Ida Maxwell "Maxie" Wells, who was a secretary at the Democratic National Committee (DNC) for a short time in 1972, filed a defamation action against G. Gordon Liddy stemming from his advocation of an alternative theory explaining the purpose of the June 17, 1972, Watergate break-in. During several public appearances and on a world wide web site Liddy stated that the burglars' objective during the Watergate break-in was to determine whether the Democrats possessed information embarrassing to John Dean.1 More specifically, Liddy asserted that the burglars were seeking a compromising photograph of Dean's fiance that was located in Wells's desk among several photographs that were used to offer prostitution services to out of-town guests.

Upon Liddy's motion for summary judgment, the district court determined that Wells was an involuntary public figure who could not prove actual malice by clear and convincing evidence. Additionally, the district court determined that Louisiana law applied to all of Wells's defamation counts and that Louisiana law would require even a private figure to prove actual malice. On the basis of these rulings, the district court entered judgment in Liddy's favor. Because we determine that Wells is not a public figure for purposes of the ongoing public debate regarding Watergate and we also conclude that Louisiana law does not apply to two of Wells's defamation counts, we reverse the district court's grant of summary judgment and remand for further proceedings consistent with this opinion.

In February of 1972, the then-twenty-three-year-old Wells moved from her hometown of Jackson, Mississippi to Washington, D.C. and began work at the DNC as the secretary to Spencer Oliver, Executive Director of the Association of State Democratic Chairmen. Wells continued in the employ of the DNC and Oliver until late July 1972. Throughout her employment, the DNC offices were located in the Watergate complex.

A few months after Wells started her job at the DNC, Frank Wills, a security guard, noticed a piece of tape propping open the door to the DNC offices while making his routine rounds during the early morning hours of June 17, 1972. See David Behrens, Day by Day, Newsday, June 17, 1992, at 63. Wills removed the tape. See id. When he made his next scheduled rounds, however, the tape had been returned to the doorway. See id. Suspecting that something was afoot, Wills called the police. See id. Shortly thereafter, the police arrived and apprehended five men: James W. McCord, Frank Sturgis, Eugenio R. Martinez, Virgilio R. Gonzalez, and Bernard L. Barker. See Alfred E. Lewis, Five Held in Plot to Bug Democrats' Office Here, Wash. Post, June 18, 1972, at A1. Of these five, one was a recent CIA retiree, three were Cuban emigres, and the fifth had trained Cuban exiles for possible guerrilla activity after the failed Bay of Pigs invasion. See id. The men were wearing business attire and surgical gloves. They were carrying $2,300 in sequentially numbered one hundred dollar bills, sophisticated electronic surveillance equipment, lock picks, door jimmies, one walkie-talkie, a short wave receiver, forty rolls of thirty five millimeter film, three pen-sized tear gas guns, see id., and the White House phone number of E. Howard Hunt.2 When initially asked about the events at the Watergate, White House spokesman Ronald Ziegler dismissed the incident as "A third-rate burglary attempt." Gaylord Shaw, Watergate Third Rate Burglary, Newsday, June 17, 1992, at 62.

In the wake of the burglary, the FBI determined that Spencer Oliver's telephone conversations were being electronically monitored from a listening post located in room 723 of the Howard Johnson's Motor Inn across the street from the Watergate. Because Wells often used Oliver's phone to make personal calls, some of her conversations were intercepted.3 Additionally, a drawer of Wells's desk was opened during the break-in. As a result, she was questioned by the FBI. Although there is some factual dispute between the parties over whether the FBI informed Wells of the discovery, the FBI also determined that a key found in a burglar's possession fit the lock on Wells's desk.

In September of 1972, Wells was subpoenaed to appear as a witness before the federal grand jury investigating the break-in. On September 15, 1972, the grand jury indicted the five burglars as well as the two men who allegedly had coordinated the break-in, E. Howard Hunt, a White House aide, and G. Gordon Liddy, counsel for the Committee to Reelect the President. Watergate Chronology, News & Observer (Raleigh, N.C.), June 17, 1992, at A4. Appearing before Judge John Sirica in United States District Court for the District of Columbia in early January of 1973, the five Watergate burglars pleaded guilty to a variety of burglary, conspiracy, and wiretapping charges. See John Berlau & Jennifer G. Hickey, List of Jailbirds is Long, but Sentences are Short, Insight Mag., June 23, 1997, at 10. Each of the five burglars was sentenced to a prison term.4 See A Watergate Scorecard, Wall St. J., Jan. 26, 1998, at A19. E. Howard Hunt also pleaded guilty to six counts of burglary, conspiracy, and wiretapping. See Berlau & Hickey, supra . As a result, he was imprisoned for thirty-three months. See A Watergate Scorecard,. Liddy neither pleaded guilty nor cooperated with the prosecution. He was tried on multiple counts of burglary, conspiracy, and interception of wire and oral communications, was found guilty, and received a sentence of six to twenty years imprisonment. See Berlau & Hickey, supra. Liddy served fifty-two months in jail as a result of his convictions. See Watergate Scorecard, supra .

Shortly after pleading guilty, James McCord wrote a letter from prison stating that he had been pressured to plead guilty and to lie during the district court proceedings relating to the Watergate incident. See Watergate Timeline, Cin. Enquirer, June 17, 1997, at A6. In his letter, McCord implicated John Dean, the president's counsel, and John Mitchell, the Attorney General, as the individuals who had been pressuring the Watergate burglars to withhold information. See Watergate TimeLine (visited April 29, 1999), < http://vcepolitics.com/wgate/timeline.htm >.

As a result of McCord's revelations implicating high level administration officials, in February of 1973 the United States Senate voted (77-0) to establish a Select Committee on Presidential Campaign Activities to be chaired by Senator Sam Ervin of North Carolina. See id. Wells, who had by this time relocated to Atlanta, Georgia, returned to Washington on June 20, 1974 to testify before the Committee. Wells's testimony was not part of the televised Watergate hearings. During its investigation, the Senate Committee discovered a campaign of political "dirty tricks" of which the Watergate break-in was a part. The White House's effort to cover up its involvement led to the imprisonment of several high ranking White House officials and ultimately to the resignation of President Nixon in August of 1974.

Wells returned to Washington in 1976 and served as a secretary to President Carter. After she left that post, she entered a Ph.D. program in English at Louisiana State University and at the time of this lawsuit planned to pursue a career as a college professor. Liddy was released from prison in 1977, and since that time he has become a successful radio talk show personality. He has also published his autobiography, Will, and is a frequent speaker on the lecture circuit.

In 1991, Len Colodny and Robert Gettlin authored a book entitled Silent Coup: The Removal of a President. Len Colodny & Robert Gettlin, Silent Coup: The Removal of a President (1991). In Silent Coup, Colodny and Gettlin discussed new evidence regarding the Watergate break-in and concluded that the purpose of the break-in was not simply to replace a malfunctioning listening device that had been installed in an earlier break-in at the DNC in May 1972.5 Rather, Colodny and Gettlin concluded that John Dean had personally authorized the Watergate break-in to protect his own reputation and the reputation of his now-wife, Maureen Biner.6

In Silent Coup, Colodny and Gettlin assert that an attorney, Phillip Mackin Bailley, assisted a woman named Erica L."Heidi" Rikan expand her preexisting call-girl operation located at the Columbia Plaza apartments, near the Watergate, by promoting Rikan's services to Bailley's DNC connections. The book also notes that Maureen Biner was a close friend of Rikan. According to Silent Coup, when Bailley came to visit the DNC, he asked for Spencer Oliver, but because Oliver was out of the office at the time, his secretary, Wells, gave him a tour of the DNC facilities. As a result of Bailley's contact with the DNC, Silent Coup reports that one client per day was referred to Rikan from DNC headquarters. Colodny and Gettlin state that meetings with call girls were arranged on Oliver's phone while he was out of the office, and that Oliver's telephone was the target of the first, May 1972, Watergate break-in during which the wiretaps were initially installed. According to Silent Coup Bailley was eventually arrested and indicted for violations of the Mann Act (transporting under-age females across state lines for immoral purposes), extortion, blackmail, pandering, and procuring. As a result, Bailley's address books were seized. Silent Coup also notes that Maureen Biner's name appeared in Bailley's address books.

After news of Bailley's arrest appeared in the newspaper, together with information regarding a Capitol Hill call-girl ring staffed by secretaries, office workers, and a White House secretary, Silent Coup reports that John Dean called the Assistant United States Attorney investigating the Bailley case and summoned him to the White House. During the meeting, Dean reportedly told the Assistant United States Attorney that he thought the Democrats had leaked the prostitution ring story. Thereafter, Dean made a photocopy of Bailley's address books and proceeded to compare the names from the book to a list of White House staff. Colodny and Gettlin state that Dean immediately would have recognized Maureen Biner's name as well as the alias of her good friend Rikan during this examination.

The implication of Colodny and Gettlin's narrative is that the June 17, 1972, Watergate break-in was ordered by Dean so that he could determine whether the Democrats had information linking Maureen Biner to the Bailley/Rikan call-girl ring and whether they planned to use such information to embarrass him. After the break-in was ordered, Alfred Baldwin, the man who was operating the listening post at the Howard Johnson's motel, visited DNC headquarters in order to "case" the layout of the offices. Because he posed as a friend of Oliver to gain admittance to the office, he was referred by the receptionist to Wells, who gave him a tour of the facility. During the visit, Silent Coup concludes "Baldwin either somehow obtained a key from Wells, or stole one." Colodny & Gettlin, supra at 149. Colodny and Gettlin contend that the purloined key was found on Watergate burglar Martinez. Although Silent Coup posits the question, "Why would a Watergate burglar have a key to Wells's desk in his possession and what items of possible interest to a Watergate burglar were maintained in Wells's locked desk drawer?" id. at 159, the book never proffers a specific answer.7

Liddy had extensive conversations with Colodny regarding the theory of the break-in promulgated in Silent Coup beginning in 1988. By 1991, Liddy had reached the conclusion that Colodny and Gettlin's theory was correct. As a result, in 1991 Liddy published a special paperback edition of his autobiography Will that included a discussion and endorsement of the Silent Coup theory. On June 3, 1991, Liddy had a meeting with Phillip Mackin Bailley, during which Bailley discussed his involvement with the Rikan prostitution ring. During the meeting, Bailley told Liddy that tasteful photographs of the Rikan call-girls wearing see-through negligees were kept in a desk at the DNC in the Oliver/Wells/Governors area. According to Bailley, various personnel at the DNC would show the photos to DNC visitors and would arrange rendezvous. Bailley also stated that several DNC employees were compensated for making referrals.

After he reissued his autobiography, Liddy began routinely incorporating Colodny and Gettlin's Silent Coup theory, including the additional information garnered from Bailley, into his public speeches. He would do so either by informing the listeners of the recent developments in the Watergate case as part of his prepared remarks or in response to questions raised by audience members during a question-and-answer period at the end of the program. Several of Liddy's public appearances during which he presented this theory are the subject of Wells's defamation suit. He delivered one such speech at James Madison University in Harrisonburg, Virginia on April 2, 1996 (JMU speech).

