PUBLISHED

                           UNITED STATES COURT OF APPEALS
                               FOR THE FOURTH CIRCUIT


                                       No. 18-1762


In re: MURPHY-BROWN, LLC,

               Petitioner.

------------------------------

AMERICAN FARM BUREAU FEDERATION; NORTH CAROLINA FARM
BUREAU     FEDERATION;    NATIONAL    ASSOCIATION    OF
MANUFACTURERS; NORTH CAROLINA CHAMBER LEGAL INSTITUTE;
NORTH CAROLINA PORK COUNCIL; NATIONAL PORK PRODUCERS
COUNCIL; REPORTERS COMMITTEE FOR FREEDOM OF THE PRESS;
AMERICAN SOCIETY OF NEWS EDITORS; THE ASSOCIATED PRESS;
ASSOCIATION OF ALTERNATIVE NEWSMEDIA; THE MCCLATCHY
COMPANY; THE MEDIA LAW RESOURCE CENTER, INC.; NATIONAL
PRESS PHOTOGRAPHERS ASSOCIATION; THE NEWS MEDIA
ALLIANCE; NORTH CAROLINA PRESS ASSOCIATION; SOCIETY OF
PROFESSIONAL JOURNALISTS; TULLY CENTER FOR FREE SPEECH;
WRAL-TV,

               Amici Supporting Petitioner.



On Petition for Writ of Mandamus from the United States District Court for the Eastern
District of North Carolina, at Wilmington. W. Earl Britt, Senior District Judge. (7:14-
cv-00180-BR; 7:14-cv-00182-BR; 7:14-cv-00183-BR; 7:14-cv-00185-BR; 7:14-cv-
00237-BR)


Argued: September 25, 2018                                  Decided: October 29, 2018


Before WILKINSON and AGEE, Circuit Judges, and James P. JONES, United States
District Judge for the Western District of Virginia, sitting by designation.
Petition granted by published opinion. Judge Wilkinson wrote the opinion, in which
Judge Agee and Judge Jones joined.


ARGUED: Stuart Alan Raphael, HUNTON ANDREWS KURTH LLP, Washington,
D.C., for Petitioner. John S. Hughes, WALLACE & GRAHAM, PA, Salisbury, North
Carolina, for Respondents. ON BRIEF: Robert M. Tata, Washington, D.C., Kevin S.
Elliker, HUNTON ANDREWS KURTH LLP, Richmond, Virginia, for Petitioner. Ellen
Steen, AMERICAN FARM BUREAN FEDERATION, Washington, D.C.; H. Julian
Philpott, Jr., Phillip Jacob Parker, Jr., NORTH CAROLINA FARM BUREAU
FEDERATION, Raleigh, North Carolina; Timothy S. Bishop, Michael B. Kimberly,
MAYER BROWN LLP, Washington, D.C.; Eugene Volokh, UCLA SCHOOL OF LAW,
Los Angeles, California, for Amici American Farm Bureau Federation and North
Carolina Farm Bureau Federation.          Peter C. Tolsdorf, Leland P. Frost,
MANUFACTURERS’ CENTER FOR LEGAL ACTION, Washington, D.C.; Catherine
E. Stetson, Sean Marotta, HOGAN LOVELLS US LLP, Washington, D.C., for Amici
National Association of Manufacturers and North Carolina Chamber Legal Institute. C.
Amanda Martin, STEVENS MARTIN VAUGHN & TADYCH PLLC, Raleigh, North
Carolina; Matthew Nis Leerberg, SMITH MOORE LEATHERWOOD, LLP, Raleigh,
North Carolina, for Amici North Carolina Pork Council and National Pork Producers
Council.     Michael C. Formica, NATIONAL PORK PRODUCERS COUNCIL,
Washington, D.C., for Amicus National Pork Producers Council. Bruce D. Brown, Katie
Townsend, Caitlin Vogus, Sarah Matthews, THE REPORTERS COMMITTEE FOR
FREEDOM OF THE PRESS, Washington, D.C., for Amici The Reporters Committee for
Freedom of the Press and 11 Other Media Organizations. Kevin M. Goldberg,
FLETCHER, HEALD & HILDRETH, PLC, Arlington, Virginia, for Amici American
Society of News Editors and Association of Alternative Newsmedia. Karen Kaiser,
General Counsel, THE ASSOCIATED PRESS, New York, New York for Amicus The
Associated Press. Juan Cornejo, THE MCCLATCHY COMPANY, Sacramento,
California, for Amicus The McClatchy Company. George Freeman, MEDIA LAW
RESOURCE CENTER, New York, New York, for Amicus The Media Law Resource
Center, Inc. Mickey H. Osterreicher, Buffalo, New York, for Amicus National Press
Photographers Association.     Kurt Wimmer, COVINGTON & BURLING LLP,
Washington, D.C., for Amicus The News Media Alliance. Bruce W. Sanford, Mark I.
Bailen, BAKER & HOSTETLER LLP, Washington, D.C., for Amicus Society of
Professional Journalists.




