Christiana Tah v. Global Witness Publishing, Inc.

United States Court of Appeals

FOR THE DISTRICT OF COLUMBIA CIRCUIT

Argued September 14, 2020 Decided March 19, 2021

No. 19-7132

CHRISTIANA TAH AND RANDOLPH MCCLAIN,

APPELLANTS

v.

GLOBAL WITNESS PUBLISHING, INC. AND GLOBAL WITNESS,

APPELLEES

Consolidated with 19-7133

Appeals from the United States District Court

for the District of Columbia

(No. 1:18-cv-02109)

Rodney A. Smolla argued the cause for appellants/cross-

appellees. With him on the briefs was Arthur V. Medel.

Chad R. Bowman argued the cause for appellees/cross-

appellants. With him on the briefs were David A. Schulz, Mara

J. Gassmann, and Maxwell S. Mishkin.

Gregory M. Lipper was on the brief for amici curiae Non-

Governmental Organizations in support of appellees/cross-

appellants.

2

Bruce D. Brown and Katie Townsend were on the brief for

amici curiae Reporters Committee for Freedom of the Press

and 26 Media Organizations in support of appellees/cross-

appellants.

Before: SRINIVASAN, Chief Judge, TATEL, Circuit Judge,

and SILBERMAN, Senior Circuit Judge.

Opinion for the Court filed by Circuit Judge TATEL.

Opinion dissenting in part filed by Senior Circuit Judge

SILBERMAN.

TATEL, Circuit Judge: In this defamation action, two

former Liberian officials allege that Global Witness, an

international human rights organization, published a report

falsely implying that they had accepted bribes in connection

with the sale of an oil license for an offshore plot owned by

Liberia. The district court dismissed the complaint for failing

to plausibly allege actual malice. For the reasons set forth in

this opinion, we affirm. The First Amendment provides broad

protections for speech about public figures, and the former

officials have failed to allege that Global Witness exceeded the

bounds of those protections.

I.

Because this appeal comes to us from a dismissal pursuant

to Federal Rule of Civil Procedure 12(b)(6), “[w]e accept facts

alleged in the complaint as true and draw all reasonable

inferences from those facts in the plaintiffs’ favor.” Hancock v.

Urban Outfitters, Inc., 830 F.3d 511, 513–14 (D.C. Cir. 2016).

The dispute in this case traces its roots to an Atlantic

Ocean plot owned by Liberia and thought to have potentially

significant oil reserves. Compl. ¶ 18. The National Oil

3

Company of Liberia (NOCAL), responsible under Liberian law

for awarding oil licenses, first issued a license for the plot,

known as “Block 13,” in 2007 to a company called Broadway

Consolidated PLC (BCP). Id. ¶¶ 19–21. That transaction was

marred by “rumors of corruption,” and when BCP failed to

fulfill its obligations under its production sharing contract,

Liberia began arranging to sell Block 13 to a different oil

company. Id. ¶¶ 21–22.

ExxonMobil, a multinational oil company, was interested

in purchasing Block 13 but wary of buying the license directly

from BCP given the rumors of corruption surrounding the 2007

transaction. Accordingly, Exxon got a third-party, Canadian

Overseas Petroleum Limited, to buy the Block 13 license and

resell it to Exxon. In exchange, Exxon paid $120 million, of

which $50 million went directly to Liberia—the most Liberia

had ever received in a single natural resources deal. Id. ¶ 22.

Unlike in the BCP transaction, Liberia was represented in these

negotiations by the Hydrocarbon Technical Committee (HTC),

a six-member government entity created to “superintend []

negotiations” between oil companies and NOCAL. Id. ¶¶ 23–

24. Plaintiffs Christiana Tah and Randolph McClain, Liberia’s

Minister of Justice and NOCAL’s CEO respectively, were

HTC members during the transaction.

After the deal was consummated, the Liberian President

directed NOCAL’s board to pay bonuses to those responsible

for the new agreement as a “reward for exceptionally well-done

service.” Id. ¶¶ 25–26, 28. But before the board determined the

size of the bonuses, McClain “asked two of the HTC members,

the President’s Legal Advisor, Seward Cooper, and Minister of

Justice, Christiana Tah, if payment of such bonuses would be

legally permissible.” Id. ¶ 26. Cooper and Tah “concluded”

that they were legal for “two . . . independent reasons.” Id. ¶ 28.

First, the pertinent Liberian anti-corruption law had “expired

4

and was no longer legally operative.” Id. And second, even had

the law remained in force, the bonuses would still pass muster

because they “had come at the initiative of [the] President” and

“[n]o prospective recipient of the bonuses claimed to have

demanded any such bonus payments.” Id.

NOCAL’s board then authorized “approximately

$500,000” worth of bonuses. Id. ¶ 29. Each “member[] of the

HTC, including . . . Tah and . . . McClain, received . . .

$35,000,” and each of “five consultants were . . . sent bonuses

of $15,000.” Id. The rest of the funds were split among the

remaining NOCAL employees, including drivers and custodial

workers. Id. The $35,000 payments to Tah and McClain are the

focus of this case.

According to Global Witness’s report, Catch me if you

can, the organization first learned of Exxon’s Block 13 deal

from the Liberian Extractive Industries Transparency Initiative

(LEITI), a semi-autonomous Liberian agency that publishes

information about payments made by energy companies to the

Liberian government. Because of NOCAL’s “tarnished track

record of corrupt deals, Global Witness saw there was a risk of

bribery and began its investigation.” Catch me if you can

(“Report”) at 9; see also Compl. ¶ 42. Global Witness focused

on Block 13 in order to highlight the “critical information”

provided by section 1504 of the Dodd-Frank Act, see 15 U.S.C.

§ 78m(q), which “[l]ike LEITI, . . . requires all oil, gas, and

mining companies to report the payments they make to

governments.” Report at 9.

Catch me if you can addresses Block 13’s background and

the corruption surrounding the BCP deal. For example, it states

BCP was “likely part-owned by [now-former Liberian]

government officials with the power to influence the award of

oil licenses,” and that the award of Block 13 to BCP therefore

5

violated Liberian law. Id. at 12. The report also claims that the

BCP license was approved due to bribery. It then explains how

Exxon structured its transaction to alleviate its concerns about

the BCP deal.

The report principally addresses the $35,000 payments in

a section titled “Monrovia, 2013: Awash in Cash.” Id. at 30–

31. This section discusses what are repeatedly described as

“unusual, large” payments made to HTC members, referencing

Tah and McClain by name. Id. at 30. It states that NOCAL

characterized the payments as “bonuses,” using scare quotes

whenever it repeats the word “bonus,” and claims that the

payments “appear . . . to be linked to the HTC’s signing of

Block 13.” Id.

In support of its claim that the payments were “large” and

“unusual,” the report states that “there is no sign of equivalent

bonuses during” the surrounding years, “except for smaller

yearly bonuses paid shortly before Christmas[;]” that “the

payments represented a 160 percent increase on the reported

highest salary paid to a Liberian minister[;]” and that one HTC

member who was supposedly working for free nonetheless

received a payment. Id. The report then gives the definition of

bribery under Liberian law and references some of the

corruption surrounding the 2007 BCP deal—specifically,

payments NOCAL made to members of the Liberian legislature

to ensure approval of that earlier license, which NOCAL

deemed “lobbying fees,” and which the Liberian Government’s

General Auditing Commission later “classified as bribes.” Id.

A few weeks before Global Witness issued the report, it

sent letters to HTC members informing each that “we believe

that the payment made by NOCAL to you was most likely a

bribe, paid as a reward to ensure that [Block] 13 was negotiated

successfully,” and asking for a response. Compl. ¶¶ 91–92.

6

Several HTC members, including Tah, denied that the

payments were bribes, insisting they were bonuses authorized

by NOCAL’s board that were “appropriately earned given the

extraordinary success of the Exxon negotiations,” and pointing

out that all NOCAL employees received bonuses. Id. ¶¶ 93–95.

Global Witness included excerpts from these denials in the

report. Report at 30.

The report also discusses Exxon’s relationship to these

payments. It characterizes them as evidence of Exxon’s

possible “complicit[y]” in “Liberia’s corrupt oil sector,”

declaring that “Exxon should have known better.” Id. at 32–33.

According to the report, “Exxon . . . knew it was buying a

license with illegal origins” and the payments were “in effect

. . . likely made with Exxon’s money.” Id. Although stating that

“Global Witness believes that Exxon should have considered it

possible that money the company provided to NOCAL could

have been used as bribes in connection with Exxon’s Block 13

deal,” the report acknowledges that “Global Witness has no

evidence that Exxon directed NOCAL to pay Liberian officials,

nor that Exxon knew such payments were occurring.” Id. at 31–

32.