During the JMU speech, an audience member asked Liddy: Mr. Liddy, I have a question . . . I want your in put [sic] on one of the theories surrounding the mystery of Watergate. [I]t specifically related to James McCord. There are some who believe that maybe he wasn't working along with you, he had ulterior motives. And what gives . . . credit to this theory is that an ex-CIA agent . . . made two critical mistakes that really . . . caused you all to be caught. What do you think about that?

(J.A. at 996.) In response to the question, Liddy began to explain the Silent Coup theory of Watergate to the audience. During the explanation, he noted that the Howard Johnson's listening post "looked directly down at a desk of a secretary named Maxine Wells, and her telephone. And they had a telescopic lens camera pointed at that. And that is where the wiretap was subsequently found by the democrats on that phone." (J.A. at 998.) After explaining the Bailley/Rikan prostitution ring and Maureen Biner's connection to the ring, Liddy stated:

[S]ome members of the DNC were using the call girl ring as an asset to entertain visiting firemen. And to that end they had a manila envelope that you could open or close by wrapping a string around a wafer. And in that envelope were twelve photographs of an assortment of these girls and then one group photograph of them. And what you see is what you get. It was kept he said in that desk of Ida Maxine Wells. Thus, the camera [and] all the rest of it. And what they were doing is as these people would be looking at the brochure, if you want to call it that, and making the telephone call to arrange the assignation that was being wiretapped, recorded and photographed.

(J.A. at 998-99.) Liddy gave a similar speech while on a Mediterranean cruise (cruise ship speech) in August 1997.8

Liddy also discussed Watergate during an appearance on the Don and Mike Radio show on April 25, 1997. During the Don and Mike broadcast, Don's son Bart, who was doing research for a school project, asked Liddy questions about Watergate: Bart: I was wondering what was your role in the Watergate breakup [sic] scandal?

Liddy: Okay. I was the political intelligence chieftain, as well as the general counsel of the Committee to Reelect the President. . . . Now what I did not know is that John Dean did not trust me any more than I trusted him. And so my men were told, although I was not, that they were to go in there and, what, the telephone that was wired was not Mr. O'Brien's but was the telephone that was on the desk of a woman named . . . Ida Maxwell Wells . . . and she was the secretary to a man named R. Spencer Oliver. . . . .

Liddy: Well next door to the Watergate was a place called the Columbia Plaza Apartments and operating in there was what is known as a call girl ring and the lawyer who represented those girls was arrested by the FBI and they found his address book that had the names of his clients and also that included the call girl and there was a woman in there whose code name was "clout." . . . .

Liddy: Now to make a long story short. That was kind of what it was all about and if you want a secondary source on Watergate, you know to read about what was going on and everything. There is a book called Silent Coup .

(J.A. at 1021, 1022, 1023.)

The fact of Liddy's belief in the Colodny and Gettlin Watergate theory also appeared on the Accuracy in Media site on the world wide web9 in a review of the Oliver Stone directed film Nixon. Nixon (Cinergi, Hollywood Pictures, Illusion Entertainment 1995). The Accuracy in Media review criticized Stone for failing to seize an opportunity to adopt the Silent Coup theory of the Watergate break-in and for speculating that the Watergate burglars were looking for information linking Nixon to the Bay of Pigs invasion and the assassination of President Kennedy. In support of its argument that Stone should have pursued the Silent Coup theory, the web site characterized the theory as plausible and provided Liddy's explanation of the value of the Silent Coup theory:

Not until Colodny and Gettlin wrote Silent Coup did Liddy realize that the true objective of this second raid was to get into the desk of Maxie Wells, Spencer Oliver's secretary, said to be the key figure in arranging dates with the call girls. Unknown to Liddy at the time, one of the burglars carried a key to Wells'[s] desk.

(J.A. at 1016.)

Based upon the foregoing statements, Wells filed a defamation suit in the United States District Court for the District of Maryland on April 1, 1997.10 Wells asserted that Liddy defamed her by stating to public audiences on several occasions that she acted as a procurer of prostitutes for men who visited the DNC. Particularly, Wells asserted that Liddy defamed her during the JMU speech, during the cruise ship speech, on the Don and Mike Show, and in the Accuracy in Media web site.11 The complaint sought one million dollars in damages for injury to reputation, one million dollars in damages for mental suffering and three million dollars in punitive damages. Liddy filed his Answer on April 28, 1997, and the case proceeded to discovery.

Discovery did not progress smoothly, and Wells filed several motions to compel interrogatory answers and document production. The motions to compel asserted that Liddy had given inadequate answers to the interrogatories, failed to sign the interrogatory answers, and failed to provide a privilege log for the requested documents. Wells voluntarily dismissed the motions to compel based upon Liddy's interrogatory answers, and the district court held a hearing on Wells's remaining motion to compel document production by conference call on October 16, 1997. At the close of the hearing, the district court denied Wells's motion because the initial document request was overly broad and not narrowly tailored.12

Liddy filed a summary judgment motion on October 10, 1997. After various responses, replies and surreplies were filed, the summary judgment motion was ripe for disposition. After a hearing, the district court issued an opinion granting summary judgment to Liddy on April 13, 1998. See Wells v. Liddy, 1 F. Supp. 2d 532 (D. Md. 1998). In its opinion the district court first applied Maryland's lex loci delicti tort choice-of-law rule and concluded that the law of Louisiana, Wells's domicile, should apply to all of Wells's defamation claims. See id. at 536-37. The district court then reviewed each of the alleged defamatory statements to determine whether each one was capable of defamatory meaning under Louisiana law. See id. at 53739. After considering each of the statements as a whole, the context in which each was made, and the effect each would have upon the listener, see id. at 537 (citing Kosmitis v. Bailey, 685 So. 2d 1177, 1180 (La. Ct. App. 1996)), the district court held that only the JMU speech was capable of defamatory meaning, see id. at 537-38.

Next, the district court considered whether Wells was required to prove that Liddy acted with actual malice. See id. at 539. In evaluating that question, the district court examined Wells's situation under the standard established in Gertz v. Robert Welch, Inc., 418 U.S. 323 (1974), that, inter alia, established a First Amendment-driven public figure doctrine for determining which level of a defendant's culpability a defamation plaintiff must prove. After rejecting Liddy's contention that Wells had voluntarily injected herself into a public controversy, the district court concluded that Wells's participation in Watergate was such that she was one of the rare involuntary public figures envisioned by Justice Powell in Gertz . See Wells, 1 F. Supp. 2d at 540-41. Applying this reasoning, the district court held that Wells was required to prove actual malice. In the alternative, the district court examined Louisiana law and determined that Louisiana required all plaintiffs, whether public or private figures, to prove actual malice in defamation cases against a media defendant when the publication was on a matter of public concern. See id. at 541-42.

Turning to the application of the actual malice standard, the district court determined that Wells was unable to prove by clear and convincing evidence that Liddy knew that the information regarding her connection to a prostitution ring was false or that he recklessly disregarded the truth or falsity of the information. See id. at 542, 545. Specifically, the district court ruled that although Bailley, the sole source of Liddy's information that Wells had prostitution-related pictures in her desk, was extremely unreliable, there was sufficient factual verification of his information to preclude Wells, as a matter of law, from establishing that Liddy acted with actual malice. See id. at 543-45.

As a result of these rulings, the district court granted Liddy's summary judgment motion and entered judgment on his behalf. See id. at 545. Wells immediately filed a motion for reconsideration of the ruling under Rule 59 of the Federal Rules of Civil Procedure. (J.A. at 2076.) In the motion, Wells contended that she had projected sufficient evidence to raise a genuine issue of material fact on the actual malice issue, and that the district court had erred in granting summary judgment to Liddy. The district court considered the motion, and issued a memorandum opinion and order in which it confirmed the grant of summary judgment to Liddy. The district court noted that Wells had not pointed to error in its previous ruling, but rather merely raised the argument that the district court drew the wrong legal conclusion on the actual malice question. As a result, the district court denied Wells's motion to reconsider. Wells appealed.

On appeal Wells makes several assignments of error: (1) that the district court erred in applying Louisiana libel law to all counts of the complaint; (2) that the district court erred in denying her motion to compel document production; (3) that the district court erred in ruling that Wells was an involuntary public figure; (4) that the district court erred in ruling that Liddy's cruise ship speech, Don and Mike show statements, and Accuracy in Media web site statements were not capable of defamatory meaning; and (5) that the district court erred when it ruled that Wells failed to forecast sufficient evidence from which a reasonable jury could conclude that Liddy acted with actual malice when publicizing Wells's connection with the DNC prostitution ring during the JMU speech.

Summary judgment is appropriate when a party, who would bear the burden on the issue at trial, does not forecast evidence sufficient to establish an essential element of the case, see Laughlin v. Metropolitan Washington Airports Auth., 149 F.3d 253, 258 (4th Cir. 1998), such that "there is no genuine issue as to any material fact and the moving party is entitled to a judgment as a matter of law," Fed. R. Civ. P. 56(c). Viewing the facts in the light most favorable to the nonmoving party, we review a grant of summary judgment de novo. See Laughlin, 149 F.3d at 258. When, as here, the non-moving party must produce clear and convincing evidence to support its claim, that higher evidentiary burden is considered as part of the summary judgment calculus. See Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 244 (1986) (noting that the New York Times Co. v. Sullivan requirement of clear and convincing evidence of malice must be considered on a motion for summary judgment). Additionally, because defamation claims raise First Amendment issues, we have "an obligation to make an independent examination of the whole record in order to make sure that the judgment does not constitute a forbidden intrusion on the field of free expression." Milkovich v. Lorain Journal Co., 497 U.S. 1, 17 (1990) (internal quotation marks omitted). These principles guide our evaluation of Wells's claims.

An individual's interest in protecting his good reputation from being falsely impugned, the interest at the core of modern defamation law, has been carefully guarded from time immemorial. See Rodney A. Smolla, Law of Defamation § 1.01 (1998) (citing Exodus 20:16, "Thou shalt not bear false witness against thy neighbour"). The common law "has afforded a cause of action for damage to a person's reputation by the publication of false and defamatory statements" since the second half of the sixteenth century. Milkovich v. Lorain Journal Co., 497 U.S. 1, 11 (1990). Since the founding of the United States, state governments primarily have been responsible for developing defamation law adequate to protect citizens' reputational interests; "[t]he protection of private personality, like the protection of life itself, is left primarily to the individual States under the Ninth and Tenth Amendments." Rosenblatt v. Baer, 383 U.S. 75, 92 (1966) (Stewart, J., concurring) (cited with approval in Gertz v. Robert Welch, Inc., 418 U.S. 323, 341 (1974)). "Prior to 1964, the common law of defamation strongly favored the State's interest in preventing and redressing injuries to individuals' reputations, and the prevailing view gave little or no weight to First Amendment considerations." Foretich v. Capital Cities/ABC, Inc., 37 F.3d 1541, 1551 (4th Cir. 1994).