                                         2
WILKINSON, Circuit Judge:

       Petitioner challenges the United States District Court for the Eastern District of

North Carolina’s gag order imposing stringent restrictions on participants and potential

participants in a series of nuisance suits brought against the hog industry in North

Carolina. For the reasons that follow, we find the district court’s order defective in

multiple respects. We direct the district court to vacate the gag order and allow the parties

to begin anew under the guidelines set forth below, and even then, only if exceptional

circumstances warrant.

                                              I.

       Petitioner (and defendant below) seeks relief from a gag order controlling

interrelated civil nuisance suits. The plaintiffs in those suits allege that nearby hog farms

associated with Murphy-Brown, LLC are a private nuisance. Hog farming is a predictably

messy business, but one with central economic importance to the state of North Carolina.

Plaintiffs claim that the hog farms give rise to undue amounts of odors, insects, and pests,

not to mention the noise and debris generated by raising hogs and shipping them for sale.

In total, the Master Case Docket includes more than 20 lawsuits and more than 500

plaintiffs. Plaintiffs from five cases in the initial discovery pool have been grouped for

eleven seriatim trials—three have already delivered large verdicts in plaintiffs’ favor. The

cases likely will take years to resolve.

       Respondents (and plaintiffs below) support, in part, the gag order in this case. That

order was based on a “significant increase in trial publicity” after the first two trials. Gag

Order at 2 (J.A. 616). The hog nuisance suits touch on important matters of local and

                                              3
regional concern. Sizable damages awards in the first two cases have generated

legislative responses. It is no wonder that the press and the public have taken an interest,

and that a lively, indeed passionate, debate has ensued.

       The jury pools were, naturally, exposed to the publicity. Three juries have

considered the hog nuisance suits. Of the jury pool for the first trial, two of the fifty

potential jurors had been exposed to the issues of the case in some way. Of the second

jury pool, eleven of fifty. Of the third, twenty-three of fifty. The majority of potential

jurors with prior exposure to the case testified that they could impartially serve on one of

the juries—two did in fact serve on the second one. Moreover, during the second trial,

one juror conducted outside research and shared the results with other jurors, but the trial

continued when jurors stated that they could remain impartial.

       The present case addresses a gag order that the district court issued sua sponte

while the second jury deliberated on June 27, 2018. The court found that “the volume and

scope of prejudicial publicity observed” led to a “substantial risk of additional publicity

tainting or biasing future jury pools.” Gag Order at 3 (J.A. 617). The gag order was a

sweeping one. It prohibited all parties and their lawyers, representatives, and agents, as

well as “all potential witnesses,” from:

       giv[ing] or authoriz[ing] any extrajudicial statement or interview to any
       person or persons associated with any public communications media or that
       a reasonable person would expect to be communicated to a public
       communications media relating to the trial, the parties or issues in this case
       which could interfere with a fair trial or prejudice any plaintiff, the
       defendant, or the administration of justice and which is not a matter of
       public record. Statements of information intended to influence public
       opinion regarding the merits of this case are specifically designated as
       information which could prejudice a party.

                                             4


order provides for limited exceptions. “[W]ithout elaboration or any kind

of characterization,” covered individuals may discuss “the general nature of an allegation

or defense”; “information contained in the public record of this case”; “scheduling

information”; “any decision made or order issued by the court which is a matter of public

record”; and “the contents or substance of any motion or step in the proceedings, to the

extent such motion or step is a matter of public record.” 

(J.A. 617-18).

       On July 6, 2018, Murphy-Brown petitioned this court for mandamus relief from

the gag order. We promptly ordered expedited review. On August 9—after partial

appellate briefing and forty-three days after the gag order—respondents filed a motion in

the district court asking it to clarify, modify, or vacate the gag order. The district court

required parties to submit briefing on that motion two days after the expedited appellate

briefing concluded. On August 31, the district court vacated its gag order. It further

instructed parties to seek a new gag order for the third trial, if desired, by September 18,

one week before we heard oral argument. Respondents filed a contested motion to

dismiss the mandamus petition on mootness grounds on September 10. We held that

motion in abeyance pending oral argument.

