BOUDIN, Circuit Judge.
Robert K. Gray was for many years active in Republican politics and a leading figure in public relations in Washington, D.C. He served in the Eisenhower administration in various roles
(e.g.,
Secretary to the Cabinet), worked in the 1980 Reagan-Bush presidential campaign and served between 1961 and 1981 as the head of the Washington office of, and eventually as vice chairman of, Hill and Knowlton, a major public relations and lobbying firm. He founded his own firm in 1981, sold it to Hill and Knowlton in 1986, and served for a period as a member of the board of directors and chairman of a division of the latter.
In July 1992, St. Martin’s Press, Inc., published a book by Susan Trento, entitled
The Power House: Robert Keith Gray and the Selling of Access and Influence in Washington.
Focusing on Gray’s career, the book sought to show the influence of powerful and well-connected lobbyists on the federal government. In June 1995, Gray brought suit both against St. Martin’s Press and Trento in the federal district court in New Hampshire, claiming that eight separate statements made in the book were defamatory. The eight statements are set forth in an appendix to this opinion.
After two years of discovery, St. Martin’s Press moved for summary judgment. On March 5, 1998, the district court granted partial summary judgment to St. Martin’s Press, ruling that three of the eight statements — (designated (b), (f), and (h))— were non-actionable statements of opinion. More discovery was conducted and both defendants filed a second summary judgment motion. On May 19, 1999, the district court ruled that Gray was a “limited purpose public figure,” requiring Gray to show “actual malice” in order to prevail. The district court granted summary judgment for defendants as to one statement (statement (c)), finding that there was no basis for the jury to find actual malice.
The trial on the remaining four statements began on June 7,1999. On June 22, 1999, the jury returned special verdicts in favor of St. Martin’s and Trento. As to each of the four remaining statements in issue ((a), (d), (e), and (g)), the jury found that Gray had not proved that the defendants had published to third parties statements that were false and defamatory as to Gray; separately, the jury found that Gray failed to prove actual malice by either defendant as to any of the four statements.
Gray has now appealed. In this court he contests the dismissal before trial of four of the statements, a discovery ruling upholding a claim of privilege asserted by Trento that pertains to one of the statements considered by the jury, and the denial before trial of a motion by Gray to amend his complaint to add twenty additional statements to the eight already charged. We consider the issues in this order, applying the standard of review pertinent to the issue in question.
Under state law, defendants in this case would be hable for damages for libel if, as a result of the failure to exercise reasonable care, they published false and defamatory facts about the plaintiff to a third party, assuming that no valid privilege applies.
Independent Mechanical Contractors, Inc. v. Gordon T. Burke & Sons, Inc.,
138 N.H. 110, 118, 635 A.2d 487
(N.H.1993);
The Gazette, Inc. v. Harris,
229 Va. 1, 8, 15, 325 S.E.2d 713 (1985),
cert. denied,
472 U.S. 1032, 105 S.Ct. 3513, 87 L.Ed.2d 643 (1985).
See generally Restatement (Second) of Torts
§ 558 (1977). However, the Supreme Court has read the First Amendment, made binding on the states through the Fourteenth, to impose additional limitations in defamation cases, whether or not they are also part of state law. Two of these limitations are significant in this case — one dealing with scien-ter and the other with opinion.
Pertinently, the Court has held that a “public figure” may recover only if the false and defamatory statement was made with “actual malice,” meaning (in the Supreme Court’s non-literal usage) either that defendant knew that the statement was false or showed a “reckless disregard” as to its truth or falsity.
New York Times Co. v. Sullivan,
376 U.S. 254, 279-80, 84 S.Ct. 710, 11 L.Ed.2d 686 (1964);
see also Gertz v. Robert Welch, Inc.,
418 U.S. 323, 335-37, 94 S.Ct. 2997, 41 L.Ed.2d 789 (1974). A “public figure” may be one of such fame as to be so in all contexts
(e.g.,
the President) or a “limited-purpose public figure” as to a particular episode or subject; in the latter case, only the statements about the person in that context require a showing of actual malice.
Gertz,
418 U.S. at 351-52, 94 S.Ct. 2997;
Pendleton v. City of Haverhill,
156 F.3d 57, 67
&
n. 7 (1st Cir.1998).
The Court has also held that only statements that present or imply the existence of facts that can be proven true or false are actionable under state defamation law.
Milkovich v. Lorain Journal Co.,
497 U.S. 1, 18-20, 110 S.Ct. 2695, 111 L.Ed.2d 1 (1990). But to say “I think” is not enough to turn fact into opinion,
Milkovich,
497 U.S. at 18-19, 110 S.Ct. 2695, where what is supposedly “thought” is, or implies, a proposition of fact,
id.; Levinsky’s v. Wal-Mart Stores, Inc.,
127 F.3d 122, 127 (1st Cir.1997). Rather, the cases are likely to protect a statement as “opinion” where it involves expressions of personal judgment, especially as the judgments become more vague and subjective in character.