In 1964, however, the landmark decision New York Times Co. v. Sullivan, 376 U.S. 254 (1964), changed the course of defamation law. In New York Times Co., the Court first determined that the First Amendment limits state law remedies available to a defamation plaintiff. See id. at 269. The Court announced that the First Amendment, as incorporated and applied to the states through the Fourteenth Amendment, prohibited "a public official from recovering damages for a defamatory falsehood relating to his official conduct unless he proves that the statement was made with `actual malice' -that is, with knowledge that it was false or with reckless disregard of whether it was false or not." Id. at 279-80. The limits of the New York Times Co. standard have been further explored and defined in a series of High Court defamation cases: Curtis Publ'g Co. v. Butts, 388 U.S. 130 (1967); Rosenbloom v. Metromedia, Inc., 403 U.S. 29, 91S.Ct. 1811, 29 L.ed.2d 296 (1971), overruled by Gertz v. Robert Welch, Inc., 418 U.S. 323 (1974); Time, Inc. v. Firestone, 424 U.S. 448 (1976); Wolston v. Reader's Digest Ass'n, Inc., 443 U.S. 157 (1979); Hutchinson v. Proxmire, 443 U.S. 111 (1979); and Philadelphia Newspapers, Inc. v. Hepps, 475 U.S. 767 (1986).

Because a Constitutional defamation jurisprudence has developed, the state law of defamation has been displaced to the extent that the state law conflicts with Constitutional law. The primary framework of a defamation claim, however, continues to be a state law tort claim. This commingling of state law and Constitutional law has created a complex jurisprudence of interlocking and overlapping Constitutional and state law inquiries. On appeal, Wells has raised questions of both state and Constitutional defamation law. We first address the state law issues.

Wells's state law questions are two-fold. First, Wells appeals the district court's ruling on choice of law. Second, Wells asserts that the district court erred when it concluded that Liddy's statements made during the cruise ship speech, the Accuracy in Media web site, and the Don and Mike show appearance were not capable of defamatory meaning. After a brief review of choice-of-law principles, we will address each of Wells's four defamation claims in turn to review the district court's conclusion regarding the applicable law and each statement's possible defamatory meaning.

Wells asserts that the district court misapplied Maryland's lex loci delicti choice-of-law rule for tort claims when it determined that the JMU speech claim, the cruise ship speech claim, and the Don and Mike show claim were subject to the law of Wells's domicile, Louisiana, rather than to the law of Virginia.13 During the proceedings below, the district court applied, as a default rule, the law of Louisiana, the place of Wells's domicile, to all four of her claims of defamation. The district court noted that "[i]dentifying the place of injury is somewhat problematic in a case such as this where the plaintiff does not allege concrete harm." Wells, 1 F. Supp. 2d at 536. Therefore, it determined that under Maryland's choice-of-law rules the presumptive place of the harm in a defamation action should be the plaintiff's domicile. Wells asserts that the district court erred in applying Louisiana law to the JMU speech claim, the cruise ship speech claim, and the Don and Mike show claim. Instead, Wells asserts that Virginia law, for varying reasons, should have been applied to each of these claims.

A federal court sitting in diversity must apply the choice-of-law rules from the forum state. See Klaxon Co. v. Stentor Elec. Mfg. Co., 313 U.S. 487, 496-97 (1941). For tort claims, Maryland adheres to the First Restatement of Conflict of Laws rule, lex loci delicti commissi, or the law of the place of the harm, to determine the applicable substantive law. See Naughton v. Bankier, 691 A.2d 712, 716 (Md. Ct. Spec. App. 1997). Under the First Restatement, the place of the harm is defined as "the state where the last event necessary to make an actor liable for an alleged tort takes place." Restatement (First) of Conflict of Laws § 377 (1934); see generally Eugene F. Scoles & Peter Hay, Conflict of Laws 571 (2d ed. 1992) (discussing lex loci delicti); Robert L. Felix, Leflar in the Courts: Judicial Adoptions of Choice-Influencing Considerations, 52 Ark. L. Rev. 35 (1999) (surveying states' decisions to move away from lex loci delicti regime). In defamation actions, the place of the harm has traditionally been considered to be the place where the defamatory statement was published, i.e., seen or heard by non-parties. See Restatement (First) of Conflicts § 377 n.5 ("[W]here harm is done to the reputation of a person, the place of wrong is where the defamatory statement is communicated."); Lapkoff v. Wilks, 969 F.2d 78, 81 (4th Cir. 1992) (applying lex loci delicti rule and concluding that when defamatory statements occurred in Virginia, Virginia law applied); Clair v. Righter, 250 F. Supp. 148, 150 (W.D. Va. 1966) (stating that the place of publication is the last event necessary to render the tort-feasor liable in a defamation action); see also James R. Pielemeier, Constitutional Limitations on Choice of Law: The Special Case of Multistate Defamation, 133 U. Pa. L. Rev. 381, 393-94 (1985) (noting that as a general rule the place of publication is the place of the harm).

With this framework in mind, we address each of Wells's four defamation claims and apply the correct law to the legal question of whether the statement is capable of a defamatory meaning.

Applying Maryland's choice-of-law analysis to the JMU speech is quite straightforward. Liddy delivered the JMU speech on April 2, 1996, in Harrisonburg, Virginia. The record indicates that the speech was never broadcast by any means and was heard only by the audience at JMU. Therefore, publication of the speech occurred solely in Virginia. Applying Maryland's traditional lex loci delicti rule as discussed above, it is clear that the law of Virginia should have been applied to this claim. The district court erred when it applied the law of Louisiana.

The district court ruled that under Louisiana law the JMU speech was capable of defamatory meaning. Applying Virginia law to the same question, we also conclude that the JMU speech is capable of conveying a defamatory meaning. Under Virginia law, the question of whether a statement is capable of having a defamatory meaning is a legal question. See Yeagle v. Collegiate Times , 497 S.E.2d 136, 138 (Va. 1998) (noting that trial court had responsibility to determine as a matter of law whether an allegedly defamatory phrase was capable of defamatory meaning).

The Virginia standard for determining whether words are capable of defamatory meaning derives from the common law:

At common law defamatory words which are actionable per se are: (1) Those which impute to a person the commission of some criminal offense involving moral turpitude, for which the party, if the charge is true, may be indicted and punished. (2) Those which impute that a person is infected with some contagious disease, where if the charge is true, it would exclude the party from society. (3) Those which impute to a person unfitness to perform the duties of an office or employment of profit, or want of integrity in the discharge of the duties of such an office or employment. (4) Those which prejudice such person in his or her profession or trade.

Carwile v. Richmond Newspapers, Inc., 82 S.E.2d 588, 591 (Va. 1954).14 Additionally, Virginia recognizes that "[a]ll other defamatory words which, though not in themselves actionable, occasion a person special damage are actionable." Id., accord Fleming v. Moore, 275 S.E.2d 632, 635 (Va. 1981).

Virginia law requires that the potential defamatory meaning of statements be considered in light of the plain and ordinary meaning of the words used in context as the community would naturally understand them. See Old Dominion Branch No. 496 v. Austin, 192 S.E.2d 737, 742 (Va. 1972), rev'd on other grounds, 418 U.S. 264 (1974). We look not only to the actual words spoken, but also to "inferences fairly attributable to [them]," Yeagle , 497 S.E.2d at 138, and consider whether the words have the potential to hurt the plaintiff's reputation among the "important and respectable" parts of the community. Weaver v. Beneficial Finance Co., 106 S.E.2d 620, 622 (Va. 1959).

During the JMU speech, Liddy made the following statements pertaining to Wells:

[The surveillance camera at the listening post in the Howard Johnson's] looked directly down at a desk of a secretary named Maxine Wells, and her telephone. And they had a telescopic lens camera pointed at that. And that is where the wiretap was subsequently found by the democrats on that phone.

(J.A. at 998.)

[S]ome members of the DNC were using the call girl ring as an asset to entertain visiting firemen. And to that end they had a manila envelope that you could open or close by wrapping a string around a wafer. And in that envelope were twelve photographs of an assortment of these girls and then one group photograph of them. And what you see is what you get. It was kept . . . in that desk of Ida Maxine Wells. Thus, the camera [and] all the rest of it. And what they were doing is as these people would be looking at the brochure, if you want to call it that, and making the telephone call to arrange the assignation that was being wiretapped, recorded and photographed.

(J.A. at 998-99.)

Ascribing to these words their plain meaning and understanding them as the community would naturally understand them in their context, we conclude that the actual words spoken by Liddy are capable of defamatory meaning, namely, that Wells was a participant in a scheme to procure prostitutes. Liddy mentioned Wells during the JMU speech as the focus of illicit surveillance activities linked to the effort to get information on a prostitution ring operating out of the DNC. He mentioned her specifically because the prostitution "brochure" was kept in her desk and her phone was used to arrange meetings with prostitutes. Additionally, a fair inference from Liddy's statements is that Wells was not only involved in prostitution activities but was the primary DNC contact for the prostitution ring. Based upon Liddy's statements, a reasonable listener could readily conclude that Wells's desk was the center of prostitution activity at the DNC. It follows that if Wells's desk was the primary location of illicit activity at the DNC, then Wells herself may have been deeply involved.

Liddy's words, and a fair implication thereof, "impute to [Wells] the commission of some criminal offense involving moral turpitude," Carwile, 82 S.E.2d at 591 -procuring prostitution services. Therefore, under Virginia law, Liddy's statements made during the JMU speech are capable of conveying a defamatory meaning. See id.

Wells also challenges both the district court's choice-of-law ruling and its defamatory meaning ruling pertaining to the cruise ship speech claim. Liddy allegedly defamed Wells while giving a speech on a seven-day Mediterranean cruise.15 Wells asserts that it was error for the district court to apply Louisiana law, the law of her domicile, to this claim. Instead, for reasons that are not entirely clear from her argument, she suggests that Virginia law should have been applied. Wells also avers that the district court erred when it ruled that Liddy's cruise ship statements were not capable of defamatory meaning because "testimony d[id] not establish the exact nature of the statements sufficiently to demonstrate that the statements were capable of defaming Wells." Wells, 1 F. Supp. 2d at 538. We address the choiceof-law issue and the defamatory meaning issue in turn and conclude that the district court erred in both respects.

The district court's application of Louisiana law to a case of defamation at sea was incorrect. "All cases involving a tort committed on navigable water, whether brought under federal admiralty jurisdiction, in state court under the saving-to-suitors clause, or in federal court under diversity jurisdiction, are governed by admiralty law." Byrd v. Byrd, 657 F.2d 615, 617 (4th Cir. 1981); accord Kermarec v. Compagnie Generale Transatlantique, 358 U.S. 625, 628 (1959); see Pryor v. American President Lines, 520 F.2d 974, 977 (4th Cir. 1975) (noting that under either federal diversity or admiralty jurisdiction tort claim would be rooted in maritime law); Scott v. Eastern Air Lines, Inc., 399 F.2d 14, 25 (3d Cir. 1968) (opinion on rehearing) ("[M]aritime principles will govern the tort aspects of the case, since admiralty standards define liability for a maritime tort, whether the proceeding is instituted in admiralty or on the law side of the court."); see also Theodore F. Stevens, Erie R.R. v. Tompkins and the Uniform General Maritime Law, 64 Harv. L. Rev. 246, 269 (1950) (noting that federal courts should apply general maritime law to maritime cases brought under diversity jurisdiction). In this case, the alleged defamation of Wells occurred when Liddy delivered a speech on a ship sailing on the high seas.16 Hence, the governing law is not the common law of any single state, but rather is the general maritime law as interpreted and applied by the courts of the United States. See Knickerbocker Ice Co. v. Stewart, 253 U.S. 149, 159 (1920); cf. Guidry v. Durkin, 834 F.2d 1465, 1470 (9th Cir. 1987) (noting that tort of defamation is cognizable under general maritime law when prima facie elements occur on the high seas). Were Maryland choiceof-law rules applicable here, they too would point us to the general maritime law. Maryland's lex loci delicti principles as outlined above counsel that the law of the place of publication governs defamation claims. Here, the place of publication was the high seas.