                                             II.

       We first turn to the motion to dismiss the mandamus petition. Respondents

contend that the district court’s August 31 order rescinding the gag order somehow moots

the mandamus petition. We disagree. Not only was the August 31 order invalid, but the


                                             5
ordinary application of mootness doctrine comfortably allows this court to reach the

merits of the gag order.

                                              A.

       The mandamus petition is not moot because the August 31 order exceeded the trial

court’s authority vis-à-vis the court of appeals. Petitioner urges us to say that the district

court lacked jurisdiction to declare its earlier order ineffective. But we need not speak

broadly or rigidly in jurisdictional terms. It is enough to observe that in the circumstances

here the district court’s decision to revisit its June 27 gag order was ill-advised.

       In analogous circumstances on direct appeal, district courts formally lose

jurisdiction over the questions being appealed. For example, we have held “that a district

court loses jurisdiction to amend or vacate its order after the notice of appeal has been

filed . . . .” Lewis v. Tobacco Workers’ Int’l Union, 

, 1139 (4th Cir. 1978).

Or more generally, a district court loses jurisdiction when the court of appeals assumes

jurisdiction. But writs of mandamus function differently from ordinary appeals, and we

have no need at this time to extend Lewis’s holding categorically to all issues challenged

in mandamus petitions.

       Indeed, there are good reasons not to do so. For example, many mandamus

petitions seek to require district courts to rule on matters which, in the petitioner’s view,

have been subject to an unwarranted district court delay. See, e.g., In re Richardson, 734

Fed. App’x 901 (4th Cir. 2018) (per curiam); In re Gallardo, 709 Fed. App’x 179 (4th

Cir. 2018) (per curiam). A mandamus petition may prompt the district court to turn its

attention to the case. Those situations are quite different from the one before us and

                                               6
indicate why, in light of such disparate scenarios, a broad jurisdictional rule addressing

district court actions during the pendency of a mandamus petition would be

inappropriate.

       None of this, however, forecloses our review of the district court’s actions and its

reassertion of jurisdiction in the circumstances of this case. The analogy to Lewis

illustrates just how infirm was the district court’s decision to rescind its initial gag order

in favor of an option to request a new or modified gag order for the third nuisance trial. In

Lewis, the court of appeals had merely received notices of appeal when the district court

. This case involved circumstances well beyond that. The

court of appeals had already scheduled the case for expedited hearing, the parties had

already written and filed appellate briefs, and the court was poised to hear the case within

weeks. The district court should have allowed our responsibilities to run their due course

rather than revisiting its order right before plenary appellate review. For the district court

to take up an issue squarely before the court of appeals on the very eve of oral argument

risks a duplication of resources through simultaneous consideration of the very same

order by two courts, even as it invites confusing or conflicting dispositions of the issue.

       The mischief of the trial court’s action should be apparent. Allowing the August

31 rescission of the June 27 gag order to stand would invite district courts to track cases

on the appellate court’s docket, and when a reversal seemed possible or imminent, to pull

the rug out from under the appellate court and the parties. This sets up an endless game of

cat and mouse. One need only contemplate the inadvisability of a court of appeals



                                              7
revisiting its opinions sua sponte on the eve of a Supreme Court argument to understand

the inappropriateness of the course of action below.

       The All Writs Act empowers this court to “issue all writs necessary or appropriate

in aid of [its] respective jurisdiction[] and agreeable to the usages and principles of law.”

28 U.S.C. § 1651(a) (2012). Mandamus, while appropriate only in rare circumstances,

plainly extends to invalidating district court orders that threaten to undermine the very

system of review that the All Writs Act protects. For those reasons, the district court’s

order of August 31 was in error. Since the August 31 order so plainly undercut the court

of appeals in the orderly exercise of its own jurisdiction, it cannot be allowed to moot the

petition.

                                             B.

       Even if the district court had properly rescinded the gag order, this case still would

not be moot. While several aspects of mootness doctrine might be relevant here,

including voluntary cessation of erroneous conduct, * we need hold only that this case is

capable of repetition yet evading review. “A dispute qualifies for that exception only if

(1) the challenged action is in its duration too short to be fully litigated prior to its

cessation or expiration, and (2) there is a reasonable expectation that the same

complaining party will be subjected to the same action again.” United States v. Sanchez-


*
  We note that the Supreme Court recently applied the voluntary cessation doctrine when
the Southern District of California altered its challenged pretrial restraint policy before
the Supreme Court reviewed the case. United States v. Sanchez-Gomez, 

,
1537 n.1 (2018).


