Opinion for the court filed by Circuit Judge STARR and Senior Circuit Judge J. SKELLY WRIGHT.

Opinion concurring in the judgment filed by Chief Judge WALD.

Concurring opinion filed by Circuit Judge RUTH BADER GINSBURG.

Dissenting opinion filed by Senior Circuit Judge MacKINNON.

STARR, Circuit Judge and J. SKELLY WRIGHT, Senior Circuit Judge:

William Tavoulareas and his son Peter brought suit for injury to reputation after The Washington Post published a story that said, among other things, that Tavoulareas had used his influence as president of Mobil Corporation to "set up" Peter as a partner in a shipping firm whose business included a multi-million dollar management services contract with Mobil. After a jury trial in the United States District Court for the District of Columbia, Judge Oliver Gasch awarded judgment notwithstanding the verdict to the Post defendants. 567 F.Supp. 651 (D.D.C.1983). A divided panel of this court reinstated the jury's verdict, 759 F.2d 90 (D.C.Cir.1985), but the full court vacated that portion of the panel opinion and set the case for rehearing en banc, 763 F.2d 1472, 1481 (D.C.Cir.1985).

After a careful review of the entire record in the light most favorable to plaintiff, we are convinced that the only reasonable inference to be drawn is that the "set up" allegation was substantially true. We further hold that Tavoulareas is a limited purpose public figure who can recover for defamation only upon clear and convincing proof that defendants acted with actual malice. Because insufficient evidence exists in the record to support a finding of constitutional malice with respect to any of the defendants, we affirm entirely the District Court's decision.

At trial, the parties presented conflicting evidence concerning Mobil's and Tavoulareas' involvement with Peter's shipping firm. The following account adopts the undisputed facts and Tavoulareas' version of disputed events.

William Tavoulareas was at all relevant times president and chief operating officer of Mobil Corporation, the Nation's second largest oil company and its third largest industrial corporation. In his position at Mobil, Tavoulareas took an active role in the public debate during the 1970's over the manner in which the United States should respond to the rise of OPEC and the ensuing energy crisis. In particular, Tavoulareas vigorously defended the performance of the oil industry against critics who called for sweeping reforms in the structure and management of the industry, and he publicly advocated less governmental regulation of the industry as the solution to the energy shortage. In addition, he was an important proponent of Mobil's "Saudi strategy" of dependence on Arab oil supplies at a time when increasing American energy independence became a significant public policy objective.

In 1973, a group of influential Saudis approached Mobil with a plan for a jointly owned shipping company. The group included the Alirezas, a prominent merchant family in Saudi Arabia that had other business relationships with Mobil.1 The plan called for Mobil to furnish the capital for the proposed company, which would transport Saudi crude in its ships. At that time, it was widely anticipated that Saudi Arabia would soon establish a preference system requiring that Saudi crude be shipped in Saudi vessels. If Mobil, which depended heavily on Saudi supplies, were forced to use Saudi-owned ships, much of Mobil's own large fleet would then be idled. Mobil initially rejected the offer, whereupon the Saudis found another partner, Fairfield-Maxwell. The resulting venture was called Samarco--the Saudi Maritime Company. Mobil reassesed its situation in early 1974, reversed its field, and decided to join the Saudi venture.

Under the Samarco arrangement, ships owned by Mobil would be "bareboat-chartered" to Samarco--that is, leased without crews or provisions. Samarco would then "time-charter" the ships back to Mobil with full crews, supplies, and fuel.

Mobil decided that Samarco's ships should be operated by an independent management company, Atlas Maritime Company. Mobil believed that Atlas could operate the ships at less cost than Mobil2 and that this arrangement would avoid a conflict of interest among the Samarco partners.

Atlas was established by George Comnas in 1974. Comnas met with Tavoulareas in January 1974 and explained that he had just started his own business after leaving his position as managing director of C.M. Lemos & Co., one of the largest Greek shipping concerns. One of Comnas' assistants at Lemos was Tavoluareas' son Peter. Peter, 24 and fresh from business school, was working at his first job in the shipping business as a $14,000 per year employee.3 Comnas informed Tavoulareas at the January meeting that he, Comnas, wanted Peter to join his new venture as a principal and that Comnas would like to explore the possibility of doing business with Mobil. Tavoulareas then informed Rawleigh Warner, chairman of Mobil's board of directors, and George Birrell, chairman of Mobil's Conflicts of Interest Committee, that Peter might join Comnas in performing services for Samarco. Soon thereafter, sometime in the spring of 1974, Tavoulareas personally recruited Comnas to manage Samarco's ships through Atlas.

In August 1974, Peter left Lemos to become an equity partner at Atlas. Ares Emmanuel, a more experienced but similarly youthful co-worker from Lemos, had also joined Atlas but was provided with a much smaller equity interest in the firm.4 When Peter joined Atlas, Tavoulareas sent a memorandum to Paul Wolfe, executive vice-president of Mobil, stating that he "would no longer be involved with anything as to Atlas and Samarco." Record Excepts (RE) at 2440; see RE at 2339. Mobil later told the Post that "[f]rom the date Peter Tavoulareas joined Atlas, Mr. Tavoulareas divorced himself from involvement in matters involving business transactions between Mobil and/or SAMARCO with Atlas" and that "[t]his is what Mr. Warner reported to [Mobil's] Board." RE at 2344.5 The undisputed record, however, reveals that Tavoulareas involved himself in Samarco-Atlas matters on numerous occasions after Peter joined Atlas. For example, at trial Tavoulareas admitted that he attended and participated in meetings in Geneva in August 1974 and Saudi Arabia in November 1974 at which substantive discussions took place that helped produce the final agreement between Atlas and Samarco. Tavoulareas conceded that in these meetings he was "representing George Comnas' [and hence Peter's] position," Transcript (Tr.) at 1712, urging the Saudis to accept terms sought by Atlas.

As its inaugural project, Atlas began operating two Mobil-owned ships under contract with Samarco at an annual fee of more than $600,000, with the prospect of additional ships in the future. No other bids were solicited or received for the ship-management contract. By the time the Post published its story, Atlas had received more than $4.5 million in management fees from Samarco.

Shortly after Atlas began operations, Mobil grew disenchanted with Comnas, although there is disagreement among Tavoulareas' witnesses about whether the dissatisfaction arose from Comnas' business performance or some alleged misconduct on his part. What is undisputed is that Tavoulareas participated in a meeting, held in Tavoulareas' office, at which senior Mobil executives decided to seek Comnas' removal from Atlas. It is also undisputed that Tavoulareas and two other Mobil shipping executives flew to London to notify Comnas of this decision. Comnas was offered and accepted a $30,000 a year, three-year consultancy arrangement with Mobil in return for his resignation from Atlas.

When notified by Mobil of Comnas' resignation, the Saudi partners in Samarco, who had never been enthusiastic about using an independent management company, expressed promptly their view that the Samarco-Atlas management contract was terminated. Tavoulareas personally urged the Samarco partners to retain Atlas and successfully resisted the Saudi partners' attempts to take over some of the departed Comnas' equity interest. Most of that equity soon went to Peter.

After Comnas resigned, Atlas was left without an experienced hand at the helm. Harmon Hoffmann, a senior and highly respected Mobil executive, took over management of Atlas for the next six months, assisting Peter and the latter's fellow Lemos alumnus, Ares Emmanuel. Peter, whose share of Atlas increased from 40 to 75 percent in the wake of Comnas' departure, then assumed the management duties along with Emmanuel at what had become Peter's shipping firm.6

In November 1976, more than two years after Peter joined Atlas and over a year after Peter became its principal owner, Tavoulareas and Warner decided to send a letter to Mobil's quarter-million stockholders informing them that Tavoulareas had not been "actively involved in the planning, negotiation and direction" of Atlas and did "not participate in any decisions regarding the [Samarco-Atlas] relationship because his son, Peter, [was] one of the principals of that marine management firm." RE at 2650. Although Tavoulareas had formally recused himself from such matters, several of Mobil's outside directors raised objections to Peter's involvement in Atlas. Tavoulareas' potential conflict of interest also attracted the attention of the Securities and Exchange Commission, which conducted an investigation into the Mobil-Samarco-Atlas arrangements but took no enforcement action. The House Subcommittee on Energy and Power and several well-known journalists also investigated the matter.

On November 30, 1979, the Post published a frontpage story stating that

Mobil Oil Corp. president William P. Tavoulareas set up his son five years ago as a partner in a London-based shipping management firm that has since done millions of dollars in business operating Mobil-owned ships under exclusive, no-bid contracts.

The story, which is set forth in full in the Appendix, went on for eighty-five paragraphs to describe the Mobil-Samarco-Atlas arrangements in detail. The article stated, among other things, that Mobil's board of directors had been assured that Tavoulareas "was not involved in his son's venture in any way," but that unidentified sources said that Tavoulareas had nonetheless (1) recruited the shipping executive who set up Atlas; (2) helped negotiate the Samarco-Atlas contract; (3) "personally urged" Comnas to accept Peter as a partner; (4) "played a personal role" in Comnas' resignation; and (5) "dispatched one of his senior shipping executives, Herman [sic] F. Hoffmann, to London to help run Atlas" after Comnas' departure. At key points, the story quoted Tavoulareas' own account of the incidents and included his denials of disputed assertions; in all, more than thirty paragraphs of the article reported Mobil's version of the events in question.

The Post first became aware of the connection between Atlas and Mobil in 1976 when reporter Robert Woodward received an anonymous note describing the Samarco arrangement. Woodward made some preliminary inquiries, but for a variety of reasons put the matter aside. In 1979, during a period of rising fuel prices, gasoline lines, and increased interest in the Nation's energy situation, Woodward, by then metropolitan editor of the Post, revived the Atlas matter. He assigned it to Patrick Tyler, a relatively new reporter on the metropolitan staff who had written a few months earlier about the oil shortage (and in a manner which prompted Mobil's disapprobation).

