PUBLISHED
UNITED STATES COURT OF APPEALS FOR THE FOURTH CIRCUIT
STEVEN J. HATFILL, Plaintiff-Appellant, v. THE NEW YORK TIMES COMPANY, Defendant-Appellee, No. 04-2561
and NICHOLAS KRISTOF, Defendant. Appeal from the United States District Court for the Eastern District of Virginia, at Alexandria. Claude M. Hilton, Chief District Judge. (CA-04-807-A)
Argued: May 24, 2005
Decided: July 28, 2005
Before WILKINS, Chief Judge, and NIEMEYER and SHEDD, Circuit Judges.
Reversed and remanded by published opinion. Judge Shedd wrote the majority opinion, in which Chief Judge Wilkins joined. Judge Nie- meyer wrote a dissenting opinion.
COUNSEL
ARGUED: Christopher J. Wright, HARRIS, WILTSHIRE & GRAN- NIS, L.L.P., Washington, D.C., for Appellant. David A. Schulz, 2 HATFILL v. THE NEW YORK TIMES LEVINE, SULLIVAN, KOCH & SCHULZ, L.L.P., New York, New York, for Appellee. ON BRIEF: Thomas G. Connolly, Mark A. Grannis, HARRIS, WILTSHIRE & GRANNIS, L.L.P., Washington, D.C., for Appellant. Jay Ward Brown, LEVINE, SULLIVAN, KOCH & SCHULZ, L.L.P., New York, New York; David E. McCraw, THE NEW YORK TIMES COMPANY, New York, New York, for Appel- lee.
OPINION
SHEDD, Circuit Judge:
Dr. Steven J. Hatfill sued The New York Times Company ("The Times") and columnist Nicholas Kristof, alleging claims under Vir- ginia law for defamation and intentional infliction of emotional dis- tress. Hatfill’s claims arise from The Times’ publication of a series of Kristof’s columns concerning the federal investigation into the mailing of letters laced with anthrax in the fall of 2001. The district court dismissed Hatfill’s complaint under Fed. R. Civ. P. 12(b)(6), and Hatfill now appeals. We conclude that Hatfill has adequately pled the elements of his claims under Virginia law, and we reverse the rul- ing of the district court.
I.
In the fall of 2001, shortly after the terrorist attacks on the World Trade Center and the Pentagon, someone mailed letters laced with anthrax to several news organizations and members of Congress.1 At least five people died as a result of contact with these letters, and the federal government launched an investigation to identify and capture the responsible party or parties. By May 2002, the Federal Bureau of Investigation ("FBI") had not made any arrests.
Kristof writes a regular column for the editorial page of The Times. 1 Because we are reviewing an order granting dismissal under Rule 12(b)(6), we accept as true the allegations in Hatfill’s complaint. See Papasan v. Allain, 478 U.S. 265, 283 (1986). HATFILL v. THE NEW YORK TIMES 3 During the spring and summer of 2002, Kristof wrote several columns criticizing the FBI’s investigation. From May through July 2002, Kristof focused his attention on the FBI’s handling of information related to a man he called "Mr. Z." According to Kristof, circumstan- tial evidence pointed to Mr. Z, who was widely suspected by other scientists of involvement in the anthrax mailings. In Kristof’s opinion, the FBI had not moved aggressively enough against Mr. Z. In August 2002, Kristof identified Mr. Z as Dr. Steven J. Hatfill, a research sci- entist employed by the Department of Defense.
A.
Kristof’s columns expressed opinions about the progress of the FBI’s investigation based on factual assertions concerning Hatfill. In a column published on May 24, 2002, Kristof urged his readers to "light a fire" under the FBI in its investigation of the anthrax mailings since "[e]xperts in the bioterror field are already buzzing about a handful of individuals who had the ability, access and motive to send the anthrax." According to Kristof, these experts suspected "one middle-aged American who has worked for the United States military biodefense program and had access to the labs at Fort Detrick, Md. His anthrax vaccinations are up to date, he unquestionably had the ability to make first-rate anthrax, and he was upset at the United States government in the period preceding the anthrax attack." According to Kristof, the FBI had been "painstakingly slow in its investigation" of this person and unnamed others. J.A. 22.
Kristof repeated this theme in a column published on July 2, 2002, writing that "the bureau’s lackadaisical ineptitude in pursuing the anthrax killer continues to threaten America’s national security by permitting him to strike again or, more likely, to flee to Iran or North Korea." As to the identity of this killer, Kristof offered the following:
Some in the biodefense community think they know a likely culprit, whom I’ll call Mr. Z. Although the bureau has poly- graphed Mr. Z, searched his home twice and interviewed him four times, it has not placed him under surveillance or asked its outside handwriting expert to compare his writing to that on the anthrax letters. 4 HATFILL v. THE NEW YORK TIMES ....
. . . People in the biodefense field first gave Mr. Z’s name to the bureau as a suspect in October, and I wrote about him elliptically in a column on May 24.
He denies any wrongdoing, and his friends are heartsick at suspicions directed against a man they regard as a patriot. Some of his polygraphs show evasion, I hear, although that may be because of his temperament.
If Mr. Z were an Arab national, he would have been imprisoned long ago. But he is a true-blue American with close ties to the U.S. Defense Department, the C.I.A. and the American biodefense program. On the other hand, he was once caught with a girlfriend in a biohazard "hot suite" at Fort Detrick, surrounded only by blushing germs.
Kristof argued that the FBI’s handling of this information reflected a casual approach to the investigation. "With many experts buzzing about Mr. Z behind his back, it’s time for the F.B.I. to make a move: either it should go after him more aggressively, sifting thoroughly through his past and picking up loose threads, or it should seek to exculpate him and remove this cloud of suspicion." J.A. 23.