                                             8
Gomez, 

, 1540 (2018) (internal quotation marks omitted). This court and

the Supreme Court have both applied this exception to vindicate First Amendment rights.

See, e.g., Neb. Press Ass’n v. Stuart, 

, 546-47 (1976); Baltimore Sun Co. v.

Goetz, 

, 63 (4th Cir. 1989). We follow those examples here.

       First, it is self-evident that the challenged order was undone too quickly for

appellate review. Indeed, this court granted expedited review to address the serious First

Amendment concerns raised by petitioner in a timely manner. Yet the district court

attempted to rescind its order shortly before oral argument and only two months after

issuing the order. See Baltimore 

(less than “eight months” after

mandamus filing was too short to be fully litigated). This sort of withdrawal risks a series

of orders too short to be reviewed but too damaging to be ignored.

       Second, petitioner must demonstrate a reasonable expectation that the district

court will subject it to another gag order. In Nebraska Press, the Supreme Court held that

a First Amendment violation was capable of repetition because the Nebraska Supreme

Court could subsequently reverse the underlying conviction and order a new 

. For petitioner here, future trials are more than a contingency as in Nebraska

Press. They are a near certainty. Many hundreds of plaintiffs await their day in court, and

petitioner will likely be defending against these nuisance suits for years to come. It is

easy to see this scenario repeating itself. Plaintiffs may seek another gag order before a

future trial (as the district court invited them to do prior to a future trial). Or the district

court may issue another gag order sua sponte, only to vacate such an order on the eve of



                                               9
appellate argument. This is the sort of case for which “capable of repetition yet evading

review” is made.

       The gag order has already inflicted serious harm on parties, advocates, and

potential witnesses alike. It has muted political engagement on a contested issue of great

public and private consequence. It has hamstrung the exercise of First Amendment rights.

Even in short doses, these harms are hostile to the First Amendment. Indeed, as petitioner

notes, the threat of a gag order will hang over these trials as a sort of “Sword of

Damocles,” dangling by the thread of the court’s displeasure. Opposition to Respondents’

Motion to Dismiss Petition as Moot at 10. Its chilling effect is evident and unacceptable;

the mandamus petition is not moot.

       Respondents’ motion to dismiss the petition is accordingly denied.

                                            III.

       We now turn to the merits. Petitioner challenges the gag order in the form of a writ

of mandamus. We reiterate that “mandamus is a ‘drastic’ remedy that must be reserved

for ‘extraordinary situations.’” Cumberland Cty. Hosp. Sys., Inc. v. Burwell, 

,

52 (4th Cir. 2016) (quoting Kerr v. U.S. Dist. Court, 

, 402 (1976)). Courts

provide mandamus relief only when (1) petitioner “ha[s] no other adequate means to

attain the relief [it] desires”; (2) petitioner has shown a “clear and indisputable” right to

the requested relief; and (3) the court deems the writ “appropriate under the

circumstances.” Cheney v. U.S. Dist. Court, 

, 380-81 (2004) (internal

quotation marks omitted).



                                             10
       Respondents contend that petitioner should have sought reconsideration in the

district court before petitioning for mandamus relief in this court. We disagree. A motion

for reconsideration would not have been an “adequate” means of attaining relief from the

gag order. The trial court had already considered the legal issues surrounding the gag

order. Moreover, petitioner’s “loss of First Amendment freedoms . . . unquestionably

constitutes irreparable injury.” Elrod v. Burns, 

, 373 (1976) (plurality

opinion); see Johnson v. Bergland, 

, 995 (4th Cir. 1978) (“Violations of

[F]irst [A]mendment rights constitute per se irreparable injury.”). Parties need not endure

repeated and irreparable abridgments of their First Amendment rights simply to afford the

district court a second chance. And the public need not suffer the prolonged deprivation

of robust discussion on an issue that affected so many in the community so tangibly.

       Indeed, this court has recognized that mandamus plays an important and necessary

role in protecting First Amendment freedoms. For example, we have long held “that

mandamus is the preferred method of review for orders restricting press activity . . . .” In

re Wash. Post Co., 

, 388 (4th Cir. 1986); see also In re State-Record Co.,

, 126-27 (4th Cir. 1990) (per curiam) (expressing preference for mandamus).

