The Normative Fourth Amendment

University of Minnesota Law School University of Minnesota Law School

Scholarship Repository Scholarship Repository

Minnesota Law Review

2019

The Normative Fourth Amendment The Normative Fourth Amendment

Matthew Tokson

Follow this and additional works at: https://scholarship.law.umn.edu/mlr

Part of the Law Commons

Recommended Citation Recommended Citation

Tokson, Matthew, "The Normative Fourth Amendment" (2019).

Minnesota Law Review

. 3241.

https://scholarship.law.umn.edu/mlr/3241

This Article is brought to you for free and open access by the University of Minnesota Law School. It has been

accepted for inclusion in Minnesota Law Review collection by an authorized administrator of the Scholarship

Repository. For more information, please contact [email protected].

741

Article

The Normative Fourth Amendment

Matthew Tokson

†

INTRODUCTION

The concept of a Fourth Amendment “search” is important

to both law enforcement officers and the citizens they may sur-

veil. The Amendment classically requires officers to obtain a

warrant before engaging in a search,

and even the exceptions to

this rule typically demand probable cause.

By contrast, when

an investigative practice is not a search, the government can use

it to investigate any citizen without meaningful constitutional

regulation.

Yet the definition of a “search” has changed dramatically

over time and remains contested today. Currently, searches are

largely defined by the Katz test, which looks to whether a person

had a “reasonable expectation of privacy” in the thing searched.

†

Associate Professor, University of Utah S.J. Quinney College of Law.

Thanks to Shima Baradaran Baughman, Jeffrey Bellin, Paul Cassell, Dan Epps,

Chad Flanders, Aya Gruber, Eve Hanan, Cathy Hwang, Margot Kaminsky,

Orin Kerr, Michael Mannheimer, Sandra Mayson, Cliff Rosky, Stephen Ross,

Christopher Slobogin, Lawrence Solum, James Stern, Lior Strahilevitz, and all

workshop participants at Vanderbilt, Ohio State, Utah, the Ruth Bader Gins-

burg Clerks in Academia Workshop, the Rocky Mountain Junior Scholars Work-

shop, and the CrimFest Conference for helpful comments and suggestions. Spe-

cial thanks to Christian Clark and Jennifer Joslin for excellent research

. See, e.g., Mitchell v. Wisconsin, 139 S. Ct. 2525, 2533 (2019) (asserting

that the Fourth Amendment “normally requires a warrant for a lawful search”).

. See, e.g., United States v. Ross, 465 U.S. 798, 799 (1982) (“[A] warrant-

less search of an automobile stopped by police officers who had probable cause

to believe the vehicle contained contraband was not unreasonable within the

meaning of the Fourth Amendment.” (citing Carroll v. United States, 267 U.S.

132 (1925))).

. For example, the government lawfully gathered millions of citizens’ di-

aled phone numbers following Smith v. Maryland, 442 U.S. 735, 745–46 (1979),

which held that obtaining such information was not a search.

. E.g., New York v. Class, 475 U.S. 106, 112 (1986) (quoting Katz v.

742 MINNESOTA LAW REVIEW [104:741

This expectations-based test expanded the scope of the Fourth

Amendment beyond physical things and helped solve the prob-

lem of rampant government wiretapping in the mid-twentieth

century.

But it has given rise to a host of new problems and has

become one of the most widely disparaged tests in all of Ameri-

can law.

The test is tautological,

incoherent,

ignores im-

portant Fourth Amendment values,

gives judges free reign to

impose their policy preferences,

and, as a practical matter, is

notoriously unhelpful.

It has failed to protect privacy in many

digital forms of information,

will shrink the Fourth Amend-

ment’s scope as knowledge of privacy threats increases,

and is

increasingly useless in the Internet age.

These problems stem

from a core deficiency: societal expectations are difficult to assess

and offer a shaky foundation for the Fourth Amendment’s pro-

tections. Katz, in short, is poorly suited to regulating government

United States, 389 U.S. 347, 360 (1867) (Harlan, J., concurring)). Investigations

involving the physical touching of property for information-gathering purposes

typically require a warrant under a new and evolving sub-rule. See Florida v.

Jardines, 569 U.S. 1, 7–10 (2013); United States v. Jones, 565 U.S. 400, 404–06

(2012).

. Matthew Tokson, Automation and the Fourth Amendment, 96 IOWA L.

REV. 581, 583–84, 584 n.13 (2011).

. See, e.g., Anthony G. Amsterdam, Perspectives on the Fourth Amend-

ment, 58 MINN. L. REV. 349, 384 (1974); Craig M. Bradley, Two Models of the

Fourth Amendment, 83 MICH. L. REV. 1468, 1468 (1985); Morgan Cloud, Rube

Goldberg Meets the Constitution: The Supreme Court, Technology and the

Fourth Amendment, 72 MISS. L.J. 5, 28–29 (2002); Sherry F. Colb, What Is a

Search? Two Conceptual Flaws in Fourth Amendment Doctrine and Some Hints

of a Remedy, 55 STAN. L. REV. 119, 120–21 (2002); Amitai Etzioni, Eight Nails

into Katz’s Coffin, 65 CASE WESTERN L. REV. 413, 413 (2014); Jed Rubenfeld,

The End of Privacy, 61 STAN. L. REV. 101, 103 (2008); Scott E. Sundby, “Every-

man” ’s Fourth Amendment: Privacy or Mutual Trust Between Government and

Citizen?, 94 COLUM. L. REV. 1751, 1791 (1994).

. William Baude & James Y. Stern, The Positive Law Model of the Fourth

Amendment, 129 HARV. L. REV. 1821, 1824–25 (2016).

. Daniel J. Solove, Fourth Amendment Pragmatism, 51 B.C. L. REV. 1511,

1511 (2010).

. William J. Stuntz, Privacy’s Problem and the Law of Criminal Proce-

dure, 93 MICH. L. REV. 1016, 1021 (1995).

. Minnesota v. Carter, 525 U.S. 83, 97 (1998) (Scalia, J., concurring).

. Id.; see Solove, supra note 8, at 1522–24.

. E.g., Colb, supra note 6, at 132–39; Etzioni, supra note 6, at 421–22.

. Matthew Tokson, Knowledge and Fourth Amendment Privacy, 111 NW.

U. L. REV. 139, 187 (2016).

. Paul Ohm, The Fourth Amendment in a World Without Privacy, 81

MISS. L.J. 1309, 1325–26 (2012).

2019] NORMATIVE FOURTH AMENDMENT 743

surveillance in the modern world. The Supreme Court itself has

begun to recognize the deficiencies of the current regime, holding

in Carpenter v. United States that the Fourth Amendment pro-

tects against cell phone location tracking despite the fact that

cell phone location data is not “private” and is exposed to third-

party companies.

As the Court starts to move beyond the stric-

tures of the Katz test, the time is right to rethink how the Fourth

Amendment applies to modern surveillance practices.

But while critiques of the Katz test are legion, concrete al-

ternatives are rare. There is a growing recognition that the ques-

tion of the Fourth Amendment’s scope is inescapably normative;

it requires courts to make a value judgment about when the

Fourth Amendment should protect citizens’ privacy rather than

simply determining whether citizens generally expect privacy.

A number of scholars have accordingly argued that courts should

take a more normative approach to the Fourth Amendment.

But little progress has been made towards developing an actual

normative test, beyond simply calling for courts to create one.

This Article takes a different approach. It develops a spe-

cific, detailed normative model for determining the scope of the

Fourth Amendment. The model is grounded in contextual theo-

ries of surveillance, which focus on the specific activities and

communications that surveillance disrupts. It draws on Fourth

Amendment precedents that reflect many of the same concerns,

which are sometimes lost in the futile search for societal expec-

tations. And it addresses a literature that has received relatively

little attention in Fourth Amendment scholarship, encompass-

ing numerous studies of the measurable harms of surveillance to

its targets.

. 138 S. Ct. 2206, 2217–20 (2018).

. Note that the terms “citizens” or “people” used below encompass resi-

dent aliens, although the Supreme Court has not directly ruled that the Fourth

Amendment applies to such persons. Cf. United States v. Verdugo-Urquidez,

494 U.S. 259, 271 (1990) (holding that the Fourth Amendment does not apply

to nonresident aliens but noting that similar protections apply to residents);

INS v. Lopez-Mendoza, 468 U.S. 1032, 1050 (1984) (assuming without deciding

that the Fourth Amendment applied to an undocumented immigrant present in

the United States).

. See infra Part I.A.

. See infra Part I.B.2.c.

744 MINNESOTA LAW REVIEW [104:741

Drawing on these sources, the normative model breaks out

surveillance harms into three categories: (1) avoidance of activi-

ties because of fear of surveillance; (2) harm to relationships and

communications; and (3) direct psychological or physical harm.

These harms are measurable and often well-documented.

Yet

they are also easier for judges to intuit in difficult cases than

concepts like societal knowledge or expectations.

On the other side of the balance are the benefits of crime

detection and prevention. This inquiry would consider, for in-

stance, whether a surveillance technique would primarily be

used in the early stages of an investigation in order to build prob-

able cause, and whether it would be likely to reveal criminal ac-

tivity that would otherwise be impossible to detect.

A norma-

tive test would also examine whether the same information

might be obtained through less invasive means.

Considering

these factors, if a surveillance practice causes harms to individ-

uals that outweigh the benefits from enhanced law enforcement,

courts should hold that the Fourth Amendment requires the po-

lice to obtain a warrant, or satisfy an exception to the warrant

requirement, before conducting the surveillance.

The goal of the proposal is to move past mere critique of the

Katz test and towards formulating a workable replacement, one

that is better able to address the ever-changing landscape of

modern surveillance. Like any legal regime, the normative

model is hardly perfect, and potential objections to it are ad-

dressed in detail below.

But there are numerous theoretical

and practical reasons to favor a normative approach.

A norma-

tive balancing test reflects the values at the heart of Fourth

Amendment jurisprudence more fully and effectively than other

approaches. It is likewise consistent with the text and history of

the Fourth Amendment.

Indeed, both the leading originalist

interpretation of the Amendment and less formalist theories of

construction support a balancing approach to the crucial ques-

. See, e.g., infra text accompanying notes 128–33.

. See, e.g., infra text accompanying note 95.

. See infra Part I.B.1.

. See infra Part I.B.3.

. See infra Part IV.

. See infra Part III.

. See infra Part II.

2019] NORMATIVE FOURTH AMENDMENT 745

tion of when the government can engage in suspicionless surveil-

lance.

The functional advantages of the normative test are sub-

stantial, and arguably essential, for addressing modern surveil-

lance practices. The test is, for example, adaptable to new sur-

veillance technologies and new social contexts. It takes into

account harms that other approaches ignore, including coercion

and discrimination. It is far better suited to addressing program-

matic surveillance and data analysis. And it directly considers

the normative values at stake in Fourth Amendment cases,

avoiding the false targets and arbitrariness of alternative tests.

Moreover, the test can be applied to a variety of Fourth

Amendment questions that courts and scholars struggle with

under current law. It can offer clear answers in frontier cases

such as those involving internet browsing data, smart home

technology, or email content. The normative approach can also

help rehabilitate some widely criticized cases that have plausible

outcomes but dubious reasoning. Finally, the test can help iden-

tify flawed cases that are ripe for reversal, where the normative

balance tilts sharply in favor of privacy or surveillance but cur-

rent law leads courts to the opposite outcome.

The Article proceeds in five Parts. Part I describes the nor-

mative model in detail and traces its lineage in Fourth Amend-

ment precedent and surveillance theory. Part II discusses the

textual, historical, and theoretical foundations of a balancing ap-

proach to the Fourth Amendment’s scope. Part III examines the

many practical advantages of the normative approach. Part IV

addresses potential objections to the normative test and to bal-

ancing tests in general. It also examines an alternative approach

that looks to positive law as the basis for the Fourth Amend-

ment’s protections. Part V applies the normative model to re-

solve frontier cases, provide firmer support for poorly reasoned

cases, and identify deeply flawed cases suitable for reversal.

. See infra Part II.B.

. Additional advantages are discussed infra Part III.

. See infra Part V.

746 MINNESOTA LAW REVIEW [104:741

I. TOWARDS A NEW MODEL OF THE FOURTH

AMENDMENT

A. THE KATZ TEST AND THE NEED FOR NORMATIVITY

The Supreme Court established that a Fourth Amendment

search occurs when a government act violates an individual’s

“reasonable expectation of privacy.”

This standard derives

from Justice Harlan’s solo concurrence in the 1967 case Katz v.

United States.

The Court has largely failed to clarify what

makes an expectation of privacy “reasonable,” and the rationales

of its Katz cases are often contradictory.

In some cases, the

Court looks primarily to the probability of detection by the po-

lice, while in others it looks to positive law or amorphous policy

considerations.

The law of the Katz test has, to date, been

largely case-dependent and unpredictable.

. E.g., Oliver v. United States, 466 U.S. 170, 178 (1984). The Supreme

Court has recently adopted a sub-test that finds a Fourth Amendment search

when a government official physically intrudes on property for the purposes of

gathering information. See Florida v. Jardines, 569 U.S. 1, 7–10 (2013); United

States v. Jones, 565 U.S. 400, 404–06 (2012). This has, thus far, added little to

the Katz test, and the Supreme Court cases where it has been employed would

likely have reached the same outcome under Katz. Jardines, 569 U.S. at 12–16

(Kagan, J., concurring); Jones, 565 U.S. at 418–31 (Alito, J., concurring in judg-

ment). It has also rapidly become confusing and difficult to apply, as the Court

has had to determine the extent of an implied social license to enter the curti-

lage of a home—a question bound up in a social norms inquiry even more amor-

phous and confusing than the Katz test. Jardines, 569 U.S. at 10; George M.

Dery III, Failing to Keep “Easy Cases Easy”: Florida v. Jardines Refuses to Rec-

oncile Inconsistencies in Fourth Amendment Privacy Law by Instead Focusing

on Physical Trespass, 47 LOY. L.A. L. REV. 451, 471–79 (2014).

. 389 U.S. 347, 361 (1967) (Harlan, J., concurring). This approach was

quickly adopted by lower courts and the Supreme Court as the definitive test.

E.g., Terry v. Ohio, 392 U.S. 1, 9 (1968) (“We have recently held that . . . wher-

ever an individual may harbor a reasonable ‘expectation of privacy,’ he is enti-

tled to be free from unreasonable government intrusion.” (quoting Katz, 389

U.S. at 361 (Harlan, J., concurring))); United States v. Guadalupe-Garza, 421

F.2d 876, 878 (9th Cir. 1970) (considering whether defendant had a “reasonable

‘expectation of privacy’” when crossing the border from Mexico to California

(quoting Terry, 392 U.S. at 9)).

. E.g., Orin S. Kerr, Four Models of Fourth Amendment Protection, 60

STAN. L. REV. 503, 504 (2007); Matthew Tokson, The Emerging Principles of

Fourth Amendment Privacy, 88 GEO. WASH. L. REV. (forthcoming 2020).

. Kerr, supra note 32, at 50722.

. See Ronald J. Allen & Ross M. Rosenberg, The Fourth Amendment and

the Limits of Theory: Local Versus General Theoretical Knowledge, 72 ST. JOHNS

L. REV. 1149, 115358, 1166 (1998).

2019] NORMATIVE FOURTH AMENDMENT 747

Criticism of the Katz test began not long after its adoption

and has only grown in volume and intensity over the years.

Critics argue that a test based on expectations is unworkable

and tautological.

They note the potential for circularity, as so-

cietal expectations about privacy may be shaped by government

practices and judicial decisions.

They point out that courts are

poorly situated to assess societal views about privacy.

Moreo-

ver, an expectations-based Fourth Amendment will shrink over

time as knowledge of privacy threats increases.

For decades, and increasingly often in recent years, scholars

have called upon courts to take a more normative approach.

Such an approach would focus on the level of privacy that citi-

zens should have rather than how much privacy they expect.

Calls for a normative approach to the Fourth Amendment

sometimes follow broad critiques of the Katz test,

but they also

arise in narrower works examining new surveillance technolo-

gies.

These analyses are generally insightful. Yet these diver-

gent writings share a profound humility regarding the content

. See, e.g., Joseph D. Grano, Foreword, Perplexing Questions About Three

Basic Fourth Amendment Issues: Fourth Amendment Activity, Probable Cause,

and the Warrant Requirement, 69 J. CRIM. L. & CRIMINOLOGY 425, 429 (1978);

sources cited supra note 6.

. Baude & Stern, supra note 7, at 1824–25.

. E.g., Rubenfeld, supra note 6, at 13233.

. Solove, supra note 8, at 152122.

. E.g., Tokson, supra note 13, at 187.

. See, e.g., Catherine Hancock, Warrants for Wearing a Wire: Fourth

Amendment Privacy and Justice Harlan’s Dissent in United States v. White, 79

MISS. L.J. 35, 3638 (2009); Justin Holbrook, Communications Privacy in the

Military, 25 BERKELEY TECH. L.J. 831, 903 (2010); Neil Richards, The Third-

Party Doctrine and the Future of the Cloud, 94 WASH. U. L. REV. 1441, 148788

(2017); Gavin Skok, Establishing a Legitimate Expectation of Privacy in Click-

stream Data, 6 MICH. TELECOMM. & TECH. L. REV. 61, 82–83 (2000); Olivier

Sylvain, Failing Expectations: Fourth Amendment Doctrine in the Era of Total

Surveillance, 49 WAKE FOREST L. REV. 485, 522 (2014); James J. Tomkovicz,

Beyond Secrecy for Secrecy’s Sake: Toward an Expanded Vision of the Fourth

Amendment Privacy Province, 36 HASTINGS L.J. 645, 698 (1985).

. E.g., Aya Gruber, Garbage Pails and Puppy Dog Tails: Is That What

Katz Is Made of?, 41 U.C. DAVIS L. REV. 781, 795 (2008) (“At some level the

constitutional inquiry must concern not just what society actually believes is

private, but what we ought to be able to regard as private . . . .”).

. See, e.g., Tomkovicz, supra note 40, at 698.

. See, e.g., Skok, supra note 40, at 8283. Justice Harlan himself called

for a more normative approach, repudiating in part the Katz test that he had

created, in a case involving an undercover government agent’s recording of a

748 MINNESOTA LAW REVIEW [104:741

of a normative test. They note “[t]he difficulty [in] determining

the right normative formula,”

clarify that the general norma-

tive approach they favor is “fact-driven” and imprecise,

or ex-

plain that “[i]n this initial effort it would be futile to attempt to

provide closure on the subject of possible grounds” for a norma-

tive test.

More commonly, they simply urge courts to take a

normative approach and reach the correct results in various

cases, without explaining what such an approach would entail.

A few scholars have taken a descriptive approach, examining

federal and state cases post-Katz and identifying factors that

seem to correlate with Fourth Amendment violations (such as

intrusiveness) or that are generally relevant to privacy (such as

the nature of the information sought).

But these correlates

have not yielded a test, except perhaps a “totality of the circum-

stances” test that directs courts to weigh any relevant normative

considerations and reach the best outcome.

conversation. United States v. White, 401 U.S. 745, 768 (1971) (Harlan, J., dis-

senting).

. Gruber, supra note 41, at 836.

. Mary Graw Leary, Reasonable Expectations of Privacy for Youth in a

Digital Age, 80 MISS. L.J. 1035, 1091 (2011).

. Tomkovicz, supra note 40, at 703.

. See, e.g., Skok, supra note 40, at 8283; Sylvain, supra note 40, at 522.

. See, e.g., Susan Freiwald, First Principles of Communications Privacy,

2007 STAN. TECH. L. REV. 3 (2007) (discussing intrusive searches); Melvin Gut-

terman, A Formulation of the Value and Means Models of the Fourth Amend-

ment in the Age of Technologically Enhanced Surveillance, 39 SYRACUSE L. REV.

647, 722–23 (1988) (discussing generalities relevant to privacy); Stephen E.

Henderson, Beyond the (Current) Fourth Amendment: Protecting Third-Party

Information, Third Parties, and the Rest of Us Too, 34 PEPP. L. REV. 975, 985–

1014 (2007) (listing considerations relevant to privacy).

. Henderson, supra note 48, at 985–1014, 1025 (noting several nondis-

positive considerations relevant to privacy and affirming the importance of a

“totality of the circumstances” approach to the Fourth Amendment). Paul Ohm

has described Carpenter v. United States as radically changing the Katz test

itself and virtually replacing it with the standard for cell phone data set out in

Carpenter, which looks to the “the deeply revealing nature of [cell phone data],

its depth, breadth, and comprehensive reach, and the inescapable and auto-

matic nature of its collection.” Paul Ohm, The Many Revolutions of Carpenter,

32 HARV. J.L. & TECH. 357, 361–63 (2019); see Carpenter v. United States, 138

S. Ct. 2206, 2217–18, 2223 (2018). Even assuming that this standard is now

controlling in the third-party doctrine context, it is unlikely that the Court in-

tended it to modify Katz. Indeed, the Court took pains to avoid providing any

guidance on future Fourth Amendment issues, emphasizing that “[o]ur decision

today is a narrow one” and listing several Fourth Amendment issues (including

those closely related to historical cell phone data) on which the Court expressed

2019] NORMATIVE FOURTH AMENDMENT 749

What explains the reluctance to specify how courts should

normatively determine the scope of the Fourth Amendment?

