The Commercial Appeal of Memphis, Tennessee included the following item in its “People” column on September 8, 1972:
FLICKERING FLAME: Back in 1957 Anita Wood, who came from Jackson, Tenn., to Memphis to sing on TV, was Elvis Presley’s “No. 1 girl.” This week as Elvis closed his month-long show at the Las Vegas Hilton, Miss Wood stopped by the hotel for what appeared to be a “reunion” of two old friends. Elvis recently filed for divorce from his wife of five years, Priscilla. Miss Wood is divorced from former Ole Miss football star Johnny Brewer.
A photograph with the caption “Anita Wood” appeared near the top of the column. On April 25, 1973 John Brewer contacted the newspaper by letter alleging that the article was false and defamatory.1 On May 12, 1973 the newspaper published the following in its “People” column:
CORRECTION — On September 8 of last year an item in this column reported that the former Anita Wood, now Mrs. John Brewer, had been in Las Vegas, Nev., a few days previously and had stopped by the Las Vegas Hilton to visit Elvis Presley, an old friend, who was entertaining there. That apparently was a case of mistaken identity, because Mrs. Brewer says she was not in Las Vegas on or anywhere near that date. The item also said she and former Ole Miss football star Johnny Brewer were divorced, which was incorrect. The Commercial Appeal regrets the error.
John and Anita Brewer brought separate defamation actions against Memphis Publishing Company in United States District Court for the Southern District of Mississippi on June 23, 1973 and September 6, 1973, respectively. Jurisdiction was based on diversity2 and the suits were consolidated.
In 1974 the first jury found for the Brewers and awarded each plaintiff $400,000. *1241The district judge found the amount of the verdicts unconscionable and granted a new trial on damages alone. The second jury awarded Anita Brewer $250,000 and John Brewer $150,000; they accepted a remittitur to $100,000 and $50,000, respectively. The defendant appealed and plaintiff cross-appealed. This court reversed and remanded, holding that the finding of liability in the first trial, held before the decision in Gertz v. Robert Welch, Inc., 418 U.S. 323, 94 S.Ct. 2997, 41 L.Ed.2d 789 (1974), was infected by several errors: (1) the question of whether plaintiffs were public figures or private persons was never directly addressed; (2) the jury was instructed that defendant might be liable without fault and that plaintiffs might recover without showing any injury because the article was libelous per se and because the law presumes that damages flow from written publication of such a defamation. In addition to remanding for proper fault and damage instructions, the court stated:
The initial question is, of course, whether Anita Wood Brewer or John Brewer, or both, are public figures within the context of . Supreme Court cases. An important question relevant to this issue left open by the presently decided cases concerns the extent to which a person who has been a public figure in the past can retreat to private status and obtain the benefit of the lesser standard of proof required for damages for injury from libel.
Brewer v. Memphis Pub. Co., Inc., 538 F.2d 699, 703 (5th Cir. 1976).
The third trial also resulted in verdicts for the plaintiffs, $150,000 to Anita Brewer and $60,000 to John Brewer. On appeal defendant asserts: (1) under Mississippi law (a) the article did not defame either plaintiff; (b) if defamatory, the article was at most libel per quod requiring plaintiffs to plead and prove special damages with particularly which neither did; and (c) liability was permitted for simple negligence while state law requires proof of actual malice for libel liability; and (2) the first amendment (a) requires the trial court to find that both plaintiffs were public figures; and (b) dictates a finding that the evidence did not prove “actual malice” with convincing clarity.3
Ultimately our holding in this case is based upon federal constitutional issues. We reach those constitutional issues after first attempting to decide whether we can dispose of the appeal under state law. We should first seek such a resolution.4 Here we are required to predict how the Missis*1242sippi Supreme Court would have decided several issues. Our predictions are that as to each state issue the Mississippi Supreme Court would conclude that appellees, and each of them had valid claims. While this excursion into Mississippi law is a necessary prelude to our reaching the constitutional issues, we find it not determinative of the rights and liabilities of the parties. Of course, in the further development of the law of libel in the state of Mississippi its Supreme Court may decide quite differently from the predictions we make here.
State Law
After careful consideration of defendant’s arguments, we conclude, not without some difficulty,5 that both plaintiffs would be permitted to recover some amount6 in damages from the defendant under Mississippi libel law.7
Appellant argues that the article did not defame, either plaintiff. Anita Brewer *1243claimed in her complaint that by this article the defendant intended to convey that she was divorced and “openly involved in a relationship with a married man.” John Brewer claimed that the defendant intended to convey that he was divorced from his present wife and that his wife was openly involved in a relationship with a married man. Both asserted in their complaints that readers so understood the words of the article and that their reputations were thereby injured and that they suffered severe “shame, embarrassment, humiliation, mental anguish and emotional distress and strain” because of the article. Essentially, then, plaintiffs claim the article defamed them both by stating that they were divorced, defamed John by conveying that he had been cuckolded, and defamed Anita by conveying that she was involved in an immoral or illegal relationship.
Under Mississippi common law
*1244any written or printed language which tends to injure one’s reputation, and thereby expose him to public hatred, contempt or ridicule, degrade him in society, lessen him in public esteem or lower him in the confidence of the community is actionable per se.
Henry v. Collins, 253 Miss. 34, 158 So.2d 28 (1963), rev’d on other grounds, 380 U.S. 356, 85 S.Ct. 992, 13 L.Ed.2d 892 (1965); Natchez Times Pub. Co. v. Dunigan, 221 Miss. 320, 72 So.2d 681 (1954), Conroy v. Breland, 185 Miss. 787, 189 So. 814 (1939). We discuss the per se aspect of this definition below and view the remainder of the quotation as comprising Mississippi’s definition of libel. Appellant urges that defamation in Mississippi is limited to attacks on integrity and moral character (plus business reputation), citing a passage to that effect in Manasco v. Walley, 216 Miss. 614, 63 So.2d 91 (1953), and argues that a statement that a person is divorced or is cuckolded does not reflect on integrity or moral character and thus is not defamatory. Manasco, however, did not involve common law defamation but a statute requiring a newspaper to print the reply of a candidate for public office when it had earlier printed a story that reflected on the candidate’s honesty, integrity, or moral character. According to the Supreme Court of Mississippi, the legislature intended to limit the statute
to cases in which the editorial comment or the news story is defamatory, and defamatory in a particular manner, that is to say, where the editorial or news story reflects upon the honesty or integrity or moral character of the candidate, (emphasis added)
Id. at 627, 63 So.2d at 95. Both Natchez and Henry, supra were decided after Manasco and neither adopts the statute’s limitation to particular sorts of defamation.
While the parties cite no Mississippi case involving statements that a person is divorced or is a cuckold, and we have found none, such statements appear to fit within the state’s definition of defamation, i. e., such statements may well injure one’s reputation and thereby expose him to public contempt or ridicule, degrade him in society, or lessen him in public esteem, while not necessarily reflecting on his integrity or moral character. We also note that other courts have sustained defamation actions based on statements about marital discord. See, e. g., Gariepy v. Pearson, 207 F.2d 15 (D.C.Cir.1953), Thackrey v. Patterson, 157 F.2d 614 (D.C.Cir.1946). See generally Annot., 9 A.L.R. 1128 (1934).
Appellant asserts that John Brewer may not base his defamation claim on what was said about Anita Brewer. It is said that a general principle of libel law is that only the person defamed, the person to whom the article refers, has a cause of action; his or her relatives do not.8 It is also true, however, that certain defamations of one person also defame another who may not be named.9 In our view this is such a defamation. John Brewer did not assert his wife’s claim but, instead, claimed that this statement about her also defamed him.
Clearly, the defendant could not escape liability for printing such a statement by including the additional defamatory statement that John Brewer was divorced; such a rule of law would not only be facially inequitable but also would permit a newspaper to limit its potential liability for one *1245statement by adding a statement that a person is divorced even when it positively knew that was untrue.
In her complaint, Anita Brewer claimed that the article conveyed that she was “openly involved in a relationship with a married man.” On appeal she urges that the article “unmistakably meant that an affair between Elvis Presley and Anita Brewer had occurred at the Las Vegas Hilton Hotel.” Brief for Appellee at 20. Appellant argues that the article described an “entirely innocent and proper” reunion of two old friends. Brief for Appellant at 20. Under Mississippi law, the trial court must initially determine whether language is actionable, and if the language is “ambiguous, of doubtful import, or susceptible of two or more interpretations, its actionability must ordinarily be decided by the jury under appropriate instructions from the court.” Cameron Bros. v. Posey, 237 Miss. 432, 115 So.2d 138 (1959). That is, the court must decide whether the language is capable of the defamatory meaning asserted by the plaintiff; if so, the jury decides whether readers understood the article to mean that, id. Bearing in mind that a statement may be defamatory under Mississippi law although it does not attack the integrity or moral character of the plaintiff, we find that this article was susceptible to a defamatory meaning. Whether or not it indicated that Anita Brewer was having an affair with a married man as she argues, it is susceptible to the interpretation that her encounter with a married man was romantic.10
Appellant asserts that Mississippi defamation law distinguishes libel per quod from libel per se and requires a plaintiff to plead and prove special damages with particularity in the former case while presuming damages in the latter. According to appellant, under Mississippi law special damages include only material and pecuniary damages, not emotional injury, and plaintiffs neither pled nor proved special damages. Assuming that plaintiffs did not plead and prove special damages, we must determine whether Mississippi law would bar recovery. Appellant vigorously asserts that the article in question is at best libel per quod; appellee responds by arguing at length that it is libel per se. Appellant cites Holliday v. Maryland Casualty Co., 115 Miss. 56, 75 So. 764 (1917), and Barton v. Barnett, 226 F.Supp. 375 (N.D.Miss.1964), for the proposition that Mississippi law distinguishes between these two types of libel. Barton relies on Holliday as the sole source of Mississippi law for such a proposition. Holliday may bear such an interpretation, but three years after it ruled in Holliday, the Mississippi Supreme Court clearly stated that all libel is actionable per se:
It may be stated generally that any written or printed language which tends to injure one’s reputation, and thereby expose him to public hatred, contempt, or ridicule, degrade him in society, lessen him in public esteem, or lower him in the confidence of the community, is actionable per se .