A short time later, Sandy Golden, a reporter for a suburban newspaper who was ambitious for a job on the Post, telephoned Woodward to offer a lead he hoped to develop into a story about how the president of Mobil set up his son to be an "overnight millionaire." Upon returning Golden's call on Woodward's behalf, Tyler learned that Golden had encountered Dr. Philip Piro, a young Johns Hopkins physician and estranged son-in-law of Tavoulareas, and that Piro had offered to discuss Tavoulareas' involvement with Peter's rise in the business world. Tyler and Golden met with Piro at The Owl restaurant in Baltimore and interviewed him for several hours. Tyler concluded that Piro had little direct knowledge of Mobil's business affairs, and that he obviously harbored a grudge against the Tavoulareas family. In view of Piro's perceived limited value as a source, Tyler informed Golden that he would continue to pursue the story on his own; Golden said he would do likewise.7

Tyler's research led him to George Comnas, who told the reporter that Tavoulareas recruited him to help establish Samarco and Atlas. Comnas also stated that Tavoulareas requested him to bring Peter into Atlas. Comnas described the Samarco-Atlas arrangements in detail and discussed Tavoulareas' role in persuading the Saudi partners to accept Atlas as Samarco's management company. Tavoulareas, according to Comnas, personally asked him to resign from Atlas. Following his own interviews, Tyler learned that Comnas had related substantially the same account to investigators from the House Subcommittee on Energy and Power, who had questioned Comnas in detail and concluded that his story was accurate.

Tyler also interviewed John Kousi, a New York lawyer who served as Fairfield-Maxwell's representative on the Samarco board of directors. Kousi told Tyler that Peter had limited experience and ability, and that, in Kousi's opinion, Peter's partnership in Atlas was "a nepotistic act" on the part of Tavoulareas. Kousi also said that Tavoulareas personally negotiated on Atlas' behalf to obtain the agreement of Samarco's Saudi partners to the Atlas arrangement. Kousi further stated that Mobil was the moving force in both hiring and firing Comnas.

In preparing the article, Tyler sought repeatedly but unsuccessfully to interview Tavoulareas and other Mobil officials. At trial, Mobil officials explained the company's refusal to cooperate with Tyler on the ground that in his story earlier in 1979 Tyler had misrepresented statements by Mobil executives. Eventually, however, at Woodward's urging, Mobil agreed to answer written questions from Tyler. Those answers confirmed the basic outlines of the Samarco arrangement. The responses further stated that, although Mobil's Board of Directors was told that "Tavoulareas divorced himself from involvement in matters involving business transactions between Mobil and/or SAMARCO with Atlas," he had in fact been involved in hiring Comnas and played what Mobil called "a minor role" in firing him. RE at 2344-45. Moreover, Mobil confirmed that Tavoulareas had participated in setting up Atlas as an independent management firm, but stated that he had not initiated Peter's association with Atlas.

Tyler also obtained a transcript of Tavoulareas' sworn testimony before the SEC's Enforcement Division in 1977. In that testimony, Tavoulareas acknowledged his key role in the creation of Atlas. He admitted that he knew that Comnas and Peter had discussed going into business together when he, Tavoulareas, recruited Comnas. This transcript, together with the interviews of Comnas and Kousi, the report of the Congressional investigators, and the written responses from Mobil, comprised the Post's principal sources for its story of November 30, 1979.

Tyler spent one month working on the story, tracking down leads, confirming information, and writing the final article. The story underwent scrutiny from senior editors, including Woodward and national news editor William Greider. The editors questioned Tyler about his sources. Counsel for the Post conducted a line-by-line review. Managing editor Howard Simons and executive editor Benjamin Bradlee examined the story before it was cleared for publication. When a copy editor, Christine Peterson, raised questions about the story, Tyler responded with a detailed memorandum that satisfied the Post's editors that the story was accurate. The Post vigorously contends to this day that it published the story in the belief that it was truthful; at oral argument on rehearing en banc, the Post reaffirmed its continued belief in the story's accuracy.

William and Peter Tavoulareas brought libel actions against the Washington Post Company, Benjamin Bradlee, Robert Woodward, Patrick Tyler, and Sandy Golden.8 Plaintiffs claimed they were falsely defamed both by the November 30 article and by a follow-up Post story on December 1. The suit against the Post was consolidated with a separately filed action brought against Dr. Piro for slander. The District Court subsequently determined that William Tavoulareas was a limited public figure for purposes for the case but that Peter was not. RE at 696-97, 709-29.

On July 30, 1982, a jury found that Bradlee and Woodward were not liable to either of the plaintiffs and that the Post, Tyler, and Golden were not liable to William Tavoulareas for the December 1 article and not liable to Peter for either article. The jury did, however, return a general verdict against the Post, Tyler, and Golden for the November 30 article, awarding William Tavoulareas $250,000 in compensatory and $1.8 million in punitive damages. The jury also found Piro liable to both Tavoulareases for slander.

The District Court subsequently entered judgment notwithstanding the verdict (j.n.o.v.) for Piro and for the Post defendants against William Tavoulareas. Appeals were then brought to this court challenging the judgments n.o.v. and attacking the trial court's determination that William Tavoulareas was a limited public figure.

In New York Times Co. v. Sullivan, 376 U.S. 254, 84 S.Ct. 710, 11 L.Ed.2d 686 (1964), the Supreme Court first recognized that traditional actions for defamation might interfere with First Amendment rights of free expression. Discerning in the First Amendment a demand that writers and speakers enjoy enough "breathing space" to avoid self-censorship and encourage "debate on public issues [that is] uninhibited, robust, and wide open," id. at 270, 84 S.Ct. at 721, the Court held that a public official could recover damages for libel only by showing that the allegedly defamatory statement was made with " 'actual malice'--that is, with knowledge that it was false or with reckless disregard of whether it was false or not." Id. at 279-80, 84 S.Ct. at 726. In Curtis Publishing Co. v. Butts, 388 U.S. 130, 87 S.Ct. 1975, 18 L.Ed.2d 1094 (1967), this constitutional protection was applied to speech concerning "public figures" who were not government officials, but who nonetheless "often play an influential role in ordering society." Id. at 164, 87 S.Ct. at 1996 (Warren, C.J., concurring). But in Gertz v. Robert Welch, Inc., 418 U.S. 323, 94 S.Ct. 2997, 41 L.Ed.2d 789 (1974), the Court declined to extend the actual malice rule to speech about private individuals.

Gertz held that the extent of the constitutional privilege varies with the strength of the government's legitimate interest in protecting individual reputation. This interest is strongest when the state protects private citizens whose actions reflect a decision to shield their lives from public scrutiny. Such persons are typically "more vulnerable to injury than public officials and public figures [and] more deserving of recovery." Id. at 345, 94 S.Ct. at 3010. The state's interest is correspondingly weaker with respect to government officials and others who "assume[ ] [a] role of especial prominence in the affairs of society ... [that] invite[s] attention and comment." Id. This is true partly because public figures "usually enjoy significantly greater access to the channels of effective communication and hence have a more realistic opportunity to counteract false statements than private individuals normally enjoy." Id. at 344, 94 S.Ct. at 3009. More important, "by reason of the notoriety of their achievements or the vigor and success with which they seek the public's attention," id at 342, 94 S.Ct. at 3008, public figures voluntarily "run[ ] the risk of closer public scrutiny than might otherwise be the case." Id. at 344, 94 S.Ct. at 3009. Moreover, "[o]ur citizenry has a legitimate and substantial interest in the conduct of such persons, and freedom of the press to engage in uninhibited debate about their involvement in public issues and events is as crucial as it is in the case of 'public officials.' " Butts, 388 U.S. at 164, 87 S.Ct. at 1996. (Warren, C.J., concurring).

The Supreme Court has identified two classes of public figures in addition to government officials: general purpose and limited purpose public figures. "In some instances, an individual may achieve such pervasive fame or notoriety that he becomes a public figure for all purposes and in all contexts. More commonly, an individual voluntarily injects himself or is drawn into a particular public controversy and thereby becomes a public figure for a limited range of issues." Gertz, 418 U.S. at 351, 94 S.Ct. at 3013. The extent of the individual's participation in public affairs and assumption of the risk of adverse publicity determines the weight of the government's interest in protecting his or her reputation.

Whether (and to what extent) a person is a public figure is a matter of law for the court to decide. Rosenblatt v. Baer, 383 U.S. 75, 88, 86 S.Ct. 669, 677, 15 L.Ed.2d 597 (1966); Waldbaum v. Fairchild Publications, Inc., 627 F.2d 1287, 1293 n. 12 (D.C.Cir.), cert. denied, 449 U.S. 898, 101 S.Ct. 266, 66 L.Ed.2d 128 (1980). This is a difficult and sensitive exercise unsusceptible to the application of rigid or mechanical rules. See, e.g., Rosanova v. Playboy Enterprises, Inc., 411 F.Supp. 440, 443 (S.D.Ga.1976) ("Defining public figures is much like trying to nail a jellyfish to the wall."), aff'd, 580 F.2d 859 (5th Cir.1978). In this circuit, our determination is guided by Waldbaum.

