness standard;

340

create a new exception to the

individualized suspicion requirement and tailor that exception for

antiterrorism cases; or conclude that suspicionless antiterrorism

searches should be unconstitutional. There are problems with the first

two options: an unchecked slippery slope, an ominous but inexorable

(and probably irreversible) shift in perceptions of what constitutes a

reasonable search by the government, and ultimately allowing the

terrorists themselves to determine the scope of Fourth Amendment

liberties.

341

In short, the first two options are essentially abandoning

principled limitations at a time when those principles are needed the

most. Therefore, it is at least worth asking: what if these searches

were found to be unconstitutional?

A. Abolishing Antiterrorism Searches

The simplest option would be to abolish antiterrorism searches

altogether. At first, this possibility seems unthinkable: surely no court

today would forbid the long-standing suspicionless searches at

airports and public courthouses. The very fact that this last option

seems unthinkable is itself deeply troubling: in only thirty years, the

notion of law enforcement agents subjecting individuals and their

belongings to suspicionless searches has become not just

constitutional and acceptable, but also thought to be indispensable.

The searches have become a part of our way of life, a part of our

culture, so that we have come to accept that there is no other way to

keep us safe on airplanes or in courthouses. In this sense, it may be

instructive to consider how “unthinkable” it was to many Americans

when the Supreme Court desegregated public schools,

342

or when the

Supreme Court required suppression of any confession made without

340. The reasonableness test will apply even if law enforcement tries to rely on implied

consent to justify its searches, due to the doctrine of unconstitutional conditions. See supra notes

311–17 and accompanying text.

341. See supra notes 285–307 and accompanying text.

342. See Brown v. Bd. of Educ., 394 U.S. 294, 299 (1955) (“Full implementation of these

constitutional principles may require solution of varied local school problems.”).

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 913

an informed, express waiver of a right to counsel.

343

Indeed, the

dissent in Miranda v. Arizona

344

predicted that “a good many criminal

defendants who would otherwise have been convicted . . . will now,

under this new version of the Fifth Amendment, either not be tried at

all or will be acquitted.”

345

Of course, nothing of the kind actually

occurred—instead, the Court merely forced law enforcement to

adhere to certain fundamental requirements of the right against self-

incrimination, and law enforcement officers were able to adapt their

tactics so that defendants were still convicted at the same rate.

346

As with Miranda, the actual effect of declaring such searches

unconstitutional might be far less severe than it first appears. Even

without suspicionless searches, the police would be far from

powerless to detect and apprehend potential terrorists. As Professor

William Stuntz has pointed out, even if suspicionless searches were

not allowed, police could use existing authority to great effect.

347

According to Stuntz, the combination of Atwater v. City of Lago

Vista

348

(which held that the police could make an arrest—and

therefore an attendant search—for any crime, regardless of how

trivial

349

), and Whren v. United States

350

(which held that an officer’s

motive in conducting a search is irrelevant

351

) already give police

broad powers to conduct de facto suspicionless searches in almost any

343. See Miranda v. Arizona, 384 U.S. 436, 492 (1966) (holding a criminal defendant’s

confession to police inadmissible without a “knowing and intelligent waiver” of Fifth

Amendment rights).

344. Miranda v. Arizona, 384 U.S. 436 (1966).

345. Id. at 542 (White, J., dissenting). Justice White continued rather dramatically:

In some unknown number of cases the Court’s rule will return a killer, a rapist or

other criminal to the streets and to the environment which produced him, to repeat

his crime whenever it pleases him. As a consequence, there will not be a gain, but a

loss, in human dignity. The real concern is not the unfortunate consequences of this

new decision on the criminal law as an abstract, disembodied series of authoritative

proscriptions, but the impact on those who rely on the public authority for protection

and who without it can only engage in violent self-help with guns, knives and the help

of their neighbors similarly inclined.

Id.

346. See, e.g., R

ICHARD A. LEO & GEORGE CONNER THOMAS, THE MIRANDA DEBATE:

LAW, JUSTICE, AND POLICING 56 (1998).

347. See Stuntz, supra note 307, at 2141. Professor Stuntz argues that the current system,

which allows police to conduct suspicionless antiterrorism searches in some contexts and

promises to allow them in many more as time goes on, represents a “healthy bribe” to the police

to prevent them from engaging in more intrusive and discriminatory searches. Id.

