Circuit Judge, concurring in part and dissenting in part:
TABLE OF CONTENTS
Page
I. Background............................................. ^ 124
II. Legal Standard......................................... 125
A. Overview............................................. 125
B. The Constitutional Responsibility of Independent Review 126
III. Actual Malice........................................... 130
A. The Challenged Statements............................ 130
B. Factors Lacking Any Probative Weight................ 132
1. Defendants’ motivation............................ 132
(a) The Post ..................................... 134
(b) Tyler......................................... 135
*124Page
(c) Golden ................................................ 136
(d) Piro ................................................... 136
2. The Post's editorial decisions ............................... 136
(a) Failure to include certain information ................... i 136
(b) Resolution of ambiguities............................... 137
3. The Checket conversation and the “dispatch” of Hoffmann ... 138
4. Other indicia of actual malice............................... 139
C. Potentially Probative Factors................................... 140
1. The Peterson memorandum................................. 140
2. Reliability of sources....................................... 141
(a) Comnas................................................ 141
(b) Piro................................................... 142
IV. Tavoulareas’ Status As a Limited Purpose Public Figure........ 142
V. Golden ........................................................... 145
VI. Piro .............................................................. 145
VII. Conclusion....................................................... 146
“[W]e consider this case against the background of a profound national commitment to the principle that debate on public issues should be uninhibited, robust, and wide-open * * New York Times Co. v. Sullivan, 376 U.S. 254, 270, 84 S.Ct. 710, 721, 11 L.Ed.2d 686 (1964). In my judgment, the majority opinion strays far from this fundamental commitment, vastly increases journalistic liability, and mistakenly rejects the carefully reasoned opinion of Judge Gasch. I dissent from almost all of the majority opinion; I concur only in the judgment affirming the judgment notwithstanding the verdict for defendant Golden.
I. Background
It is helpful to review the salient facts that are not in dispute. In 1974 Mobil became a substantial partner with prominent Saudis and another company in a joint venture called SAMARCO, a shipping concern. William Tavoulareas, Mobil’s president and the advocate of close cooperation with the Saudis, played a major role in conceiving and preparing Mobil’s participation in SAMARCO. Mobil and its partners agreed that Atlas Trading Company, which had been formed, according to Mobil, “in anticipation of managing SAMARCO’s business,” Record Excerpts (RE) 2344, would manage SAMARCO’s ships; many of those ships would be leased from Mobil. In the spring of 1974 William Tavoulareas personally recruited George Comnas to manage SAMARCO’s ships through Atlas. At the time Tavoulareas knew that his son, Peter, worked for Comnas. Soon after Comnas took control at Atlas, Peter left his $14,000 a year job to become a major partner at Atlas. All of these facts are undisputed.
After researching the story for a month, Washington Post reporter Patrick Tyler wrote an 85-paragraph article about the Mobil-SAMARCO-Atlas relationship and Tavoulareas’ role in it. More than 20 paragraphs of the article reported Mobil’s version of the events in question. Tavoulareas challenges three statements and one implication in the article: the statement that he “set up” his son and “personally urged” his inclusion as a partner, the statement that he “personally dispatched” an executive to Atlas after Comnas’ departure, the statement that Mobil “provided” ships to Atlas, and the possible implication that Tavoulareas engineered the transaction solely to benefit his son.
Assuming that Tavoulareas is a public figure, see discussion in text at 162-165 infra, the question before this court is whether these statements and the claimed implication are stripped of their First Amendment protection because they were false, defamatory, and published with knowledge or reckless disregard of their falsity. Since I find that the evidence in *125the record falls far short of establishing clear and convincing proof of knowledge or reckless disregard of falsity, I view that question as dispositive. For, even if the statements are false and defamatory, without such clear and convincing proof of knowledge or recklessness, thfe First Amendment unquestionably protects the challenged expression.
II. Legal Standard
A. Overview
The majority opinion achieves no less than an ambitious, wide-ranging revision of libel jurisprudence. The majority accomplishes this revision in two ways. First, it counts, as elements of actual malice, factors that are completely impermissible considerations in reaching the daunting conclusion that the First Amendment does not apply to the challenged expression. These factors include criticism of a newspaper’s emphasis on attention-getting stories, majority opinion (maj. op.) at 120-121; condemnation of the career ambitions of young journalists, id. at 119-120; challenges to a newspaper’s resolution of reasonable ambiguities, id. at 121-123; and rejection of editorial judgments about the information to be included in an article that clearly reports both sides of a disputed story, id. at 123-125. Indeed, the majority appears to criticize what it takes to be the general climate in journalism today. It goes so far as to attach some opprobrium to a newspaper’s policy of “ ‘hard-hitting investigative journalism,’ ” maj. op. at 121, and, too, emphasizes the need to deter “the least responsible journalists” who might “scoop their more careful colleagues.” Id. at 131 n. 50.
In my judgment, we do not sit, even in reviewing a libel verdict, as some kind of journalism review seminar, offering our observations on contemporary journalism and journalists. Our mission is to see that the First Amendment is vigorously protected and that libel verdicts not supported by clear and convincing evidence do not stand. The majority’s use of these impermissible factors as evidence of actual malice seriously undermines that mission. The majority stresses that we should suspend judgment on each of its factors because, after all, “ ‘a brick is not a wall.’ ” Maj. op. at 122 (citation omitted). But when a particular factor is so insubstantial and so inappropriate to the weighty First Amendment matters before us, it clearly is not even a brick.
Second, the majority dramatically narrows our well-established constitutional responsibility to conduct an independent review of the record in these First Amendment cases. The nature of that review is critically important, and the majority seeks to reduce it to a pro forma, mechanical ratification to be undertaken only after the pieces of evidence have already been uncritically labelled “indicia of actual malice” and “cumulated.” The majority’s determined effort to constrict our constitutionally mandated inquiry will be discussed in detail below.
Through these two devices, the majority manages to reinstitute a jury verdict of libel in this case. In a brief aside at the last moment, the majority dangles the possibility that this excessive $2.05 million verdict may ultimately be reduced in further proceedings. Maj. op. at 137. And that is at least some comfort, given that these damages seem clearly disproportionate to any possible rendering of the Post’s conduct and Tavoulareas’ injury. But it is, of course, the libel verdict itself, not merely the amount of damages, that chills First Amendment expression. Indeed, a majority of this court recently recognized that a libel suit itself has the potential to chill. See Ollman v. Evans, 750 F.2d 970, 991 n. 44 (D.C.Cir.1984) (en banc); id. at 993 (Bork, J., concurring). And in a statement that aptly summarizes the problems with the majority’s approach in this case, three members of the Oilman court joined Judge Bork’s statement that “[i]t is not merely the size of damage awards but an entire shift in the application of libel laws that raises problems for press freedom.” Id. at 996-997. In short, our central inquiry is whether there is clear and convincing evi*126dence of reckless falsity; the possibility of reduced damages does not remedy the problems in the majority’s decision, or .the distortion that it works on our constitutionally mandated independent inquiry.
This is not a casé that lends itself to easy capsulization. The record — which we are constitutionally charged to review — includes some nine volumes and 3400 pages in the Joint Appendix submitted to us, and that appendix contains only record excerpts. The issues are many and varied, touching on countless subdoctrines in the often baroque field of libel jurisprudence. But despite the vastness of the record and the number and complexity of the issues, our responsibility is fundamental, all-important, and inescapable: we must determine whether the First Amendment protects the challenged expression.
To discharge this responsibility, it is necessary for the court — and this dissent — to examine the varied issues and the complex record with extreme care. With respect to the majority’s use of impermissible factors, I will discuss each impermissible factor in turn as part of a systematic analysis of the elements that, in the majority’s view, are sufficient to reinstate the verdicts. However, the nature of the independent review that is our constitutional responsibility— and that the majority transforms into a pale imitation of its former self — must be considered in detail at the outset.
B. The Constitutional Responsibility of Independent Review
In Bose Corp. v. Consumers Union of United States, Inc., 466 U.S. 485, 10S.Ct. 1949, 80 L.Ed.2d 502 (1984) (hereinafter cited only to 104 S.Ct.), the Supreme Court vigorously reaffirmed the New York Times requirement of independent appellate review for determinations of actual malice. The Court stated unequivocally,
“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.’ ” Id. at 1965. Specifically, the Court noted that this responsibility extends to “an independent assessment * * * of the evidence germane to the actual malice determination.” Id. at 1967 n. 31.
This case concerns the exercise of that independent review in considering a challenge to a trial judge’s decision to grant a j.n.o.v. In the majority’s rendering, our review in this situation becomes a process in which we resolve all inferences in favor of the plaintiff, reject conflicting evidence supporting the defendant, reflexively label inferences “indicia of actual malice,” “cumulate” these malice-laden inferences, and then ask whether actual malice has been clearly and convincingly established. In my judgment, this approach assumes its conclusion, narrowly cabins our review, and deprives the media of the protection that New York Times and Bose sought to confer — a vigorous judicial review to ensure that verdicts based on less than clear and convincing evidence of actual malice will not stand. It thus reduces the safeguards for the “journalistic independence” that this court has stressed. McBride v. Merrell Dow & Pharmaceuticals, Inc., 717 F.2d 1460, 1466 (D.C.Cir.1983).1
According to the majority, New York Times and Bose create some kind of distinction between the “manner” and “standard” of evaluating evidence. This distinction, articulated in two Ninth Circuit decisions, is, of course, nowhere to be found in the Supreme Court’s frequent admonitions to “make an independent examination of the whole record,” see, e.g., Old Dominion Branch No. 496, Nat’l Ass’n of Letter *127Carriers v. Austin, 418 U.S. 264, 282, 94 S.Ct. 2770, 2780, 41 L.Ed.2d 745 (1972). Since the majority seeks to enthrone it as the law of the circuit, however, the distinction bears some attention.
In the majority’s view, “ ‘[t]he standard against which the evidence rnr^st be examined is that of New York Times and its progeny. But the manner in which the evidence is to be examined in the light of that standard is the same’ ” as in all other challenges to j.n.o.v. determinations. Maj. op. at 105 (quoting Guam Federation of Teachers, Local 1581 v. Ysrael, 492 F.2d 438, 441 (9th Cir.), cert. denied, 441 U.S. 962, 95 S.Ct. 132, 42 L.Ed.2d 111 (1974) (emphasis by the Guam court)). Thus the majority concludes that “the evidence and the reasonable inferences derived from it” should, be “examined in the light most favorable to the plaintiffs,” id. at 106, and that the j.n.o.v. posture of the case requires that “all reasonable inferences must be granted to the plaintiff.” Id. at 108. Critically, this presumption in granting inferences even applies to inferences that the majority has determined are “indicia of actual malice” — reliance on unreliable sources, resolution of ambiguities against the plaintiff, and so forth. These inferences are then “cumulated.” Only after these inferences have been granted in favor of the plaintiff and cumulated, the majority reasons, does the New York Times/Bose responsibility of independent review come into play, and the question is whether these cumulated inferences prove actual malice by clear and convincing evidence.