"Drawn from state and federal sources, the general maritime law is an amalgam of traditional common-law rules, modifications of those rules, and newly created rules." East River Steamship Corp. v. Transamerica Delaval, Inc., 476 U.S. 858, 864-65 (1986). The role of state law in maritime cases is significant and complex.17 "State law may . . . supplement federal maritime law, as in the exercise of its police powers or in the additional maritime tort remedy; state law may not, however, conflict with federal maritime law, as it would be redefining the requirements or limits of a remedy available at admiralty." Powell v. Offshore Navigation, Inc. , 644 F.2d 1063, 1065 n.5 (5th Cir. 1981). State law is said to conflict with general maritime law when it negatively impacts upon admiralty's foremost goal -uniformity. See Maryland Dep't of Natural Resources v. Kellum, 51 F.3d 1220, 1227 (4th Cir. 1995); Lewis v. Timco, Inc. , 716 F.2d 1425, 1428 (5th Cir. 1983) ("[M]aritime law traditionally resists doctrinal change that might balkanize its uniformity and generality."); Byrd, 657 F.2d at 617. Thus, "courts applying maritime law have repeatedly rejected choice of law notions that would reference state tort doctrines." Lewis, 716 F.2d at 1428.

After a thorough review, it appears that there is no well-developed body of general maritime law of defamation. In such a situation, it is clear that the general maritime law may be supplemented by either state law, see Bell v. Tug Shrike, 332 F.2d 330, 334 (4th Cir. 1964), or more general common law principles, see Marastro Compania Naviera, S.A. v. Canadian Maritime Carriers, Ltd. , 959 F.2d 49, 53 (5th Cir. 1992) (applying the Restatement (Second) of Torts in the absence of general maritime law of trespass); Vickers v. Chiles Drilling Co., 822 F.2d 535, 538 (5th Cir. 1987) (applying Restatement (Second) of Torts for principles of product liability law); Nissan Motor Corp. in U.S.A. v. Maryland Shipbuilding & Drydock Co., 544 F. Supp. 1104, 1110-111 (D. Md. 1982) (applying Restatement (Second) of Torts for law of trespass and nuisance). Because great diversity exists among the states' defamation laws, we conclude that it would be more appropriate to apply general common law tort principles rather than the specific law of a single state. Application of a single state's defamation law would "impair the uniformity and simplicity which is a basic principle of the federal admiralty law." Nissan Motor Corp., 544 F. Supp. at 1111 (internal quotation marks omitted). Accordingly, we determine that the common law as compiled in the Restatement (Second) of Torts should control our evaluation of Wells's claim of shipboard defamation. See Marastro Compania Naviera, 959 F.2d at 53 (applying Restatement in absence of clear general maritime law on point); Vickers , 822 F.2d at 538 (same); Nissan Motor Corp., 544 F. Supp. at 1111 (same).

Thus, we will apply the standards of the Restatement (Second) of Torts to determine whether Liddy's cruise ship speech is capable of defamatory meaning.18 Under the Restatement, the questions of "whether a communication is capable of bearing a particular meaning" and "whether that meaning is defamatory" are questions of law. Restatement (Second) of Torts § 614 (1977). Further, the Restatement states that "[a] communication is defamatory if it tends so to harm the reputation of another as to lower him in the estimation of the community or to deter third persons from associating or dealing with him." Id. § 559. In determining whether words are capable of conveying a defamatory meaning the court must "take into account all of the circumstances surrounding the communication of the matter complained of as defamatory." Id. § 614 cmt. d. "A communication to be defamatory need not tend to prejudice the other in the eyes of everyone in the community or of all of his associates, nor even in the eyes of a majority of them. It is enough that the communication would tend to prejudice him in the eyes of a substantial and respectable minority of them." Id. § 559 cmt. e. "On the other hand it is not enough that the communication would be derogatory in the view of a single individual or a very small group of persons." Id. We turn now to the task of determining whether Liddy's cruise ship speech is capable of defamatory meaning.

As evidence of what statements were made on the Mediterranean cruise, we have Liddy's own deposition testimony regarding the events that transpired aboard the ship:

Q: I was told that in August you were scheduled to do some boat trip around the Mediterranean or something. Did that take place?

Liddy: Yes. . . . .

Q: Now, when you are on a trip like that would you tell the Watergate story as you now believe it to be, that you tell in your book?

Liddy: Yes. Sometimes, there are sometimes when I have given a speech that was a bring up to date, what we have learned through the Dean case of Watergate. Other times the speech has absolutely nothing to do with Watergate .. . a motivational speech. Typically, at the end of a motivational speech I will have a question and answer period and sometimes the subject of Watergate comes up, sometimes it does not.

Q: But on the boat trip, did you discuss the Watergate theory as you now believe it? . . . .

Q: Did you mention the name of Ida Maxwell Wells on the boat? Liddy: I think I did.

Q: Did you say to the people on the boat that she kept pictures of prostitutes in her desk? . . . . Liddy: My best recollection is that I said that the target of the second break-in was the desk in the Oliver/Wells area that was assigned to one Ida Maxie Wells.

Q: And did you say there were pictures of prostitutes in that desk, and that's why Martinez had the key and went after it? . . . .

Liddy: I said that the evidence indicates that there were photographs of women in the desk who were available for prostitution activities.

Q: But did you say that to the people on the boat in August 1997, that you believe there were pictures of prostitutes in Maxie Wells'[s] desk?

Liddy: In the desk of Maxie Wells. I would also mention that in addition to Miss Wells having a key, Mr. Martinez had a key . . . .

Q: Did you say that in the speech in August 1997 that visitors to the DNC would be shown these pictures of prostitutes?

Liddy: I said the photographs would act as a brochure.

(J.A. at 1131-34.)

The district court determined that Liddy's testimony"d[id] not establish the exact nature of the statements sufficiently to demonstrate that the statements were capable of defaming Wells" because "Liddy was not even certain that he had mentioned Wells's name on the cruise ship." Wells, 1 F. Supp. 2d at 538.

We disagree. Reviewing Liddy's deposition testimony in context, that testimony indicates that he gave a speech on the cruise ship during which he discussed the Watergate break-in and can be viewed as capable of defamatory meaning, specifically, that Wells's desk contained pictures of prostitutes whose services would be offered to men visiting the DNC. Liddy's deposition was taken in November of 1997, only three months after the cruise ship speech, so the fact that some of his answers are based upon his "best recollection" is not particularly alarming. Nor do we share the district court's concern that Liddy's testimony "did not expressly differentiate between what he said on the boat and what he generally advances as his theory." Wells, 1 F. Supp. 2d at 538. The majority of Liddy's answers directly address his memory of the speech he gave during the cruise. The district court based its conclusion that Liddy was not discussing a specific recollection of the cruise ship speech upon one statement that used the phrase "I would also mention." Id. Given that we are obligated to review the evidence in its full context and view it in the light most favorable to Wells, the nonmoving party, to determine whether a statement is capable of defamatory meaning, we believe the district court took too narrow a view, discounting several more definite statements of what transpired on the basis of one slightly unclear statement. Further, Liddy's statement that he "would also mention" does not necessarily indicate that he was speaking of his general Watergate presentation to the exclusion of what he said on the cruise ship. Rather, it is just as likely that Liddy was communicating to the questioner the fact that he always said the same thing during his speeches, including the speech delivered on the cruise ship.

Liddy's own testimony supports the conclusion that he told the cruise ship audience that Wells's locked desk held pictures of prostitutes that were shown to visiting men. Such a statement can be viewed as implying that Wells was involved procuring prostitution services, and as a result that Wells was involved in immoral criminal acts. A statement implicating Wells in prostitution activities "tends so to harm [her] reputation . . . as to lower h[er] in the estimation of the community or to deter third persons from associating or dealing with h[er]." Restatement (Second) of Torts § 559. Although a jury, upon further factual development, may not conclude that Wells was actually defamed during the cruise ship speech, Liddy's testimony supports the conclusion that he made statements during the cruise ship speech that meet the legal test for defamatory meaning.

Next, Wells asserts that the district court improperly applied Louisiana law to the Don and Mike show broadcast and erred when it determined that the statements on the radio show were not capable of defamatory meaning. Finding no error in the district court's rulings, we affirm.

The Don and Mike radio show is a nationally syndicated daily show that can be heard throughout the United States. As such, any defamatory content is published simultaneously in multiple state jurisdictions. Because of the widespread simultaneous publication of the allegedly defamatory statement in many different jurisdictions, application of the traditional lex loci delicti rule becomes cumbersome, if not completely impractical. See James R. Pielemeier, Constitutional Limitations on Choice of Law: The Special Case of Multistate Defamation, 133 U. Pa. L. Rev. 381, 393-94 (1985) (noting choice of law difficulties in First Restatement jurisdictions). Maryland has not yet promulgated a unique choice of law rule made applicable to multistate defamation cases to correct the obvious deficiency of the lex loci delicti rule in this context. See Fornshill v. Ruddy, 891 F. Supp. 1062, 1069 (D. Md. 1995).

As a court sitting in diversity, we have an obligation to interpret the law in accordance with the Court of Appeals of Maryland, or where the law is unclear, as it appears that the Court of Appeals would rule. See Liberty Mut. Ins. Co. v. Triangle Indus., 957 F.2d 1153, 1156 (4th Cir. 1992) (holding that if state law is unclear federal courts must predict the decision of the state's highest court); Brendle v. General Tire & Rubber Co., 505 F.2d 243, 245 (4th Cir. 1974). To forecast a decision of the state's highest court we can consider, inter alia: canons of construction, restatements of the law, treatises, recent pronouncements of general rules or policies by the state's highest court, well considered dicta, and the state's trial court decisions. See Liberty Mut., 957 F.2d at 1156.

In recent years, the Court of Appeals of Maryland has indicated its willingness to apply more flexible choice-of-law rules from the Second Restatement in situations when the First Restatement rules have become unworkable. Specifically, in Hauch v. Connor, 453 A.2d 1207 (Md. 1983), the Court of Appeals of Maryland moved away from a strict lex loci delicti approach in the field of workers' compensation law, and instead adopted an approach considering the Second Restatement's most significant relationship test. See id. at 1214. Similarly, in American Motorists Ins. Co. v. ARTRA Group, Inc., 659 A.2d 1295 (Md. 1995), the Court of Appeals adopted a limited version of the concept of renvoi from the Second Restatement and in so doing displaced the strict lex loci contractus rule of the First Restatement. See id. at 1301. Because multistate defamation is a tort for which the lex loci delicti rule fails to reach a satisfactory result on the choice of applicable substantive law, we believe, based upon the cases cited above, that the Court of Appeals of Maryland would consider the rule outlined in the Restatement (Second) of Conflict of Laws to select the applicable substantive law. See Fornshill, 891 F. Supp. at 1069 (applying Second Restatement to choice of laws issue in Maryland venue multistate defamation claim); Crowley v. Fox Broadcasting Co., 851 F. Supp. 700, 702 (D. Md. 1994) (same).