One of the earliest and most illuminating calls for a normative

approach, from James Tomkovicz’s 1985 article, suggests that

the difficulty of formulating a normative test stems in part from

the difficulty of conceptualizing the harms that government sur-

veillance can cause.

Tomkovicz offers no test and notes that

there are “no ready guides” for value judgments regarding citi-

zens’ privacy, but posits that as theories of privacy and related

constitutional values develop, courts could incorporate their con-

clusions into a normative approach.

Several decades later, the time has come to incorporate the

insights of privacy and surveillance theory into a concrete

Fourth Amendment test. Such theory has made enormous pro-

gress over the past thirty years and in a variety of fields, includ-

ing law, sociology, philosophy, and information science. Among

other developments, privacy theory has largely shifted from

identifying abstract principles of privacy towards focusing on the

specific practices, communications, and freedoms that privacy

enables.

Scholars have offered various general theories of privacy, in-

cluding privacy as control over information,

limited exposure

no opinion. Carpenter, 138 S. Ct. at 2220. Still, Ohm’s point is well taken that

Carpenter might serve as a basis for a rethinking of the Katz test. See Ohm,

supra, at 36163. I have elsewhere argued that the Carpenter and United States

v. Jones opinions reflect the Court’s recognition of factors that have long dic-

tated its application of Katz. See Tokson, supra note 32, at 18–20.

. Tomkovicz, supra note 40, at 70102.

. Id. at 70103.

. E.g., ALAN F. WESTIN, PRIVACY AND FREEDOM 7 (1967) (“Privacy is the

claim of individuals . . . to determine for themselves when, how, and to what

extent information about them is communicated to others.”); Charles Fried, Pri-

vacy, 77 YALE L.J. 475, 482 (1968) (“Privacy is not simply an absence of infor-

mation about us in the minds of others; rather it is the control we have over

information about ourselves.”).

750 MINNESOTA LAW REVIEW [104:741

to others,

intimacy,

bodily integrity,

and as a precondition

to self-development.

Yet theorists have increasingly recognized

that the meaning of privacy is rarely fixed or universal, and its

value often depends on the social contexts in which it can protect

individuals from coercion, condemnation, and other harms.

social practices and norms change, different aspects of privacy

can become more or less important. For instance, control over

data may be increasingly important in the Internet era, while

limiting exposure to others may be less of a concern in an age of

larger houses and increasing social isolation. Moreover, some as-

pects of privacy may be crucial in some contexts and irrelevant

in others.

In order to develop a more complete account of privacy

harm, theories of contextual privacy have looked to the norms

that govern information exchange in a wide variety of social con-

texts and relationships.

When people offer their information in

a certain context, the exchange of information is generally gov-

erned by implicit agreements regarding its use.

These agree-

ments and norms might dictate, for instance, that the parties

. E.g., SISSELA BOK, SECRETS: ON THE ETHICS OF CONCEALMENT AND

REVELATION 10–11 (1983) (“[P]rivacy [is] the condition of being protected from

unwanted access by others . . . .”); Ruth Gavison, Privacy and the Limits of Law,

89 YALE L.J. 421, 423 (1980).

. E.g., JULIE C. INNESS, PRIVACY, INTIMACY, AND ISOLATION 140 (1992)

(“[P]rivacy is the state of possessing control over a realm of intimate deci-

sions . . . .”); Tom Gerety, Redefining Privacy, 12 HARV. CIV. RTS.-CIV. LIBER-

TIES L. REV. 233, 268 (1977).

. E.g., Richard B. Parker, A Definition of Privacy, 27 RUTGERS L. REV.

275, 283–84 (1974).

. E.g., Anita L. Allen, Coercing Privacy, 40 WM. & MARY L. REV. 723,

739–40 (1999); Julie E. Cohen, Examined Lives: Informational Privacy and the

Subject as Object, 52 STAN. L. REV. 1373, 1377 (2000); Paul M. Schwartz, Pri-

vacy and Democracy in Cyberspace, 52 VAND. L. REV. 1609, 1653 (1999).

. See, e.g., HELEN NISSENBAUM, PRIVACY IN CONTEXT: TECHNOLOGY,

POLICY, AND THE INTEGRITY OF SOCIAL LIFE 8089 (2010); Adam Barth et al.,

Privacy and Contextual Integrity: Framework and Applications, 2006 IEEE

SYMP. ON SECURITY & PRIVACY 184, 184–86 (2006).

. For example, keeping one’s marital or health status private may be im-

portant in the context of a job interview but unimportant in the context of social

interactions with friends or a conversation with a doctor. See NISSENBAUM, su-

pra note 57, at 143–44.

. Id.

. Id. at 124–25.

2019] NORMATIVE FOURTH AMENDMENT 751

restrict further information flow or maintain anonymity by de-

clining to link the data with personally identifiable infor-

mation.

Violations of these context-dependent norms lead to

identifiable harms, as parties’ preferences are ignored and their

interests adversely affected.

Thus a clinical worker who dis-

closed a patient’s treatment for addiction would violate norms of

behavior specific to the treatment context, causing harms to the

patient’s reputation, psychological well-being, employment pro-

spects, etc.

Relatedly, pragmatic privacy theories focus on how the lack

of privacy deters and interrupts specific social and personal prac-

tices.

They posit that the value of privacy depends on the prac-

tices that it protects, which include activities as varied as politi-

cal activism, shopping, communication, research, nudity, and

intimacy.

Likewise, the concept of intellectual privacy has

called attention to the importance of privacy to expressive activ-

ities, personal communications, and freedom of thought itself.

It reveals a particularly important set of practices and cognition

that surveillance has the potential to disrupt. These and other

recent theoretical movements offer a deeper, more specific, and

more practical understanding of the harms of surveillance. Their

insights can help provide a foundation for a workable normative

approach to the Fourth Amendment.

This Article’s analysis of the harms of government surveil-

lance can help to further develop and refine contextual and prag-

matic privacy theories. The Article examines in detail a particu-

larly important privacy context: surveillance by police or other

government officials of private citizens. It identifies the most

fundamental disruptions and harms caused by such surveil-

lance. More broadly, the Article develops an analytical approach

that can be used to evaluate private intrusions and government

surveillance alike.

. See id. at 186–87.

. See id. at 212.

. See Daniel J. Solove, Conceptualizing Privacy, 90 CAL. L. REV. 1087,

112632 (2002).

. Id. at 1143, 114654.

. See Neil M. Richards, Intellectual Privacy, 87 TEX. L. REV. 387, 41226

(2008).

752 MINNESOTA LAW REVIEW [104:741

The following Sections propose a concrete, normative test for

the Fourth Amendment’s scope and trace the lineage of each fac-

tor of the test in surveillance theory, constitutional practice, or

both. Part II then discusses the test’s doctrinal, historical, and

theoretical foundations.

B. A NORMATIVE TEST

An effective normative test for the Fourth Amendment’s

scope would balance the benefits of warrantless government sur-

veillance against its costs. However, a test that merely directs

courts to weigh all benefits to law enforcement against all harms

to citizens is not sufficiently detailed or rigorous. Such a stand-

ard would require each individual court to determine how best

to theorize and assess the various harms of surveillance, likely

resulting in extreme inconsistency and prohibitively high deci-

sion costs.

Courts require a more concrete, workable test. But, if it is to

reflect the normative balance inherent in the Fourth Amend-

ment, such a test must also incorporate essential categories of

law enforcement benefit and social harm. The following proposal

attempts to fulfill these goals and strike a middle ground be-

tween including important categories of surveillance harm and

remaining concise. Its aim is not only to offer a workable test,

but to shift the focus of Fourth Amendment debate away from

criticisms of Katz and toward actual alternatives.

The test can be described as follows:

The normative test asks whether a surveillance practice’s

value to law enforcement in terms of crime detection and preven-

tion outweighs three fundamental harms: (1) the avoidance of

lawful activity because of fear of surveillance; (2) the harm to

relationships and communications caused by observation; and

(3) the concrete psychological or physical harm suffered due to

surveillance. The test then asks whether the same law enforce-

ment goals could be achieved via a less invasive practice. If, con-

sidering these factors, the total harm to citizens from a type of

surveillance outweighs the total benefit from enhanced law en-

forcement, courts should hold that the Fourth Amendment re-

quires police to obtain a warrant, or to satisfy an exception to the

warrant requirement, before conducting the surveillance. If the

benefit to law enforcement outweighs the harm, then the police

2019] NORMATIVE FOURTH AMENDMENT 753

should be able to conduct the surveillance without Fourth

Amendment regulation.

These three categories of harm are derived not only from

basic Fourth Amendment ideals like privacy, liberty, and secu-

rity, but also from a consideration of the functional and practical

values these ideals protect.

The Fourth Amendment is de-

signed to prevent arbitrary government surveillance,

a valua-

ble goal not only in itself but also because such surveillance pre-

vents us from acting freely, stifles our relationships and freedom

of association, and does harm to us both as individuals and as

citizens of a democracy. These practical values are embodied in

the proposed test. Each of the factors has a basis in existing

Fourth Amendment jurisprudence, well-developed theories of

privacy and police coercion, or both. The following Sections dis-

cuss the factors in more detail, as well as their doctrinal and the-

oretical foundations.

1. Crime Detection and Prevention

The first factor of the test examines a warrantless surveil-

lance practice’s benefits to law enforcement, which can primarily

be expressed in terms of enhanced crime detection and enhanced

deterrence.

Because detection and prevention are generally

linked, the test combines them in a single inquiry.

. For a discussion of historical Fourth Amendment ideals, see, for exam-

ple, Boyd v. United States, 116 U.S. 616, 630 (1886) (“It is not the breaking of

[a man’s] doors, and the rummaging of his drawers, that constitutes the essence

of the offense; but it is the invasion of his indefeasible right of personal security,

personal liberty, and private property . . . .”), abrogated by Warden, Md. Peni-

tentiary v. Hayden, 387 U.S. 294 (1967); Morgan Cloud, Searching Through

History; Searching for History, 63 U. CHI. L. REV. 1707, 1726 (1996) (“[T]he his-

torical record suggests that objections to general warrants and general searches

alike rested upon broad concerns about protecting privacy, property, and liberty

from unwarranted and unlimited intrusions.”).

. E.g., Thomas Y. Davies, Recovering the Original Fourth Amendment, 98

MICH. L. REV. 547, 74445 (1999).

. It would also encompass evidence collection for the purposes of convic-

tion, which would have benefits related to detection, deterrence, incapacitation,

and retribution.

. See, e.g., Daniel S. Nagin & Greg Pogarsky, Integrating Celerity, Impul-

sivity, and Extralegal Sanction Threats into a Model of General Deterrence: The-

ory and Evidence, 39 CRIMINOLOGY 865, 883–84 (2001) (studying drinking and

driving trends among college students and finding that the certainty of punish-

ment was a stronger deterrent than the severity of punishment). Courts might

optionally prefer to analyze these facts of law enforcement separately, breaking

754 MINNESOTA LAW REVIEW [104:741

This factor essentially asks, how valuable to law enforce-

ment would it be to be able to engage in a certain type of war-

rantless surveillance? A court might consider whether a surveil-

lance technique would primarily be used in the early stages of

investigations, before probable cause has been developed, and

whether the warrantless use of the technique would likely reveal

criminal activity that would otherwise go undetected.

For ex-

ample, if obtaining certain financial records without a warrant

would allow police to identify white collar crimes that would oth-

erwise be difficult to detect, that would weigh in favor of exclud-

ing such records from Fourth Amendment regulation.

Relat-

edly, courts could consider studies examining the effects of

limiting a particular surveillance technique. Research indicating

that limits on certain kinds of surveillance would reduce police

ability to build probable cause

or to deter certain crimes

may

help to quantify the value of the surveillance to law enforcement

this factor out into two separate factors on the law enforcement side of the bal-

ance.

. Courts could also consider relevant studies examining the effects of lim-

iting various surveillance techniques. One recent study, for instance, found that

subjecting telephone call logs to a warrant requirement resulted in fewer appli-

cations for wiretaps and a decrease in the duration of permitted wiretaps. ANNE

E. BOUSTEAD, POLICE, PROCESS, AND PRIVACY: THREE ESSAYS ON THE THIRD

PARTY DOCTRINE 18–20 (2016), https://www.rand.org/pubs/rgs_dissertations/

RGSD384.html [https://perma.cc/3KFX-U299]. Its findings suggest that regu-

lating the acquisition of call log data reduces police officers’ ability to obtain

sufficient probable cause for Wiretap Act applications. Id.

. See Orin S. Kerr, An Equilibrium-Adjustment Theory of the Fourth

Amendment, 125 HARV. L. REV. 476, 509 (2011) (explaining that the Supreme

Court eliminated the warrant requirement for financial records following the

rise of difficult-to-detect white-collar crimes); see also David Gray et al.,

Fighting Cybercrime After United States v. Jones, 103 J. CRIM. L. & CRIMINOL-

OGY 745, 777–78, 798 (2013) (discussing types of digital evidence that are espe-

cially helpful in detecting healthcare fraud and cyber harassment).

. See BOUSTEAD, supra note 70, at 1820.

. See Paul G. Cassell & Richard Fowles, What Caused the 2016 Chicago

Homicide Spike? An Empirical Examination of the ‘ACLU Effect’ and the Role

of Stop and Frisks in Preventing Gun Violence, 2018 U. ILL. L. REV. 1581, 1665

(2018) (noting an increase in gun violence in the year following the cessation of

programmatic stop-and-frisk searches in Chicago); see also Gary T. Marx, See-

ing Hazily (But Not Darkly) Through the Lens: Some Recent Empirical Studies

of Surveillance Technologies, 30 L. & SOC. INQUIRY 339, 348–49 (2005) (discuss-

ing the deterrent effects of video monitoring in interrogation rooms on violence

by both detainees and the police).

2019] NORMATIVE FOURTH AMENDMENT 755

goals. Reports issued by agencies tasked with independent eval-

uation, such as the Privacy and Civil Liberties Oversight Board,

may also be helpful in assessing law enforcement efficacy.

The consideration of law enforcement effectiveness is

grounded in Fourth Amendment caselaw, although courts’ treat-

ment of it has been haphazard and unstructured. The Supreme

Court has explicitly considered benefits to law enforcement in

cases concerning the Fourth Amendment’s scope,

and such

benefits implicitly justify the results in countless other scope

cases.

This consideration also helps determine the effective

scope of the Amendment by shaping and limiting its remedies.

. See Daphna Renan, The Fourth Amendment as Administrative Govern-

ance, 68 STAN. L. REV. 1039, 111821 (2016) (explaining that the Privacy and

Civil Liberties Oversight Board has access to classified information and can con-

sult a wide breath of institutional and public opinion).

. See Arizona v. Gant, 556 U.S. 332, 346–47 (2009) (mentioning the evi-

dentiary interests of the police as a justification for maintaining a broad scope

for the vehicular search incident to arrest doctrine); Hudson v. Palmer, 468 U.S.

517, 526–27 (1984) (discussing the need for decreased privacy rights for inmates

in their cells due to the importance of detecting inmate crimes in a prison set-

ting); United States v. Miller, 425 U.S. 435, 442–44 (1976) (citing 12 U.S.C.

§ 1829b(a)(1) (1970)) (noting the “high degree of usefulness in criminal tax, and

regulatory investigations and proceedings” of bank record availability for law

enforcement).

. See, e.g., Minnesota v. Carter, 525 U.S. 83, 9091 (1998) (holding that

the Fourth Amendment does not apply to temporary house guests who are not

personal friends of the homeowner, in a case involving a drug sale); California

v. Ciraolo, 476 U.S. 207, 213–14 (1986) (concluding that police officers are enti-

tled to view a house’s curtilage from any place where citizens can lawfully go,

including airspace); see also United States v. Mohamud, 843 F.3d 420, 43941

(9th Cir. 2016) (holding that no warrant is required to collect a U.S. citizen’s

emails to an overseas foreign national, when that foreign national is subject to

a lawful search, in a case involving allegations of terrorism).

. For example, the “good-faith exception” cases limit the application of

the exclusionary rule in large part because of the rule’s detrimental effects on

law enforcement and criminal deterrence. See, e.g., Pa. Bd. of Prob. & Parole v.

Scott, 524 U.S. 357, 36465 (1998) (internal quotation marks omitted) (“[T]he

exclusionary rule . . . allows many who would otherwise be incarcerated to es-

cape the consequences of their actions . . . . [T]he rule’s costly toll upon truth-

seeking and law enforcement objectives presents a high obstacle for those urg-

ing application of the rule.”). Courts grant qualified immunity to law enforce-

ment officers for violations of the Fourth Amendment for similar reasons. See,

e.g., Harlow v. Fitzgerald, 457 U.S. 800, 814 (1982) (justifying qualified immun-

ity in part on concerns that a lack of immunity would deter law enforcement

officers from performing their duties to the full extent).

756 MINNESOTA LAW REVIEW [104:741

At the level of theory, some concern for effective law enforce-

ment is inherent in the existence of criminal laws. The theoreti-

cal justifications for criminal law enforcement are largely iden-

tical to those that justify criminal laws and punishments—the

utilitarian benefits of deterrence, public safety, and rehabilita-

tion;

the deontological values of justice and retribution;

or a

pragmatic mixture of both.

Any normative balancing approach

to regulating law enforcement must take law enforcement effec-

tiveness into account.

2. Harms to Individuals

As discussed above, a workable normative test must capture

the most substantial harms caused by government surveillance

and be sufficiently administrable that judges can effectively ap-

ply the test.

Contextual and pragmatic theories of surveillance

point the way towards a test that can meet both needs. They fo-

cus on the particular practices and relationships disrupted by

surveillance.

This practical emphasis has several benefits.

First, it can unify various theories of privacy and other Fourth

Amendment values like liberty and trust by emphasizing their

shared practical concerns rather than their abstract theoretical

differences.

Second, the practical harms of surveillance are eas-

ier for judges to address than are esoteric theories of privacy or

trust.

The normative test proposed here combines a focus on dis-

rupted practices and relationships with another category of fun-

damental harms: measurable psychological or physical harms

suffered by the subjects of government investigations. By incor-

porating these factors, the test can capture the primary harms

. E.g., Joel Feinberg, The Classic Debate, in PHILOSOPHY OF LAW 727,

729 (Joel Feinberg & Jules Coleman eds., 6th ed. 2000).

. E.g., Michael T. Cahill, Retributive Justice in the Real World, 85 WASH.

U. L. REV. 815, 826–28 (2007) (discussing “desert-based punishment” and ne-

cessity of punishment for “those who deserve it”).

. E.g., Stephen P. Garvey, Lifting the Veil on Punishment, 7 BUFF. CRIM.

L. REV. 443, 449–50 (2004) (“The mixed theory . . . unites the purpose of utili-

tarianism with the limits of retribution.”).

. See supra Part I.B.

. See supra notes 6571 and accompanying text.

. See generally Solove, supra note 63 (discussing the theoretical differ-

ences between the leading privacy theories).

. Id. at 1090 (discussing the difficulties judges face when conceptualizing

privacy).

2019] NORMATIVE FOURTH AMENDMENT 757

to individuals from government surveillance without requiring

judges to grapple with abstract theories or societal expectations.

Although the test focuses on the pragmatic harms of surveil-

lance, its focus is necessarily broad, addressing the surveillance

technique used in the relevant case as a whole rather than in

isolation. It does so by hypothesizing that the surveillance tech-

nique has become widespread and well-known, and asking how

people’s behavior would change as a result. This comprehensive

approach is necessary for several reasons. First, a broad ap-

proach to the harms of surveillance is necessary to match the

broad consideration of law enforcement benefits. The Supreme

Court frequently considers the general benefits of surveillance

to law enforcement, benefits that go beyond those realized in the

instant case.

Courts should likewise consider the widespread

harms of surveillance when evaluating potential Fourth Amend-

ment searches.

Second, predicting the exact future prevalence

of a surveillance technique or determining the likely extent of

societal knowledge would be very difficult, especially for courts

addressing novel surveillance technologies.

Finally, a broad as-

sessment better aligns courts’ analyses with the potential conse-

quences of their decisions.

Fourth Amendment cases nearly al-

ways have broad implications. When a court rules that the police

may dig through one defendant’s trash bags without a warrant,

the police can thereafter dig through the trash bags of any per-

son in the court’s jurisdiction.

By assessing surveillance tech-

niques as a whole, the normative test appropriately focuses

courts’ attention on the actual impacts of their decisions.

. See supra notes 75, 77.

. See discussion infra Parts I.B.2.bc.

. See Tokson, supra note 13, at 16479 (discussing the difficulties of

measuring societal knowledge in even the most favorable circumstances).

. See, e.g., infra notes 23941 and accompanying text.

