Amicus Curiae Brief — Actavis Holdco U.S., Inc., et al., Petitioners v. Connecticut, et al.

No. 19-1010

IN THE

Supreme Court of the United States

_______________

ACTAVIS HOLDCO U.S., INC., ET AL.,

Petitioners,

v.

CONNECTICUT, ET AL.,

Respondents.

_______________

On Petition for a Writ of Certiorari to the

United States Court of Appeals

for the Third Circuit

_______________

MOTION FOR LEAVE TO FILE BRIEF FOR

AMICI CURIAE AND BRIEF OF TWELVE

COMPANIES AND ROBERT D. OWEN AS AMICI

CURIAE IN SUPPORT OF PETITIONERS

_______________

ROBERT D. OWEN

COUNSEL OF RECORD

EVERSHEDS SUTHERLAND (US) LLP

900 N. Michigan Avenue

Suite 1000

Chicago, IL 60611-6521

(312) 724-9006

RobertOwen@eversheds-sutherland.com

Counsel for Amici Curiae

i

MOTION

Pursuant to Rule 37.2(b) of the rules of this

Court, Robert D. Owen and the below-listed

companies respectfully move for leave to file a brief as

amici curiae in support of Petitioners in the abovecaptioned case:

Microsoft Corporation

GlaxoSmithKline PLC

American International Group, Inc.

Oppenheimer & Co., Inc.

American Electric Power Company, Inc.

Kason Industries, Inc.

3M Company

Oracle America, Inc.

Genentech, Inc.

Deere & Company

Baxter International Inc.

Equitable Holdings, Inc.

Amici tender their proposed brief with this motion.

Counsel for amici have provided notice and

sought consent from all parties to this action. All

Respondents who are Plaintiffs-Respondents below

have given their consent, but we have not received

consent individually from Respondents who are nonpetitioning defendants, possibly because of the

disruption of personal schedules that the COVID-19

virus is today causing in our country. No party has

refused to give consent, but considering the number of

parties and in the interest of time and efficiency, amici

respectfully move for leave to file the proposed brief.

ii

Amici corporations are regular litigants in the

federal courts. Amici corporations and Mr. Owen, who

is an ediscovery practitioner and litigator, are

invested in the development of the Federal Rules of

Civil Procedure and civil discovery jurisprudence. The

Order of the Special Master and the District Court in

this case (the “Order”), as affirmed 2-1 by the Third

Circuit, deeply concerns amici. Specifically, the Order

is the most recent, and most distressing, example of a

growing trend whereby requesting parties seek access

to massive amounts of data without the protection of

a relevance review by the producing parties. Amici are

concerned about their own rights as non-parties who

might share confidential and proprietary information

with Petitioners, and about the privacy rights of

innocent bystanders whose personal and confidential

information, wholly irrelevant to the dispute, will be

produced to Plaintiffs-Respondents and accessed by

opposing counsel, document review vendors, and

potentially hackers or others with malevolent

intentions. Amici have a substantial interest in

protecting these privacy interests in the present

dispute and in future litigation.

Amici respectfully submit that the attached

brief setting forth their views will be helpful to the

Court in its consideration of these important issues

and request that the Court grant leave to file the brief

tendered with the motion.

iii

Respectfully submitted,

Robert D. Owen

COUNSEL OF RECORD

EVERSHEDS SUTHERLAND (US) LLP

900 North Michigan Avenue, Suite 1000

Chicago, Illinois 60611-6521

(312) 724-9006

RobertOwen@eversheds-sutherland.com

Stacey M. Mohr

Michelle McIntyre

EVERSHEDS SUTHERLAND (US) LLP

999 Peachtree Street, NE, Suite 2300

Atlanta, Georgia 30309-3996

(404) 853-8000

StaceyMohr@eversheds-sutherland.com

MichelleMcIntyre@eversheds-sutherland.com

Counsel to Amici Curiae

iv

TABLE OF CONTENTS

MOTION ............................................................................. i

TABLE OF AUTHORITIES ...................................... vi

INTERESTS OF THE AMICI CURIAE .................... 1

SUMMARY OF THE ARGUMENT ........................... 3

ARGUMENT ............................................................... 8

I.

WHY IS THE ORDER AND OTHERS

LIKE IT SO CONCERNING TO AMICI

CURIAE? .......................................................... 8

A.

The Order and Others Like It

Threaten Real Harm to the

Proprietary and Privacy Interests

of Amici Curiae. ..................................... 9

B.

The Order and Others Like It

Threaten Real Harm to the

Privacy Interests of Companies

and Individuals Unrelated to the

Action ................................................... 11

C.

The Mere Entry of a Protective

Order Does Not Protect Against

the Theft of Valuable Information

Resident in Produced Emails. ............ 13

D.

The Order and Others Like It

Threaten Conflict With Foreign

Data Production Laws and Real

Harm to International Relations. ....... 15

v

II.

III.

THE ORDER CONTRAVENES

EXISTING RULES IN MULTIPLE

WAYS. ............................................................ 17

A.

FED. R. CIV. P. 26(b)(1) ........................ 17

B.