Having called the FBI to account for the slow pace of its investiga- tion, Kristof put a series of rhetorical questions to the FBI concerning Mr. Z particularly:
Do you know how many identities and passports Mr. Z has and are you monitoring his international travel? I have found at least one alias for him, and he has continued to travel abroad on government assignments, even to Central Asia.
Why was his top security clearance suspended in August, less than a month before the anthrax attacks began? This move left him infuriated. Are the C.I.A. and military intelli- gence agencies cooperating fully with the investigation? HATFILL v. THE NEW YORK TIMES 5 Have you searched the isolated residence that he had access to last fall? The F.B.I. has known about this building, and knows that Mr. Z gave Cipro to people who visited it. This property and many others are legally registered in the name of a friend of Mr. Z, but may be safe houses operated by American intelligence.
Have you examined whether Mr. Z has connections to the biggest anthrax outbreak among humans ever recorded, the one that sickened more than 10,000 black farmers in Zimba- bwe in 1978-80? There is evidence that the anthrax was released by the white Rhodesian Army fighting against black guerillas, and Mr. Z has claimed that he participated in the white army’s much-feared Selous Scouts. Could rogue elements of the American military have backed the Rhodesian Army in anthrax and cholera attacks against blacks? Mr. Z’s resume also claims involvement in the for- mer South African Defense Force; all else aside, who knew that the U.S. Defense Department would pick an American who had served in the armed forces of two white-racist regimes to work in the American biodefense program with some of the world’s deadliest germs?
J.A. 23.
In his July 12, 2002 column, Kristof suggested that Mr. Z might have been involved in a previous attack against B’nai B’rith offices in April 1997:
When someone expert in bio[ ]warfare mailed anthrax last fall, it may not have been the first time he had struck.
So while the F.B.I. has been unbelievably lethargic in its investigation so far, any year now it will re-examine the package that arrived on April 24, 1997, at the B’nai B’rith headquarters in Washington. The package contained a petri dish mislabeled "anthracks." The dish did not contain anthrax. But a Navy lab determined that it was bacillus cereus, a very close, non-toxic cousin of anthrax used by the Defense Department. 6 HATFILL v. THE NEW YORK TIMES Anybody able to obtain bacillus cereus knew how to spell "anthrax." An echo of that deliberate misspelling came last fall when the anthrax letters suggested taking "penacillin."
The choice of B’nai B’rith probably was meant to suggest Arab terrorists, because the building had once been the tar- get of an assault by Muslim gunmen. In the same way, F.B.I. profilers are convinced that the real anthrax attacks last year were conducted by an American scientist trying to pin the blame on Arabs.
In a column on July 2 I wrote about "Mr. Z," an American bio[ ]defense insider who intrigues investigators and whose career has been spent in the shadowy world of counterterror and intelligence. He denies any involvement in the anthrax attacks.
On the date the perpetrator chose for the B’nai B’rith attack, a terrorism seminar was under way in the Washing- ton area and Mr. Z seemed peeved that neither he nor any other bio[ ]defense expert had been included as a speaker. The next day, Mr. Z sent a letter to the organizer saying that he was "rather concerned" at the omission and added: "As was evidenced in downtown Washington D.C. a few hours later, this topic is vital to the security of the United States. I am tremendously interested in becoming more involved in this area. . . ."
Over the next couple of years, Mr. Z has used the B’nai B’rith attack to underscore the importance of his field and his own status within it. "Remember B’nai B’rith," he noted at one point. In examples he gave of how anthrax attacks might happen, he had a penchant for dropping Arab names.
The F.B.I. must be on top of the B’nai B’rith episode, right? Well, it was told about it months ago. But B’nai B’rith says it hasn’t been asked about the incident by the F.B.I. HATFILL v. THE NEW YORK TIMES 7 Kristof also suggested that Mr. Z might have been involved with another set of "anthrax hoaxes" in February 1999:
In contrast to the 1997 package with fake anthrax gelatin, the 1999 letters each contained a teaspoon of fake anthrax powder (roughly the same amount as of real anthrax in 2001). That’s interesting because as of 1997, U.S. bio[ ]defense scientists were working basically only with wet anthrax, while by 1999 some had experimented with making powders.
For example, Mr. Z apparently learned about powders during those two years. His 1999 resume adds something missing from the 1997 version: "working knowledge of wet and dry BW biological warfare agents, large-scale produc- tion of bacterial, rickettsial and viral BW pathogens and tox- ins."
J.A. 24-25.
The next week Kristof wrote that Mr. Z had been interviewed by the FBI four times and that his home had been searched twice during the course of the investigation. Kristof noted that the Army had hired Mr. Z in 1997 to work with Ebola and Marburg viruses, even though he had previously worked with the armed forces of Rhodesia and apartheid South Africa. J.A. 27.
Finally, on August 13, 2002, Kristof identified his Mr. Z as Dr. Steven J. Hatfill:
It’s time for me to come clean on "Mr. Z."
Since May, I’ve written periodically about a former U.S. Army scientist who, authorities say privately, has become the overwhelming focus of the investigation into the anthrax attacks last fall. I didn’t name him.
But over the weekend, Mr. Z named himself. He is Steven J. Hatfill, 48, a prominent germ warfare specialist who for- 8 HATFILL v. THE NEW YORK TIMES merly worked in the Army labs at Fort Detrick, Md. Hatfill made a televised statement on Sunday, describing himself as "a loyal American" and attacking the authorities and the media for trying "to smear me and gratuitously make a wasteland of my life."
The first thing to say is that the presumption of innocence has already been maimed since 9/11 for foreign Muslims, and it should not be similarly cheapened with respect to Dr. Hatfill. It must be a genuine assumption that he is an inno- cent man caught in a nightmare. There is not a shred of tra- ditional physical evidence linking him to the attacks.