. See California v. Greenwood, 486 U.S. 35, 40 (1988). Resource con-

straints may prevent police departments from engaging in costly surveillance

on a grand scale. For lower-cost types of surveillance or for national security

matters, however, the government might actually surveil most or all citizens.

Thus, courts might safely assume that the use of a costly surveillance technique

would be less widespread than that of a cheap technique, potentially affecting

the extent of the harm caused. For a detailed argument regarding surveillance

costs and the importance of assessing surveillance technologies as a whole, see

David Gray & Danielle Citron, The Right to Quantitative Privacy, 98 MINN. L.

REV. 62, 10103 (2013).

758 MINNESOTA LAW REVIEW [104:741

a. Deterring Lawful Activities

The first harm factor asks whether a given type of surveil-

lance would cause people to avoid lawful activities. People en-

gage in all manner of potentially sensitive, embarrassing, or con-

troversial activities, like visiting a psychiatrist, researching

sensitive subjects online, purchasing certain drugs or medical

equipment, joining a substance-abuse support group, or criticiz-

ing government or social elites. These lawful activities can be

deterred by the threat of surveillance. For example, Google

searches for especially controversial or embarrassing terms de-

creased significantly following Edward Snowden’s disclosure of

an NSA program capable of capturing internet information.

Likewise, researchers documented a reduction in a wide variety

of religious and social activities at New York mosques due to in-

creased police surveillance after the September 11 attacks.

Courts may assess deterrence of lawful activities by using

studies that show reduced activity following increased aware-

ness of surveillance.

Empirical studies on chilling effects have

Further, courts applying a normative test would primarily focus on the do-

mestic law enforcement context but could also consider the domestic anti-ter-

rorism context if doing so is helpful. By contrast, foreign intelligence surveil-

lance may be exempt from the warrant requirement in any event, potentially

making the question whether such surveillance is a “search” irrelevant. See gen-

erally United States v. Truong Dinh Hung, 629 F.2d 908, 913–15 (4th Cir. 1980)

(concluding that “the courts should not require the executive to secure a warrant

each time it conducts foreign intelligence surveillance,” but noting that those

reasons do not justify warrantless domestic surveillance).

. See Alex Marthews & Catherine Tucker, Government Surveillance and

Internet Search Behavior 7, 3536 (Feb. 17, 2017) (unpublished manuscript),

https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2412564 (explaining em-

barrassing search terms were identified by poll respondents and included “abor-

tion,” “gender reassignment,” “police brutality,” and “tax avoidance”).

. See DIALA SHAMAS & NERMEEN ARASTU, MUSLIM AM. CIVIL LIBERTIES

COAL. ET AL., MAPPING MUSLIMS: NYPD SPYING AND ITS IMPACT ON AMERICAN

MUSLIMS 1215 (2013), http://www.law.cuny.edu/wp-content/uploads/page

-assets/academics/clinics/immigration/clear/Mapping-Muslims.pdf [https://

perma.cc/8PCY-8583].

. See, e.g., Marthews & Tucker, supra note 90; Jonathon W. Penney,

Chilling Effects: Online Surveillance and Wikipedia Use, 31 BERKELEY TECH.

L.J. 117, 14657 (2016) (finding that views of Wikipedia articles on sensitive

topics decreased significantly following the Snowden revelations); see also MI-

CHAEL MCCAHILL, THE SURVEILLANCE WEB: THE RISE OF VISUAL SURVEIL-

LANCE IN AN ENGLISH CITY 145 (2002) (discussing the effects of video monitoring

2019] NORMATIVE FOURTH AMENDMENT 759

become increasingly common in recent years.

Courts may also

rely on expert witnesses or amicus briefs from professional asso-

ciations noting the lawful activities that a type of surveillance

may discourage, as the Supreme Court did in Ferguson v. City of

Charleston.

Moreover, judges are likely to be able to assess deterrence of

lawful activities even in situations where there are no directly

relevant studies. Whether surveillance would deter a person

from engaging in lawful activities is a question that judges can

fruitfully address through reasoning and intuition: “if I were be-

ing surveilled by government agents using the technique at issue

in this case, would I be likely to forego certain activities?” For

example, a judge assessing long-term video monitoring by drones

might recognize that she would likely curtail her activities in

public and in the back yard of her home because of the monitor-

ing. This likely reduction in lawful activity would weigh in favor

of requiring a warrant for long-term drone surveillance. Judges

are likely to be more successful in forming intuitions about how

their own activities would be impacted by surveillance than

grappling with abstract theories of privacy or attempting to cal-

culate societal expectations.

on the social behavior of mall security guards); Darhl M. Pedersen, Psychologi-

cal Functions of Privacy, 17 J. ENVTL. PSYCHOL. 147, 15052 (1997) (presenting

survey results evaluating everyday activities that depend upon privacy).

. See, e.g., PEN AM. CTR, CHILLING EFFECTS: NSA SURVEILLANCE

DRIVES U.S. WRITERS TO SELF-CENSOR 3 (2013) (reporting that 28% of surveyed

writers had curtailed social media activities out of concerns about surveillance,

while 16% had avoided writing or speaking about certain subjects); Jonathon

W. Penney, Internet Surveillance, Regulation, and Chilling Effects Online: A

Comparative Case Study, 6 INTERNET POL’Y REV. 1, 1–3 (2017) (reporting sur-

vey evidence that government surveillance of the internet would reduce online

speech, make speakers more guarded in terms of the content of their online

speech, and chill online searching).

. 532 U.S. 67, 84 n.23 (2001) (noting that the American Medical Associa-

tion and other groups filing amicus briefs agreed that drug testing of pregnant

patients’ urine would deter women who use drugs from seeking prenatal care).

. One might object that judges applying the Katz test can already use

personal intuitions about whether they would expect privacy. Aside from the

myriad problems with using anyone’s expectations as a barometer for Fourth

Amendment protection, see supra notes 6–9 and accompanying text, judicial in-

tuition regarding privacy expectations is likely to be systematically biased

against privacy interests. See Tokson, supra note 13, at 17273 (describing var-

ious cognitive biases that may affect judicial intuitions about privacy expecta-

tions). Expectations of privacy are inextricably linked to knowledge regarding

surveillance and privacy threats. See id. at 149–50 (“What a person expects is

760 MINNESOTA LAW REVIEW [104:741

Judicial intuitions are, of course, not infallible and are sub-

ject to inaccuracy and bias.

Social science studies provide more

objective evidence but are likewise imperfect and prone to mis-

interpretation.

This Article does not argue that judges will em-

ploy either source of information perfectly. It does contend that

judicial intuition is better suited for assessing surveillance’s

dampening effects on activities and relationships than for intu-

iting the state of societal expectations of privacy.

Moreover,

there is an extensive social science literature on surveillance

harms that can aid judges in their assessments.

Courts are likely able to evaluate surveillance’s potential

impact on lawful activities—indeed, they have already done so

in several cases. In Zurcher v. Stanford Daily, the Court exam-

largely a function of what they know.”). Judges will generally have unusually

high levels of knowledge regarding the surveillance technique at issue—the par-

ties will have informed them at length about the technology in their pleadings

and briefs. Thus, they may expect less privacy in a given context than the vast

majority of people. Further, judges’ acquired knowledge is likely to bias their

intuitive judgments about societal knowledge in general. Individuals tend to

automatically and irrationally impute their own knowledge to other people,

even when those people are extremely unlikely to know it. See Boaz Keysar et

al., States of Affairs and States of Mind: The Effect of Knowledge of Beliefs, 64

ORGANIZATIONAL BEHAV. & HUM. DECISION PROCESSES 283, 284 (1995)

(“[O]nce people know what speakers intend, they believe that addressees will

perceive the same intention—even when addressees lack the crucial piece of

information which is necessary to understand the speakers’ intention.”).

. See Tokson, supra note 13, at 17273 (explaining that by the time

judges decide a case, they have so much knowledge about the issue they may

intuitively overestimate societal knowledge).

. See J. Alexander Tanford, The Limits of a Scientific Jurisprudence: The

Supreme Court and Psychology, 66 IND. L.J. 137, 145 (1990).

. See supra note 95.

. There is a smaller but growing collection of surveys about surveillance

and privacy expectations that can assist judges in assessing such expectations

under Katz. See Brief of Amici Curiae Empirical Fourth Amendment Scholars

in Support of Petitioner at 4–10, Carpenter v. United States, 138 S. Ct. 2206

(2018) (No. 16-402), 2018 WL 3073916 (discussing studies that ask respondents

about their expectations of privacy). Courts have thus far been reluctant to em-

ploy such data, and people’s reported expectations may not match their behavior

or may be more aspirational than actual. See Tokson, supra note 13, at 180.

Nonetheless, the use of empirical studies of societal expectations and knowledge

would likely improve the accuracy of courts’ decisions under the Katz test. Id.

However, the many conceptual flaws of the Katz test itself recommend aban-

doning the test even if courts were able to adjudicate it perfectly. See, e.g., id. at

18187.

2019] NORMATIVE FOURTH AMENDMENT 761

ined whether police searches of newspaper offices would inter-

fere with the newspaper’s operations, dissuade confidential

sources from coming forward, motivate editors to suppress con-

troversial news stories, or “intrude into or to deter normal edito-

rial and publication decisions.”

100

Likewise, in cases involving

searches and seizures of expressive materials, the Court has em-

phasized the need for the rigorous application of Fourth Amend-

ment protections to prevent the stifling of legitimate book distri-

bution or movie displays.

101

Nor has this principle been limited

to cases involving expressive activities. In Ferguson v. City of

Charleston, the Supreme Court held that a public hospital’s pro-

gram of drug testing pregnant women’s urine violated the

Fourth Amendment, noting that medical professionals appar-

ently agreed that such programs “discourag[ed] women who use

drugs from seeking prenatal care.”

102

A concern with the deterrence of legitimate activities also

has roots in pragmatic theories of privacy. Pragmatic theories

explicitly focus on concrete practices and conceive of privacy as

a constitutive part of such practices.

103

Accordingly, they define

privacy harms in terms of disruptions to practices.

104

In a similar

vein, the theory of intellectual privacy emphasizes surveillance’s

ability to chill activities of intellectual development and expres-

sion, from reading library books to web-surfing to writing and

speaking.

105

These theories provide a compelling account of the

potential chilling effects of surveillance and the value of privacy-

100

. 436 U.S. 547, 566 (1978).

101

. See, e.g., Roaden v. Kentucky, 413 U.S. 496, 50405 (1973) (expressing

concerns about police searches and seizures suppressing legitimate displays of

movies); Quantity of Copies of Books v. Kansas, 378 U.S. 205, 21113 (1964)

(holding that an overbroad warrant was unconstitutional in part because of its

potential for deterring the publication of legitimate books). Justice Sotomayor

recently expressed concern about the potential for surveillance to “chill[]. . . ex-

pressive freedoms.” United States v. Jones, 565 U.S. 400, 416 (2012) (So-

tomayor, J., concurring).

102

. Ferguson v. City of Charleston, 532 U.S. 67, 84 n.23 (2001).

103

. Solove, supra note 63, at 112730.

104

. See id. at 1129. An essential characteristic of a pragmatic theory is that

it “focus[es] on the specific types of disruption and the specific practices dis-

rupted rather than looking for the common [theoretical] denominator that links

all of them.” Id. at 1130.

105

. See Richards, supra note 65, at 389, 421 (“Intellectual privacy is the

ability . . . to develop ideas and beliefs away from the unwanted gaze or inter-

ference of others.”).

762 MINNESOTA LAW REVIEW [104:741

dependent practices. There are, however, other fundamental

harms caused by government surveillance that a Fourth Amend-

ment normative model must incorporate.

106

b. Harm to Relationships

The second harm factor asks whether a surveillance practice

would interfere with or diminish interpersonal relationships.

Surveillance might harm such relationships by compromising in-

timate communications, deterring relationship formation, or di-

minishing the depth or quality of intimate relationships via the

threat of observation.

Relationships with others are both extremely important to

people’s well-being and particularly dependent on privacy to

flourish.

107

An important aspect of personal relationships is “the

sharing of information about one’s actions, beliefs or emotions

which one does not share with all.”

108

By protecting such per-

sonal information from general observation, “privacy creates the

moral capital which we spend in friendship and love.”

109

Surveil-

lance can easily disrupt personal relationships by deterring un-

fettered communication,

110

disrupting intimacy,

111

inducing self-

consciousness and self-censorship,

112

or causing social embar-

rassment or condemnation.

113

106

. See infra Parts I.B.2.bc.

107

. See Lior Jacob Strahilevitz, A Social Networks Theory of Privacy, 72 U.

CHI. L. REV. 919, 923–24 (2005).

108

. CHARLES FRIED, AN ANATOMY OF VALUES: PROBLEMS OF PERSONAL

AND SOCIAL CHOICE 142 (1970).

109

. Id.

110

. See Richards, supra note 65, at 424 (“Our confidants are a source of new

ideas and information, but without confidentiality they may be reluctant to

share subversive or deviant thoughts with us lest others overhear.”).

111

. Robert S. Gerstein, Intimacy and Privacy, in PHILOSOPHICAL DIMEN-

SIONS OF PRIVACY: AN ANTHOLOGY 268–69 (Ferdinand D. Schoeman ed., 1984)

(explaining that surveillance disrupts the development and experience of inti-

macy by creating pressure of observation).

112

. Id.

113

. See James Rachels, Why Privacy Is Important, in PHILOSOPHICAL DI-

MENSIONS OF PRIVACY: AN ANTHOLOGY, supra note 111, at 29396 (“[S]epara-

tion allows us to behave with certain people in the way that is appropriate to

the sort of relationship we have with them.”); see also Helen Nissenbaum, Pri-

vacy as Contextual Integrity, 79 WASH. L. REV. 119, 13839 (2004) (“[N]orms of

appropriateness dictate what information about persons is appropriate, or fit-

ting, to reveal in a particular context.”).

2019] NORMATIVE FOURTH AMENDMENT 763

If a surveillance technique is likely to prevent people from

expressing private, provocative, or intimate thoughts to each

other, then that would weigh in favor of finding a Fourth Amend-

ment search. Courts may assess a surveillance technique’s im-

pacts on relationships by, for instance, examining studies show-

ing that the technique decreases or diminishes personal

communications.

114

Judges can also usefully intuit the impact of

outside surveillance on relationships. The effects of observation

by others on personal communications are generally easy to com-

prehend. Virtually everyone has had the experience of moderat-

ing or ceasing a conversation due to potential overhearing by an-

other such as a parent, teacher, stranger, or co-worker.

The Supreme Court has not expressly analyzed interference

with personal relationships in the Fourth Amendment context,

but it has repeatedly protected personal communications from

government surveillance and emphasized the importance of un-

fettered discourse. In the majority opinion in Katz, the Court

subjected telephone conversations to a warrant requirement,

grounding its holding in its recognition of “the vital role that the

public telephone has come to play in private communication.”

115

In one of the earliest Supreme Court Fourth Amendment cases,

the Court declared that sealed letters could not be inspected

without a search warrant.

116

Recently, a Sixth Circuit case con-

cluded that the Fourth Amendment should generally protect the

contents of emails, lest it “prove an ineffective guardian of pri-

vate communication, an essential purpose it has long been rec-

ognized to serve.”

117

This essential purpose has been obscured to

114

. See Carl Botan, Communication Work and Electronic Surveillance: A

Model for Predicting Panoptic Effects, 63 COMM. MONOGRAPHS 293, 307, 309–

10 (1996) (finding that workers under surveillance engaged in fewer personal

communications); R.H. Irving et al., Computerized Performance Monitoring Sys-

tems: Use and Abuse, 29 COMM. ACM 794, 799 (1986) (finding computer moni-

toring to be correlated with a decrease in the quality of peer relationships).

115

. Katz v. United States, 389 U.S. 347, 352 (1967).

116

. Ex parte Jackson, 96 U.S. 727, 733 (1877).

117

. Warshak v. United States, 631 F.3d 266, 286 (6th Cir. 2010) (recogniz-

ing the Fourth Amendment’s protection of “conversational privacy” (citing

United States v. U.S. Dist. Court, 407 U.S. 297, 313 (1972))).

764 MINNESOTA LAW REVIEW [104:741

some degree by the confusions of the Katz test, but courts con-

tinue to protect personal communications even when current

doctrine seems to suggest doing otherwise.

118

Outside of Fourth Amendment law, the Supreme Court has

recognized the importance of intimate relationships to human

well-being and has vigorously protected these relationships from

unnecessary state interference.

119

Laws that might adversely af-

fect marriages, parent-child relationships, non-marital romantic

relationships, cohabitation, and others have been struck down

as unconstitutional infringements on intimate relationships.

120

The Court’s longstanding recognition of the importance of these

relationships provides another basis for weighing harm to such

relationships in a normative Fourth Amendment analysis.

Intimacy and personal relationships have long been a cen-

tral focus of privacy theory, and more recent developments in

surveillance theory have specifically examined the potential for

surveillance to disrupt relationships. Scholars have explored in-

timacy as an important component of privacy since the 1970s,

121

developing various accounts of the values of private relation-

ships and the perniciousness of judgmental or exploitative obser-

vation.

122

Recently, contextual theories of privacy have explored

the disparate norms of information flow that govern various re-

lationships.

123

Surveillance can harm these associations not only

118

. See id. at 287 (refusing to create a bright-line rule protecting emails

and noting that protecting emails is somewhat in tension with the reasoning of

Miller v. United States, 425 U.S. 435 (1976)).

119

. Roberts v. U.S. Jaycees, 468 U.S. 609, 619 (1984) (collecting cases).

120

. See, e.g., Carey v. Population Servs. Int’l, 431 U.S. 678, 68486 (1977)

(non-marital intimacy); Moore v. City of East Cleveland, 431 U.S. 494, 50304

(1977) (plurality opinion) (cohabitation); Wisconsin v. Yoder, 406 U.S. 205, 232

(1972) (parent-child); Griswold v. Connecticut, 381 U.S. 479, 485–86 (1965)

(marriage); see also NAACP v. Alabama ex rel. Patterson, 357 U.S. 449, 46062

(1958) (striking down, on due process grounds, a law likely to deter citizens from

associating with others for the purposes of advocacy).

121

. See, e.g., FRIED, supra note 108, at 142 (discussing intimacy as it relates

to love and friendship and the necessity of privacy for those relationships to

flourish); Gerety, supra note 54, at 268, 273 (compiling cases).

122

. See, e.g., JULIE C. INNESS, PRIVACY, INTIMACY, AND ISOLATION 57–58,

61–63 (1992) (exploring the connection between intimate information, access,

and decision making); Gerstein, supra note 111, at 267–69.

123

. Nissenbaum, supra note 113, at 13839 (“Generally, these norms cir-

cumscribe the type or nature of information about various individuals that,

within a given context, is allowable, expected, or even demanded to be revealed.

2019] NORMATIVE FOURTH AMENDMENT 765

when these norms are violated and information is spread too

widely, but also when the fear of observation prevents the com-

munication necessary to maintain these relationships.

124

Inti-

macy, privacy, and communication are essential components of

personal relationships, and our understanding of the roles they

play has grown substantially in recent years.

c. Psychological and Physical Injury

The third factor asks whether people will suffer psychologi-

cal or physical injury as a result of surveillance. The impact of

surveillance goes beyond the substantial effects it can have on

people’s activities and relationships. Even in the absence of such

effects, the targets of surveillance can suffer personal harm from

the observation, judgment, fear, and occasional use of physical

force associated with government investigations.

125

Under this factor, evidence that a surveillance technique

will likely cause stress, depression, or physical harm would

weigh in favor of Fourth Amendment protection. The injuries

captured here include not only violations of privacy but also a

variety of other important harms, including discrimination, po-

lice coercion, and physical harm.

126

In many cases, judges may

be able to reason about or intuit such harms. For instance, they

can conclude that constant visual monitoring of a subject will re-

sult in stress or that stop-and-frisk techniques will be associated

with aggressive physical force. There are, moreover, an increas-

ing number of studies and reports that demonstrate measurable

psychological and physical harms from surveillance.

127

The rich and growing social science literature on the per-

sonal harms of surveillance has been largely ignored in existing

Fourth Amendment scholarship. Yet it can provide a way for

judges to concretize and measure internal privacy harms in a

normative Fourth Amendment analysis. For example, studies of

computer keystroke monitoring, telephone monitoring, and re-

lated practices have found a variety of psychological harms suf-

In medical contexts, it is appropriate to share details of our physical condi-

tion . . . among friends we may pour over romantic entanglements . . . .”).

124

. See id.

125

. See infra notes 13945.

126

. See infra Part III.C.

127

. See infra notes 12833 and accompanying text.

766 MINNESOTA LAW REVIEW [104:741

fered by the targets of surveillance, including stress, anger, fa-

tigue, depression, irritation, and infantilization.

128

Researchers

have also measured the physical and psychosomatic harms pro-

duced by surveillance, such as muscle pain and headaches.

129

Studies of video monitoring show that subjects feel discomfort

and agitation, as well as a feeling of being mistrusted.

130

Re-

search into stop-and-frisks and related police investigations

demonstrate that a history of police contact is correlated with

higher anxiety and stress, while stop-and-frisk frequency and in-

vasiveness is correlated with symptoms of PTSD.