A person becomes a general purpose public figure only if he or she is "a well-known 'celebrity,' his name a 'household word.' " Waldbaum, 627 F.2d at 1294. Such persons have knowingly relinquished their anonymity in return for fame, fortune, or influence. They are frequently so famous that they "may be able to transfer their recognition and influence from one field to another." Id. at 1294 n. 15. Thus, it is reasonable to attribute a public character to all aspects of their lives. William Tavoulareas is a highly prominent individual, especially in business circles, but his celebrity in society at large does not approach that of a well-known athlete or entertainer--apparently the archetypes of the general purpose public figure. See, e.g., [Johnny] Carson v. Allied News Co., 529 F.2d 206 (7th Cir.1976); Chuy v. Philadelphia Eagles Football Club, 431 F.Supp. 254 (D.C.Pa.1977), aff'd, 595 F.2d 1265 (3d Cir.1979) (en banc). The standard as generally applied is a strict one; the Supreme Court has not found anyone to be a general public figure since Butts. In view of the stringent standards applicable to this category of public figure, we have no difficulty in upholding the District Court's conclusion that the plaintiff was not a general purpose public figure.

Although few persons attain the level of notoriety to be public figures in all contexts, many individuals may be public figures for the more limited purpose of certain issues or situations. Waldbaum sets out a three-step inquiry to identify these limited-purpose public figures. First, we isolate the controversy at issue, because the scope of the controversy in which the plaintiff involves himself defines the scope of the public personality. The controversy must be public both in the sense that "persons actually were discussing" it, 627 F.2d at 1297, and that "persons beyond the immediate participants in the dispute [are likely] to feel the impact of its resolution." Id. Second, we examine the plaintiff's role in the controversy, to be sure that it is more than "trivial or tangential." Id. An individual does not forfeit the full protection of the libel laws merely by stating a position on a controversial issue if he or she is not a principal participant in the debate or is unlikely to have much effect on its resolution. See Gertz, 418 U.S. at 351-52, 94 S.Ct. at 3012-13. Finally, we determine if the alleged defamation was germane to the plaintiff's participation in the controversy. Waldbaum, 627 F.2d at 1298.

The Waldbaum inquiry provides a uniform approach to identifying those individuals whose voluntary participation in public life and whose access to the media have reduced the state's interest in protecting them from the risk of defamation. Waldbaum thus provides us with useful analytic tools; nevertheless, the touchstone remains whether an individual has "assumed [a] role[ ] of especial prominence in the affairs of society ... [that] invite[s] attention and comment." Gertz, 418 U.S. at 345, 94 S.Ct. at 3009.

Applying these principles to the case at hand, we first consider the possibility that Tavoulareas was a limited public figure for the purpose of the controversy over whether the management and structure of the United States' private oil industry was in need of alteration or reform. Examining the oil shortages of the 1970's, numerous public officials, pundits, and commentators criticized both the performance and integrity of the major, integrated oil companies. Many reform proposals were publicly advanced and considered, including measures to break up or divest the large oil companies, increase their taxes, install government representatives on their boards of directors, and subject them to more intensive federal regulation.

We have held in Waldbaum that "[b]eing an executive within a prominent and influential company does not by itself make one a [limited purpose] public figure." 627 F.2d at 1299. We reaffirm that principle today. But that is not to say that an individual's position as president and chief operating officer of one of the world's largest multinational corporations, with a quarter-million stockholders, is irrelevant to whether that person has "invite[d] attention and comment" with respect to public issues affecting his business dealings. This is especially true when that industry--and the company itself--is at the center of a vigorous public debate touching on a vital national interest.

More specifically, Tavoulareas avowedly attempted to "thrust [Mobil and himself] to the forefront" of the national controversy over the state of the oil industry. In November 1979, for example, Tavoulareas made the following telling observations in a speech:

As you know, Mobil has gotten the reputation of being probably the most outspoken company on public issues, through our newspaper advocacy advertising and other public statements. It's not always comfortable being in the limelight, particularly when the President of the U.S. calls us perhaps the most irresponsible company in the country. But we think the effort has been worth making and we're going to go on with it. We see positive results in the press and in our meetings with elected officials.

RE at 302; cf. RE at 470. Even more than Eric Waldbaum (whom this court, speaking through Judge Tamm, found to be a limited public figure with respect to his business dealings), Tavoulareas was not "merely a boardroom president whose vision was limited to the balance sheet. He became an activist, projecting his own image and that of [Mobil] far beyond the dollars and cents aspects of marketing." 627 F.2d at 1300.

Mobil and Tavoulareas played substantial roles in spearheading a public counterattack on the movement for reform of the oil industry. The 500-page "Public Figure Index"--a collection of news clippings and the like, submitted by the Post--attests to the undisputed fact that Tavoulareas was outspoken in defending the oil industry's performance, see, e.g., RE at 455, in blaming the oil crisis on government regulation and interference with the free market, and in advocating rejection of efforts to further regulate or alter the oil industry. He made speeches, testified before Congress, RE at 304-67, published articles, RE at 378-97, and through Mobil's publicity apparatus enjoyed continuing access to the media, RE at 419-63. Even more than his counterparts in the industry, Tavoulareas sought and received public attention on the management and operations of the Nation's oil companies.9 Having "stepp[ed] into the public spotlight ... he must take the good with the bad." Waldbaum, 627 F.2d at 1294-95.

Public policy toward the oil industry was clearly a controversial subject that "was being debated publicly and ... had foreseeable and substantial ramifications for non-participants." Id. at 1297. Although only a "few [participants] can have the necessary impact," id. at 1297 n. 27, on such a broadly-defined controversy, Tavoulareas' widely-reported, influential public role in the debate as president of Mobil eminently qualified him as a public figure.

Waldbaum 's final requirement, namely germaneness of the publication to the controversy, was also satisfied by the Post article. The story of Mobil's president and his junior-executive son, who achieved great business success early in life, sought to provide the public with "a rare glimpse into corporate behavior at the top of one of the largest publicly held international oil companies." p 12. The alleged nepotism by Tavoulareas was not "wholly unrelated" to a public controversy where the credibility and integrity of representatives of the oil industry had become an issue. 627 F.2d at 1298; see, e.g., RE at 570. In our view, the Waldbaum criteria are abundantly satisfied here.

In addition, Judge Gasch found Tavoulareas to be a public figure for purposes of a more narrowly defined controversy concerning the Mobil-Samarco-Atlas arrangement. Tavoulareas acknowledges that the Post article was germane to that arrangement, but disputes whether it was a "public" controversy within the meaning of Gertz, Waldbaum, and Firestone.10 Like the District Court, we conclude that it was.

The involvement of Mobil, Tavoulareas, and Peter in the Samarco-Atlas arrangement attracted the attention of journalists and government officials long before the November 30, 1979 article appeared. Mobil publicly announced the Samarco-Atlas deal in 1974, albeit without mentioning Peter's involvement. RE at 629. After consultation with Tavoulareas, Mobil officials subsequently decided to publicize Peter's involvement in the arrangement by releasing the information to a "widely read oil industry periodical." Tr. at 1880-81. Thereafter, a number of journalists, including reporters from The New York Times and Jack Anderson's office, contacted Mobil about Samarco and Atlas. RE at 1420-22; see also RE at 2464; Tr. at 2435.11

In light of various inquiries, Tavoulareas, along with Mobil's Chairman Rawleigh Warner, decided to disclose in a letter to Mobil's shareholders that Peter was a principal in Atlas and that, as a result, Tavoulareas had formally recused himself from Samarco-Atlas matters. RE at 2440-42, 2650.12 In presenting his side of the story to Mobil's shareholders, Tavoulareas reached an audience of more than a quarter-million people, a group larger than the circulation of most daily newspapers.

Shortly after the Mobil shareholder letter, the SEC launched an investigation into Tavoulareas' relationship to Samarco and Atlas. Tavoulareas vigorously defended his conduct in sworn testimony before the SEC staff in March of 1977. RE at 2399-2460. Ten days prior to publication of the Post article, Congressman John Dingell, Chairman of the House Subcommittee on Energy and Power, wrote the Chairman of the SEC requesting the Commission to reopen its investigation on the basis of information unearthed by Congressional investigators indicating that "Tavoulareas may have been actively involved in the formation and operation of both Samarco and Atlas, notwithstanding the potential conflict of interest and his denials of participation." RE at 2466.

Not only were "persons actually ... discussing" the extent and propriety of Tavoulareas involvement in Samarco-Atlas matters prior to the Post article, Waldbaum, 627 F.2d at 1297, but "a reasonable person would have expected persons beyond the immediate participants in the dispute to feel the impact of its resolution," id. (footnote omitted). Highly respected outside members of Mobil's board of directors--including Ambassador George McGhee, Fred Borch, chairman of General Electric, and Albert Williams, president of IBM--specifically objected to the arrangements Mobil's management had made with respect to Peter, and had raised their concerns directly with Mr. Warner. Tr. at 3585-86; see also Tr. at 1814-15. Mobil's outside directors were worried that the appearance of a conflict of interest would hurt Mobil's public image and thereby have an adverse impact on Mobil's shareholders. GE's Borch, for example, "urged [Warner] to urge Tavoulareas Sr. to get out of his relationship with his son and put him in a different business. If he wanted to put him in a business, put him in one that does not affect Mobil...." Tr. at 3551. Ambassador McGhee opposed Peter's involvement in Atlas from the outset, expressing the view that the arrangement was "bad policy and precedent for the company" and would "expose [Tavoulareas and his son] to the risks of possible adverse criticism, even investigations by Congressional committees." Tr. at 1795-96 (emphasis added); see also RE at 2670. Having opened these matters for public scrutiny by going forward in the face of considered objections by leading businessmen serving on his own board, and then publicizing the affair to the press and to Mobil's shareholders, Tavoulareas could scarcely expect all ensuing commentary on these arrangements to be favorable.13

In sum, we find abundant evidence of the already "public" nature of this subject in the activities of both government and corporate officials. When, in Judge Tamm's words, we "look at the facts, taken as a whole, through the eyes of a reasonable person," Waldbaum, 627 F.2d at 1293, we conclude that Tavoulareas was a public figure for purposes of this publication, and that the Post was therefore "entitled to act on the assumption that [he] voluntarily exposed [himself] to increased risk" of critical comment and publicity. Gertz, 418 U.S. at 345, 94 S.Ct. at 3010.