Wrought Iron Range Co. v. Boltz, 123 Miss. 550, 86 So. 354 (1920). Subsequent cases have reaffirmed the view stated in Boltz. Heralds of Liberty v. Rankin, 130 Miss. 698, 94 So. 849 (1922); Hodges v. Cunningham, 160 Miss. 576, 135 So. 215 (1931); Conroy v. Breland, 185 Miss. 787, 189 So. 814 (1939) (also citing but not quoting Restatement (Second) of Torts § 569, which states:
Liability Without Proof of Special Harm-Libel
One who falsely publishes matter defamatory of another in such a manner as to make the publication a libel is subject to liability to the other although no special harm results from the publication;
Morehead v. United States Fidelity & Guaranty Co., 187 Miss. 55, 192 So. 300 (1939); Natchez Times Pub. Co. v. Danigan, 221 Miss. 320, 72 So.2d 681 (1954); Cameron Bros. v. Posey, 237 Miss. 432, 115 So.2d 138 (1959) (slander case).
*1246(Defamatory statements, to be actionable, do not have to be charged in a direct, positive and open manner. Indirect imputations and insinuations may suffice . . A distinction seems to be made between oral and written imputations . . It is frequently stated broadly that defamatory words, when spoken, are ordinarily not actionable per se unless they impute a crime, but that written or printed words are actionable when they subject the person to disgrace, ridicule, odium or contempt in the estimation of his friends and acquaintances or the public,
quoting 33 AmJur., Libel and Slander § 9); Henry v. Collins, 253 Miss. 34, 158 So.2d 28 (1963), rev’d on other grounds, 380 U.S. 356, 85 S.Ct. 992, 13 L.Ed.2d 892 (1965). Although in some cases the Court uses the phrase “libel per se,” it does not do so to distinguish that from libel per quod but, instead, as a way of stating that libel is actionable per se, that damages are presumed and the plaintiff need not plead and prove special damages with particularity if in fact the statements constitute a libel. The question of whether an article is libel per se is identical to the question we addressed above, whether the article is defamatory.
As this court noted in the earlier appeal of this case, Brewer v. Memphis Publishing Co., 538 F.2d 699, 702-03 (5th Cir. 1976), in Gertz v. Robert Welch, Inc., 418 U.S. 323, 94 S.Ct. 2997, 41 L.Ed.2d 789 (1974), the Supreme Court held that states may not permit recovery of presumed damages unless the plaintiff shows defendant’s knowledge of falsity or reckless disregard for the truth. (We find, infra, that plaintiffs have not made such a showing.) The Court in Gertz elaborated the constitutional overlay on the state law of damages in defamation actions:
It is necessary to restrict defamation plaintiffs who do not prove knowledge of falsity or reckless disregard for the truth to compensation for actual injury. We need not define “actual injury,” as trial courts have wide experience in framing appropriate jury instructions in tort actions. Suffice it to say that actual injury is not limited to out-of-pocket loss. Indeed, the more customary types of actual harm inflicted by defamatory falsehood include impairment of reputation and standing in the community, personal humiliation, and mental anguish and suffering. Of course, juries must be limited by appropriate instructions, and all awards must be supported by competent evidence concerning the injury, although there need be no evidence which assigns an actual dollar value to the injury.
In its instructions the trial court in this case followed the mandates of Gertz and the prior opinion of this court. We find that competent evidence concerning injury to reputation and personal humiliation, mental anguish, and suffering supported an award for damages in this case.11
Defendant asserts that before Gertz Mississippi law would permit recovery for defamation only if the defendant acted with malice, though malice was presumed in libel cases. Defendant argues that Mississippi, responding to the ruling in Gertz that presumed fault is unconstitutional, would choose to retain the malice standard and require that the plaintiff prove malice; the state would not change its fault standard to negligence. Here the jury was instructed on a negligence standard which, according to defendant, was contrary to state law.
Mississippi courts appear to have followed the general common law of defamation, stating that malice is the gist of an action for libel or slander, see, e.g., Tipps Tool Co. v. Holifield, 218 Miss. 670, 67 So.2d 609 (1953), but that malice is presumed where the words are actionable per se, see, e.g., Travis v. Hunt, 224 Miss. 193, 79 So.2d 734 (1955). As the prior panel of this court held, Gertz prohibits the states from allow*1247ing liability without fault, from “presuming” fault. Brewer, supra at 702. The overwhelming majority of state courts responding to this mandate of Gertz have adopted negligence as the rule of liability. See Memphis Pub. Co. v. Nichols, 569 S.W.2d 412 (Tenn.1978), and cases cited therein and Note, State Court Reactions to Gertz v. Robert Welch, Inc., 29 Vand.L.Rev. 1431 (1976), and cases cited therein. We reject appellant’s contention that following Gertz Mississippi would require a plaintiff in a libel action to prove malice; we agree with appellee that Mississippi would adopt a negligence standard.12
Constitutional Law
The trial judge submitted to the jury the issues of whether either plaintiff, or both, was a public figure at the time the article appeared and, if so, whether the evidence showed with convincing clarity either that the defendant acted with reckless disregard of whether the article was false or not or that the defendant had a high degree of awareness of its probable falsity. Responding to special interrogatories the jury found that neither plaintiff was a public figure on that date and that clear and convincing proof showed that the defendant published the article either knowing it to be false or with knowledge of its probable falsity.
It was for the trial judge, not the jury, to determine whether the evidence showed that either plaintiff was a public figure.13 After a review of the law and the record, we find for the reasons discussed below, that undisputed portions of the evidence required the trial judge to find both Anita and John Brewer public figures for *1248the purposes of this suit14 and that defendant’s knowledge of falsity or reckless disregard for the truth is not supported by clear and convincing evidence.
Exhibits and testimony showed that Anita Brewer was, at various and sometimes overlapping times, a well known entertainer locally (in the Jackson and Memphis areas), regionally (in the mid-South), and nationally. She also attained national fame through her relationship with Elvis Presley.
In the early to mid 1950’s Anita won talent and beauty contests and worked as a disc jockey on the “Antics of Anita” show with a radio station in Jackson, Tennessee and the “Anita Wood Show” in Memphis. In 1957 she won a “Hollywood Star Hunt” contest and was awarded a motion picture contract. The studio provided her with a talent agent. She never appeared in a motion picture. She did, however, perform and make promotional appearances in New Orleans and Chicago plus cities in Texas, Ohio, and Mississippi. As a vocalist she made several recordings on various labels. In addition, she appeared in plays and television commercials in Memphis. In the mid 1950’s she appeared on television first in Jackson and then in Memphis where she was co-hostess of a program for approximately one year. From 1962 to 1964 she sang weekly on another Memphis television show. Her nationwide television exposure consisted of two appearances on the “Jack Paar Tonight Show,” an appearance on Bert Parks’ “Country Fair,” four or five appearances on the “Andy Williams Show,” sometimes as a singer.
Anita testified that she dated Elvis Presley and that their relationship lasted for five or six years, until approximately 1960 or 1961. This relationship generated very extensive nationwide publicity, including articles in Time and Life magazines. Several articles quoted Presley as saying that she was his “number one girlfriend.”
In 1964 Anita Wood married John Brewer and, with the exception of possibly one television and one newspaper interview shortly thereafter, did not seek media attention or continue as a professional entertainer from that time to the filing of this suit.
We note that her relationship with Presley occurred during her active entertainment career and that publications after her marriage referred to her past relationship with him. There is some conflict in the evidence as to whether she ever arranged to be photographed with Presley, Record, Vol. I, at 190; Vol. IV, at 1054. Anita testified, “I don’t think anyone ever had to call anybody. They [photographers] automatically turned up.” One reporter testified that while Anita never asked him to write anything, “she was co-operative and she wasn’t unreluctant for me to write stories. She was ambitious,” Record, Vol. Ill, at 775, and that most of his stories about her were in connection with her relationship with Presley “but also often included matters about her talent as an individual herself,” id. at 770. The exhibits in this case clearly demonstrate that press coverage of her career was tied to coverage of her relationship with Presley. In fact, according to the articles, her success was due in large part to the relationship.