131

Studies of

civilians subjected to consent searches of their vehicles reported

persistent negative thoughts and attitudes about the encounter,

as well as feelings of violation and bitterness.

132

These reports

can be augmented with the numerous studies in which respond-

ents rate the perceived invasiveness of various surveillance prac-

tices including location tracking, social media monitoring, and

128

. See, e.g., Lawrence M. Schleifer et al., Mood Disturbance and Musculo-

skeletal Discomfort Effects of Electronic Performance Monitoring in a VDT Data-

Entry Task, in ORGANIZATIONAL RISK FACTORS FOR JOB STRESS 195 (Steven L.

Sauter & Lawrence R. Murphy eds., 1995) (psychological and musculoskeletal

strain); Carl Botan & Mihaela Vorvoreanu, “What Are You Really Saying to

Me?” Electronic Surveillance in the Workplace, CERIAS TECH REPORT, June

2000, at 9–10, http://www.antoniocasella.eu/nume/Botan_2000.pdf [https://

perma.cc/J68P-ELEM] (distrust and lack of motivation); Irving, supra note 114,

at 799 (stress and “decrease in the quality of relationships with peers, supervi-

sors, and senior management”); M. J. Smith et al., Employee Stress and Health

Complaints in Jobs with and Without Electronic Performance Monitoring, 23

APPLIED ERGONOMICS 17, 2122 (1992) (anxiety, depression, anger, health com-

plaints, and fatigue).

129

. E.g., Schleifer et al., supra note 128, at 195; Smith et al., supra note

128, at 2122.

130

. Emmeline Taylor, I Spy with My Little Eye: The Use of CCTV in Schools

and the Impact on Privacy, 58 SOC. REV. 381, 39193 (2010); see also MCCAHILL,

supra note 93, at 15–16 (discussing the discriminatory harms that CCTV facil-

itates).

131

. Amanda Geller et al., Aggressive Policing and the Mental Health of

Young Urban Men, 104 AM. J. PUB. HEALTH 2321, 232324 (2014); Abigail A.

Sewell et al., Living Under Surveillance: Gender, Psychological Distress, and

Stop-Question-and-Frisk Policing in New York City, 159 SOC. SCI. & MED. 1, 2,

67 (2016).

132

. Janice Nadler, No Need to Shout: Bus Sweeps and the Psychology of

Coercion, 2002 SUP. CT. REV. 153, 21213 (2002). One respondent noted that

the police encounter produced “an empty feeling, like you’re nothing.” Id. at 212.

Another said, “I feel really violated . . . . I feel really bitter about the whole

thing.” Id.

2019] NORMATIVE FOURTH AMENDMENT 767

internet data collection.

133

Together, these studies constitute a

detailed and wide-ranging account of the internal harms of sur-

veillance.

This is not to say that every surveillance technique found to

cause stress or discomfort in a study should be considered a

search. Rather, these and similar studies can help to quantify

the harms of surveillance and are accordingly relevant to the

question of the Fourth Amendment’s scope. The fact that a sur-

veillance technique is linked to stress or pain is just one factor of

several in the proposed normative test, and the relevant social

science will rarely be definitive in any event. Moreover, not every

surveillance situation confronted by courts will have been ad-

dressed in an existing study of surveillance’s concrete harms.

Yet, courts can usefully test their intuitions about the harms

caused by surveillance against the available evidence, taking

empirical data into account as they have in a wide variety of con-

stitutional and other cases, including Brown v. Board of Educa-

tion,

134

Roper v. Simmons,

135

and countless others.

136

Courts can

133

. Tamara Dinev et al., Internet Privacy Concerns and Beliefs About Gov-

ernment Surveillance—An Empirical Investigation, 17 J. STRATEGIC INFO. SYS.

214, 223 (2008); Yongick Jeong & Erin Coyle, What Are You Worrying About on

Facebook and Twitter? An Empirical Investigation of Young Social Network Site

Users’ Privacy Perceptions and Behaviors, 14 J. INTERACTIVE ADVERT. 51, 55

(2014); Laurel A. McNall & Jeffrey M. Stanton, Private Eyes Are Watching You:

Reactions to Location Sensing Technologies, 26 J. BUS. PSYCHOL. 299, 304

(2011); Christopher Slobogin, Government Data Mining and the Fourth Amend-

ment, 75 U. CHI. L. REV. 317, 335 (2008); Christopher Slobogin, Public Privacy:

Camera Surveillance of Public Places and the Right to Anonymity, 72 MISS. L.

J. 213, 27576 (2002); Christopher Slobogin & Joseph E. Schumacher, Reason-

able Expectations of Privacy and Autonomy in Fourth Amendment Cases: An

Empirical Look at “Understandings Recognized and Permitted by Society,” 42

DUKE L.J. 727, 73739 tbl.1 (1993); Tokson, supra note 5, at 622–26.

134

. 347 U.S. 483, 494 n.11 (1954).

135

. 543 U.S. 551, 569–70 (2005).

136

. See, e.g., Riley v. California, 134 S. Ct. 2473, 2489–91 (2014) (discussing

phone usage statistics to demonstrate that searching a cell phone is different

and more invasive than other types of searches); Graham v. Florida, 560 U.S.

48, 68–69 (2010) (discussing scientifically established psychological differences

between adults and minors); Hodgson v. Minnesota, 497 U.S. 417, 469–70 (1990)

(discussing studies and testimony regarding the effects of involuntary disclo-

sure between parents and children in analyzing a law requiring teens to consult

with their parents before obtaining access to abortions); City of Renton v. Play-

time Theatres, Inc., 475 U.S. 41, 50–52 (1986) (holding that the City was per-

mitted to rely on studies performed by other cities to demonstrate the impact of

768 MINNESOTA LAW REVIEW [104:741

also draw useful comparisons between known surveillance

harms and those likely to be suffered in analogous cases. More-

over, judges and juries already conduct a somewhat similar in-

quiry in personal injury cases, where they assess damages for

psychological pain and suffering.

137

3. Less Invasive Means

Finally, the normative test incorporates a requirement that

courts consider whether there is a less invasive practice that

could reveal roughly the same information as the challenged

practice. If a surveillance technique is invasive or affects an en-

tire population, and a less invasive, feasible alternative exists,

that would weigh in favor of finding a Fourth Amendment

search. If alternative techniques would not be as effective or

would be prohibitively costly, that would weigh against finding

a search.

Courts currently apply a similar, albeit stricter, standard in

cases involving the Wiretap Act, which directs the government

to show that it has attempted less invasive surveillance before

applying for a wiretap.

138

In Fourth Amendment law, the Su-

preme Court has expressly considered the availability of less in-

vasive means when assessing the constitutionality of conducting

blood tests on suspected drunk drivers.

139

This factor is also

adult theaters on the community before enacting its adult theater zoning ordi-

nance). Social science and other scientific research are also routinely analyzed

in administrative law cases. See, e.g., Lorillard Tobacco Co. v. Reilly, 533 U.S.

525, 559–60 (2001) (explaining FDA studies of the effects of smokeless tobacco

and cigar use on teens).

137

. Tokson, supra note 13, at 199; see Sean Hannon Williams, Self-Altering

Injury: The Hidden Harms of Hedonic Adaptation, 96 CORNELL L. REV. 535,

543–44 n.42 (2011) (collecting cases involving hedonic damages). The inquiry

proposed here would likely be substantially easier, as the psychological harm

from surveillance need only be situated somewhere on the general scale from

low to high and would not have to be translated into a precise money value. Fact

finders tend to be far more consistent in performing the former calculation than

the latter. Cass R. Sunstein et al., Assessing Punitive Damages (with Notes on

Cognition and Valuation in Law), 107 YALE L.J. 2071, 2097103, 2099 tbl.1

(1998) (finding that mock jurors assessing various hypothetical cases tend to

give consistent rankings of blameworthiness but very different damages

awards).

138

. See 18 U.S.C. § 2518(1)(c) (2012); United States v. Carter, 449 F.3d

1287, 1293 (D.C. Cir. 2006).

139

. Birchfield v. North Dakota, 136 S. Ct. 2160, 2184, (2016) (holding that

2019] NORMATIVE FOURTH AMENDMENT 769

based in part on the intermediate scrutiny test in free speech

law, which directs courts to approve restrictions on certain types

of speech only if the restrictions do not burden substantially

more speech than is necessary to serve a significant government

interest.

140

Similarly, the existence here of a potentially less re-

strictive alternative would not definitively render a surveillance

technique unlawful, but it would be a factor that favors applying

a warrant requirement.

141

C. OMITTED FACTORS

The proposed test, like any Fourth Amendment test, cannot

incorporate every potential surveillance harm or every abstract

Fourth Amendment value without devolving into a “totality of

the circumstances” standard.

142

Accordingly, the test does not

analyze every circumstance or examine every theory that might

bear on the normative assessment of a surveillance practice.

Conceptually, it emphasizes pragmatic and contextual theories

of surveillance rather than more abstract theories that center on

control over information, autonomy, or personality develop-

ment.

143

The latter theories operate at too high a level of abstrac-

tion to be useful in a legal test. The normative approach proposed

here focuses on the more concrete harms of surveillance in order

to remain workable for judges and capable of consistent applica-

tion.

police officers cannot warrantlessly conduct blood tests incident to arrest be-

cause “[b]lood tests are significantly more intrusive [than breath tests], and

their reasonableness must be judged in light of the availability of the less inva-

sive alternative of a breath test”).

140

. See Ward v. Rock Against Racism, 491 U.S. 781, 799 (1989).

141

. See Bd. of Trs. of the State Univ. of N.Y. v. Fox, 492 U.S. 469, 480 (1989)

(explaining that intermediate scrutiny requires only a reasonable fit between

means and ends and does not require that the government select the least re-

strictive means possible).

142

. This is not to say that totality of the circumstances tests are entirely

foreign to Fourth Amendment law. See e.g., Birchfield, 136 S. Ct. at 2189 (as-

sessing the exigent circumstances exception to the warrant requirement using

a totality of the circumstances test).

143

. E.g., Anita L. Allen, Coercing Privacy, 40 WM. & MARY L. REV. 723, 738

(1999) (“Privacy has value relative to normative conceptions of spiritual person-

ality, political freedom, health and welfare, human dignity, and autonomy.”);

Charles Fried, Privacy, 77 YALE L.J. 475, 482 (1968) (defining privacy as “the

control we have over information about ourselves”).

770 MINNESOTA LAW REVIEW [104:741

Yet the test’s focus on foregone activities and the psycholog-

ical harms of surveillance can also capture many of the concerns

that drive the more abstract theories of privacy. Consider theo-

ries of privacy and autonomy, which focus on the need to pre-

serve a private zone within which individuals can develop and

choose free of social coercion.

144

Such coercion can cause the tar-

gets of surveillance to conform their behavior to perceived social

norms by foregoing legitimate but potentially embarrassing ac-

tivities.

145

And the social pressures inherent in many forms of

surveillance can result in psychological stress and harm.

146

These foregone activities and psychological harms would be cap-

tured by the normative test. Likewise, the test’s consideration of

physical harms resulting from police investigatory activity is in

accord with theories of privacy that focus on bodily integrity and

personal dignity.

147

These more abstract values are captured, at

least in part, by the proposed test, even though they are not

overtly included.

In any event, the impossibility of capturing every surveil-

lance harm in a single test mirrors the impossibility of capturing

every facet of law enforcement benefit. Both the deterrence ef-

fects and the retributivist values served by law enforcement are

unlikely to be fully captured, for instance. Any workable balanc-

ing test will elide some quantum of harm and benefit on both

sides. One of the virtues of such tests is that they typically leave

out far less than other types of legal standards.

148

The normative test also reflects a variety of the more ab-

stract Fourth Amendment values identified by courts and schol-

ars, such as privacy, liberty, or security.

149

One advantage of a

pragmatic approach is that the practical harms of surveillance

144

. See Cohen, supra note 56, at 1377, 1424.

145

. Julie E. Cohen, Privacy, Visibility, Transparency, and Exposure, 75 U.

CHI. L. REV. 181, 186 (2008); Richards, supra note 65, at 403–04.

146

. See supra Part I.B.2.c.

147

. See, e.g., Radhika Rao, Property, Privacy, and the Human Body, 80 B.U.

L. REV. 359, 388 (2000).

148

. See infra Part III.A.

149

. See Christopher Slobogin, A Defense of Privacy as the Central Value

Protected by the Fourth Amendment’s Prohibition on Unreasonable Searches, 48

TEX. TECH L. REV. 143, 157–62 (2015) (collecting studies identifying different

but closely related Fourth Amendment principles).

2019] NORMATIVE FOURTH AMENDMENT 771

are often common denominators among the various abstract the-

ories of Fourth Amendment principles.

150

Indeed, to the extent

that courts and historians have identified a single general pur-

pose of the Fourth Amendment, that purpose is itself more func-

tional than abstract: to protect citizens from arbitrary govern-

ment intrusions.

151

Given this shared practical foundation, it is

unsurprising that the various theories of Fourth Amendment

values overlap more than they conflict.

152

The common func-

tional goals of these various Fourth Amendment “search” theo-

ries are largely captured by the proposed test.

153

II. DOCTRINAL AND THEORETICAL FOUNDATIONS OF

FOURTH AMENDMENT BALANCING

In setting out the normative model, the previous Part dis-

cussed some of the legal and theoretical foundations of its fac-

tors. This Part briefly examines doctrinal, historical, and theo-

retical support for a normative balancing approach in general.

The Fourth Amendment’s text and its broader purposes are con-

sistent with the balancing of law enforcement benefits against

the costs of surveillance.

154

The language and history of the

Amendment evince a concern with effective law enforcement as

150

. See, e.g., Thomas P. Crocker, From Privacy to Liberty: The Fourth

Amendment After Lawrence, 57 UCLA L. REV. 1, 18 (2009) (discussing state in-

trusions on same-sex intimacy and noting the link between principles of liberty

and the protection of intimate relationships); Radhika Rao, Reconceiving Pri-

vacy: Relationships and Reproductive Technology, 45 UCLA L. REV. 1077, 1101–

07 (1998) (discussing state intrusions on private decisions surrounding relation-

ships and arguing that the right of privacy is fundamentally a right of protection

of personal relationships).

151

. See, e.g., United States v. Verdugo-Urquidez, 494 U.S. 259, 266 (1990)

(“[T]he purpose of the Fourth Amendment was to protect the people of the

United States against arbitrary action by their own Government.”); Wolf v. Col-

orado, 338 U.S. 25, 28–29 (1949) (holding that the Fourth Amendment’s protec-

tion against arbitrary intrusion by the police is part of the Due Process guaran-

teed by the Fourteenth Amendment); Davies, supra note 67, at 556 (discussing

“the larger purpose for which the Framers adopted the text; namely to curb the

exercise of discretionary authority by officers”).

152

. John D. Castiglione, Human Dignity Under the Fourth Amendment,

2008 WIS. L. REV. 655, 675 (2008); Slobogin, supra note 149, at 152–54.

153

. See U.S. CONST. amend. IV. The Fourth Amendment also protects

against unreasonable “seizures,” a separate prohibition than the one addressed

here and one that embodies the values of protection of property and freedom

from arrest.

154

. See infra Part II.A.

772 MINNESOTA LAW REVIEW [104:741

well as citizen privacy.

155

Moreover, both the leading originalist

interpretation of the Amendment and less formalist theories of

construction point to a balancing approach.

A. THE FOURTH AMENDMENT’S BALANCE

Balancing is inherent in Fourth Amendment law, as re-

flected in the Amendment’s history, language, and purposes. The

very concept of warrants supported by “probable cause”

156

con-

templates a balancing between law enforcement interests and

citizen privacy.

157

The government can obtain a search warrant

only if it has sufficient cause to believe the search will uncover a

crime.

158

Once the government has sufficient cause, it can search

citizens and their property despite the considerable harms to pri-

vacy and liberty that might result.

159

Indeed, the police can enter

the house of a totally innocent person to arrest a criminal or seize

contraband possessed by a houseguest.

160

Neither the interests

of individuals in avoiding government intrusions nor the inter-

ests of law enforcement are absolute.

Founding-era practices likewise evinced a non-absolutist ap-

proach to searches and seizures. Unlawful searches were ad-

dressed with civil liability rather than the exclusion of evi-

dence.

161

The trespass actions that provided a basis for Fourth

Amendment protection were themselves tempered by doctrines

155

. See infra Part II.B.

156

. See infra Part II.B.

157

. See infra Part II.B.

158

. See Illinois v. Gates, 462 U.S. 213, 24346 (1983).

159

. See Michigan v. Summers, 452 U.S. 692, 703 (1981) (noting that a war-

rant gives the police “a special authorization to thrust themselves into the pri-

vacy of a home” as well as the authority to physically detain occupants of the

home while it is searched for contraband).

160

. See Steagald v. United States, 451 U.S. 204, 213, 222 (1981) (stating

that the police could enter the house of an innocent third party to arrest a felon

if they had a search warrant or probable cause and exigency).

161

. See, e.g., Entick v. Carrington, [1765] 95 Eng. Rep. 807 (K.B.).

2019] NORMATIVE FOURTH AMENDMENT 773

of necessity, which allowed trespasses when necessary to pre-

vent public or private harm.

162

Unwarranted invasions were gen-

erally excused if contraband was discovered.

163

In each situation,

citizens’ protections against government intrusions were coun-

terweighted by other values and defeasible in cases involving

probable cause, public or private necessity, or actual guilt.

164

The balancing inherent in Fourth Amendment law does not

dictate that courts must balance when examining the scope of

the Amendment—perhaps balancing should be confined to other

aspects of Fourth Amendment law or eschewed altogether.

165

But a normative balancing test for scope is consistent with the

structure and traditional practice of the Fourth Amendment.

B. TEXT, ORIGINALISM, AND DETERMINACY

The Fourth Amendment prohibits “unreasonable searches,”

a phrase that is not defined and is susceptible to a wide variety

of meanings.

166

The dominant view of the Fourth Amendment is

that its text and history are of little or no help in determining its

scope.

167

Yet, a number of scholars contend that the scope of the

Amendment is determinable by reference to the original public

meaning of the relevant phrase.

168

162

. See, e.g., Campbell v. Race, 61 Mass. (7. Cush.) 408, 41011 (1851) (col-

lecting American and English sources describing the common law rule that en-

croachment on private property was permitted when a highway becomes im-

passable); Mouse’s Case, [1608] 77 Eng. Rep. 1341 (K.B.) 1342 (finding that

property may be trespassorily destroyed if necessary to save lives).

163

. See, e.g., Gelston v. Hoyt, 16 U.S. 246, 310 (1818); Akhil Reed Amar,

Fourth Amendment First Principles, 107 HARV. L. REV. 757, 767 (1994) (collect-

ing sources).

164

. See supra notes 16063.

165

. See discussion infra Part IV.C.

166

. See Matthew Tokson, Blank Slates, 59 B.C. L. REV. 591, 62730 (2018).

167

. See, e.g., Amsterdam, supra note 6, at 395; Orin S. Kerr, The Curious

History of Fourth Amendment Searches, 2012 SUP. CT. REV. 67, 70.

168

. See, e.g., Brief of Scholars of the History and Original Meaning of the

Fourth Amendment as Amici Curiae in Support of Petitioner at 2, Carpenter v.

United States, 138 S. Ct. 2206 (2018) (No. 16-402), 2017 WL 3530961 [herein-

after Originalist Scholars Amicus Brief]; DAVID GRAY, THE FOURTH AMEND-

MENT IN AN AGE OF SURVEILLANCE 251 (2017); Amar, supra note 163, at 767.

There are other forms of originalist interpretation, including “original methods

originalism,” which recommend interpreting the Constitution by reference to

the methods of legal interpretation used at the time of the Founding. See, e.g.,

John O. McGinnis & Michael B. Rappaport, Original Methods Originalism: A

New Theory of Interpretation and the Case Against Construction, 103 NW. U. L.

774 MINNESOTA LAW REVIEW [104:741

This Article does not undertake to resolve this debate, be-

cause it need not resolve it—both major views of the determin-

ism of the Fourth Amendment’s text are consistent with the nor-

mative balancing approach. Indeed, both counsel weighing the

harms of surveillance against law enforcement justifications in

order to determine which investigations the police can perform

without any quantifiable suspicion.

169

This Section explores the-

ories regarding the determinacy of the Fourth Amendment and

shows how they provide further support for Fourth Amendment

balancing.

1. Fourth Amendment Searches as Textually Determinate

Several Fourth Amendment scholars have argued that the

term “search” in the context of the Fourth Amendment gives spe-

cific guidance as to the scope of the Amendment.

170

They contend

that the Amendment applies to any “search” in the broadest

sense of that term, meaning any act of seeking, gathering infor-

mation, or looking at something.

171

Thus a government official

looking at a house or a crowd of people would be conducting a

warrantless Fourth Amendment search.

172

Many such searches

would be lawful, however, because they would be “reasonable.”