Even when such orders are challenged via both direct appeal and mandamus, we have

elected to provide mandamus relief. See In re Charlotte Observer, 

, 852 (4th

Cir. 1989). And we have even treated an “appeal as a petition for mandamus” when

reviewing a sealed access order challenged on First Amendment grounds. United States v.

Appelbaum, 

, 289 (4th Cir. 2013). This court has reviewed gag orders in

criminal cases via a mandamus petition. In re Russell, 

, 1008 (4th Cir.

                                            11
1984); In re Wall St. Journal, 601 F. App’x 215, 218 (4th Cir. 2015) (per curiam)

(reviewing sealing and gag order). While gag orders in civil cases are quite rare (there

being no Sixth Amendment right at issue), the First Amendment interests in an open and

public civil justice system are no less salient, and we see no reason to depart from

longstanding mandamus procedure when reviewing the civil gag order in this case.

                                            IV.

       Given that a mandamus petition is the appropriate mechanism for challenging the

gag order, 

, we next ask whether petitioner has shown a “clear

and indisputable” right to the requested relief. 

As we explain below, petitioner

has met its burden because the gag order contravened the First Amendment in basic

respects.

       Under the challenged gag order, the parties, their lawyers, and all potential

witnesses were not allowed to “give or authorize any extrajudicial statement . . . relating

to the trial, the parties or issues in this case” if such statement could reasonably reach

“public communications media” and could “interfere with a fair trial or prejudice any

plaintiff, the defendant, or the administration of justice.” Gag Order at 3 (J.A. 617). The

gag order designated as prejudicial any statement “intended to influence public opinion

regarding the merits of this case.” 

could, however, discuss some

basic information “without elaboration or any kind of characterization.” 

First Amendment claims, gag orders warrant a most rigorous form of

review because they rest at the intersection of two disfavored forms of expressive

limitations: prior restraints and content-based restrictions. Like all “court orders that

                                            12
actually forbid speech activities,” Alexander v. United States, 

, 550 (1993),

gag orders are prior restraints. Prior restraints bear “a heavy presumption against [their]

constitutional validity.” Bantam Books, Inc. v. Sullivan, 

, 70 (1963). Prior

restraints upend core First Amendment principles because “a free society prefers to

punish the few who abuse rights of speech after they break the law [rather] than to

throttle them and all others beforehand.” Se. Promotions, Ltd. v. Conrad, 

,

559 (1975).

       Similarly, gag orders are presumptively unconstitutional because they are content

based. Nat’l Inst. of Family and Life Advocates v. Becerra, 

, 2371 (2018)

(presumption against content-based restraints). Content-based restrictions target

“particular speech because of the topic discussed or the idea or message expressed.” Reed

v. Town of Gilbert, 

, 2227 (2015). Gag orders inherently target speech

relating to pending litigation, a topic right at the core of public and community life. But

the “government has no power to restrict expression because of its message, its ideas, its

subject matter, or its content.” Ashcroft v. ACLU, 

, 573 (2002) (internal

quotation marks omitted).

       In light of these twin presumptions, gag orders must survive strict scrutiny. 

(strict scrutiny for content-based restrictions). We review orders

limiting First Amendment rights de novo. Charlotte 

. And we

require trial courts to support them with reasoning “specific enough to enable the

reviewing court to determine” whether the order survives rigorous scrutiny. Wash. Post

; see State-Record 

(“[W]e have required

                                            13
specific reasons and findings on the record, because without such findings, de novo

review is difficult.” (internal citation omitted)). Applying searching review, we hold that

petitioner has demonstrated a “clear and indisputable” right to relief from the gag order

because it breached basic First Amendment principles in both meaningful and material

ways. 

.

                                             A.

       Strict scrutiny first requires that a gag order serve a “compelling” public interest.

See 

. “Our system of justice properly requires that civil litigants

be assured the right to a fair trial.” Hirschkop v. Snead, 

, 373 (4th Cir. 1979)

(en banc) (per curiam). Ensuring fair trial rights is a compelling interest, however, only

when there is a “reasonable likelihood” that a party would be denied a fair trial without

the order under challenge. 

.

       While respondents here attempt to analogize the gag order’s restrictions to the

rules of professional ethics (and the arguably less demanding standards applied to them),

see Gentile v. State Bar of Nevada, 

(1991), this particular order extended

far beyond the legal profession. The Supreme Court’s decision in Nebraska Press thus

guides our understanding of when a party would be denied a fair trial in the absence of a

gag order under review. Specifically, a gag order may issue only if there is a likelihood

that “publicity, unchecked, would so distort the views of potential jurors that [enough]

could not be found who would, under proper instructions, fulfill their sworn duty to

render a just verdict exclusively on the evidence presented in open court.” Neb. Press

Ass’n v. Stuart, 

, 569 (1976).