It is now well settled that, in the face of the countervailing demands of the First Amendment, a public figure such as Tavoulareas may recover for injury to reputation "only on clear and convincing proof" that the defamatory falsehood was made with "actual malice." Gertz, 418 U.S. at 342, 94 S.Ct. at 3008. It is equally well established that the standard of actual malice requires proof not merely that the defamatory publication was false, but that the defendant either knew the statement to be false or that the defendant "in fact entertained serious doubts as to the truth of his publication." St. Amant v. Thompson, 390 U.S. 727, 731, 88 S.Ct. 1323, 1325, 20 L.Ed.2d 262 (1968); see also Philadelphia Newspapers, Inc. v. Hepps, 475 U.S. 767, 106 S.Ct. 1558, 1567, 89 L.Ed.2d 783 (1986) (Stevens, J., dissenting) ("For [the actual malice] standard to be met, the publisher must come close to wilfully blinding itself to the falsity of its utterance.") (footnote omitted). Moreover, the burden of proof imposed on public figures is significantly more onerous than the usual preponderance of the evidence standard. See, e.g., Long v. Arcell, 618 F.2d 1145, 1148 (5th Cir.1980), cert. denied, 449 U.S. 1083, 101 S.Ct. 869, 66 L.Ed.2d 808 (1981); Nader v. De Toledano, 408 A.2d 31, 49 (D.C.1979), cert. denied, 444 U.S. 1078, 100 S.Ct. 1028, 62 L.Ed.2d 761 (1980). The requirement of "clear and convincing" proof of actual malice, initially articulated by the Supreme Court in Sullivan, "administers an extremely powerful antidote to the inducement to media self-censorship of the common-law rule of strict liability for libel and slander." Gertz, 418 U.S. at 342, 94 S.Ct. at 3008.

On the basis of its careful review of the entire record, the District Court concluded that there was not "sufficient evidence in the record from which a jury could reasonably find, by clear and convincing proof, that the [Post] defendants published the November 30 article with actual malice." 567 F.Supp. at 653-64 (footnote omitted). In reaching this decision, Judge Gasch applied the traditional standard for overturning a jury verdict: "A trial court may grant [j.n.o.v.] only when 'the evidence, together with all inferences that can reasonably be drawn therefrom[,] is so one-sided [in favor of the moving party] that reasonable men could not disagree on the verdict.' " Id. at 652 (quoting Vander Zee v. Karabatsos, 589 F.2d 723, 726 (D.C.Cir.1978), cert. denied, 441 U.S. 962, 99 S.Ct. 2407, 60 L.Ed.2d 1066 (1979)).14 The District Court considered all the evidence in the light most favorable to Tavoulareas and abstained from assessing the credibility of witnesses or weighing the evidence. We follow the same course. Based upon our review of the entire record in this case, we agree with Judge Gasch's dispositive conclusion.

In reaching this conclusion, we apply a standard of review that all the parties, amici, and the District Court have agreed is correct. Specifically, by virtue of the First Amendment nature of this litigation, the jury verdict must be measured, on the basis of an independent examination, against the heavy burden of proof imposed on a plaintiff who is a public figure:

In reviewing a defamation verdict, courts must exercise particularly careful review. They "must 'make an independent examination of the whole record,' ... so as to assure [themselves] that the judgment does not constitute a forbidden intrusion on the field of free expression." New York Times v. Sullivan, 376 U.S. 254, 285, 84 S.Ct. 710, 729, 11 L.Ed.2d 686 (1964) (quoting Edwards v. South Carolina, 372 U.S. 229, 235, 83 S.Ct. 680, 683, 9 L.Ed.2d 697 (1963)).

Supplemental Brief for Appellant at 9; see also Supplemental Brief for Appellees at 6-7; Brief Amicus Curiae of the American Legal Foundation in Support of Appellant at 7; Brief of the Reporters Committee for Freedom of the Press and Radio Television News Directors Association as Amici Curiae in Support of Appellees at 22-23; 567 F.Supp. at 653-54.

The parties, at our instance, have devoted enormous effort to the difficult enterprise of discerning the precise meaning of Bose Corp. v. Consumers Union, 466 U.S. 485, 104 S.Ct. 1949, 80 L.Ed.2d 502 (1984) (mandating independent judicial review of evidence of actual malice, but leaving unresolved the extent of that review). The District Court was, of course, not faced with this inquiry; the decision granting j.n.o.v. antedated Bose by almost exactly one year. From Bose, the parties have argued sharply divergent positions. Under one view, Bose 's mandate of de novo review means precisely that, with no deference at all to be accorded any jury finding germane to actual malice. See Supplemental Brief for Appellees at 6-16. Under the contrary view, Bose does not alter the traditional rules governing the review of jury verdicts and thus judicial deference is constitutionally mandated to presumed jury findings of underlying facts, evaluations of credibility, and the drawing of inferences. See Supplemental Brief for Appellant at 9-19. Notwithstanding these divergent positions, however, it is undisputed that Bose at least reaffirmed Sullivan 's mandated duty of independent review. See, e.g., Supplemental Brief for Appellant at 9 ("In Bose ... the Court reaffirmed that 'Judges, as expositors of the Constitution, must independently decide whether the evidence in the record is sufficient to cross the constitutional threshold that bars the entry of any judgment that is not supported by clear and convincing proof of "actual malice." ' ") (quoting Bose, 466 U.S. at 511, 104 S.Ct. at 1965).

The somewhat murky dispute that has consumed this litigation in its later phases is whether Bose went further and sanctioned independent review as to findings of underlying facts, evaluations of credibility, and the drawing of inferences. It is this issue--the existence and application of this possible additional aspect of Bose --that we believe, upon reflection, we need not decide. We have concluded that sailing into these uncharted waters is unnecessary to the proper and principled disposition of this case. Bearing in mind the exacting standard of clear and convincing proof of actual malice, we are persuaded that, when we faithfully apply the traditional j.n.o.v. standards in this First Amendment context, Judge Gasch was manifestly correct.

The dissent complains that we have evaded this complex constitutional issue only by "reject[ing] entirely plausible (if not inevitable) interpretations of the article." Dissent at 810. But in determining what defamatory meaning the article is capable of bearing, we perform the quintessential function of the court in defamation actions, as the dissent forthrightly acknowledges. See Dissent at 817 ("It is for the court to determine whether a particular communication is capable of bearing a defamatory meaning.") (emphasis in original).

The dissent vehemently insists that we abandon the role of the court and invade the jury's function in upholding the District Court's grant of j.n.o.v. The dissent accuses us of "discard[ing] those traditional [j.n.o.v.] standards," by "proceed[ing] to make [our] own assessment of witness credibility" and "disregard[ing] controlling reasonable inferences in favor of the verdict." Dissent at 810-11. These accusations are repeated sporadically throughout our dissenting colleague's opinion. But they are ultimately without foundation. We are satisfied that the District Court did not trench on the jury's function, and we have been mindful ourselves to avoid any such usurpation.

The only example that the dissent presents of our alleged "reassessment" of witness credibility is in our determination that the "set up" allegation is substantially true. See infra pp. 783-86. But this allegation fails to recognize that the fact that Tavoulareas "set up" Peter in Atlas--the defamatory sense that we attribute to the article as a matter of law--is clear beyond reasonable dispute. Indeed, as will become evident, we rely on Tavoulareas' own testimony and on facts undisputed by him to make this finding.

The dissent's final grievance--that we disregard reasonable inferences drawn by the jury--is also echoed in Chief Judge Wald's concurrence, which notes that "throughout its opinion [the majority] scrutinizes the inferences to be drawn from evidence of actual malice far more rigorously than [the] ordinary judgment n.o.v. standard would permit." Concurring Opinion at 804. Chief Judge Wald suggests that we tackle the knotty constitutional issue regarding what constitutes independent review under Bose, so we could properly reach the conclusion as to the absence of actual malice. Both the dissent and the concurrence, with all respect, misapprehend the inquiry we undertake. We consistently grant to Tavoulareas all reasonable inferences that could be drawn from the evidence, including all issues of witness credibility. As all agree, long before Bose and without the benefit of any construction of its unelaborated standard, reviewing courts carefully and independently reviewed findings of fact and rejected jury verdicts when all favorable inferences did not constitute clear and convincing evidence of actual malice. See Edwards, 372 U.S. at 235, 83 S.Ct. at 683 (court must "make an independent examination of the whole record" in First Amendment cases); Sullivan, 376 U.S. at 285, 84 S.Ct. at 728-29 (same, citing Edwards ); St. Amant, 390 U.S. 727, 88 S.Ct. 1323 (finding insufficient evidence of actual malice to support jury verdict). That traditional undertaking in defamation jurisprudence is precisely the task before us today.

Accordingly, we have at every turn accepted only undisputed facts or Tavoulareas' version of disputed facts, avoided any evaluations of credibility, and credited all permissible inferences the jury may have drawn favorably to Tavoulareas. Having done so, we nonetheless are constrained to conclude that the evidence is insufficient to constitute clear and convincing evidence of actual malice.