The Second Restatement contains a specific section addressing multistate defamation that provides:

(1) The rights and liabilities that arise from defamatory matter in any . . . broadcast over radio or television . . . or similar aggregate communication are determined by the local law of the state which, with respect to the particular issue, has the most significant relationship to the occurrence and the parties under the principles stated in § 6.

(2) When a natural person claims that he has been defamed by an aggregate communication, the state of most significant relationship will usually be the state where the person was domiciled at the time, if the matter complained of was published in that state.

Restatement (Second) of Conflict of Laws § 150 (1971). Wells notes that the Don and Mike show was broadcast in Louisiana, the state of her domicile. Therefore, following the Second Restatement, Louisiana law applies to the Don and Mike show claim.

Under Louisiana law, whether a statement is capable of having a defamatory meaning is a question of law. See Ryan v. Shreveport Times Publ'g Co., 344 So. 2d 114, 117 (La. Ct. App. 1977) ("Whether the words used are capable of a defamatory meaning is a question of law for a determination by the court, and not a question of fact to be decided by a jury. If the court concludes, as a matter of law, the words are capable of having a defamatory meaning, it then becomes a jury question whether the words, as applied to plaintiff, in fact defamed him.") To determine whether words are capable of defamatory meaning, Louisiana applies the following standard:

Defamatory words are those that would expose a person to contempt or ridicule, or cause a person to be shunned or avoided. Defamatory words include almost any language which on its face has a tendency to injure a person's reputation. Consideration must be given to the entire statement that was made and the circumstances of its publication. Words which impute criminal action to another are defamatory per se.

Tonubbee v. River Parishes Guide, 702 So. 2d 971, 974 (La. Ct. App. 1997), cert. denied, 119 S. Ct. 142 (1998).

In deciding whether the words are capable of a defamatory meaning, the publication must be read as a whole. The words must be construed according to the meaning that will be given them by reasonable individuals of ordinary intelligence and sensitivity. Their significance must not be distorted to give an unusual meaning, and they are to be understood only in the context in which they were used and in the manner shown by the circumstances under which they were used.

Brown v. News-World Publ'g Corp., 245 So. 2d 430, 433 n.1 (La. Ct. App. 1971) (internal quotation marks and citations omitted). We turn now to applying the Louisiana law to Liddy's statements on the Don and Mike show. Wells asserts that the following statements are capable of defamatory meaning:

Liddy: Okay, I was the political intelligence chieftain, as well as the general counsel of the Committee to Reelect the President. . . . Now what I did not know is that John Dean did not trust me any more than I trusted him. And so my men were told, although I was not, that they were to go in there and, what, the telephone that was wired was not Mr. O'Brien's but was the telephone that was on the desk of a woman named . . . Ida Maxwell Wells . . . and she was the secretary to a man named R. Spencer Oliver. . . . .

Liddy: Well next door to the Watergate was a place called the Columbia Plaza Apartments and operating in there was what is known as a call girl ring and the lawyer who represented those girls was arrested by the FBI and they found his address book that had the names of his clients and also that included the call girl and there was a woman in there whose code name was "clout." . . . .

Liddy: Now, to make a long story short. That was kind of what it was all about and if you want a secondary source on Watergate, you know, to read about what was going on and everything. There is a book called Silent Coup .

(J.A. at 1021, 1022, 1023.)

We cannot conclude that these statements when read in context and given their ordinary and common meaning would be understood "by reasonable individuals of ordinary intelligence and sensitivity," as impugning Wells's reputation. Brown, 245 So. 2d at 433 n.1. Although the statements indicate that the telephone on Wells's desk was bugged during Watergate, the statements do not indicate her participation in that activity or any other criminal acts, including prostitution activities. Establishing only a connection between Wells and Watergate is not sufficient to create defamatory meaning. Liddy's comments on the Don and Mike show do not take the extra step of establishing a direct connection between Wells, or even her desk, and the prostitution ring to which the DNC was allegedly referring visitors.

Here, Liddy merely referred listeners to Silent Coup for the details of the Watergate break-in. Wells argues that the Silent Coup reference is sufficient to alert listeners to the alleged connection between her and the DNC call-girl ring. We cannot agree. Silent Coup itself does not explicitly state that Wells was personally involved in prostitution activities. Under Louisiana law, we must assess the defamatory meaning of a claim in terms of the reasonable listener of ordinary sensitivity. See Brown, 245 So. 2d at 433 n.1. A reasonable Don and Mike listener familiar with Silent Coup would not necessarily reach the conclusion that Wells was linked to prostitution activities. Thus, fair inferences from the Don and Mike show broadcast do not negatively impact Wells's reputation. Accordingly, we conclude that the statements are not capable of defamatory meaning, and we affirm the district court's grant of summary judgment on this claim.

Finally, we assess the final allegedly defamatory statement, the comments that appeared on the Accuracy in Media web site. Wells does not challenge the district court's application of Louisiana law to this claim. Louisiana law is applicable because publication from a world wide web site is another example of multistate defamation, and, therefore, following our analysis in the previous part, we apply the law of Wells's domicile.

For an unknown period of time, the Accuracy in Media web site contained the following statement:

Not until Colodny and Gettlin wrote Silent Coup did Liddy realize that the true objective of this second raid was to get into the desk of Maxie Wells, Spencer Oliver's secretary, said to be the key figure in arranging dates with the call girls. Unknown to Liddy at the time, one of the burglars carried a key to Wells'[s] desk.

(J.A. at 1016.) Of the four statements that Wells has claimed to be defamatory, this statement most clearly associates Wells with criminal activity. As a result, the statement is capable of defamatory meaning. There are evidentiary problems, however, that preclude attributing the statement to Liddy, and, therefore, the district court's grant of summary judgment was appropriate.

The record is devoid of any proof that Liddy ever spoke to Accuracy in Media regarding the Watergate break-in19 or authorized them to use his theory. Thus, we cannot conclude that Wells has met an essential element of a defamation cause of action under Louisiana law -publication. See Wiggins v. Creary, 475 So. 2d 780, 782 (La. Ct. App. 1985) ("The elements of [a] cause of action in defamation are (1) defamatory words, (2) publication to a third party, (3) falsity of the statement, and (4) injury."). Further, assuming the Accuracy in Media web site is a republication of Liddy's views on Watergate, Louisiana law holds as a general rule that the original author/speaker is not liable for the voluntary republication of his statements by others. See id. An exception to this rule exists when the republication is the "natural and probable consequence of the defendant's act." Giordano v. Tullier, 139 So. 2d 15, 19 (La. Ct. App. 1962). Here, however, Wells has not pleaded or put forth any proof on this point of Louisiana law. Because Wells has not raised a genuine issue of material fact on this matter, we hold that the district court's grant of summary judgment on this claim was correct and must be affirmed.

In sum, we hold that the district court's grant of summary judgment was proper on the Don and Mike show and Accuracy in Media web site claims. We conclude, however, that the district court erred when it determined that Louisiana law applied to Wells's JMU speech and cruise ship speech claims. Applying Maryland's lex loci delicti rule, the substantive defamation law controlling the JMU speech must be the law of Virginia. The cruise ship speech claim, asserting that defamation occurred shipboard during a Mediterranean cruise, points us to general maritime law of defamation and consequently to the Restatement (Second) of Torts. Applying the correct substantive law, we determine that both the JMU speech claim and the cruise ship claim are capable of defamatory meaning. Having resolved these state law matters, we turn next to Wells's public figure status.

Wells next appeals the district court's conclusion that for purposes of the public debate on Watergate she is an involuntary public figure, and therefore required to prove that Liddy acted with actual malice before she can recover compensatory damages for actual injury. Liddy argues that the district court's conclusion that Wells is an involuntary public figure is correct and asserts in the alternative that Wells's participation in Watergate-related dialogue is sufficient to qualify her as a voluntary limited-purpose public figure. Because the level of culpability the plaintiff must prove to recover compensatory damages under state law is dependent upon the plaintiff's public or private figure status, we must address this issue, which is a matter of law. See Fitzgerald v. Penthouse Int'l, Ltd., 691 F.2d 666, 669-70 (4th Cir.1982)

The public figure inquiry is a First Amendment-influenced area of defamation jurisprudence. As we noted briefly above, prior to New York Times Co. v. Sullivan, 376 U.S. 254 (1964), defamation law was entirely an area of state concern and was immune from First Amendment restraints. In New York Times Co., however, the Supreme Court announced that the First and Fourteenth Amendments prohibit "a public official from recovering damages for a defamatory falsehood relating to his official conduct unless he proves that the statement was made with `actual malice' -that is, with knowledge that it was false or with reckless disregard of whether it was false or not." Id. at 27980. Additionally, New York Times Co. indicated that actual malice must be demonstrated by clear and convincing evidence. Id. at 28586; see Gertz v. Robert Welch, Inc., 418 U.S. 323, 342 (1974). In Curtis Publishing Co., v. Butts, 388 U.S. 130 (1967), the New York Times Co. actual malice standard was extended beyond public officials to encompass "public figures." See id. at 162-65 (Warren, C.J., concurring in the result). The parameters of public figure status were not fully defined in the Curtis decision.

After Curtis, the Court took a short-lived step away from classifying plaintiffs by their public or private figure status as a means to determine what level of defendant's fault was required for recovery. In Rosenbloom v. Metromedia, Inc., 403 U.S. 29 (1971), a plurality of the Court determined that plaintiffs must prove actual malice when their claim rested upon any "discussion and communication involving matters of public or general concern, without regard to whether the persons involved are famous or anonymous." Id. at 44.

The Rosenbloom plurality's standard, however, was overruled in Gertz v. Robert Welch, Inc., 418 U.S. 323 (1974). In reversing course and holding that to recover compensatory damages private individuals need not prove that the defendant acted with actual malice even when the defamation occurred during discussion of matters of public concern, the Court enunciated classifications of defamation plaintiffs and the corresponding levels of defendant's culpability that must be proven. See id. at 342-48. Justice Powell, writing for the majority, distinguished between public figures and private figures and justified their different requirements for proof of defendant's fault on two grounds: first, the greater ability of public figures to resort to self-help on account of their better access to the media and channels of communication, and second, the voluntary assumption of the risk of publicity undertaken by public figures. See id. at 344. Among the class of public figures, the Court made further distinctions:

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. Thus, we have interpreted Gertz as creating three distinct types of public figures:

(1) "involuntary public figures," who become public figures through no purposeful action of their own; (2) "all-purpose public figures," who achieve such pervasive fame or notoriety that they become public figures for all purposes and in all contexts; and (3) "limited-purpose public figures," who voluntarily inject themselves into a particular public controversy and thereby become public figures for a limited range of issues.

Foretich v. Capital Cities/ABC, Inc., 37 F.3d 1541, 1551-52 (4th Cir. 1994).20

Since Gertz, it has been clear that the First Amendment sets limits on a public figure's ability to recover for defamation. A public figure "may recover for injury to reputation only on clear and convincing proof that the defamatory falsehood was made with knowledge of its falsity or with reckless disregard for the truth." Gertz, 418 U.S. at 342. After Gertz, however, the level of defendant's fault that must be proved by private figures to recover compensatory damages21 in defamation actions is left to the states22 even when the defamatory communication touches on matters of public concern, with the caveat that strict liability schemes run afoul of the First Amendment. See id. at 347-50.