John Brewer was a member of the Ole Miss football team the year it was ranked number one in the nation. He was named to an All-SEC and at the time of trial still held an Ole Miss record. He played professional football with the Cleveland Browns from approximately 1960 to 1967 and was with the New Orleans Saints from 1968 until near the end of the 1970 season. In a published interview he was quoted as saying that his football career had made his name well-known enough to open business opportunities for him for the rest of his life; at trial he agreed he had made the statement. After retiring from professional football he: ran an advertisement in a New *1249Orleans paper for a week or two saying that he was selling real estate in Mississippi and inviting his New Orleans “friends” to view these properties; made repeated references to his successful sports career in his campaign literature when he ran for a seat in the Mississippi House of Representatives (he placed fourth of four candidates); opened a restaurant prominently displaying a large sign bearing the words “Johnny Brewer [sic] Catfish House” and containing pictures and memorabilia of his sports career (he also sought copies of his wife’s records for the restaurant’s jukebox); spoke at high school banquets where the press covered his appearances; and organized a school sports program and coached a team.
Appellant argues that plaintiffs in this suit were public figures who failed to prove that the article was published with “malice" as defined in New York Times. Appellees argue in the alternative that (1) according to the standards of Gertz v. Robert Welch, Inc., 418 U.S. 323, 94 S.Ct. 2997, 41 L.Ed.2d 789 (1974), Anita and John were not public figures for “all purposes,” even at the height of their careers; that they were not public figures for purposes of an article that dealt with their private lives not “public questions”; (2) even if plaintiffs once were public figures they had both retreated from the public spotlight and were no longer public figures, at least for purposes of an article that described supposed events that occurred during retirement and that bore on their private lives; and (3) clear and convincing evidence supports the jury finding of New York Times malice.
We must decide (1) whether Anita, as an entertainer and/or as Elvis Presley’s “number one girlfriend,” or John, as a football player and/or as Anita’s husband, was ever a public figure for purposes of this article; (2) if so, whether either plaintiff could and did cease to be a public figure for purposes of the article, and (3) if the plaintiffs were required to show malice by clear and convincing evidence, whether they did.
Plaintiffs argue that even at the height of their careers they were neither public figures “for all purposes” nor for any “limited range of issues,” the two types of public figures described in Gertz, supra at 351, 94 S.Ct. at 3012. According to plaintiffs, both types of public figures must have “assume[d] special prominence in the resolution of public questions,” id. According to them, neither had ever done so and, further, the article in question did not concern a matter of public interest or a public question, but rather, concerned their private lives. Defendant, focusing on different statements in Gertz, argues that both plaintiffs are public figures because they vigorously and successfully sought public attention. Defendant cites several cases where courts have found that entertainers and sports figures were public figures. We have reviewed several federal appellate court opinions and one state court opinion dealing with defamation of entertainers or sports figures and find that, for various reasons, none is especially helpful in determining whether Anita or John Brewer should be treated as a public figure.15 Our analysis of this issue, will, therefore, concentrate on *1250relevant Supreme Court opinions. The newspaper emphasizes the connection between this article and one cause of Anita’s fame, her relationship with Presley, and argues that the article therefore did not concern a purely private matter. Finally, defendant cites instances where the Supreme Court stated that courts should not undertake to distinguish between expressions of timely important ideas and other sorts of printed articles. Indeed each side finds some support for its position in opin*1251ions dealing with public figures. Bearing in mind the facts of the present case, we shall briefly trace the evolution of immunity for those who without “malice” print articles that defame public figures, to determine whether this defendant is entitled to that constitutional protection.
The Supreme Court in New York Times v. Sullivan, 376 U.S. 254, 84 S.Ct. 710, 11 L.Ed.2d 686 (1964), emphasized this nation’s profound historical commitment to robust debate on political issues, especially press criticism of the stewardship of public officials, and found that the constitution required a public official plaintiff to prove “malice” in a libel action in order to allow sufficient breathing space for such criticism.16
In 1967 the Court extended the malice standard to an action under a state right of privacy statute where the plaintiff was not a public official and where the article reported on a matter of public interest, Time, Inc. v. Hill, 385 U.S. 374, 87 S.Ct. 534, 17 L.Ed.2d 456 (1967),17 and to public figures in libel actions, Curtis Publishing Co. v. Butts, 388 U.S. 130, 87 S.Ct. 1975, 18 L.Ed.2d 1094 (1967).18 Hill did not involve defamation19 but it does contain broad statements that the first amendment protects more than reports on political and other important public issues; it protects
the vast range of published matter which exposes persons to public view, both private citizens and public officials. Exposure of the self to others in varying degrees is a concomitant of life in a civilized community. The risk of this exposure is an essential incident of life in a society which places a primary value on freedom of speech and of press. 385 U.S. at 388, 87 S.Ct. at 542.20
*1252The article in Hill reported on the opening of a new play that was linked to an incident involving the plaintiff. The Court had “no doubt that” the subject of the article was a matter of public interest, quoting Winters v. People of State of New York, 333 U.S. 507, 510, 68 S.Ct. 665, 667, 92 L.Ed. 840 (1948): “the line between the informing and the entertaining is too elusive for the protection of [freedom of the press].”
Butts is the only Supreme Court opinion dealing with sports and public figure status. The defendant published an article accusing Butts, then-Athletic Director at the University of Georgia, of fixing a football game. He was not a public official (the court noted that he was not employed by the state but by a private athletic association) but he was a public figure “involved in issues in which the public has a justified and important interest,” id. 388 U.S. at 134, 87 S.Ct. at 1980. The plaintiff in a companion case had attained public figure status by thrusting his personality into the vortex of an important public controversy. While “Butts may have achieved that status by position alone; both [plaintiffs] commanded continuing public interest and access to the media for counterargument.” Id. at 154-55, 87 S.Ct. at 1991. It is not clear, then, whether Butts’ connection with collegiate sports alone accounted for his public figure status.21
In Rosenbloom v. Metromedia, Inc., 403 U.S. 29, 91 S.Ct. 1811, 29 L.Ed.2d 296 (1971), the plurality opinion extended New York Times protection to all publications involving matters “of public or general interest,” even those that defame a private individual who did not voluntarily choose to become involved with the matter reported. Three years later in Gertz v. Robert Welch, Inc., 418 U.S. 323, 345-46, 94 S.Ct. 2997, 3009-10, 41 L.Ed.2d 789 (1974), the Court expressly rejected the Rosenbloom extension. The Court’s reasons for this shift are important for our purposes. The Court determined that the Rosenbloom “public or general interest” test “inadequately served both of the competing values at stake” the legitimate state interest in compensating individuals injured by defamation (and the private individual’s interest) and the interests of the press protected by the first amendment, specifically the interests of
a publisher or broadcaster of a defamatory error which a court deems unrelated to an issue of public or general interest,
id. at 346, 94 S.Ct. at 3010. Further, such a test
would occasion the additional difficulty of forcing state and federal judges to decide on an ad hoc basis which publications address issues of “general or public interest” and which do not — to determine, in the words of Mr. Justice Marshall, ‘what information is relevant to self-government.’ Rosenbloom v. Metromedia, Inc., 403 U.S. at 79, 91 S.Ct. at 1837. We doubt the wisdom of committing this task to the conscience of judges.
Id. We must, therefore, construe the Court’s definition of public figures in Gertz in light of the Court’s warning in the same opinion that the press is not adequately protected by a rule that allows a court to determine whether a published article is relevant to an issue of public or general *1253interest, and that judges ought not make such ad hoc determinations.
Gertz held that the New York Times standard applies to public officials and
[t]hose who by reason of the notoriety of their achievements or the vigor and success with which they seek the public’s attention, are properly classed as public figures,
Id. at 342, 94 S.Ct. at 3008. Both groups must prove “malice” because (1) they “usually enjoy greater access to channels of effective communication and hence have a more realistic opportunity to counteract false statements than private individuals normally enjoy;” because they are less vulnerable to injury, the state’s interest in protecting them is less; and (2) the more important normative consideration: those who seek public office must accept certain necessary consequences of involvement in public affairs, specifically the risk of closer public scrutiny including public interest in his or her personal attributes that ought touch on fitness for office. Similarly public figures, for the most part, act voluntarily.
Those classed as public figures stand in a similar position. Hypothetically, it may be possible for someone to become a public figure through no purposeful action of his own, but the instances of truly involuntary public figures must be exceedingly rare. For the most part, those who attain this status have assumed roles of especial prominence in the affairs of society. Some occupy positions of such persuasive power and influence that they are deemed public figures for all purposes. More commonly, those classed as public figures have thrust themselves to the forefront of particular public controversies in order to influence the resolution of the issues involved. In either event, they invite attention and comment. Id. at 345, 94 S.Ct. at 3009.
Having already determined that a case-by-case approach would lead to unpredictable results and uncertain expectations, and that it must devise broad rules of general application, id. at 343, 94 S.Ct. at 3008, the Court stated,
Even if the foregoing generalities do not obtain in every instance, the communications media are entitled to act on the assumption that public officials and pub- ' lie figures have voluntarily exposed themselves to increased risk of injury from defamatory falsehood concerning them .
whereas a private individual “has relinquished no part of his interest in the protection of his own good name,” id. In deciding that the plaintiff was not a public figure, the Court restated the definition:
That designation may rest on either of two alternative bases. In some instances an individual may achieve such persuasive fame or notoriety, that he becomes a public figure for all purposes and in all contexts. More commonly, an individual voluntarily injects himself or is drawn into a particular public controversy and thereby becomes a public figure for a limited range of issues. In either case such persons assume special prominence in the resolution of public questions. Id. at 351, 94 S.Ct. at 3012.