Lastly, Global Witness called on the Liberian government

to investigate the payments and, in the event such investigation

uncovers unlawful behavior, urged the U.S. Department of

Justice “to determine if the company violated the [Foreign

Corrupt Practices Act].” Id. at 32. Global Witness sent copies

of the report to the U.S. Attorney General and the Chairman of

the Securities and Exchange Commission. Compl. ¶¶ 100–02.

Following the report’s publication, the Liberian

government investigated the payments and concluded that they

did not “constitute[] bribe[s] within the context of [Liberian]

law” and were not “made so [the HTC] could undertake [an]

7

official act.” Id. ¶ 81 (internal quotation marks omitted). The

Liberian government nonetheless recommended that the HTC

members return the payments. Id. ¶ 83. Tah and McClain

refused, asserting that the payments were above-board bonuses

for a job well done. Id.

Believing that Catch me if you can falsely impugns their

integrity and reputations, Tah and McClain sued Global

Witness for defamation and false light invasion of privacy.

They dispute none of the facts contained in the report but argue

that Global Witness falsely “communicated [through

implication] . . . that . . . each took a bribe in exchange for their

roles in the Exxon purchase of Block 13.” Id. ¶ 31 (emphasis

omitted).

Global Witness responded with a special motion to dismiss

under the District of Columbia’s anti-SLAPP (strategic

lawsuits against public participation) statute, which seeks to

protect speakers from lawsuits “filed by one side of a political

or public policy debate aimed to punish or prevent the

expression of opposing points of view.” Competitive

Enterprise Institute v. Mann, 150 A.3d 1213, 1226 (D.C. 2016)

(internal quotation marks omitted). To defeat such a motion,

the plaintiff must “demonstrate[] that the claim is likely to

succeed on the merits,” even as the act severely limits

discovery. D.C. Code § 16-5502(b). A prevailing defendant

may seek an award of attorney’s fees. Id. § 16-5504(a). Global

Witness also filed a motion to dismiss for failure to state a

claim under Federal Rule of Civil Procedure 12(b)(6), arguing

that the complaint failed to plead defamation by implication,

that any defamatory implication was protected opinion, and

that, in any event, the complaint failed to plead actual malice

as required under the First Amendment.

8

The district court denied Global Witness’s special motion

because, in its view, the D.C. anti-SLAPP statute did not apply

in federal court. Tah v. Global Witness Publishing, Inc., No.

18-cv-2109 (D.D.C. June 19, 2019). The court, however,

granted Global Witness’s Rule 12(b)(6) motion, finding that

“the contents of the report are protected speech under the First

Amendment and cannot sustain a defamation claim.” Tah v.

Global Witness Publishing, Inc., 413 F. Supp. 3d 1, 3–4

(D.D.C. 2019).

Tah and McClain appeal, arguing, as they did in the district

court, that their allegations are sufficient to state a plausible

case of actual malice because Global Witness (1) began its

investigation with a preconceived story line, (2) received

denials from some of those involved, (3) harbored ill-will

toward Exxon, and (4) omitted Seward Cooper from the list of

payment recipients. Global Witness cross-appeals, arguing that

the anti-SLAPP statute applies in federal court and that the

district court’s denial of the special motion to dismiss deprived

it of the ability “to recover the expenses it has incurred in

defending against this meritless attack.” Appellees’ Br. 67.

Like the district court, we begin with the anti-SLAPP issue.

II.

Under the Supreme Court’s decision in Shady Grove

Orthopedic Associates., P.A. v. Allstate Insurance Co., to

decide whether a state (or district) law or rule—in this case the

D.C. anti-SLAPP statute—applies in a federal court exercising

diversity jurisdiction, we “first determine whether [a federal

rule of civil procedure] answers the question in dispute.” 559

U.S. 393, 398 (2010). If it does, the federal rule “governs . . .

unless it exceeds statutory authorization or Congress’s

rulemaking power.” Id.

9

Applying the Shady Grove test, our court held in Abbas v.

Foreign Policy Group, LLC that the D.C. anti-SLAPP act does

not apply in federal court. 783 F.3d 1328, 1334–37 (D.C. Cir.

2015). Without controlling guidance from the D.C. Court of

Appeals—at the time that court had yet to interpret the anti-

SLAPP act—we construed the statute’s “likely to succeed on

the merits” standard literally, finding that it “is different from

and more difficult for plaintiffs to meet than the standards

imposed by Federal Rules 12 and 56.” Id. at 1335. Accordingly,

we concluded that the D.C. anti-SLAPP statute impermissibly

“conflicts with the Federal Rules by setting up an additional

hurdle a plaintiff must jump over to get to trial.” Id. at 1334.

Global Witness argues that the D.C. Court of Appeals’s

subsequent decision in Competitive Enterprise Institute v.

Mann effectively abrogates Abbas. There, interpreting the anti-

SLAPP statute’s special motion to dismiss provision for the

first time, the Court of Appeals held, contrary to Abbas, that

the “D.C. Anti-SLAPP Act’s likelihood of success standard . . .

simply mirror[s] the standards imposed by Federal Rule 56,”

and that to decide a special motion to dismiss, the court “must

assess the legal sufficiency of the evidence” as it currently

stands at the time of the motion. Mann, 150 A.3d at 1236, 1238

n.32 (internal quotation marks omitted).

Even with this development, however, Abbas remains

circuit law and controls this case. The reason comes from Mann

itself, in which the Court of Appeals “agree[d] with Abbas that

the special motion to dismiss is different from summary

judgment” in two respects. Id. at 1238 n.32.

First, the special motion to dismiss “imposes the burden

on plaintiffs.” Id. Once a defendant makes a prima facie

showing that the lawsuit in question qualifies as a SLAPP, the

burden shifts to the plaintiff to defeat the special motion to

10

dismiss. Id. at 1237. By contrast, even a “movant” defendant

on a Federal Rule 56 summary judgment motion retains some

initial “burden of showing that there is no genuine issue of

fact.” Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 256

(1986).

Second, the Court of Appeals observed that, unlike a

summary judgment motion, a special motion to dismiss will

usually be decided “before discovery is completed.” Mann, 150

A.3d at 1238 n.32. By contrast, under Federal Rule 56,

summary judgment is typically “premature unless all parties

have had a full opportunity to conduct discovery.”

Convertino v. DOJ, 684 F.3d 93, 99 (D.C. Cir. 2012) (internal

quotation marks omitted). According to Global Witness,

however, the allowance for discovery under the anti-SLAPP

statute is identical to that under Federal Rule 56. The D.C.

Court of Appeals’s recent decision in Fridman v. Orbis

Business Intelligence Ltd., 229 A.3d 494 (D.C. 2020),

forecloses this argument. There, the court addressed the

provision of the anti-SLAPP act that stays discovery whenever

a special motion to dismiss is filed, except for “[w]hen it

appears likely that targeted discovery will enable the plaintiff

to defeat the motion and that the discovery will not be unduly

burdensome.” D.C. Code § 16-5502(c)(2). That standard, the

court explained, “is difficult to meet,” because the party

requesting discovery must show that it is actually “likely” that

“targeted discovery will enable him to defeat the special

motion to dismiss.” Fridman, 229 A.3d at 512–13. Thus,

“discovery normally will not be allowed.” Id. at 512. This

differs from Federal Rule 56, under which full discovery is the

norm, not the exception.

Although Mann may undermine some of Abbas’s

reasoning, the bottom line remains: the federal rules and the

anti-SLAPP law “answer the same question about the

11

circumstances under which a court must dismiss a case before

trial . . . differently,” and the anti-SLAPP law still “conflicts

with the Federal Rules by setting up an additional hurdle a

plaintiff must jump over to get to trial.” Abbas, 783 F.3d at

1333–34 (internal quotation marks omitted). Accordingly, the

district court properly applied Abbas to this case and denied the

special motion.

III.

“To survive a motion to dismiss, a complaint must contain

sufficient factual matter, accepted as true, to ‘state a claim to

relief that is plausible on its face.’” Ashcroft v. Iqbal, 556 U.S.

662, 678 (2009) (quoting Bell Atlantic Corp. v. Twombly, 550

U.S. 544, 570 (2007)). “We assume the truth of all well-

pleaded factual allegations and construe reasonable inferences

from those allegations in a plaintiff’s favor.” Nurriddin v.

Bolden, 818 F.3d 751, 756 (D.C. Cir. 2016). “Threadbare

recitals of the elements of a cause of action, supported by mere

conclusory statements, do not suffice.” Iqbal, 556 U.S. at 678.

In a defamation by implication case under D.C. law, “the

courts are charged with the responsibility of determining

whether a challenged statement is capable of conveying a

defamatory meaning.” White v. Fraternal Order of Police, 909

F.2d 512, 518 (D.C. Cir. 1990) (internal quotation marks

omitted). A plaintiff must show first that the “communication,

viewed in its entire context, . . . conveys materially true facts

from which a defamatory inference can reasonably be drawn,”

and second, that “the communication, by the particular manner

or language in which the true facts are conveyed, supplies

additional, affirmative evidence suggesting that the defendant

intends or endorses the defamatory inference.” Armstrong v.