The Court has revisited public figure status three times since it handed down its decision in Gertz. In Time, Inc. v. Firestone, 424 U.S. 448 (1976), Time magazine had printed an unflattering report regarding the divorce of Russell and Mary Alice Firestone. See id. at 452. Mary Alice Firestone filed a defamation action in Florida state court and won a $100,000 judgment. See id. Time appealed to the Supreme Court and argued that Mary Alice Firestone should have been required to prove actual malice because she was a public figure. See id. The Court disagreed with Time's contention. See id. at 453. The Court noted that Mary Alice Firestone was not a public figure because she "did not assume any role of especial prominence in the affairs of society, . . . and she did not thrust herself to the forefront of any particular public controversy in order to influence the resolution of the issues involved in it." Id. The Court noted that although the divorce proceedings were a matter of public record and public interest, they were not a public controversy. See id. at 454. Further, the Court held that Mary Alice Firestone's participation in the divorce proceedings was required by state law, so she had not freely chosen to publicize the issues. See id. The Court also stated that Mary Alice Firestone's appearance in a few press conferences to satisfy the press did not convert her into a public figure. See id. n.3, 96 S.ct.958

The Court next addressed the public figure issue in a pair of 1979 cases -Hutchinson v. Proxmire, 443 U.S. 111 (1979) and Wolston v. Reader's Digest Ass'n, Inc., 443 U.S. 157 (1979). Hutchinson, a research behavioral scientist, filed a defamation suit as the result of having the topic of his research, why monkeys clench their jaws, listed as a "Golden Fleece Award"23 winner. See Hutchinson, 443 U.S. at 117-18. The Court held that it was error to classify Hutchinson as a limited-purpose public figure based upon his receipt of federal research funds and his access to the media to respond to the Golden Fleece controversy, because neither factor demonstrated that Hutchinson was a public figure prior to his mention in connection with the Golden Fleece Award. See id. at 134-35. In Wolston, the Court held that the nephew of convicted spies was not a public figure on the basis of his receipt of grand jury subpoenas, failure to appear before the grand jury, and contempt conviction resulting therefrom. See Wolston, 443 U.S. at 161-63. Although fifteen newspaper stories had been written covering his contempt conviction and Wolston's name appeared in two books published at the time of the events, the Supreme Court held that Wolston's action of failing to appear at the grand jury was not sufficient to create limited-purpose public figure status when his voluntary discussion of the matter was minimal. See id. at 166. The Court reasoned that "[a] private individual is not automatically transformed into a public figure just by becoming involved in or associated with a matter that attracts public attention," id. at 167, and emphasized that Wolston had not "engaged the attention of the public in an attempt to influence the resolution of the issues involved, id. at 168, and "in no way calculated to draw attention to himself . . . to invite public comment or influence the public with respect to any issue," id. Because the Court concluded that Wolston was a private figure it was error to apply the actual malice standard to Wolston's claim for compensatory damages.

Applying the foregoing Supreme Court precedents, this Circuit has developed a five-factor test to determine whether a plaintiff is a limited-purpose public figure. Liddy has asserted that Wells meets the test for limited-purpose public figure status. Because the jurisprudence regarding limited-purpose public figures is well-established in this Circuit, we address Liddy's contentions regarding the voluntary nature of Wells's participation in the Watergate controversy first, before weighing-in on the district court's ultimate conclusion that Wells qualifies as an involuntary public figure.

Before a plaintiff can be classified, as a matter of law, as a limitedpurpose public figure, the defendant must prove that:

(1) the plaintiff has access to channels of effective communication;

(2) the plaintiff voluntarily assumed a role of special prominence in the public controversy;

(3) the plaintiff sought to influence the resolution or outcome of the controversy;

(4) the controversy existed prior to the publication of the defamatory statement; and

(5) the plaintiff retained public-figure status at the time of the alleged defamation.

See Reuber v. Food Chemical News, Inc., 925 F.2d 703, 708-10 (4th Cir. 1991) (en banc). In Foretich, we added an additional consideration to the limited-purpose public figure inquiry when we held that if the content of a defamatory statement touches upon an area that state law has traditionally considered to be defamatory per se, then the plaintiff cannot be categorized as a limited-purpose public figure solely because he makes reasonable public replies to the statement. See 37 F.3d at 1558. In evaluating an individual's limited-purpose public figure status we look at "the nature and extent of an individual's participation in the particular controversy giving rise to the defamation." Gertz, 418 U.S. at 351.

In order to prevail on his assertion that Wells's participation in the Watergate controversy merits her classification as a limited-purpose public figure Liddy must establish each of the five elements recited above. See Foretich, 37 F.3d at 1556 (noting that the defendant bears the burden on all five elements). According to Liddy, Wells's voluntary injection into the Watergate controversy, an event of unprecedented historical interest and importance, is sufficient to satisfy the five-part test because she was interviewed by the FBI, mentioned in the newspaper, subpoenaed to testify before the grand jury, and called before the Senate in the early 1970s. Additionally, Liddy points to several media exposures Wells has had since the prostitution-oriented theory of the Watergate break-in emerged in 1984. Specifically, Liddy points out that Wells published a letter to the editor in the New York Times in 1985, spoke to a reporter on the twentieth anniversary of the Watergate break-in for an article that appeared in the Washington Post in 1992, spoke to the BBC in 1993, was named in an Arts & Entertainment Network broadcast entitled"The Key to Watergate" and a Geraldo Rivera documentary "Now it Can be Told," and spoke to historian James Rosen. Although Liddy has amply demonstrated that Wells has "access to channels of effective communication," we conclude that Liddy has failed to show that Wells has "voluntarily assumed [a role] of special prominence in the . . . public controversy."24 Id. Therefore, we cannot conclude that Wells is a limited-purpose public figure.

Gertz, Firestone, and Wolston each have touched upon the characteristics that must be proven to demonstrate that a plaintiff has "voluntarily assumed a role of special prominence in a public controversy." Reuber, 925 F.2d at 709. The Supreme Court stated in Gertz that the class of limited-purpose public figures included those individuals who "thrust[ed] themselves to the forefront of particular public controversies in order to influence the resolution of the issues involved," 418 U.S. at 345, or who "voluntarily inject[ed] [themselves] . . . into a particular public controversy . . . [and] assume[d] special prominence in the resolution of public questions," id. at 351. Although Gertz, an attorney, had taken on a case that was related to a public controversy and likely to generate media exposure, the Court held that Gertz's actions were insufficient to characterize him as a public figure:

[Gertz's] participation related solely to his representation of a private client. He took no part in the criminal prosecution . . . . Moreover, he never discussed either the criminal or civil litigation with the press and was never quoted as having done so. He plainly did not thrust himself into the vortex of this public issue, nor did he engage the public's attention in an attempt to influence its outcome.

Id. at 352.

In Firestone, the Court stated that the Firestones' divorce was a private matter, not a public controversy, but even if it had been a public controversy, Mary Alice Firestone "did not thrust herself to the forefront of [that] controversy in order to influence the resolution of the issues involved in it." 424 U.S. at 453. Thus, the Court held that Mary Alice Firestone was not a limited-purpose public figure. Then-Justice Rehnquist amplified the Court's holding, noting that Mary Alice Firestone:

[had not] freely cho[sen] to publicize issues as to the propriety of her married life. She was compelled to go to court by the State in order to obtain legal release from the bonds of matrimony. . . . [I]n such an instance resort to the judicial process . . . is no more voluntary in a realistic sense than that of the defendant called upon to defend his interests in court. Id. at 454 (internal quotation marks omitted).

The Court went on to observe that litigants in general have not voluntarily stepped before the public eye, but rather are "drawn into a public forum largely against their will in order to attempt to obtain the only redress available to them or to defend themselves against actions brought by the State or by others." Id. at 457.

In Wolston, the Court built upon its holding in Firestone regarding the public figure status of those involved in litigation. Its holding extended the view that persons involved in litigation have not voluntarily sought prominence in a public controversy merely by being the subject of a subpoena to appear as a witness in a government investigation. In addressing whether Wolston had voluntarily injected himself into the controversy surrounding Soviet espionage in the 1950s, the Court stated:

It would be more accurate to say that [Wolston] was dragged unwillingly into the controversy. The Government pursued him in its investigation. [He] did fail to respond to a grand jury subpoena, and this failure, as well as his subsequent citation for contempt, did attract media attention. But the mere fact that [he] voluntarily chose not to appear before the grand jury, knowing that his action might be attended by publicity, is not decisive on the question of public figure status. . . . [Wolston] never discussed this matter with the press and limited his involvement to that necessary to defend himself against the contempt charge. It is clear that[he] played only a minor role in whatever public controversy there may have been concerning the investigation of Soviet espionage. 443 U.S. at 166-67.

Further, the Court observed that Wolston "assumed no `special prominence in the resolution of public questions,'" id. at 168 (quoting Gertz, 418 U.S. at 351), because "[h]is failure to respond to the grand jury's subpoena was in no way calculated to draw attention to himself in order to invite public comment or influence the public with respect to any issue. He did not in any way seek to arouse public sentiment in his favor and against the investigation." Id.

In light of the foregoing, viewing Wells's public exposures during and after the Watergate break-in collectively, we conclude that Wells has not "voluntarily assumed [a role] of special prominence in the . . . public controversy." Foretich, 37 F.3d at 1556. Gertz, Firestone, and Wolston establish conclusively that Wells was not a public figure during the immediate aftermath of the Watergate break-in in the 1970s. Like Wolston, Wells's involvement in the Watergate investigation was wholly involuntarily; she was "dragged unwillingly into the controversy," Wolston, 443 U.S. at 166, initially by the commission of a crime at her workplace and later by the governmental investigation that ensued. Wells's discussions with the FBI, response to the grand jury subpoena, and appearance before the Senate committee simply were not voluntary actions but rather were compelled by the force of law. See id. Even if we assume arguendo that Wells was involved in criminal activity during her employment at the DNC,25 precedent counsels that activity likely to engender publicity, even criminal activity, does not equate to taking on a role of special prominence in a public controversy. See Wolston, 443 U.S. at 168 ("reject[ing] the further contention . . . that any person who engages in criminal conduct automatically becomes a public figure for purposes of comment on a limited range of issues relating to his conviction."); Gertz, 418 U.S. at 352 (holding that an attorney who took on a case linked to a public controversy but who did not speak to the press was not a public figure).

The question then arises whether Wells voluntarily has attained special prominence in the Watergate controversy as a result of the revelation of the prostitution-related theory of the break-in. For the most part, Wells's reported role in Watergate has remained unchanged after Secret Agenda and Silent Coup -Wells is named as Spencer Oliver's secretary and a woman whose phone calls were overheard on illegal wire taps. To the extent that Wells's role has expanded, it has done so because Secret Agenda and Silent Coup revealed that Watergate burglar Martinez possessed a key to her desk, exposed prostitution activities that purportedly were occurring at the DNC, and implied that Wells may have had some connection to those illicit activities. There is no proof that any of these disclosures were the result of Wells's voluntary interaction with the authors. Instead, it is clear, as the district court concluded, that the story was divulged by a sole source, Bailley. See Wells, 1 F. Supp. 2d at 543. Because the disclosures of Secret Agenda and Silent Coup cannot be fairly attributed to Wells's voluntary participation, we cannot conclude that she sought a prominent role in the Watergate controversy as a result of being named therein.