348. Atwater v. City of Lago Vista, 532 U.S. 318 (2001).

349. See id. at 354.

350. Whren v. United States, 517 U.S. 806 (1996).

351. See id. at 813.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

914 DUKE LAW JOURNAL [Vol. 59:843

context.

352

When further combined with the low standard that justifies

a Terry search and the ease with which police can elicit consent,

police have plenty of options without resorting to suspicionless

searches.

353

In many contexts, particularly commercial air travel, private

companies would immediately pick up where law enforcement left

off. Allowing airlines to screen their own passengers and luggage

would pose no Fourth Amendment problems.

354

Public law

enforcement could focus on less intrusive methods of preventing air

piracy, such as increased use of sky marshals. New technologies could

be deployed that do not implicate the Fourth Amendment, such as

trace detection machines that only alert if illegal explosives are

detected.

355

In other contexts, such as suspicionless searches at public

buildings, the necessity argument is much weaker: is there really no

other way of protecting court personnel than to place a security

cordon around every government building and create a miniature

“green zone” on the inside? Is the threat posed by terrorists (or

ordinary criminals) so much more severe in the context of

government buildings that extraordinary search procedures are

justified there, as opposed to at a public school, or along a parade

route, or at a public sporting event?

At any rate, suspicionless searches at the entryway of public

buildings do nothing to prevent the threat of a real terrorist attack—a

car bomb or other large explosive being detonated outside, next to, or

underneath the building; currently, physical impediments (such as

bollards and setbacks from the road) prevent such crimes.

352. Stuntz, supra note 307, at 2158.

353. Given this existing broad authority, Stuntz argues that allowing an even greater range

of suspicionless searches and more carefully regulating the manner in which they are carried out

would be preferable. Id. at 2168–69.

354. Under the state action doctrine, the Fourth Amendment would be implicated as long as

these searches were required by the government, as has been the case since the early 1970s.

Thus, the government would have to abolish the requirement and allow airlines to take

responsibility for their own passengers’ safety. Consumers may or may not find this to be a

welcome development; private security guards may be more or less abusive—or more or less

effective—than government personnel.

355. A surveillance procedure that can only detect the presence or absence of illegal activity

is not considered a search under the Fourth Amendment. See United States v. Place, 462 U.S.

696, 707 (1983) (concluding that luggage sniffing by narcotics detection dogs is not a search

under the Fourth Amendment).

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 915

Furthermore, if law enforcement personnel had any reasonable

suspicion of a threat, they could still conduct Terry searches of that

individual. And lastly, the criminal law itself would provide a

powerful deterrent to any criminal activity, as it does in every other

context—in fact, the deterrence would be even more powerful than

usual because it is almost certain that anyone who committed a crime

inside a courthouse or other government building would be

apprehended, and security cameras would provide solid evidence at

any subsequent trial.

But most people would say these safeguards are not nearly

enough. At airports, private security guards may not be as well

trained as government agents, and the threat that terrorists pose to air

traffic is a matter of national security, not to be left to the cost-benefit

calculations of the private sector. At courthouses, there may be

certain times when higher security is required—during the trial of a

particularly high-profile terrorist, or after a credible threat has been

received regarding a city or public buildings generally. And as far as

the deterrent effect of the criminal sanction, many terrorists have

proven themselves to be undeterrable; thus, the prospect of being

captured and imprisoned after their crime has been committed will do

little to prevent them from carrying out their actions.

The special needs doctrine provides a ready solution to these

problems: suspicionless searches could be permitted as long as the

government is not permitted to use any fruits of the search in a

subsequent prosecution. In other words, such a search would be

perfectly legal—because it is justified by the special need of

protecting public safety—so citizens could not bring any civil suits

against government agents who conducted suspicionless searches for

terrorists. If the government agents, however, sought to use the

evidence recovered in the search in a subsequent prosecution, a court

would conclude that the search was no longer a special need search,

but rather was conducted for the purpose of crime control, and so the

search would be unconstitutional. Thus, the searches would be

allowed to prevent any terrorist attack from occurring, but the

exclusionary rule would apply so that any contraband that was

recovered would be precluded from use during any subsequent

criminal trial.