This approach renders the independent appellate review promised by Bose a mirage. Independent review of the record devolves into review of the pool of pre-selected inferences that cut against the defendant. It comes as no surprise, then, that, at the point for the majority’s exercise of independent review, the outcome has already been largely decided against the defendant. This is far from the independent review envisioned by Bose “to preserve the precious liberties established and ordained by the Constitution.” 104 S.Ct. at 1965.
The majority’s method of analyzing the evidence and permissible inferences conflicts with Supreme Court decisions requiring an independent review of the record. In Bose itself the decision turned on the inference to be drawn from the finding that a witness was not telling the truth in failing to understand the difference between two phrases: the Supreme Court rejected the inference, drawn by the trier of fact, that the witness recognized the difference at the time of publication, and the Court drew its own inference that the testimony was part of the witness’ “capacity for rationalization.” 104 S.Ct. at 1966. As the majority notes, maj. op. at 107, the Bose Court relied on the “purely factual findings” of the trial judge, 104 S.Ct. at 1967; as the majority fails to note, the Court rejected, as part of its independent inquiry, an inference drawn by the trial judge that was freighted with actual malice implica- • tions.
Bose was far from the first case in which the Court emphasized that the constitutional responsibility of independent review encompasses far more than the majority’s exercise in ritualistic inference granting. In Time v. Pape, 401 U.S. 279, 91 S.Ct. 633, 28 L.Ed.2d 45 (1971), the Court drew its own inference about the character of a government report described by the defendant. Id. at 286-290, 91 S.Ct. at 637-639. Though reviewing a directed verdict — which has the same standard as a j.n.o.v., Vander Zee v. Karabatsos, 589 F.2d 723, 726 (D.C.Cir.1978) — the Court never suggested that its independent review of the evidence in the record should be structured by automatic resolution of malice-laden inferences against the moving party. In its independent evaluation of actual malice determinations, moreover, the Supreme Court has repeatedly stressed that its review is “an independent examination of the record as a whole," Beckley Newspapers Corp. v. Hanks, 389 U.S. 81, 82, 88 S.Ct. 197, 198, 19 L.Ed.2d 248 (1967) (emphasis added), not merely the bits and pieces which support “indicia of actual malice.” Finally, in other constitutional areas in which the Supreme Court has exercised *128independent review the Court has not undertaken the kind of selective reading of the record and automatic inference assumption that the majority employs.2
Nor has the Court relied on a purported distinction between the “manner” and “standard” of evaluating evidence in reviewing the record. In Bose the question before the Court was the very similar question of whether Fed.R.Civ.P. 52(a) — which provides that findings of the trial court3 may only be set aside if “clearly erroneous” — governed a determination of actual malice. The Court did not turn to some manner/standard distinction. Rather, in emphasizing the differences between questions of fact and questions of mixed fact and law the Bose Court emphasized the delicate judgments involved and the necessity of giving a narrow reading to the realm of pure fact in certain circumstances. “Where the line [between fact and law] is drawn varies according to the nature of the substantive law at issue. Regarding certain largely factual questions in some areas of the law, the stakes — in terms of impact on future cases and future conduct — are too great to entrust them finally to the judgment of the trier of fact.” 104 S.Ct. at 1960 n. 17.4 Instead of relying on a supposed distinction between the manner and *129standard of evaluating evidence, the Court framed the judicial responsibility more forthrightly — “an independent assessment * * * of the evidence germane to the actual malice determination.” Id. at 1967 n. 31.5
Similarly, the majority’s emphasis on some great divide between “each separate fact determination that forms the basis for the ultimate conclusion of ‘actual malice’ ” and “the ultimate conclusion of clear and convincing proof of ‘actual malice,’ ” maj. op. at 107, also distorts Bose. The majority’s central contention is that the Bose emphasis on the difference between the scope of Rule 52(a) and the scope of our independent, constitutional review corresponds to the majority’s own distinction between “preliminary factual determinations” and “ultimate” facts. Maj. op. at 108. But that is simply not how Bose delineates the difference between Rule 52(a) (which applies to factual questions) and our independent review (which applies to constitutionally mixed questions of fact and law). The Court emphasized that Rule 52(a) applies to the “many findings of fact [in a libel case] that are irrelevant to the constitutional standard of New York Times v. Sullivan,” 104 S.Ct. at 1967 n. 31 (emphasis added). Our independent review, in contrast, extends to “the evidence germane to the actual malice determination.” Id. Clearly, the inferences labelled “indicia of actual malice” are relevant and germane to the constitutional inquiry; indeed, they are essential to it. Thus they are appropriately within the scope of our independent appellate review, rather than within the scope of Rule 52(a). Far from “vesting] appellate courts with original jurisdiction in libel actions,” maj. op. at 108, that Supreme Court-identified “rational stopping point,” id., ensures that we may discharge our constitutional responsibility and see that the decision not to extend the First Amendment to challenged expression is limited, in these cases, to those instances in which there is clear and convincing evidence supporting the verdict.
I believe that the Bose Court meant what it said, and that we must undertake an independent assessment of the evidence germane to the actual malice determination — not merely a ratification of the preselected inferences supporting an actual malice conclusion. To be sure, we must recognize that “the presumption that attaches to factual findings is stronger in some cases than others.” 104 S.Ct. at 1959. Credibility of a witness may be the most compelling case for deferring to the factfinder; applying inferences that are labelled “indicia of actual malice,” however, may be the least, for it directly affects the independent assessment of the evidence germane to actual malice that Bose requires.
In my judgment, Judge Gasch’s well-reasoned opinion, see Tavoulareas v. Washington Post Co., 567 F.Supp. 651 (D.D.C. 1983), decided before Bose, is far closer to Bose’s conception of independent review and sensitivity to First Amendment values than the majority’s opinion.6 For it seems *130clear to me that if evidence in the record is to be considered probative of actual malice, the judge must answer two questions. First, can this be the type of evidence which the Supreme Court has found may be probative of actual malice? Second, if so, does the evidence in this case contribute to a finding of knowledge of falsity or reckless disregard of truth or falsity? Failing to address these two questions seems to me to shirk this court’s responsibility to undertake an independent examination and assessment of the record.7
With these principles in mind, and without the blinders of automatic assumption of malice-laden inferences and their consequent cumulation, I review the evidence in the record to determine if it contains clear and convincing evidence of actual malice.
III. Actual Malice
As noted, our constitutional responsibility requires a careful evaluation of “the evidence germane to actual malice.” Bose, supra, 104 S.Ct. at 1967 n. 31. In this case, and in view of the majority’s opinion, that obligation requires an unusually extended discussion. Nonetheless, an independent examination of the record reveals that most of the factors cited by the majority fail to raise even a substantial question about knowledge of falsity or reckless disregard of truth or falsity. Two factors do pose a more substantial question — the Peterson memorandum and the reliability of sources. Even those factors, however, do not, in my mind, rise to the level of clear and convincing evidence of knowledge of falsity, or reckless disregard of truth or falsity.
Before considering the lengthy list of factors that comprises the majority’s actual malice analysis, it is helpful to focus on the statements that are the putative basis for Tavoulareas’ lawsuit and the $2.05 million verdict, and on the materials that were before the Post defendants at the time of the article. For the majority’s list of factors is only relevant insofar as it contributes to an understanding of whether the defendants published those statements with knowledge or reckless disregard of their falsity.
A. The Challenged Statements
For each of the three statements and the implication that Tavoulareas challenges, I will briefly review the statement and the Post’s supporting materials.8 This relationship, never adequately explored by the majority, is an important backdrop for the majority’s wide-ranging conclusions about the reportage at issue. The ultimate question for purposes of this actual malice analysis, it should be remembered, is the Post defendants’ state of mind in reviewing this material — the degree to which the defendants acted with knowledge or reckless disregard of falsity — not the accuracy of the statements themselves.9
*1311. Tavoulareas “set up” his son and “personally urged” his inclusion in Atlas. Tavoulareas challenges two descriptions of his role in establishing the MobilSAMARCO-Atlas relationship: in the headline and the first paragraph the Post reported that Tavoulareas “set up his son”; in the 23rd and 52nd paragraphs it reported that Tavoulareas “personally urged” his son’s inclusion as a partner in Atlas.
In support of the first characterization Tyler and the Post had various information before them at the time of publication. As Judge Gasch observed, “[W]hile [Tavoulareas] may contest the ultimate truth of some of this information, he does not dispute the fact that Tyler had this material in his possession at the time he wrote the story.” 567 F.Supp. at 659.10 Tyler had obtained a copy of Tavoulareas’ 1977 SEC testimony in which Tavoulareas testified (1) that, in early 1974, he knew that his son worked as an assistant to George Comnas, (2) that he played a major role in the creation of Mobil’s role in SAMARCO, (3) that he went to London in the spring of 1974 to interest Comnas in managing SAMARCO’s ships through Atlas, and (4) that his son soon became a major partner in Atlas. RE 2051-2053, 2398-2460. Judge Gasch was so impressed with the significance of this information that he concluded, “[Tavoulareas’] own 1977 SEC testimony provided Tyler with a sufficient basis for this allegation.” 567 F.Supp. at 659. In addition to the SEC testimony, former SAMARCO director John Kousi, whose credibility is not in dispute, told Tyler that the hiring of Peter Tavoulareas was “clearly a nepotistic act.” RE 1163-1166, 1237, 1927-1928. Mobil executive vice president Paul Wolfe also confirmed in response to Tyler’s questions that Tavoulareas was involved in planning SAMARCO and Atlas and in recruiting Comnas. RE 2344-2345. Finally, George Comnas, whose credibility is fiercely disputed, told Tyler that Tavoulareas asked him to bring Peter Tavoulareas into Atlas and that Tavoulareas later asked him to resign in return for a consultant’s position with Mobil. RE 981, 989.
In support of the second characterization Tyler had Comnas’ statement and the contextual corroboration provided by Tavoulareas, Kousi, and Wolfe about Tavoulareas’ deep involvement in the formation of the Mobil-SAMARCO-Atlas relationship, as well as his personal recruitment of Comnas at a time when he knew Comnas was in business with his son. The Tavoulareas SEC testimony and the Wolfe letter did deny that Tavoulareas personally urged Peter’s inclusion; the Post reported this denial four times, including both points at which the “personally urged” statement was made. ¶¶ 23, 25, 32, 53.