Therefore, the question of whether Wells has "voluntarily assumed [a role] of special prominence in the . . . public controversy," Foretich, 37 F.3d at 1556, ultimately, on this record, revolves around four media contacts: (1) her letter-to-the-editor of the New York Times, published in 1985; (2) her 1992 interview with a Washington Post reporter on the twentieth anniversary of the Watergate break-in; (3) her interview with the BBC in 1993; and (4) her discussions with historian James Rosen. The letter-to-the-editor, made in response to a book review of Secret Agenda that named her as a figure involved in arranging dates with prostitutes, clearly falls within the category of self-help and reasonable response to reputation-injuring statements of the type that we approved in Foretich. See 37 F.3d at 1558. Therefore, Wells's letter-to-the-editor should not contribute to the public figure analysis.

Regarding Wells's three other media contacts, Firestone makes clear that voluntary discussion of events with the press does not per se indicate that a defamation plaintiff has "thrust herself to the forefront of [a public] controversy." 424 U.S. at 454 n.3. Rather, when an individual has had contact with the press, the proper questions are whether he has attempted to influence the merits of a controversy, see id., or has "draw[n] attention to himself in order to invite public comment," Wolston, 443 U.S. at 168, or "invited that degree of public attention and comment . . . essential to meet the public figure level," Hutchinson v. Proxmire, 443 U.S. 111, 135 (1979). In this case, there is no proof that Wells spoke to the press in any role other than as a witness to history in response to their repeated requests. Wells, as a DNC worker at the time of Watergate, has a story to tell about the facts of the break-in and the circumstances surrounding the investigation that has independent value. Telling that story, however, is simply giving an eye-witness account of events that are no longer controversial. For example, whether the Watergate break-in occurred on June 17, 1972 is not generally regarded as an open question. How Wells felt as an individual caught up in the investigation is a strictly personal observation. Responding to requests from inquiring reporters about these matters does not convert Wells into a public figure when her statements cannot be interpreted to have an impact upon the merits of the ongoing Watergate controversy. See Firestone, 424 U.S. at 454 n.3 (stating that holding press conferences does not create public figure status when held in response to reporter inquiries and in a situation when it is not possible to influence the merits).

Twenty-seven years after the event, the still-controversial aspects of Watergate revolve around bigger issues such as why the break-in occurred, who was responsible, whether our Constitutional checks and balances are adequate, whether the American populace lost confidence in politicians as a result of the scandal, etc. Liddy has failed to point to evidence demonstrating that Wells has made any attempt to become a spokesperson on any of these types of matters. Therefore, we cannot conclude that Wells has voluntarily undertaken a position at the forefront of the Watergate controversy. The record establishes that Wells has conducted only a handful of interviews over the course of twenty-seven years and each one has been in response to inquiries from reporters requesting her eye-witness account. As a result, we conclude that Liddy has not met his burden of proving each of the five elements of this circuit's limited-purpose public figure test, and we must determine that Wells is not a limited-purpose public figure under our jurisprudence.

The district court also concluded that Wells did not meet the fivepart test governing limited-purpose public figure status. Rather, the district court concluded that Wells was an involuntary public figure. See Wells, 1 F. Supp. 2d at 540. Wells appeals this ruling. We conclude that Wells was not an involuntary public figure and reverse.

The concept of an involuntary public figure has its origins in one sentence from Gertz: "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." 418 U.S. at 345. So rarely have courts determined that an individual was an involuntary public figure that commentators have questioned the continuing existence of that category. See, e.g., Rodney A. Smolla, Law of Defamation § 2.14 (1998) (questioning continuing vitality of involuntary public figure). Although we have acknowledged that involuntary public figures constitute one of the three classes of public figures categorized in Gertz, we have never explored the parameters of the involuntary branch of the public figure typography. See Foretich, 37 F.3d at 1551-52 (stating that Gertz created three categories of public figures, one of which was involuntary). Thus, Wells's challenge to the district court's conclusion that she is an involuntary public figure presents a novel question of law.

The district court ruled, applying Dameron v. Washington Magazine, Inc., 779 F.2d 736 (D.C. Cir. 1985), that Wells was an involuntary public figure because she had the "misfortune" of being "drawn by a series of events into the Watergate controversy." Wells, 1 F. Supp. 2d at 540-41. Because we conclude that "misfortune" is but one aspect of the considerations that should be weighed before concluding that an individual is an involuntary public figure, we are not persuaded by the district court's analysis.

The Dameron case, the leading case on involuntary public figure status, involved an air-traffic controller who had been the sole controller on duty during a 1974 airline crash near Dulles Airport in Northern Virginia. See 779 F.2d at 738. In 1982, following the crash of Air Florida Flight 90 into the Potomac River on takeoff from National Airport in Washington, D.C., Washingtonian magazine published a story that listed, inter alia, plane crashes that had occurred in the Washington, D.C. metropolitan area and were attributable to controller error. See id. The magazine article stated that controller error was partially to blame for the 1974 Dulles crash. See id. As a result, the air traffic controller filed a defamation action. See id.

The D.C. Circuit, recognizing that the air traffic controller had not voluntarily injected himself into a public controversy, and therefore could not satisfy the court's definition of a limited-purpose public figure, concluded that the air traffic controller could nevertheless be considered a public figure. See id. at 741. The court applied two of the three parts of its Waldbaum test for limited-purpose public figures and evaluated whether there was a public controversy and whether the allegedly defamatory statements concerned the public controversy. See id. at 741 (citing Waldbaum v. Fairchild Publications, Inc., 627 F.2d 1287, 1296-98 (D.C. Cir. 1980)). Instead of inquiring whether the air traffic controller had voluntarily entered the fray, however, the D.C. Circuit concluded that "[b]y sheer bad luck" the air traffic controller had become a prominent figure, central to the resolution of a public question. Id. at 742. On that basis, the D.C. Circuit held that the air traffic controller was an involuntary public figure. See id. at 743.

We are hesitant to rest involuntary public figure status upon "sheer bad luck." Gertz tells us that involuntary public figures "must be exceedingly rare," 418 U.S. at 345, and, unfortunately, bad luck is relatively common. The Dameron definition of an involuntary public figure, someone who by bad luck is an important figure in a public controversy, runs the risk of returning us to the Rosenbloom plurality's conception of defamation law. Under Rosenbloom, all defamation plaintiffs were required to prove actual malice when the allegedly defamatory statements occurred during "discussion and communication involving matters of public or general concern, without regard to whether the persons involved are famous or anonymous." 403 U.S. at 44. The Supreme Court expressly repudiated the "public interest" test in Gertz, see 418 U.S. at 346, and further disavowed it in Wolston, see 443 U.S. at 168 ("To accept such reasoning would in effect reestablish the doctrine advanced by the plurality opinion in Rosenbloom . . . which concluded that the New York Times standard should extend to defamatory falsehoods relating to private persons if the statements involved matters of public or general concern. We repudiated this approach in Gertz and in Firestone, however, and we reject it again today."). Because Dameron has not narrowly tailored the class of possible involuntary public figures, it has created a class of individuals who must prove actual malice that is equivalent to the class in Rosenbloom. Under either Dameron or Rosenbloom all individuals defamed during discourse on a matter of public concern must prove actual malice. In light of the Supreme Court's repeated rejection of Rosenbloom, we are unwilling to adopt an approach that returns us to an analysis that is indistinguishable. See David L. Wallis, Note, The Revival of Involuntary Limited-Purpose Public Figures-Dameron v. Washington Magazine, Inc., 1987 B.Y.U. L. Rev. 313, 323 (criticizing the Dameron approach as overinclusive and as a return to Rosenbloom). In order to flesh out the boundaries of who may constitute an involuntary public figure, a return to Gertz, the only Supreme Court case to mention the concept, is in order. In Gertz , Justice Powell iterated two rationales for concluding that public figure status must be the determinative inquiry in the balance of interests between the plaintiff and the First Amendment in a defamation case. First, the public figure can take better advantage of the free press and has an easier time resorting to self-help because notoriety guarantees better access to the media and channels of communication. See 418 U.S. at 344. Second, the public figure has taken actions through which he has voluntarily 47 assumed the risk of publicity. See id. We conclude that in order to ensure that the balance between states' and individuals' interests in protection of reputation and First Amendment freedoms at the heart of Constitutional defamation law is maintained, any standard for determining who is an involuntary public figure must be mindful of both of these underpinnings. See Khawar v. Globe Int'l, Inc., 965 P.2d 696, 702 (Cal. 1998) (holding that the characterization of involuntary public figure must be reserved for those individuals who satisfy both of Gertz's supporting grounds), cert. denied, 119 S. Ct. 1760 (1999). Yet, because the usual and natural conception of a public figure encompasses a sense of voluntary participation in public debate, and because to do otherwise would threaten a return to Rosenbloom, the class of involuntary public figures must be a narrow one, so as to encompass only "exceedingly rare" cases. Gertz, 418 U.S. at 345.

With Gertz's two supporting rationales and the need for a narrow class of involuntary public figures in mind, we believe that the following considerations are warranted. First, to prove that a plaintiff is an involuntary public figure the defendant must demonstrate to the court that the plaintiff has become a central figure in a significant public controversy and that the allegedly defamatory statement has arisen in the course of discourse regarding the public matter. To prove that the plaintiff is a central figure in the controversy, the defendant must put forth evidence that the plaintiff has been the regular focus of media reports26 on the controversy. A significant public controversy is one that touches upon serious issues relating to, for example, community values, historical events, governmental or political activity, arts, education, or public safety. Second, although an involuntary public figure need not have sought to publicize her views on the relevant controversy, she must have nonetheless assumed the risk of publicity. Therefore, the defendant must demonstrate that the plaintiff has taken some action, or failed to act when action was required, in circumstances in which a reasonable person would understand that publicity would likely inhere.27See Reuber v. Food Chemical News, Inc., 925 F.2d 703, 709 (4th Cir. 1991) (en banc) ("[E]ven involuntary participants can be public figures when they choose a course of conduct which invites public attention."). Unlike the limited-purpose public figure, an involuntary public figure need not have specifically taken action through which he has voluntarily sought a primary role in the controversy to influence the outcome of debate on the matter.

To summarize, an involuntary public figure has pursued a course of conduct from which it was reasonably foreseeable, at the time of the conduct, that public interest would arise. A public controversy must have actually arisen that is related to, although not necessarily causally linked, to the action. The involuntary public figure must be recognized as a central figure during debate over that matter. Further, we retain two elements of the five-part Reuber test, specifically: (1) the controversy existed prior to the publication of the defamatory statement; and (2) the plaintiff retained public-figure status at the time of the alleged defamation. See id. at 710. Additionally, to the extent that an involuntary public figure attempts self-help, the Foretich rule must apply with equal strength. See 37 F.3d at 1558.

We believe that the foregoing test captures that"exceedingly rare" individual who, although remaining mute during public discussion of the results of her action, nevertheless has become a principal in an important public matter. Further, this test excludes from the category of involuntary public figures those individuals who by happenstance have been mentioned peripherally in a matter of public interest or have merely been named in a press account. Also, the foregoing analysis avoids resurrecting Rosenbloom because this conception of the involuntary public figure does not cast too broad a net and encompass all individuals who become linked in the media to a matter of public concern. Every plaintiff who is allegedly defamed during discussion of a public controversy will not necessarily be required to prove actual malice to recover compensatory damages under this test.