Some of the descriptions or definitions of the public figure category in Gertz clearly indicate that the plaintiffs in this case at least at the height of their fame, were in that category; other formulations, especially those referring to power, influence, the “affairs of society,” and public controversies, raise questions about whether entertainers, “girlfriends” of entertainers, or sports figures are properly so classified.
The evidence in this case shows that both plaintiffs at one time or another vigorously and successfully sought the public’s attention or gained notoriety for their own achievements. Both achieved “pervasive fame or notoriety,” at least regionally; whether it was “such pervasive fame or notoriety that he becomes a public figure for all purposes and in all contexts” (emphasis added) is a more difficult question. The Supreme Court in Gertz delineated two possible public figure categories, all-pur*1254pose22 and limited; the second, however, the Court described in terms of involvement in public controversies. A sports or entertainment career clearly does not necessarily involve any public controversy, nor would we so categorize a romantic relationship with a famous entertainer. In our view, however, the Court in Gertz did not define all subcategories of the public figure classification. Rather, after determining that the plaintiff in that case was not an all-purpose public figure, the Court set forth the more limited basis that might arguably have applied to that plaintiff, an attorney who had become involved in a local controversy over relations between police and minorities. We note that in describing two subcategories that Court used words like “some,” “more commonly,” “for the most part.” In describing the whole class of public figures the court included those who seek public attention or whose achievements gain notoriety, and commented that all public figures “invite attention and comment” and, later, that they “assume special prominence in the resolution of public questions.” To determine plaintiffs’ status based on the latter formulation, however, would require us. to decide what is and is not a public question, a task the Court has stated is inappropriate for judges, and to make an ad hoc determination when, according to the Court, such determinations provides too little protection for freedom of the press. We therefore focus instead on plaintiffs’ actions in seeking publicity or voluntarily engaging in activities that necessarily involve the risk of increased exposure and injury to reputation. Both entered professions that by their nature require public appearances and invite press attention. Anita’s relationship with Presley was not, and probably could not have been, kept out of the press. This relationship coincided with a portion of her career as an entertainer and it appears that during this time she invited press coverage, at least of her career. Further, as discussed above, her relationship with Presley advanced her career and it is clear from the exhibits that much press coverage focused both on the relationship and on her career, some items detailing a purported connection between these two aspects of her life.
The dual rationale for special treatment of public figures articulated in Gertz, access to the media for rebuttal and voluntary assumption of risk, was first set forth in Butts, supra, where Justice Harlan turned for guidance in setting a standard for public figures in libel actions to the considerations involved in devising the common law of torts in general and libel in particular, noting that in determining the degree of fault that will subject a defendant to liability, the common law looks to the importance of the defendant’s activities as well as to the plaintiff’s prior activities and his means for avoiding harm or mitigating damages:
In the cases we decide today none of the particular considerations involved in New York Times is present. These actions cannot be analogized to prosecutions for seditious libel. Neither plaintiff has any position in government which would permit a recovery by him to be viewed as a vindication of governmental policy. Neither was entitled to a special privilege protecting his utterances against accountability in libel. We are prompted, therefore, to seek guidance from the rules of liability which prevail in our society with respect to compensation of persons injured by the improper performance of a legitimate activity by another. Under *1255these rules, a departure from the kind of care society may expect from a reasonable man performing such activity leaves the actor open to a judicial shifting of loss. In defining these rules, and especially in formulating the standards for determining the degree of care to be expected in the circumstances, courts have consistently given much attention to the importance of defendants’ activities. Prosser, The Law of Torts § 31, at 151. The courts have also, especially in libel cases, investigated the plaintiff’s position to determine whether he has a legitimate call upon the court for protection in light of his prior activities and means of self-defense. See Brewer v. Hearst Publishing Co., 7 Cir., 185 F.2d 846; Flanagan v. Nicholson Publishing Co., 137 La. 588, 68 So. 964, L.R.A.1917E, 510. We note that the public interest in the circulation of the materials here involved, and the publisher’s interest in circulating them, is not less than that involved in New York Times. And both Butts and Walker commanded a substantial amount of independent public interest at the time of the publications; both, in our opinion, would have been labeled “public figures” under ordinary tort rules.
Butts, supra at 154. In our view, the plaintiffs in this case were, at least at some time and for purposes of some articles, public figures who, as such, must prove defendant’s “malice,” not because they are to be punished for having sought press attention but, rather, because the first amendment requires that the press be afforded the protection of the “malice” standard vis-a-vis those who have sought its coverage, either through direct invitation or by participating in activities whose success depends in large part on publicity. In performing its role the press covers not only political events and public controversies, but also sports and entertainment. Entertainers (and sports figures as entertainers) typically put more of themselves in the public view than their particular performances; both plaintiffs in this case certainly did. Anita was in the spotlight not only because of her entertaining but also as the entertainer who was Presley’s “girlfriend.” John was portrayed in his campaign literature as one who had worked hard to become successful in football and would work hard if elected. As the Court of Appeals for the District of Columbia recently stated, public figures must prove malice in defamation actions in order to give the media “breathing space . in fulfilling its role of reporting, analyzing, and commenting on well-known persons” who have “invit[ed] attention and comment.” Waldbaum v. Fairchild Publications, Inc., 627 F.2d 1287, at 1291-1292 (D.C.Cir. 1980).
Gertz, then, indicates that these plaintiffs could be classified as public figures at least at the height of their fame and at least for purposes of some articles. We must still determine whether any later Supreme Court opinion modifies this determination, and whether the defendant in this case is entitled to New York Times, “malice” protection for this article published when Anita was no longer a professional entertainer or romantically involved with Presley and when John was no longer playing professional football.
In Time, Inc. v. Firestone, 424 U.S. 448, 96 S.Ct. 958, 47 L.Ed.2d 154 (1976), the Court held that plaintiff, the former wife of “the scion of one of America’s wealthier industrial families,” id. at 450, 96 S.Ct. at 963, was not a public figure. She had not “assume[d] any role of special prominence in the affairs of society, other than perhaps Palm Beach society” and had not “thrust herself to the forefront of any particular public controversy to influence the resolution of the issues involved in it,” id. at 453, 96 S.Ct. at 964, (her celebrated divorce proceeding did not constitute the sort of “public controversy” to which Gertz referred, id., at 454, 96 S.Ct. at 965). As we have discussed, the public figure status of plaintiffs in the present case does not rest on the public controversy prong of Gertz but, instead, on their fame. The Court in Firestone did not further discuss the fame of the plaintiff before the divorce proceedings *1256that were the subject of the article.23 Evidence in the present case of plaintiffs’ extensive public exposure and media coverage distinguish this from a case where a plaintiff’s fame may be limited to prominence in a local community.
In Wolston v. Reader's Digest Assoc., 443 U.S. 157, 99 S.Ct. 2701, 61 L.Ed.2d 450 (1979), the Court held that the plaintiff had not voluntarily thrust himself into the controversy over the espionage activities of the Soviet Union in the 1950’s by failing to respond to a grand jury subpoena and pleading guilty to a contempt charge. Neither the district court nor the appellate court for the District of Columbia, nor the defendant, had relied on any claim that the plaintiff was a public figure for all purposes. The Court noted that he had led a “thoroughly private existence” before the grand jury inquiry and “returned to a position of relative obscurity” after his sentencing, and that he
achieved no general fame or notoriety and assumed no role of special prominence in the affairs of society as a result of his contempt citation or because of his involvement in the investigation of Soviet espionage in 1958.
As detailed above, plaintiffs in this case were certainly not “thoroughly private” or “relatively obscure” people. Wolston, then, basically restated the Gertz position that overruled the Rosenbloom rationale and the facts in Wolston did not present an opportunity for much further comment on public figures who have not thrust themselves into public controversies to influence their resolution.