2. Tavoulareas “dispatched” Hoffmann to run Atlas. In the 82nd paragraph the Post reported that, after Comnas' departure from Atlas, “the elder Tavoulareas dispatched one of his senior shipping executives, Herman F. Hoffmann to London to help run Atlas.” As Judge Gasch found, Tavoulareas “apparently participated to some extent in the discussion that led to Hoffmann’s temporary assignment at Atlas.” 567 F.Supp. at 661. Mobil had informed Tyler of this participation. Wolfe wrote that Tavoulareas personally negotiated the settlement for Comnas’ departure, that Tavoulareas played a role, which Wolfe described as “minor,” in the arrangements after Comnas’ departure, and that an unspecified “[w]e” sent Hoffman to Atlas. RE 2345. In its discussion of the “dispatched” statement as part of its list of factors, the majority dismisses this evidence. See text at 158-159 infra. In addition to the Wolfe letter, Tyler had Piro’s comments and his own version of the Checket conversation, the credibility of both of which is vigorously disputed.
*1323. Mobil “provided” ships to Atlas. Tavoulareas claims that the Post’s reference to ships “provided” by Mobil to Atlas was published with reckless disregard or knowledge of its falsity. His objection is apparently rooted in the view that use of the word “provided” suggested a direct relationship between Mobil and Atlas instead of the actual relationship — that Mobil bareboat-chartered vessels to SAMARCO and time-chartered them back and that Atlas managed the vessels. However, the Post article did describe the full, complex transaction. 1111 6, 58-59, 74. Furthermore, Kousi had emphasized to Tyler that control of ships effectively passed from Mobil to Atlas. RE 1940. Finally, Wolfe wrote that “Atlas was created in anticipation of managing SAMARCO’s business” and that Mobil beneficially owned four of the seven ships operated by Atlas on behalf of SA-MARCO. RE 2344-2346.
4. The implication that Tavoulareas engineered the entire Mobil-SAMARCOAtlas arrangement for the benefit of his son. Tavoulareas also argues that “the article’s primary implication was * * * that plaintiff misused Mobil assets by engineer;ing the entire Mobil-Samarco-Atlas arrangement for the benefit of his son.” Brief for appellant at 25-26. Judge Gasch rejected this contention because “[t]he article * * * did not say that William Tavoulareas put together the Mobil-Atlas-Samarco arrangement solely for the sake of his son. * * * [Tjhe article specifically stated * * * that Samarco and Atlas were created for certain political and economic reasons. The nepotism discussed in the article clearly had its origin in a legitimate business opportunity.” 567 F.Supp. at 660 (footnote omitted). It is undisputed that the article reported Mobil’s business reasons for entering into the SAMARCO-Atlas relationship — its belief that “a shipping partnership with the Saudis might yield preferential treatment * * * [and] ‘favorable financing from Saudi Arabian sources.’ ” Ml 40-41 (quoting 1976 Mobil message to stockholders).
These, then, are the statements and supporting materials that form the core of the $2.05 million verdict and the majority’s determination that the First Amendment does not protect the Post’s article.
B. Factors Lacking Any Probative Weight
With these challenged statements and their supporting materials in mind, we can turn to the numerous factors that constitute the majority’s actual malice analysis. Most of the factors cited by the majority completely lack probative weight for a finding of knowledge or reckless disregard of falsity. These insubstantial factors include the majority’s comments about the defendants’ motivation, the defendants’ editorial process (the claimed omission of information and resolution of ambiguities), the disputed Checket conversation, and sundry “other indicia of actual malice” (the defendants’ refusal to retract, the lack of deadline pressure, and the knowledge of harm).
It is important to consider these factors in detail, for they are so insubstantial and inappropriate that they should not receive any weight. The time and effort required to analyze them, by the majority and by this dissent, should not give them a weight that they lack on their merits. These factors simply fail to amount to bricks for the wall that the majority seeks to build.
1. Defendants’ motivation. The majority recognizes that “common law malice is not the equivalent of actual malice in the defamation context,” but argues that “the presence of common law malice or evidence that a newspaper followed a sensationalistic policy, because it provides a motive for knowing or reckless falsehood, is evidence of actual malice.” Maj. op. at 117 (emphasis in original). The majority elaborates that a journalistic policy of “ ‘hard hitting investigative journalism,’ or * * * ‘sophisticated muckraking,’ * * * certainly is relevant to the inquiry of whether a newspaper’s employees acted in reckless disregard of whether a statement is false or not.” Maj. op. at 121-22.
The notion that a policy of muckraking and hard-hitting investigative journalism is *133somehow evidence of actual malice is deeply troubling, contrary to the nation’s tradition of a free and aggressive press, and entirely unsupported by precedent. I will address this startling revision of the actual malice notion in connection with the majority’s use of it in its analysis of the Post.
A more fundamental problem, however, is whether ill will may be considered evidence of actual malice. The Supreme Court has emphasized that relying on the ill will of the speaker to establish actual malice in a libel case has a chilling effect on vibrant discussion in a free society. “Debate on public issues will not be uninhibited if the speaker must run the risk that it will be proved in court that he spoke out of hatred; even if he did speak out of hatred, utterances honestly believed contribute to the free interchange of ideas and the ascertainment of truth.” Garrison v. Louisiana, 379 U.S. 64, 73, 85 S.Ct. 209, 215, 13 L.Ed.2d 125 (1964). See also Greenbelt Cooperative Publishing Ass’n, 398 U.S. 6, 10-11, 90 S.Ct. 1537, 1540, 26 L.Ed.2d 6 (1970) (same). On numerous occasions the Supreme Court has also emphasized that “ill will toward the plaintiff, or bad motives, are not elements of the New York Times standard,” Rosenbloom v. Metromedia, Inc., 403 U.S. 29, 52 n. 18, 91 S.Ct. 1811, 1824 n. 18, 29 L.Ed.2d 296 (1971) (Brennan, J., plurality opinion); Old Dominion Branch No. 496, Nat'l Ass’n of Letter Carriers v. Austin, supra, 418 U.S. at 281, 94 S.Ct. at 2779, and that jury instructions that include ill will as an element of actual malice are “constitutionally insufficient.” Rosenblatt v. Baer, 383 U.S. 75, 84, 86 S.Ct. 669, 675, 15 L.Ed.2d 597 (1966). See Beckley Newspapers Corp. v. Hanks, supra, 389 U.S. at 82, 88 S.Ct. at 198; Henry v. Collins, 380 U.S. 356, 357, 85 S.Ct. 992, 993, 13 L.Ed.2d 892 (1965).
In the majority’s view, the force and reach of these emphatic holdings are limited to establishing that, although ill will alone does not establish actual malice, it is nonetheless probative of actual malice. In my judgment, such a view conflicts with the important principle articulated in Garrison — that debate will be inhibited if trials and liability may turn on the extent to which the speaker’s feelings toward the subject of the speech may have strayed from a fiercely guarded neutrality. The difference between the recognized chilling effect of the speaker’s “risk that it will be proved in court that he spoke out of hatred,” Garrison, supra, when the proof is equivalent to actual malice, and the effect of such risk when the proof is merely evidence of actual malice seems to me insubstantial. The powerful insight underlying Garrison and subsequent decisions suggests that ill will should not, in any circumstances, be considered probative of actual malice. See Ryan v. Brooks, 634 F.2d 726, 731 n. 4 (4th Cir.1980) (“malevolence, ill-will or spite * * * is irrelevant in First Amendment cases”).11
*134Even if ill will may be probative of actual malice in certain circumstances, however, I am convinced that the evidence which the majority claims to rely on in this case is utterly unable to provide any probative force for a finding of knowledge of falsity or reckless disregard of truth or falsity.
(a) The Post. The majority describes the Post as “a newspaper which seeks, among other things, hard-hitting investigative stories.” Maj. op. at 121. For the majority, this journalistic concern becomes the beginning of evidence of actual malice: “Regardless of whether one chooses to characterize this policy as conducive to ‘hard-hitting investigative journalism,’ or * * * ‘sophisticated muckraking,’ it certainly is relevant to the inquiry of whether a newspaper’s employees acted in reckless disregard of whether a statement is false or not.” Id. at 121.
This conclusion represents a sharp departure from the principles of free and vigorous discussion that have been the touchstone of First Amendment jurisprudence. It is a conclusion fraught with the potential to shrink the First Amendment’s “majestic protection,” Bose, supra, 104 S.Ct. at 1961. In our society speech may be controversial and contentious; words may be intended to arouse, disturb, provoke, and upset. For a primary “function of free speech under our system of government is to invite dispute. It may indeed best serve its high purpose when it induces a condition of unrest, creates dissatisfaction with conditions as they are, or even stirs people to anger.” Terminiello v. Chicago, 337 U.S. 1, 4, 69 S.Ct. 894, 896, 93 L.Ed. 1131 (1949).
Muckraking — a term developed when writers like Lincoln Steffens, Ida Tarbell, and Upton Sinclair relentlessly exposed pervasive corruption12 — may be seen to serve that high purpose even if it offends and startles; so, too, hoping that readers will react in colorful terms to front page stories fits comfortably within the traditions of a free press and a nation that has nothing to fear from an aroused citizenry. As Justice Brandéis observed, “Like the course of the heavenly bodies, harmony in national life is a resultant of the struggle between contending forces.” Gilbert v. Minnesota, 254 U.S. 325, 338, 41 S.Ct. 125, 129, 65 L.Ed. 287 (1920) (Brandeis, J., dissenting).
Scant authority supports the majority’s remarkable innovation in First Amendment jurisprudence. The majority cites Chief Justice Warren’s concurring opinion in Curtis Publishing Co. v. Butts, 388 U.S. 130, 162, 87 S.Ct. 1975, 1995, 18 L.Ed.2d 1094 (1967) (Warren, C.J., concurring in result), and his reference to “ ‘sophisticated muckraking,’ ” id. at 169, 87 S.Ct. at 1999 (iquoting Saturday Evening Post announcement). This is a slim reed indeed to support the majority’s expansive construction. Chief Justice Warren’s comment is ambiguous and of dubious precedential value.13 No federal court has ever relied on *135muckraking as an indication of actual malice, and the Supreme Court has not even mentioned it since Butts.