FED. R. CIV. P. 34(b) ............................ 20

CERTIORARI IS URGENTLY

NECESSARY. ................................................ 23

CONCLUSION ......................................................... 23

vi

TABLE OF AUTHORITIES

CASES

In re Actavis Holdco U.S., Inc.,

No. 19-3549 (3d Cir. Dec. 6, 2019) ........................ 6

In re Air Cargo Shipping Servs.

Antitrust Litig.,

278 F.R.D. 51 (E.D.N.Y. 2010) ............................ 16

Burton Mech. Contractors, Inc. v.

Foreman, 148 F.R.D. 230 (N.D. Ind.

1992) .................................................................... 10

Caves v. Beechcraft Corp., Case No. 15CV-125, 2016 WL 355491 (N.D.

Okla., Jan. 29, 2016) ............................................. 5

City of Ontario, Cal. v. Quon, 560 U.S.

746 (2010) ............................................................ 12

In re Ford Motor Co.,

345 F.3d 1315 (11th Cir. 2003) ............................. 3

In re Lithium Ion Batteries Antitrust

Litig., No. 13MD02420, 2015 WL

833681 (N.D. Cal. Feb. 24, 2015) .......................... 5

Regan-Touhy v. Walgreen Co., 526 F.3d

641 (10th Cir. 2008) ............................................ 22

Sanderson v. Winner,

507 F.2d 477 (10th Cir. 1974) ............................... 9

vii

Schlagenhauf v. Holder,

379 U.S. 104 (1964) ............................................. 19

Seattle Times Co. v. Rhinehart, 467 U.S.

20 (1984) .............................................................. 10

Société Nationale Industrielle

Aérospatiale v. U.S. District Court for

the Southern District of Iowa,

482 U.S. 522 (1987) ................................. 15, 16, 17

U.S. ex rel. Chandler v. Cook Cty., Ill.,

277 F.3d 969 (7th Cir. 2002) ............................... 12

Youngevity Int'l Corp. v. Smith, No.

16CV00704, 2017 WL 6541106 (S.D.

Cal. Dec. 21, 2017)............................................. 4, 5

RULES & STATUTES

CAL. CIV. CODE § 1798.155 ........................................ 13

CAL. CIV. CODE § 1798.175 ........................................ 12

FED. R. CIV. P. Rule 26 ....................................... 14, 19

FED. R. CIV. P. 26(b)(1) ......................... 3, 6, 17, 18, 23

FED. R. CIV. P. 26(b)(2)(C) .................................... 8, 18

FED. R. CIV. P. Rule 34 ............................................. 20

FED. R. CIV. P. 34(b)(1)(a) ..................................... 5, 21

FED. R. CIV. P. 34(b)(2)(B) ........................................ 20

viii

SUPREME COURT Rule 37.6 ......................................... 1

OTHER AUTHORITIES

Alistair Barr, “Mozilla CEO Brendan

Eich Steps Down,” WALL ST. J. (Apr.

3, 2014) ................................................................ 14

Michael Bahar, et al., The state of US

data privacy and cybersecurity laws

in 2019 (updated Dec. 2019),

https://us.eversheds-sutherland.com

/NewsCommentary/Articles/227590/T

he-state-of-US-data-privacy-andcybersecurity-laws-in-2019. .......................... 12, 13

Committee Notes on Rules – 2015

Amendment. ........................................................ 20

Samantha Cutler, The Face-Off Between

Data Privacy and Discovery: Why

U.S. Courts Should Respect EU Data

Privacy Law When Considering the

Production of Protected Information,

59 B.C. L. Rev. 1513 (2018) ................................ 16

J. Randolph Evans and Shari L. Kleven,

Cybersecurity: You Can’t Afford to

Ignore It Anymore, DAILY REPORT

(Apr. 25, 2016) ....................................................... 7

ix

Glenn Lammi, “Supreme Court Must

Take Action on Lawless Discovery

Order in Generic-Drug Antitrust

MDL,” FORBES (March 6, 2020)

https://www.forbes.com/sites/wlf/202

0/03/06/supreme-court-must-takeaction-on-lawless-discovery-order-ingeneric-drug-antitrustmdl/#3b0f8c037b3e ................................................ 7

Nicole Hong & Robin Sidel, Hackers

Breach Law Firms, Including

Cravath and Weil Gotshal, WALL ST.

J., Mar. 29, 2016. ................................................. 14

David Kessler, et al., Protective Orders

in the Age of Hacking, N.Y.L.J., Mar.

16, 2015................................................................ 13

Kirk Roberts, et al., “Overview of the

TREC 2018 Precision Medicine

Track,” Text Retrieval Conference,

https://trec.nist.gov/pubs/trec27/pape

rs/Overview-PM.pdf .......................................... 5, 6

Roberts, C.J., “2015 Year-End Report on

the Federal Judiciary” .................................... 4, 19

Katherine Rosman, They Paid $42 for a

SoulCycle Ride, Not for Trump, N.Y.