B. Allowing the Searches, but Precluding the Evidence

Although at first this proposal sounds radical (not to mention

politically unpalatable), it finds strong support in criminal procedure

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

916 DUKE LAW JOURNAL [Vol. 59:843

jurisprudence. Most importantly, it comports with the legitimate

justification for special needs searches. Originally, suspicionless

searches were allowed only if the government agents were conducting

the search for a purpose other than generalized crime control. This

threshold requirement has been watered down to nonexistence over

the decades, culminating in the rather absurd—but consistently

repeated—assertion that preventing terrorist actions is not a law

enforcement purpose. Not every special needs search suffered from

this fatal weakness: some—the original administrative searches—

sought information for a legitimate regulatory purpose. Others—for

example, the cases involving drug testing of school children or public

employees—involved searches for unambiguously criminal activity,

but did not use the results for criminal prosecutions.

356

If the

government did not use the results of suspicionless antiterrorist

searches in a subsequent prosecution, these searches would fall

squarely into the second category of special needs searches, alongside

the drug tests of Acton, Skinner, and Von Raab.

357

To carry the

analogy with the drug testing cases one step further, under the current

doctrine, suspicionless antiterrorist searches at airports are really the

same as the invalidated drug tests in Ferguson, in which the public

hospital tested pregnant mothers—supposedly to protect the health of

the mother and fetus—and then turned the results over to law

enforcement.

358

If the fruits of the search were barred from use in

future criminal prosecution, these searches would become more like

those of schoolchildren in Acton and public train drivers in Skinner.

Essentially, barring the government from using the fruits of the

search in a subsequent prosecution would accomplish very simply

what many courts have been trying to establish throughout the

tortured history of antiterrorism searches: ensure that the actual

purpose of the search is to prevent terrorist actions, rather than to

detect, apprehend, and prosecute criminals. The court in MacWade,

for example, concluded that the New York subway search program

served a need other than law enforcement because the officers

“search only those containers capable of carrying explosive devices,

356. See supra Part IV.A.

357. See supra Part IV.A.

358. See Ferguson v. City of Charleston, 532 U.S. 67, 71–73 (2001); see also supra notes 242–

51 and accompanying text.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 917

and they may not intentionally search for other contraband.”

359

Likewise, some of the first courts to review suspicionless airport

searches relied on the implied consent of the suspect being searched

as evidence that the search was narrowly tailored to prevent

terrorism, and therefore did not serve a general crime-control

purpose.

360

But these arguments do not really show that the searches

serve a need other than law enforcement; they merely perpetuate the

fiction that detecting and arresting people who seek to blow up

subway cars or airplanes is not a law enforcement purpose.

361

If the

results of a search were not used in a future criminal prosecution,

there would be no doubt that “the essential purpose of the scheme is

not to detect weapons or explosives or to apprehend those who carry

them, but to deter persons carrying such material from seeking to

board at all.”

362

Another way of looking at this rule is to say that if law

enforcement officials decide to use the fruits of the search in a

criminal prosecution, it would create a conclusive presumption that

the search was conducted for traditional law enforcement purposes,

thus triggering the Fourth Amendment’s individualized suspicion

requirements. This would create a simple solution to the problem that

the Supreme Court created with Edmond and Ferguson: how to

determine when a search was conducted for traditional law

enforcement purposes.

Adopting this rule would mean that the focus of suspicionless

antiterrorist searches would truly be on prevention, not on

apprehension and subsequent punishment. Of course, applying the

exclusionary rule would make these searches less effective at

359. MacWade v. Kelly, 460 F.3d 260, 270 (2d Cir. 2006).

360. See, e.g., United States v. Davis, 482 F.2d 893, 910–11 (9th Cir. 1973) (“To meet the test

of reasonableness, an administrative screening search must be as limited in its intrusiveness as is

consistent with satisfaction of the administrative need that justifies it. It follows that airport

screening searches are valid only if they recognize the right of a person to avoid search by

electing not to board the aircraft. It is difficult to see how the need to prevent weapons and

explosives from being carried aboard the plane could justify the search of a person who had

elected not to board.” (footnotes omitted)), abrogated by United States v. Aukai, 497 F.3d 955

(9th Cir. 2007) (en banc).

361. Professor Bascuas makes a similar, if more sophisticated argument by proposing to

exclude any contraband that is “unrelated to the justification for suspicionless searches.”

Bascuas, supra note 2, at 787–88. This rule would ensure that antiterrorism searches are not

used as pretexts for generalized crime-control searches, which is one of Bascuas’s chief

concerns. See id. at 758–69. But it does not go far enough because it does not address the

fundamental hypocrisy of categorizing antiterrorism searches as special needs searches.