In stark contrast to fundamental First Amendment principles and the clear and convincing evidence of actual malice required by Supreme Court cases from New York Times to Bose, the majority is prepared to give at least some probative weight to a newspaper’s emphasis on attention-getting stories. “Woodward, as editor, wanted from his reporters the same kind of stories on which he built his own reputation: high-impact investigative stories of wrongdoing. * * * From Woodward’s testimony it was proper for the jury to infer that the Post put some pressure on its reporters to come up with ‘holy * * * ’ stories.” Maj. op. at 120-21. Such is the basis for the majority’s arresting conclusion that the Post’s policy contributes to a finding that its “employees acted with reckless disregard of whether a statement is false or not.” Id. at 122. This deep hostility to an aggressive press is directly contrary to the mandates of the Supreme Court and the spirit of a free press. Indeed, the majority itself acknowledges that a central objective of New York Times is served by a newspaper “acting as watchdog for the public.” Maj. op. at 122 n. 39. Yet that recognition is flatly inconsistent with the majority’s holding that the Post’s policy of “hard-hitting investigative journalism” somehow helps to establish recklessness. Id. at 121. In my judgment, neither a newspaper’s muckraking policy nor its hard-hitting investigative journalism should ever be considered probative of actual malice.
(b) Tyler. Similarly, even if motive may be probative of actual malice, the facts in the record do not support the majority’s finding that Tyler revealed a motive that would represent an indication of actual malice.
Conceding that Tyler “had little reason for personal animus,” the majority concludes that Tyler wanted “to ‘get’ Mobil and the Tavoulareases * * * for the sake of a big story and because of [his] general bias against the oil industry.” Maj. op. at 118 (emphasis in original). The majority relies on four statements of Tyler and the fact that Mobil had publicly taken issue with a story Tyler had previously written. None of these bits of evidence is substantial enough to suggest a motive indicative of actual malice. Three of the four statements seem well within the scope of a working journalist’s vernacular. He speculated that the impact of the story might be to “knock off one of the seven sisters,” RE 789, he referred to his article in part as a “case against Tavoulareas,” RE 2489, and he characterized a session between Tavoulareas and Post editors as one in which the Post “blew [Tavoulareas] out of the water.” RE 857-858. These statements do reveal that Tyler, in conversation and in a memorandum, sometimes betrayed a less than Diogenes-like demeanor, but that is hardly sufficient to find the statements probative of actual malice. The fourth statement — that Tyler asked Piro about a source who could gain access to Tavoulareas’ safe — is more troubling. But however reprehensible a journalistic practice, such a request is in no sense indicative of a motive suggesting a knowledge of falsity or a reckless disregard of truth or falsity; indeed, if he was not joking it is evidence that Tyler sought documentary support, however objectionably. Finally, to hold that objections by a major company to a reporter’s coverage of it is probative of a motive suggesting actual malice serves to chill the kind of aggressive journalism that the First Amendment should foster and that the majority itself recognizes as “a vital service * * * protected by the New York Times actual malice standard.” Maj. op. at 121 n. 39. The record contains evidence that Tyler’s assignment to the piece was rooted in the fact that his area *136of expertise was the energy business, rather than any motive to “get" Tavoulareas and Mobil. RE 2157-2158.
(c) Golden. The majority also found that evidence indicating Golden’s “primary motive in working on the story was to get a byline on a Post story,” represents “evidence of motive [that] has considerable bearing on whether or not he acted with actual malice.” Maj. op. at 120 (emphasis in original). The question before this court is whether Golden acted with knowledge of falsity, or reckless disregard of truth or falsity. A desire to succeed and advance, in the journalistic or any profession, does not pose an inherent conflict with the requirements of professional rigor. Even if Golden was, as the majority asserts, “an ambitious young reporter looking for his big break,” id. at 120, that ambition seems to me completely irrelevant to the question of knowledge or reckless disregard of falsity; the majority’s holding places any reporter who reveals personal ambition in a more vulnerable position. This is not “the breathing space that gives life to the First Amendment.” Bose, supra, 104 S.Ct. at 1967.
(d) Piro. Piro’s animosity toward the Tavoulareas family is clear. To the extent that the majority focuses solely in this factor on Piro’s motivation, it is, of course, only relevant to the determination of Piro's actual malice. To the extent that the majority also means for this motivation to be relevant to the Post defendants’ state of mind, the majority seeks to count Piro’s motivation in its “cumulation” twice: as discussed below, the majority also counts it in its “reliance on sources” section, see text at 162 infra; maj. op. at 129.
2. The Post’s editorial decisions. The majority continues its reliance on unsupported- and impermissible factors when it offers two conclusions about the Post’s editorial policy as indicia of actual malice— that the Post failed to include certain information available to it and that the Post consistently resolved ambiguities in the light most damaging to Tavoulareas. As Judge Gasch found, according these conclusions probative weight conflicts sharply with the record in this case and with controlling Supreme Court precedent.
(a) Failure to include certain information. The majority rejects Judge Gasch’s conclusion that the supposed omission of certain information failed to “show that [Tyler] acted in reckless disregard of the truth,” 567 F.Supp. at 659. See maj. op. at 125. But the majority misconstrues the holding below, for it writes as though the court had held that suppression of information is never relevant to an actual malice inquiry. See id. at 125-26. Judge Gasch explicitly recognized that “if the defendants possessed information that showed that some of the article was incorrect, ignoring it would constitute actual malice.” 567 F.Supp. at 657. Judge Gasch carefully and properly examined the claimed omissions and determined that the disputed material had been substantially included; as a result, in this case, the claimed omissions were not probative of actual malice.
An independent review of the record confirms this conclusion. The majority maintains that three sets of facts were omitted: comments by Mobil outside director Lewis Lapham, information about the benefits of the SAMARCO-Atlas relationship, and internal Mobil memoranda. On the Lapham comments Judge Gasch found: “Despite the fact that the November 30 article did not contain this exact information, it did include at least three paragraphs that conveyed almost everything that Lapham had said.” 567 F.Supp. at 658. The differences that the majority points to, maj. op. at 123-24, are hardly substantial enough to be indicative of knowledge of falsity or reckless disregard of truth or falsity. The differences largely amount to whether the Post should have included a board member’s report of his belief in the Mobil chairman’s assurances about the SA-MARCO-Atlas relationship along with its report of the Mobil chairman’s assurances. To hold such a difference probative of actual malice — even probative “cumulatively”— not only is illogical but thrusts this court *137far too deeply into the nuances of editorial-discretion.
The majority’s reliance on the other two claimed omissions is similarly misplaced. The absence of information on the benefits of the SAMARCO transaction lacks any probative weight as an indication of actual malice. “The November 30 article simply did not relate whether the establishment of Atlas was a good or bad business decision.” 567 F.Supp. at 659. The majority emphasizes that the Post alluded to “ ‘millions of dollars in business’ ” and “ ‘exclusive, no-bid contracts,’ ” maj. op. at 124; both allusions were undeniably true. Similarly, the substance of the Mobil memoranda — the Mobil executive vice president’s announcement of arm’s length dealings with SA-MARCO and Tavoulareas’ statement that he would not be involved in SAMARCO-Atlas matters — were, as Judge Gasch found, “essentially summarizefd]” in the Post article. 567 F.Supp. at 659 n. 15. Instead of adding that the Mobil executive vice president wrote that dealings were arm’s length, the Post said only that the Mobil chairman told the Board about the arrangements and Tavoulareas’ non-involvement; instead of adding that Tavoulareas wrote the executive vice president to disassociate himself from SAMARCO-Atlas, it reported Tavoulareas’ statement that he had disassociated himself. Such variations seem well within the editor’s “inevitable] * * * set of choices,” Time, Inc. v. Pape, 401 U.S. 279, 286, 91 S.Ct. 633, 637, 28 L.Ed.2d 45 (1971), and it would greatly shrink the “constitutional zone of protection,” id. at 291, 91 S.Ct. at 640, afforded by the New York Times principles to hold such judgments even partially probative of actual malice.
(b) Resolution of ambiguities. Similarly, the argument that the Post consistently resolved ambiguities in the light most damaging to the plaintiff and that this process was at least partially probative of actual malice is supported neither by the governing precedents and principles nor by an independent examination of the record.
The majority premises its discussion about the resolution of ambiguities on its view that, “to the extent that the facts known to the defendants * * * were susceptible of more than one possible interpretation, the numerous instances in which the defendants selected the most damaging— and most sensation — alternative is some evidence which a jury, and an appellate court on review, can evaluate in determining whether publication was made with reckless disregard of whether it was false or not.” Maj. op. at 123 (emphasis in original).
It is a remarkable feat to reinterpret Time, Inc. v. Pape, supra, to establish that reasonable resolution of ambiguities may be probative of actual malice, so long as the resolution is not the sole piece of evidence. The majority accurately quotes the penultimate sentences from Pape, maj. op. at 122. Despite the majority’s italicization of one word, however, Pape contains no hint that reasonable resolution of ambiguities should have some probative weight in establishing knowledge of falsity or reckless disregard of truth or falsity. The majority’s reinterpretation undermines the central thrust of Pape — the deference to be accorded “possible rational interpretations” of a situation that “bristle[s] with ambiguities,” 401 U.S. at 290, 91 S.Ct. at 639,14 because of the “almost infinite variety of shadings” and “inevitable] * * * set of choices” with confront publishers. Id. at 286, 91 S.Ct. at 637.
Nor does invocation of cumulation provide the support that Pape does not. Indeed, in Bose the Court specifically refused to give probative weight to a rational interpretation of an ambiguous situation even in conjunction with other pieces of evidence: “adoption of the language chosen was ‘one of a number of possible rational interpretations’ of an event ‘that bristled with ambiguities' and descriptive challenges for the writer. * * * The choice of such lan*138guage, though reflecting a misconception, does not place the speech beyond the outer limits of the First Amendment’s broad protective umbrella.” Bose, supra, 104 S.Ct. at 1966 (quoting Pape). See also Ryan v. Brooks, supra, 634 F.2d at 733 (finding resolution of ambiguities in light most damaging to plaintiff not probative of actual malice even in conjunction with other pieces of evidence). Pape and Bose thus establish that rational interpretations of ambiguous situations do not contribute to a finding of actual malice.15
Whether or not the majority’s view is accepted, however, it is at least clear that some sustained consideration of the relevant interpretations and the underlying events to which they refer is required to evaluate the resolution of ambiguities. The majority makes no such serious effort to evaluate the interpretations at issue; indeed, it does not even identify them. Rather, the majority is content to assert that there was a “consistent selection of the inferences most damaging to the Tavoulareases,” maj. op. at 121 (emphasis in original), without specifying which inferences and what they were inferring. Such a blanket, unexplained assertion is far from the rigorous, independent evaluation of the record required by New York Times and subsequent decisions.16
3. The Checket conversation and the “dispatch” of Hoffmann. The majority cites a conflict between the Tyler and Checket accounts of the substance of their conversation as an indicium of actual malice. Maj. op. at 125-26. According to the majority, this conflict is probative of actual malice because the Checket conversation is the only source besides Piro for the article’s statement that Tavoulareas “personally dispatched” Hoffmann to Atlas. The majority concludes, “[I]t is clearly possible that Tyler deliberately falsified his notes.” Id. at 126. The majority thus rejects Judge Gasch’s conclusion that “Tyler may * * * have been negligent in transcribing Checket’s statement but plaintiff introduced no evidence that would show that Tyler deliberately falsified this document.” 567 F.Supp. at 661 n. 18.