TIMES (Aug. 8, 2019),

https://www.nytimes.com/2019/08/08/

us/politics/soulcycle-equinoxboycott.html ......................................................... 14

x

The Sedona Conference, International

Principles on Discovery, Disclosure &

Data Protection in Civil Litigation,

https://thesedonaconference.org/publi

cation/International_Litigation_Prin

ciples .................................................................... 16

The Sedona Conference, Practical inHouse Approaches for Cross-Border

Discovery & Data Protection, 17

SEDONA CONF. J. 397 (2016)................................ 17

Wright & Miller, 8A FEDERAL PRACTICE

AND PROCEDURE § 2211 ....................................... 22

1

INTERESTS OF THE AMICI CURIAE1

Amici Curiae corporations are regular litigants all

over the country:

Microsoft Corporation

GlaxoSmithKline PLC

American International Group, Inc.

Oppenheimer & Co., Inc.

American Electric Power Company, Inc.

Kason Industries, Inc.

3M Company

Oracle America, Inc.

Genentech, Inc.

Deere & Company

Baxter International Inc.

Equitable Holdings, Inc.

Amicus Curiae Robert D. Owen is a Partner at

Eversheds Sutherland (US) LLP. He is a nationally

recognized expert in e-discovery and an experienced

litigator. He is President of the Electronic Discovery

1 Pursuant to Rule 37.6, Amici Curiae state that no counsel

for a party authored this brief in whole or in part, and no person

or entities other than Amici Curiae, their members or counsel

made a monetary contribution to the preparation of submission

of the brief. Counsel for Amici Curiae provided the requisite 10

days’ notice to all parties and received consent from Petitioners,

State Attorney General Respondents, Direct Purchaser

Respondents, End-Payor Respondents, Indirect-Reseller Respondents, and Direct Action Respondents as well as some but not all

of the Respondents who are non-petitioning defendants.

2

Institute and editor-in-chief of EDI’s The Federal

Judges’ Guide to Discovery (3rd ed. 2017).

Amici respectfully submit this brief in support of

Petitioners Actavis Holdco, Inc., et al. urging this

Court to grant Petitioners’ petition for certiorari

seeking review of the decision of the United States

Court of Appeals for the Third Circuit. That court

improperly held, 2-1, that the district court could order

Petitioners to produce irrelevant information via a

flawed, one-round keyword search methodology that

(i) gathered a vastly overinclusive collection of emails

containing search term “hits” and (ii) prohibited

Petitioners from removing irrelevant emails prior to

production, the latter being the central error and clear

departure from decades of practice to the contrary.

All amici defend claims all over the Nation, and

some are seeing an increasing prevalence of requests

to produce all documents in custodians’ email

accounts, or at least those containing keyword search

terms. There is also a real danger that the confidential

and valuable commercial information of some amici

will exist in the document productions of defendants

pursuant to the Order. All amici are deeply concerned

about the effect of this ruling on actions in which they

are parties, view the Order at issue here with alarm,

and respectfully urge herein that the Court arrest this

trend’s progress.

3

SUMMARY OF THE ARGUMENT

Amici Curiae urge the Court to take up a recurring

issue that has arisen because of the data explosion of

the last few decades: Can parties be ordered to

produce – in violation of the plain limitations of Rule

26(b)(1) – vast collections of documents without being

afforded the opportunity to remove their own – and

third parties’ – irrelevant personal, private, trade

secret, or confidential material?

Our Nation’s system of full pretrial disclosure – so

idealistic and workable in 1938 when the

transformative Federal Rules of Civil Procedure were

adopted – never anticipated a reality in which

potentially disclosable material in a civil action could

comprise millions or billions of pages. Nor could the

architects of full disclosure have anticipated that

properly discoverable documents might exist literally

alongside and intermingled with irrelevant material

containing commercial information of extraordinary

value to parties as well as nonparties, or the highly

personal information of nonparties who are but

bystanders to the dispute.

Over time, the evolving reality of astronomical

data volumes required courts, lawyers, and parties to

adapt discovery practices to the new reality. Modes of

coping came into being, a central tenet of which is that

producing parties have the right and a reasonable

opportunity to identify and remove irrelevant

documents or content from collections of materials

being produced to requesting parties.

4

Full pretrial discovery is of course a necessary

component of our system for resolving civil disputes,

but it comes at a cost, sometimes to the personal

privacy of individuals and the confidential

information of businesses. We tolerate the cost

because it facilitates the peaceful resolution of civil

disputes, but discovery is made subject to strict

limiting rules to ensure that no more privacy is

compromised than necessary. Our “civil tribunals . . .

must be governed by sound rules of practice and

procedure.” Roberts, C.J., “2015 Year-end Report on

the Federal Judiciary,” at 2 (comparing our civil

justice system to the “inherently uncivilized” practice

of dueling). The instant case brings Chief Justice

Roberts’ admonition into sharp focus.

The district court order at issue (the “Order”)

requires defendants to produce every single email of

hundreds of custodians that contain any one of

hundreds of keyword search terms. Keyword

searching of the type ordered here is an accepted

method for pulling potentially responsive documents

from a large collection of data. Yet, it is also well

accepted that keywords inevitably pull huge volumes

of emails having no connection to the case that, of

course, contain much irrelevant information. Keyword

hits and relevance “are not synonymous.”2 “Search

terms are an important tool parties may use to

identify potentially responsive documents in cases

involving substantial amounts of ESI. Search terms do

2 Youngevity Int'l Corp. v. Smith, No. 16CV00704, 2017 WL

6541106, at *10 (S.D. Cal. Dec. 21, 2017).