Still, Dr. Hatfill is wrong to suggest that the F.B.I. has casually designated him the anthrax "fall guy." The authori- ties’ interest in Dr. Hatfill arises from a range of factors, including his expertise in dry biological warfare agents, his access to Fort Detrick labs where anthrax spores were kept (although he did not work with anthrax there) and the ani- mus to some federal agencies that shows up in his private writings. He has also failed three polygraph examinations since January, and canceled plans for another polygraph exam two weeks ago.
So far, the only physical evidence is obscure: smell. Spe- cially trained bloodhounds were given scent packets pre- served from the anthrax letters and were introduced to a variety of people and locations. This month, they responded strongly to Dr. Hatfill, to his apartment, to his girlfriend’s apartment and even to his former girlfriend’s apartment, as well as to restaurants that he had recently entered (he is under constant surveillance). The dogs did not respond to other people, apartments or restaurants.
Putting aside the question of Dr. Hatfill and the anthrax, there are two larger issues.
First is the F.B.I.’s initial slowness in carrying out the anthrax investigation. Why did it take nine months to call in the bloodhounds, or to read Dr. Hatfill’s unpublished novel, HATFILL v. THE NEW YORK TIMES 9 "Emergence," which has been sitting in the copyright office since 1998 and draws on his experiences in South Africa and Antarctica to recount a biological warfare attack on Congress?
Second is the need for much greater care within the U.S. biodefense program. Dr. Hatfill’s resume made claims (a Ph.D. degree, work with the U.S. Special Forces, member- ship in Britain’s Royal Society of Medicine) that appear false, but they were never checked.
Moreover, what was a man like Dr. Hatfill who had served in the armed forces of two white racist governments (Rhodesia and South Africa) doing in a U.S. Army lab working with Ebola? . . .
....
To its credit, in the last few months, the bureau has finally picked up its pace. Its experts in Quantico are belatedly examining anthrax hoax letters sent in 1997 and 1999 that bear fascinating resemblances to the real anthrax letters. Investigators are looking at another hoax letter with intrigu- ing parallels to the real one; that hoax letter was sent to Sen- ator Tom Daschle from London in mid-November, when Dr. Hatfill was visiting a biodefense center in England.
Partly because of the newfound energy, the F.B.I. has lately been enjoying genuine progress in its anthrax investi- gation. People very close to Dr. Hatfill are now cooperating with the authorities, information has been presented to a grand jury, and there is reason to hope that the bureau may soon be able to end this unseemly limbo by either exculpat- ing Dr. Hatfill or arresting him.
J.A. 28-29. As this column illustrates, Kristof’s argument about the progress of the FBI’s investigation of the anthrax mailings had much to do with specific allegations concerning Hatfill. 10 HATFILL v. THE NEW YORK TIMES B.
On June 18, 2003, Hatfill filed suit against Kristof and The Times in Virginia state court. The complaint in that case alleged that "Defen- dants’ false and reckless public identification of Dr. Hatfill as the likely anthrax mailer imputed homicidal activity to Dr. Hatfill and impugned his good name as a physician and bio[ ]medical researcher, and thereby constituted, separately, defamation, defamation per se, and defamation by false light." In addition, the complaint alleged that Kristof’s "intentional public denunciation of Dr. Hatfill as the likely anthrax murderer regardless of whether Dr. Hatfill was guilty or inno- cent" constituted intentional infliction of emotional distress. Hatfill sought at least $1 million in compensatory damages, the same amount in punitive damages, and costs. Hatfill never served this complaint on the defendants and instead took a voluntary nonsuit on March 9, 2004.
Hatfill filed this lawsuit on July 13, 2004, asserting claims for defa- mation and intentional infliction of emotional distress. Count One alleges that "Defendants’ false and reckless public identification of Dr. Hatfill with the anthrax mailings, both directly and by implication from the manner in which his personal and professional background were presented in the ‘Mr. Z’ columns, constituted a false factual alle- gation of terrorist and homicidal activity and impugned Dr. Hatfill’s good name as a citizen, a physician and a bio[ ]medical researcher to a reasonable reader." Count Two alleges that each of eleven discrete factual statements in Kristof’s columns "constituted defamation per se that, in the mind of the reasonable reader, would tend to incriminate Dr. Hatfill in the anthrax mailings." Count Three alleges that Kristof’s suggestion that Hatfill was responsible for the anthrax attacks of 2001 constituted intentional infliction of emotional distress. Hatfill seeks an unspecified amount of compensatory and punitive damages.
Hatfill voluntarily dismissed Kristof as a defendant in this case when it became clear that the district court lacked personal jurisdic- tion over him. The district court later dismissed Hatfill’s complaint against The Times under Rule 12(b)(6). According to the district court, Count One failed as a matter of law because Kristof’s columns, when read in their entirety and in context, could not reasonably be read as accusing Hatfill of being responsible for the anthrax attacks. HATFILL v. THE NEW YORK TIMES 11 The columns merely reported on an ongoing investigation that tar- geted Hatfill, and Kristof was careful to disavow any conclusion of Hatfill’s guilt. Count Two failed because the claims concerning dis- crete statements in Kristof’s columns were not preserved in the state- court complaint and thus are time-barred. Even if they were timely, the district court ruled that none of these discrete factual allegations is sufficient to constitute defamation. Finally, the district court dis- missed Count three of the complaint on the grounds that publication of commentary on a matter of public concern is not outrageous con- duct and Hatfill had failed to allege sufficiently severe emotional dis- tress. This appeal followed.
II.
We review de novo the district court’s dismissal of Hatfill’s com- plaint. A complaint should not be dismissed under Rule 12(b)(6) "un- less it appears beyond doubt that the plaintiff can prove no set of facts in support of his claim which would entitle him to relief." Conley v. Gibson, 355 U.S. 41, 45-46 (1957). We must accept as true all well- pleaded allegations and view the complaint in the light most favorable to Hatfill. See Papasan, 478 U.S. at 283.