The majority’s emphasis on the Checket conversation is misguided. For the undisputed evidence in the record reveals that the Checket conversation — under either version — was far from Tyler’s sole source for the conclusion that Tavoulareas “personally dispatched” Hoffmann. As found by Judge Gasch, Tavoulareas “apparently participated to some extent in the discussions that led to Hoffmann’s temporary assignment at Atlas.” Id. at 661. The majority simply ignores this finding. Yet the evidence in the record fully supports *139Judge Gasch’s determination. Wolfe wrote to Tyler that “Mr. Tavoulareas played a minor role * * * in the arrangements made when G. Comnas departed from Atlas. Since it was on Mobil’s recommendation (including Mr. Tavoulareas’), because of long personal acquaintance with G. Comnas, that he negotiate with SAMARCO it was logical that Mr. Tavoulareas participate to the extent of assuring a settlement that was fair and equitable to both parties * * *. Since Mobil had originally recommended G. Comnas to SAMARCO, we felt an obligation to SAMARCO to maintain quality management of their operations. We made H. R. Hoffmann * * * available as an interim manager of Atlas.” RE 2345.17 In the face of this undisputed statement by Mobil — acknowledging that Tavoulareas played a role, albeit “minor,” in Comnas’ departure, and then shifting to “we” without distinguishing Tavoulareas— the majority’s emphasis on the Checket conversation is plainly mistaken.18
4. Other indicia of actual malice. The majority cites three “other indicia of actual malice” — the Post’s failure to retract, the lack of deadline pressure, and the knowledge of the harm the article would cause. All three are of questionable legal force; even if all three were valid indicia of actual malice, the evidence in the record does not support a finding that they were indicia of actual malice in this case.
In considering the Post’s refusal to retract the majority argues that “[t]he correct view” is that “ ‘[ujnder certain circumstances evidence [of a refusal by a publisher to retract a statement after it has been demonstrated to him to be both false and defamatory] ... might be relevant in showing recklessness at the time the statement was published.’ ” Maj. op. at 132 (brackets in original; citation omitted). The majority brushes aside the Supreme Court’s view in New York Times that a refusal to retract was not probative of actual malice by arguing that the Court’s use of the word “adequate” meant only that refusal to retract could not be the sole evidence. Id. The majority fails to mention, however, that the Court was considering the refusal to retract along with other claimed pieces of evidence. See New York Times, supra, 376 U.S. at 286-87, 84 S.Ct. at 729.19 Similarly, the majority does not mention that, in Bose, the Court did not even consider the publisher’s refusal to retract, which it had mentioned in its statement of facts, 104 S.Ct. at 1953, when it considered all of the possible facts which might lead to a determination of actual malice. Id. at 1965-67.
Even if the majority’s view is accepted, however, there is no evidence in the record to support the view that there had been the necessary demonstration to the Post of falsity and defamation. The majority does not pause to explain when and how such a demonstration — essential, under the majority’s formulation, if refusal to retract is to be considered an indicium of actual malice — was made. In addition, the Post reported Mobil’s response in two articles (on December 1 and December 4) as well as additional facts provided by Mobil (on December 7). RE 2559, 2561, 2562.
The majority also argues that the absence of deadline pressure was “evidence that the jury could fairly consider” because “[t]he defendants had ample time to search for corroboration of Comnas’ statements.” *140Maj. op. at 131. Even if such a relationship between deadlines and constitutional principles were established,20 the evidence in the record does not support a finding that the timing of this article was probative of actual malice. First, as will be discussed infra, Comnas’ statements had largely been corroborated by other sources. Second, Tyler’s investigative efforts over the course of the month that he spent researching this article establish considerable efforts to corroborate. Third, since Mobil gave no indication of further cooperation, it is difficult to see what further efforts Tyler should have undertaken. Fourth, despite the majority’s comment about the “[ajbsence of time pressure,” maj. op. at 131, undisputed evidence in the record suggests substantial time pressure — Tyler had learned both that a congressional committee would soon go public with the relevant information and that the New York Times was also investigating the story. RE 1100, 1126. Under the majority’s own terms, then, the timing of publication in this case does not amount to evidence of actual malice.
Finally, the majority alludes to “[knowledge of the harm likely to follow publication” as an indication of actual malice. Maj. op. at 131. In contrast, in a series of cases in which serious harm was obvious the Supreme Court has not even mentioned knowledge of harm as a possible factor probative of actual malice. See, e.g., St. Amant v, Thompson, 390 U.S. 727, 88 S.Ct. 1323, 20 L.Ed.2d 262 (1968) (allegation of corruption); Time, Inc. v. Pape, supra, 401 U.S. 279, 91 S.Ct. 633, 28 L.Ed.2d 45 (allegation of police brutality). Furthermore, if knowledge of harm has any probative weight it must be slight, for an inquiry into knowledge of harm diverts from the central inquiry into knowledge of falsity.
C. Potentially Probative Factors
The majority argues that we must consider the indicia of actual malice as cumulative circumstantial evidence, but since the indicia analyzed above are unsupported, either in the governing legal principles or in the record, cumulation of them is equally without weight. Our inquiry must focus, therefore, on the two potentially probative sources — the Peterson memorandum and the reliability of sources. These factors may be bricks, but they utterly fail to build the majority’s wall: they fail to establish, by clear and convincing evidence, knowledge of falsity or reckless disregard of truth or falsity.
1. The Peterson memorandum. Judge Gasch considered the Peterson memo at length. The majority, however, misconstrues his analysis of the inferences that could reasonably be drawn from it. Undertaking the required review of the record, Judge Gasch concluded that “[tjhere is no evidence whatsoever to show that Ms. Peterson had any ‘inside’ information about Tyler’s sources or research that would enable her to know, or even suspect, that the story was inaccurate. She was merely expressing her uninformed opinion that she found the story hard to believe.” 567 F.Supp. at 655 (emphasis added). The majority fails to offer a shred of evidence which contradicts this finding. Instead, it seeks to emphasize the jury’s right to disbelieve Peterson’s testimony. But even if Peterson’s testimony is disbelieved, the rejection of her testimony clearly fails to provide affirmative evidence that her opinion was informed, rather than uninformed. See Bose, supra, 104 S.Ct. at 1966 (“discredited testimony is not considered a sufficient basis for drawing a contrary conclusion”); Moore v. Chesapeake & Ohio R. Co., 340 U.S. 573, 576, 71 S.Ct. 428, 429, 95 L.Ed. 547 (1951).
There is no evidence in the record to support the view that Peterson’s opinion was informed. Indeed, the memo as a whole makes clear that Peterson had no knowledge of sources and research beyond what she had read in the article. RE 2486-2487. Furthermore, the undisputed evidence in the record does establish that Ty*141ler produced a point-by-point rebuttal to the Peterson memo for Greider, the senior editor ultimately responsible for the article. RE 2487-2491. Under these circumstances the thrust of the trial court’s analysis was eminently correct; the evidence in the record did not support the view that Peterson’s memo proved the Post’s knowledge of falsity, or reckless disregard of truth or falsity. At most, it shows that the editor with ultimate decisionmaking responsibility had before him two paragraphs raising doubts from an editor — in whatever role— who had no independent knowledge of the subject or sources and the lengthy rebuttal, as well as the article itself, from a reporter who had previously written about energy issues and who had spent a month of research on the subject and sources. The inference of actual malice — or even a substantial indication of actual malice — that the majority seeks to draw is unwarranted.
To be sure, the majority does not quite claim that the Peterson memo itself provides clear and convincing proof of knowing or reckless falsity. Plainly, the memo could not bear such interpretation. Instead, the majority seeks to impute some additional weight to the memo through the coloration of other pieces of evidence. Maj. op. at 116. But, as discussed, the majority’s “other evidence,” id., consists largely of totally impermissible and insubstantial considerations — a newspaper’s reputation, a journalist’s ambition, the reasonable resolution of ambiguity, and the like. These inappropriate factors cannot bolster the Peterson memo and give it some added force; the memo must be evaluated without the distortion of such unacceptable considerations.
Since the evidence in the record does not support a view that Peterson’s opinion was informed, and since the evidence in the record establishes that Tyler and the Post editors engaged in far more extensive preparation and investigation than Peterson, the Peterson memo fails to assume the probative weight adopted by the majority and properly rejected below.
2. The reliability of sources. The majority correctly notes that “[rjeliance on the unsupported word of obviously biased sources for key allegations is evidence which a jury is entitled to consider in evaluating the existence of actual malice.” Maj. op. at 131 (emphasis in original). The validity of this general proposition is clear. See St. Amant v. Thompson, supra, 390 U.S. at 732, 88 S.Ct. at 1326. But in this case the evidence in the record does not support the necessary finding that the defendant relied on “the unsupported word of obviously biased sources for key allegations.” Id.
(a) Comnas. “[Rjeliance upon George Comnas,” Judge Gasch concluded, “* * * does not come close to approaching the level of recklessness required by the Supreme Court.” 567 F.Supp. at 656. The majority rejects this conclusion.
The majority argues that the evidence in the record supports a finding that Tyler had reason to doubt Comnas’ veracity. Relying primarily on Tyler’s awareness of allegations that Comnas left Atlas because of dissatisfaction over his business performance, it thus rejects Judge Gasch’s conclusion that such awareness did not necessarily constitute reliance on an obviously unreliable source. True to its method, the majority discards the evidence in the record mitigating the view that reliance on Comnas was an act of recklessness. As Tyler knew, Tavoulareas had told the SEC that “our people had every confidence in the world” in Comnas. RE 2446. Similarly, on Mobil’s behalf Paul Wolfe wrote Tyler of Tavoulareas’ “long personal acquaintance with G. Comnas” and “Mobil’s recommendation” of Comnas. RE 2345. Tyler had also read Comnas’ testimony to congressional investigators, which corresponded to his statements to Tyler. RE 2129-2131, 2467-2745. The majority maintains that Tyler must have known of rumors about Comnas being involved in scandals. However, except for Piro’s statement the majority is unable to point to a shred of evidence suggesting Tyler’s knowledge of these rumors. Instead, it asserts that Kousi must have told Tyler of rumors involving Com*142nas even though Kousi never offered such testimony. Maj. op. at 129. As Judge Gasch found, Tyler’s knowledge of Comnas’ background, placed in the context of the evidence in the record rather than the majority’s skewed presentation, fails to assume the great probative weight that the majority seeks. Cf. St. Amant v. Thompson, supra, 390 U.S. at 733, 88 S.Ct. at 1326 (reliance on a dissident union member’s comments about the union leadership did not amount to “evidence in the record of [the dissident’s] reputation for veracity”).