173

Reasonableness would no longer require a warrant supported by

probable cause as a default rule; instead, reasonableness would

be a more general inquiry into whether a search had a “good and

sufficient justification” and was not “greater than is fit” or “im-

moderate.”

174

Although the reasonableness inquiry is an amor-

phous, “common sense” sort of analysis,

175

it would consider how

REV. 751, 78687 (2009). The predominant originalist approach in the Fourth

Amendment context focuses on original public meaning, and that is the ap-

proach discussed in this section.

169

. See infra Parts II.B.12.

170

. See supra note 168.

171

. Originalist Scholars Amicus Brief, supra note 168, at 67; GRAY, supra

note 168, at 251; Amar, supra note 163, at 76869.

172

. See, e.g., Originalist Scholars Amicus Brief, supra note 168, at 13;

Amar, supra note 163, at 768.

173

. Amar, supra note 163, at 769.

174

. Originalist Scholars Amicus Brief, supra note 168, at 14–15.

175

. Amar, supra note 163, at 801.

2019] NORMATIVE FOURTH AMENDMENT 775

intrusive a search is,

176

and whether the search is excessive in

light of its justifications.

177

This interpretation of the Fourth Amendment is notably

consistent with the balancing approach proposed in Part I. Both

approaches would resolve the question of when the government

can engage in warrantless surveillance by making a normative

inquiry to determine whether such surveillance is justified.

There are differences, of course. This Article’s approach is more

specific and less reliant on distant historical analogy than the

originalist approaches.

178

It would also conduct its balancing at

the “scope” stage rather than the “reasonableness” stage of a

Fourth Amendment case, preserving the longstanding role of

warrants and probable cause in regulating police behavior.

179

The warrant requirement, unlike the Katz test, has not come un-

der widespread attack by scholars or commentators.

180

Indeed,

many have argued for strengthening the requirement by limit-

ing its various exceptions, and empirical data indicates that war-

ranted searches are far more likely than unwarranted probable-

cause searches to actually produce evidence of crime.

181

The nor-

176

. Id.

177

. Originalist Scholars Amicus Brief, supra note 168, at 15. Note that this

is not the only originalist interpretation of Fourth Amendment reasonableness.

Laura Donohue has argued that “unreasonable” in the Fourth Amendment’s

text refers to something “against the reason of the common law,” including war-

rantless entry into a home. See Laura Donohue, The Original Fourth Amend-

ment, 83 U. CHI. L. REV. 1181, 1192 (2016). This approach to Fourth Amend-

ment reasonableness would be less consistent with the normative balancing

approach proposed above and would likely be more focused on government ac-

tions violating the common law.

178

. Cf. Originalist Scholars Amicus Brief, supra note 168, at 34 (analyzing

cell phone signal data collection by reference to the general warrant cases of the

pre-Founding era).

179

. Under the Fourth Amendment, once a government action is found to be

a “search,” it can still be justified as “reasonable” if, for instance, the govern-

ment has obtained a warrant or qualified for an exception to the warrant re-

quirement. The prevailing originalist approach to reasonableness would simply

ask whether a search was generally reasonable, rather than requiring a war-

rant as a default. Id. at 15; Amar, supra note 163, at 801.

180

. Cf. CHRISTOPHER SLOBOGIN, PRIVACY AT RISK: THE NEW GOVERNMENT

SURVEILLANCE AND THE FOURTH AMENDMENT 4445 (2007) (suggesting that

courts preserve ex ante review but advocating for the issuance of warrants on

less than probable cause).

181

. See, e.g., Phyllis T. Bookspan, Reworking the Warrant Requirement: Re-

suscitating the Fourth Amendment, 44 VAND. L. REV. 473, 481 (1991); Wayne D.

776 MINNESOTA LAW REVIEW [104:741

mative test would avoid overturning more than a century of war-

rant-requirement precedents and undermining effective ex ante

judicial review of police surveillance.

182

But the inquiry would be

conceptually similar to the originalist inquiry. Moreover, it

would make little difference to a police officer whether looking

at a house without probable cause is lawful because it is not reg-

ulated by the Fourth Amendment or because it is “reasona-

ble.”

183

The normative test proposed here is congruous with the

predominant originalist interpretation of the Fourth Amend-

ment.

184

2. Fourth Amendment Searches as Textually Indeterminate

The majority of scholars who have written on the Fourth

Amendment’s scope consider its text and history to be indeter-

minate, or at best, profoundly underdeterminate.

185

Not only is

Holly, The Fourth Amendment Hangs in the Balance: Resurrecting the Warrant

Requirement Through Strict Scrutiny, 13 N.Y. L. SCH. J. HUM. RTS. 531, 532

(1997); Max Minzner, Putting Probability Back into Probable Cause, 87 TEX. L.

REV. 913, 92325 (2009).

182

. See, e.g., Riley v. California, 134 S. Ct. 2473, 2482 (2014); Bond v.

United States, 529 U.S. 334, 33839 (2000); Oliver v. United States, 466 U.S.

170, 170 (1984); Katz v. United States, 389 U.S. 347, 347 (1967); United States

v. Lefkowitz, 285 U.S. 452, 467 (1932); Perlman v. United States, 247 U.S. 7, 7

(1918); Boyd v. United States, 116 U.S. 616, 616 (1886).

183

. See Amar, supra note 163, at 769.

184

. Indeed, to the extent that originalism incorporates values of stare deci-

sis, the normative test may be the optimal originalist approach because it avoids

overturning longstanding precedents. See supra note 182.

185

. See supra note 167. For a discussion of underdeterminacy and construc-

tion in legal interpretation, see Lawrence B. Solum, Originalism and Constitu-

tional Construction, 82 FORDHAM L. REV. 453, 458 (2013). Some originalist

scholars have argued that underdeterminate text can be clarified by reference

to the spirit of the constitutional provision at issue, i.e. its original function or

purpose. See Randy E. Barnett & Evan D. Bernick, The Letter and the Spirit: A

Unified Theory of Originalism, 107 GEO. L.J. 1, 3 (2018). In this context, the

generally acknowledged purposes of the Fourth Amendment are fairly abstract

and may not substantially clarify the scope of the Fourth Amendment. See Tok-

son, supra note 166, at 635 & n.279 (noting that historians generally consider

the “bedrock purpose of the Fourth Amendment” to be the protection of “privacy,

property, and liberty from undue intrusions by government officers,” and quot-

ing several historians). Assuming historians are correct that a core purpose of

the Amendment was to protect values like privacy and liberty against govern-

ment oppression, the test proposed here is likely congruent with an original-

purpose-based approach.

2019] NORMATIVE FOURTH AMENDMENT 777

the term “search” ambiguous and capable of multiple mean-

ings,

186

but the Supreme Court’s time-honored interpretation of

“reasonable” as typically requiring a warrant or at least some

articulable suspicion means that not every investigative act can

be a search.

187

The crucial question of when the police can con-

duct suspicion-less surveillance is not answered in the text or

history.

188

What should courts do when addressing an indeterminate

law? General theories of legal indeterminacy typically conceive

of judges who fill legal gaps as acting in a legislative capacity

and attempting to reach optimal outcomes via a normative-style

inquiry.

189

This inquiry might entail the consideration of moral

values, policy judgments, or personal experiences.

190

Judges

might accordingly weigh these types of considerations in ad-

dressing the Fourth Amendment’s scope in the absence of formal

guidance. These broad prescriptions do not mandate a balancing

test, but they are certainly consistent with the use of normative

balancing when addressing indeterminate law.

Further, theories of indeterminacy that focus on how courts

should formulate legal tests in the absence of determinate law

directly support the use of a balancing test in the Fourth Amend-

ment context.

191

The issue of the Fourth Amendment’s scope is

normatively complex, covers a wide variety of government con-

duct, and has been repeatedly destabilized by technological and

social change.

192

Non-balancing standards may therefore fail to

186

. Kerr, supra note 167, at 70.

187

. Tokson, supra note 166, at 640.

188

. Id. at 628–29.

189

. See, e.g., JOSEPH RAZ, THE AUTHORITY OF LAW 197–99 (1979); Thomas

W. Merrill, The Common Law Powers of Federal Courts, 52 U. CHI. L. REV. 1,

43 (1985). Ronald Dworkin takes a philosophically different approach to doctri-

nal indeterminacy that ultimately offers similar advice. See RONALD DWORKIN,

TAKING RIGHTS SERIOUSLY 124, 128 (1977) (describing the central role of polit-

ical and personal convictions in Dworkinian adjudication). Dworkin argues that

judges should address difficult legal questions by choosing the outcome that fits

best with the overarching narrative or theory of law and with political morality.

RONALD DWORKIN, A MATTER OF PRINCIPLE 138–43 (1985); DWORKIN, supra, at

107.

190

. See MELVIN ARON EISENBERG, THE NATURE OF THE COMMON LAW 148

(1988); RICHARD A. POSNER, HOW JUDGES THINK 82–83, 106–08 (2008); DAVID

A. STRAUSS, THE LIVING CONSTITUTION 38 (2010).

191

. See supra note 80.

192

. Tokson, supra note 166, at 614–15, 643–44.

778 MINNESOTA LAW REVIEW [104:741

capture the fundamental values underlying the issue and may

not be much simpler to apply than a direct balancing test.

193

Moreover, courts are increasingly likely to be able to obtain the

information they need to effectively balance in the Fourth

Amendment context.

194

In such a situation, a balancing test is

likely to be the optimal approach for courts faced with legal in-

determinacy.

195

General theories of legal indeterminacy are con-

sistent with a normative balancing approach, and more detailed

theories directly support such an approach.

III. THE CASE FOR A NORMATIVE BALANCING MODEL

The previous Parts have set out a normative balancing

model for the Fourth Amendment’s scope, traced the lineage of

its various factors, and given an account of its doctrinal, histori-

cal, and theoretical foundations. This Part details the normative

balancing model’s more practical advantages: directness, adapt-

ability to social and technological change, inclusion of non-pri-

vacy harms, harmonization of doctrine with practice, and ap-

plicability to broad surveillance programs. In a society where

surveillance technology consistently advances and expectations

of privacy continually shrink, these benefits may be indispensa-

ble.

196

A. DIRECTNESS

A prominent advantage of the normative balancing ap-

proach is that it directly addresses the normative values at issue.

Courts need not use “false targets” or proxies that stand in for

essential Fourth Amendment interests;

197

they would examine

those interests directly. If judges can administer a balancing test

193

. Cloud, supra note 6, at 28–36; Tokson, supra note 166, at 644–45.

194

. Tokson, supra note 166, at 645. See generally John R. Aiello & Kathryn

J. Kolb, Electronic Performance Monitoring and Social Context: Impact on

Productivity and Stress, 80 J. APPLIED PSYCHOL. 339, 339 (1995) (studying

stress among the targets of surveillance); Christopher Slobogin, Government

Data Mining and the Fourth Amendment, 75 U. CHI. L. REV. 317, 335 (2008)

(setting out the average perceived intrusiveness of various types of searches);

Daniel J. Solove & Danielle Keats Citron, Risk and Anxiety: A Theory of Data

Breach Harms, 96 TEX. L. REV. 737, 773–77 (2018) (discussing how courts might

quantify damages from privacy breaches).

195

. Tokson, supra note 166, at 613–16.

196

. See id. at 614.

197

. See id. at 60911.

2019] NORMATIVE FOURTH AMENDMENT 779

effectively, then its outcomes should maximize societal welfare

relative to other tests.

Of course, the other side of this coin is that balancing tests

are generally difficult to administer, as discussed below.

198

But

a balancing approach is likely to be more effective than a nar-

rower standard in this context. Because the question of the

Fourth Amendment’s scope is conceptually complex, broad, and

subject to constant disruption by new technologies, it is unlikely

that a narrow standard can effectively capture the fundamental

values at stake.

199

A balancing test, though hardly without draw-

backs, avoids this fatal error.

Relatedly, the normative approach embodies the balance

that is inherent in the Fourth Amendment.

200

It squarely ad-

dresses the purposes of the Fourth Amendment, directly as-

sessing the harms of arbitrary government intrusions and the

practical values of security, liberty, and privacy.

201

It hews far

more closely to traditional Fourth Amendment goals than does,

for instance, the Katz test, which focuses on current societal ex-

pectations about privacy.

202

B. ADAPTABILITY TO SOCIAL AND TECHNOLOGICAL CHANGE

The normative approach is especially adaptable to new cir-

cumstances and new surveillance technologies. It looks to law

enforcement benefits and practical privacy harms, no matter

how those benefits and harms may manifest in a given surveil-

lance context. Alternative tests are often more rigid and prone

to destabilization by changing circumstances.

203

Changes in surveillance practices and technologies have re-

peatedly undermined narrower Fourth Amendment tests in the

past. In the early twentieth century, the Supreme Court held

that the Fourth Amendment’s protections were limited to the

specific types of property enumerated in the “persons, houses,

papers, and effects” clause of the Amendment.

204

This property-

based approach exposed telephone and other conversations to

198

. See infra Part IV.A.

199

. See supra Part II.B.2.

200

. See supra Part II.A.

201

. See supra note 66.

202

. Katz v. United States, 389 U.S. 347, 361 (1967) (Harlan, J., concurring).

203

. See Tokson, supra note 13, at 18187.

204

. See Olmstead v. United States, 277 U.S. 438, 476 (1928).

780 MINNESOTA LAW REVIEW [104:741

pervasive government monitoring, leading to egregious privacy

violations and political abuses.

205

The Court eventually adopted

the Katz test, which expanded the Fourth Amendment’s cover-

age to intangible things and based it on expectations of pri-

vacy.

206

Yet the Katz test has itself been rapidly destabilized as

threats to privacy proliferate, knowledge of such threats gradu-

ally spreads, and the cost-per-citizen of surveillance drops pre-

cipitously.

207

In a society where the government can collect huge

databases of personal information held by commercial third par-

ties,

208

engage in constant visual monitoring via drones or satel-

lites,

209

or mine email metadata to reveal intimate details about

people’s lives,

210

the concept of an expectation of privacy not

grounded in legal protections is increasingly obsolete.

Adaptability is especially important given the outsized role

that social and technological change plays in Fourth Amendment

law. A normative balancing approach allows courts to take ac-

count of a novel surveillance context without depending on soci-

etal expectations or waiting for Congress to pass a law—a wait

that might take decades.

211

The normative test is resilient to the

changes that have undermined previous and current Fourth

Amendment tests.

C. DISCRIMINATION-BASED HARMS

Many of the harms of surveillance are related to the loss of

privacy that occurs when a subject is observed by others. But the

205

. See, e.g., U.S. SELECT COMM. TO STUDY GOVERNMENTAL OPERATIONS

WITH RESPECT TO INTELLIGENCE ACTIVITIES, FINAL REPORT, S. REP. No. 94-755,

at 18384, 198201 (1976) [hereinafter FINAL REPORT OF THE SELECT COMM.].

206

. Katz, 389 U.S. at 361 (Harlan, J., concurring).

207

. See Tokson, supra note 13, at 18187.

208

. E.g., Chris J. Hoofnagle, Big Brother’s Little Helper: How ChoicePoint

and Other Commercial Data Brokers Collect and Package Your Data for Law

Enforcement, 29 N.C. J. INT’L. L. & COM. REG. 595, 63537 (2004).

209

. Robert Draper, They Are Watching You—And Everything Else on the

Planet, NAT’L GEOGRAPHIC, Feb. 2018, https://www.nationalgeographic.com/

magazine/2018/02/surveillance-watching-you [https://perma.cc/769W-LDCP].

210

. Barton Gellman & Ashkan Soltani, NSA Collects Millions of E-mail Ad-

dress Books Globally, WASH. POST (Oct. 14, 2013), https://www.washingtonpost

.com/world/national-security/nsa-collects-millions-of-e-mail-address-books

-globally/2013/10/14/8e58b5be-34f9-11e3-80c6-7e6dd8d22d8f_story.html.

211

. The Electronic Communications Privacy Act (“ECPA”) has not yet been

meaningfully updated since it became law in 1986, despite massive advances

and changes in email technology. See 18 U.S.C. §§ 2516, 2703 (2012).

2019] NORMATIVE FOURTH AMENDMENT 781

Katz test’s exclusive focus on informational privacy fails to cap-

ture some of the most harmful aspects of government surveil-

lance: discrimination, coercion, intimidation, and physical

harm.

212

Routine police encounters on public sidewalks or roads,

for instance, may have little impact on informational privacy but

nonetheless may harm individuals through coercion or the

threat of violence.

213

The normative test takes a broader view of

Fourth Amendment privacy and protection, one that considers

the personal harms of surveillance whether they arise from ob-

servation or from more direct tactics of intimidation or coer-

cion.

214

Non-privacy harms may be especially important when sur-

veillance reflects discrimination against certain groups or other-

wise expresses societal condemnation of surveillance targets.

State surveillance can have a powerful expressive component,

conveying the message that its targets are low-status members

of society, unworthy of trust, or inherently dangerous.

215

Dis-

crimination itself, including discrimination associated with po-

lice practices, can cause serious short-term psychological and

physical effects, including stress, depression, elevated heart

rate, and high blood pressure.

216

Over the long term, such dis-

crimination is correlated with a variety of health problems, such

as heart attacks and strokes.

217

Surveillance programs that tar-

get or disproportionately affect a particular demographic group

may cause serious harms to individuals that should be taken into

account in a Fourth Amendment analysis. The normative test

allows courts to directly consider such harms when assessing a

government surveillance practice.

212

. See Stuntz, supra note 9, at 1065–66.

213

. Id.

214

. See supra notes 128–32 and accompanying text.

215

. See Craig Konnoth, An Expressive Theory of Privacy Intrusions, 102

IOWA L. REV. 1533, 156367 (2017).

216

. Pamela J. Sawyer et al., Discrimination and the Stress Response: Psy-

chological and Physiological Consequences of Anticipating Prejudice in Inter-

ethnic Interactions, 102 AM. J. PUB. HEALTH 1020, 1022 tbl.1 (2012); Abigail A.

Sewell & Kevin A. Jefferson, Collateral Damage: The Health Effects of Invasive

Police Encounters in New York City, 93 J. URB. HEALTH 542, 543 (2016).

217

. Sawyer et al., supra note 216, at 1020.

782 MINNESOTA LAW REVIEW [104:741

D. HARMONIZING PRACTICE AND DOCTRINE

The Katz test directs courts to assess society’s expectations

of privacy, and many courts faithfully attempt to do so. Lower

courts especially tend to address novel Fourth Amendment scope

questions by attempting to calculate societal knowledge and ex-

pectations about surveillance practices.

218

The Supreme Court

frequently does the same, looking explicitly to our “everyday ex-

pectations of privacy”

219

and what people “typically know”

220

determining the scope of the Fourth Amendment.

221

The results

and reasoning of such cases are frequently criticized, but one

might at least admire these courts’ fidelity to governing prece-

dent.

222

Yet many Fourth Amendment cases, especially at the Su-

preme Court level, appear to be driven by normative concerns

rather than doctrinal ones.

223

Consider the third-party doctrine,

which states that people waive their Fourth Amendment rights

in things that they voluntarily disclose to a third party.

224

This

infamous doctrine threatens privacy in a vast swath of personal

data in the internet age.

225

Yet, even before it was expressly lim-

ited in Carpenter v. United States,

226

the third-party doctrine

218

. See Tokson, supra note 13, at 154, 15658, 16163 (describing numer-

ous examples of lower courts attempting to assess the extent of societal

knowledge in order to determine societal expectations of privacy).

219

. Minnesota v. Olson, 495 U.S. 91, 98 (1990).

220

. Smith v. Maryland, 442 U.S. 735, 743 (1979).

221

. See, e.g., Bond v. United States, 529 U.S. 334, 33839 (2000) (“[A] bus

passenger clearly expects that his bag may be handled. He does not expect that

other passengers or bus employees will . . . feel the bag in an exploratory man-

ner.”); California v. Greenwood, 486 U.S. 35, 40 (1988) (“It is common knowledge

that plastic garbage bags left on or at the side of the road are readily accessible

to . . . scavengers, snoops, and other members of the public.”); California v. Car-

ney, 471 U.S. 386, 392 (1985) (“The public is fully aware that it is accorded less

privacy in its automobiles . . . .”).

222

. See, e.g., SLOBOGIN, supra note 180, at 15164 (2007); Lewis R. Katz, In

Search of a Fourth Amendment for the Twenty-First Century, 65 IND. L.J. 549,

564–66 (1990).

223

. See Carpenter v. United States, 138 S. Ct. 2206, 2236 (2018) (Thomas,

J., dissenting); Kerr, supra note 32, at 519.

224

. See Smith, 442 U.S. at 744; United States v. Miller, 425 U.S. 435,

44243 (1976).

225

. Tokson, supra note 5, at 585.

226

. See, e.g., Miller, 425 U.S. at 442 (holding that bank records are not pro-

tected by the Fourth Amendment because they are exposed to bank employees

in the ordinary course of business).

2019] NORMATIVE FOURTH AMENDMENT 783

seemed to disappear whenever it would produce a particularly

unjust outcome.

227

In a typical third-party doctrine case, expo-

sure of something to a third party’s employees eliminates Fourth

Amendment protection in that thing.