The theme of the November 30, 1979 article was set forth in its headline and lead sentence: that William Tavoulareas "set up" Peter as a partner in a business managing Mobil-owned ships. Peter, the article reported, had in 1974 been a 24-year-old, $14,000 per year shipping clerk with the Lemos firm, but "with the help of Mobil" five years later owned 45 percent of a newly formed shipping management firm, Atlas Maritime Co. "[T]he overall thrust of the article was that William Tavoulareas improperly set up his son in business and made sure the business would prosper." Reply Brief for Appellant at 21-22; see Brief for Appellant at 70; 567 F.Supp. at 660.

Tavoulareas advances several possible defamatory interpretations of this "set up" allegation.15 His primary contention, and one adopted by the dissent, Dissent at 820-21, is that "[t]he November 30 article was reasonably understood to mean that plaintiff misused Mobil assets by engineering the entire Mobil-Samarco-Atlas arrangement for the benefit of his son, to the detriment of Mobil shareholders."16 Brief for Appellant at 26 (emphasis added). The District Court squarely rejected as untenable that construction of the article. 567 F.Supp. at 660. We agree.17

In a libel case, it is the role of the court to determine whether the challenged statement is "capable of bearing a particular meaning" and whether "that meaning is defamatory." Restatement (Second) Of Torts Sec. 614(i), at 311 (1977); see McBride v. Merrell Dow and Pharmaceuticals, Inc., 717 F.2d 1460, 1463 (D.C.Cir.1983). In making this determination, a court is to consider both the words themselves and the entire context in which the statement occurs. See Ollman v. Evans, 750 F.2d 970, 982-83 (D.C.Cir.1984) (en banc), cert. denied, 471 U.S. 1127, 105 S.Ct. 2662, 86 L.Ed.2d 278 (1985). The jury's proper function, in turn, is to determine whether a statement, held by the court to be capable of a defamatory meaning, was in fact attributed such a meaning by its readers. Restatement (Second) Of Torts Sec. 614(2).

The statement that "a father set up his son in business" would ordinarily mean to a reasonable reader that the father provided the son with the means or opportunity by which the latter could assume a position of responsibility in a business venture or commercial firm. See Webster's Third New International Dictionary 2079 (unabridged ed. 1981). In our view, when the term "set up" is employed in a familial context, it implies that one family member provided an opportunity to another family member on the basis of kinship, not merit. Accordingly, we hold that the article, as a matter of law, can reasonably be interpreted as capable of bearing a defamatory meaning, namely that Tavoulareas, as president of Mobil, made it possible for Peter to become a partner in Atlas and then helped to ensure that the business would prosper because Peter was his son. This, in our view, is the normal, everyday reading of the article. The headline and lead sentence, generally reliable indicators of an article's content, clearly convey this meaning.18 This meaning is defamatory in that it accuses Tavoulareas of nepotism--furthering his son's business career--which might "tend[ ] to injure [him] in his trade, profession or community standing, or lower him in the estimation of the community." Afro-American Publishing Co. v. Jaffe, 366 F.2d 649, 654 (D.C.Cir.1966) (footnote omitted).

As we have observed, Tavoulareas' proffered interpretation is much broader. He (and the dissent) contend that the article is capable of bearing the interpretation that Tavoulareas "set up" the entire Mobil-Sarmarco-Atlas relationship to benefit Peter. Thus, before passing to a consideration of the falsity vel non of the "set up" allegation, we first consider this asserted, much different, broader reading. For the reasons that follow, we reject Tavoulareas' construction and interpret the language according to its common usage. As did Judge Gasch, we hold as a matter of law that the article is incapable of bearing the interpretation Tavoulareas advances.19

As the District Court correctly observed, the Post 's allegation that Tavoulareas "set up" his son in Atlas is entirely different from the claim that the Post asserted that the creation of the entire Mobil-Samarco-Atlas relationship was a nepotistic act. 567 F.Supp. at 660 & n. 16. The article clearly says the former, not the latter. The article simply will not reasonably bear Tavoulareas' interpretation, as evidenced by his failure to cite anything in the article itself to support this reading. See Brief for Appellant at 25-29; Reply Brief for Appellant at 16.20 Indeed, the article discussed at length Mobil's legitimate business reasons for participating in Samarco, namely its anticipation of Saudi flag preference requirements and favorable Saudi financing. paragraphs 38-44, 51.

The dissent ignores the specifics of the article, complaining instead that "[t]he entire discussion of the legitimate business reasons in the Post article is perjorative." Dissent at 821. In our view, the dissent's reliance on the "tone" of the article is entirely misplaced. The article expressly buttressed the credibility of Mobil's stated business reasons for joining Samarco. The story specifically reports that the other non-Saudi partner in Samarco, Fairfield-Maxwell, joined on the basis of the same "anticipated benefits." p 43. Needless to say, Fairfield-Maxwell had no interest in participating in Samarco to benefit Peter. Rather, as the article itself reported, Fairfield-Maxwell was of the view that Peter's involvement in Atlas, although nepotistic, would not preclude the successful and profitable operation of Samarco. p 27. We cannot accept the dissent's tortured attempt to discern some dark, hidden meaning in the article when the plain words of the piece explicitly rebut that meaning.21

Tavoulareas relies heavily on internal unpublished Post memoranda to establish the meaning of the article. This will not do. Nothing in law or common sense supports saddling a libel defendant with civil liability for a defamatory implication nowhere to be found in the published article itself.22 In addition to the internal memoranda, the dissent cites two other items of extrinsic evidence as relevant to a determination of the article's meaning.23 First, much is made of the "public interpretation" of the Post article. Dissent at 819 n. 25. Assuming arguendo that such evidence bears on our legal determination, the items earmarked by the dissent are utterly incapable of supporting Tavoulareas' interpretation; if anything, these items support our interpretation of the article.24 Second, the dissent pounces upon a single passage from the Post 's closing argument at trial and attempts by that maneuver to characterize the Post as in fact accepting Tavoulareas' interpretation of the article. See Dissent at 822-23.25 This is grasping at straws. From the time the article was published to its appearance before this court en banc, the Post has steadfastly interpreted the article as saying that Peter was "set up" in Atlas, not that the entire Mobil-Samarco-Atlas arrangement was set up for Peter.26 This is hardly surprising, for as we have held as a matter of law, the article is incapable of bearing Tavoulareas' and the dissent's distorted interpretation. Because the piece is capable, however, of bearing the narrower but nonetheless defamatory meaning that Tavoulareas "set up" his son in Atlas, we proceed to consider the truth or falsity of this allegation.

Tavoulareas, it should be noted, would pretermit entirely our consideration of truth or falsity, maintaining that the court should not consider the question in reviewing the jury verdict in his favor. Brief for Appellant at 17a-17b. We emphatically disagree.27 Tavoulareas was required to prove falsity at trial in order to prevail. Philadelphia Newspapers, 106 S.Ct. at 1563; Garrison v. Louisiana, 379 U.S. 64, 74, 85 S.Ct. 209, 215-16, 13 L.Ed.2d 125 (1964). Moreover, the defendants never conceded this bedrock element of plaintiff's case; quite to the contrary, they have adamantly maintained throughout these proceedings that the evidence of the story's truth (or substantial truth) precludes any reasonable inference of actual malice. See 567 F.Supp. at 654 n. 9; Brief for Appellees at 18. In turn, Tavoulareas himself has relied on evidence of falsity in his effort on appeal to establish actual malice, see, e.g., Brief for Appellant at 54, and the District Court found the evidence of the story's truth highly relevant to its decision to grant j.n.o.v. on the issue of actual malice. See 567 F.Supp. at 659-61. More fundamentally, the truth or falsity of allegedly defamatory speech must be considered if we are properly to balance the individual interest "in vindicating a reputation that is wrongly sullied," Brief for Appellant at 17c (emphasis added), against the community's interest in free speech. See Philadelphia Newspapers, 106 S.Ct. 1558. We reject the extravagant suggestion that we ignore evidence of truth in reviewing the reasonableness of the jury's finding that the Post article was unprotected by the First Amendment.

The undisputed evidence at trial, including plaintiff's own testimony, precludes any reasonable inference that the central allegation of the challenged article--that Tavoulareas "set up" Peter in Atlas--was false. Tavoulareas repeatedly testified at trial that by December 1973 he knew that Peter's supervisor at Lemos, Comnas, was planning to leave Lemos and form a joint venture company. Tr. at 1275-76, 1293, 1433, 1635, 1647. According to his own testimony, Tavoulareas also knew by late 1973 that Comnas had offered Peter a partnership position in that venture. Tr. at 1293; see also Tr. at 1278, 1455.28 What is more, Tavoulareas testified that he thought it likely that Comnas made the offer to Peter not because of Peter's modest qualifications as a young and inexperienced junior executive but in order "to make sure he got business from Mobil." Tr. at 1648; see also Tr. at 1295-96, 1455, 1516; cf. Tr. at 1656 (testimony of Tavoulareas that as of 1974 Peter was young and needed more training in shipping business); Tr. at 3080 (uncontradicted testimony of Kousi, a Samarco director, to same effect).

In the spring of 1974, with Tavoulareas fully aware that Comnas was trying to "curry favor" with him by offering Peter a share of Comnas' venture, Tr. at 1516, Tavoulareas flew to London to ask Comnas "if [Comnas] was interested in taking over the management" of Samarco's ships. RE at 2443-46; see also RE at 2344-45, Tr. at 1534 ("I [Tavoulareas] more than anybody else was responsible for bringing [Comnas] into [the management of Samarco's ships] ..."); Tr. at 1289-91, 3423, 3349. But cf. Dissent at 826-27 (recounting evidence that Comnas was recruited by Mobil as a corporate entity, including actions of its chairman, Rawleigh Warner, rather than by Tavoulareas, notwithstanding Tavoulareas' own uncontradicted testimony).29 Thereupon, Comnas created Atlas to manage Samarco's ships, and Peter--until then an assistant at Lemos--joined Atlas as a part owner. RE at 2421-22, 2426-28; 3141. By itself, the undisputed fact that Tavoulareas personally recruited Comnas to manage Samarco's ships (ultimately through the vehicle of Atlas Maritime) shortly after learning that Peter had an outstanding offer from Comnas goes far toward justifying the charge that Tavoulareas "set up" his son in Atlas. See 567 F.Supp. at 659 (finding that Tavoulareas' 1977 SEC testimony, describing this sequence of events, "provided Tyler with a sufficient basis for [his set up] allegation").