Both the district and appellate courts in Wolston had held that plaintiff was a public figure and had rejected his argument that, even if he once was, the passage of time had restored him to the status of private person for first amendment purposes. He abandoned this argument in the Supreme Court. For that reason and because the Court found he was never a public figure, it specifically did not decide “whether or when an individual who was once a public figure may lose that status by the passage of time,” id. at 166 n.7, 99 S.Ct. at 2707 n.7.24 Six Justices joined in the opinion of the Court; Justice Brennan dissented; and Justice Blackmun wrote an opinion concurring in the result, in which Justice Marshall joined, stating that assuming arguendo that plaintiff was a public figure in 1958 when he became involved in the espionage controversy, he had clearly lost that distinction by the time the article was published in 1974. According to Justice Blackmun, the passage of time will often be relevant to determining a plaintiff’s access to the means of counter argument and the risk of public scrutiny he may fairly be said to have assumed, the bases for public figure treatment articulated in Gertz. Justice Blackmun focused on the plaintiff’s conscious efforts to regain anonymity during the ensuing sixteen year period; these efforts “negated” any inference that he had invited press commentary by ignoring a subpoena in 1958. Id. at 170-71, 99 S.Ct. at 2709. Finally, Justice Blackmun stated that first amendment values would not be violated if historians, who have greater opportunities to investigate and time to reflect, are held to a higher standard of care than those *1257reporting on contemporaneous events. Id. at 171, 99 S.Ct. 2709. Even if we were to follow Justice Blackmun’s analysis, it would not necessarily apply to the present case where: plaintiffs’ public figure status is not tied to a particular public controversy; Anita Brewer may have sought anonymity eight years before the publication but her name continued to appear in articles and books about Presley; and the article was not an historical reflection but a newspaper article that purported to relate a contemporaneous event.25
The Fourth Circuit in Time, Inc. v. Johnston, 448 F.2d 378 (4th Cir. 1971), held that the plaintiff remained a public figure for purposes of an article that described events that had occurred twelve years earlier during his career as a professional basketball player. He had retired from professional ball nine years before the publication but had remained in organized ball until two years before the article appeared. At the time of publication he was a college coach. The article described the spectacular debut of Bill Russell, whose career was still phenomenal at the time of publication, and stated that Russell had psychologically destroyed the plaintiff’s playing. The court stated that the vast majority of courts had rejected arguments by plaintiffs in invasion of privacy cases that withdrawal or the passage of time insulated them from the New York Times rule for public figures.
The article in the present case similarly related to one whose career was still phenomenal, Presley. While it did not describe an event during plaintiffs’ active period, its subject did relate to one cause of Anita’s fame, her relationship with Presley. Plaintiffs urge this court to hold that this defendant is not protected by the New York Times standard for publishing an article about the private affairs of people who had effectively retreated from public figure status and which purported to describe an event that occurred after their retreat, not while they were in the public spotlight. We have determined, however, that the first amendment requires that this defendant be protected by the New York Times standard for publishing this article whose content relates to Anita’s fame as Presley’s “number one girlfriend.” It might be that during the “active” public figure period a wider range of articles, including those only peripherally related to the basis of the public figure’s fame, are protected by the malice standard and that the passage of time or intentional retreat narrows the range of articles so protected to those directly related to the basis for fame. Even under such a rule, and even viewing Anita as one who had retreated several years before this article, she would be required to prove malice in her suit based on this article. The article does not relate to John’s fame as a football player. To hold that he might therefore recover on a showing of negligence would, however, strip the required protection from the press to write such a story about Anita. To summarize, we find that Anita Wood Brewer who had gained media exposure and fame through her career and her romantic relationship with Presley, an extremely well-known entertainer, and whose name continued to appear in stories about Presley after her retirement, was required to prove malice in this suit based on an article that dealt primarily with that romantic relationship and incidentally with her marital status. Since, in our view, the first amendment requires “malice” protection for the press, which has repeatedly covered both these aspects of her life thereby advancing her career, in publishing this story, the same standard must be applied to John Brewer’s suit. Even where the defamatory portions of an article do not relate to the basis for one public figure’s fame, he may not, by marrying another public figure, *1258reduce the constitutional protection afforded the press to publish stories about his spouse.
We hold that both plaintiffs were required to present evidence that showed with convincing clarity a high degree of awareness by the defendant of the article’s falsity, or probable falsity.26 This court must make an independent examination of the whole record to determine whether plaintiffs made such a showing.27
The record shows that this article was prepared through the combined efforts of a reporter, James Kingsley, and the “People” column writer, Hope Pectol.28 On September 7,1972, Mr. Kingsley telephoned his sister Ann Otis who lived in Las Vegas to inquire about a family matter. During this conversation, Otis told her brother that during the previous weekend she had been in the dress shop of the Hilton International Hotel where a salesclerk told her that “a girl” had been in the shop, saying she was Anita Wood from the Memphis area and that she was there to see the Elvis Presley show. Otis testified that the clerk was her friend and that she had no reason to doubt the truth of the information this person presented. Otis had provided reliable information to her brother and to a reporter for another publication which had been the basis of published articles; she had done so without solicitation or compensation.
Kingsley had been with the Commercial Appeal in its Memphis office since 1960. He was a childhood acquaintance of Elvis Presley and had specialized in entertainment coverage, especially of Presley and including coverage of Anita Wood. Hope *1259Pectol had recently been encouraged to expand coverage of local and regional figures in the “People” column and to rely less on wire service information; she had relayed this interest to Kingsley. Kingsley told her that he had a story that she might want for her column. She asked him to bring her his notes. When he did, she located the paper’s files on Anita Wood and Presley which contained articles about the entertainment careers of each and about their relationship.. She noted that Kingsley had reported on Presley’s marital difficulties, asked him about this, and he confirmed that Presley had recently filed for a divorce. Based on the clippings she composed the “Flickering Flame” caption. She wrote the “reunion” portion based on Kingsley’s notes. Pectol then asked Kingsley if he knew whether Anita had ever been married. He told her that Anita had married John Brewer but that as far as he knew they were divorced and that he would check further on this and report back. She also asked him to call his sister to verify the original information.
Kingsley called his sister; while he waited, she telephoned the clerk on another phone and then confirmed the statement to him. Kingsley asked two people in his office about the Brewers’ marital status. One, who had written a book on John Brewer’s coach at Ole Miss, told him he did not know. Another, a photographer for the paper who had taken several pictures of Anita, told Kingsley that he had heard that the Brewers were separated. At trial Kingsley testified that he had thought that the Brewers were divorced. Although he could not identify the source of this information, one of his regular business and social acquaintances testified that she had spoken with John Brewer when he was separated from his wife and John had told her that he was separated from his wife and that they were getting divorced. Editors at the paper who reviewed the piece before publication, testified that they did not then suspect that it was false. They relied on Kingsley’s and Pectol’s past reliability, one adding that he had seen Pectol checking files and asking questions during preparation.
It does not appear, then, that Kingsley, Pectol, or those who reviewed the story after them entertained a serious doubt, or reasonably should have, about the story’s accuracy. Kingsley’s belief that the Brewers were divorced was not totally unfounded, though he could not later recall the specific communication upon which this belief was based. In fact, Anita Brewer had twice brought actions for separate maintenance from her husband in Louisiana and had once been granted a decree of separation there on the ground of abandonment. Kingsley did not testify that he had direct knowledge of this legal action but it is certainly likely that he obtained information on which to base his belief through covering Anita, working with a superior who had written two books about John Brewer’s college football team, or having recurring contacts with a person who was John Brewer’s friend. Further, the fact that Kingsley questioned two people in the office, one connected with John, one with Anita, about his belief, and that neither disabused him of this notion (and one confirmed that the couple had separated), does not indicate that he did not originally believe that the Brewers were divorced. Instead, it shows that he made an honest effort to test his belief.
Otis had proven to be a reliable source in the past. It is true that the salesclerk, not Otis, was the original information source for the piece and that she had not earlier provided the paper with information. Otis, however, knew and believed her and the record indicates no reason for disbelief; she did not, for example, contact the paper directly to provide the information and apparently had no “axe to grind” on the matter. We certainly ought not require of the press the degree of reliability that we must require of the police when several informer links provide them with information on which they seek a search warrant.
The record as a whole therefore fails to show at all, let alone with “convincing clarity,” that Kingsley, Pectol, or those who reviewed the piece after it was written, knew it was false or recklessly disregarded whether it was false.
*1260Because plaintiffs in this case were required and failed to show that the defendant acted with “malice” in printing this article, the verdict and judgments must be set aside and judgment entered here for the defendant.
REVERSED.
. Mississippi law requires such notification, at least ten days prior to bringing any civil action for publication, to afford a newspaper the opportunity to make corrections prior to suit. The statute further provides that if it appears . at trial that the article was published in good faith, that its falsity was due to an honest mistake of facts, that there were reasonable grounds for believing the statements in the article, and if the jury finds that the newspaper published a “full and fair correction, apology and retraction” within ten days of such notice, then the plaintiff shall recover “only actual damages.” Miss.Code Ann. § 95-1-5.
. Plaintiffs were Mississippi residents when they filed suit and when the article was published. In their complaints they alleged that *1241defendant, a foreign corporation, had its principal place of business in Tennessee but was “doing business” in Mississippi where service of process could be obtained upon its agent. Defendant waived any question of personal jurisdiction that may have existed, see Breckenridge v. Time, Inc., 253 Miss. 835, 179 So.2d 781 (1965), and Lee v. Memphis Pub. Co., 195 Miss. 264, 14 So.2d 351 (1943). In its answer it admitted for purposes of these suits only that it was doing business in Mississippi, that process could be served on its agent there, and that jurisdiction was vested in the district court. Appellant appeared in that court to defend these suits.
. Appellant also argues that (1) the article’s content did not make substantial damage to reputation apparent, that its contents did not warn a reasonably prudent editor of its defamatory potential and that the Supreme Court in dicta has indicated that in such cases even a private individual, i. e., one who is neither a public official nor a public figure, might not be permitted to recover; and (2) the awards here are constitutionally impermissible because they are not supported by competent evidence of actual injury and, instead, amount to presumed or penalty damages. Because of our resolution of other issues here presented, we need not and do not address the merits of these arguments.