                                             14
       The district court—without adequate factual findings or development of a

record—found that a “significant increase in trial publicity” supported the gag order. Gag

Order at 2 (J.A. 616). But the fact of publicity is hardly dispositive. Publicity often

accompanies trials, including trials in which the public has a keen and understandable

interest. The judicial process does not run and hide at those moments when public

appraisal of its workings is most intense. An impartial jury, moreover, need not be wholly

unaware of information—including potentially prejudicial information—outside the

record. “Prominence does not necessarily produce prejudice, and juror impartiality . . .

does not require ignorance.” Skilling v. United States, 

, 381 (2010).

       Jurors are not that fragile. Under admonishment from the court, the jury must

confine its consideration to what it saw and heard in the courtroom. Moreover, courts

often tell jurors to disregard statements at trial that should not have been made or

evidence that should not have been admitted. Courts also routinely instruct jurors to

consider potentially prejudicial evidence, when admitted, only for a non-prejudicial

purpose. To take but one example, Federal Rule of Evidence 404(b) permits evidence of a

“crime, wrong, or other act” to show, inter alia, motive, opportunity, or intent. Trial

courts instruct jurors not to consider such evidence as establishing a person’s bad

character. See, e.g., United States v. Faulls, 

, 508 (4th Cir. 2016) (affirming

admission of evidence under Federal Rule of Evidence 404(b) as proof of motive after

trial judge gave an instruction “reminding the jury that it should not consider the evidence

to prove [defendant’s] character or his propensity to commit the charged offenses”).



                                            15
       The question, therefore, is neither whether a case has garnered public attention nor

whether public discussion of it risks revealing potentially prejudicial information.

Guidance is the critical concept. The question is whether the judge finds it likely that he

or she will be unable to guide a jury to an impartial verdict. If judges can guide the jury to

an impartial verdict, then no gag order may issue.

       The record here demonstrates that the trial court assembled impartial juries each

time it needed to do so. Plenty of potential jurors remained unfamiliar with the case, and

most of those with prior exposure remained able to serve impartially. Two even did so.

There was no showing here that the judge could not guide a jury to an impartial verdict.

       Further and crucially, a gag order must actually “operate to prevent the threatened

danger.” Neb. 

. Sometimes the inevitability of publicity

surrounding civil proceedings will render a gag order entirely superfluous. On other

occasions parties and counsel may even mitigate any prejudice by disseminating accurate

and timely information to the public. By prohibiting only certain kinds of information

from selected sources, a gag order can actually warp and distort discussion, thereby

enhancing prejudice rather than mitigating it.

       It is anything but clear that this gag order would accomplish its intended purpose

of reducing risks to fair trial rights. See also infra Part V (discussing a variety of the gag

order’s unintended and costly effects). Amici identified a torrent of potentially prejudicial

news coverage released after the gag order had taken effect. Brief Amici Curiae of the

American Farm Bureau Federation and the North Carolina Farm Bureau Federation at 15-

18, app. Recent news coverage included florid characterizations of farmers and hog

                                             16
farms, discussions of excluded testimony from an odor expert, and attacks on the

attorneys at trial. 

news articles). It is simply not clear why this

gag order would not work selective and uneven impositions. Those could prove even less

desirable than a fuller and more balanced airing of the viewpoints surrounding the hotly

disputed question of the benefits and problems that hog farming has brought to eastern

North Carolina.

       For these reasons, the gag order did not further a compelling interest based on the

record before us.

                                             B.

       Even when a gag order furthers a compelling interest, it must be the “least

restrictive means” of furthering that interest, 

. The law

empowers trial courts to ensure fair jury trials using a number of tools short of gag

orders. These tools include enlarged jury pools, voir dire, changes to a trial’s location or

schedule, cautionary jury instructions, and, in more unusual circumstances, sequestration.

The gag order failed to explain if and why these alternatives had proven ineffective. See

Neb. 

(holding that “the record [was] lacking in evidence to

support” a finding that alternative “measures might not be adequate”).