Critically, whatever Tyler’s knowledge of Comnas’ background, the majority finds an “absence of * * * corroboration” for Comnas’ statements. Maj. op. at 129. However, as Judge Gasch found, “[M]uch 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, the majority directs its attention to Comnas as the source of the Post statement that “Tavoulareas ‘set up’ Peter by ‘personally urging’ his inclusion in Atlas.” Maj. op. at 129. Yet the evidence in the record is clear that former SAMARCO director Kousi, whose credibility has not been challenged, told Tyler that the hiring of Peter was a “nepotistic act.” RE 1165, 1237, 1927-1928. The majority summarily and unjustifiably rejects this evidence because Kousi’s deposition contained “no supporting facts” for this conclusion. Maj. op. at 129 n. 45. Given Kousi’s position, knowledge, and undisputed veracity, his comment should be credited as corroboration of Comnas; it is at least a significant piece of evidence that compels greater attention and weight than the majority accords it. Furthermore, as noted, Comnas’ statement was also at least partially corroborated by Tavoulareas’ SEC testimony about his recruitment of Comnas and knowledge of Comnas’ business relationship with his son. RE 2424-2425. On behalf of Mobil, Wolfe also confirmed that Tavoulareas had played a major role in establishing SAMARCO and Atlas and in recruiting Comnas. RE 2344-2345. The record thus shows that Tyler corroborated the major pieces of information provided by Comnas. The use of questionable sources does not contribute to a finding of actual malice when the writer “used nothing from it that was not also found in his other sources.” Ryan v. Brooks, supra, 634 F.2d at 733.
(b) Piro. Judge Gasch found that Piro “was not a primary, or even a secondary, source for the November 30 article.” . 567 F.Supp. at 657. In contrast, the majority points to three statements for which it claims that Piro was an important source. The various sources for two of these statements — that Tavoulareas “personally dispatched” Hoffmann and that he “personally urged” Peter’s inclusion as a partner— have been discussed at length. The third statement — that Tavoulareas expressed an interest in “giving Peter a little nudge to get him along” — was specifically attributed to Piro; Piro’s estrangement from the family was clearly noted; and Tavoulareas’ denial was fully reported. 11 29. Moreover, the record is clear that Tyler chose not to use information provided by Piro which he could not corroborate from other sources. Under these circumstances, the quoting of Piro failed to establish a reliance on unreliable sources that carries significant probative weight in establishing actual malice by clear and convincing evidence.
Summary
Tyler and the Post editors had various sources of information before them supporting a belief in each of the challenged statements. Most of the factors cited by the majority are insubstantial or inappropriate; the Peterson memo and the reliance on sources simply do not establish clear and convincing evidence of knowledge or reckless disregard of falsity. In short, the majority’s “wall” is a house of cards that collapses under the rigorous scrutiny demanded by precedent and principle.
IV. Tavoulareas’ Status as a Limited Purpose Public Figure
The actual malice standard, of course, must be met only if Tavoulareas is a public *143figure, or for punitive damages if he is a private person. Gertz v. Robert Welch, Inc., 418 U.S. 323, 342, 350, 94 S.Ct. 2997, 3008, 3012, 41 L.Ed.2d 789 (1974). Gertz also established two types of public figures: general purpose, those of “general fame or notoriety in the community, and pervasive in the affairs of society,” id. at 352, 94 S.Ct. at 3013, and limited purpose, those who “have thrust themselves to the forefront of particular public controversies in order to influence the resolution of the issues involved.” Id. at 345, 94 S.Ct. at 3009.
Judge Gasch determined, on the basis of this circuit’s controlling opinion in Waldbaum v. Fairchild Publications, Inc., 627 F.2d 1287 (D.C.Cir.1980), that Tavoulareas was a limited purpose public figure. Though the majority finds it unnecessary to reach the issue because of its conclusion that the defendants acted with actual malice, it expresses reservations about this determination. See maj. op. at 103 n. 12. I believe that Judge Gasch’s analysis was eminently sound and fully supported by the evidence in the record.
Waldbaum sets forth the standards for determining public figure status. “[T]he general public figure is a rare creature,” 627 F.2d at 1292, because general public figure status must be supported by “ ‘clear evidence of general fame or notoriety in the community, and pervasive involvement
in the affairs of society.’ ” Id (quoting Gertz v. Robert Welch, Inc., supra, 418 U.S. at 352, 94 S.Ct. at 3013). Limited public figure status is determined by a three-prong test: (1) “the court must isolate a public controversy,” id. at 1296, (2) “the court * * * must analyze the plaintiff’s role in it,” id. at 1297, and (3) “the alleged defamation must have been germane to the plaintiff’s participation in the controversy.” Id. at 1298.
Judge Gasch determined that all three prongs of the Waldbaum test were met. For the first prong — the public controversy requirement — he concluded, “[T]he Court is convinced that defendants have established the existence of a public controversy over Mobil’s (and William Tavoulareas’) business dealings in Saudi Arabia and also over Mobil’s (and William Tavoulareas’) role in SA-MARCO and its business — through SA-MARCO — with Atlas prior bo the publication of The Post’s two articles on the subject.” Memorandum Opinion of District Court in Civil Action No. 80-3032 filed July 26, 1982 (Mem.Op.) at 7, RE 715 (emphasis in original). The evidence submitted by the defendant leaves no doubt that Tavoulareas eagerly sought and obtained the public spotlight to advocate views on relations with Saudi Arabia,21 and to feature Mobil’s efforts in Saudi Arabia, including the SA-MARCO relationship.22 ' This controversy— *144Mobil’s dealings with Saudi Arabia in general and SAMARCO in particular — thus clearly met the Waldbaum standard for establishing a public controversy. “[T]he issue was being debated publicly and * * * it had foreseeable and substantial ramifications for non-participants,” 627 F.2d at 1297 — as Tavoulareas repeatedly stressed in his public remarks. “In coming forward to play a prominent role in a heated controversy he rendered himself a public figure— someone who might expect public commentary both admiring and, as is not uncommon when passions run high and substantial interests are at stake, of a hurtful and perhaps unfair nature.” McBride v. Merrell Dow & Pharmaceuticals, Inc., supra, 717 F.2d at 1466.
The second and third prongs are sufficiently indisputable as to merit only brief discussion. The second prong concerns Tavoulareas’ role in the public controversy: “William Tavoulareas undeniably played an important role at least in the conception of Samarco as a joint American-Saudi Arabian Shipping concern and in the formulation of the idea for an independent ship management company in the nature of Atlas. He also is the chief architect of Mobil’s special relationship with the Saudi Arabians.” Mem. Op. at 7, RE 715. Similarly, the third prong is also clearly satisfied: “[T]he alleged defamation unquestionably concerned one aspect of the overall Mobil-Saudi Arabian relationship or, more narrowly, of the specific Mobil-Atlas-Samarco relationship.” Id.23
Furthermore, Tavoulareas’ leadership on a broad range of public issues is at least noteworthy. He has taken great pride in the role of public advocate and has urged others to follow his, and Mobil’s, lead.24 This leadership role has been especially conspicuous on energy issues.25 Such leadership is particularly significant because *145Tavoulareas commands media attention and space virtually at will. Indeed, Tavoulareas has relished Mobil’s extensive publicity operation, including its ability to purchase response space in major newspapers. See RE 393. He thus has the capacity for “self-help” that the Supreme Court has recognized as an important attribute of a public figure. Gertz v. Robert Welch, Inc., supra, 418 U.S. at 344, 94 S.Ct. at 3009.
It is instructive to compare Tavoulareas and Mobil with the plaintiff and his company in Waldbaum. The court judge Waldbaum to be a limited purpose public figure because Waldbaum “ ‘did not become merely a boardroom president whose vision was limited to the balance sheet. He became an activist, projecting his own image and that of the cooperative far beyond the dollars and cents aspects of marketing.’ ” 627 F.2d at 1300 (quoting District Court). On a far broader scale and with far more success, Tavoulareas also became an activist and projected an image far beyond the dollars and cents of oil marketing.26
V. Golden
I concur in the judgment affirming the j.n.o.v. with respect to Golden. I concur, however, not only because I share the publication concerns raised in the majority opinion, but also because I am convinced that the record does not contain clear and convincing evidence of actual malice on Golden’s part.
VI. Piro
Tavoulareas claims that three of Piro’s statements are false and defamatory: (1) that William Tavoulareas set up his son as a partner in Atlas; (2) that William Tavoulareas arranged for Peter’s first job with C.M. Lemos & Company; and (3) that Tavoulareas dispatched a senior executive to help run Atlas. Brief for appellant at 29.27
The majority again misconstrues Judge Gasch’s opinion. Judge Gasch, while noting that Piro poses “a more difficult question of the existence of actual malice” than the Post defendants, Memorandum-Order of District Court in Civil Action No. 80-2387 filed May 29, 1983, at 2 n. 2, RE 766, properly emphasized the lack of clear and convincing evidence of knowledge or reckless disregard of falsity.
In stating that Tavoulareas set up his son, Piro knew that Peter Tavoulareas had received a lucrative job in a concern that was involved in a business relationship with Mobil, which William Tavoulareas headed. RE 907. The majority ignores this knowledge. As a result the majority misconceives the meaning of Piro’s ignorance of the complexities of the Mobil-SA-MARCO-Atlas transactions. For the point is not that statements made in ignorance are automatically protected; rather, the point is that the ignorance, in conjunction with his knowledge, did not provide evi*146dence of knowledge or reckless disregard of falsity.
For the statement about Peter’s job, Piro knew that Lemos was a friend of the Tavoulareas family and that Peter obtained a job at his firm. RE 899. Indeed, in his SEC testimony Tavoulareas testified that he played no role in Peter getting the job, but conceded that Lemos “was a personal friend” and “must have known he was my son.” RE 2414. Though Piro did not know any further details, this ignorance, considered with his knowledge, does not provide the necessary affirmative evidence of actual malice.
Finally, the statement about dispatching an executive poses the hardest question, but it also fails to establish actual malice. Piro said that he overheard Tavoulareas telling Checket he had personally dispatched Hoffmann. RE 2089-2090. Checket testified that he did not recall this conversation. RE 1844. Yet even if Checket's testimony is believed and Piro’s disbelieved, there is no affirmative evidence that Piro knew of or recklessly disregarded falsity. As Bose stresses, 104 S.Ct. at 1966, disbelieving Piro’s testimony is not affirmative evidence of the contrary proposition. The record contains no clear and convincing evidence that Piro’s statement about the Checket conversation was based on actual malice, rather than on any number of possible other explanations— such as a misunderstanding or a mistake. Thus the third statement also fails to meet the high standard for actual malice emphasized in Bose. Though it is a far closer case for Piro than for the Post defendants, Tavoulareas has failed to prove by clear and convincing evidence that Piro’s statements were made with knowledge or reckless disregard of falsity.