5

not, however, replace a party’s requests for

production,”3 which must be stated with specificity in

the first instance. 4

In one recent study conducted by the National

Institute of Standards and Technology (“NIST”), an

agency of the U.S. Commerce Department, even the

most carefully designed keyword search achieved only

44.6% precision, meaning 55.4% of the “hits” were

nonresponsive.5 While keyword searches of the type

specified in the Order are commonly used in modern

litigation, because of their imprecision as evidenced by

the NIST study, they are only the first step in a

methodology that also requires a substantive review of

the “hits” to separate relevant information from the

many false-positive retrievals. The Order improperly

omits this critical step from the keyword methodology

3 Id. (citing In re Lithium Ion Batteries Antitrust Litig., No.

13MD02420, 2015 WL 833681, at *3 (N.D. Cal. Feb. 24, 2015)

(keyword searches “often are overinclusive, that is, they find

responsive documents but also large numbers of irrelevant

documents.”)

See Fed.R.Civ.P. 34(b)(1)(A) (“The request: must describe

with reasonable particularity each item or category of items to be

inspected.”); Caves v. Beechcraft Corp., Case No. 15-CV-125, 2016

WL 355491 (N.D. Okla., Jan. 29, 2016) (denying motion to compel

and sustaining defendant’s objections to document requests that

do not identify with reasonable particularity what is being sought

because: “Neither Defendants nor the Court should have to guess

what Plaintiff is really seeking.”)

4

5 Kirk Roberts, et al., “Overview of the TREC 2018 Precision

Medicine Track,” Text Retrieval Conference, https://trec.nist.gov

/pubs/trec27/papers/Overview-PM.pdf (last accessed Mar. 15,

2020.)

6

and flies in the face of studies like NIST that prove

keywords without a substantive review of “hits”

actually identify as much – or more – irrelevant

information as relevant: in the NIST study, the

median precision of all searches was 26.7%, so that

73.3% of the material tagged by search terms was

irrelevant.6

Egregiously, and of urgent concern to these amici,

the Order forbids the producing parties from removing

any non-responsive emails from those to be produced

to plaintiffs-respondents, so that “[t]here is no dispute

that the order compels the production of a volume of

non-responsive and irrelevant emails.”7

The Order thereby disregards the discovery limits

of FED. R. CIV. P. 26(b)(1), and requires production of

documents and information concededly unnecessary

for the resolution of the dispute, undoubtedly

including material in emails that the senders thought

would remain forever private: e.g., trade secrets,

personal health information protected by HIPAA,

Social Security numbers, expert consultants’

communications and reports, login and password

credentials, valuable personally identifiable information, or even intensely personal, amorous

communications. Under the Order, all of this material

must be produced to plaintiffs and be exposed to

innumerable persons at plaintiffs’ law firms and

document review vendors, not to mention any

6 Id.

7 In re Actavis Holdco U.S., Inc., No. 19-3549 (3d Cir. Dec. 6,

2019)(order denying mandamus) (Phipps, J., dissenting).

7

malevolent hackers who may gain access to their data

stores.8 Amici’s proprietary and confidential information may also be included in that production.

As an article concerning this case published

recently by Forbes states, “Employees’ private

conversations, personal business transactions, and

financial data, irrelevant for the lawsuits but

otherwise potentially compromising, could show up in

produced emails.”9 This presents a severe problem for

nonparty entities such as amici, not to mention the

parties to this case and individual persons. The sweep

of the Order also compromises the ability of our civil

litigation system to coexist with the personal privacy

regimes of other countries, most notably in the

European Union.

Requests by plaintiffs for unfiltered production of

data stores are becoming widespread nationally.

Consequently, five other circuits have granted

mandamus of similar orders – having found that

expanding the scope of discovery is not within the

It is now generally understood that law firms are

particularly vulnerable to data breaches and are being

specifically targeted for attacks. J. Randolph Evans and Shari L.

Kleven, Cybersecurity: You Can’t Afford to Ignore It Anymore,

DAILY REPORT at 1 (Apr. 25, 2016) (noting U.S. Department of

Justice investigation in wake of Russian hacker attacks on 48 law

firms and aftermath of the Panama Papers hacking attack),

available at http://tiny.cc/u2x3kz.)

8

Glenn Lammi, “Supreme Court Must Take Action on

Lawless Discovery Order in Generic-Drug Antitrust MDL,”

FORBES (March 6, 2020), https://tinyurl.com/se5crdr.

9

8

discretion of the district court10 – and none have

allowed such orders. The Order represents by far the

most expansive application of this developing

discovery thrust and if left in place it will inevitably

encourage more such attempts in matters large and

small.

As national conceptions of personal privacy evolve

and tighten, and requests for unfiltered productions of

confidential business information become more

prevalent, the need grows for clarity around the production of personal or otherwise confidential emails in

civil action pretrial discovery. Review of the Order by

this Court is urgently needed.

ARGUMENT

I.

WHY IS THE ORDER AND OTHERS LIKE

IT SO CONCERNING TO AMICI CURIAE?