       In particular, our system places abundant trust in voir dire, which is a process

“within the province of the trial judge.” Ristaino v. Ross, 

, 595 (1976)

(internal quotation marks omitted). Trial judges possess great discretion over the “depth

or breadth of voir dire,” 

, and they routinely use that discretion to

identify jurors unable to render a fair verdict. Voir dire is effective because of the general

                                             17
desire of jurors to express themselves honestly, the desire of lawyers to winnow out

partial jurors through questions (where permitted) and strikes, and the desire of trial

courts in assessing credibility and bias to ensure a fair trial. For these reasons, voir dire

“can serve in almost all cases as a reliable protection against juror bias however

induced.” Charlotte 

. That maxim holds true here. Voir dire has

already thrice proven effective in assembling juries in the McKiver, McGowan, and Artis

trials. The gag order should not have come before less restrictive alternatives including

voir dire were somehow found or proven to be ineffective.

       Strict scrutiny also demands that First Amendment restraints be “narrowly

tailored” to serve their intended purpose. 

(2015). Each

restriction must be no “greater than necessary.” In re Morrissey, 

, 140 (4th

Cir. 1999). The overbreadth doctrine permits litigants to challenge First Amendment

restrictions even so far as they also impinge others’ First Amendment rights, see

Broadrick v. Oklahoma, 

, 611-13 (1973), and we thus consider whether this

gag order was narrowly tailored as it applies to all covered parties.

       The gag order was not narrowly tailored. It included no findings specific to the

various individuals it restricted. It treated lawyers no differently from parties, who in turn

were treated the same as potential witnesses. It assumed all covered individuals were

identically situated vis-à-vis pending and future litigation. Moreover, the gag order

applied blanket restrictions to more than twenty cases that will be tried over a period of

years. Participants in any given trial appear to remain restricted until the resolution of the

final case.

                                             18
       What is more, the unclear yet extraordinary reach of the phrase “potential witness”

could include almost anyone who is knowledgeable or conversant about the issues in

these cases. This breadth alone cannot help but impair legitimate news gathering

activities that in and of themselves underlie the proper functioning of the First

Amendment. See Branzburg v. Hayes, 

, 681 (noting that news gathering

qualifies for First Amendment protections because “without some protection for seeking

out the news, freedom of the press could be eviscerated”). The gag order’s blanket

restrictions thus exceed what would be “essential to the preservation of a fair trial.”

.

       Gag orders should be a last resort, not a first impulse. The skimpy findings

supporting the gag order here frustrate the sort of fulsome, substantive appellate review

that shields First Amendment freedoms from unjustifiable restraints. The lack of narrow

tailoring and the availability of less-restrictive alternatives are among the further defects

that accompanied the imposition of the challenged order in this case.

                                             C.

       We turn finally to petitioner’s vagueness challenge. In Hirschkop this court held,

in the company of many others, that vague restrictions on speech “offend” the First

. For that reason, restraints on speech must include “some

sensible basis for distinguishing what may come in from what must stay out.” Minn.

Voters Alliance v. Mansky, 

, 1888 (2018). Speakers deserve to know. This

core requirement of clarity avoids twin problems. For one, “[t]he interpretive process

itself would create an inevitable, pervasive, and serious risk of chilling protected speech

                                             19
pending the drawing of fine distinctions that, in the end, would themselves be

questionable.” Citizens United v. FEC, 

, 327 (2010). Vague restraints also

pose the risk of discriminatory or arbitrary enforcement. See 

.

      This gag order was unconstitutionally vague because it forced individuals to

“guess at its contours,” 

To take but one example, it permitted statements

describing “the general nature of an allegation or defense,” but only “without elaboration

or any kind of characterization whatsoever.” Gag Order at 3 (J.A. 617). The Supreme

Court has expressly found nearly identical language permitting discussion of the “general

nature” of a defense “without elaboration” to be vague. 

.

This gag order’s uses of “general” and “elaboration” were impermissibly vague, too.

Since “‘general’ and ‘elaboration’ are both classic terms of degree,” a person subject to

the gag order has “no principle for determining when his remarks pass from the safe

harbor of the general to the forbidden sea of the elaborated.” 

-

49. Moreover, the phrase “public communications media,” Gag Order at 3 (J.A. 617),

contains no hint of how it would apply to social media. When and how are social media

posts “public communications media?” Does it turn on whether a post is public or

private, on the account’s number of followers, or on whether a reporter is among those

followers? The gag order simply does not provide an answer.

      The gag order also only vaguely specified who it covered. Individuals and

organizations are led to guess whether they are “potential witnesses,” a term the gag

order does not even define. To illustrate, when a lawyer for the North Carolina Pork

Council inquired whether her non-party client had to comply with the gag order, the

                                            20
district court called the order “self-[explanatory]” and said that “[y]ou’ll just have to try

to figure it out or file a motion or something. I can’t expand on what it covers.” J.A. 620-

21. To the contrary, the pool of potential witnesses was unimaginably large and anything

but self-explanatory. If read to its full extent, the pool of possible witnesses included

nearly everyone related to the controversy over pork farming in eastern North Carolina. A

gag order must make clear to whom it applies; this gag order failed to do so.