362. See Davis, 482 F.2d at 908.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

918 DUKE LAW JOURNAL [Vol. 59:843

prevention. Currently the risk of getting caught and ultimately going

to prison is a strong deterrent to at least some potential terrorists, and

incapacitation through incarceration can help to ensure that any

particular criminal who gets caught will no longer be a threat for the

time he is in prison. But many potential terrorist suspects are

impossible to deter by any means, and so the absence of a threat of

prison is unlikely to alter their behavior. And although they cannot be

incapacitated through incarceration based on the recovered items, the

government is not without options once a suspect is caught. Law

enforcement agents could interview the suspect to learn more about

his plans. They could begin surveillance on the suspect.

363

In many

cases, they would be able to institute deportation proceedings against

the suspect. These options would not be as effective as using the

results of the search in the criminal prosecution—but again, the focus

of the suspicionless searches is by definition not bringing the suspect

to justice; rather, it is preventing the terrorist act from occurring.

In dissent, Justice Marshall argued in favor of this sort of rule in

the Fifth Amendment context, pointing out that if a police officer

wished to protect public safety, he or she was free to ignore the

Miranda rules—but that this should not somehow render the

compelled testimony admissible:

If a bomb is about to explode or the public is otherwise

imminently imperiled, the police are free to interrogate suspects

without advising them of their constitutional rights. Such

unconsented questioning may take place not only when police

officers act on instinct but also when higher faculties lead them to

believe that advising a suspect of his constitutional rights might

decrease the likelihood that the suspect would reveal life-saving

information.

364

Marshall acknowledged that in certain situations, this would

mean the defendant would go free if there were no other way to

prosecute him for the crime. But “however frequently or infrequently

363. Although the “fruit of the poisonous tree” doctrine would prevent law enforcement

officials from using any evidence that was discovered as a direct result of the suspicionless

search (for example, a confession resulting from the subsequent arrest), it would not preclude

law enforcement from beginning to monitor the public activities of the suspect and gather

evidence of any new crimes against him. See generally Wong Sun v. United States, 371 U.S. 471,

487–88 (1963) (describing operation of the “fruit of the poisonous” tree doctrine in the context

of illegal police actions).

364. New York v. Quarles, 467 U.S. 649, 686 (1984) (Marshall, J., dissenting).

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 919

such cases arise, their regularity is irrelevant. The Fifth Amendment

prohibits compelled self-incrimination.”

365

Of course, the Court rejected Justice Marshall’s argument in the

Fifth Amendment context, creating a public safety exception to the

Miranda rule.

366

Yet the Supreme Court has confirmed that—outside

the public safety context—law enforcement officers are faced with a

choice: they are free to ignore Miranda as long as they do not use any

subsequent statements at a criminal trial.

367

Allowing suspicionless

antiterrorism searches but precluding the evidence from trial would

give the officers the same choice.

Many would argue that the analogy with Fifth Amendment

jurisprudence is fallacious. The Fifth Amendment precludes

compelled self-incrimination, which naturally leads courts to conclude

that the Amendment does not apply if the statements are not used in

a criminal case.

368

The Fourth Amendment’s text is much broader; it

precludes all “unreasonable searches,” regardless of how the

government uses these searches. And the Supreme Court has said

numerous times that the Fourth Amendment applies with equal force

to civil and criminal cases, whether or not the evidence is used in a

subsequent trial.

369

365. Id. at 687.

366. Id. at 657–58 (majority opinion). The Court employed reasoning quite similar to what is

used to justify suspicionless antiterrorism searches in the Fourth Amendment context, id.,

noting that “we do not believe that the doctrinal underpinnings of Miranda require that it be

applied in all its rigor to a situation in which police officers ask questions reasonably prompted

by a concern for the public safety,” id. at 658 n.7.

367. See, e.g., Chavez v. Martinez, 538 U.S. 760, 766 (2003) (reasoning that Fifth

Amendment rights are not implicated when statements by a subject of a police investigation are

not admitted in a criminal proceeding). Just as with suspicionless antiterrorism searches, the

courts seem unwilling to apply this doctrine when there is a vague risk to public safety—even

though the existence of a danger to public safety has no bearing on whether or not a statement

is the result of a coercive interrogation. See Parry, supra note 3, at 816–18 (noting that

“interrogation issues . . . are litigated along two tracks,” and using this as an example of how a

“new criminal process”—more like war and less like crime control—“restricts the space in

which constitutional rights operate to the courtroom alone”).