The district court stated that "[t]his standard is to be applied with particular care" in reviewing defamation claims. To the extent that the district court applied a stricter standard to Hatfill’s complaint than the ordinary standards under Rule 12(b)(6), that was error. A defamation complaint, like any other civil complaint in federal court, must pro- vide "a short and plain statement of the claim showing that the pleader is entitled to relief," Fed. R. Civ. P. 8(a)(2), sufficient to "give the defendant fair notice of what the plaintiff’s claim is and the grounds upon which it rests," Conley, 355 U.S. at 47. While the Fed- eral Rules of Civil Procedure require more specific pleading in certain cases, defamation cases are not among them. See Fed. R. Civ. P. 9(b). Thus, the usual standards of notice pleading apply in defamation cases such as this one. See Swierkiewicz v. Sorema N.A., 534 U.S. 506, 510-14 (2002) (rejecting the argument that a heightened pleading standard should apply in employment discrimination cases); Leather- man v. Tarrant County Narcotics Intelligence & Coordination Unit, 507 U.S. 163, 168 (1993) (rejecting the argument that a heightened pleading standard should apply in § 1983 actions against municipali- 12 HATFILL v. THE NEW YORK TIMES 2 ties). We review the allegations of Hatfill’s complaint to determine whether they state claims upon which relief may be granted under Virginia law.3
Count One alleges that The Times’ publication of Kristof’s col- umns defamed Hatfill by implying that Hatfill was involved in the anthrax mailings. Under Virginia law, a plaintiff seeking to recover for defamation per se must allege a publication of false information concerning the plaintiff that tends to defame the plaintiff’s reputation. See Chapin v. Knight-Ridder, Inc., 993 F.2d 1087, 1092 (4th Cir. 1993) (citing The Gazette, Inc. v. Harris, 325 S.E.2d 713 (Va. 1985)). The district court dismissed Count One because "[t]he columns are not reasonably capable of being understood to convey either an accu- sation that plaintiff is the anthrax mailer, or an intention by the author to make such an accusation."4 2 The leading commentators have recognized that "[t]he notion that a federal court can use a stricter pleading standard on a motion to dismiss a disfavored action has been cast in serious doubt — indeed, it may well have been rendered invalid by the Supreme Court’s 1993 decision in [Leatherman] . . . and its 2002 decision in [Swierkiewicz] . . . . In both cases the Court makes it clear that Rule 8 announces a pleading standard that is applicable to all cases except those governed by Rule 9(b) or a heightened pleading requirement in a federal statute." 5B Charles A. Wright & Arthur R. Miller, Federal Prac. & Proc.: Civil 3d § 1357, at 732 (3d ed. 2004). 3 Although there was some question whether Hatfill’s claims were gov- erned by the law of Virginia or New York, the parties proceeded on the assumption that Virginia law governs. The district court applied Virginia law, and the parties do not suggest that we do otherwise. 4 We take as true Hatfill’s allegations that both the implication of involvement in the anthrax mailings and the particular factual assertions (identified in Count Two) from which that implication arises are false. See Chapin, 993 F.2d at 1092 ("On a motion to dismiss a libel suit because of no actionable statement, the court must of course credit the plaintiff’s allegation of the factual falsity of a statement."). Moreover, we note that all of the columns at issue "concern" Hatfill, even though only one of the columns actually identifies him by name. See The Gazette, HATFILL v. THE NEW YORK TIMES 13 The question whether a statement is capable of having a defama- tory meaning is a question of law to be decided by the court. Yeagle v. Collegiate Times, 497 S.E.2d 136, 138 (Va. 1998). Under Virginia law, the following kinds of statements are actionable as defamation per se: (1) statements that "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) statements that "impute that a person is infected with some contagious disease, where if the charge is true, it would exclude the party from society," (3) statements that "impute to a person unfitness to perform the duties of an office or employment of profit, or want of integrity in the dis- charge of duties of such an office or employment," and (4) statements that "prejudice such person in his or her profession or trade." Carwile v. Richmond Newspapers, Inc., 82 S.E.2d 588, 591 (Va. 1954).
It is not always clear whether particular words actually charge a person with a crime of moral turpitude or unfitness for employment or the like, but the general rule of interpretation is that "allegedly defamatory words are to be taken in their plain and natural meaning and to be understood by courts and juries as other people would understand them, and according to the sense in which they appear to have been used." Id. at 591-92. A defamatory charge may be made expressly or by "inference, implication or insinuation." Id. at 592. In short, "it matters not how artful or disguised the modes in which the meaning is concealed if it is in fact defamatory." Id. Accordingly, we have stated that courts applying Virginia defamation law should con- sider not only the words themselves but also the "inferences fairly attributable" to them. Wells v. Liddy, 186 F.3d 505, 523 (4th Cir. 1999).
Inc., 325 S.E.2d at 738 (stating that a plaintiff alleging defamation "need not show that he was mentioned by name in the publication" and that it is sufficient to show that "the publication was in its description or identi- fication such as to lead those who knew or knew of the plaintiff to believe that the article was intended to refer to him"); see also WJLA-TV v. Levin, 564 S.E.2d 383, 390 (Va. 2002) (noting that a plaintiff may rely upon statements made before his actual identification, so long as such statements were made by the same defendant concerning the same sub- ject or event over a short period of time). 14 HATFILL v. THE NEW YORK TIMES "In determining whether the words and statements complained of . . . are reasonably capable of the meaning ascribed to them by innu- endo, every fair inference that may be drawn from the pleadings must be resolved in the plaintiff’s favor." Carwile, 82 S.E.2d at 592. Even so, "the meaning of the alleged defamatory language cannot, by innu- endo, be extended beyond its ordinary and common acceptation. The province of the innuendo is to show how the words used are defama- tory, and how they relate to the plaintiff, but it cannot introduce new matter, nor extend the meaning of the words used, or make that cer- tain which is in fact uncertain." Id.