228

Yet in Ferguson v. City

of Charleston, the Court held that a state hospital’s program of

surreptitiously testing patients’ urine for cocaine violated the

Fourth Amendment, despite the fact that patients voluntarily

turned over their urine to hospital employees.

229

And the Court

held in Stoner v. California that the police must obtain a search

warrant to enter a hotel room despite the fact that “maids, jani-

tors, or repairmen” routinely enter and observe the room in the

normal course of business.

230

Recently, in Carpenter, several dis-

senting Justices reasonably complained that the Court’s decision

to extend the Fourth Amendment to cell phone location data ap-

peared driven by normative considerations rather than the lit-

eral Katz test.

231

Policy considerations, rather than societal ex-

pectations, seem to dictate the outcomes of several other Fourth

Amendment cases as well.

232

Indeed, they appear to drive the

outcomes of some cases that purport to turn on neutral concepts

like trespass and property.

233

227

. Neil Richards notes a similar phenomenon in Richards, supra note 40,

at 146873, contending that the Supreme Court was always more concerned

with the unrevealing nature of the information at issue in the third-party doc-

trine cases than with the fact of disclosure to third parties.

228

. Miller, 425 U.S. at 442.

229

. 532 U.S. 67, 84–85 (2001). The Court granted certiorari only on the is-

sue of whether the testing fit within the special needs exception and assumed a

lack of patient consent. Id. at 76. But the dissenting Justices noted that the

patients’ consent was obvious and provided a clear basis to resolve the case. Id.

at 9296 (Scalia, J., dissenting).

230

. 376 U.S. 483, 489 (1964).

231

. Carpenter v. United States, 138 S. Ct. 2206, 2236 (2018) (Thomas, J.,

dissenting); id. at 2265 (Gorsuch, J., dissenting).

232

. See Illinois v. Caballes, 543 U.S. 405, 410 (2005) (holding that dog sniffs

for contraband are not searches regardless of people’s expectations of privacy);

United States v. Jacobsen, 466 U.S. 109, 124 (1984) (holding that testing sub-

stances for contraband is not a search); Hudson v. Palmer, 468 U.S. 517, 536

(1984) (expressly considering the benefits and costs of permitting warrantless

searches of prison cells); Kerr, supra note 32, at 51922.

233

. See Florida v. Jardines, 569 U.S. 1, 7–10 (2013) (finding a Fourth

Amendment search under the Jones trespass test despite the absence of a tres-

pass, based largely on novel claims about the social norms that govern ap-

proaching a doorstep).

784 MINNESOTA LAW REVIEW [104:741

The normative test directs courts to give an account of the

core normative considerations that appear to drive a substantial

portion of the Supreme Court’s cases. It would have the benefit

of making the Court’s actual rationales for its decisions visible

and subject to scrutiny. When the Supreme Court reaches an es-

sentially normative decision but obscures its reasoning behind

one Katz doctrine or another, observers are less able to predict

future cases, detect judicial bias, or understand existing law. The

normative test would better align the outcomes of Fourth

Amendment cases with their actual rationales, promoting trans-

parency and judicial credibility.

E. AGGREGATION AND SPILLOVER

The normative test would help courts to address the Fourth

Amendment issues raised by aggregated programs of surveil-

lance. Wide-ranging surveillance programs can yield massive

databases of citizens’ information.

234

These vast collections of

data can be analyzed to reveal far more than would be revealed

by any single act of investigation.

235

Aggregated surveillance

programs are increasingly problematic as the cost-per-citizen of

surveillance and analysis decreases.

236

Current Fourth Amendment approaches are largely blind to

the dangers of aggregate surveillance. Courts have rightly been

criticized for their transactional, non-systematic approach to

Fourth Amendment questions.

237

Although courts occasionally

look to the future impacts of their decisions, they generally as-

sess each investigatory act in isolation rather than considering

surveillance programs as a whole.

238

This is problematic be-

cause, in practice, Fourth Amendment decisions that permit the

government to surveil one specific individual effectively grant

the government the power to surveil citizens en masse.

234

. Renan, supra note 74, at 105860.

235

. Id. at 1056.

236

. See Tokson, supra note 32.

237

. Barry Friedman & Cynthia Benin Stein, Redefining What’s “Reasona-

ble”: The Protections for Policing, 84 GEO. WASH. L. REV. 281, 298 (2016); Renan,

supra note 74, at 1053.

238

. Renan, supra note 74, at 1053. At times, Fourth Amendment analyses

are overtly narrow, for example, looking to the specific terms of a particular

defendant’s privacy policy. United States v. Warshak, 631 F.3d 266, 287 (6th

Cir. 2010).

2019] NORMATIVE FOURTH AMENDMENT 785

In several situations, the government has done just that.

The Supreme Court’s holding that the government may collect

Michael Lee Smith’s dialed telephone numbers provided the le-

gal basis for surveillance programs targeting millions of citizens.

These include the NSA’s collection of citizens’ dialed phone num-

bers and the DEA’s decades-long program of collecting telephone

metadata on all calls from the United States to other coun-

tries.

239

Likewise, the Court’s ruling that address information on

a postal letter is unprotected by the Fourth Amendment eventu-

ally became the basis for a government program of scanning

every mailed envelope into a massive database of postal commu-

nications.

240

These aggregate programs of surveillance have a ca-

pacity to infringe on citizens’ privacy that is greater than the

sum of their parts, and they raise questions that the Court does

not even contemplate under traditional Fourth Amendment

tests.

241

The normative test is better suited for addressing wide-

spread surveillance and the collection of large databases of citi-

zens’ personal information. It directs courts to assess surveil-

lance at a programmatic level under the presumption that the

government will pursue unregulated surveillance as broadly as

resource constraints allow, as it has repeatedly done in the mod-

ern era.

242

Thus it has the benefit of aligning courts’ assessments

with the likely consequences of their decisions.

A related problem in Fourth Amendment law is that of spill-

over, meaning that information collected for one purpose may

later be used for another, more invasive or problematic pur-

pose.

243

For instance, section 702 of FISA authorizes intelligence

agencies to monitor the phone calls and electronic communica-

tions of non-U.S. persons.

244

The intelligence program, however,

239

. Smith v. Maryland, 442 U.S. 735, 745–46 (1979); see Renan, supra note

74, at 1055.

240

. Ex Parte Jackson, 96 U.S. 727, 733 (1877); Ron Nixon, Report Reveals

Wider Tracking of Mail in U.S., N.Y. TIMES (Oct. 27, 2014), https://www.nytimes

.com/2014/10/28/us/us-secretly-monitoring-mail-of-thousands.html.

241

. See Renan, supra note 74, at 1056.

242

. See supra notes 239–40; see also Renan, supra note 74, at 1059 (discuss-

ing uses of license plate scanning to monitor people’s movements).

243

. See Renan, supra note 74, at 106067.

244

. See, e.g., Erin Kelly, What Is the Section 702 Surveillance Program and

Why Should You Care?, USA TODAY (Jan. 11, 2018), https://www.usatoday.com/

story/news/politics/2018/01/11/what-section-702-surveillance-program-and

-why-should-you-care/1025582001 [https://perma.cc/5U3H-ZCN2].

786 MINNESOTA LAW REVIEW [104:741

also collects the data and communications of U.S. citizens com-

municating with non-U.S. citizens.

245

This information is then

accessible by the FBI for domestic law enforcement purposes,

and the FBI uses it “[w]ith some frequency” for purely domestic

law enforcement.

246

Similar problems arise with data collected

by private parties that is then purchased or obtained by the gov-

ernment for more invasive or de-anonymized uses.

247

Although secondary uses of information are difficult to reg-

ulate under any standard, the normative test is more compatible

with judicial scrutiny of, say, transfers of data between different

government agencies or between private data brokers and gov-

ernment officials.

248

While Katz’s expectations-of-privacy analy-

sis is largely incompatible with the concept of regulating law en-

forcement collection of already-gathered information,

249

the

normative approach could allow courts to determine that a

transfer of information to law enforcement entities is regulated

by the Fourth Amendment based on its substantial potential for

additional surveillance harms.

250

IV. OBJECTIONS AND ALTERNATIVES

Any test for the Fourth Amendment’s scope will have draw-

backs as well as advantages. Normative balancing’s advantages

are arguably essential to an effective Fourth Amendment test.

Yet objections might be raised that counsel against adopting nor-

mative balancing nonetheless. This Part responds to some po-

tential objections to a Fourth Amendment balancing test. In the

course of doing so, it touches on the deficiencies of the current

test, which carries many of the same drawbacks as the norma-

tive test with virtually none of the benefits. This Part also dis-

cusses two potential alternatives to the Katz test: the positive

245

. Id.

246

. PRIVACY & CIVIL LIBERTIES OVERSIGHT BD., REPORT ON THE SURVEIL-

LANCE PROGRAM OPERATED PURSUANT TO SECTION 702 OF THE FOREIGN INTEL-

LIGENCE SURVEILLANCE ACT 59 (2014), https://www.pclob.gov/library/702

-Report.pdf [https://perma.cc/S687-SBWH].

247

. See, e.g., Renan, supra note 74, at 106263.

248

. See generally Ric Simmons, The Mirage of Use Restrictions, 96 N.C. L.

REV. 133 (2017) (discussing the difficulty of creating effective use restrictions

on government agencies).

249

. See, e.g., United States v. Jacobsen, 466 U.S. 109, 117–18 (1984).

250

. See generally supra Part I.B.2 (discussing harms to individuals).

2019] NORMATIVE FOURTH AMENDMENT 787

law approach and an Olmstead-like, textualist approach. In do-

ing so, it develops another argument for the normative test—

even accounting for its disadvantages, the normative test is su-

perior to the alternatives.

A. ADMINISTRABILITY AND INSTITUTIONAL CAPACITY

One potential objection to the normative balancing test con-

cerns its administrability. Multifactor balancing standards tend

to be more complex and have higher decision costs than other

potential tests.

251

Relatedly, courts may lack the institutional ca-

pacity to effectively apply a balancing test. The normative ap-

proach asks judges to consider the likely effects of legal regula-

tion on police and citizen behavior, a policy inquiry that may be

better suited to a legislature.

252

Although balancing is a funda-

mental practice of courts, and the central metaphor of judging

involves balance scales, judges may be more effective applying

narrower standards or bright-line rules.

253

The normative test is designed to mitigate some of the ad-

ministrability issues and decision costs inherent in balancing

tests. It focuses on actual practices and communications as well

as measurable internal harms rather than abstract concepts of

privacy or security. It is also designed to allow judges to consult

intuitions about the potential effects of surveillance on their own

behaviors.

254

Thus it is likely to be more administrable than

many balancing tests commonly used in other areas of law.

255

Further, balancing tests in general are well suited to “rulifica-

251

. See Louis Kaplow, Rules Versus Standards: An Economic Analysis, 42

DUKE L.J. 557, 572–86 (1992). Concerns about decision costs may be mitigated

somewhat by the fact that stare decisis will resolve the vast majority of Fourth

Amendment decisions under any standard. See Allen & Rosenberg, supra note

34, at 115358.

252

. See, e.g., Orin Kerr, The Fourth Amendment and New Technologies:

Constitutional Myths and the Case for Caution, 102 MICH. L. REV. 801, 857–87

(2004).

253

. See T. Alexander Aleinikoff, Constitutional Law in the Age of Balancing,

96 YALE L.J. 943, 944 (1987).

254

. See, e.g., supra note 95 and accompanying text.

255

. See, e.g., Pike v. Bruce Church, Inc., 397 U.S. 137, 142 (1970) (weighing

the interests of states against the burdens placed on interstate commerce); Pick-

ering v. Bd. of Educ., 391 U.S. 563, 568 (1968) (weighing a government em-

ployee’s interest in free speech against the interests of the government in effi-

ciently providing public services).

788 MINNESOTA LAW REVIEW [104:741

tion,” or the gradual development of sub-rules that govern par-

ticular situations.

256

Rulification is likely to reduce decision costs

and increase administrability over time.

257

More broadly, courts appear able to effectively apply balanc-

ing tests that consider the effects of legal regulation in a wide

variety of contexts. First Amendment law is famously a domain

of balancing tests, which allow courts to robustly protect free

speech without unduly hampering legitimate government activ-

ities.

258

Similar balancing tests are also employed in the law of

equal protection,

259

procedural due process,

260

the Fifth Amend-

ment,

261

the dormant Commerce Clause,

262

torts,

263

and confi-

dentiality.

264

Although a definitive analysis of balancing in these

areas would require thousands of pages, the ubiquity of balanc-

ing tests suggests that courts are hardly incapable of applying

them.

Finally, although administrability is a concern with any bal-

ancing test, such a test could hardly be less administrable than

Katz.

265

Although Katz’s reasonable expectation of privacy test

is confusing enough on its face, the test in practice is even more

complex and puzzling. Frustrated by the failures of the Katz test

to embody important Fourth Amendment principles, courts have

expanded and modified the test haphazardly.

266

As Orin Kerr fa-

mously described, courts have created multiple, conflicting ver-

sions of the test, sometimes applying it literally, sometimes look-

ing to positive law for guidance, sometimes emphasizing the

256

. See Frederick Schauer, The Tyranny of Choice and the Rulification of

Standards, 14 J. CONTEMP. LEGAL ISSUES 803, 805–06 (2005).

257

. See Tokson, supra note 166, at 652.

258

. See Joseph Blocher, Categoricalism and Balancing in First and Second

Amendment Analysis, 84 N.Y.U. L. REV. 375, 386 (2009).

259

. Loving v. Virginia, 388 U.S. 1, 11 (1967).

260

. Mathews v. Eldridge, 424 U.S. 319, 326 (1976).

261

. New York v. Quarles, 467 U.S. 649, 65657 (1984).

262

. Pike v. Bruce Church, 397 U.S. 137, 142 (1970).

263

. United States v. Carroll Towing Co., 159 F.2d 169, 173 (2d Cir. 1947).

264

. See Ellen E. Deason, Predictable Mediation Confidentiality in the U.S.

Federal System, 17 OHIO ST. J. DISP. RESOL. 239 (2002).

265

. See, e.g., Baude & Stern, supra note 7, at 1825, 1860 (noting Katz’s no-

torious lack of administrability); see also Minnesota v. Carter, 525 U.S. 83, 97

(1998) (Scalia, J., concurring); Etzioni, supra note 6, at 420–21; Solove, supra

note 8, at 1511.

266

. See Tokson, supra note 166, at 64748.

2019] NORMATIVE FOURTH AMENDMENT 789

thing being investigated, and sometimes focusing mostly on pol-

icy considerations.

267

Lower courts applying Katz in cases of first

impression must choose between these various conflicting mod-

els, yet there is no law or norm that tells them how to make this

crucial decision.

268

Unsurprisingly, in novel cases, Fourth

Amendment law under the Katz test is unpredictable and cha-

otic.

269

By contrast, the normative test directs a court to overtly

weigh the normative considerations at issue and to explain its

actual reasons for reaching its conclusions. Not only is this a

more rigorous approach, it is a more honest one, and it can help

facilitate the development of efficient sub-rules over time.

270

B. UNPREDICTABILITY

A potential objection to normative approaches in general is

that they may be unpredictable and inconsistent across cases.

Different judges may reach conflicting normative conclusions or

may frame policy questions differently, leading to splits among

lower courts.

271

Police officers using new surveillance techniques

or facing novel situations may have difficulty determining

whether they can lawfully surveil without a warrant.

272

Ideally,

a Fourth Amendment test would be predictable and simple

enough for courts and police officers alike.

273

There are several reasons to think that unpredictability is

not as significant a problem as it may seem, however. First,

while police officers can simply follow established law in most

cases, they are unlikely to be able to resolve difficult Fourth

Amendment questions of first impression under any viable test,

267

. Kerr, supra note 223, at 507–08.

268

. Although Orin Kerr has argued that certain patterns might help guide

lower court behavior, courts appear unaware of these patterns and any such

guidelines as to model selection appear to be faint and inconsistent. See Lior

Strahilevitz & Matthew Tokson, Should Fourth Amendment Law Pay Attention

to What People Expect? If So, How?, CONCURRING OPINIONS (Nov. 27, 2017),

https://concurringopinions.com/archives/2017/11/should-fourth-amendment

-pay-attention-to-what-people-expect-if-so-how.html [https://perma.cc/EYM7

-EZTN].

269

. See Carter, 525 U.S. at 97 (Scalia, J., concurring); Solove, supra note 8,

at 1519–20; Strahilevitz & Tokson, supra note 268.

270

. Tokson, supra note 166, at 619.

271

. Kerr, supra note 223, at 53637.

272

. Amsterdam, supra note 6, at 403–04; Ohm, supra note 14, at 1333–34.

273

. Wayne R. LaFave, ‘Case-by-Case Adjudication’ Versus ‘Standardized

Procedures’: The Robinson Dilemma, 1974 SUP. CT. REV. 127, 141–42 (1974).

790 MINNESOTA LAW REVIEW [104:741

normative or otherwise. The Fourth Amendment’s remedial doc-

trines already take ample consideration of this difficulty. Quali-

fied immunity limits officers’ liability to those cases where offic-

ers violate clearly established law,

274

and the good faith doctrine

prevents the exclusion of evidence where officers rely on law that

is later overturned.

275

Even if these doctrines were to disappear,

the indemnification of police officers would prevent officers from

facing personal consequences for non-egregious legal viola-

tions.

276

Moreover, tests that are simple enough to permit police

officers to reliably answer novel Fourth Amendment questions

may be profoundly deficient in other respects, such as drastically

under-protecting privacy or protecting it in an extremely arbi-

trary manner.

277

Second, under any Fourth Amendment test, a large majority

of cases will be governed by precedent and stare decisis. The Su-

preme Court has already resolved how the Fourth Amendment

applies in a wide variety of familiar surveillance contexts, in-

cluding houses, cars, investigatory stops, inventory searches,

searches incident to arrest, border stops, and many forms of elec-

tronic surveillance.

278

These precedents should continue to guide

courts and police officers under a normative test, even as courts

discard the Katz test which provided the nominal basis for many

of their outcomes. The values of stare decisis counsel preserving

the results of these cases, upon which law enforcement officials

have long relied.

279

In addition, normative considerations often

drove the results of these cases far more than Katz’s ambiguous

“expectation of privacy” inquiry.

280

A few existing cases should

be overturned under the new test, but stare decisis suggests

overturning only cases that are especially flawed.

281

274

. Harlow v. Fitzgerald, 457 U.S. 800, 818 (1982).

275

. Illinois v. Krull, 480 U.S. 340, 353–54 (1987).

276

. Joanna C. Schwartz, Police Indemnification, 89 N.Y.U. L. REV. 885, 936

(2014).

277

. See, e.g., Olmstead v. United States, 277 U.S. 438, 464 (1928) (holding

that the Fourth Amendment is limited to physical trespasses against tangible

things).

278

. See Allen & Rosenberg, supra note 34, at 115358.

279

. See, e.g., Dickerson v. United States, 530 U.S. 428, 443 (2000).

280

. See Kyllo v. United States, 533 U.S. 27, 34, 40 (2001); Delaware v.

Prouse, 440 U.S. 648, 66263 (1979).

281

. Arizona v. Rumsey, 467 U.S. 203, 212 (1984); See infra Part V.C.

2019] NORMATIVE FOURTH AMENDMENT 791

Finally, the normative test would perform no worse than the

current test in terms of predictability and consistency. For the

reasons discussed above,

282

it is very difficult to predict how any

case of first impression will be resolved under Katz.

283

Lower

courts facing novel Fourth Amendment questions frequently

produce splits

284

or rule unanimously only to see their rulings

rejected by the Supreme Court.

285

A normative test grounded in

the analysis of actual surveillance harm and law enforcement

benefit, aided by studies of the measurable effects of surveil-

lance, would, if anything, be more consistent than the multi-

model Katz regime.

C. REDUNDANCY

Another potential argument against a balancing test for the

Fourth Amendment’s scope is that it would be redundant in

some cases because the Court sometimes uses a balancing ap-

proach in determining whether a Fourth Amendment search or

seizure is “reasonable.”

286

A test that balances to determine

whether something is a Fourth Amendment search and then

sometimes balances to determine whether that search is reason-

able would be partially redundant and could impose high deci-

sion costs on courts.

Yet courts applying a balancing test for the Fourth Amend-

ment’s scope would not have to balance again at the reasonable-

ness stage, even in the subset of cases that use a reasonableness

balancing test. First, although courts in Fourth Amendment

cases often weigh the policy implications of their rulings, overt

balancing tests are relatively rare in Fourth Amendment law,

especially in cases regulating law enforcement.

287

Courts tend to

282

. See supra notes 267–69 and accompanying text.

283

. See supra notes 267–69 and accompanying text.

284

. Wayne A. Logan, Constitutional Cacophony: Federal Circuit Splits and

the Fourth Amendment, 65 VAND. L. REV. 1137, 1195–1203 (2012) (listing nearly

forty unresolved circuit splits on Fourth Amendment issues).