368. The Fifth Amendment is also distinct from the Fourth Amendment in that the former

provides a right and a remedy together—effectively stating that compelled testimony is

inadmissible in a criminal proceeding. The Fourth Amendment provides a right but no remedy,

which has led to courts creating the exclusionary rule.

369. See, e.g., Camara v. Mun. Court, 387 U.S. 523, 530 (1967) (“But we cannot agree that

the Fourth Amendment interests at stake in these inspection cases are merely ‘peripheral.’ It is

surely anomalous to say that the individual and his private property are fully protected by the

Fourth Amendment only when the individual is suspected of criminal behavior.” (quoting Frank

v. Maryland, 359 U.S. 360, 367 (1959), overruled by Camara, 387 U.S. 523)).

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

920 DUKE LAW JOURNAL [Vol. 59:843

But the special needs doctrine itself is proof that this is not in fact

how the Fourth Amendment has been applied: if evidence that is

found as a result of a search is not used for a criminal prosecution, the

Court is willing to suspend the individualized suspicion requirement.

As it turns out, this is a perfectly sensible distinction. Shifting the

focus from the purpose of the search to the use of the results helps to

resolve a glaring inconsistency in Fourth Amendment law.

The Supreme Court has made it clear that judges should not

delve into the subjective intentions of law enforcement officers who

conduct a search

370

—yet the purpose of the search is a critical issue in

deciding whether the special needs doctrine applies. If the fruits of

the search are never used in a subsequent criminal prosecution, it is

easy to determine that the purpose of the search was other than law

enforcement. But otherwise it is almost impossible to distinguish

between the purpose of the search (the determinative question in

special needs cases) and the subjective intentions of those who

conduct the search (an improper inquiry under current Fourth

Amendment jurisprudence).

One commentator tries to resolve this inconsistency by

distinguishing between the “motive” of a search and the “fringe

benefits” of the search,

371

or by inquiring whether or not crime control

was the “causative or substantive factor” in the decision to conduct

the search.

372

The Supreme Court itself in Ferguson tried to draw a

distinction between the immediate purpose of the search and the

ultimate purpose of the search—or whether law enforcement was a

means to an end or a happy side effect of the search.

373

But none of

this matters to the suspect who is being searched; his reasonable

expectation of privacy is being infringed upon to the same degree

regardless of the motive of the search.

In other words, one should pay less attention to why the search

was conducted—which requires courts to delve into the mind of the

police officers or police officials, and probably has multiple

370. See Whren v. United States, 517 U.S. 806, 813 (1996) (declining to analyze the

subjective intent of police officers in the context of the Fourth Amendment).

371. Brooks Holland, The Road ‘Round Edmond: Steering Through Primary Purposes and

Crime Control Agenda, 111 P

ENN ST. L. REV. 293, 305 (2006).

372. Id. at 309.

373. See supra notes 242–51 and accompanying text.

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 921

answers

374

—and instead focus on how the results of the search are

used once they are acquired: are they used as evidence in a criminal

prosecution or merely to establish regulatory violations, maintain

school discipline, or prevent a bomb from coming on board an

airplane? This distinction is logical: an individual who is being

searched by a law enforcement officer does not care why the officer is

searching him—a checkpoint and visual search of a car pursuant to a

drunk-driving checkpoint

375

is just as intrusive as a checkpoint and

visual search of a car by officers looking for information on a recent

hit and run.

376

But a subject is likely to care quite a bit about whether

the results of the search will later be used in a criminal prosecution

against him.

In short, one of the primary reasons for limiting the power of the

state when it conducts searches is the state’s ability to use the fruits of

the search to prosecute individuals. What the special needs cases

indicate is that these searches become less intrusive—and thus

possibly constitutional—if the law enforcement agent does not use

the results of the search in a subsequent criminal prosecution.

C. Responding to Potential Criticisms

A rule allowing suspicionless antiterrorism searches would find

plenty of opposition on both ends of the political spectrum. To

conservatives, the idea that a police officer could apprehend a

criminal attempting to blow up an airplane or shoot a judge and then

not be allowed to use the recovered evidence against the criminal

might seem ludicrous. But of course this is already the law in every

other Fourth Amendment context: if law enforcement officers search

someone without any individualized suspicion and recover

contraband of any kind, the exclusionary rule precludes the

contraband from being used in court. There is no reason why certain

crimes should be exempt from the exclusionary rule, however serious

they may be. Of course, the exclusionary rule itself has been the

374. Of course, it is somewhat easier to determine the purpose of a search if courts look to

the programmatic purpose of the search, as opposed to the individual motive of the officer

conducting the search, but by either definition any given search would probably have multiple

purposes.