Hatfill contends that Kristof’s columns defamed him by imputing to him the commission of crimes of moral turpitude, namely, the mur- ders of five people who were exposed to the anthrax letters. If the col- umns fairly can be read to make such a charge, then they are defamatory per se. Cf. Schnupp v. Smith, 457 S.E.2d 42, 45-47 (Va. 1995) (holding that a charge of aiding and abetting in the possession of narcotics was defamatory per se); Great Coastal Express, Inc. v. Wellington, 334 S.E.2d 846, 849 (Va. 1985) (holding that a charge of commercial bribery was defamatory per se). "In determining whether or not the language does impute a criminal offense the words must be construed in the plain and popular sense in which the rest of the world would naturally understand them." Schnupp, 457 S.E.2d at 46 (inter- nal quotations omitted). Thus, while Kristof’s columns need not "make the charge in express terms," they must "naturally and presum- ably be understood by those who [read] them as charging a crime." Id. (internal quotations omitted).
The plaintiff in Carwile was an attorney who made charges of graft in the Richmond, Virginia police department. 82 S.E.2d at 589. After the grand jury declined to return an indictment, the defendant newspa- per reported that two city officials declined to comment on whether they were considering a recommendation to the Virginia State Bar that Carwile be disciplined for making the accusation. Id. The news- paper then reported that "[u]nder the State Code, the State Bar as an administrative agency of the Supreme Court of Appeals may request a court of competent jurisdiction to disbar an attorney for violation of the ethical code governing the professional conduct of attorneys." Carwile, 82 S.E.2d at 589-90. The newspaper did not argue explicitly HATFILL v. THE NEW YORK TIMES 15 that the state bar should take action, only that it was authorized by law to do so. The Supreme Court of Virginia ruled that
it is a reasonable implication of this language, read in con- nection with the whole article, that the plaintiff is guilty of unethical and unprofessional conduct for his charges made against the Police Department; for which conduct the defen- dant suggests in a veiled but pointed way that the plaintiff could and should be subjected to disbarment proceedings . . . . While the defamatory language does not in express terms charge the plaintiff with a breach of his professional honor, yet, when aided by the innuendo, operating within the scope of its legitimate functions, it does impute conduct tending to injure him in his profession.
Id.
The plaintiff in Schnupp alleged defamation arising from a police officer’s statements to the plaintiff’s employer suggesting that the plaintiff had committed a crime punishable by imprisonment. 457 S.E.2d at 43. During a surveillance in a high-drug-crime area, the officer had observed what he believed were narcotics transactions occurring near a company van driven by the plaintiff. Id. at 43-44. The officer later contacted the plaintiff’s employer and reported what he had seen. Specifically, the officer told the employer that he had seen the plaintiff driving a van to a particular location, that his pas- senger got out of the van, and that his passenger then made some type of exchange. Id. at 45. The officer argued that his statements merely described his observation that the plaintiff was present in the van and did not suggest that the plaintiff had actually bought or sold drugs. Since mere presence at the scene of a crime is not itself a crime of moral turpitude, his statements could not be defamatory. Id. Rejecting this argument, the Supreme Court of Virginia held that the officer’s statement to the employer, "unaided by innuendo but assisted by the reasonable inferences to be drawn from the words used, while not charging a criminal offense in express terms, did impute to [the plain- tiff] the commission of the offense of aiding and abetting in the pos- session of narcotics." Schnupp, 457 S.E.2d at 46.
We applied this Virginia law in Wells, where a former secretary at the Democratic National Committee sued radio personality and 16 HATFILL v. THE NEW YORK TIMES Watergate conspirator G. Gordon Liddy, alleging that he had defamed her by telling public audiences that she helped procure prostitutes for visitors to party headquarters in Washington. 186 F.3d at 512-18. During one speech, Liddy stated that the surveillance camera posi- tioned at the nearby Howard Johnson’s motel
looked directly down at a desk of a secretary named Maxine Wells, and her telephone. They had a telescopic lens camera pointed at that. And that is where the wiretap was subse- quently found by the [D]emocrats on that phone. . . . Some members of the DNC were using the call girl ring as an asset to entertain visiting firemen. And to that end that 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.
Id. at 523. We concluded that these words were "capable of defama- tory meaning, namely, that Wells was a participant in a scheme to procure prostitutes." Id. Liddy had not explicitly accused Wells of complicity in the alleged call-girl operation, but he implied as much by asserting that the manila envelope was kept in her desk.