The Order is far from a “mere discovery order”

unworthy of review by this Court, as Respondents

suggest. Respondents Opp. to Stay, at 1. It threatens

real harm to non-parties such as amici, parties, and

individuals that protective orders cannot prevent, and

typifies a growing trend of openly seeking production

10 FED. R. CIV. P. 26(b)(2)(C) states unequivocally that “the

court must limit . . . discovery [that] is outside the scope

permitted by Rule 26(b)(1).” This was echoed by the dissent in the

Third Circuit: “nothing in the civil rules permits a court to compel

production of non-responsive and irrelevant documents at any

time.” App. D at 3 n.1

9

of irrelevant material. The Order and others like it

should be decisively rebuffed now.

A. The Order and Others Like It Threaten

Real Harm to the Proprietary and

Privacy Interests of Amici Curiae.

In the normal course of their businesses, amici

curiae companies, like all companies, regularly

exchange with other entities highly confidential and

valuable information via email. Because of (i) the

excessive sweep and inherent limitations of one-round

keyword searching, (ii) the fact that the emails of

literally hundreds of custodians will be searched, and

(iii) the fact that the producing parties are forbidden

from removing irrelevant emails from the productions

the Order requires them to make, there is real danger

that some of the nonparty amici’s confidential and

valuable commercial information will exist in the

document productions of defendants pursuant to the

Order.

Thus, the procedure specified in the Order could

lead to the unnecessary production in this case and

others of irrelevant but valuable trade secrets and

other confidential information (e.g., confidential

pricing, bank account credentials and information,

financial statements). See, e.g., Sanderson v. Winner,

507 F.2d 477 (10th Cir. 1974) (mandamus in antitrust

action to prevent disclosure of private financial

records). To make matters worse, the third parties

such as amici may never discover that their

proprietary and confidential information was

disclosed.

10

Although the producing parties have a limited

period of time within which to clawback emails they

claim are irrelevant, nonparties including amici (i)

will have no role at all in selecting emails for clawback

and (ii) even if ultimately clawed back their

proprietary and confidential information will have

been exposed to all parties in this MDL for a period of

time. In Seattle Times Co. v. Rhinehart, 467 U.S. 20,

35 (1984), the Court recognized that incidental release

of third parties’ irrelevant information could damage

their privacy interests, and courts’ processes should be

managed to prevent such abuse:

“[R]elevant information in the hands of third

parties may be subject to discovery.

“There is an opportunity, therefore, for

litigants to obtain — incidentally or purposefully — information that not only is irrelevant

but if publicly released could be damaging to

reputation and privacy. The government

clearly has a substantial interest in

preventing this sort of abuse of its processes.”

Moreover, parties do not have standing to protect

proprietary or confidential business information of

third parties, see Burton Mech. Contractors, Inc. v.

Foreman, 148 F.R.D. 230, 234 (N.D. Ind. 1992)

(collecting cases), so it is up to those third parties, like

amici here, or the courts to protect those privacy

interests.

11

B. The Order and Others Like It Threaten

Real Harm to the Privacy Interests of

Companies and Individuals Unrelated to

the Action.

It is undeniable that individuals across all

businesses routinely use their business email accounts

to send personal material to others. Social Security

numbers, logins and passwords, protected health

information, even amorous communications – there is

an endless list of sensitive information that can be

found in business email repositories. The Court has

noted, in discussing the new era of electronic

communication, that “many employers expect or at

least tolerate personal use of [electronic communication] equipment by employees because it often

increases worker efficiency.” City of Ontario, Cal. v.

Quon, 560 U.S. 746, 759 (2010).

Producing any and all “hits” resulting from

hundreds of search terms applied to hundreds of

custodian email accounts ensures that there will be

production of the custodians’ highly personal,

sensitive, and revealing emails to strangers they will

never know.11 The custodians’ email correspondents –

11 Keywords Cannot Prevent Private or Protected Information

From Being Retrieved. Keyword searching like that at issue here

is designed to assemble an overinclusive set of data from which

responsive material can be pulled and produced, but it is not

designed to prevent nonresponsive or private material from

appearing in the tagged data set. A simple Google search (which

is simply a keyword search like that at issue here) using just one

of the hundreds of search terms proposed (see App. H to

Petitioners Application for Stay Pending Certiorari) quickly

makes concrete how innocent-sounding keywords can lead to

12

third parties outside the producing party as well as

countless employees within it – will also have their

emails exposed to anonymous persons. See, e.g., In re

Ford Motor Co., 345 F.3d 1315, 1316 (11th Cir. 2003)

(mandamus “[i]n the context of discovery orders” is

appropriate to prevent “invasion of privacy rights.”);

U.S. ex rel. Chandler v. Cook Cty., Ill., 277 F.3d 969,

981 (7th Cir. 2002) (mandamus appropriate to prevent

“serious harm to patients’ privacy rights”). The more

copies of this valuable or private data that are

distributed widely, the more chances there are for it to

be accessed without authorization, or in the present

age of hacking, simply stolen.