. In a recent libel case, the Supreme Court reiterated the principle that “dispositive issues of statutory and local law are to be treated before reaching constitutional, issues” and interpreted a footnote in the opinion of the Court of Appeals for the District of Columbia that “cast substantial doubt on” an interpretation of local law that would have been dispositive as an indication of the view of the Court of Appeals that the interpretation was inaccurate and that, “therefore, the appeal could not be decided without reaching the constitutional question.” Wolston v. Reader’s Digest Assn., Inc., 443 U.S. 157, 160 n.2, 99 S.Ct. 2701, 2704 n.2, 61 L.Ed.2d 450 (1979). Because defendant *1242raises points of state law that, if correct would obviate the need for this court to reach the constitutional questions, we address them first.
We recognize that where local law would permit recovery in a libel case, it is not “dispositive.” In our view, however, the Supreme Court’s pronouncements in this area, e. g., Wolston, supra; Dillard v. Industrial Commission of Virginia, 416 U.S. 783, 784, 94 S.Ct. 2028, 2030, 40 L.Ed.2d 540 (1974); Alma Motor Co. v. Tinken-Detroit Axle Co., 329 U.S. 129, 136, 67 S.Ct. 231, 233, 91 L.Ed. 128 (1946); Siler v. Louisville & Nashville R. Co., 213 U.S. 175, 193, 29 S.Ct. 451, 455, 53 L.Ed. 753 (1909), require us at least to take a broad look at potentially dispositive state law issues to determine whether we must rule on constitutional issues. Judge Godbold’s special concurrence points up some of the futility, and peril to the state’s development of its jurisprudence, in our excursion into unresolved state law issues. The rights and liabilities of the parties are not, as it turns out, affected by our resolution, largely by prediction, of the questions of state law. Much of what we find it necessary to write in this opinion may be likened unto deciding whether or not a base runner touched third when it is clear that he was thrown out at home plate.
. This difficulty arises, in part, from the fact that we could locate no Mississippi case where a plaintiff claimed to be defamed by erroneous reports that he was divorced or cuckolded not any reported state decisions re-evaluating common law defamation in light of the damage and fault mandates of Gertz v. Robert Welch, Inc., 418 U.S. 323, 94 S.Ct. 2997, 41 L.Ed.2d 789 (1974), and its progeny. On August 1, 1980, well after oral argument in this case took place, the Supreme Court of Mississippi adopted a rule allowing the Supreme Court of the United States and federal circuit courts of appeals to certify questions and propositions of state law to the Supreme Court of Mississippi. Supreme Court Rule 46. This rule, however, limits certification to “questions or propositions of law of this state which are determinative of said cause independently of any other questions involved in said case” and thus appears not to allow certification of the state law questions raised here which could only, if resolved in defendant’s favor, pretermit our resolution constitutional issues; the state law issues would not determine the cause “independently of any other questions [here constitutional] involved.” The state law issues might be dispositive, but might not. We shall not impose upon the Mississippi Supreme Court, as our first acceptance of its welcome invitation to certify, such an “iffy” issue.
. Because of our resolution of the first amendment issues in this case, we need not decide whether the awards are excessive.
Defendant also argues that the verdict and award may not stand because the trial judge erred in excluding certain evidence, gave instructions that were erroneous in part and erred in overruling its motion for a mistrial based on the judge’s comments during trial. Because we find that the first amendment would not permit a verdict for plaintiffs in this case, we decline to examine these grounds which, if resolved in defendant’s favor, could only result in a pointless fourth trial.
. The district court sitting in Mississippi was required to apply Mississippi choice-of-law in the event of a conflict of laws. Defendant’s principal place of business was Tennessee and it is there that defendant’s allegedly faulty activities occurred. In their complaints plaintiffs did not particularize the location of publication (the place where the statement was received by a third person and, hence, where the injury to reputation occurred). Anita’s evidence consisted of the testimony of five witness, all Tennessee residents; they testified about their own reactions to the article and their observations of its effect on others in Tennessee. Anita testified that she received several telephone calls inquiring about the article, including one from a Mississippi resident who heard about the article from a relative in Tennessee. Both Anita and John testified that Anita became and remained upset over the article; this occurred in Mississippi. John testified that several coworkers (in Mississippi) asked him about the article and this humiliated and embarrassed him. He also stated that he received telephone calls from as far away as Arkansas. Two Ar*1243kansas residents testified that they read the article in that state and described their reactions and those of people in the Jonesboro, Arkansas area. John is a Mississippi native who lived in several other states during his professional football career and for a time thereafter. Anita is a Tennessee native and resided there until her marriage to John; while married she lived in several states before moving to Mississippi.
In Mitchell v. Craft, 211 So.2d 509 (1968), the Supreme Court of Mississippi referred to the Restatement (Second) of Conftict of Laws and to Leflar, Choice-Influencing Considerations in Conflicts Law, 41 N.Y.U.L.Rev. 267 (1966), for help in deciding a choice-of-law question. The latter work presents the more modern analytical approach and criticizes the Restatement (Second). Restatement § 150(a) states that the state of the plaintiff’s domicile will usually be the state with the most significant relationship to the occurrence and the parties, and, hence, the state whose law should be applied in the event of conflict, if the defamatory statement is published to third parties in more than one state including the state of plaintiffs domicile. Section 149, on the other hand, emphasizes the state where publication to a third party occurred (and not the state of plaintiffs domicile) when such publication is confined to one state. In Rose, Interstate Libel for the Future, 30 Hastings L.J. 1515 (1979), the author argues that the series of first amendment libel cases starting with New York Times Co. v. Sullivan, 376 U.S. 254, 84 S.Ct. 710, 11 L.Ed.2d 686 (1964), have focused on the defendant’s fault rather than the plaintiffs reputation, and thus call for focusing on the place of the defendant’s actions rather than on the place of publication to a third party.
Neither party raised this issue at any of the trials in this case or in appellate briefs and the district court apparently did not consider the issue. At oral argument, a member of this court asked appellant whether any thought had been given either to whether the “single publication rule” applied to this fact situation (appellant answered no) or to the application of Mississippi law, when the record showed all of about fourteen copies of the newspaper were distributed in the county of plaintiffs residence, instead of the law of Tennessee, the situs of the publisher’s operation. (Appellant responded that this had never been raised.) The paper’s managing editor testified that total circulation in Mississippi on September 8, 1972 was approximately 49,000 and that total circulation in the mid-South that day was approximately 220,000. Record, vol. IV, at 919.
We are mindful that we are to treat dispositive issues of local law before reaching constitutional issues. We are also acutely aware of the fact that this case is before this court for the second time following three trials, that the propriety of this court’s raising this issue sua sponte (and either deciding it or remanding for further proceedings that might result in yet a fourth trial and third appeal if Tennessee law conflicted with Mississippi law and if the former should be applied) is less than crystal clear. In Pierce v. Capital Cities Communications, Inc., 576 F.2d 495, (3d Cir. 1978), cert. denied, 439 U.S. 861, 99 S.Ct. 181, 58 L.Ed.2d 170, neither party had addressed the question of whether applicable state defamation law gave rise to a cause of action for the broadcast in question. At oral argument both agreed that the law of Pennsylvania governed; according to the court, that position
apparently rests on the fact that the business activities of the television station and the reporter charged with defamation are based in Philadelphia and, as a result, that jurisdiction has an abiding interest in the lawsuit. In the absence of any suggestion that there exists here a true conflict of law — that is, that Pennsylvania law differs from that of another interested forum having a strong competing stake in the case arguably outweighed Pennsylvania interest — there is no occasion to diverge from the parties’ position.
Id. at 501-02 (footnote omitted).
Here, similarly, there has been no suggestion that Tennessee and Mississippi law would conflict on any issue in the case; identification of such a conflict is, of course, the first step in a choice-of-law analysis. Further, because of our resolution of the first amendment issues, such a use of the courts’ and parties’ resources at this late date is not necessary.
. E. g., Wilson v. Retail Credit Co., 325 F.Supp. 460, 463 (S.D.Miss.1971); Restatement (Second) of Torts § 564 & Comment (e); 53 C.J.S. Libel and Slander §§11 and 145.
In Wilson, however, the court recognized that Mississippi law on this point was uncertain; no case directly addressed the issue. We have located no subsequent Mississippi cases treating this issue.
. Comment (e) to § 564 of the Restatement, supra, includes the following illustrations;
2. A calls B a cuckold in the presence of C. A has defamed B’s wife.
3. A states to B that C is an illegitimate child. A has defamed C’s mother.
The fact that the unnamed person in both examples is defamed through an imputation of unchastity, a traditional slander per se category, while John Brewer asserts that the present article defamed him by calling him a cuckold, does not affect the determination of whether the latter is a defamation.
. In Interstate v. Garnett, 154 Miss. 325, 122 So. 373 (1929), the court held that orally calling a woman a “bitch” constituted slander per se; in the circumstances of the case, the statement imputed unchastity.
. See the description, supra note 6, of the testimony of plaintiffs and other witnesses. We have expressly declined, however, to decide whether the evidence supported the awards in this case, supra note 5.