Applying this formulation to Wells, we determine that she is not an involuntary public figure. Wells simply has not been a central figure in media reports on Watergate. Liddy has been able to point to very few published reports on Watergate even mentioning Wells by name. Prior to the revelation of the call-girl ring theory, Liddy has shown that Wells was mentioned by name only in an International Herald Tribune article noting that her conversations, and those of her boss Spencer Oliver, had been overheard on a listening device. Since the emergence of the call-girl ring theory, in those instances where the media has mentioned Wells, she has been a very minor figure in the discussion of the primary actors in the Watergate affair -the burglars, Dean, Hunt, Liddy, and, of course, President Nixon. We cannot say that in any of the reports contained in the record Wells is portrayed as a central figure in the Watergate controversy. The focus has always been on the roles of other people.

This is true even of Secret Agenda and Silent Coup, the two books that are the primary proponents of the call-girl theory. From the excerpts that are contained in the record, it appears that Wells is mentioned very briefly in Secret Agenda. In the midst of several hundred pages explaining a complex theory regarding CIA manipulation, Wells's name appears on three pages that recount her testimony before the Senate Select Committee and note in passing that Martinez had a key to her desk. See Jim Hougan, Secret Agenda: Watergate, Deep Throat and the CIA 176-79 (1984). Similarly, in Silent Coup, Wells is mentioned on six pages of an approximately five hundred page book that focuses on John Dean. On those pages the contents of Wells's Senate Committee testimony are reiterated, her position as Oliver's secretary is mentioned, and questions about Martinez's possession of her key are posed. See Len Colodny & Robert Gettlin, Silent Coup: The Removal of A President 134, 138, 148, 149, 157, 159 (1991).

Liddy argues that Wells is an important figure in the Watergate controversy because a Watergate burglar possessed a key to her desk. We agree that Martinez's possession of that key is an interesting mystery, and it immediately raises the questions posed in Silent Coup: "Why would a Watergate burglar have a key to Wells's desk in his possession and what items of possible interest to a Watergate burglar were maintained in Wells's locked desk drawer?" Colodny & Gettlin, supra at 159 (emphasis omitted). But, however intriguing the factual revelation regarding Martinez's possession of Wells's key may be, it doesn't establish involuntary public figure status for Wells. The touchstone of involuntary public figure status cannot be mere potential public interest. Designating someone as a public figure in a sense makes her name and her reputation public property. Defamation jurisprudence has managed to strike a delicate, but fair, balance. In order to prevent a chilling effect upon the media's investigation of public events, the media need not even be reasonable in reporting on a public figure. All we require is that the media not be reckless or state knowing falsehoods. The typical public figure has championed her views on a matter of public interest and thereby has assumed the risk of publicity. The public's potential interest in an unknown cannot serve as surrogate to that voluntary engagement in public affairs. Were we to reach such a holding, an individual's interest in privacy and protecting her reputation would be erased from the balance of defamation law.

There is a great temptation when evaluating a controversy as longstanding and significant as Watergate to allow the controversy itself to take precedence in the analysis and let it convert all individuals in its path into public figures. The Supreme Court has admonished us strongly against allowing the public event with which the individual is connected to be the determinative factor governing an individual's public figure designation. In the great wealth of materials on Watergate, Wells is, at most, a footnote. Even within materials promulgating the call-girl theory of the June 17, 1972, break-in, Wells is a minor figure. As a result, we must conclude that she is not an involuntary public figure.28

Based upon the preceding analysis, we must conclude that Wells is a private figure. She need not prove actual malice to recover compensatory damages under the applicable law. Therefore, this case must be remanded for the district court to reconsider Wells's claims under the lesser standard. Because, however, punitive and presumed damages hinge on a finding of actual malice, we proceed below to consider Wells's final assignment of error.

Wells lodges her final challenge against the district court's conclusion that she would be unable to prove that Liddy made the allegedly defamatory statements with knowledge of their falsity or reckless disregard for their falsity, i.e., with actual malice, see New York Times Co. v. Sullivan, 376 U.S. 254, 279-80 (1964), by clear and convincing evidence as a matter of law. Wells claims that the district court erred when it relied on seven pieces of corroborating factual evidence to determine that she could not prove actual malice because there was no proof that Liddy had relied upon those items. Additionally, Wells argues that the unreliability of the primary source for the information coupled with Liddy's personal knowledge of the Watergate affair should be sufficient to create a genuine issue of material fact.

Our review of the evidence adduced below indicates that one can fairly infer that there was only one source for the specific content of Liddy's allegedly defamatory statements. The record amply supports the district court's conclusion that the single source was unreliable. Further, it is apparent that Liddy understood that the source may not be trustworthy. On those bases, we agree with Wells that when the evidence is viewed in the light most favorable to her, she has raised a genuine issue of material fact on the actual malice question. Accordingly, we reverse.

In evaluating whether Wells raised a genuine issue of material fact on the question of whether Liddy had acted with actual malice, the district court thoroughly reviewed the great quantity of evidence submitted to it throughout the course of the proceedings below. As a result of that review, the district court first determined that Bailley was the sole source for all of the variations of the prostitution ring theories of the June 17, 1972, Watergate break-in including both Silent Coup and Secret Agenda. See Wells, 1 F. Supp. 2d at 543. The district court also stated that Bailley, who is a disbarred attorney and convicted felon with a long history of substance abuse and mental illness, had changed his story about the prostitution ring several times and was not a reliable source. See id. Nevertheless, the district court concluded that there were seven pieces of evidence corroborating Bailley's story, and the existence of that evidence prevented Wells from establishing that Liddy acted with actual malice. See id. at 54344.

The district court concluded that the following evidence sufficiently corroborated Bailley's story: (1) although Bailley changed his story several times, he consistently stated that a DNC secretary was involved in the call-girl ring; (2) in 1976 public reports circulated that intimate phone calls were occurring on DNC phones that led to unconfirmed rumors that the phones were being used in a call-girl ring; (3) the FBI found a tap on Oliver's phone; (4) Martinez possessed a key to Wells's desk; (5) it is entirely unclear why anyone would want to break into Wells's desk; (6) Bailley's sister said that Wells had a relationship with Bailley; and (7) there were rumors circulating among DNC staffers after the Watergate break-in regarding a call-girl ring. See id. at 543-45. Because this evidence has a tendency to verify the call-girl theory generally, rather than Wells's participation in the call-girl ring specifically, and because Liddy has confirmed that he does not find believable the sole source connecting Wells specifically to prostitution activities, we cannot conclude that these seven pieces of evidence prevent Wells from proving that Liddy acted with actual malice.

The district court, in summarizing the seven points of corroborating evidence overlooked a very significant fact. Bailley is the only individual who has ever made the claim that there were pictures of prostitutes in a desk at the DNC. Bailley told Liddy during a meeting in June 1991 that "[t]he photographs were kept in a manila envelope that was opened and closed with a string wound around a disc on the flap. The envelope, [which] Bailley stated he saw physically, was kept in a locked desk drawer in the DNC in the Oliver/Wells/Governors area." (J.A. at 1074.) The transcript of the JMU speech shows that Liddy recited this information practically verbatim. The only difference between the report of Liddy's 1991 meeting with Bailley and Liddy's recitation at JMU was that, rather than noting that the photographs of prostitutes were in any desk in the Oliver/Wells/Governors area, Liddy stated during the JMU speech that those photos were in Wells's personal desk.

The seven points of corroborating evidence relied upon by the district court, although they may tend to corroborate the overall Silent Coup theory of the break-in, do not sufficiently corroborate the central issue in this case -whether Wells was personally involved in prostitution activities. Liddy's statements from the JMU speech and the cruise ship speech are capable of defamatory meaning as a matter of law only because they can be fairly understood to connect Wells directly to prostitution activity. If Wells's defamation claims were premised solely upon the fact that Liddy told the Silent Coup story, summary judgment for Liddy would have been appropriate. Liddy, using material that came from Bailley, however, expanded upon the Silent Coup theory in his speeches and said enough so that the ordinary listener could conclude that he explicitly connected Wells personally to prostitution activities.

During his deposition testimony Liddy verified that he was aware that Bailley was not a credible individual. Liddy was aware of Bailley's history of mental illness prior to the publication of Silent Coup in 1991. Liddy had been advised by counsel that he should not rely on Bailley as a sole source for any information because Bailley had "difficulty differentiating between reality and nonreality." (J.A. at 1161.) Liddy also knew that in 1992 Bailley had gotten into an altercation with a security guard at the Library of Congress during which Bailley told the guard that he was Star Caesar and that Bailley believed that the Green River murderers from Washington State were chasing him. In as much as the actual malice standard inquires about the defendant's mental state at the time of publication -whether the defendant knew the statement was false or recklessly disregarded the truth or falsity of the statement -Liddy's knowledge of Bailley's condition is very significant. See Church of Scientology Int'l v. Daniels, 992 F.2d 1329, 1332 (4th Cir. 1993) (defining actual malice). The district court underemphasized Liddy's knowledge of Bailley's unreliability.

Wells's alleged direct involvement in DNC prostitution activities is not sufficiently corroborated by reliable sources to the extent necessary to warrant summary judgment against her. While a defamation plaintiff must meet a clear and convincing standard of proof of actual malice, during summary judgment proceedings initiated by the defendant the court must draw all possible inferences in the plaintiff's favor. See Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 255-56 (1986). Based upon the evidence in the record viewed in the light most favorable to Wells, the inference that Bailley is the only person who has ever linked Wells to prostitution activities is a fair one. Although Secret Agenda and Silent Coup imply that Wells may have been involved in the call-girl operation allegedly catering to DNC visitors, those books do not explicitly state that Wells was directly involved in prostitution-related activities. Further, as the district court noted, for both Secret Agenda and Silent Coup, the source of information on the call-girl ring is Bailley. There is no question that Liddy was aware that Bailley was the primary source for those books. Additionally, during the JMU speech, Liddy quoted virtually verbatim from his report of his 1991 meeting with Bailley, to the extent that he specifically told the audience that the envelope in Wells's desk was one that opened and closed by winding a string around a disc on the flap. The inference that Bailley was Liddy's only source directly connecting Wells to prostitution activity combined with Liddy's acknowledgment that he knew Bailley was not reliable is sufficient to create a genuine issue of material fact regarding whether Liddy acted with actual malice. See Hudnall v. Sellner, 800 F.2d 377, 382 (4th Cir. 1986).

When the evidence is properly viewed in the light most favorable to Wells, the nonmoving party, she is able to raise a genuine issue of material fact regarding whether Liddy acted with actual malice when he published statements that can be fairly understood as linking her to prostitution activities at the DNC. As a result, the district court erred in granting summary judgment to Liddy.

In sum, two statements identified by Wells as potentially defamatory are capable of defamatory meaning: the JMU speech, governed by the substantive law of Virginia, and the cruise ship speech, governed by the maritime common law. Because Wells is a private figure, these two claims must be remanded for trial under the applicable negligence standards for Liddy's culpability. Finally, Wells has succeeded in raising a genuine issue of material fact on the issue of Liddy's actual malice. For these reasons we remand this case for further proceedings consistent with this opinion.