285

. See Carpenter v. United States, 138 S. Ct. 2206, 2223 (2018) (rejecting

the unanimous holding of several federal courts of appeal that cell phone loca-

tion information is not protected by the Fourth Amendment).

286

. Matthew B. Kugler & Lior Jacob Strahilevitz, Actual Expectations of

Privacy, Fourth Amendment Doctrine, and the Mosaic Theory, 2015 SUP. CT.

REV. 205, 237 (2015).

287

. See, e.g., City of Indianapolis v. Edmond, 531 U.S. 32, 37 (2000) (noting

that “search or seizure is ordinarily unreasonable in the absence of individual-

ized suspicion of wrongdoing” and that the Court has “recognized only limited

792 MINNESOTA LAW REVIEW [104:741

balance in “special needs” cases that are likely to involve “mini-

mal” privacy interests and government interests other than the

traditional investigation of crime.

288

To date, special needs cases

virtually always involve seizures or very clear searches such as

building inspections.

289

The only issue is their reasonableness;

the Fourth Amendment’s application to the situation is obvious.

It might be objected that reasonableness balancing is not

limited to special needs cases, even if those cases are the only

ones that regularly employ balancing tests.

290

Courts occasion-

ally weigh competing considerations when addressing novel

questions of reasonableness.

291

But such cases almost always in-

volve obvious seizures, such as car stops and Terry stops, and

thus do not address the test for Fourth Amendment searches in

any event.

292

In addition, these cases are rare; the default rule

for searches still requires a valid warrant,

293

and the vast major-

ity of cases that depart from that rule simply apply a suspicion-

circumstances in which the usual rule does not apply”); Chandler v. Miller, 520

U.S. 305, 313 (1997) (“To be reasonable under the Fourth Amendment, a search

ordinarily must be based on individualized suspicion of wrongdoing.”). See gen-

erally Riley v. California, 134 S. Ct. 2473, 2482 (2014) (holding that a police

search of the contents of a cell phone incident to arrest was unreasonable under

the Fourth Amendment and therefore required a warrant); California v.

Acevedo, 500 U.S. 565, 574 (1991) (holding that the police can search a container

in an automobile without a warrant only if they have probable cause).

288

. Skinner v. Ry. Labor Execs.’ Ass’n, 489 U.S. 602, 624 (1989) (“In limited

circumstances, where the privacy interests implicated by the search are mini-

mal, and where an important governmental interest furthered by the intrusion

would be placed in jeopardy by a requirement of individualized suspicion, a

search may be reasonable despite the absence of such suspicion.”); see also, e.g.,

Nat’l Treasury Emps. Union v. Von Raab, 489 U.S. 656, 665 (1989) (stating that

balancing is appropriate “where a Fourth Amendment intrusion serves special

governmental needs, beyond the normal need for law enforcement”).

289

. See, e.g., Mich. Dep’t of State Police v. Sitz, 496 U.S. 444, 455 (1990);

Camara v. Mun. Ct., 387 U.S. 523, 53031 (1967).

290

. Excessive force claims are generally evaluated under a totality of the

circumstances test that may incorporate balancing. See Graham v. Connor, 490

U.S. 386, 396 (1989). These cases inherently involve seizures, thus a normative

balancing test for searches is unnecessary.

291

. See Thomas K. Clancy, The Fourth Amendment’s Concept of Reasona-

bleness, 2004 UTAH L. REV. 977, 1012 (collecting cases); Sam Kamin & Justin

Marceau, Double Reasonableness and the Fourth Amendment, 68 U. MIAMI L.

REV. 589, 602–03 (2014) (discussing the “increasingly freewheeling form of rea-

sonableness balancing” in the context of investigative stops).

292

. For additional examples, see Michigan v. Summers, 452 U.S. 692, 704–

05 (1981).

293

. See, e.g., Riley v. California, 134 S. Ct. 2473, 2482 (2014).

2019] NORMATIVE FOURTH AMENDMENT 793

based standard, such as probable cause or reasonable suspi-

cion.

294

What if a case were to someday arise that presented both a

difficult “search” question and a special needs issue or novel rea-

sonableness question that might require the court to weigh pol-

icy interests while fashioning a new rule? Even then, the norma-

tive balancing test proposed above would displace or at least

strongly inform any balancing performed at the reasonableness

stage. If a surveillance technique caused concrete harms that

outweighed its law enforcement benefits such that it required

Fourth Amendment regulation, then both that fact and the ex-

tent of the harms and benefits would inform the Court’s reason-

ableness inquiry. Most likely, no additional balancing would be

required. Even in the rarest hypothetical case, it is unlikely that

redundant balancing would be an issue.

D. BALANCING AND BIAS

Finally, a potential objection to a balancing test for the

Fourth Amendment’s scope is that such a test will be biased in

favor of the government. Several scholars have noted that courts

applying overt balancing tests to determine the reasonableness

of a seizure or search often favor the government.

295

One might

extrapolate that balancing inherently favors the government in

the Fourth Amendment context.

296

Although the government often prevails in cases where the

court departs from the default warrant requirement and engages

in balancing, it is unlikely that the balancing is to blame. Courts

typically engage in reasonableness balancing after identifying a

case as unique—as a “special needs” case rather than a normal

294

. See, e.g., City of Indianapolis v. Edmond, 531 U.S. 32, 37 (2000) (noting

that “search or seizure is ordinarily unreasonable in the absence of individual-

ized suspicion of wrongdoing” and that the Court has “recognized only limited

circumstances in which the usual rule does not apply”).

295

. Shima Baradaran, Rebalancing the Fourth Amendment, 102 GEO. L.J.

1, 1 (2013); Eve Brensike Primus, Disentangling Administrative Searches, 111

COLUM. L. REV. 254, 296–97 (2011).

296

. Richard Re, Fourth Amendment Fairness, 116 MICH. L. REV. 1409, 1419

(2018); Sundby, supra note 6, at 1765.

794 MINNESOTA LAW REVIEW [104:741

one.

297

Special needs cases are generally outside the realm of tra-

ditional law enforcement, involving non-criminal administrative

enforcement,

298

children in school,

299

non-criminal drug test-

ing,

300

and similar scenarios.

301

The paradigm special needs case

involves “privacy interests that are minimal” and “important

governmental interests.”

302

By classifying a case as special

needs, the court has largely already determined that its intru-

sions are minimal and the government’s needs unique, even be-

fore reasonableness is assessed.

303

It is little wonder that the bal-

ancing in such cases is usually resolved in the government’s

favor.

There is, in other words, a strong selection effect at work

here. Courts overtly balance only in those cases where they feel

that a default warrant requirement is inappropriate.

304

And yet,

even in this unique subset of cases, courts do not universally fa-

vor the government. For instance, the Supreme Court has ruled

against the government in cases where the justifications for a

drug testing program failed to outweigh its privacy intrusions,

305

where a blood test incident to arrest was too invasive,

306

and

where the sanctity of the home outweighed the government’s in-

terest in drunk driving enforcement.

307

In addition, overt balancing at the reasonableness stage

may favor the government in some cases because of the Court’s

lax and poorly defined reasonableness-balancing approach. It

sometimes focuses on government interests writ large and com-

pares them to the one-off harms imposed on the single defendant

297

. Fabio Arcila Jr., Special Needs and Special Deference: Suspicionless

Searches in the Modern Regulatory State, 56 ADMIN. L. REV. 1223, 1227–31

(2004).

298

. Camara v. Mun. Ct., 387 U.S. 523, 53031 (1967).

299

. New Jersey v. T.L.O., 469 U.S. 325, 33839 (1985).

300

. Nat’l Treasury Emps. Union v. Von Raab, 489 U.S. 656, 66566 (1989).

301

. See, e.g., Vernonia School Dist. 47J v. Acton, 515 U.S. 646, 65657

(1995).

302

. Skinner v. Ry. Labor Execs.’ Ass’n, 489 U.S. 602, 624 (1989).

303

. See id.

304

. See Terry v. Ohio, 392 U.S. 1, 20 (1968).

305

. Chandler v. Miller, 520 U.S. 305, 318 (1997) (striking down a statute

mandating drug testing of candidates for certain state officers).

306

. Birchfield v. North Dakota, 136 S. Ct. 2160, 2184–85 (2016); see also

Winston v. Lee, 470 U.S. 753, 766 (1985) (ruling in favor of the defendant in a

reasonableness balancing case despite the presence of a warrant).

307

. Welsh v. Wisconsin, 466 U.S. 740, 750–53 (1984).

2019] NORMATIVE FOURTH AMENDMENT 795

challenging the seizure or search.

308

This can create a sort of “im-

balancing” test that favors the government by aggregating gov-

ernment interests while failing to do the same for citizens.

309

But

the far more concrete test developed above specifically directs

courts to assess harms to citizens in the aggregate.

310

The sur-

veillance technique at issue is hypothesized to be widespread

and its targets numerous, as frequently happens when surveil-

lance goes unregulated by the Fourth Amendment.

311

A more

symmetrical balance should produce more symmetrical results.

E. POSITIVE LAW ALTERNATIVES

One model that courts have used when applying Katz looks

to positive law to determine when people have a reasonable ex-

pectation of privacy. In recent years, some scholars have sug-

gested that courts should apply this model exclusively, basing

the scope of the Fourth Amendment on what other sources of law

permit or prohibit.

312

The leading positive law proposal envisions

a test in which the Fourth Amendment applies whenever a gov-

ernment officer’s investigative action would be a violation of law,

a tort, or a use of the government’s unique legal authority.

313

Alt-

hough the positive law test offers some advantages, it has sev-

eral flaws that render it undesirable as a determinant of the

Fourth Amendment’s scope.

A positive law approach would be more predictable than

most other approaches, at least in the subset of cases where pos-

itive law is clear. For instance, if a town had an ordinance pro-

hibiting anyone but the licensed trash collecting company from

collecting people’s trash, then the police would not be able to ex-

amine trash in that town without a warrant.

314

There will be nu-

merous other cases, however, when government surveillance

presents an issue that is unresolved in existing statutes or prec-

edents. Many government surveillance practices, like the use of

308

. See Baradaran, supra note 295, at 1521.

309

. Id.

310

. See supra text accompanying notes 8789.

311

. See supra text accompanying notes 23940.

312

. See Baude & Stern, supra note 7, at 183132; Michael J. Zydney Mann-

heimer, Decentralizing Fourth Amendment Search Doctrine, 107 KY. L.J. 169

(2018).

313

. See Baude & Stern, supra note 7, at 183132.

314

. Id. at 1882.

796 MINNESOTA LAW REVIEW [104:741

drug-sniffing dogs, arise rarely, if ever, in litigation between pri-

vate parties.

315

Even those that do arise in litigation commonly

rest on open-ended standards like “reasonableness,” which are

often less developed in the context of privacy torts than they are

in Fourth Amendment law.

316

In a variety of cases, a positive law

test may simply move from a hard Fourth Amendment question

to an even harder tort question.

317

Perhaps the most serious flaw in the positive law approach

is the arbitrariness of its protections. The Fourth Amendment

would often rest on considerations that have nothing to do with

citizens’ privacy, security, or freedom from government intru-

sion.

318

Consider the trash collection example. A person’s trash,

which can reveal intimate details about activities inside their

home, would be protected in a town where laws establish a local

trash-collection monopoly and entirely unprotected in a town

without a monopoly.

319

The protection of citizens’ privacy at

home should not turn on such irrelevant details. Likewise, it

makes little difference whether the government monitors a citi-

zen by attaching a GPS device to her car or by tracking the car

with a lawfully operated drone. Yet the former would presump-

tively require a warrant, while the latter would be wholly unreg-

ulated by the Fourth Amendment.

320

It is the constant monitor-

ing of individuals, not the de minimis touching of a car, that

315

. Richard M. Re, The Positive Law Floor, 129 HARV. L. REV. F. 313, 320

(2016).

316

. See id.

317

. Id.

318

. Protecting citizens’ privacy against arbitrary government intrusion is a

fundamental purpose of the Fourth Amendment. See Boyd v. United States, 116

U.S. 616, 630 (1886) (“It is not the breaking of [a man’s] doors, and the rummag-

ing of his drawers, that constitutes the essence of the offense; but it is the inva-

sion of his indefeasible right of personal security, personal liberty, and private

property . . . .”); WILLIAM J. CUDDIHY, THE FOURTH AMENDMENT: ORIGINS AND

ORIGINAL MEANING 602–1791, at 766 (2009) (“Privacy was the bedrock concern

of the amendment, not general warrants.”); Cloud, supra note 66, at 1726 (not-

ing that “the historical record suggests that objections to general warrants and

general searches alike rested upon broad concerns about protecting privacy,

property, and liberty from unwarranted and unlimited intrusions”); Davies, su-

pra note 67, at 744–45 (arguing that “it is certainly the case that the Framers

intended to preserve a personal and domestic sphere that would be meaning-

fully protected against undue intrusions by government officers”).

319

. See Baude & Stern, supra note 7, at 1882.

320

. State laws may vary, but the Restatement (Second) of Torts § 217 sug-

gests that touching a chattel without permission would be unlawful, even if the

2019] NORMATIVE FOURTH AMENDMENT 797

invades people’s privacy and raises concerns about government

oppression. But under a positive law test, only the touching mat-

ters.

A positive law regime would have the benefits of increased

legislative control over criminal procedure, such as institutional

competence and comprehensiveness.

321

But an enhanced legisla-

tive role would also have significant drawbacks in the Fourth

Amendment context. A regime that significantly relies on legis-

lative action to address new surveillance questions would likely

be systematically under-protective of privacy.

322

The high and

growing enactment costs of legislation and the preferences of en-

trenched interest groups result in a powerful bias in favor of leg-

islative inaction.

323

Law enforcement agencies are likely to use

invasive surveillance technologies long before legislatures regu-

late them via statute.

324

A core function of the Fourth Amendment is to limit the abil-

ity of the political branches of government to compromise citi-

zens’ privacy.

325

The positive law approach would eliminate such

limits so long as legislatures allow private parties as well as of-

ficials to engage in surveillance.

326

Under the positive law model,

a determined government could permit its officials to engage in

owner could not maintain an action for damages. RESTATEMENT (SECOND) OF

TORTS § 217 (AM. LAW INST. 1965); see Laurent Sacharoff, Constitutional Tres-

pass, 81 TENN. L. REV. 877, 90607 (2014).

321

. See John Rappaport, Second-Order Regulation of Law Enforcement, 103

CAL. L. REV. 205, 23234 (2015); see also Orin S. Kerr, The Fourth Amendment

and New Technologies: Constitutional Myths and the Case for Caution, 102

MICH. L. REV. 801, 870, 875 (2004).

322

. This is especially the case for surveillance techniques that do not fit

neatly into existing privacy tort categories, such as location tracking or the col-

lection of communications metadata.

323

. See FRANK R. BAUMGARTNER ET AL., LOBBYING AND POLICY CHANGE:

WHO WINS, WHO LOSES, AND WHY 24–26, 45 (2009); Tokson, supra note 13, at

193.

324

. See Daniel J. Solove, Fourth Amendment Codification and Professor

Kerr’s Misguided Call for Judicial Deference, 74 FORDHAM L. REV. 747, 768–71

(2005). Likewise, statutes regulating evolving technologies tend to become ob-

solete quickly, and Congress has historically failed to amend such laws to ac-

commodate technological change. See id.

325

. See supra note 151.

326

. See Re, supra note 315, at 33031. Re notes that citizens will often be

unable or unwilling to engage in such surveillance, and thus often do not pre-

sent a substantial barrier to privacy-eliminating laws. Id.

798 MINNESOTA LAW REVIEW [104:741

any type of surveillance without judicial check. Relatedly, a pos-

itive law approach could result in law enforcement and national

security interest groups lobbying for diminished protections

against private surveillance.

327

This would both increase the

scope of permissible government monitoring and reduce existing

protections against intrusions by private parties.

Several other substantial objections to the positive law test

have been raised. Private intrusions and government investiga-

tions are very different, and the law has regulated them differ-

ently.

328

Treating them as the same threatens to ignore the

greater harms of government investigation in many cases and

the greater justifications for government investigation in oth-

ers.

329

Depending on how it is applied, the positive law test might

also produce absurd results. For instance, a positive law ap-

proach may find a Fourth Amendment violation when a CDC re-

searcher violates an FDA safety regulation while conducting a

blood test.

330

Significant problems also arise in cases involving

data held by third parties.

331

Ultimately, the fundamental arbi-

trariness and under-protectiveness of the positive law approach

make it an unappealing alternative to the normative model.

F. OLMSTEAD ALTERNATIVES

Justice Thomas’s dissent in Carpenter suggested an alterna-

tive approach to determining the Fourth Amendment’s scope,

grounded in text and historical practice.

332

Thomas essentially

proposed returning to the rule of Olmstead v. United States,

which held that the Fourth Amendment only applied to the tan-

327

. See id. at 329.

328

. See id. at 32124.

329

. See id.

330

. See id. at 318.

331

. Under the leading positive law approach, for example, it is difficult to

separate out uses of the government’s unique authority (which are searches)

from informal government coercion (which is not). See id. at 323. The govern-

ment’s ability to obtain information held by third parties, perhaps the central

issue of modern Fourth Amendment law, would be largely determined by the

efficacy of informal pressure to persuade telecommunications service providers

to share data. See Tokson, supra note 13, at 191 n.307.

332

. See Carpenter v. United States, 138 S. Ct. 2206, 2238 (2018) (Thomas,

J., dissenting).

2019] NORMATIVE FOURTH AMENDMENT 799

gible things mentioned in the Amendment: “persons, houses, pa-

pers, and effects.”

333

Under this approach, as Olmstead ruled,

there would be no constitutional impediment to the government

wiretapping its citizens’ phone calls.

334

Moreover, a citizen could

only assert a Fourth Amendment right in herself or her property

and would have no protectible interest in papers or property

owned by others.

335

Justice Thomas’s approach has recently at-

tracted some scholarly support.

336

Perhaps the best argument against this approach was writ-

ten in 1928, by Justice Brandeis in his influential Olmstead dis-

sent.

337

Brandeis noted that, in a world of constantly advancing

surveillance technology, a Fourth Amendment that only ad-

dressed the property-based surveillance common in the late

1700s would be “impotent and lifeless,” incapable of any mean-

ingful protection of citizens’ rights.

338

His warnings proved pres-

cient in the decades following Olmstead, when the federal gov-

ernment engaged in a widespread and flagrantly abusive

program of bugging and wiretapping citizens, including civil

rights leaders, political activists, attorneys and their clients,

journalists, and members of Congress, among others.

339

These

practices were declared unconstitutional only in 1967, when the

Supreme Court reversed Olmstead in Katz v. United States.

340

333

. Id.; see also Olmstead v. United States, 277 U.S. 438 (1928).

334

. See Olmstead, 277 U.S. at 466.

335

. See Carpenter, 138 S. Ct. at 224142 (Thomas, J., dissenting). This lim-

itation on the Fourth Amendment’s scope is often framed as textualist, but the

text of the Fourth Amendment uses plural terms, referring to a right against

unreasonable searches vested in “the people” and “their” persons, houses, pa-

pers, and effects. U.S. CONST. amend. IV. While historical practice is consistent

with a Fourth Amendment limited to trespasses on an individual claimant’s

property, the text itself is consistent with a broader, collective right. See gener-

ally David Gray, Collective Standing Under the Fourth Amendment, 55 AM.

CRIM. L. REV. 77 (2018) (arguing that the Fourth Amendment was intended to

provide broader privacy protection than was recognized by late twentieth cen-

tury Supreme Court rulings).

336

. See Jeffrey Bellin, Fourth Amendment Textualism, 118 MICH. L. REV.

(forthcoming Nov. 2019) https://papers.ssrn.com/sol3/papers.cfm?abstract_

id=3309688 (arguing also on textual grounds that a “search” should be defined

as a reasonably detailed examination of something).

337

. Olmstead, 277 U.S. at 471 (Brandeis, J., dissenting).

338

. Id. at 473.

339

. See Tokson, supra note 5, at 58384.

340

. 389 U.S. 347, 353 (1967); see also Tokson, supra note 5, at 584 n.13

(discussing the closure of the government’s wiretapping and bugging programs

800 MINNESOTA LAW REVIEW [104:741

Indeed, the most substantial flaw of an Olmstead-like ap-

proach to the Fourth Amendment’s scope would be its lack of

protection against modern forms of surveillance. Under such an

approach, the government could constitutionally wiretap citi-

zens’ phone calls, record their conversations taking place outside

the home, constantly track their locations via cell phone signal,

monitor their web surfing habits, monitor their television watch-

ing habits, record their search terms, analyze their urine, blood,

or DNA samples given to doctors or other parties, and obtain re-

cordings inside their homes from internet-connected devices like

the Amazon Echo—all without a warrant or any quantum of sus-

picion.

341

The Fourth Amendment would also not apply to gov-

ernment searches of non-residential real property, which is nei-

ther a “house” nor an “effect.”