375. See supra notes 234–41 and accompanying text.

376. See supra note 241. Compare City of Indianapolis v. Edmond, 531 U.S. 32, 48 (2000)

(holding the Indianapolis drunk driving checkpoint system unconstitutional under the Fourth

Amendment), with Illinois v. Lidster, 540 U.S. 419, 427–28 (2004) (upholding the

constitutionality of police checkpoints used to ask motorists about a hit-and-run incident).

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

922 DUKE LAW JOURNAL [Vol. 59:843

subject of a significant amount of criticism,

377

but for the moment it

remains the centerpiece of enforcement for criminal procedure

violations, and so in a sense this Article merely proposes that the rule

should be applied to every search, regardless of the type of crime

being investigated.

378

Liberals, meanwhile, might worry about abuse on the part of law

enforcement: if police no longer need to worry about violating the

Fourth Amendment, what will prevent them from conducting overly

intrusive searches when searching for terrorists?

379

One response to

this concern is to take a realistic look at the state of the suspicionless

search doctrine. Many courts have held that antiterrorism searches

meet the regulatory purpose of protecting public safety. And most

courts that have rejected the special needs test for antiterrorism

searches (such as Johnston or Seglen) are willing to change their

minds if there is credible evidence of a terrorist threat—thus opening

the door to widespread, indiscriminate suspicionless searches in the

name of preventing terrorism. The only other check under current

law on the government’s ability to conduct such searches is that they

must be reasonable—and as this Article demonstrates, this is at best

an unpredictable test, with the government more and more likely to

win as the terrorist threat (or the perceived terrorist threat) increases.

Thus, under current law the government does not face many obstacles

in creating suspicionless antiterrorism searches—and if the

government agents overstep their bounds and conduct a search that is

unreasonable, the usual remedy is simply that the evidence cannot be

used in court—which would be no different from the result under the

proposed rule.

380

377. See H. Mitchell Caldwell, Fixing the Constable’s Blunder: Can One Trial Judge in One

County in One State Nudge a Nation Beyond the Exclusionary Rule?, 2006 BYU

L. REV. 1, 1–2

(surveying criticism of the exclusionary rule). See generally Amar, supra note 33, at 785–800

(critiquing the exclusionary rule as a remedy for violations of constitutional rights).

378. Even if the exclusionary rule is eventually abolished by the Supreme Court, the

rationale for excluding contraband recovered in special needs searches remains because the

search can only be honestly termed “special needs” if the results of the search are not used

against the suspect in a criminal trial.

379. Judge Friendly, for example, warned against unbridled searches if police were no

longer concerned with the Fourth Amendment. See Henry J. Friendly, The Bill of Rights as a

Code of Criminal Procedure, 53 C

AL. L. REV. 929, 949 (1965) (analyzing the relationship of the

exclusionary rule to the Fourth Amendment).

380. It is true, of course, that individuals whose Fourth Amendment rights are violated have

the right to bring a civil rights suit under 42 U.S.C. § 1983. See, e.g., Torbet v. United Airlines,

298 F.3d 1087, 1088 (9th Cir. 2002) (presenting factual background for a statutory claim of a

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 923

Another response to those worried about overly intrusive

searches is that even if the Fourth Amendment allowed

indiscriminate suspicionless antiterrorism searches, the Due Process

Clause would still apply to limit the search method used by the police

and prevent abuses of this power.

381

There is a long history of case law

that defines the due process limits of police activity, leading to a well-

established test: police activity that is “so brutal and so offensive to

human dignity” that it “shocks the conscience” violates an

individual’s due process rights.

382

If an antiterrorism search was so

extremely abusive that it met this high standard, the subject of the

search would be able to recover damages from the law enforcement

officer.

383

For example, a brief pat-down search of every person who enters

the subway system at a certain station could be termed a special needs

search. As long as the police do not seek to use any contraband

Fourth Amendment rights violation). These lawsuits are rare, however, and are not the primary

mechanism for enforcing the Fourth Amendment.