. We also reject appellant’s argument that under Mississippi law this article is qualifiedly privileged (in which case proof of malice would be required) as a news report about a public or quasi-public figure or a matter of legitimate public interest. This conditional privilege for “fair comment” is limited to criticism in the form of an opinion. Edmonds v. Delta Democrat Publishing Co., 230 Miss. 583, 93 So.2d 171 (1957). This article clearly made statements that purported to be facts, not opinions.
Appellant also argues that the trial court permitted plaintiffs to recover mental distress damages for negligence absent proof of any physical injury and thus contrary to Mississippi tort law which allows mental damages absent physical injury only for malicious or intentional torts. If we were to agree with appellant, essentially holding that while substituting negligence for presumed malice after Gertz the Mississippi courts would then import this tort law rule into libel for mental pain only, we would be required to remand for yet a fourth trial on damages because the jury did not indicate in its answers to the special interrogatories how much, if any, of this award was for mental pain. Such a remand would enable us to avoid ruling on the constitutional issues in the case only in the event that another jury were to find no reputation injury. Noting that several witnesses, see note 6 supra, testified about damage to plaintiffs’ reputations and the consequent unlikelihood that this jury’s verdict was, or a future jury’s verdict would be, based solely on emotional damages, and given our resolution of the constitutional issues, we decline to rule on this issue of Mississippi law.
. Defendant objected to submitting this issue to the jury. The district judge, however, stated that he interpreted the statement in the panel’s opinion in the first appeal of this case that
Defendant is entitled to have a jury, not the judge, resolve the liability and damage issues under proper instructions as to the law,
Brewer, supra at 703, as requiring him to submit this question for jury determination.
In an earlier case, a panel of this court declined to decide whether in all cases it was the duty of the trial judge to determine whether the plaintiff is a public figure; the panel did recognize, however, that the majority of courts have treated this question as one for the court and not the jury. Rosanova v. Playboy Enterprises, Inc., 580 F.2d 859, 861-62 (5th Cir. 1978), and cases cited therein. In Rosenblatt v. Baer, 383 U.S. 75, 76, 88, 86 S.Ct. 669, 671, 677, 15 L.Ed.2d 597 (1966), the Supreme Court remarked
as is the case with questions of privilege generally it is for the trial judge in the first instance to determine whether the proofs show respondent to be a ‘public official’15 Clearly the same considerations apply to determinations of public figure and public official status.
1 Harper & James, Torts § 5.29 p. 823 (3d ed. 1964), Restatement Torts, § 619. Such a course will both lessen the possibility that a jury will use the cloak of a general verdict to punish unpopular ideas or speakers, and assure an appellate court the record and findings required for review of constitutional decisions. Cf. Speiser v. Randall, 357 U.S. 513, 525, 78 S.Ct. 1332, 1341, 2 L.Ed.2d 1460; New York Times, 376 U.S. at 285, 84 S.Ct. at 728.
. The Court in Rosenblatt, supra note 12, remanded the case noting that the trial was held before New York Times v. Sullivan, 376 U.S. 254, 84 S.Ct. 710, 11 L.Ed.2d 686 (1964), therefore plaintiff did not shape his presentation to bring his claim outside the New York Times rule and the record left open the possibility that he could have adduced proofs to do so or to present a jury question of “malice,” 383 U.S. at 87, 86 S.Ct. at 676.
. Three such cases were decided after the Supreme Court’s ruling in Curtis Publishing Co. v. Butts, 388 U.S. 130, 87 S.Ct. 1975, 18 L.Ed.2d 1094 (1967), but before the Court’s decision in Gertz V. Robert Welch, Inc., 418 U.S. 323, 94 S.Ct. 2997, 41 L.Ed.2d 789 (1974). Cepeda v. Cowles Magazines, 392 F.2d 417, cert. denied, 393 U.S. 840, 89 S.Ct. 117, 21 L.Ed.2d 110 (9th Cir. 1968), held that a top major league baseball player was a public figure. According to the court, the Butts public figure definition (persons “involved in issues in which the public has a justified and important interest”) would “of course . . . include artists, athletes, business people, dilettantes, anyone who is famous or infamous because of who he is or what he has done,” id. at 419. We are not as confident that the Butts definition would “of course” include all members of the listed groups. The opinion does not include any discussion of how the court reached its conclusion and therefore does not aid our resolution of somewhat analogous issues raised by the Gertz definition. (See our identification and discussion of such issues in text at 24-29.) In Vandenburg v. Newsweek, 441 F.2d 378 (5th Cir. 1971), this court stated that “there is no dispute that” plaintiff, a track coach at the University of Texas, was a public figure but similarly did not discuss the point. The published article in that case referred to relations between the coach and black athletes; it is possi*1250ble that this would qualify as an important public issue to which Butts referred. In Time, Inc. v. Johnston, 448 F.2d 378, 380 (4th Cir. 1971), the court, citing Cepeda, supra, stated that there “can be no dispute that” plaintiff was a public figure when he was a member of the Philadelphia Warriors basketball team. By offering his services to the public as a paid performer he had, according to the court, assumed the risk of good or bad publicity “so far as it concerned his public performance.” The assumption of risk rationale does not appear to be linked to the Butts definition.
We have also reviewed four entertainer or sports figure cases decided after Gertz. In Vandenburg v. Newsweek, 507 F.2d 1024, 1025 n.1 (5th Cir. 1975), the court simply noted that in its earlier opinion Vandenburg, supra, where it remanded the case for a trial, it had decided that plaintiff was a public figure. The court stated that trial testimony supported the decision that he was a public figure, “at least for a ‘limited range of issues,’ ” citing Gertz, supra, at 351, 94 S.Ct. at 3012. The panel did not specify how it limited that range of issues; the range might have been limited to issues relating to track coaching or limited to include only race relations in the track program. The court does not address the statements in Gertz that may require the existence of a public controversy for limited issue public figures; on the facts of Vandenburg, however, such a controversy probably existed. In Carson v. Allied News Co., 529 F.2d 206, 209-10 (7th Cir. 1976), the court stated that Johnny Carson and his wife Joanna Holland had admitted that they were public figures. The opinion goes on to label Carson an “all-purpose” public figure. Earlier the court had stated that according to Gertz, “[s]ome persons occupy positions of such pervasive power and influence that they are deemed public figures for all purposes and in all contexts,” id at 209. In discussing Carson, the court referred to planitiffs’ brief, which descrioed Carson as an entertainer with an international reputation and “one of the more popular and outstanding practitioners of his profession,” and to their reply brief where they directly stated that both were public figures. [The opinion’s statement that the wife of such a public figure “more or less automatically becomes at least a part-time public figure herself,” id. at 210, quite possibly would not survive the ruling in Time, Inc. v. Firestone, 424 U.S. 448, 96 S.Ct. 958, 47 L.Ed.2d 154 (1976).] The court never directly states that Carson’s “power and influence” was “pervasive,” but possibly the opinion implies this. Chuy v. Philadelphia Eagles Football Club, 595 F.2d 1265, 1280 (3d Cir. en banc 1979), involved an article that discussed a professional football player’s physical condition, contractual dispute, and retirement. After noting that Gertz described two types of public figures, all-context and limited-issue, the opinion appears to use what it identifies as the basis for the former (“pervasive fame or notoriety”) to determine that plaintiff was a limited issue public figure, a “public figure, at least with respect to his ability to play football.” We find this analysis somewhat confused. Further, Chuy does not hold that professional football players as such are public figures for purposes of all articles but, instead, limits its holding to articles concerning playing ability and goes on to note the publicity surrounding the plaintiff’s contractual dispute. In James v. Gannett, 40 N.Y.2d 415, 386 N.Y.S.2d 871, 353 N.E.2d 834 (1976), the court identified the essential element of public figure status as taking “an affirmative step to attract public attention.” id 386 N.Y.S.2d at 876, 353 N.E.2d at 840. After quoting portions of the Gertz public figure definition, and citing Cepeda, supra, for the proposition that the category includes “without doubt” professional athletes, singers, and actors, id., 386 N.Y.S.2d at 875, 353 N.E.2d at 839, the court concludes that a professional belly dancer who performed in cabarets was a public figure with respect to accounts of her stage performances, id., 386 N.Y.S.2d at 876, 353 N.E.2d at 840. The opinion does not precisely state how this performer fits into either of the two Gertz categories.
In sum, the post-Gertz cases do not address what we view as certain problems of fitting entertainer and sports figures into the Gertz definition of public figures. Also, the published articles in Vandenburg, Chuy, and James commented on the public performances (or ability to perform) of the entertainer or sports figure. The defamatory portions of the article in the present case do not. The article in Carson concerned his personal life but (1) he admitted that he was a public figure and (2) the court may have found his international reputation indicated that his power and influence were sufficiently pervasive to qualify him as an all-purpose public figure. The Brewers’ power and influence never were as pervasive. The pre-Gertz cases could have, but did not, address aspects of the Butts definition that are similar to the problematic aspects of the Gertz definition.
. In his opinion concurring in the result, Justice Goldberg, joined by Justice Douglas, stated that the constitution would not protect defamatory statements about the private conduct of a public official as such defamation would have “little to do with the political ends of a self-governing society,” id. at 301-02, 84 S.Ct. at 731. But see Gertz v. Robert Welch, Inc., 418 U.S. 323, 344-45, 94 S.Ct. 2997, 3009, 41 L.Ed.2d 789, commenting on society’s interest in officials’ personal attributes that touch on fitness for office.