342

While most traditional methods

of surveillance and their analogues (such as email inspections)

in the late 1960s and their clear illegality thereafter). Congress passed a law in

1934 purporting to regulate wiretapping, but the legislation was narrow and

ineffective. See Tokson, supra, at 59192. A year after Katz was decided and

several decades after the advent of widespread government wiretapping, Con-

gress passed the Wiretap Act, which (in addition to Katz) has effectively regu-

lated government and private wiretapping. See Omnibus Crime Control and

Safe Streets Act of 1968, Pub. L. No. 90-351, tit. III, 82 Stat. 212.

341

. See Bellin, supra note 336, at 58 (noting that the government could ob-

tain recordings from any Amazon Echo so long as they obtain them from Ama-

zon’s servers). Telecommunications companies and internet service providers

may have their own Fourth Amendment rights in user-related data. But users

would have no rights in such information, and the government could easily ob-

tain it by request, subpoena, or other means. The heavily regulated companies

that control such data would have strong incentives to comply with government

requests, and users would have no constitutional standing to challenge any il-

legal searches. Cf. United States v. Payner, 447 U.S. 727, 732 (1980) (holding

that a bank customer lacked standing to challenge the government’s unlawful

theft of a bank employee’s briefcase containing documents relevant to the cus-

tomer’s finances).

342

. Although the Court has traditionally protected non-residential property

under the Fourth Amendment, it would presumably not be protected under Jus-

tice Thomas’s text-and-history-based approach. See Oliver v. United States, 466

U.S. 170, 177 n.7 (1984) (noting that “[t]he Framers would have understood the

term ‘effects’ to be limited to personal, rather than real, property”); Maureen E.

Brady, The Lost “Effects” of the Fourth Amendment: Giving Personal Property

Due Protection, 125 YALE L.J. 946, 98287 (2016) (detailing how drafting history

and Founding-era sources indicate that “effects” are synonymous with personal,

chattel property); cf. Bellin, supra note 336, at 24–27 (noting the Court’s anti-

textualist precedents).

2019] NORMATIVE FOURTH AMENDMENT 801

would still be prohibited,

343

numerous forms of modern surveil-

lance would be unchecked by the Fourth Amendment.

To be sure, legislatures would eventually fill some of these

gaps.

344

But legislative protections would necessarily be defeasi-

ble and non-constitutional, subject to rapid repeal if the political

winds were to change. Justifying a general shift away from con-

stitutional protection and towards greater legislative control of

public policy would require a much broader theoretical argument

than advocates of Olmstead-based approaches have offered to

date.

345

Moreover, legislative regulation of surveillance must over-

come powerful institutional obstacles.

346

It is likely to occur, if at

all, years or decades after new technologies are used to monitor

citizens.

347

Yet legislation would be the only protection available

for most modern forms of information under Justice Thomas’s

approach. In short, an Olmstead-like regime is likely to be se-

verely under-protective of citizen privacy relative to other ap-

proaches.

V. APPLYING THE NEW MODEL

The normative approach requires courts to overtly examine

the concrete benefits and harms of government surveillance.

This direct analysis will often clarify what the “reasonable ex-

pectation of privacy” test obscures. The normative approach can

resolve novel cases more effectively and clearly than the Katz

test, which struggles with new technologies and social prac-

tices.

348

It can also provide a better foundation for cases with

sound outcomes but dubious rationales. Finally, the normative

model can reveal existing cases that are seriously flawed and

ripe for reversal.

343

. Bellin, supra note 336, at 61.

344

. Id. at 60.

345

. Strong arguments along these lines have been made by scholars criti-

cizing judicial review of statutes, although such arguments thus far have re-

ceived relatively little traction outside of the academy. E.g., Jeremy Waldron,

The Core of the Case Against Judicial Review, 115 YALE L.J. 1346 (2006).

346

. See supra notes 32224.

347

. See supra notes 323, 340; see also Tokson, supra note 13, at 193.

348

. See supra notes 22023 and accompanying text.

802 MINNESOTA LAW REVIEW [104:741

A. DECIDING FRONTIER CASES

A primary virtue of the normative test is that it can resolve

with relative ease many cases that are difficult to assess under

the Katz regime. It decides cases involving new forms of surveil-

lance effectively without bogging down in a futile inquiry about

societal expectations towards novel technologies. Indeed, the

Katz approach can leave Fourth Amendment protection for a

new technology unresolved long after its adoption by the general

public.

349

Several decades after the popularization of email, the Su-

preme Court has yet to determine whether the contents of emails

and other text-based electronic communications are protected by

the Fourth Amendment.

350

Further, the leading appeals court

case on emails declined to reach a definitive ruling, instead hold-

ing that protection for emails is dependent on the specifics of

email service privacy policies and user agreements.

351

The

Fourth Amendment would not apply, for example, to emails gov-

erned by a privacy policy that allows a service provider to inspect

or monitor a user’s emails.

352

This echoed a previous en banc de-

cision, which stated that “the expectation[ ] of privacy that com-

puter users have in their emails . . . assuredly shifts from inter-

net-service agreement to internet-service agreement,” depend-

ing on the specific terms of each agreement.

353

Whether emails are protected under the Fourth Amendment

remains unresolved outside of the Sixth Circuit, and even in that

circuit, it is unknown whether third-party email services that

electronically inspect user emails strip those emails of Fourth

Amendment protection.

354

The normative test would resolve

349

. See e.g., Kyllo v. United States, 533 U.S. 27, 40 (2001) (addressing in-

frared technology).

350

. To be sure, dicta in Carpenter suggests that the Justices intuitively fa-

vor extending Fourth Amendment protection to emails. See Carpenter v. United

States, 138 S. Ct. 2206, 2220 (2018). But the Justices have not assessed email

collection in any depth nor addressed the user agreements and electronic in-

spection issues that threaten to undermine Fourth Amendment protection for

emails.

351

. United States v. Warshak (Warshak III), 631 F.3d 266, 287 (6th Cir.

2010).

352

. Id.

353

. Warshak v. United States (Warshak II), 532 F.3d 521, 526–27 (6th Cir.

2008) (en banc).

354

. See Dana T. Benedetti, How Far Can the Government’s Hand Reach In-

side Your Personal Inbox?: Problems with the SCA, 30 J. MARSHALL J. INFO.

2019] NORMATIVE FOURTH AMENDMENT 803

these open issues definitively. The harms to individuals of wide-

spread government inspection of the contents of emails are po-

tentially enormous. There would be a profound chilling effect on

both the volume and the content of personal communications,

especially intimate or controversial communications. The scope

and vigor of the ideas conveyed via email would decrease, politi-

cal activism would be hampered, and personal relationships

would be harmed and, in some cases, substantially diminished.

At first glance, the law enforcement benefits of allowing the

government to read every citizen’s emails might also seem sub-

stantial, albeit not great enough to outweigh the enormous costs.

But the benefits to law enforcement may be far less extensive

than they initially appear. The vast majority of crimes—rob-

beries, car thefts, drug crimes, murders, assaults, etc.—are un-

likely to be discussed via email either before or after the crime.

The volume of intimate communications captured or chilled by

government observations would be exponentially higher than

the volume of emails remotely relevant to legitimate law enforce-

ment. Moreover, there is an ironic benefit to law enforcement in

confining email observation to those cases where the police have

probable cause. In a world where the police review virtually eve-

ryone’s emails, even unsophisticated criminals will avoid dis-

cussing their crimes via email or take care to securely encrypt

their emails. By contrast, the currently low probability that any

given email will be read encourages criminals to occasionally use

email in the course of their crimes. The very difficulty of gener-

ating probable cause helps ensure that, when the police do have

probable cause, they often find evidence.

355

For all of these rea-

sons, the normative test would universally protect citizens from

the routine government inspection of personal communications,

rather than leaving them unprotected or basing protection on

the unread fine print of their software user agreements.

A similar analysis could be performed for newer technolo-

gies such as smart homes and voice-controlled home speakers

like the Amazon Alexa. The chilling effects and psychological

harms inflicted by government monitoring of in-home cameras

and microphones would be massive. The benefits to law enforce-

ment would be dwarfed by such harms, and a substantial

TECH. & PRIVACY L. 75, 91 (2013).

355

. See Minzner, supra note 181, at 92325.

804 MINNESOTA LAW REVIEW [104:741

amount of criminal activity would simply be relocated to the

basement or the back yard.

B. FIXING CASES WITH UNSOUND RATIONALES

Many cases decided under the Katz test are poorly rea-

soned—full of incoherent statements about societal expectations

or unworkable standards that make a muddle of future cases.

356

Yet many of the same cases reach sound or at least defensible

outcomes.

357

The normative test can provide a more coherent jus-

tification of these outcomes and avoid the perils of expectation-

based rationales.

For example, the Court in United States v. Miller held that

the Fourth Amendment did not apply to bank records, such as

checks and deposit slips, relating to an individual’s bank ac-

count.

358

The Court dubiously asserted that customers lose any

expectation of privacy in their bank records because the records

are voluntarily conveyed to the banks and are “exposed to their

employees in the ordinary course of business.”

359

This reasoning

has been criticized extensively.

360

But the outcome of Miller, at

least as applied to account balances, checks, and deposit slips

rather than more revealing data like credit card purchase infor-

mation, is defensible under the normative test and likely unde-

serving of reversal.

To summarize, allowing the government to access bank rec-

ords is unlikely to harm interpersonal relationships or intimate

communications. There appears to be little in the psychological

literature on harmful effects of government scrutiny of one’s fi-

356

. E.g., Kyllo v. United States, 533 U.S. 27, 34 (2001) (noting that infrared

scans of houses might be permissible were “the technology in question . . . in

general public use”); California v. Ciraolo, 476 U.S. 207, 21314 (1986) (finding

no reasonable expectation of privacy against overflight observation of the back-

yard of a home despite the homeowner enclosing the yard with high double

fences).

357

. See, e.g., California v. Greenwood, 486 U.S. 35, 37 (1988).

358

. See United States v. Miller, 425 U.S. 435, 440 (1976).

359

. Id. at 442.

360

. See, e.g., Gerald G. Ashdown, The Fourth Amendment and the “Legiti-

mate Expectation of Privacy,” 34 VAND. L. REV. 1289, 131314 (1981); Katherine

J. Strandburg, Home, Home on the Web and Other Fourth Amendment Implica-

tions of Technosocial Change, 70 MD. L. REV. 614, 662 n.247 (2011) (noting that

the Miller court might have been wrong in analyzing bank records as business

records of the banks).

2019] NORMATIVE FOURTH AMENDMENT 805

nances, although surveys indicate that people find it fairly inva-

sive.

361

The potential for substantial harm may be limited, how-

ever, as the information disclosed in one’s deposit slips, negotia-

ble instruments, and account balances is unlikely to reveal

drastically more than citizens already reveal to the government

in the course of paying their taxes. Government scrutiny of bank

records may also chill certain legitimate activities in rare cases.

These might include the transfer of money to activist groups, for-

eign entities, or other lawful groups disfavored by the state.

These harms are non-trivial, albeit less profound than those at

issue in cases involving email searches or searches of the home.

Yet the criminal enforcement benefits of obtaining bank records

are substantial and unique. As the Court briefly noted in Miller,

bank records have a “high degree of usefulness in criminal, tax,

and regulatory investigations and proceedings.”

362

White-collar

investigations are unique in that they typically lack physical ev-

idence or neutral witnesses.

363

A rule that law enforcement must

have probable cause before accessing financial records “would

end many white-collar criminal investigations before they had

begun.”

364

Thus a court could hold that subpoenaing bank rec-

ords other than detailed credit card records is not a Fourth

Amendment search, reasoning that such records are not espe-

cially sensitive and their benefits to law enforcement are exten-

sive. The normative test provides a basis for Miller’s holding that

avoids the Court’s privacy-eroding rationale, which declared

that any sharing of information with a third party eliminates

Fourth Amendment protection.

365

A similar rethinking could benefit cases like Kyllo v. United

States, which held that the infrared scanning of a house was a

Fourth Amendment search.

366

Kyllo limited its holding to sur-

veillance technologies that were not in “general public use,” as

people would presumably have no expectation of privacy against

361

. Slobogin & Schumacher, supra note 133, at 73738 tbl.1. It was rated

as more invasive than questioning someone on the sidewalk for ten minutes,

but less invasive than searching a corporation’s computer. Id. The study did not

examine the harms of such surveillance, if any.

362

. Miller, 425 U.S. at 443 (quoting 12 U.S.C. § 1829b(a)(1) (1970)).

363

. See William J. Stuntz, O.J. Simpson, Bill Clinton, and the Transsub-

stantive Fourth Amendment, 114 HARV. L. REV. 842, 859–60 (2001).

364

. Id. at 860; see also Kerr, supra note 71, at 509.

365

. See Miller, 425 U.S. at 44243.

366

. Kyllo v. United States, 533 U.S. 27, 40 (2001).

806 MINNESOTA LAW REVIEW [104:741

a technology that was widely used to observe or record their ac-

tivities.

367

The normative test could resolve the issue without the

ambiguous “general public use” limitation, on the basis that in-

frared scanning to detect activities occurring inside private

homes could cause serious harms and chilling effects on a variety

of private activities within the home, and the benefits of detect-

ing mostly low-level drug crimes do not come close to justifying

such an intrusion.

368

C. IDENTIFYING AND REVERSING FLAWED CASES

The normative approach can also identify existing cases that

are especially flawed and ripe for reversal. In California v.

Greenwood, for instance, the Court held that opening citizens’

trash bags left on the curb and examining their trash is not a

Fourth Amendment search.

369

The Court reasoned that the de-

fendants had no expectation of privacy because “[i]t is common

knowledge that plastic garbage bags left on or at the side of a

public street are readily accessible to animals, children, scaven-

gers, snoops, and other members of the public” and thus “re-

spondents exposed their garbage to the public sufficiently to de-

feat their claim to Fourth Amendment protection.”

370

Although unsubstantiated claims about societal knowledge

might support the Court’s holding, the normative test does not.

A homeowner’s trash is especially revealing of the activities that

occur inside of a home, likely more revealing than the infrared

heat scan in Kyllo. If trash surveillance were to become wide-

spread, the intimate activities of the home would be exposed to

the observation and judgment of others.

371

Such observation can

lead to chilling effects or to significant psychological harm.

372

367

. Id.

368

. See id. at 37 (noting that all activities occurring within the home are

intimate activities).

369

. California v. Greenwood, 486 U.S. 35, 40 (1988).

370

. Id.

371

. Trash inspection has not yet become widespread, but the government

could lawfully embark on a program to inspect every citizen’s trash at any time,

without legal check. As discussed above, previously unthinkable programs of

surveillance often arise as the costs of information collection and processing de-

crease. See supra note 240.

372

. See Slobogin & Schumacher, supra note 133, 73738 tbl.1; sources cited

supra notes 92, 130.

2019] NORMATIVE FOURTH AMENDMENT 807

Trash surveillance also threatens intimate relationships by ex-

posing them to invasive scrutiny. The sexual and other intima-

cies of a home are revealed in its trash, and the relationships

involved may be deterred or diminished by outside observa-

tion.

373

Even the considerable law enforcement benefits of examin-

ing citizens’ trash are insufficient to justify such invasive sur-

veillance. In a world of pervasive trash inspection, criminals are

less likely to throw away incriminating documents or evidence;

the hassle of shredding or burning such evidence would be well

worth avoiding imprisonment.

374

Even setting these dynamic ef-

fects aside, trash surveillance is likely to be most effective at de-

tecting discarded drug paraphernalia, as in the Greenwood

case.

375

Not only may there be less value in pursuing low-level

drug crimes, but police may be able to investigate more serious

drug-trafficking crimes by other means. The police in Green-

wood, for instance, may have had probable cause to search

Greenwood’s house even without the trash inspection, having ob-

served heavy vehicular traffic at night, cars visiting the house at

night for only a few minutes, and a truck that drove from the

house to a narcotics-trafficking location.

376

The police might

have also generated additional proof by pulling over the visiting

cars based on reasonable suspicion of drug possession.

377

short, the normative test counsels in favor of overturning Green-

wood, an especially egregious application of the Katz test that

allows police to dig through any person’s trash without suspi-

cion. The normative approach would protect the intimate details

of people’s activities inside their homes from arbitrary govern-

ment scrutiny.

The normative model might also spur a rethinking of Ari-

zona v. Hicks, where a divided Court held that moving stereo

equipment in order to view its serial number was a Fourth

Amendment search.

378

It is likely that the de minimis harm

caused by such inspection is outweighed by the potential benefits

373

. See Gerstein, supra note 111, at 268–69; Nissenbaum, supra note 113,

at 13839.

374

. See supra text accompanying note 355.

375

. Greenwood, 486 U.S. at 38.

376

. Id. at 37.

377

. See United States v. Sharpe, 470 U.S. 675, 682 (1985).

378

. Arizona v. Hicks, 480 U.S. 321, 324–25 (1987).

808 MINNESOTA LAW REVIEW [104:741

of deterring crime and recovering stolen property, such that no

warrant should be required.

Finally, the normative approach may counsel rejecting the

emerging appeals court consensus that the Fourth Amendment

does not apply to the Internet Protocol (IP) addresses that a user

visits while surfing the internet.

379

These addresses can reveal

the content or at least the subject matter of the websites that a

user visits.

380

Such surveillance is likely to deter legitimate in-

ternet communications and research, potentially stunting intel-

lectual development and exploration.

381

Further, the evidence

generated by such monitoring is likely to be weak and circum-

stantial, while evidence of internet-based crimes can likely be

generated through less invasive means.

382

CONCLUSION

Fourth Amendment law has undergone several dramatic

shifts over the course of its history, as courts have struggled to

preserve citizens’ privacy in the face of new surveillance prac-

tices and technologies.

383

The Katz test was a particularly im-

portant shift. It allowed courts to regulate non-physical surveil-

lance practices by focusing on people’s “reasonable expectations

of privacy,” rather than on property.

384

But the test has been

deeply flawed from the start, and it is rightly criticized today as

incoherent, tautological, and arbitrary. Increasingly, as

379

. See, e.g., United States v. Ulbricht, 858 F.3d 71, 9798 (2d Cir. 2017);

United States v. Forrester, 512 F.3d 500, 510 (9th Cir. 2008). IP addresses are

sequences of numbers assigned to each computer or other Internet-enabled de-

vice that is active on a network.

380

. Matthew J. Tokson, The Content/Envelope Distinction in Internet Law,

50 WM. & MARY L. REV. 2105, 214750 (2009).

381

. See, e.g., Richards, supra note 65, at 389, 421; Marthews & Tucker, su-

pra note 90, at 37.

382

. For example, the police could obtain a warrant to capture the IP ad-

dresses that communicate with a website trafficking in child pornography or

selling illegal goods.

383

. See Olmstead v. United States, 277 U.S. 438, 464 (1928) (holding that

the Fourth Amendment is limited to tangible things); Katz v. United States, 389

U.S. 347, 353 (1967) (declaring that the Fourth Amendment’s scope is not based

on physical intrusion but is determined by expectations of privacy); United

States v. Jones, 132 S. Ct. 945 (2012) (holding that the Fourth Amendment’s

scope is also determined by trespass concepts); Florida v. Jardines, 133 S. Ct.

1409 (2013) (abandoning the trespass concept for a concept based on physical

touching and social norms).

384

. Katz, 389 U.S. at 362 (Harlan, J., concurring).

2019] NORMATIVE FOURTH AMENDMENT 809

knowledge of threats to privacy grows and an ever-greater pro-

portion of our data is made accessible to third parties, societal

expectations are unable to serve as an adequate foundation for

the Fourth Amendment’s protections.

The normative test offers a better approach to determining

the Fourth Amendment’s scope. It is both more consistent with

the historical purposes of the Amendment and far more resilient

to social and technological change. Its factors capture the funda-

mental harms of government surveillance and are firmly

grounded in precedent and pragmatic surveillance theory. Fur-

ther, its analysis is direct and transparent, avoiding false targets

and arbitrary distinctions. It is better able to address the wide-

spread surveillance programs that increasingly pose the greatest

threats to citizen security. And it provides a superior method for

deciding frontier cases and resolving controversies about exist-

ing decisions.

The Supreme Court has been slow to adopt new Fourth

Amendment paradigms in the past. It took the Court nearly forty

years to overrule Olmstead v. United States,

385

which ruled that

the Fourth Amendment did not apply to microphones or wire-

taps. During those decades, the government engaged in a mas-

sive program of bugging and wiretapping private citizens.

386

used these recordings to monitor and undermine political

groups, intimidate members of Congress, and threaten civil

rights leaders, among numerous other abuses.

387

These abuses

did not come to light until long after the damage had been

done.

388

Fourth Amendment law is in need of another paradigm

shift, one that will enable courts to protect privacy in a world of

ever-changing and expanding surveillance technologies. If his-

tory is any guide, the time for such a change is now. The norma-

tive test, like any legal test, has both advantages and drawbacks.

But in the world of modern surveillance, it offers the best way

forward for Fourth Amendment law.

385

. Olmstead, 277 U.S. at 438 (1928).

386

. See, e.g., Tokson, supra note 133, at 583.

387

. Id.

388

. See FINAL REPORT OF THE SELECT COMM., supra note 205, at 18385,

198201.