But that is not all. The record abounds with uncontradicted evidence of nepotism in favor of Peter. The record unmistakably reveals that Tavoulareas remained personally involved in the Samarco-Atlas arrangement after Peter left Lemos and took on his partnership position at Atlas. In August 1974, only days after Peter joined Atlas as a partner, Tavoulareas took Peter--without Comnas--to Geneva to meet the Alirezas. Tr. at 1305-06. This father-and-son trip to Geneva commenced after Tavoulareas "sent a memorandum to Paul Wolf[e] to tell him [Tavoulareas] would no longer be involved with anything as to Atlas and Samarco" and thus to bypass him on all Atlas matters beyond Wolfe's authority in favor of Rawleigh Warner, Mobil's chairman. RE at 2440; see RE at 2339; see also Tr. at 1332-33, 1464-65; cf. RE at 2344 (Mobil's pre-publication letter sent to Tyler claiming that "[f]rom the date Peter Tavoulareas joined Atlas, Mr. Tavoulareas divorced himself from involvement in matters involving business transactions between Mobil and/or SAMARCO with Atlas.") At a luncheon gathering in Geneva, Tavoulareas and a Mobil subordinate engaged in substantive discussions with the Alirezas regarding the tentative Samarco-Atlas contract. During that conversation, Tavoulareas and his Mobil colleague argued in favor of Atlas' position that Atlas should be independent of Samarco and that its compensation should include both a minimum fee and an equity interest in the ships it managed. Tr. at 1712. See generally RE at 2587-2590 (Peter's written summary of the meeting). Although the Alirezas were of the view that Samarco should have "clean cut control of the management group[,] ... [t]his course of action was opposed by ... Mobil" and Mobil's pro-Atlas position on this fundamental issue prevailed. RE at 2587.

Thereafter, Tavoulareas' personal involvement in building up Atlas, with Peter then an equity partner in the firm, continued unabated. In November 1974, at a social gathering in Saudi Arabia, Tavoulareas attempted to convince Ahmed Alireza to accept Atlas' position on the final sticking point between Samarco and Atlas over the terms of the latter's compensation. Tr. at 1312-13, 1725-27. Based on his conversation with Alireza, Tavoulareas recommended to Comnas and Peter that they accept Alireza's counter-offer. Tr. at 1312-13, 1725-27. Atlas accepted Tavoulareas' advice, and the Samarco-Atlas deal was subsequently consummated.

Having helped Atlas secure its management agreement with Samarco, Tavoulareas then--by his own testimony--personally participated in the series of events whereby Comnas in short order resigned from Atlas, with Peter becoming its 75 percent owner. According to uncontradicted testimony, the decision to discharge Comnas was made in April 197530 at a meeting of high-level Mobil officials in Tavoulareas' own office at Mobil headquarters. Not only was Tavoulareas present for this meeting, but he participated fully in the discussion leading directly to Comnas' removal. Indeed, the meeting commenced with Tavoulareas as the highest ranking Mobil officer present, without Mobil's chairman, Mr. Warner, who only later joined the ongoing conclave. Tr. at 1185-87, 1331-33, 1534.31 Far from divorcing himself from this matter, Tavoulareas, according to Mr. Warner's uncontradicted testimony, stated in that critical meeting:

I more than anybody else was responsible for bringing [Comnas] into this and I think that I should be involved in helping to handle the situation.

Tr. at 1534.

A few days later, Tavoulareas flew to London from New York with two Mobil subordinates to explain to Comnas that "we weren't satisfied with [his] services." Tr. at 1333. After one of Tavoulareas' subordinates met with Comnas "to find out what terms he would agree to for leaving," Tr. at 1195, Comnas asked to meet with Tavoulareas. Tr. at 1336-37; see also Tr. at 3349. Tavoulareas reviewed a draft agreement of terms for Comnas' departure and then met with Comnas. Tr. at 1336-37. At Mobil's unilateral insistence, made without even notifying or consulting its other partners in Samarco, Comnas left Atlas immediately thereafter. Tr. at 1190.

In addition to the trial testimony of Tavoulareas and his witnesses, Mobil's own answers to Tyler's written questions prior to the article's publication expressly conceded that Tavoulareas personally participated "in the arrangements made when G. Comnas departed from Atlas ... to the extent of assuring a settlement that was fair and equitable." RE at 2345. It is also beyond dispute that as part of the settlement resulting in Comnas' resignation Mobil agreed to put Comnas on its payroll as a consultant.32

Tavoulareas also played a pivotal role in helping Atlas not only to survive but to prosper after Comnas left, when Atlas thereby became the firm of Peter and his youthful colleague from Lemos, Ares Emmanuel. Tr. at 1835; see also RE at 2053, 2435. Tavoulareas personally participated in Mobil's internal deliberations that resulted in the decision to make Harmon Hoffmann, a senior and highly respected Mobil executive, available to Atlas as an interim manager. See Tr. at 1193, 2848-49; cf. RE at 2345. Furthermore, Tavoulareas was personally and directly involved in persuading the Alirezas to retain Atlas as Samarco's independent management firm upon Comnas' departure. Tavoulareas flew to Saudi Arabia and personally informed the Alirezas of Mobil's discharge of Comnas, Tr. at 3286-88. Subsequently, the Alirezas took the position that the Samarco-Atlas contract was terminated by virtue of Comnas' departure, Tr. at 3292, and that to continue collaboration, Atlas would have to assign partial control of its stock and management to Samarco. Tr. at 3292-93; see also Tr. at 1867, 2602, 3441. The Alirezas later withdrew these demands antithetical to Atlas (and Peter) only after Tavoulareas and one of his subordinates flew once again to Jeddah and assured Ahmed Alireza that the Samarco-Atlas contract remained valid, Tr. at 3294, 3298-99, 3302, and that Mobil "would support the needs of [Atlas]." Tr. at 3304.

Given this plentiful, undisputed evidence of Tavoulareas' personal involvement in the establishment and operation of Atlas to Peter's manifest benefit, we conclude that no reasonable jury could, on this record, find that the "set up" allegation was false.33

Although the central thrust of the story was not proven false at trial, it is still possible that the story contains defamatory falsehoods. Cf. Afro-American Publishing, 366 F.2d at 655 ("[T]he defamer may be [all] the more successful when he baits the hook with truth."). But cf. Restatement (Second) Of Torts Sec. 581A comment f, at 237 ("Slight inaccuracies of expression are immaterial provided that the defamatory charge is true in substance."). We must, in consequence, carefully consider both the veracity and defamatory character of the three challenged statements in the Post article besides the fundamental "set up" charge.34

Tavoulareas first seeks to premise liability on statements in the article creating the impression that there was a "direct link between Mobil and Atlas." Brief for Appellant at 25. Given the overwhelming proof at trial of precisely such a link, this argument collapses at the outset. To recap briefly, it is undisputed that Mobil recruited Comnas to form Atlas; that Mobil argued--successfully--in favor of Atlas' position with the other Samarco partners on several critical occasions; that Mobil removed Comnas as head of Atlas without consulting its Samarco partners; and that Mobil made a senior Mobil executive available to Atlas as an interim replacement for Comnas. Furthermore, it is beyond cavil that Mobil provided office space and direct financial assistance to Atlas and that Atlas managed the ships that Mobil bareboat chartered to Samarco. See, e.g., Tr. at 1142-48. Thus, even if the Post article failed to make clear the formal, corporate relationship between Mobil, Samarco, and Atlas, but see paragraphs 6, 58-59, 74, the defendants cannot in reason and in law be held liable for accurately reporting the direct link that undisputedly did exist between Mobil and Atlas.

Tavoulareas also challenges the allegation at the end of the article that when Comnas left Atlas, Tavoulareas "dispatched one of his senior shipping executives, Herman [sic] F. Hoffmann, to London to help run Atlas." p 82. Tavoulareas does not contest the fact that Mobil indeed sent Hoffmann to London on that very mission. Rather, he contends that the Post published a defamatory falsehood by suggesting that he "personally ordered a Mobil executive to London to bail out his son's company." Reply Brief for Appellant at 14 (emphasis added). As we have seen, Tavoulareas does not and could not dispute the Post 's allegation that he played a personal role in arranging Comnas' departure from Atlas. Moreover, the article was also undisputedly correct in reporting that Comnas' removal made it " 'natural for Mobil to step forward' ... 'to maintain quality management of [Samarco's] operations.' " paragraphs 83-84 (quoting Kousi as well as Mobil's own statement). Nor is it contested, as noted above, that Tavoulareas personally participated in the Mobil discussions in which it was decided that Hoffmann would replace Comnas. Tr. at 1193. Nevertheless, Tavoulareas asserts that the jury could reasonably have found the "dispatch" allegation to be actionable.