VII. Conclusion
The central question is whether the Constitution’s guarantee of free speech protects the disputed statements published by the Post. Tyler knew that Tavoulareas personally recruited his son’s business associate to run a Mobil-conceived company and that Tavoulareas’ son ultimately landed a lucrative partnership at the company; he wrote that Tavoulareas “set up” his son. Tyler knew that Tavoulareas personally participated in the arrangements for a transition in leadership at the Mobil-conceived company; he wrote that Tavoulareas “personally dispatched” the Mobil executive who assumed command of the company. Tyler knew that the Mobil plans called for Mobil ships to be managed by the Mobil-conceived company after being leased to SAMARCO; he wrote that Mobil “provided” ships to the company. These statements emerged in the course' of an extensively researched 85-paragraph story, fully a fourth of which recounted Mobil’s version of the disputed events. I have no doubt that the Constitution protects these statements in the circumstances under which they were made.
This is an extremely important First Amendment case. If this excessive jury verdict on these mundane, flimsy facts is upheld, the effect on freedom of expression will be incalculable. The message to the media will be unmistakable — steer clear of unpleasant news stories and comments about interests like Mobil or pay the price. Thus we will have created a class of untouchables and abandoned our “profound national commitment to the principle that debate on public issues should be uninhibited, robust, and wide-open.” New York Times Co. v. Sullivan, supra, 376 U.S. at 270, 84 S.Ct. at 721.
I respectfully dissent.
. See, e.g., Cox v. Louisiana, 379 U.S. 536, 546-548 & n. 7, 85 S.Ct. 453, 459-460 & n. 7, 13 L.Ed.2d 471 (1965) (review of record leads Court to independent inferences about ambiguous events); Haynes v. Washington, 373 U.S. 503, 515-516, 83 S.Ct. 1336, 1344, 10 L.Ed.2d 513 (1963) ("we cannot avoid our responsibilities by permitting ourselves to be ‘completely bound by state court determination of any issue essential to decision of a claim of federal right, else federal law could be frustrated by distorted fact finding’ ”) (quoting Stein v. New York, 346 U.S. 156, 181, 73 S.Ct. 1077, 1091, 97 L.Ed.2d 1522 (1953)); Moore v. Michigan, 355 U.S. 155, 164, 78 S.Ct. 191, 196, 2 L.Ed.2d 167 (1957) (review of record leads to different "inference of fact’’); Pierre v. Louisiana, 306 U.S. 354, 358, 59 S.Ct. 536, 538, 83 L.Ed. 757 (1939) (in race discrimination cases Court has "solemn duty to make independent inquiry and determination of the disputed facts’’); Herndon v. Lowry, 301 U.S. 242, 249, 57 S.Ct. 732, 735, 81 L.Ed. 1066 (1937) (Court rejects “inference” that might save trial court verdict); Norris v. Alabama, 294 U.S. 587, 590, 55 S.Ct. 579, 580, 79 L.Ed. 1074 (1935) ("whenever a conclusion of law of a state court as to a federal right and findings of fact are so intermingled that the latter control the former, it is incumbent upon us to analyze the facts in order that the appropriate enforcement of the federal right may be assured"); Fiske v. Kansas, 274 U.S. 380, 385-386, 47 S.Ct. 655, 656-657, 71 L.Ed. 1108 (1927) (Court reaches own "inference” about IWW preamble; Court must review state fact-finding "where a conclusion of law as to a Federal right and a finding of fact are so intermingled as to make it necessary, in order to pass upon the Federal question, to analyze the facts”).
That many of these decisions arose in review of state court decisions is irrelevant to the principles of appellate review they enunciate: "surely it would pervert the concept of federalism for this Court to lay claim to a broader power of review over state court judgments than it exerrises in reviewing the judgments of intermediate federal courts.” Bose Corp. v. Consumers Union of United States, Inc., - U.S. -, 104 S.Ct. 1949, 1959, 80 L.Ed.2d 502 (1984).
. In Bose the Court rejected the notion that a different standard might apply to a jury trial. 104 S.Ct. at 1964, n. 27. Notably, two of the three Bose dissenters emphasized that their dissent was from the Court's conclusion about a bench trial. They conceded that a jury verdict of actual malice — as in the instant case — would compel independent appellate review. See id. at 1969 n. 2 (Rehnquist, J., dissenting) ("The fact-finding process engaged in by a jury rendering a general verdict is much less evidence to the naked eye and thus more suspect than the fact-finding process engaged in by a trial judge who makes written findings * * *.”).
Appellant’s argument that the j.n.o.v. violates the Seventh Amendment is meritless. The constitutionality of j.n.o.v. determinations has long been settled. See Neely v. Martin K. Eby Construction Co., 386 U.S. 317, 322, 87 S.Ct. 1072, 1076, 18 L.Ed.2d 75 (1967); Baltimore & Carolina Line v. Redman, 295 U.S. 654, 660, 55 S.Ct. 890, 893, 79 L.Ed.2d 1636 (1935).
. Bose repeatedly stressed the wide berth that must be accorded the actual malice inquiry in drawing the line between fact and mixed questions of fact and law. The Court emphasized that libel, like fighting words, incitement, obscenity, and child pornography, is an area in which "the limits of the unprotected category, as well as the unprotected character of particular communications, have been determined by the judicial evaluation of special facts that have been deemed to be of constitutional significance.” 104 S.Ct. at 1962. See also id. at 1964 n. 27 (“ ‘the simple fact is that First Amendment questions of “constitutional fact” compel this Court’s de novo review' ”) (quoting Rosenbloom v. Metromedia, Inc., 403 U.S. 29, 54, 91 S.Ct. *1291811, 1825, 29 L.Ed.2d 296 (1971) (Brennan, J., plurality opinion)).
. In contrast to the apparent ease of the majority's distinction between the "manner" and "standard" of evaluating evidence, the Bose Court recognized the "vexing nature” of the necessary distinction between questions of fact and mixed questions of fact and law. 104 S.Ct. at 1960. See also Pullman-Standard v. Swint, 456 U.S. 273, 288, 102 S.Ct. 1781, 1789, 72 L.Ed.2d 66 (1982) (commenting on the absence of "any * * rule or principle that will unerringly distinguish a factual finding from a legal conclusion”); Baumgartner v. United States, 322 U.S. 665, 671, 64 S.Ct. 1240, 1243, 88 L.Ed. 1525 (1944) (observing that the "recognized scope of appellate review is usually differentiated from review of ordinary questions of fact by being called review of a question of law, but that is often not an illuminating test and is never self-executing”).
. Despite the majority's assertion to the contrary, I do not assume that the District Court’s conclusions and findings "somehow control our review." Maj. op. at 109 n. 19. The references to Judge Gasch's conclusions and findings are, of course, references to them as findings of law. Unquestionably, we undertake an independent review, just as Judge Gasch did. The references to Judge Gasch's analysis are useful only because of the logic, force, and acuity of his insights.
.Even if Bose did not impose a special responsibility of independent review in this case, I might find that the majority’s approach violates longstanding principles for construing and evaluating inferences. "Inference is capable of bridging many gaps," Galloway v. United States, 319 U.S. 372, 386, 63 S.Ct. 1077, 1085, 87 L.Ed. 1458 (1943), but the inference must be reasonable and supported in the record. Id. See also Bechtel v. United States, 101 U.S. (11 Otto) 597, 600, 25 L.Ed. 1019 (1879) ("It would be a long stride in dialectics, and one we are not prepared to take, from this fact to the inference * * *.”). In relying on inference "[sjpeculation cannot take over where the proofs fail.” In Re Sawyer, 360 U.S. 622, 628, 79 S.Ct. 1376, 1379, 3 L.Ed.2d 1473 (1959) (Brennan, J., plurality opinion). In "cumulating" inferences, moreover, a court should be especially vigilant in determining whether the inferences are% supported by the record. Cf. People v. Galbo, 218 N.Y. 283, 294, 112 N.E. 1041 (1916) (Cardozo, J.) ("We may multiply inferences at times, but, in multiplying them, we must not refine and rarefy them beyond measure.”).
. As the majority notes, Tavoulareas also points to securities law allusions as false and defamatory, but, since those allusions were not specifically pleaded, they are not before the court as possibly libellous statements. See maj. op. at 127; see also note 16 infra.
. The majority, of course, necessarily discusses and decides falsity as well as actual malice. In doing so, it ”assume[s] without deciding” that falsity, like the other elements of the constitutional inquiry, must be shown by clear and *131convincing proof. Maj. op. at 112. The majority's reticence is puzzling. Like actual malice, falsity must be shown by clear and convincing evidence. See Franklin & Bussell, The Plaintiff's Burden in Defamation: Awareness and Falsity, 25 Wm. & Mary L.Rev. 825, 864 (1984).
. I emphasize again that Judge Gasch’s opinion is instructive, not because of a legal principle of deference, but because of its analytical insight and clarity. See note 6 supra.
. Indeed, Justice Brennan, the author of the "actual malice” standard in New York Times Co. v. Sullivan, 376 U.S. 254, 84 S.Ct. 710, 11 L.Ed.2d 686 (1964), once suggested that jury instructions be framed strictly in terms of "knowing or reckless falsity" to avoid confusion of "actual malice” in the First Amendment sense with common law malice. Rosenbloom v. Metromedia, Inc., supra note 4, 403 U.S. at 52 n. 18, 91 S.Ct. at 1824 n. 18 (Brennan, J., plurality).
The majority’s contention that the question is analogous to the distinction between motive and intent in a murder prosecution, maj. op. at 118 n. 34, utterly misses the point. For the point in excluding ill will as an element of proof in actual malice is the Garrison principle that the admission of such proof chills free speech. It thus rests on important First Amendment concerns. No such principle bears in the murder analogy.