See Levinsky’s,
127 F.3d at 130 (store was “trashy”). As Chief Judge Posner put the matter in
Haynes v. Alfred A. Knopf, Inc.,
8 F.3d 1222, 1227 (7th Cir.1993):
[I]f it is plain that the speaker is expressing a subjective view, an interpretation, a theory, conjecture, or surmise, rather than claiming to be in possession of objectively verifiable facts, the statement is not actionable.
On this appeal, Gray first questions the district court’s March 5, 1998, ruling that statements (b), (f), and (h) are not actionable because they are not factual statements capable of being proven false. The determination was made on summary judgment and in any event the courts treat the issue of labeling a statement as verifiable fact or as opinion as one ordinarily decided by judges as a matter of law.
Bose Corp. v. Consumers Union of United States, Inc.,
466 U.S. 485, 510-11, 104 S.Ct. 1949, 80 L.Ed.2d 502 (1984). Thus from either vantage, our review is
de novo.
As it happens, we agree with the district court’s reasoning as well as its result and so treat this issue briefly.
Statement (b) is the view, attributed to an unnamed Gray and Company senior executive, that Gray’s “closeness to the President [Reagan] and others was often faked. ‘He completely faked his closeness with a number of senior administration officials.’ ” There are various vantages from which the statement could be attacked as false
(e.g.,
that no such view was expressed by the unnamed executive), but the bite here is in the claim that “closeness” was “faked” and Gray’s position is that he could show at trial that he was quite close to President Reagan and other senior officials and was not “faking” these relationships.
Whether calling something a “fake” is or is not protected opinion depends very much on what is meant and therefore on
context. To say that a dollar'bill is a fake would, in most situations (but perhaps not all), be taken to mean that it was a counterfeit; and to say that the defendant was knowingly passing a fake dollar bill would surely be actionable, if false. At the other extreme, where there were two productions of
Phantom of the Opera,
and the defendant called one of them “fake” and “phony,” this court held that the adjectives were subjective aesthetic judgments protected as opinion.
Phantom Touring, Inc. v. Affiliated Publications,
953 F.2d 724, 728 (1st Cir.),
cert. denied,
504 U.S. 974; 112 S.Ct. 2942, 119 L.Ed.2d 567 (1992).
In this case, Gray might have a claim if defendants had said that Gray claimed to know President Reagan or other high officials but did not in fact know them; whether or not he knew them is an objective fact. However, the book made quite clear that Gray did have contacts at the highest levels; the word “fake” was used to imply that Gray was exaggerating his “closeness.” This is just the kind of subjective judgment that is only minimally about “what happened” but expresses instead a vague and subjective characterization of what happened. As we read thé case law, the statement is protected opinion.
Statement (f) is in essence several different statements: in it Trento asserted that a number of Washington lobbyists said that Gray and Company “ultimately failed because it offered very little real substance.” Gray’s quarrel is not with the claim that lobbyists (quite possibly competitors) had expressed such views but with the assertions that the company had “failed” and offered “little real substance.” The latter judgment, where the product is an intangible service like lobbying and criteria for success are debatable, is surely one of opinion so we direct our attention to the charge that Gray and Company “failed.”
If the book had said or even implied that Gray and Company went bankrupt or did not make a profit, these would be statements of fact that could be proved true or false. ' Instead, the book made clear that in 1986 Gray sold his company to JWT Group, Inc., which made the company part of its subsidiary, Hill and Knowlton, for about $16 million, of which Gray himself got at least $9 million (the purchase price had been $21 million but the buyer withheld about $4.6 million to cover possible liabilities). Gray does not dispute that the figures are given in the book and those figures make clear that Gray’s company did not fail in any absolute sense.
Indeed, in explaining that Gray and Company “failed,” Trento said that the sale was “profitable” but it “shattered” Gray’s dream of owning the world’s largest public communications firm. Gray does not challenge the latter statement; recall that he left Hill and Knowlton to found Gray and Company as an independent enterprise to compete with Hill and Knowl-ton. Some might think it a success, rather than a failure, that his former employer found him competitive and competént enough to buy him out for millions and place him on its board of directors; but what is “success” in a situation like this one is very much a matter of opinion.
The next statement, designated (h), reads as follows:
Robert Crowley believed that “Casey may have asked Gray to take on these controversial clients — for the very purpose of spying on them.” If that were so it would explain why Gray considered countries like Libya, and took clients like Angola.
Robert Crowley was a former senior CIA official and William Casey was the CIA director in the early to mid-1980s. Once again, Gray does not dispute that Crowley may have so believed but he does challenge the balance of the statements.