Thompson, 80 A.3d 177, 184 (D.C. 2013) (emphasis omitted)

(quoting White, 909 F.2d at 520). Where, as here, plaintiffs

qualify as public officials—as Tah and McClain concede they

12

do—the First Amendment requires that they also allege that the

defamatory statement “was made with actual malice.” New

York Times Co. v. Sullivan, 376 U.S. 254, 279–80 (1964)

(internal quotation marks omitted). The First Amendment, the

Supreme Court long ago observed, enshrines “a profound

national commitment to the principle that debate on public

issues should be uninhibited, robust, and wide-open.” Id. at

270. The actual malice standard reflects the cornerstone First

Amendment principle that “speech relating to public officials

and public figures, as distinct from private persons, enjoys

greater protection.” Jankovic v. International Crisis Group

(Jankovic III), 822 F.3d 576, 584 (D.C. Cir. 2016).

The actual malice standard is famously “daunting.”

McFarlane v. Esquire Magazine, 74 F.3d 1296, 1308 (D.C.

Cir. 1996). A plaintiff must prove by “clear and convincing

evidence” that the speaker made the statement “with

knowledge that it was false or with reckless disregard of

whether it was false or not.” Jankovic III, 822 F.3d at 589–90

(second part quoting New York Times Co., 376 U.S. at 279–80).

“[A]lthough the concept of reckless disregard cannot be fully

encompassed in one infallible definition,” the Supreme Court

has “made clear that the defendant must have made the false

publication with a high degree of awareness of probable

falsity,” or “must have entertained serious doubts as to the truth

of his publication.” Harte-Hanks Communications, Inc. v.

Connaughton, 491 U.S. 657, 667 (1989) (alteration omitted)

(internal quotation marks omitted); see also id. at 688 (using

these formulations interchangeably). The speaker’s failure to

meet an objective standard of reasonableness is insufficient;

rather the speaker must have actually “harbored subjective

doubt.” Jankovic III, 822 F.3d at 589.

The dissent thinks this is an easy case. “In Global

Witness’s story,” the dissent asserts, “Exxon was the briber,”

13

Dissenting Op. at 1, yet the report admits that “Global Witness

ha[d] no evidence that Exxon directed NOCAL to pay Liberian

officials, nor that Exxon knew such payments were occurring,”

Report at 31.

Critically, however, neither Tah nor McClain advances

this theory—in their briefing to us, they never even mention the

sentence on which the dissent relies. They make four specific

arguments in support of their claim that Global Witness

possessed actual malice, supra at 8, not one of which is that

Global Witness had no evidence that Exxon was the briber, and

for good reason. At most, the report implies that NOCAL, not

Exxon, was the briber, thus rendering any lack of evidence as

to Exxon’s direction or knowledge of the payments totally

irrelevant. See Report at 32 (stating that Exxon “knew the risk”

and “should have considered it possible that money the

company provided to NOCAL could have been used as bribes

in connection with Exxon’s Block 13 deal” (emphasis added));

id. (noting that Global Witness asked Exxon for comment on

any “safeguards the company may have put in place to prevent

the possible misuse of its funds by NOCAL” (emphasis added)).

Contrary to the dissent, see Dissenting Op. at 6, a generic

statement accusing someone of acting with reckless

disregard—here, Tah and McClain’s claim that “Global

Witness subjectively knew that it had not been able to

determine whether the payments of $35,000 to Christiana Tah

and Randolph McClain were corrupt bribery payments,”

Appellants’ Br. 36—simply cannot be read to shoehorn in

every conceivable actual malice theory. Indeed, when our

dissenting colleague surfaced his theory at oral argument, it

was so foreign to appellants’ counsel that our colleague had to

spoon-feed him after he failed to get the initial hint. See Oral

Arg. Tr. at 10 (“Well, no, it’s worse. Isn’t it stronger than that,

counsel? We have no evidence.”). As our dissenting colleague

himself has made clear, “we do not consider arguments not

14

presented to us.” Diamond Walnut Growers, Inc. v. NLRB, 113

F.3d 1259, 1263 (D.C. Cir. 1997) (en banc). Or put another

way, “appellate courts do not sit as self-directed boards of legal

inquiry and research, but essentially as arbiters of legal

questions presented and argued by the parties before them.”

Carducci v. Regan, 714 F.2d 171, 177 (D.C. Cir. 1983).

We turn, then, to the “legal questions presented and

argued” by Tah and McClain. They advance what the district

court described as “several interlocking theories to support the

allegation of actual malice.” Tah, 413 F. Supp. 3d at 12. We

agree with the district court that these theories fail to support a

plausible claim that Global Witness acted with actual malice.

Tah and McClain first allege that Global Witness began

their investigation with “a preconceived story line” that they

argue “is plainly probative of actual malice.” Appellants’ Br.

19 (emphasis omitted). In support, they point out that the letters

in which Global Witness asked for comment state that the

$35,000 payments were “most likely” bribes. Compl. ¶¶ 91–

92.

Our court, however, has made clear that “preconceived

notions” or “suspicion[s]” usually do “little to show actual

malice.” Jankovic III, 822 F.3d at 597. After all, virtually any

work of investigative journalism begins with some measure of

suspicion. Thus, “concoct[ing] a pre-conceived storyline” by

itself is “not antithetical to the truthful presentation of facts.”

Id. at 597 (internal quotation marks omitted). Moreover,

because Global Witness sent the letters toward the end of its

investigative process and just a few weeks before publication,

the letters provide no support at all for the notion that Global

Witness’s conclusion was preconceived. As the district court

correctly observed, “[t]hat Global Witness had arrived at its

conclusion, right or wrong, by the time it reached out for

15

comment and shortly before publication is commonplace and

no surprise.” Tah, 413 F. Supp. 3d at 13. Finally, seeking

comment in advance of publication is a standard journalistic

practice. See, e.g., Responses, Associated Press,

http://www.ap.org/about/news-values-and-principles/telling-

the-story/responses (“We must make significant efforts to

reach anyone who may be portrayed in a negative way in our

stories, and we must give them a reasonable amount of time to

get back to us before we move the story.”). Drawing a

pernicious inference from adherence to such professional

standards would turn First Amendment case law on its head. In

any event, even an “extreme departure from professional

standards” is insufficient to prove actual malice on its own.

Harte-Hanks, 491 U.S. at 665.

Next, Tah and McClain seek to draw an inference of actual

malice from Global Witness’s failure to credit their denials.

This too finds no support in our First Amendment case law. A

publisher “need not accept ‘denials, however vehement; such

denials are so commonplace in the world of polemical charge

and countercharge that, in themselves, they hardly alert the

conscientious reporter to the likelihood of error.’” Lohrenz v.

Donnelly, 350 F.3d 1272, 1285 (D.C. Cir. 2003) (quoting

Harte-Hanks, 491 U.S. at 691 n.37). Although consistent with

each other, the denials contain no “evidence that could be

readily verified” of the sort that would provide “obvious

reasons to doubt the veracity of [Global Witness’s]

publication.” Id. (internal quotation marks omitted). As the

district court pointed out, the denials “fail” even to “contest the

facts that are [stated] in the Report.” Tah, 413 F. Supp. 3d at

13.

According to the dissent, our description of the law is

“obviously fallacious,” Dissenting Op. at 10, an odd accusation

given that we have done nothing more than quote from our

16

court’s decision in Lohrenz. Undaunted, the dissent attempts to

distinguish Lohrenz on the ground that Global Witness “had

‘no evidence’—and no witnesses—to contradict the six

denials.” Id. at 11 (quoting Report at 31). But that quotation

comes from the same sentence in Catch me if you can that the

dissent relies on for the proposition—irrelevant to the

arguments made by Tah and McClain, see supra at 13—that

Global Witness had no evidence that Exxon had paid bribes.

Contrary to the dissent, moreover, nothing in the six

denials comes close to the kind of “readily verifi[able]”

evidence, Lohrenz 350 F.3d at 1285, needed to support a

plausible—and we emphasize the word plausible—case that

Global Witness published with a “high degree of awareness of

probable falsity,” Harte-Hanks, 491 U.S. at 688 (alteration

omitted) (internal quotation marks omitted). See Dissenting

Op. at 12. To be sure, as the dissent points out, the denials state

that more than 140 others, such as NOCAL drivers and janitors,

also received bonuses. But as the report explains, “the vast

majority of these payments were smaller . . . by two orders of

magnitude” and “were not made to people who signed the

Exxon deal.” Report at 32. Indeed, the denials themselves

characterized the payments the HTC received as rewards for

those “who performed exceptionally in conducting the

negotiations on the Exxon Contract,” a rationale obviously

inapplicable to payments to other company employees. Compl.