Considered in light of Carwile, Schnupp, and Wells, Hatfill’s com- plaint adequately alleges that Kristof’s columns, taken together, are capable of defamatory meaning. The columns did not describe any other actual or potential target of investigation, and they recounted detailed information pertaining to Hatfill alone. Once Kristof named Hatfill as Mr. Z (and perhaps even before that time), a reasonable reader of his columns could believe that Hatfill had the motive, means, and opportunity to prepare and send the anthrax letters in the fall of 2001; that he had particular expertise with powder forms of anthrax, the type used in the mailings; that his own anthrax vaccina- tions were current; that he was the prime suspect of the biodefense HATFILL v. THE NEW YORK TIMES 17 community as well as federal investigators; that he had failed numer- ous polygraph examinations; that specially trained bloodhounds had "responded strongly" to Hatfill, his apartment, and his girlfriend’s apartment but not to anyone else or any other location; and that Hatfill was probably involved in similar anthrax episodes in recent years. Based on these assertions, a reasonable reader of Kristof’s columns likely would conclude that Hatfill was responsible for the anthrax mailings in 2001.5
Notwithstanding Kristof’s attribution of certain allegations to unnamed sources, or his caution that readers should entertain a pre- sumption of Hatfill’s innocence, or even his statement that the FBI should "end this unseemly limbo by either exculpating Dr. Hatfill or arresting him," the unmistakable theme of Kristof’s columns is that the FBI should investigate Hatfill more vigorously because all the evi- dence (known to Kristof) pointed to him. Just as a defendant cannot escape liability for making a false assertion of fact by prefacing that assertion with the words "in my opinion," Milkovich v. Lorain Jour- nal Co., 497 U.S. 1, 18-19 (1990),6 neither can it escape liability sim- 5 At this stage of the litigation, there is no evidence to show whether or to what extent Kristof’s columns were, as the district court stated, "ac- curate report[s] of [an] ongoing investigation." A court considering whether a statement is capable of defamatory meaning must confine itself to the allegedly defamatory statement and the fair inferences and innuendoes derived from it. See Schnupp, 457 S.E.2d at 46; Carwile, 82 S.E.2d at 591-92. In this case, the columns themselves show that Kris- tof’s thesis was not that the official investigation targeted Hatfill; rather, it was that the official investigation should be targeting Hatfill more vig- orously, if not exclusively, because the available evidence pointed to him. In describing all this evidence, Kristof’s columns did not merely report others’ suspicions of Hatfill; they actually generated suspicion by asserting facts that tend to implicate him in the anthrax murders. 6 The Court in Milkovich made clear that a statement may be defama- tory even if it seems merely to state an opinion: If a speaker says, "In my opinion John is a liar," he implies a knowledge of facts which lead to the conclusion that Jones told an untruth. Even if the speaker states the facts upon which he bases his opinion, if those facts are either incorrect or incom- plete, or if his assessment of them is erroneous, the statement 18 HATFILL v. THE NEW YORK TIMES ply by pairing a charge of wrongdoing with a statement that the subject must, of course, be presumed innocent, see Carwile, 82 S.E.2d at 592 (stating that "it matters not how artful or disguised the modes in which the meaning is concealed it if is in fact defamatory").
For purposes of Rule 12(b)(6), the question is simply whether Kris- tof’s columns are capable of defamatory meaning under Virginia law, i.e., whether they imputed to Hatfill the commission of a crime involving moral turpitude. They did, and much more forcefully than the statements at issue in Carwile and Wells. Because Kristof’s col- umns, taken together, are capable of defamatory meaning under Vir- ginia law, the district court erred in dismissing Count One.7
may still imply a false assertion of fact. Simply couching such statements in terms of opinion does not dispel these implications; and the statement, "In my opinion Jones is a liar," can cause as much damage to reputation as the statement, "Jones is a liar." 497 U.S. at 18-19. Although the Court declined to create a rigid distinc- tion between statements of fact and expressions of opinion, it did note that "a statement on matters of public concern must be provable as false before there can be liability under state defamation law, at least in situa- tions . . . where a media defendant is involved." Id. at 19-20. The alleg- edly defamatory charge in this case — that Hatfill was responsible for the anthrax mailings — is provable as false and thus may be the subject of a defamation claim under Milkovich. 7 Contrary to the district court’s assertion, it is immaterial whether Kristof actually intended to defame Hatfill. We stated in Chapin that "a libel-by-implication plaintiff must make an especially rigorous showing where the expressed facts are literally true. The language must not only be reasonably read to impart the false innuendo, but it must also affirma- tively suggest that the author intends or endorses the inference." 993 F.2d at 1092-93 (emphasis added). In this case, Hatfill alleges both that the inference — that he was responsible for the anthrax mailings — is false and that the factual assertions from which that inference arises are false. Since the district court was required to accept Hatfill’s assertion that the facts upon which Kristof based his defamatory charge were false, id. at 1092, Chapin is inapposite. HATFILL v. THE NEW YORK TIMES 19 B.
Count Two alleges that each of eleven discrete factual assertions contained in Kristof’s columns separately defamed Hatfill by incrimi- nating him in the anthrax mailings. The district court dismissed this count on the grounds that (1) the statute of limitations barred any claims other than the claim asserted in Count One and (2) none of the eleven statements is independently capable of defamatory meaning.
1.
Hatfill’s defamation claims are subject to a one-year statute of limi- tations. See Va. Code Ann. § 8.01-247.1. Hatfill’s claims accrued when Kristof’s columns were published — between May and August 2002. See Jordan v. Shands, 500 S.E.2d 215, 218 (Va. 1998). Thus, Hatfill had until August 2003 (at the latest) to assert his defamation claims. Yet he did not file this complaint until July 2004.
Under Virginia law, however, a plaintiff who voluntarily with- draws a state-court action and recommences it in federal court within six months may toll the statute of limitations as of the date he filed the initial state-court action. Va. Code Ann. § 8.01-229(E)(3).8 Hatfill filed a lawsuit in state court on June 18, 2003; took a nonsuit on March 9, 2004; and commenced this action in federal court on July 13, 2004. Thus, the one-year statute of limitations was tolled as of June 18, 2003.
The district court ruled, however, that this tolling provision saved 8 Section 8.01-229(E)(3) provides as follows: "If a plaintiff suffers a voluntary nonsuit as prescribed in § 8.01-380, the statute of limitations with respect to such actions shall be tolled by the commencement of the nonsuited action, and the plaintiff may recommence his action within six months from the date of the order entered by the court, or within the original period of limitation, or within the limitation period as provided by subdivision (B)(1), whichever period is longer. This tolling provision shall apply irrespective of whether the action is originally filed in a fed- eral or state court and recommenced in any other court, and shall apply to all actions irrespective of whether they arise under common law or statute." 20 HATFILL v. THE NEW YORK TIMES only the claim relating to the overall implication of Kristof’s columns and not the claim relating to discrete false statements. According to the district court, since the initial complaint asserted only a single cause of action for defamation alleging that "[d]efendants’ false and reckless public identification of Dr. Hatfill as the likely anthrax mai- ler imputed homicidal activity to Dr. Hatfill and impugned his good name," only that claim was saved by the tolling provision.