As authority the majority cites two decisions from other circuits, two state cases, and a commentator. Maj. op. at 117-118. One of the decisions from another circuit — Bose Corp. v. Consumers Union of United States, Inc., 692 F.2d 189, 196 (1st Cir.1982), aff’d, 466 U.S. 485, 104 S.Ct. 1949, 80 L.Ed.2d 502 (1984) — did not concern any allegation of ill will. To the extent that the decisions, and the commentary that recites them, conflict with the reading of Garrison set out above, I necessarily disagree with their interpretation and agree with Ryan v. Brooks, 634 F.2d 726 (4th Cir.1980). Furthermore, one of the most cited opinions of this circuit on actual malice emphasized the difference between common law malice and the requisite New York Times knowledge of falsity or reckless disregard of truth or falsity. "Malice, under the pre-Times practice, was equated with hostility, vindictiveness or negligent disregard *134of reputation. Under the Times test false statements made with these motives alone are not actionable; malic[e] may be shown only through knowledge of falsity or reckless disregard of truth or falsity." Washington Post Co. v. Keogh, 365 F.2d 965, 967 (D.C.Cir.1966), cert. denied, 385 U.S. 1011, 87 S.Ct. 708, 17 L.Ed.2d 548 (1967). Nothing in Keogh suggested that pre-Times malice should be considered probative of actual malice under Times.
The majority also cites Curtis Publishing Co. v. Butts, 388 U.S. 130, 87 S.Ct. 1975, 18 L.Ed.2d 1094 (1967). Butts relates to the muckraking point, which is discussed in detail in text at 153-55 infra.
. See S. Morrison. The Oxford History of the American Pf.ople 814 (1965) ("Stephens and his fellow writers (such as Ida Tarbell who showed up Standard Oil, Upton Sinclair of The Jungle fame, and Ray Stannard Baker) * * * muckraked to good purpose, exposing the evils of city and state governments, unions, business, the drug trade, and whatever was curably wrong in divers segments of American life.").
. Chief Justice Warren’s comment provides little support for several reasons. First, the Chief Justice was quoting the magazine’s self-description rather than setting forth a standard for evaluating probativeness of actual malice. Second, in reciting the Saturday Evening Post’s self-definition he did not explicitly find such a policy probative of actual malice, and he subsequently recounted other evidence clearly probative of actual malice. Third, whatever the meaning of his reference, it is unclear how many Justices shared his view. No other Justice joined that part of his opinion. Justice Harlan’s *135plurality opinion mentioned "sophisticated muckraking,” 388 U.S. at 158, 87 S.Ct. at 1993, but that analysis is structured by a standard of reasonable care, rather than actual malice. Justices Brennan and White agreed that Chief Justice Warren’s opinion presented evidence supporting a judgment of actual malice, but their agreement was in very general terms. Id. at 172, 87 S.Ct. at 2000 (Brennan, J„ concurring in result).
. Pape refers to a document that bristles with ambiguity, but Bose makes clear that Pape applies to ambiguous situations and events as well. 104 S.Ct. at 1966.
. The majority’s reliance on Rebozo v. Washington Post Co., 637 F.2d 375 (5th Cir.1981), is misplaced. Rebozo is ill-conceived, contrary to Pape and Bose, and of weak precedential value. In concluding that resolution of ambiguities may be probative of actual malice, Rebozo did not even cite, much less consider, Pape, the governing Supreme Court decision on the subject. Instead, it offered cites to pages in New York Times Co. v. Sullivan, supra note 11, and St. Amant v. Thompson, 390 U.S. 727, 88 S.Ct. 1323, 20 L.Ed.2d 262 (1968), that merely recite the actual malice standard, 637 F.2d at 382.
. The only interpretation to which the majority specifically alludes is the reference to securities laws. Maj. op. at 126-27. The majority apparently attaches some weight to this reference, but declines to elaborate on the significance this reference is supposed to assume within the majority's grand scheme of “cumulation.’’ Clearly, the reference adds nothing to a finding of knowing or reckless falsity. Virtually all of the sentences are completely accurate. The parties dispute whether the relevant securities regulation would have applied to Tavoulareas. It is undisputed, however, that a congressional hearing specifically addressing this question of securities law resulted in a debate between top SEC officials about whether full disclosure requirements would have attached even without the son living in the same household. See Securities Law and Corporate Disclosure Regulations, 1982: Hearings Before the Subcommittee on Oversight and Investigations of the House Committee on Energy and Commerce, 97th Cong., 2d Sess. 261-62 (1982). It is also undisputed that the Post reported Mobil’s claim that it had fully compiled with securities law. ¶ 11. The Post's description of a point of securities law so murky that top SEC officials could not agree on it hardly amounts to the significant role the majority seeks — evidence that contributes to a finding of clear and convincing proof of actual malice.
. At trial Wolfe also testified that Tavoulareas was present during the discussion about Comnas’ departure. RE 1385. Wolfe was ambiguous about whether Tavoulareas had actually participated in the decision to dispatch Hoffmann. RE 1390-1394.
. The majority claims that Tyler testified that Piro and Checket were his only sources for the "personally dispatched” allegation. Maj. op. at 125. Yet Tyler specifically pointed to Paul Wolfe’s confirmation as well. RE 2089. Furthermore, Mobil's response to Tyler is indisputably part of the record that is within the scope of our review, and the Post article itself reported much of the statement. ¶¶ 79, 84.
.Indeed, New York Times specifically left open the question whether a failure to retract might ever be probative of actual malice, and limited its holding to determining that a failure to retract under the circumstances of the case clearly was not probative of actual malice. 376 U.S. at 286-88, 84 S.Ct. at 729-30.
. Cf. R. Sack, Libel, Slander, and Related Problems 216 n. 175 (1980) ("Whether or not a communication is ‘hot news’ ought not to be overemphasized.’’).
. See, e.g., RE 268, 271 (Tavoulareas' remarks to National Society of Security Analysts, Nov. 24, 1975: "Now, I’d like to talk about Saudi Arabia. You may have wondered why Mobil is giving so much attention to that country. * * * Saudi Arabia is where the oil is."); RE 288, 300 (Tavoulareas’ remarks to Nomads Luncheon, Nov. 12, 1979: "All consuming governments should demonstrate appreciation to those moderate OPEC governments — particularly Saudi Arabia — that consistently play a restraining role in OPEC deliberations. This appreciation should take the form of public statements and preferential mutual-assistance programs."); RE 304, 309 (Tavoulareas’ testimony before Subcommittee on Multinationals, Senate Foreign Relations Committee, June 16, 1974: "A government negotiated arrangement is vulnerable to a weakness which is not present with private negotiations. * * * Aramco has been able to continue operating in Saudi Arabia during the period when exports from Saudi Arabia to the United States were totally embargoed."); RE 326, 334 (Tavoulareas’ testimony before Energy Subcommittee, Joint Economic Committee, June 2, 1976: "Generally, the Saudis have adopted more moderate positions on price than other OPEC members. * * * It is of vital importance to the interests of the United States and the rest of the free world that Saudi Arabia continue to feel justified in increasing production, even though it will thereby also continue to amass dollar surpluses."); RE 427, 429 (Tavoulareas quoted in Mobil press release, June 29, 1977: ”[B]y providing services to the Saudis through participation in their projects, we expect to secure long-term access to additional crude and products.").
. See, e.g., RE 274, 275-276 (Tavoulareas’ remarks to Security Analysts Seminar, Sept. 28, 1977: "Mobil has taken a broad range of steps to improve its access to [Saudi Arabia’s] vast reserves of hydrocarbons. * * * Mobil is a *14445% partner with private Saudi interests in SA-MARCO, a shipping company that transports crude and products under the Saudi flag.”); RE 382, 386 (Tavoulareas’ article in Columbia Journal of World Business, Summer 1977: "Mobil has shown a long-standing interest and commitment to Saudi Arabia through its involvement in Aramco and SAMARCO [and other projects]. Responding positively to Saudi Arabia’s development plans and objectives, Mobil is working on a variety of projects that will benefit both Saudi Arabia and Mobil as well as the free world. * * [SAMARCO and another marine venture] respond to Saudi Arabia’s desire to develop its own shipping industry.’’); RE 430, 433 (Tavoulareas quoted in Mobil press release. Sept. 28, 1977: "Our Saudi Arabian ventures are economically attractive, besides strengthening our identity in that country.’’).
. Having decided "not [to] reach the issue,” maj. op. at 103, the majority cannot resist offering, in wholly conclusory terms, an assumption that flies in the face of precedent and logic. The majority's apparent view is that the third prong of Waldbaum is probably not satisfied because the controversy over SAMARCO is not germane to Tavoulareas’ prominent public role as an outspoken advocate of Mobil’s special relationship with Saudi Arabia. Id. at 103, n.12. This is akin to arguing that a vigorous public advocate of, for instance, night flights into National Airport would not be a public figure for purposes of a controversy involving a particular night flight for which the advocate bore substantial responsibility. Clearly, in this case and in the night flights example, the specific incident bears a sufficient relationship to the particular controversy to satisfy the Waldbaum germaneness prong.
. See, e.g., RE 288, 302 (Tavoulareas’ remarks to Nomads Luncheon, Nov. 12, 1979: "I’d like to ask you to join us in an effort we have been making throughout the ’70’s to overcome the obstacles created by our government and special-interest groups. 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.”); RE 391, 393 (Tavoulareas’ article in St. John’s Alumni Quarterly, Spring 1979: "[W]e’ve come to rely on print ads to make our points, though we also try everything we can (like sending our executives out around the country). Still, it’s hard to make ourselves believed, or even heard. * * * After all our years of talking, there is finally evidence that, at last, somebody is listening. We’re cautiously encouraged because newspaper and broadcast editorials have begun to see merit in our arguments. * * * Washington may not be heeding the expert yet, but we’re hopeful." (emphasis in original)).
. See, e.g., Re 488-582 passim (articles reporting Tavoulareas’ views on oil price decontrol).
. In Waldbaum this court noted that, although "[b]eing an executive within a prominent and influential company does not by itself make one a public figure!,] * * * [s]ometimes position alone can make one a public figure.” 627 F.2d at 1299 & n. 36. The court elaborated:
The position itself may be so prominent that any occupant unavoidably enters the limelight and thus becomes generally known in the community — a general public figure. Similarly, the responsibilities of a position may include decisionmaking that affects significantly one or more public controversies, in which case the occupant becomes a limited public figure for those controversies. Courts should avoid generalizing, however, for label-ling certain positions as always being public is tantamount to making subject-matter classifications, forbidden under the case law. * * *
Id. at 1299-1230 n. 36 (emphasis added).
An analysis of this particular case suggests that Tavoulareas’ position — president of one of the world’s largest corporations whose decisions play a major role in national energy policy — is precisely one of such public power and prominence that the position itself confers at least limited public figure status with respect to energy issues. Given the public controversy about Mobil’s efforts in Saudi Arabia, including SA-MARCO, such a determination is unnecessary to this case.
. As the majority points out, maj. op. at 135, Tavoulareas’ complaint included a fourth Piro statement — the "little nudge” comment. Since Tavoulareas does not pursue this comment in his appeal, brief for appellant at 29, and since the context for this comment is similar to the context of the "set up" comment, I do not address the "little nudge” comment separately.