Protecting the privacy of employee data is

increasingly important for amici companies and

others. California recently enacted the California

Consumer Privacy Act (CCPA), a sweeping privacy

law that echoes the EU’s GDPR and regulates

companies’ use of personal data. CAL. CIV. CODE

§ 1798.175. The CCPA is likely only the first of many

such laws to come in the United States intended to

protect the personal and private data of individuals.12

documents that are embarrassing and could lead to the

inadvertent release of private information. For example,

searching on one of those search terms, “play nice,” returns a

variety of websites, including many having pornographic

overtones.

12 New York, Nevada, and Massachusetts are among the many

states enacting data privacy legislation. Michael Bahar, et al.,

The state of US data privacy and cybersecurity laws in 2019

(updated Dec. 2019), https://us.eversheds-sutherland.com

13

Broad discovery orders, like the Order here,

contravene the purpose of regulations like the CCPA

and the growing trend of protecting privacy rights,

hamper companies’ compliance with statutes like the

CCPA, and subject them to penalties. CAL. CIV. CODE

§ 1798.155.

Discovery is a serious intrusion into the privacy

rights of the parties as well as others having nothing

to do with the dispute. As a society, we allow it because

it is necessary for the truth-seeking needs of the

judicial system, but it necessarily has its carefully

crafted limits and they should be respected. The Order

does not do so.

C. The Mere Entry of a Protective Order

Does Not Protect Against the Theft of

Valuable

Information

Resident

in

Produced Emails.

The mere entry of protective orders does not

protect against cybersecurity breaches or other risks

of disclosure or misuse.13 Moreover, such orders do not

ensure there will be no disloyalty among document

/NewsCommentary/Articles/227590/The-state-of-US-dataprivacy-and-cybersecurity-laws-in-2019.

13 See, e.g., David Kessler, et al., Protective Orders in the Age

of Hacking, N.Y.L.J., Mar. 16, 2015 (“As discovery has become

predominately digital, producing parties must now face the

threat of third parties stealing highly sensitive information not

just from their and their advisor's computer systems, but their

opponents' data systems as well.”).

14

reviewers given access to the documents produced

pursuant to the Order.

The cybersecurity risk is real. At a time when even

blue chip corporate law firms –e.g., Cravath and Weil

Gotshal14 – have been hacked and bad actors

worldwide are continuing relentlessly to target major

US entities, the risks that a sensationalistic target

like personal emails of hundreds of custodians will be

vulnerable to hacking secure should not be

underestimated.15

Nor does the 120-day clawback feature ameliorate

any of the foregoing risks. Rule 26 strictly forbids the

production of irrelevant material because requesting

parties and their counsel, like the Petitioners, are not

entitled even to see such material, and once seen, the

Nicole Hong & Robin Sidel, Hackers Breach Law Firms,

Including Cravath and Weil Gotshal, WALL ST. J., Mar. 29, 2016.

14

15 In addition, at a time when partisan political emotions are

particularly pronounced, there can be no guarantee that every

one of the dozens of persons given access to the personal

information of the defendants’ custodians will honor their

obligation to preserve confidentiality if the reviewer opposes

what he or she comes to learn about the custodian’s political

views while reviewing emails. See, e.g., Alistair Barr, Mozilla

CEO Brendan Eich Steps Down, WALL ST. J. (Apr. 3, 2014)

(Mozilla CEO steps down following criticism over political

donation); Katherine Rosman, They Paid $42 for a SoulCycle

Ride, Not for Trump, N.Y. TIMES (Aug. 8, 2019),

https://www.nytimes.com/2019/08/08/us/politics/soulcycleequinox-boycott.html (customers boycott and fitness chains do

damage control after news of stakeholder holding fundraiser for

President Trump).

15

bell cannot be unrung.16 A clawback procedure does

not cure or mitigate this error.

D. The Order and Others Like It Threaten

Conflict With Foreign Data Production

Laws and Real Harm to International

Relations.

The Order creates even more harm because

Petitioners include multinational companies that do

business globally. Not only does this mean that

Petitioners do confidential business outside the

United States that is irrelevant to the claims and

defenses of the instant matter, but they have

employees, vendors and clients who live and operate

outside the United States and who of course

communicate by email with colleagues in this country.

The personal information of some of these employees,

vendors, and clients will be protected by non-U.S. data

protection laws, such as the European Union’s

General Data Protection Regulation (GDPR), that can

conflict with United States discovery. See. e.g. Société

Nationale Industrielle Aérospatiale v. U.S. District

Court for the Southern District of Iowa, 482 U.S. 522

(1987).

It is already difficult for businesses to comply with

their discovery obligations in U.S. courts while

complying with their data protection obligations

16 This is all the more problematic as, increasingly, the larger

firms bring a larger share of class actions. Their exposure to

irrelevant, confidential information, which cannot be erased from

their minds, cannot be waved away as inconsequential.

16

abroad. See generally., The Sedona Conference, International Principles on Discovery, Disclosure & Data

Protection in Civil Litigation.17 Compelling the

unnecessary production of admittedly irrelevant

documents almost certainly conflicts with the GDPR

because the legitimate interest in producing

irrelevant information arguably does not outweigh the

data subject’s interest in not having its personal

information disclosed.18

Allowing the discovery practice ordered by the

District Court to become prevalent would surely risk

further disapproval of our practices and exacerbate

issues of international comity. As this Court stated in

Aerospatiale:

“American courts . . . should exercise

special vigilance to protect foreign litigants

from the danger that unnecessary, or unduly

burdensome, discovery may place them in a

17https://thesedonaconference.org/publication/International_

Litigation_Principles (last visited Mar. 15, 2020).