Section 8.01-229(E)(3), the tolling provision at issue here, applies when the plaintiff "suffers a voluntary nonsuit as prescribed in § 8.01- 380." Section 8.01-380 authorizes a plaintiff to take a nonsuit on any "cause of action," and the Virginia Supreme Court defines a "cause of action" as "a set of operative facts which, under the substantive law, may give rise to one or more rights of action," Roller v. Basic Constr. Co., 384 S.E.2d 323, 326 (Va. 1989).9
When Hatfill took a voluntary nonsuit in his original state-court action, he did so with respect to the set of operative facts underlying his complaint, namely, the defendants’ publication of Kristof’s col- umns. When he filed this lawsuit in federal court, the tolling provision saved all rights of action arising from that cause of action, including the right of action alleged in Count Two. See Odeneal v. Thompson, 2003 WL 22518523, at *2 (Va. Cir. Ct. Aug. 6, 2003). Thus, the dis- trict court erred in ruling that the statute of limitations barred Count Two of Hatfill’s complaint.
2.
The district court further erred in ruling that none of the eleven statements identified by Hatfill is independently capable of defama- tory meaning. Broadly grouped, Hatfill complains about five sets of 9 The Virginia Supreme Court distinguishes causes of action from rights of action: "A right of action is a remedial right to presently enforce a cause of action. There can be no right of action until there is a cause of action." Stone v. Ethan Allen, Inc., 350 S.E.2d 629, 631 (Va. 1986) (internal citations omitted). "Unless otherwise provided by statute, tradi- tional statutes of limitations begin to run, not when a wrongful act is done, but when injury or damage results from it and the cause of action has thus ripened into a right of action." Roller, 384 S.E.2d at 327-28. HATFILL v. THE NEW YORK TIMES 21 factual assertions in Kristof’s columns. First, Hatfill complains about Kristof’s assertion that he had the ability, access, and motive to make and send the anthrax. Second, Hatfill complains about Kristof’s state- ment that he had access to an "isolated residence" where he gave Cipro to visitors. Third, Hatfill complains about the allegation that he had up-to-date anthrax vaccinations himself. Fourth, Hatfill com- plains about Kristof’s charge that he failed three polygraph examina- tions in 2002. Finally, Hatfill complains about the allegation that he "was once caught with a girlfriend in a biohazard ‘hot suite’ at Fort Detrick . . . surrounded only by blushing germs."
Taken in the context of the columns in which they appear, and con- sidered in light of Carwile, Schnupp, and Wells, all of these state- ments but one are capable of incriminating Hatfill in the anthrax mailings. With the exception of the final statement — that Hatfill had been caught with a girlfriend in a biohazard hot suite at Fort Detrick — these statements link Hatfill to anthrax generally and the investiga- tion specifically and give rise to an inference that he was involved in the anthrax mailings. By contrast, the statement describing Hatfill’s activities with his girlfriend does not connect Hatfill to anthrax in any way. Although this statement might be capable of some other defama- tory meaning, it does not incriminate Hatfill in the anthrax mailings and so is not actionable under this complaint.
C.
Count Three alleges that the publication of Kristof’s columns con- stituted intentional infliction of emotional distress. The district court dismissed this count on two grounds: first, that publication of defama- tory charges is not sufficiently outrageous conduct, and second, that Hatfill failed to allege severe emotional distress.
A plaintiff seeking to recover for intentional infliction of emotional distress under Virginia law must allege that (1) the defendant’s con- duct was intentional and reckless; (2) the conduct was outrageous and intolerable; (3) the conduct caused the plaintiff’s emotional distress; and (4) the plaintiff’s distress was severe. Russo v. White, 400 S.E.2d 160, 162 (Va. 1991); Womack v. Eldridge, 210 S.E.2d 145, 148 (Va. 1974). Hatfill plainly alleged that The Times engaged in intentional and reckless conduct and that such conduct caused his emotional dis- 22 HATFILL v. THE NEW YORK TIMES tress. The Times argues, however, that the complaint fails to allege conduct that qualifies as "outrageous" under Virginia law and emo- tional distress so severe that Virginia law would permit recovery.
Virginia law limits the kinds of conduct that will support a claim for intentional infliction of emotional distress. The Supreme Court of Virginia has stated that a defendant may be liable for intentional infliction of emotional distress "only where the conduct has been so outrageous in character, and so extreme in degree, as to go beyond all possible bounds of decency, and to be regarded as atrocious, and utterly intolerable in a civilized community." Russo, 400 S.E.2d at 162. It is not enough that the defendant acted with tortious or even criminal intent. Id. Nor does liability extend to "mere insults, indigni- ties, threats, annoyances, petty oppressions, or other trivialities." Gai- ters v. Lynn, 831 F.2d 51, 53 (4th Cir. 1987) (applying Virginia law) (internal quotations omitted). The question whether the defendant’s conduct is so extreme and outrageous as to permit recovery is a ques- tion of law for the court. Womack, 210 S.E.2d at 148.
The district court’s conclusion that "[p]ublishing news or commen- tary on matters of public concern" can never be sufficiently extreme or outrageous to support a claim for intentional infliction of emotional distress sweeps too broadly. Depending upon the circumstances sur- rounding the publication and the nature of the defamatory charge, a defamatory publication could be "so outrageous in character, and so extreme in degree, as to go beyond all possible bounds of decency." Russo, 400 S.E.2d at 162. Accepting Hatfill’s allegations as true, The Times intentionally published false charges accusing him of being responsible for anthrax mailings that resulted in five deaths, without regard for the truth of those charges and without giving Hatfill an opportunity to respond. Given the notoriety of the case, the charge of murder, and the refusal to permit comment by Hatfill’s counsel, we conclude that the alleged misconduct is extreme or outrageous under Virginia law. Compare Baird v. Rose, 192 F.3d 462, 473 (4th Cir. 1999) (applying Virginia law) (reversing dismissal of an intentional- infliction-of-emotional-distress claim where the plaintiff alleged that her teacher intentionally humiliated her, knowing that she suffered from clinical depression). HATFILL v. THE NEW YORK TIMES 23 The Times argues that Hatfill cannot use an intentional-infliction- of-emotional-distress claim to avoid constitutional limitations on def- amation actions. We are confident, however, that the relevant consti- tutional limitations cannot be avoided as easily as The Times imagines. If Hatfill ultimately cannot prevail on his defamation claims because he is unable to satisfy constitutional requirements for recov- ery, then he likely will be unable to prove that The Times’ miscon- duct was intentional or reckless or that such misconduct was sufficiently outrageous to warrant recovery. At this stage of litigation, our sole concern is whether Hatfill’s allegations, taken as true, describe intentional and outrageous misconduct. We conclude that they do.