381. As noted above, Fifth Amendment jurisprudence already has a similar two-tier system:

if the government agents want to use an individual’s statements against him in a criminal trial, it

must comply with Miranda and all the other Fifth Amendment requirements. If it has no

intention of using the information at trial, it need not comply with Miranda—but it is not

completely unregulated. See, e.g., Chavez v. Martinez, 538 U.S. 760, 773 (2003) (“Our views on

the proper scope of the Fifth Amendment’s Self-Incrimination Clause do not mean that police

torture or other abuse that results in a confession is constitutionally permissible so long as the

statements are not used at trial; it simply means that the Fourteenth Amendment’s Due Process

Clause, rather than the Fifth Amendment’s Self-Incrimination Clause, would govern the inquiry

in those cases and provide relief in appropriate circumstances.”).

382. Rochin v. California, 342 U.S. 165, 172, 174 (1952); see also Breithaupt v. Abram, 352

U.S. 432, 435 (1957) (holding that a blood test taken from a criminal defendant does not “shock

the conscience”). It is true that the Supreme Court has said that if police action is covered by

the Fourth Amendment, the Due Process Clause will not apply to the search. See, e.g., Graham

v. Connor, 490 U.S. 386, 394–95 (1989) (“Because the Fourth Amendment provides an explicit

textual source of constitutional protection against this sort of physically intrusive governmental

conduct, that Amendment, not the more generalized notion of ‘substantive due process,’ must

be the guide for analyzing these claims.”). A search that is truly a special needs search (in that

the fruits of the search are not used in a subsequent prosecution), however, could be considered

to be outside the scope of the Fourth Amendment and therefore covered by the Due Process

Clause.

383. If this standard is too low, or if the threat of civil lawsuits is so weak that it is

insufficient to deter abusive police behavior, there is nothing preventing cities, states, or the

federal government from passing legislation to raise the standards, or make lawsuits easier to

file, or to prohibit certain specific egregious practices. See generally Parry, supra note 3, at 819–

20 (discussing potential legislative or judicial responses to illegal police actions). Here is one

place in which political process theory would predict intervention by the legislature, especially if

the abusive practices were used during widespread, indiscriminate suspicionless searches (and if

they were not, the subjects of the search could conceivably file an equal protection claim).

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

924 DUKE LAW JOURNAL [Vol. 59:843

recovered in such a search in a subsequent criminal trial, the search

would not violate the Fourth Amendment. And because the pat-down

search did not shock the conscience, it would not violate the Due

Process Clause, and thus would not subject the police officer to any

civil liability.

If the police officers, however, sought to use the recovered

evidence in a subsequent criminal trial, the search would no longer be

considered a special needs search and would violate the Fourth

Amendment (leading to exclusion of the evidence and potential civil

liability against the officer). And if the police conducted a blatantly

overintrusive search—say, conducting a suspicionless strip search of

some of the passengers—a court could find that the search shocked

the conscience and violated the Due Process Clause, thus opening up

the officer to civil liability.

But of course fear of civil judgments would not be the real check

on police behavior. Just as under current law, the exclusionary rule

would provide a strong incentive for police officers to find some other

way to conduct the search that would allow them to use the evidence

in court.

Essentially, this proposal would give law enforcement officers a

choice: if their true goal was to prevent an armed criminal from

boarding a plane—or boarding a subway, or entering a public

courthouse, or entering a sports arena, or attending a protest, or

driving near a dam—they could set up a suspicionless search program

to detect any explosives or other weapons, ensuring that such

contraband is confiscated and the terrorist act is prevented. If their

goal is to apprehend the armed criminal and bring him to trial—in

other words, if they have a criminal law purpose—they would need to

design a search regime that comported with the usual Fourth

Amendment requirements, which would include some form of

individualized suspicion.

Thus, the rule would have several benefits. First, it would give

law enforcement the flexibility to decide which priority is more

important for the given context. If there is a credible threat that must

be responded to regarding, say, a specific trial or protest, law

enforcement officers could develop a suspicionless search regime,

emphasizing prevention at all costs and abandoning their crime

detection function. But in the absence of a credible threat, law

enforcement officers would decide to revert to legitimate surveillance

procedures, the results of which could be used in court. The decision

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 925

of which type of search to use would be made by the individuals in the

best position to make such a decision: law enforcement officers

themselves. And unlike the searches at airports and courthouses—

which became legally and thus culturally entrenched, and are now

permanently part of American life—intrusive security procedures

that are put into place because of the increased threat would only be

temporary, until the specific, credible threat receded and the police

decided to shift priorities to attempt to apprehend and prosecute

perpetrators.

Second, the rule would give law enforcement more powers to

search if it were truly intending only to prevent a terrorist attack.