¶ 93 (quoting Tah’s denial); see also id. ¶ 94 (quoting

McClain’s denial, which stated “[a]ny bonus given by our

superiors was in acknowledgement of the Team’s

extraordinary work after the completion of the landmark

Contract”). It is also true that the denials explain, as does the

report, that the pot of money used to make the payments was

“negotiated” as part of the larger Exxon deal, Dissenting Op. at

12; Report at 32, but we fail to see how that contradicts the idea

that the payments were bribes from NOCAL. The dissent refers

17

to the NOCAL board resolution approving the payments,

presumably because the denials claim that the payments were

“authorized by NOCAL’s Board of Directors.” Compl. ¶ 93

(quoting Tah’s denial). But according to the report, Global

Witness “requested, but had not yet received” a copy of the

board resolution, and the complaint alleges nothing to the

contrary. Report at 31.

Tah and McClain next argue that Global Witness harbored

ill-will and desired “to catch Exxon and [CEO Rex] Tillerson

in scandal.” Appellants’ Br. 25. In support, they rely on the fact

that the report is critical of Exxon and that Global Witness

subsequently sent letters reiterating the report’s conclusions to

the Department of Justice and the Securities and Exchange

Commission. Our court, however, has made clear that evidence

of ill will “is insufficient by itself to support a finding of actual

malice.” Tavoulareas v. Piro, 817 F.2d 762, 795 (D.C. Cir.

1987) (en banc); see also Harte-Hanks, 491 U.S. at 665

(“Petitioner is plainly correct in recognizing . . . that a

newspaper’s motive in publishing a story . . . cannot provide a

sufficient basis for finding actual malice.”). Regardless, neither

the report’s critical nature nor the letters sent to the Attorney

General and the SEC plausibly supports an ill-will theory. As

the district court aptly put it, the report’s “conclusion is not

evidence of its conception.” Tah, 413 F. Supp. 3d at 13.

The implications of Tah and McClain’s theory are

breathtaking: they would find support for an inference of actual

malice in a wide swath of investigative journalism that turns

out to be critical of its subject. “It would be sadly ironic for

judges in our adversarial system to conclude . . . that the mere

taking of an adversarial stance is antithetical to the truthful

presentation of facts.” Tavoulareas, 817 F.2d at 795.

18

Finally, Tah and McClain argue that “the stunning failure

of Global Witness to include . . . Seward Cooper, as among the

[HTC] members who received a $35,000 bonus” reveals actual

malice because Cooper, along with Tah, determined that the

payments were legal. Appellants’ Br. 34. We do not see how

this omission shows awareness of falsity or reckless disregard

for the truth. If anything, that one of the lawyers responsible

for conducting a legal analysis of the payments was himself in

line to receive one makes the payments even more suspicious.

For all these reasons, Tah and McClain have failed to

plausibly allege that Global Witness acted with actual malice.

This deficiency proves fatal not only to their defamation claims

but to their false light claims as well. See Farah v. Esquire

Magazine, 736 F.3d 528, 540 (D.C. Cir. 2013) (explaining that

a “plaintiff may not use related causes of action to avoid the

constitutional requisites of a defamation claim” and that “[t]he

First Amendment considerations that apply to defamation

therefore apply also to [plaintiffs’] counts for false light”

(internal quotation marks omitted)).

IV.

We affirm the district court’s dismissal of the complaint,

as well as its denial of the anti-SLAPP motion.

So ordered.

SILBERMAN , Senior Circuit Judge, dissenting in part:

Global Witness (Appellee) falsely insinuated that former

Liberian officials (Appellants) took bribes from Exxon. It

admitted that it had no evidence that Exxon had contacted

Appellants, directly or indirectly, with respect to the alleged

payments. And the evidence Global Witness did have

suggested the payments at issue were proper staff bonuses, not

bribes. Nevertheless, the Majority creates a whole new theory

of the case—one not advanced by any Party—that the

Appellants were bribed not by Exxon, but by their own

principal, the National Oil Company. According to the

Majority, its new narrative is so unassailable that, even at the

12(b)(6) stage, it precludes an inference that Global Witness

harbored subjective doubts as to the implied accusation of

bribery.

I

As Global Witness explained, “this is a story of

bribery.” J.A. 58. Bribery, as it is commonly understood,

involves a quid pro quo. See McDonnell v. United States, 136

S. Ct. 2355, 2372 (2016); accord J.A. 82 (“[A] payment given

so a public servant will undertake an official act.”). As such,

bribery has three necessary components: A briber, a bribee,

and an exchange. In Global Witness’s story, it seems obvious

that Exxon was the briber, Appellants were the bribees, and the

trade was $35,000 to ensure the deal goes through. Without

one element, there is obviously no bribery. In other words, if

no briber—or no bribe—then no bribee.

In its cross-appeal, Global Witness contends that its

Report was not even defamatory—it simply raised questions.

Of course, Appellants disagree, claiming that the Report, Catch

me if you can, falsely insinuated that they took bribes from

Exxon to approve the Block 13 deal.

The district court easily determined that Global

Witness’s story contained the defamatory implication that

2

Appellants took bribes from Exxon. Tah v. Glob. Witness

Publ’g, Inc., 413 F. Supp. 3d 1, 10 (D.D.C. 2019) (“[T]he

import of the Report [is] that there was bribery—either by [the

National Oil Company], Exxon, or both—in connection with

the post-negotiation payments to Liberia’s chief

representatives.”); id. at 9 (Global Witness’s Report “literally

[drew] a line between Exxon’s money, the bonuses, and the sale

of the license for Block 13”); accord Appellant Br. 15 (“The

story was that Exxon . . . brazenly engineered a Liberian oil

purchase through bribery.”). According to Global Witness,

Exxon’s payment to the National Oil Company and the

subsequent bonuses were “unusual” and “suspicious.” J.A. 82,

83, 85. Exxon, as Global Witness saw it, “was under no

obligation to pay most of the money” it transferred to the

National Oil Company; “$4 million of its $5 million payment

was characterized as a ‘bonus.’” J.A. 84 (emphasis in original).

And, as Global Witness reminds its readers, the National Oil

Company has a record of bribing officials on behalf of oil

companies. The “bonuses” then paid to the officials were,

Global Witness wrote, “unusual, large payments to officials

who signed the Exxon deal.” J.A. 85. Global Witness further

noted that these individual “bonuses” were likely derived from

the same bank account into which Exxon paid the initial $4

million “bonus” to the National Oil Company.

The court explained how Global Witness “tied

ExxonMobil’s payments [to the National Oil Company] for the

acquisition of rights in Block 13 with how [the Company]

shared some of that money with its negotiators, including

Plaintiffs.” Tah, 413 F. Supp. 3d at 9. One chart in the Report

tracked payments from Exxon to the National Oil Company to

members of the Hydrocarbon Technical Committee, including

Appellants Christiana Tah and Randolph McClain. The district

court noted that this chart, listing the amount of each official’s

bonus, “had the effect of literally drawing a line between

Exxon’s money, the bonuses, and the sale of the license for

Block 13.” Id.

3

The Report also connected Exxon to the bribes by

making repeated parallels to the 2003 transaction. In that deal,

Global Witness asserted that the National Oil Company paid

bribes to legislators on behalf of the Broadway Consolidated

oil company to secure ratification of its purchase. J.A. 68; see

also J.A. 84 (noting that the National Oil Company had also

paid bribes on behalf of Oranto Petroleum). And now, Exxon

was stepping into Broadway’s shows. As the district court

explained, “a reasonable reader easily could have understood

the Report to imply that the [earlier legislative] bribes and the

2013 [Technical Committee] bonuses were of a piece.” Tah,

413 F. Supp. 3d at 9. And, the district judge noted, “[i]f that

were not the case, the Report would have no reason to advocate

for an investigation” (since there is no direct evidence of

bribery). Id. at 10 (emphasis added).

The court also rejected the Appellee’s argument that its

story actually negated any inference of bribery because it

expressly stated that “Global Witness has no evidence that

Exxon directed [the National Oil Company] to pay Liberian

officials, nor that Exxon knew such payments were occurring.”

Id. (quoting J.A. 83). After the Report repeatedly insinuated

bribery, this disclaimer “did not negate the inference that

ExxonMobil’s money was, in part, paid as bribes to [Company]

representatives who signed the lease agreement.” Id. This

statement merely “admit[ed] that the Global Witness

suspicions and calls for investigations of ExxonMobil and [the

National Oil Company] lacked any evidence that the former

had involvement in monies paid to employees of the latter.” Id.

(emphasis in original).