Interestingly, the first disputed proposition — that Casey may have asked Gray to spy — is not necessarily protected as to defendants even if they were merely describing Crowley’s view. The reason, which is
one of policy rather than strict logic, is that it would otherwise be too easy for a writer or publisher to defame freely by repeating the defamation of others and defending it as simply an accurate report of what someone else had said.
Cianci v. New Times Publ’g Co.,
639 F.2d 54, 60-61 (2d Cir.1980) (Friendly, J.);
Cepeda v. Cowles Magazines & Broad., Inc.,
328 F.2d 869, 871 (9th Cir.),
cert. denied,
379 U.S. 844, 85 S.Ct. 51, 13 L.Ed.2d 50 (1964);
Restatement (Second) of Torts
§ 578. Thus, the first issue is whether Crowley’s speculation is actionable once it is attributed to the defendants.
Statement (h), by its terms, links the possible purpose to spy to Casey, not Gray; but the second sentence, seemingly a separate speculation by defendants, suggests that Gray may have acted on this request. Defendants say that to spy for one’s country is laudable, not defamatory. Whether a statement is
capable
of a defamatory meaning is an issue of law for the court,
Restatement (Second) of Torts
§ 614(1);
Harkaway v. Boston Herald Traveler Corp.,
418 F.2d 56, 58 (1st Cir.1969), and thus one we would review
de novo;
but the factfinder must ultimately decide whether a statement like the one here, reasonably capable of both a defamatory and non-defamatory meaning, was in fact understood as defamatory by its recipients.
We think that the statement’s implication of spying on clients is capable of bearing a defamatory meaning — it could easily harm Gray in dealing with clients — and that a reasonable jury could have found the statement in question defamatory.
See Restatement (Second) of Torts
§ 559 (“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.”);
Thomson,
119 N.H. at 373, 402 A.2d 651 (to be defamatory, a statement “must tend to lower the plaintiff in the esteem of any substantial and respectable group, even though it may be quite a small minority”) (internal quotation marks omitted);
Carwile v. Richmond Newspapers, Inc.,
196 Va. 1, 8, 82 S.E.2d 588 (1954) (a statement “which imputes to a business or professional man conduct which tends to injure him in his business or profession” is actionable as libel per se).
To determine if the Crowley statement is shielded because it is conditional (“may have”) is a more difficult question. Here, the statement may be protected “opinion” not because it is vague or judgmental but because it is speculative. The test, admittedly a very crude one, is whether the statement is properly understood as purely speculation or, alternatively, implies that the speaker or writer has concrete facts that confirm or underpin the truth of the speculation.
Levin v. McPhee,
119 F.3d 189, 197 (2d Cir.1997);
Restatement (Second) of Torts
§ 566, comment (c) at 173. The former is protected as opinion; the latter is taken as an indirect assertion of truth.
Like the district judge, we see nothing that suggests that Crowley or the defendants were relying upon undisclosed facts. Crowley’s own view is couched as a belief as to what “may have” happened. The defendants add a further supporting fact that Gray “considered countries like Libya, and took clients like Angola”; but Gray does not dispute that this is so, and where the underlying facts are disclosed, it becomes even more clear that the writer or publisher is merely speculating (“if so”)
about the inference.
Restatement (Second) of Torts
§ 566, at 174 ill. 5 (“A says to B about C, a city official: ‘He and his wife took a trip on city business a month ago and he added her expenses in as a part of his own.’ B responds: ‘If he did that he is really a thief.’ B’s expression of opinion does not assert by implication any defamatory facts, and he is not hable to C for defamation.”).
The last statement of the four statements disposed of before trial — statement (c) — quotes a former “Gray and Company senior vice president” as saying that “there’s a degree of venality on the part of [Gray] and lack of integrity which always took me aback” and “very little real basic principle and an awful lot, to me, of over charging.” During discovery, the defendants produced a transcript of Trento’s interview with Barry Zorthian, a former Senior Vice President at Gray and Company, as the source of this statement. The district court granted summary judgment as to this statement because it found that Gray was a limited-purpose public figure with respect to lobbying and that no reasonable jury could find that the charge was made by the defendants with actual malice.
On appeal, Gray first disputes the limited-purpose public figure label. This is treated as an issue of law to be resolved by the district judge and reviewed
de novo
by us.
Pendleton v. City of Haverhill,
156 F.3d 57, 68 (1st Cir.1998). We agree with the district court that, prior to and continuing up to the book’s publication — which (to avoid bootstrapping) is the pertinent time frame,
Bruno & Stillman, Inc. v. Globe Newspaper Co.,
633 F.2d 583, 591 (1st Cir.1980) — a public controversy existed as to the methods and influence of lobbyists in Washington. This was amply evidenced by materials submitted to the district court showing that from the early 1980s onward there has been a tide of concern and criticism about Washington lobbying.