The fact the company is required to produce the irrelevant

personal information under court order does not necessarily

resolve the conflict and protect the business from sanction under

E.U. law. See Samantha Cutler, The Face-Off Between Data

Privacy and Discovery: Why U.S. Courts Should Respect EU Data

Privacy Law When Considering the Production of Protected

Information, 59 B.C. L. REV. 1513, 1524-32 (2018) (describing

conflict between scope of federal discovery and expanding use of

sanctions for data privacy breaches in Europe); In re Air Cargo

Shipping Servs. Antitrust Litig., 278 F.R.D. 51, 54 (E.D.N.Y.

2010) (compelling production despite potential sanctions under

French discovery blocking statute).

18

17

disadvantageous position. Judicial supervision of discovery should always seek to

minimize its costs and inconvenience and to

prevent improper uses of discovery requests.

When it is necessary to seek evidence abroad,

however, the district court must supervise

pretrial proceedings particularly closely to

prevent discovery abuses. . . . Objections to

“abusive” discovery that foreign litigants

advance should therefore receive the most

careful consideration.”

Id. at 546. Discovery in the federal courts is already

considered very broad by those outside the United

States, The Sedona Conference, Practical in-House

Approaches for Cross-Border Discovery & Data

Protection, 17 SEDONA CONF. J. 397, 407 (2016) (“Some

civil law countries also have enacted blocking statutes

to curb the broad reach of discovery from the U.S.”),

but parties have always been able to defend it by

arguing that only documents probative to the dispute

will be produced and parties will be able to minimize

the impact on the data protection rights of non-U.S.

data subjects. The Order sharply undercuts that

defense.

II.

THE ORDER CONTRAVENES EXISTING

RULES IN MULTIPLE WAYS.

A. FED. R. CIV. P. 26(b)(1)

The Advisory Committee on Rules of Civil

Procedure has responded to the explosion in data

volumes by proposing, in successive steps over several

decades, to narrow the scope of discovery set forth in

18

this rule, and the Court has adopted those proposals.

This occurred in 1983, 1993, 2000 and 2015.

Rule 26(b)(1) now defines the proper scope of

discovery more carefully than ever:

“Unless otherwise limited by court order,

the scope of discovery is as follows: Parties

may obtain discovery regarding any

nonprivileged matter that is relevant to any

party’s claim or defense and proportional to

the needs of the case. . . .”

As the first clause makes clear, Courts have the

express power to “limit” discovery further, but the

Rule does not grant district courts the power to expand

the scope of what must be produced to a requesting

party. In fact, the rule is explicit that “the court must

limit . . . discovery [that] is outside the scope permitted

by Rule 26(b)(1).” FED. R. CIV. P. 26(b)(2)(C)(emphasis

supplied).

Not only are parties not entitled to irrelevant

documents, they are not even entitled to all relevant

documents as discovery is further limited by privilege

and proportionality. The 2015 amendments expressly

inserted the proportionality factor (“and proportional

to the needs of the case”) into the rule’s primary

definition of the scope of relevance, emphasizing anew

the need to restrain disproportionate discovery. In

requiring the production of innumerable irrelevant

documents, which inevitably increases the costs of

managing the larger resulting data set, the Order also

mandates the production of disproportionate volumes

of data, and thereby imposes significant unnecessary

costs on the petitioners. “Rule 26(b)(1) crystalizes the

19

concept of reasonable limits on discovery through

increased reliance on the common-sense concept of

proportionality.” Roberts, C.J., “2015 Year-End Report

on the Federal Judiciary,” at 6.

Discovery is not an end in itself, but merely a

means of obtaining facts so that all parties have

reasonably sufficient information to advocate their

case and resolve disputes. Discovery is costly, not just

in money, time and resources, but in its invasion of the

privacy of parties and third-parties. We accept these

costs and invasions because they help uncover truth

and resolve disputes on their merits, but these costs

inherently rein in the scope of discovery. It may seem

that the most obvious example of these limits is

proportionality, where the court weighs the expected

value of the discovery against its expected costs, but

the bright-line rule against the compelled production

of irrelevant documents is the most important.

Schlagenhauf v. Holder, 379 U.S. 104 (1964) (“The

Federal Rules of Civil Procedure should be liberally

construed, but they should not be expanded by

disregarding plainly expressed limitations.”) As

irrelevant documents add nothing to resolving

disputes, the costs to privacy, confidentiality and

resources are balanced by nothing, and should never

be required.

The Order is impermissible under these provisions

of Rule 26, as are others like it. Attempts to erode the

bright line rule against compelled production of

irrelevant documents should be decisively rejected.

20

B. FED. R. CIV. P. 34(b)

In two ways, the Order is also inconsistent with

Rule 34, governing the production of documents.