II

My disagreement with the district court is limited to the

actual malice question (my disagreement with the Majority is

much broader). In New York Times Co. v. Sullivan, 376 U.S.

254 (1964), the Supreme Court set forth the well-known rule

that, to hold a defendant liable for defaming a public figure, a

4

plaintiff must prove the defendant acted with “actual malice.”

Id. at 279–80. That is, with knowledge that the statement was

false or with reckless disregard for the truth. Id. at 280. As the

Supreme Court saw it, this scienter requirement appropriately

balanced (as a policy matter) the vindication of reputational

harms with the need to protect unintentional falsehoods that

inevitably arise as part of vibrant debate. Id. at 271–72. The

actual-malice rule makes the speaker’s state of mind the

constitutional gravamen in any defamation case brought by a

public figure.

The Majority emphasizes that actual malice is a

subjective test. Majority Op. 12 (citing Jankovic v. Int’l Crisis

Grp., 822 F.3d 576, 589 (D.C. Cir. 2016)). But it is important

not to confuse what a plaintiff must ultimately show with the

kind of evidence he may use to make that showing. It is the

rare case in which a defendant will confess his state of mind

and thus allow the plaintiff to prove actual malice with direct

evidence. Accordingly, as the Appellee concedes, actual

malice “is ordinarily inferred from objective facts.”

Washington Post Co. v. Keogh, 365 F.2d 965, 967 (D.C. Cir.

1966); accord Appellee Br. 50.

In St. Amant v. Thompson, the Supreme Court listed

three examples of objective circumstances that permit a

subjective inference of actual malice: (1) “where a story is

fabricated by the defendant . . . or is based wholly on an

unverified anonymous telephone call;” (2) “when the

publisher’s allegations are so inherently improbable that only a

reckless man would have put them in circulation;” and (3)

“where there are obvious reasons to doubt” the basis for the

story. 390 U.S. 727, 732 (1968); see also Tavoulareas v. Piro,

817 F.2d 762, 790 (D.C. Cir. 1987) (en banc). So even in the

absence of contradictory evidence, a story may be so facially

implausible or factually flimsy that the jury may infer that it

must have been published with reckless disregard for the truth.

See Hunt v. Liberty Lobby, 720 F.2d 631, 646 (11th Cir. 1983).

And even assuming a plausible story, the question remains

5

whether “the cumulative force of the evidence to the contrary”

should give the publisher obvious reasons for doubt.

McFarlane v. Sheridan Square Press, Inc., 91 F.3d 1501, 1514

(D.C. Cir. 1996); see Jankovic, 822 F.3d at 597. If the publisher

moves forward without reasonably dispelling his doubts, actual

malice may be inferred. Lohrenz v. Donnelly, 350 F.3d 1272,

1284 (D.C. Cir. 2003).

St. Amant’s examples thus suggest a straightforward

framework for evaluating contentions of actual malice. We

first assess the inherent plausibility of a defendant’s story as

well as the facts in support. And if we find the story objectively

plausible, we then ask whether evidence to the contrary creates

obvious reasons for doubt.

* * *

I turn to whether Global Witness’s accusation that

Exxon bribed the Appellants—the case before us—is facially

plausible. Appellants claim that Global Witness knew it lacked

any support for insinuating that the payments to Tah and

McClain were bribes. Thus, a jury could infer that Global

Witness subjectively doubted the truth of its Report.

I agree. In my view, because Global Witness’s story is

obviously missing (at least) one necessary component of

bribery, it is inherently improbable. Although it accused

Appellants of taking bribes from Exxon, Global Witness admits

that it had “no evidence that Exxon directed the [National Oil

Company] to pay Liberian officials, nor that Exxon knew such

payments were occurring.” J.A. 83 (emphasis added). In other

words, despite all its investigating, Global Witness uncovered

nothing to demonstrate that Exxon was the briber and nothing

to even suggest there was an agreed upon exchange. Accord

Tah, 413 F. Supp. 3d at 10 (“Global Witness’s suspicions and

calls for investigations of ExxonMobil and [the National Oil

Company] lacked any evidence that the former had

involvement in monies paid to employees of the latter.”)

6

(emphasis in original). And with no privity between Exxon and

the Technical Committee members, it is bizarre to accuse

Appellants of taking a bribe. As St. Amant teaches, it is

sufficient to infer—on this basis alone—that Appellee acted

with knowing disregard for the veracity of its publication.

The Majority’s assertion that this argument was never

made by the Appellants leads me to wonder whether we

received the same briefs. In my copy, Appellants argue that

“Global Witness subjectively knew that it had not been able to

determine whether the payments of $35,000 to Christiana Tah

and Randolph McClain were corrupt bribery payments.

Yet . . . Global Witness proceeded to present to readers the

defamatory message that in fact [] Tah and [] McClain had

taken bribes.” Appellant Br. 36 (emphasis in original). That

sounds to me a whole lot like accusing Global Witness of

publishing its story with no evidence to back it up. The

Majority, moreover, faults me for assessing the inherent

(im)plausibility of Global Witness’s story, without a specific

request from Tah and McClain to do so. But (as discussed)

“inherently implausible” is a legal standard by which we assess

Appellants’ arguments—not an argument to be advanced. See

Kamen v. Kemper Fin. Servs., Inc., 500 U.S. 90, 99 (1991); cf.

Eldred v. Ashcroft, 255 F.3d 849, 853 (D.C. Cir. 2001)

(Sentelle, J., dissenting) (“Merely because the parties fail to

advance the proper legal theory underlying their claim does

not—indeed cannot—prevent a court from arriving at the

proper legal disposition.”).

To be sure, Appellants did not quote the “no evidence”

paragraph in their brief; it was “the Court” at oral argument that

focused on this damning language. But it is hardly a new

argument; it is only evidence—although powerful evidence—

supporting Appellants’ argument. Apparently, the Majority

also recognizes the significance of the passage and wishes to

rule it out of order. But the Appellee itself injected this

statement into the controversy when it brandished it as

supposedly exculpatory evidence. Appellee Br. 19, 35.

7

Global Witness points to other facts that support its

story, but none amount to a hill of beans. It emphasizes the

obvious point: Payments were made to Appellants. Then it

notes these payments were “likely” sent from the same account

where Exxon deposited its $4 million “bonus” to the National

Oil Company itself. Although a sly suggestion of wrongdoing,

when you think about it, it’s a non sequitur. If that were support

for a bribe, any investment banker’s commission could be

illegal.

Next, Global Witness justifies its story based on a past

“history” of bribery by the National Oil Company. The

Company had previously, in connection with the 2003 bid, paid

out bribes to legislators on behalf of another oil company in

order to ratify a transaction. Our situation is quite different; in

this case, the recipients of the “bribes” were the National Oil

Company’s own agents and employees. And connecting

bribery to the 2013 circumstances from the wholly separate

2003 bid—in which Tah and McClain were not even

involved—is a grossly unfair inference. Global Witness

attempts to tar the conduct of two parties to a transaction with

the prior bad acts of entirely different people. That is entirely

illegitimate.

It is significant, moreover, that the National Oil

Company paid out bonuses to all those involved in the

negotiations—including American consultants—as well as

low-ranking employees. Yet the story focuses on members of

the Technical Committee as if they were special transgressors.

But for Global Witness’s story to be true, Exxon was somehow

spreading bribes left and right like Johnny Appleseed. The

much more obvious explanation is that the “bonuses” were in

fact bonuses paid for outstanding performance. Certainly at the

12(b)(6) stage, Appellants are entitled to that inference (Global

Witness is not entitled to its speculative inference to the

contrary). See Palin v. New York Times Co., 940 F.3d 804, 815

(2d Cir. 2019).

8

The timing and manner of the payments are further

indications of bonuses, not bribes. Recall that in connection

with Broadway’s 2003 bid for Block 13, Global Witness

explained that the National Oil Company supposedly paid most

of Broadway’s bribes to legislators before approval of the deal.

In contrast, the 2013 payments from the National Oil Company

to its own negotiators, staff, and consultants occurred after the

deal was completed. J.A. 67. The latter payments—as Global

Witness knew—were only initiated after the approval of a

board resolution authorizing up to $500,000 in bonuses. And,

as Global Witness also acknowledges, the board had the full

legal authority to take this action. These payments, openly

made, are also indicia of proper bonuses; bribes, which are

illegal everywhere, are typically made in the dark. See DiBella

v. Hopkins, 403 F.3d 102, 117 (2d Cir. 2005) (defendant knew

that payments to plaintiff were not surreptitious, which

supported the jury’s conclusion that the defendant made a

bribery accusation with actual malice).