Testimony was presented at trial that Paul Wolfe, Tavoulareas' subordinate at Mobil, actually made the decision to dispatch Hoffmann to Atlas. Tr. at 1098, 1440. Thus, viewing as we do the evidence most favorably to plaintiff, the "dispatch" allegation was false to the extent it overstated Tavoulareas' role in Mobil's sending Hoffmann to replace Comnas. The potentially defamatory inference that could be drawn from this falsehood is that Tavoulareas had not recused himself from Atlas matters despite a possible conflict of interest. But we have already held this implication not to be actionable in light of the overwhelming, undisputed evidence of Tavoulareas' personal, continual, and active involvement in Atlas' matters in ways that uncontestably redounded directly to Peter's benefit. Since the only derogatory implication of the dispatch statement is undisputedly correct, it is not actionable. See, e.g., Herbert v. Lando, 781 F.2d 298, 312 (2d Cir.) (To hold actionable a "statement whose ultimate defamatory implications are themselves not actionable, we believe, would be a classic case of the tail wagging the dog."), cert. denied, --- U.S. ----, 106 S.Ct. 2916, 91 L.Ed.2d 545 (1986); Skrocki v. Stahl, 14 Cal.App. 1, 6, 110 P. 957 (1910) ("It was sufficient if the gist or sting of the libelous charge was justified, and immaterial variances and defects of proof upon minor matters are to be disregarded if the substance of the charge be justified"); Restatement (Second) Of Torts Sec. 581A comment f, at 237.35

Third, and finally, plaintiff vigorously contests the article's assertion that he "personally urged" Comnas to accept Peter as a partner in Atlas. paragraphs 23, 52. The jury reasonably could have concluded that this charge, communicated to Tyler by Comnas, was false. See Tr. at 1293-94, 1296-97, 1425, 1433. And, because the "personally urged" allegation goes beyond the general charge that Tavoulareas "set up" his son and suggests that Tavoulareas actively pressured Comnas to hire Peter rather than merely rewarded Comnas for doing Peter a favor, see supra note 17, the jury may reasonably have found that this specific allegation carries with it an independent defamatory implication capable of causing a separate harm to plaintiff's reputation. We thus turn to the issue of actual malice.

In this branch of our inquiry, the issue is whether the Post defendants published the "personally urged" statement with actual malice--that is, with knowledge of falsity or reckless disregard for its truth. This, in particular, calls upon us to determine whether there is clear and convincing evidence "to permit the conclusion that the defendant[s] in fact entertained serious doubt as to the truth of [their] publication." St. Amant, 390 U.S. at 731, 88 S.Ct. at 1325.36 The "serious doubt" inquiry enunciated by the Supreme Court is too fact-bound to be resolved on the basis of any single factor or mechanical test. Id. at 730, 88 S.Ct. at 1325. Nevertheless, as the Supreme Court has clearly stated, courts may not abdicate to triers of fact the responsibility for developing the contours of the actual malice rule. Instead, courts are obligated to conduct "an independent examination of the whole record." Edwards, 372 U.S. at 235, 83 S.Ct. at 683; see also Sullivan, 376 U.S. at 285, 84 S.Ct. at 729. Thus, a substantial body of law has developed over the past two decades applying and delimiting the "serious doubt" standard first articulated by the Court, speaking through Justice White, in St. Amant. The jury verdict must be squared not only with the literal formulation of the actual malice rule but examined in the light of case law that has "marked out" the outer limits of constitutionally permissible libel verdicts for public figures. St. Amant, 390 U.S. at 730, 88 S.Ct. at 1325.

It is well established that the "serious doubt" standard requires a showing of subjective doubts by the defendant. It does not turn on whether a reasonably prudent person would have published under the circumstances. Id. at 731, 88 S.Ct. at 1325-26. The rejection of an objective standard of care, however, does not mean that libel defendants can defame with impunity merely by testifying that they published the challenged statements with the belief that they were true. Id. at 732, 88 S.Ct. at 1326. To the contrary, a plaintiff may prove the defendant's subjective state of mind through the cumulation of circumstantial evidence, as well as through direct evidence. See, e.g., Herbert v. Lando, 441 U.S. 153, 160, 99 S.Ct. 1635, 1640-41, 60 L.Ed.2d 115 (1979); Washington Post Co. v. Keogh, 365 F.2d 965, 968 (D.C.Cir.1966), cert. denied, 385 U.S. 1011, 87 S.Ct. 708, 17 L.Ed.2d 548 (1967).

At the same time, actual malice does not automatically become a question for the jury whenever the plaintiff introduces pieces of circumstantial evidence tending to show that the defendant published in bad faith. Such an approach would be inadequate to ensure correct application of both the actual malice standard and the requirement of clear and convincing evidence. Thus, as all parties and amici agree, the Supreme Court has directed us to "exercise particularly careful review," Supplemental Brief for Appellant at 9, and to "make an independent examination of the whole record." Edwards, 372 U.S. at 235, 83 S.Ct. at 683, quoted in Sullivan, 376 U.S. at 285, 84 S.Ct. at 729.37

This constitutionally mandated duty of independent review has been applied unflinchingly. The Supreme Court and other courts have more often than not concluded that public figure libel plaintiffs failed to adduce evidence of sufficient clarity to convincingly support a jury finding of actual malice. For example, in the leading case of St. Amant v. Thompson, 390 U.S. 727, 88 S.Ct. 1323, the Supreme Court reversed a jury finding of liability in a public official's defamation action. Because of St. Amant 's importance in actual malice analysis, we pause to recall the facts of that case. Briefly stated, Deputy Sheriff Thompson sued St. Amant, a candidate for public office, for repeating in the course of a televised speech the false allegation that Sheriff Thompson had taken bribes from a local Teamsters Union president. The record showed that St. Amant had based his allegation exclusively on information provided by an active member of a dissident faction within the Teamsters. At the time, the dissident faction was locked in a struggle for control against the faction led by the Teamsters' official alleged to have paid bribes to the Sheriff. Although he had no knowledge of the source's reputation for veracity, St. Amant failed to investigate independently the obviously serious charge of bribery and failed to seek confirmation of the information from others who might have known the facts. Notwithstanding this evidence, the Supreme Court found the record insufficient to support a finding of actual malice.

Before evaluating the specific record before it, the St. Amant Court provided examples of the kind of proof that would likely support a finding of actual malice. The examples fell into three general categories: evidence establishing that the story was (1) "fabricated"; (2) "so inherently improbable that only a reckless man would have put [it] in circulation"; or (3) "based wholly on an unverified anonymous telephone call" or some other source that the defendant had "obvious reasons to doubt." 390 U.S. at 732, 88 S.Ct. at 1326. After setting forth these illustrative examples, the Court held that the evidence before it, by comparison, was clearly inadequate. St. Amant's failure to investigate was deemed not indicative of actual malice, inasmuch as the plaintiff had not proven "a low community assessment of [the source's] trustworthiness or unsatisfactory experience with him by St. Amant." Id. at 733, 88 S.Ct. at 1326. The Court also found support for its decision in evidence tending to show that St. Amant published the charge in good faith, including St. Amant's testimony that he had verified other aspects of his source's information and evidence that the source had sworn to his answers in the presence of newsmen.

As the District Court correctly observed in the case at hand, the Supreme Court's reasoning and result in St. Amant are instructive for inferior tribunals in attempting faithfully to apply the "serious doubt" test. 567 F.Supp. at 656. The examples provided there of when a jury may reasonably infer actual malice from circumstantial evidence are by no means exhaustive, but, as numerous courts have recognized, constitute useful benchmarks for lower courts to employ in determining whether a record is sufficient to sustain a finding of constitutional malice. See, e.g., Marcone v. Penthouse International Magazine For Men, 754 F.2d 1072, 1089-90 (3d Cir.), cert. denied, --- U.S. ----, 106 S.Ct. 182, 88 L.Ed.2d 151 (1985); Hunt v. Liberty Lobby, 720 F.2d 631, 643-46 (11th Cir.1983). When the entire record in this case is scrutinized in light of St. Amant and other governing precedents, it is clear beyond cavil that Judge Gasch's decision to grant j.n.o.v. was fully justified.

Tavoulareas seeks to support the jury finding of actual malice on the basis of both evidence generic to the entire article, such as managerial pressure to produce stories, and evidence relating more specifically to various statements in the article. We turn first to the evidence of actual malice relating to the single remaining statement in the article that may be actionable.

a. Personally Urged." It is undisputed that Comnas was the sole source for the Post 's allegation that Tavoulareas "personally urged" that Peter be included as an equity partner in Atlas. Plaintiff contends that reliance on Comnas was evidence of the Post defendants' actual malice inasmuch as the record supports the conclusion that Comnas was an obviously unreliable source. Our examination of the record, however, leads us to agree completely with the District Court's contrary conclusion that "reliance upon George Comnas as a primary source does not come close to approaching the level of recklessness required by the Supreme Court." 567 F.Supp. at 656.

As did the District Court, we find it highly significant that "much of Comnas' information was independently verified by other sources whose credibility even the plaintiff does not now challenge." 567 F.Supp. at 656. For instance, Comnas told Tyler that Tavoulareas had participated on Atlas' behalf in key negotiations with the Saudi partners of Samarco after Peter had joined Atlas. Tr. at 3688, 3778-84. Mobil denied this critical contention in its pre-publication letter to the Post, claiming (erroneously, as Tavoulareas conceded at trial, Tr. at 4161-62) that "[f]rom the date Peter Tavoulareas joined Atlas, Mr. Tavoulareas divorced himself from involvement in matters involving business transactions between Mobil and/or SAMARCO with Atlas." RE at 2344. Tyler published Comnas' charge only after receiving independent confirmation from Kousi, Tr. at 3083-3100, a highly respected, knowledgeable member of the Samarco board with firsthand knowledge of Atlas' formation and engagement as Samarco's management firm. The only reasonable inference is that Tyler was justified in considering Kousi's confirmation of Comnas' statements important evidence of the latter's reliability. Tr. at 785-86, 3689, 3792.