Option To Produce Specific Documents. In 2015,

Rule 34(b)(2)(B) was amended to make express what

was well-established, by adding the following

language: “The responding party may state that it will

produce copies of documents or of electronically stored

information instead of permitting inspection.” This

addition was adopted “to reflect the common practice

of producing copies of documents or electronically

stored information rather than simply permitting

inspection.”19 As the recently revised rule makes clear,

the responding party has complete discretion to select

and produce its responsive documents instead of

allowing inspection of the documents.

“Production” in the context of this rule is where a

responding party takes reasonable steps to identify

responsive, relevant, and non-privileged documents

from its corpus of information and provides copies to

the requesting party. “Inspection,” on the other hand,

is where a responding party proffers an appropriately

encompassing selection of documents where it believes

the responsive, relevant and non-privileged

documents are likely to reside and allows the

requesting party to review them and identify the

documents it believes it is entitled to receive.

Inspections are generally less expensive for the

responding party, because they don’t require as much

19 Committee Notes on Rules – 2015 Amendment.

21

time and expense to arrange, but the trade-off is that

they allow access to irrelevant documents.

Seen in this light, the Order is clearly a compelled

inspection, as the responding party has lost the right

to exclude irrelevant documents, and contravenes the

rule. The mere fact that the inspection is limited by

certain custodians and certain search terms is a fig

leaf. First of all, the search terms are not formulated

to exclude personal and confidential emails. Second,

when utilized, the inspection method never affords

access to a responding party’s entire facility or its

complete set of documents. Some initial selection is

always made, and here that is akin to the use of

keyword search terms. A court cannot deprive a

responding party of the Rule’s alternative right to

produce rather than allow inspection, absent evidence

of serious discovery abuse, which is concededly not

present here.

Obligation on Requesting Party To Make

Particular Requests. One of the very few obligations

on requesting parties is that they must “describe with

reasonable particularity each item or category of items

to be inspected.” FED. R. CIV. P. 34(b)(1)(a). This is an

important requirement; it protects responding parties

from vague and ambiguous requests that either in

their ambiguity sweep too broadly or, in the hands of

less scrupulous requesting parties, support motions to

compel seeking sanctions for failing to produce

documents the responding party did not even

understand were requested. This rule, accordingly,

requires the requesting party to identify particularly

what it needs. Once it does so, the rule requires the

22

responding party to do a reasonable search to find and

produce those documents.

As clearly demonstrated by the district court’s

memorandum explaining its Order (Petition for Cert.,

App. C, at 16a, et seq.), the Order contravenes this

requirement by relieving Respondents of the

obligation to make specific requests and instead

compels the production of irrelevant documents. The

apparent motivation behind allowing this compelled

inspection was the court’s fear that the defendants

would not understand the relevance or responsiveness

of certain of their own documents. Respondents

Opposition to Stay, at 5-7. However, it is incumbent

on the requesting party to put “a ‘reasonable person of

ordinary intelligence’ on notice of which specific

documents or information would be responsive to the

request.” Regan-Touhy v. Walgreen Co., 526 F.3d 641,

649–50 (10th Cir. 2008) (quoting Wright & Miller, 8A

FEDERAL PRACTICE AND PROCEDURE § 2211, at 415)

(“Though what qualifies as ‘reasonabl[y] particular’

surely depends at least in part on the circumstances of

each case, a discovery request should be sufficiently

definite and limited in scope that it can be said ‘to

apprise a person of ordinary intelligence what documents are required and [to enable] the court . . . to

ascertain whether the requested documents have been

produced.’”).

Thus, if the Petitioners cannot understand what is

relevant and responsive, then it is incumbent on

Respondents to identify with more particularity what

they are seeking. The Order improperly transfers the

burden from Respondents to the Petitioners by forcing

23

them to produce irrelevant documents instead of the

Respondents better identifying the information they

need.

III. CERTIORARI IS URGENTLY NECESSARY.

Respondents attempt to portray this matter as a

“mere discovery dispute” but the Order’s extraordinary implications transcend the issues in the

instant matter and go to the heart of our system for

resolving civil disputes.

Respondents argue that this is a fact-bound

dispute unworthy of review by this Court, and in any

event the Order is well within the discretion of the

district court. But the Order transgresses what had

been a bright line: absent discovery misconduct, which

is not shown here, Rule 26(b)(1) does not authorize

orders to produce material outside that rule’s scope of

discovery.

Appeals to the district courts’ discretion to enter

such orders must remain unavailing. Just as a district

court has no discretion to compel discovery on claims

that fail as a matter of law, it also lacks discretion to

order the production of irrelevant information.

CONCLUSION

For the foregoing reasons, this Court should grant

Petitioners’ petition for certiorari seeking review of

the decision of the Court of Appeals for the Third

Circuit.

24

Respectfully submitted,

Robert D. Owen

COUNSEL OF RECORD

EVERSHEDS SUTHERLAND (US) LLP

900 North Michigan Avenue, Suite 1000

Chicago, Illinois 60611-6521

(312) 724-9006

RobertOwen@eversheds-sutherland.com

Stacey M. Mohr

Michelle McIntyre

EVERSHEDS SUTHERLAND (US) LLP

999 Peachtree Street, NE, Suite 2300

Atlanta, Georgia 30309-3996

(404) 853-8000

StaceyMohr@eversheds-sutherland.com

MichelleMcIntyre@eversheds-sutherland.com

March 16, 2020