The Times also contends that Hatfill failed to allege severe emo- tional distress sufficient to permit recovery under Virginia law. Under Virginia law, "liability arises only when the emotional distress is extreme, and only where the distress inflicted is so severe that no rea- sonable person could be expected to endure it." Russo, 400 S.E.2d at 162. A plaintiff in Virginia state court must plead "with the requisite degree of specificity" the facts giving rise to his claim of severe emo- tional distress. Jordan, 500 S.E.2d at 219. In Russo, for example, the Supreme Court of Virginia held that the plaintiff’s allegations that "she was nervous, could not sleep, experienced stress and ‘its physical symptoms,’ withdrew from activities, and was unable to concentrate at work" were insufficient to avoid a demurrer on her claim for inten- tional infliction of emotional distress. 400 S.E.2d at 163. It was important to the court that the plaintiff had not alleged "that she had any physical injury caused by the stress, that she sought medical attention, that she was confined at home or in a hospital, or that she lost income." Id.
Hatfill did not allege his emotional distress in such specific terms, but Rule 8 — applicable in this diversity case — did not require him to do so. See Swierkiewicz, 534 U.S. at 513. The complaint alleges that "[a]s a result of defendants’ defamation here at issue, Dr. Hatfill has suffered severe and ongoing loss of reputation and professional standing, loss of employment, past and ongoing financial injury, severe emotional distress and other injury." The complaint further 24 HATFILL v. THE NEW YORK TIMES alleges that publication of Kristof’s columns inflicted "grievous emo- tional distress" upon Hatfill. These allegations are sufficient under Rule 8 to give The Times "fair notice of what [Hatfill’s] claim is and the grounds upon which it rests," id. at 512 (citing Conley, 355 U.S. at 47), and they are adequate to state the final necessary element of a claim for intentional infliction of emotional distress.
III.
Hatfill’s complaint adequately alleges claims for defamation and intentional infliction of emotional distress based on The Times’ publi- cation of columns implicating Hatfill in the anthrax mailings of 2001. Accordingly, we reverse the district court’s order granting dismissal and remand this case for further proceedings consistent with this opin- ion.
REVERSED AND REMANDED
NIEMEYER, Circuit Judge, dissenting:
Dr. Steven Hatfill’s defamation complaint alleges at bottom that four columns written by columnist Nicholas Kristof and published in the New York Times during the period from July to August 2002 accused Dr. Hatfill of being the anthrax murderer. The essential ques- tion therefore is whether these columns, taken together or individu- ally, may fairly be read to accuse Dr. Hatfill of the murders. Because I can find nothing in the letter or spirit of the columns that amounts to such an accusation, I would affirm.
The columns, when read fairly, send the message:
(1) that the FBI’s investigation of the anthrax murders was lackadaisical and unimaginative;
(2) that the FBI should have begun pursuing obvious leads that created suspicion about Dr. Hatfill and that, based on circumstantial evidence, Dr. Hatfill should have been the leading suspect; and HATFILL v. THE NEW YORK TIMES 25 (3) that, while there was circumstantial evidence pointing to Dr. Hatfill, no "traditional physical evidence linking him to the attacks" existed and that there "must [have been] a genuine assumption that he [was] an innocent man caught in a nightmare." Nicholas Kristof, The Anthrax Files, N.Y. Times, Aug. 13, 2002, at A19.
These points were amplified by examples of suspicious circum- stances, but nowhere does any column accuse Dr. Hatfill of commit- ting the murders. The columns’ purpose was to put into operation prosecutorial machinery that would determine whether Dr. Hatfill committed the crimes and "end this unseemly limbo by either excul- pating Dr. Hatfill or arresting him." Id.
We, of course, must accept at this stage of the case the allegations that Kristof’s columns contained some factual inaccuracies. But whether Kristof’s descriptions of the various items of circumstantial evidence were accurate is irrelevant because (1) inaccurately report- ing the suspicious circumstances surrounding a suspect does not amount to inaccurately accusing — either expressly or impliedly — the suspect of actually committing the crime, and (2) historical cir- cumstances recounted in the columns and not disputed by Dr. Hatfill were sufficient to support the columns’ stated suspicion about him. Reporting suspicion of criminal conduct — even elaborately and sometimes inaccurately — does not amount to an accusation of crimi- nal conduct as necessary to support Dr. Hatfill’s claim.
Thus, I agree with the district court that the columns did not, under Virginia law, "impute to [Dr. Hatfill] the commission of some crimi- nal offense involving moral turpitude." Carwile v. Richmond Newspa- pers, Inc., 82 S.E.2d 588, 591 (Va. 1954). Similarly, I agree with the district court that the statements in Kristof’s columns were insuffi- ciently outrageous to support Dr. Hatfill’s claims for intentional infliction of emotional distress. See Womack v. Eldridge, 210 S.E.2d 145, 148 (Va. 1974). Accordingly, I would affirm the district court’s order dismissing the complaint under Federal Rule of Civil Procedure 12(b)(6).