For all these reasons, I consider Global Witness’s

Report inherently improbable. On its face, it’s a house of cards:

With “no evidence” supporting Exxon as a briber, Tah and

McClain could not be Exxon’s bribees. Nor is there any

evidence—not a shred—that the bonuses paid by the National

Oil Company to Appellants were themselves bribes. There is

no indication that the National Oil Company had a corrupt

motive, nor that Appellants were asked to perform an illegal or

improper task. I would therefore conclude, based on the

foregoing alone, that it is sufficient to infer actual malice at the

12(b)(6) stage since the story is inherently improbable.

* * *

There is more: Global Witness had additional, “obvious

reasons” to doubt its Report, which would also support an

inference of actual malice. St. Amant, 390 U.S. at 732 (example

three). All the eyewitnesses to the transaction that responded

to Global Witness explained precisely why it was wrong. And

9

Global Witness had no facts that would cause it to discount

these explanations.

Six individuals—four members of the Technical

Committee and two consultants—responded to Global

Witness’s accusations of bribery. Each denied that bribery

occurred. At least four offered specific, fact-based

explanations as to why Global Witness was wrong. Christiana

Tah (a Yale Law School graduate)1 explained that bonus

payments were made to all National Oil Company staff after

the Exxon deal was concluded. Robert Sirleaf, a Technical

Committee member and chair of the Company’s Board of

Directors, similarly explained that bonuses were paid to the

entire company after the Parties concluded the contract and

exchanged consideration. And, he added, a bonus was called

for: Liberia received a signing payment fifteen times larger

than in any prior transaction. Natty Davis, another Committee

member and the Chair of the National Investment Commission,

added that the decision to pay the bonuses was approved by a

formal resolution of the Company’s board. Randolph McClain

noted that the Exxon contract was “extraordinary enough to be

used as a model for all future contracts of this nature.” J.A. 44.

Two American consultants—not mentioned by the

Majority and not targeted by Global Witness’s story—also told

Global Witness that its Report was false. Each received

$15,000 bonuses. One consultant, Jeff Wood, explained that

Exxon’s payment to the National Oil Company was not

“voluntary” as Global Witness reported—it was negotiated as

part of the transaction. J.A. 45. Moreover, Wood noted, it

made no sense to equate legislative bribes paid following

Broadway’s 2003 bid to the National Oil Company’s 2013

payments to its own staff. Wood again explained that all the

employees of the National Oil Company received payments,

not just those that signed the deal. Last, Wood wrote that it was

1

That surely is not inculpatory.

10

absurd to infer bribery because no similar bonuses had been

paid in recent years. Since no other deals had been concluded,

there was no success to reward.

The Majority, however, asserts that a publisher “need

not accept denials, however vehement” as a matter of law.

Majority Op. 15 (quoting Lohrenz, 350 F.3d at 1285); see also

Appellee Br. 54 (“[D]enials do not and cannot constitute

‘evidence’ as a matter of law.”). This proposition is obviously

fallacious.2 It of course depends on the substance and context

of the denial. If denials were legally irrelevant, then any

response of the target could be ignored.3 Indeed, the Supreme

Court—even while professing in dicta that the mere existence

of a denial need not be considered—has evaluated the contents

of denials to determine whether a publisher acted with actual

malice. See Harte-Hanks Commc’ns, Inc. v. Connaughton, 491

U.S. 657, 691–92 (1989). And it has noted that certain key

denials should reasonably be expected to kill stories. Id. at 682.

To be sure, we discounted the probative value of the

denials in Lohrenz v. Donnelly. 350 F.3d 1272. Lohrenz

involved a publication that questioned the competence of a

2

The Majority protests that it has “done nothing more than

quote” from Lohrenz. Majority Op. 15–16. But those quotations

(strung together out of order) do not give an accurate impression of

our holding in that case. In Lohrenz, we held at the summary

judgment stage that “[u]nlike evidence that could be readily verified,

the Navy’s denials did not give [the defendant] ‘obvious reasons’ to

doubt the veracity of her publication.” 350 F.3d at 1285 (emphasis

added and internal citations omitted). This was because, as we

explained, those denials were mere assertions contradicted by other

evidence. Id.

3

Suppose a reporter plans to accuse X of robbery in New

York City on December 1st. But X denies the allegation, explaining

that he was in Los Angeles on that date. Obviously, this denial would

need to be considered and verified by a responsible reporter.

11

female fighter pilot. According to the story, the pilot was

substandard, should not be flying, and was only assigned to the

F-14 program on account of a “politically driven policy.” Id.

at 1284. The publisher’s source was one of the pilot’s former

training officers, who we characterized as “a knowledgeable,

non-anonymous source.” Id. Furthermore, the publisher had

obtained additional information from the Navy that confirmed

the training officer’s claims—namely, that the pilot had

received a number of accommodations during training, which

other officers agreed were “excessive.” J.A. 1285.

Nevertheless, the Navy denied the story, and officials told the

publisher that its conclusions were inaccurate.

But as we explained, the fact of a denial, in itself,

“hardly alert[s] the conscientious reporter to the likelihood of

error.” Lohrenz, 350 F.3d at 1285 (quoting Harte-Hanks, 491

U.S. at 691 n.37). Rather, the specific content of a denial may

well give the publisher obvious reasons to doubt the veracity of

the publication. See id.; Montgomery v. Risen, 197 F. Supp. 3d

219, 263 (D.D.C. 2016), aff’d, 875 F.3d 709 (D.C. Cir. 2017).

So in Lohrenz, we reasoned that the Navy’s denials were

contradicted by the publisher’s interview with the flight

instructor. 350 F.3d at 1285. In light of the evidence on both

sides of the question, the Lohrenz publisher did not have any

obvious reason to doubt its story.

Global Witness, however, did not have evidence on

both sides of the issue. It had “no evidence”—and no

witnesses—to contradict the six denials. The cumulative

balance of the evidence thus gives Global Witness obvious

reasons for doubt. See McFarlane, 91 F.3d at 1514.

I am dumbfounded by the Majority’s assertion that the

denials in our case contain “no readily verifiable

information . . . that would provide ‘obvious reasons to

doubt.’” Majority Op. 15. The denials were specific, and it

was within Global Witness’s power to easily inquire into

whether other employees received bonuses, the content of the

12

Board’s resolution approving the bonuses, and whether the $4

million to the National Oil Company was negotiated as part of

the purchase price (etc.).

The Majority discounts these facts by weighing the

evidence and drawing inferences against Tah and McClain. See

Majority Op. 17 (“[T]he dissent refers to the NOCAL board

resolution approving the payments . . . . But . . . Global Witness

‘requested, but had not yet received’ a copy . . . and the

complaint alleged nothing to the contrary.”) (emphases added);

id. at 16 (discounting the fact that bonuses were paid to all

employees because the largest bonuses were paid to the

Technical Committee). Such weighing of the evidence is, of

course, impermissible at the 12(b)(6) stage. As the Second

Circuit recently reiterated in another defamation case, “[I]t is

not the [] court’s province to dismiss a plausible complaint

because it is not as plausible as the defendant’s theory.” Palin,

940 F.3d at 815.

* * *

In sum, the dramatic indication of actual malice is the

statement in Global Witness’s story to which we have

previously referred. J.A. 83 (Global Witness had “no evidence

that Exxon directed the [National Oil Company] to pay

Liberian officials, nor that Exxon knew such payments were

occurring.”).4 As we noted, Global Witness raised this point

itself in a futile effort to rebut defamation. But Global Witness

is hoist on its own petard. As I have explained, rarely does he

4

Perhaps the lack of evidence explains why the district court

was confused as to whether the bribes came from Exxon—which is

the obvious import of the story—or the National Oil Company—

which makes no sense. See Tah, 413 F. Supp. 3d at 10 (“[T]he import

of the Report [was] that there was bribery—either by [the National

Oil Company], Exxon, or both—in connection with the post-

negotiation payments to Liberia’s chief representatives.”). Of

course, “both” implies that even though the National Oil Company

may have paid out the bribes, it did so on Exxon’s behalf.

13

who defames another actually admit doubts as to the truth of

the accusation. This statement comes as close as it gets to such

a concession. In light of that admission, Global Witness’s story

is not just implausible, it’s ridiculous.

Circumventing the devastating impact of this statement,

the Majority creates a new narrative: The Global Witness

Report accused only the National Oil Company—not Exxon—

of paying bribes. With all due respect, the Majority is

employing judicial jiu-jitsu. At no time did the Appellee even

hint—in its briefs or oral argument—that was its defense. The

Appellee argues the district court erred in finding the Report

defamatory for only three reasons: because (1) it calls for

investigations, Appellee Br. 31–34; (2) it includes a disclaimer

that Global Witness had not determined the payments were

improper, Appellee Br. 34–35; and (3) the Report never uses

the word “bribe” to describe the payments, Appellee Br. 35–36.

The Majority’s judicial refashioning of the defamatory

implication is entirely illegitimate. See Diamond Walnut

Growers, Inc. v. NLRB, 113 F.3d 1259, 1263 (D.C. Cir. 1997)

(en banc).