. It is not clear whether the “matters of public interest” rationale survived Gertz v. Robert Welch, Inc., 418 U.S. 323, 94 S.Ct. 2997, 41 L.Ed.2d 789 (1974), which rejected the “public issue” test of Rosenbloom v. Metromedia, 403 U.S. 29, 91 S.Ct. 1811, 29 L.Ed.2d 296 (1971).
. In Butts, Justice Harlan’s plurality opinion articulated a standard different from the New York Times “malice” standard; the plurality required public figures to show
highly unreasonable conduct constituting an extreme departure from the standard of investigation and reporting ordinarily adhered to by responsible publishers.
388 U.S. at 155, 87 S.Ct. at 1991. Chief Justice Warren’s concurrence urged the adoption of the New York Times standard. Id. at 163-64, 87 S.Ct. at 1995-96.
The Court has since stated, however, that Butts extended the New York Times “malice” standard to public figures. Wolston v. Reader’s Digest Assn., Inc., 443 U.S. 157, 163-64, 99 S.Ct. 2701, 2705-06, 61 L.Ed.2d 450 (1979). In Gertz v. Robert Welch, Inc., 418 U.S. 323, 336 & n.7, 94 S.Ct. 2997, 3005 & n.7, 41 L.Ed.2d 789 (1973), the Court stated that in Butts a majority of the Court agreed that the New York Times test should apply to criticism of public figures as well as public officials.
. The opinion makes specific reference to this, id. at 390-91, 87 S.Ct. at 543 and the Court in later opinions has stated that the opinion in Hill was limited to the consideration of nondefamatory matter. Curtis Publishing Co. v. Butts, 388 U.S. 130, 148 n.11, 87 S.Ct. 1975, 1988 n.11, 18 L.Ed.2d 1094 (1967); Gertz v. Robert Welch, Inc., 418 U.S. 323, 334 n. 6, 94 S.Ct. 2997, 3004 n.6, 4 L.Ed.2d 789 (1974).
. In this connection it may be that not only is such exposure a necessary incident of life in civilized society but that such reports make a valuable contribution to society and its members. Professor Richard Posner, in The Right of Privacy, 12 Ga.L.Rev. 393, 395-96 (1978), argues that gossip columns detailing the personal lives of successful or notorious people do not merely satisfy readers’ “idle curiosity” but, instead, are genuinely informational:
Gossip columns recount the personal lives of wealthy and successful people whose tastes and habits offer models — that is, yield information — to the ordinary person making consumption, career, and other decisions. The models are not always positive. The story of Howard Hughes, for example, is usually told as a morality play, warning of the pitfalls of success. Tales of the notorious and the criminal — of Profumo and of Leopold — have a similar function. Gossip columns open people’s eyes *1252to opportunities and dangers; they are genuinely informational.
. In his concurring opinion, Chief Justice Warren stated that citizens are often as concerned with public figures’ “views and actions with respect to public issues and events” as with those of public officials, id. at 162, 87 S.Ct. at 1995. In his view, persons outside government were increasingly
intimately involved in the resolution of important public questions or, by reasons of their fame, shape events in areas of concern to society at large . . . [the Court in Gertz, supra [418 U.S.] at 337 [94 S.Ct. at 3006], quoted this]. And surely as a class these “public figures” have as ready access as public officials” to mass media of communication, both to influence policy and to counter criticism of their views and activities. Our citizenry has a legitimate and substantial interest in the conduct of such persons, and freedom of the press to engage in uninhibited debate about their involvement in public issues and events is as crucial as it is in the case of “public officials.”
Id. at 164, 87 S.Ct. at 1996.
. The Court of Appeals for the District of Columbia recently described this category as including celebrities, those whose names are “household word[s]” if not nationwide, at least where they are defamed. Waldbaum v. Fairchild Publications, Inc., 627 F.2d 1287, at 1292, 1294, 1295 & n.22 (D.C.Cir. 1980). According to the court, the renouncement of anonymity or tolerance of publicity unavoidably carries with it the risk of media investigation of talent, character, and motives, id. at 1294. The court noted that well-known entertainers and athletes frequently endorse products and candidates, thereby indicating the breadth of their influence and the appropriateness of broad press scrutiny, id. at 1294 & n.15. The Court went on, however, to say that the test for all purpose public figures is strict, and found that the plaintiff there was a public figure because he had thrust himself into the public controversies concerning certain business issues, id. at 1299-1300.
. Justice Marshall in dissent did, and would find her a public figure, in part because she chose to become a member of a group “whose lives receive constant media attention,” id. at 486, 96 S.Ct. at 980.
. In Rosenblatt v. Baer, 383 U.S. 75, 86 S.Ct. 669, 15 L.Ed.2d 597 (1966), where the Court remanded the case for further proceedings to determine whether plaintiff was a public official, the Court noted that the plaintiff had not, and could not have, seriously contended that the fact that he had been discharged from his position as a commissioner six months before the article was published was significant; pub-
lie interest in the prior administration’s management of a particular project continued strong and the article, if it referred to the plaintiff at all, referred to his performance as a commissioner. The Court stated in dicta:
To be sure, there may be cases where a person is so far removed from a former position of authority that comment on the manner in which hi performed his responsibilities no longer has the interest necessary to justify the New York Times rule, id. at 87 n.14, 86 S.Ct. at 677 n.14,
but that this was not such a case.
. The district court’s analysis of this issue in Wolston v. Reader’s Digest Ass’n, Inc., 429 F.Supp. 167 (1977), focused on the need to apply the malice standard to authors and publishers of books as well as newspaper reporters. The Court of Appeals in Wolston v. Reader’s Digest Ass’n, Inc., 578 F.2d 427 (D.C.Cir.1978), focused on the fact that the subject matter of the article, Soviet espionage in 1958, continued to be a legitimate topic of debate when the article appeared particularly because the matter concerned national security. Neither discussion then, aids our analysis of the issues in this case.
. In New York Times Co. v. Sullivan, 376 U.S. 254, 280-81, 84 S.Ct. 710, 726, 11 L.Ed.2d 686 (1964), the Court held that a public official basing a libel action on falsehood relating to his official conduct must prove such statement was made
with ‘actual malice’ — that is, with knowledge that it was false or with reckless disregard of whether it was false or not.
The Court stated that the constitutional standard required that proof show actual malice with “convincing clarity,” id. at 285-86, 84 S.Ct. at 728-29, and found that the proof adduced failed to show an awareness of falsity or recklessness in that regard, id. at 286, 84 S.Ct. at 729. Failure to discover misstatements might show negligence but did not show actual malice; even failure to check its own files did not show that the state of mind required for actual malice was “brought home to” those responsible for publishing the advertisement. Id. at 287-88, 84 S.Ct. 710, 729-30.
In St. Amant v. Thompson, 390 U.S. 727, 731, 88 S.Ct. 1323, 1325, 20 L.Ed.2d 262 (1968), the Court held that New York Times malice must be shown by
sufficient evidence to permit the conclusion that the defendant in fact entertained serious doubts as to the truth of his publication.
Evidence showing that the defendant had no personal knowledge of plaintiffs activities, the record’s silence as to the source’s reputation for veracity, defendant’s failure to verify the information with those who might have known the facts, and his failure to consider whether the statement defamed plaintiff did not establish malice. Id. at 730, 88 S.Ct. at 1325. No evidence showed that the defendant was aware of the statement’s probable falsity and failure to investigate does not in itself establish malice. Id. at 732-33, 88 S.Ct. at 1326.
In Gertz v. Robert Welch, Inc., 418 U.S. 323, 332, 94 S.Ct. 2997, 3003, 41 L.Ed.2d 789 (1974), the Court, quoting St. Amant, emphasized that the defendant must act with a “high degree of awareness of . . probable falsity,” and repeated that proof of failure to investigate, without more, cannot establish New York Times recklessness.
In Herbert v. Lando, 441 U.S. 153, 99 S.Ct. 1635, 60 L.Ed.2d 115 (1979), the Court stated that a defamation plaintiff must prove defendant’s “knowing or reckless falsehood,” id. at 159, 99 S.Ct. at 1640, and that to be liable a defendant “must know or have reason to suspect that his publication is false,” id. at 160, 99 S.Ct. at 1641. The Court, however, did not announce any change in standard but, instead, cited the Gertz standard, so we do not interpret the latter statement as marking a departure from St. Amant or Gertz.
. See e. g., Vandenburg v. Newsweek, Inc., 507 F.2d 1024, 1027 & n.3 (5th Cir. 1975), and cases cited therein.
. The evidence included detailed testimony about the editorial processes at the paper, testimony from both the reporter and writer that up to the time of printing they believed the article to be true, testimony by Kingsley’s and Pectol’s superiors that both were reliable, and testimony from Kingsley and a reporter from another publication that Otis was a reliable source. Kingsley, Pectol, their superiors, and Otis testified and were subjected to cross examination. The requirements recently described in Herbert v. Lando, 441 U.S. 153, 99 S.Ct. 1635, 60 L.Ed.2d 115 (1979) have been meet.