But this was by no means all the information in Tyler's hands that buttressed Comnas' statements. Tavoulareas' own testimony before the SEC staff admitted his recruitment of Comnas to head Atlas and his knowledge of Comnas' business relationship with Peter. RE at 2443-46, 2424-25. On behalf of Mobil, Paul Wolfe confirmed Comnas' statement to Tyler that Tavoulareas was personally involved in Comnas' departure from Atlas, RE at 2345, a significant admission in light of the understanding communicated by Mobil to Tyler and outside members of the board of directors that Tavoulareas was not involved in Atlas matters after Peter's entry into the firm. Tr. at 3550, 3588-89; RE at 2344. Moreover, Kousi confirmed Comnas' general charge that Peter obtained his position with Atlas on the basis of kinship, not merit. Tr. at 3079-80 ("[T]he employment of Peter Tavoulareas in Atlas was a nepotistic act."). In sum, the undisputed evidence of record shows conclusively that Tyler corroborated much of the information provided by Comnas. Cf. St. Amant, 390 U.S. at 733, 88 S.Ct. at 1326-27, (verification of part of source's information probative of St. Amant's lack of actual malice in relying on source for other, uncorroborated information).

Besides independent, reliable corroboration of most of Comnas' story, Tyler had other, substantial reasons to believe Comnas. Of these, the most important was Tyler's uncontroverted knowledge that "Comnas provided virtually the same information to Tyler as he gave to investigators for the House Subcommittee," 567 F.Supp. at 656, including the substance of the "personally urged" allegation. RE at 2471; cf. St. Amant, 390 U.S. at 733, 88 S.Ct. at 1326-27 (St. Amant's knowledge that source repeated information to others in manner bespeaking earnestness indicative of good-faith belief in source's credibility). It is undisputed that an experienced House Subcommittee investigator, Peter Stockton, accompanied by a Subcommittee lawyer, traveled from Washington to Comnas' home in Osterville, Massachusetts. Tr. at 2770; RE at 2467. There, they conducted a formal three-hour interview, complete with elaborate notetaking; it is undisputed, moreover, that the two Subcommittee staff members were interviewing Comnas in their official capacities. This was a Congressional investigation, not an idle conversation between a passerby and a disgruntled executive.

After interviewing Comnas, Stockton confirmed Tyler's judgment that Comnas was credible, that he spoke reluctantly and carefully for fear of reprisal, and that his information was consistent with that provided by other sources. Tr. at 2816-17; cf. Tr. at 2792-93. Tyler was justifiably impressed with Comnas' "consistency" as a source, especially since, as Tyler reasonably concluded, Comnas was on "a higher level of notice" that "he better be truthful in his story" when speaking to the Subcommittee staff. Tr. at 3823-24.38

Tyler also knew that Comnas had enjoyed a long and successful career at Exxon where he eventually became president of Esso Mediterranean and Esso Africa.39 Tyler was further aware that, after leaving Exxon, Comnas had been entrusted to run the principal office of one of the largest Greek shippers (Lemos) and had been recruited by Mobil in early 1974 based upon Mobil officials' "long personal acquaintances with G. Comnas." RE at 2345; see also RE at 2446. As the District Court observed, Comnas "obviously had a substantial business career and reputation that could be damaged if he lied." 567 F.Supp. at 656; see Tr. at 637-38, 3690, 3825.40 Moreover, it is undisputed that no Mobil representatives ever gave Tyler the slightest indication that Mobil questioned Comnas' credibility or ethics.

Despite the abundant, undisputed evidence in the record supporting Tyler's decision to credit Comnas' story, Tavoulareas nevertheless argues that the jury could reasonably have concluded that "Comnas' patent lack of credibility" was apparent to Tyler. Brief for Appellant at 20. To support this contention, Tavoulareas relies heavily on the undisputed fact that Dr. Piro told Tyler that "Comnas had been caught in some fraud involving Atlas" and that the Tavoulareases had threatened to report Comnas to the tax authorities if Comnas ever started any trouble for Peter and Atlas. Tr. at 2908-09. What Piro related, of course, was his understanding from the Tavoulareases. Tr. at 2958, 2965.41

Tyler, as we have seen, tried to talk with both William and Peter Tavoulareas. William refused to talk at all with Tyler, but his sworn SEC testimony indicated that Comnas' withdrawal from Atlas had nothing whatever to do with unethical conduct. RE at 2435. Peter hung up the phone when Tyler called, having chosen, as he was free to do, not to enlighten Tyler as to the reasons for Comnas' departure from Atlas or to correct any misinformation Tyler had concerning Comnas' trustworthiness. Tr. at 2437. Mobil, of course, steadfastly refused Tyler's requests for interviews, as it was likewise free to do. Faced with Mobil's silence, Tyler checked with Kousi, who told him that it was the "dissatisfaction of Mobil," not any of the other parties, that led to Comnas' removal. Tr. at 3082-83; cf. Tr. at 2099-2100 (testimony of Harmon Hoffmann that the sole reason for Comnas' removal was dissatisfaction with his performance, not any ethical considerations). Thus, the undisputed record comports entirely with the District Court's conclusion that "it is not at all clear that Comnas left Atlas under circumstances that should have aroused Tyler's suspicions as to his motivation for speaking unfavorably about the [Tavoulareases]." 567 F.Supp. at 657 (footnote omitted).

We also agree with the District Court that the evidence of personal animus between Comnas and Tavoulareas was not indicative of the defendants' actual malice in view of their independent corroboration of Comnas' information and the other evidence in the hands of the Post defendants of Comnas' knowledge and credibility. Cf. St. Amant, 390 U.S. at 733, 88 S.Ct. at 1326-27 (reliance on a source locked in a fierce struggle with union officials closely allied to Sheriff Thompson did not permit an inference of actual malice when St. Amant had "verified other aspects" of the source's information and had other indicia of the source's credibility).

Finally, we reject the suggestion that the Post 's failure to introduce any of Comnas' deposition testimony at trial was probative of actual malice. Even assuming that Tavoulareas can advance this assertion in the first instance,42 Comnas was deposed for four days by both sides, yet neither side ultimately elected to introduce his testimony into the trial record. Since Comnas' deposition transcript was "equally available," the decision of both sides not to use it cannot serve as the basis for an adverse inference against the Post. Brown v. United States, 414 F.2d 1165, 1166-67 (D.C.Cir.1969); see also United States v. Young, 463 F.2d 934, 939-40 (D.C.Cir.1972); Tr. at 3817.

b. General Evidence of Actual Malice. Last, we consider the more general evidence in the record that, as Tavoulareas sees it, supports a determination that the Post defendants published the November 30 article with actual malice. At the outset, we observe that our examination of this evidence cannot be divorced from our earlier conclusion that the record before us cannot reasonably support a finding that the central theme of the article was false. For example, Tavoulareas argues that the Post 's refusal to retract the article or to concede its falsity at trial was evidence of actual malice in view of "the overwhelming proof of the article's falsity at trial." Brief for Appellant at 27. But this broadside cannot stand in light of our earlier analysis and conclusions with respect to the issue of falsity vel non. We therefore need only address briefly to what extent, if any, this evidence could reasonably have supported the jury finding of actual malice.

(i) The Christine Peterson Memorandum. Christine Peterson had the task of performing the final edit on the November 30 article. During Peterson's brief testimony at trial, it was established that her duties included editing the story for "style, punctuation, [and] grammar," confirming facts that could be checked with source material at hand, and writing the headlines. Tr. at 2827. The record is also clear that in editing the article at issue, she performed a "fresh read," that is, she had no knowledge of the sources or research underlying the article beyond what she read in the article itself. See Tr. at 2829; RE at 2486-87; 567 F.Supp. at 655. After editing the copy, and prior to publication, she sent a memorandum to assignment editor Peter Milius. The memorandum stated in part:

I've read the Mobil story several times, and while I'm impressed with the amount of work the reporter obviously did, I'm still left with an overwhelming sense of So What? Is there any way to give this story of high-level nepotism a dollars-and-cents angle? Did Mobil's shareholders lose anything? Mobil's customers? Parts of Tyler's case against Tavoulareas seem tenuous, and the whole--a $680,000-a-year plaything for an indulged son, at worst--just seems like a withered peanut in an 84"' gilded shell. A far more interesting angle, it seems to me, is Mobil's concern about Saudi preference shipping--a concern so profound that it led to the formation of an entire dummy corporation. It's impossible to believe that Tavoulareas alone could put together such a scheme for the sake of his son's business career, or that he would want to.

RE at 2486.

At trial, Peterson testified that "my point in the memorandum was that I thought the formation of Samarco was a more interesting angle in the story." Tr. at 2833. It was, as she put it, the seemingly unusual corporate arrangements that provided "a more important angle for the story to have as a lead." Tr. at 2834. In order to credit all permissible inferences that the jury may have drawn, we reject Peterson's trial testimony and accept Tavoulareas' proffered inference. Tavoulareas contends that the Peterson memorandum entitled the jury "to find that an editor of the Post charged with the final review of the story before publication had indeed found ... the story's theme impossible to believe" and brought home this doubt to those responsible for the article's publication. Brief for Appellant at 35. We accept this view--that the Peterson memorandum is evidence of "serious doubt" or actual malice. But defamation plaintiffs cannot show actual malice in the abstract; they must demonstrate actual malice in conjunction with a false defamatory statement. Upon closer examination, the Peterson memorandum does not qualify as such a showing and thus fails to constitute acceptable evidence of actual malice. It certainly falls short of clear and convin