The Majority cloaks its improper fashioning of a wholly

new argument by accusing me of doing the same. But if I’m

misreading Global Witness’s Report to imply that Exxon

bribed Appellants, I’m in quite good company. After all, the

district court endorsed my reading. Tah, 413 F. Supp. 3d at 10.

So did the Appellant. Appellant Br. 15, 23–24. And the

Liberian government. See J.A. 42. Even the Appellee assumes

that if the Report contains a defamatory implication, it would

be that Exxon bribed Tah and McClain. Otherwise, the

Appellee would not have argued its disclaimer, that “Global

Witness has no evidence that Exxon directed [the National Oil

Company] to pay Liberian officials,” is somehow exculpatory.

Appellee Br. 34–35. As the Majority recognizes, this statement

is irrelevant if one assumes that the National Oil Company was

the briber. In sum, the Majority’s theory not only falls out of

the clear blue sky, but it is also a sub silento overruling of the

14

district judge. After all, Exxon looms over the whole story. It

is impossible to visualize the article without Exxon playing the

part of the evil genius, orchestrating the implied corruption.

The Majority’s theory is not just a new argument—it’s a new

case.

Perhaps the Majority’s theory is not advanced by any

Party because the theory makes even less sense than if Exxon

were the briber. In the revisionist view, the National Oil

Company bribed its own agents and employees to do their jobs.

Tellingly, the Majority offers no motive for a bribe. After all,

the Liberian government ordered the sale of the Block 13

license for failure to make any progress in developing potential

oil reserves. J.A. 69. So there is no reason to think that the

Appellants had the authority to hold up the transaction. Nor is

there a reason to believe the National Oil Company would have

benefitted from the delay.

In any event, the Majority’s narrative is procedurally

inappropriate. At the 12(b)(6) stage, we accept all reasonable

defamatory readings of the Report advanced by the plaintiff.

See Weyrich v. New Republic, Inc., 235 F.3d 617, 627 (D.C.

Cir. 2001). As the district court ably explained, there can be no

doubt that one defamatory implication of the Report is that

Exxon bribed Appellants. That allegation—actually pressed by

Tah and McClain—must be analyzed for actual malice.

* * *

The Majority’s opinion creates a profoundly troubling

precedent. By fashioning a different defamatory implication on

its own, the Majority embraces a telling example of judicial

“creativity.” Still, its approach seems sui generis; I rather

doubt we will ever see its like again. On the other hand, the

Majority’s misunderstanding of the doctrinal framework of

New York Times v. Sullivan’s actual malice concept is

profoundly erroneous. And that will distort our libel law. But

perhaps most troublesome is the conflict it creates with the

15

Second Circuit (not to mention the Supreme Court) concerning

the role of a court when applying Rule 12(b)(6) in the libel

context. “The test is whether the complaint is plausible, not

whether it is less plausible than an alternative explanation.”

Palin, 940 F.3d at 815.

III

After observing my colleagues’ efforts to stretch the

actual malice rule like a rubber band, I am prompted to urge the

overruling of New York Times v. Sullivan. Justice Thomas has

already persuasively demonstrated that New York Times was a

policy-driven decision masquerading as constitutional law. See

McKee v. Cosby, 139 S. Ct. 675 (2019) (Thomas, J., concurring

in denial of certiorari). The holding has no relation to the text,

history, or structure of the Constitution, and it baldly

constitutionalized an area of law refined over centuries of

common law adjudication. See also Gertz v. Robert Welch,

Inc., 418 U.S. 323, 380–88 (1974) (White, J., dissenting). As

with the rest of the opinion, the actual malice requirement was

simply cut from whole cloth. New York Times should be

overruled on these grounds alone.

Nevertheless, I recognize how difficult it will be to

persuade the Supreme Court to overrule such a “landmark”

decision. After all, doing so would incur the wrath of press and

media. See Martin Tolchin, Press is Condemned by a Federal

Judge for Court Coverage, New York Times A13 (June 15,

1992) (discussing the “Greenhouse effect”). But new

considerations have arisen over the last 50 years that make the

New York Times decision (which I believe I have faithfully

applied in my dissent) a threat to American Democracy. It must

go.

Twenty-five years ago, I urged the overruling of a

similarly illegitimate constitutional decision, Monroe v. Pape,

365 U.S. 167 (1961). Our court was confronted with the vexing

question of whether qualified immunity shielded government

16

officials accused of constitutional torts from discovery into

their motivations. See Crawford-El v. Britton, 93 F.3d 813, 815

(D.C. Cir. 1996) (en banc), vacated, 523 U.S. 574 (1998). In a

concurring opinion, I suggested a solution to accommodate

Pape: When a defendant offers a proper motive, we should

allow an objective inquiry into only whether the proffered

motive is pretextual. Id. at 834–35; cf. Halperin v. Kissinger,

807 F.2d 180, 188 (D.C. Cir. 1986). When Crawford-El

reached the Supreme Court, four Justices agreed with my

approach. 523 U.S. at 602 (Rehnquist, C.J., dissenting); id. at

612 (Scalia, J., dissenting).

But I went even further in my concurrence: I urged the

Supreme Court to overrule Pape (and, while they’re at it,

Bivens5 as well). 93 F.3d at 832. Justices Scalia and Thomas

agreed with me. Crawford-El, 523 U.S. at 612 (Scalia, J.

dissenting). Both Pape and Bivens are prime examples of rank

policymaking by the High Court, not legitimate exercises of

constitutional interpretation. See also Hernandez v. Mesa, 140

S. Ct. 735, 752 (2020) (Thomas, J., concurring) (continuing to

call for the Court to abandon Bivens).

I recognized, however, that convincing the Court to

overrule these precedents would be an uphill battle. As I wrote,

the Court has committed itself to a constitutional Brezhnev

doctrine.6 That is, once the Court has “constitutionalized” a

new area of the law, it will never willingly retreat. The long-

5

Bivens v. Six Unknown Named Agents of Fed. Bureau of

Narcotics, 403 U.S. 388 (1971).

6

“When forces that are hostile to socialism try to turn the

development of some socialist country towards capitalism, it

becomes not only a problem of the country concerned, but a common

problem and concern of all socialist countries.” Leonid Brezhnev,

Remarks to the Fifth Congress of the Polish United Workers’ Party

(Nov. 13, 1968). Thus, one a country has turned communist, it can

never be allowed to go back.

17

term consequence of this policy is obvious: An ever-expanding

sphere of influence for the Judiciary at the expense of the

policymaking branches.

In a short concurring opinion, Justice Kennedy

lamented my criticism. He warned that “[w]e must guard

against disdain for the judicial system,” i.e., the Supreme Court.

Crawford-El, 523 U.S. at 601. In his view, criticism of the

Court is tantamount to an attack on the Constitution. He

cautioned, “if the Constitution is to endure, it must from age to

age retain ‘th[e] veneration which time bestows.’” Id. (quoting

The Federalist No. 49, at 314 (Madison) (C. Rossiter ed.,

1961)). Apparently, maintaining a veneer of infallibility is

more important than correcting fundamental missteps.

To the charge of disdain, I plead guilty. I readily admit

that I have little regard for holdings of the Court that dress up

policymaking in constitutional garb. That is the real attack on

the Constitution, in which—it should go without saying—the

Framers chose to allocate political power to the political

branches. The notion that the Court should somehow act in a

policy role as a Council of Revision is illegitimate. See 1 The

Records of the Federal Convention of 1787, at 138, 140 (Max

Farrand ed., 1911). It will be recalled that maintaining the

Brezhnev doctrine strained the resources and legitimacy of the

Soviet Union until it could no longer be sustained.

Admittedly, the context of the Times opinion made the

Court’s decision attractive as a policy matter. The case

centered on a full-page advertisement soliciting donations for

the civil rights movement and legal defense of Dr. Martin

Luther King, Jr. 376 U.S. at 256–57. The advertisement

claimed that civil rights proponents faced an “unprecedented

wave of terror” from “Southern violators” denying

constitutional guarantees to African Americans. Id. at 256. It

described “truckloads of police armed with shotguns and tear-

gas” that “ringed” a college campus in Montgomery, Alabama.

Id. at 257. It further asserted that state authorities padlocked

18

the dining hall “in an attempt to starve [the students] into

submission.” Id. Various claims in the ad were inaccurate, and

The Times eventually published a retraction. Id. at 261.

Sullivan sued, alleging the advertisement’s false

statements libeled him because, as commissioner of public

affairs, he supervised the police department. Id. at 256, 262.

After just two hours and twenty minutes of deliberation, an

Alabama jury awarded Sullivan $500,000 (the largest libel

judgment in Alabama history), and the state Supreme Court

affirmed. Anthony Lewis, Make No Law: The Sullivan Case

and the First Amendment 33, 45 (1991).