They would no longer have to worry about whether their actions

complied with the inconsistent and ever-changing case law in this

area, and they would no longer have to prove that they had a specific

and concrete threat (and they would no longer need to guess about

what a judge might think that term means). But the ban on using the

fruits of these searches would be an automatic check against police

frequently resorting to overly intrusive techniques: the fact that the

contraband that is recovered cannot be used in subsequent criminal

proceedings would provide a strong incentive for law enforcement to

conduct this type of search sparingly. And, as discussed above, the

Equal Protection Clause and Due Process Clause would still apply,

precluding any discriminatory or extraordinarily intrusive searches.

Finally, this rule would provide police with a strong incentive to

develop more effective but less intrusive methods of antiterrorism

surveillance, such as machines that could detect the presence of

explosives or firearms without revealing any other information about

the suspect and without subjecting him to a physical search.

384

noted above, banning suspicionless searches entirely would provide

an even stronger incentive, but would also compromise security in

those situations in which there is a real threat of a terrorist attack.

384. These types of searches are known as binary searches because they are designed to only

tell a law enforcement officer whether or not illegal conduct is occurring, without other

information about the suspect—essentially producing an output of either “yes” or “no.”

Examples of binary searches currently in widespread use by law enforcement are drug-sniffing

dogs and narcotics field tests. These searches do not implicate the Fourth Amendment under

the Katz standard because they do not violate a reasonable expectation of privacy. See United

States v. Place, 462 U.S. 696, 707 (1983) (holding that an inspection by a narcotics dog trained to

sniff contraband without opening passengers’ luggage does not violate the passengers’ Fourth

Amendment rights). For a further discussion of the legality of binary searches, see Ric

Simmons, Technology-Enhanced Surveillance by Law Enforcement Officials, 60 N.Y.U.

ANN.

SURVEY AM. L. 711, 718–19 (2005).

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

926 DUKE LAW JOURNAL [Vol. 59:843

This rule still encourages creativity and innovation in developing less

intrusive surveillance tools and procedures, but would ensure that law

enforcement could still provide protection when necessary.

ONCLUSION

When antiterrorism searches first began, they no doubt seemed

reasonable to the public and to the courts. In the late 1960s, air piracy

and domestic terrorism were seen as real threats to the United States’

security, and law enforcement seemed powerless to combat these

threats through ordinary means. Instituting a regime of blanket

suspicionless searches was the only effective solution—and the

searches were indeed effective. Once the problem was solved, why

would anyone want to dismantle the search regimes and revive the

old dangers and risks?

Likewise, the administrative search doctrine began innocently

enough. Health inspectors had an important job to do, and they could

not do it effectively if they needed to generate individualized

suspicion before conducting their inspections. And because their

inspections were far removed from the harm the Fourth Amendment

was meant to prevent, allowing suspicionless searches for purely

regulatory purposes seemed harmless—a thoughtful compromise

based on the reasonableness language in the Fourth Amendment.

But in both instances, this Article demonstrates that these

reasonable compromises have grown into an inconsistent tangle of

case law, justified by a broad Fourth Amendment loophole whose

premise—that detecting and preventing violent crime is not a law

enforcement purpose—borders on the absurd. It is long past time to

restore logic and principle to this area of the law, and the new wave of

suspicionless antiterrorism searches presents courts with the perfect

opportunity to do so. Precedent already exists in the special needs

doctrine—numerous drug testing cases such as Acton and Von Raab

look to how the fruits of the search are utilized as part of the test to

determine whether the search was conducted for a law enforcement

purpose. The obstacles are the thirty-five-year-old precedents,

which—against all logic—held that searching airline passengers and

courthouse visitors for weapons was a regulatory function.

These obstacles are all the more daunting because the searches

that they affirm are now a generally accepted—and even welcome—

part of everyday life. But if they are left undisturbed, the government

SIMMONS IN FINAL.DOC 1/5/2010 2:12:49 AM

2010] SEARCHING FOR TERRORISTS 927

will continue using them to justify antiterrorism suspicionless searches

in dozens of other contexts, and courts will be forced to approve of

them all or to manufacture artificial distinctions between cases,

creating an inconsistency and unpredictability that will inevitably

leach into other special needs cases. By excluding the fruits of these

searches from future criminal prosecutions, the courts will provide a

simple, principled distinction between legitimate and illegitimate

suspicionless searches—and their claims that these searches fulfill a

special need other than law enforcement will actually be true.