                                             V.

       For all the reasons highlighted above, the gag order fails First Amendment

scrutiny to such a degree that petitioner possesses a “clear and indisputable” right to relief

from the district court’s decree. 

. Like all First Amendment

infringements, the gag order’s harms are “per se irreparable” injuries. 

. Mandamus relief is “appropriate under the circumstances,” 

, because orders such as this one carry severe consequences.

       The gag order harmed petitioner. Petitioner has identified various attacks on its

business practices and public policy involvement to which it could not respond while the

gag order was in effect. Brief of Petitioner at 11-13. It also could not respond to a flood

of news coverage following the second jury verdict. 

order assertedly forced

petitioner into an untenable posture of silence, imposing real consequences above and

beyond the particulars of the legal proceedings. See Brief Amici Curiae of the National

Association of Manufacturers and North Carolina Chamber Legal Institute in Support of

Petitioner at 17 (asserting that “62% of Americans take a business’s ‘no comment’ about

pending litigation to mean that the business is covering up wrongdoing” (citing Julia

                                             21
Hood, “No Comment” Won’t Cut It, Finds Survey, PR Week (Aug. 5, 2002))). Whether

hyperbolic or not, petitioner’s assertions drew plausibly on the language of the gag order

itself.

          Gag orders like this one may in fact conflict with securities disclosure

requirements. The Securities and Exchange Commission requires corporations to disclose

specific facts in both periodic and situational filings. To take but one example,

corporations must disclose the magnitude of material litigation risks. 

A

company will often disclose risks with the disclaimer that it “believes that this claim

lacks merit and intends to defend itself vigorously against it.” 

(quoting

Christopher M. Holmes et al., When to Set a Reserve: Now, Never, or Somewhere in

Between 12 (2004)). But such a disclosure may run afoul of this gag order’s prohibition

on discussing the case with “any elaboration or any kind of characterization whatsoever.”

Gag Order at 3 (J.A. 617).

          The gag order harms farmers. North Carolina farmers claim they have been unable

to participate in industry-wide conferences discussing the nuisance suits’ potential effects

on their livelihoods. Brief Amici Curiae of the American Farm Bureau Federation and

North Carolina Farm Bureau Federation in Support of Petitioner at 5-8. They assert they

are prohibited from commenting on what they perceive as the almost existential threat

posed by large nuisance suit verdicts to an industry they claim provides some 46,000 jobs

for North Carolina. 

(citing Greg Barnes, Suits Against Smithfield Could

Leave Hog Industry in Doubt, Fayetteville Observer (June 16, 2018)). These harms are

allegedly exacerbated by the farmers’ inability to engage with policymakers who are

                                             22
considering legislative solutions at the federal, state, and local levels of government. See,

e.g., 

The gag order in their view is forcing them into a posture of passivity at a

time when they have every right to be outspoken.

       The gag order also harms nuisance suit plaintiffs. It prevents hundreds of plaintiffs

from discussing fully how the pork industry, for all its benefits, threatens their very

quality of life through alleged odors, noise pollution, waste management practices, and

the accompanying declines in property values. See, e.g., Third Amended McGowan

Complaint at 1-3, 38 (J.A. 237-39, 274). Plaintiffs describe in one complaint how “[l]arge

black flies periodically descend upon [their] properties, ruining and interfering with

family activities, cookouts and other outdoor activities.” 

They also allege the

nearby hog farms sometimes toss hog carcasses in “dead boxes,” which are “dumpsters

full of dead animals left out in the open often in plain view.” 

While legal

filings are, of course, not covered by the gag order, they point to a larger case that

plaintiffs and their supporters, many of whom may be involved in future legal

proceedings, might wish to take to the public.

       All these people care. This case is about their lives and their livelihoods. Whatever

differences the parties and their supporters have, they possess in common a passionate

First Amendment interest in debating their futures. It seems very wrong that a court

would take that from them.

       For all the reasons above enumerated, we direct the district court to vacate its June

27, 2018 gag order and August 31, 2018 order proposing unspecified tweaks and

modifications to it. The parties are to start from square one with respect to this issue

                                             23
under the guidance set forth in this opinion, and then only if exceptional circumstances

warrant.

      The petition is hereby granted.




                                          24