When the Supreme Court reversed, its decision was

seen as a “triumph for civil rights and racial equality.” E.g.,

Geoffrey Stone, New York Times Co. v. Sullivan, in The Oxford

Companion to the Supreme Court of the United States 586–87

(1992). The point of these suits had less to do with repairing

reputations and more to do with deterring the northern press

from covering civil rights abuses. Southern officials, as

Anthony Lewis succinctly explains, had thus twisted “the

traditional libel action . . . into a state political weapon to

intimidate the press”:

The aim was to discourage not false but true

accounts of libel under a system of white

supremacy: stories about men being lynched

for trying to vote, about cynical judges using

the law to suppress constitutional rights,

about police chiefs turning attack dogs on

men and women who wanted to drink a Coke

at a department-store lunch counter. It was

to scare the national press—newspapers,

magazines, the television networks—off the

civil rights story.

Lewis, Make no Law at 35.

19

Indeed, the day after the Alabama court’s verdict, the

Alabama Journal (a Montgomery paper) celebrated the result.

An editorial trumpeted that the case would cause the “reckless

publishers of the North . . . to make a re-survey of their habit of

permitting anything detrimental to the South and its people to

appear in their columns.” Id. at 34. “The Times was

summoned more than a thousand miles to Montgomery to

answer for its offense. Other newspapers and magazines face

the same prospect.” Id. Even before the Supreme Court issued

the Times decision, a second suit filed by a mayor—based on

the same ad—had already resulted in another $500,000 verdict

against The Times. Id. at 35. And three additional suits

remained pending. Id. CBS had similarly been sued for $1.5

million over a televised program that depicted the difficulties

of African Americans in registering to vote. Id. at 36. By 1964,

southern officials had filed almost $300 million in libel suits

against the northern press. Id.

One can understand, if not approve, the Supreme

Court’s policy-driven decision.7 There can be no doubt that the

New York Times case has increased the power of the media.

Although the institutional press, it could be argued, needed that

protection to cover the civil rights movement, that power is now

abused. In light of today’s very different challenges, I doubt

the Court would invent the same rule.

As the case has subsequently been interpreted, it allows

the press to cast false aspersions on public figures with near

7

It should be noted that precisely what should have been

done is a matter of debate. See, e.g., Richard A. Epstein, Was New

York Times v. Sullivan Wrong, 53 U. CHI. L. REV. 782, 791 (1986);

see also Lewis Green, The New York Times Rule: Judicial Overkill

12 VILLANOVA L. REV. 725, 735 (1967).

20

impunity.8 It would be one thing if this were a two-sided

phenomenon. Cf. New York Times, 376 U.S. at 305 (Goldberg,

J., concurring) (reasoning that the press will publish the

responses of public officials to reports or accusations). But see

Suzanne Garment, The Culture of Mistrust in American Politics

74–75, 81–82 (1992) (noting that the press more often

manufactures scandals involving political conservatives). The

increased power of the press is so dangerous today because we

are very close to one-party control of these institutions. Our

court was once concerned about the institutional consolidation

of the press leading to a “bland and homogenous” marketplace

of ideas. See Hale v. FCC, 425 F.2d 556, 562 (D.C. Cir. 1970)

(Tamm, J., concurring). It turns out that ideological

consolidation of the press (helped along by economic

consolidation) is the far greater threat.9

8

See Dun & Bradstreet, Inc. v. Greenmoss Builders, Inc.,

472 U.S. 749, 769 (1985) (White, J., concurring):

The New York Times rule thus countenances two

evils: first, the stream of information about

public officials and public affairs is polluted and

often remains polluted by false information; and

second, the reputation and professional life of the

defeated plaintiff may be destroyed by

falsehoods that might have been avoided with a

reasonable effort to investigate the facts. In

terms of the First Amendment and reputational

interests at stake, these seem grossly perverse

results.

9

We once explained why major American cities lost their

second mainframe papers due to market forces. See generally

Michigan Citizens for an Indep. Press v. Thornburgh, 868 F.2d 1285,

1288 (D.C. Cir.), aff’d, 493 U.S. 38 (1989). That second paper was

sometimes right of center, e.g., The New York Herald Tribune and

The Washington Star, leaving the residual paper in a local monopoly

position. As large American cities became heavily Democratic Party

21

Although the bias against the Republican Party—not

just controversial individuals—is rather shocking today, this is

not new; it is a long-term, secular trend going back at least to

the ’70s.10 (I do not mean to defend or criticize the behavior of

any particular politician). Two of the three most influential

papers (at least historically), The New York Times and The

Washington Post, are virtually Democratic Party broadsheets.

And the news section of The Wall Street Journal leans in the

same direction. The orientation of these three papers is

followed by The Associated Press and most large papers across

the country (such as the Los Angeles Times, Miami Herald, and

Boston Globe). Nearly all television—network and cable—is

a Democratic Party trumpet. Even the government-supported

National Public Radio follows along.

As has become apparent, Silicon Valley also has an

enormous influence over the distribution of news. And it

similarly filters news delivery in ways favorable to the

Democratic Party. See Kaitlyn Tiffany, Twitter Goofed It, The

Atlantic (2020) (“Within a few hours, Facebook announced that

it would limit [a New York Post] story’s spread on its platform

while its third-party fact-checkers somehow investigated the

information. Soon after, Twitter took an even more dramatic

bastions, so too did the local dominant paper. See Gentzkow and

Shapiro, What Drives Media Slant? Evidence from U.S. Daily

Newspapers, 78 ECONOMETRICA 35 (Jan. 2010).

10

Who can forget Candy Crowley’s debate moderation? See,

e.g., Noah Rothman, Candy Crowley’s Debate Moderation

Exemplifies Why Americans Do Not Trust Their Media, Mediaite

(Oct. 17, 2012); Dylan Byers, Crowley fact-checks Mitt, Politico

(Oct. 17, 2012).

22

stance: Without immediate public explanation, it completely

banned users from posting the link to the story.”). 11

It is well-accepted that viewpoint discrimination “raises

the specter that the Government may effectively drive certain

ideas or viewpoints from the marketplace.” R.A.V. v. City of St.

Paul, Minn., 505 U.S. 377, 387 (1992). But ideological

homogeneity in the media—or in the channels of information

distribution—risks repressing certain ideas from the public

consciousness just as surely as if access were restricted by the

government.

To be sure, there are a few notable exceptions to

Democratic Party ideological control: Fox News, The New

York Post, and The Wall Street Journal’s editorial page.12 It

should be sobering for those concerned about news bias that

these institutions are controlled by a single man and his son.

Will a lone holdout remain in what is otherwise a frighteningly

orthodox media culture? After all, there are serious efforts to

muzzle Fox News. And although upstart (mainly online)

conservative networks have emerged in recent years, their

11

Of course, I do not take a position on the legality of big

tech’s behavior. Some emphasize these companies are private and

therefore not subject to the First Amendment. Yet—even if correct—

it is not an adequate excuse for big tech’s bias. The First Amendment

is more than just a legal provision: It embodies the most important

value of American Democracy. Repression of political speech by

large institutions with market power therefore is—I say this

advisedly—fundamentally un-American. As one who lived through

the McCarthy era, it is hard to fathom how honorable men and

women can support such actions. One would hope that someone, in

any institution, would emulate Margaret Chase Smith.

12

Admittedly, a number of Fox’s commentators lean as far

to the right as the commentators and reporters of the mainstream

outlets lean to the left.

23

visibility has been decidedly curtailed by Social Media, either

by direct bans or content-based censorship.

There can be little question that the overwhelming

uniformity of news bias in the United States has an enormous

political impact.13 That was empirically and persuasively

demonstrated in Tim Groseclose’s insightful book, Left Turn:

How Liberal Media Bias Distorts the American Mind (2011).

Professor Groseclose showed that media bias is significantly to

the left. Id. at 192–197; see also id. at 169–77. And this

distorted market has the effect, according to Groseclose, of

aiding Democratic Party candidates by 8–10% in the typical

election. Id. at ix, 201–33. And now, a decade after this book’s

publication, the press and media do not even pretend to be

neutral news services.

It should be borne in mind that the first step taken by

any potential authoritarian or dictatorial regime is to gain

control of communications, particularly the delivery of news.

It is fair to conclude, therefore, that one-party control of the

press and media is a threat to a viable democracy. It may even

give rise to countervailing extremism. The First Amendment

guarantees a free press to foster a vibrant trade in ideas. But a

biased press can distort the marketplace. And when the media

has proven its willingness—if not eagerness—to so distort, it is

a profound mistake to stand by unjustified legal rules that serve

only to enhance the press’ power.

13

The reasons for press bias are too complicated to address

here. But they surely relate to bias at academic institutions.