The record also shows that Gray was a central figure in this controversy, being identified as one of the best-known of the high-level Washington public relations experts, an emblematic figure, and a self-professed defender against attacks on lobbying. Indeed, Gray’s lobbying and in particular billing practices were themselves the subject of comment and criticism in mainstream publications like
Time, Newsweek,
the
Washington Post
and the
New Republic.
This does not show that he behaved improperly as a lobbyist or overbilled his clients. It does mean that, in the complex equation for liability laid down by the Supreme Court, Gray needed to show actual malice by clear and convincing evidence.
Gertz,
418 U.S. at 351-52, 94 S.Ct. 2997;
Pendleton,
156 F.3d at 67 & n. 7.
In most cases, as in this one, the plaintiff does not have any evidence of actual malice in the literal sense but, as already noted, recklessness suffices.
New York Times,
376 U.S. at 279-80, 84 S.Ct. 710;
Masson v. New Yorker Magazine, Inc.,
501 U.S. 496, 510, 111 S.Ct. 2419, 115 L.Ed.2d 447 (1991). Recklessness is a jury issue so long as the plaintiff at the summary judgment stage produces evidence that would allow a reasonable jury to find the defendants reckless by clear and convincing evidence.
Anderson v. Liberty Lobby, Inc.,
477 U.S. 242, 257, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986). Recklessness, in this context, can be shown by proving “that the defendant actually had a ‘high degree of awareness of ... probably falsity,’ ”
Harte-Hanks Communications, Inc. v. Connaughton,
491 U.S. 657, 688, 109 S.Ct. 2678, 105 L.Ed.2d 562 (1989) (quoting
Garrison v. Louisiana,
379 U.S.
64, 74, 85 S.Ct. 209, 13 L.Ed.2d 125 (1964)), but mere negligence in conducting an investigation or weighing the evidence is not enough.
Harte-Hanks Communs.,
491 U.S. at 688, 109 S.Ct. 2678;
Masson,
501 U.S. at 510, 111 S.Ct. 2419.
The defendants assert, and the plaintiff does not dispute, that actual malice in this case must be shown separately as to each defendant.
But cf. Cantrell v. Forest City Publ’g Co.,
419 U.S. 245, 253, 95 S.Ct. 465, 42 L.Ed.2d 419 (1974) (referring to vicarious liability). With respect to Trento, she relied not only on Zorthian but also on several other sources. They, too, had connections with Gray and Company and their statements tended to back up Zorthian’s position. Against Trento’s multiple sources, Gray counters that Zorthian had an axe to grind (he had parted ways with Gray and later sued him); that few of Trento’s sources had much knowledge of the billing side of Gray’s business; and that Trento failed to publish one other source’s statement that he had “never heard” about charges of overbilling or to interview others who would have denied such overcharging.
There is no point in our going through this evidence piece by piece since we agree with the district court’s assessment. Prejudice or limited knowledge on the part of a source may suggest caution but does not preclude reliance; the fact that one witness had “never heard” about the charges counts for little; and while refusing to seek out decisive witnesses may be a mark of recklessness in some circumstances, Trento already had multiple sources and was under no obligation to exhaust every possible witness before winding up her investigation. Even assuming she was careless and reached a mistaken conclusion, that is not enough for actual malice.
Less need be said about St. Martin’s. It apparently had some doubts about the rigor of Trento’s book
'proposal.
But the proposal was followed by more than two years of research and there is no evidence that St. Martin’s employees doubted the accuracy of Trento’s final product. It is true that Gray protested to St. Martin’s prior to publication that some of the statements were untrue. However, apart from the fact that in this original protest statement (c) was not specifically identified as false, simple denials by the subject are commonplace and, absent more, are normally not enough to premise a finding of actual malice.
Edwards v. National Audubon Soc’y, Inc.,
556 F.2d 113, 120-21 (2d Cir.),
cert. denied,
434 U.S. 1002, 98 S.Ct. 647, 54 L.Ed.2d 498 (1977).
Gray’s next claim of error on appeal is that the district court erred in upholding a claim of press privilege during the discovery process. Specifically, Trento declined to reveal the name of her confidential source for statement (g), which reads: “One Gray and Company executive in a position to know said that Gray and Company was making payments to [Duke] Zeller.” According to the book, Duke Zel-ler had previously worked for Gray and at the time in question was the Director of Communications for the Teamsters Union, a lucrative client of Gray’s company. The apparent implication is that Gray made, secret payments to Zeller in order to retain the Teamsters account.
The magistrate judge, later sustained by the district judge, upheld Trento’s objection on the grounds that New Hampshire recognized a qualified confidential source privilege for reporters,
State v. Siel,
122 N.H. 254, 259-60, 444 A.2d 499 (1982);
Downing v. Monitor Publ’g Co., Inc.,
120 N.H. 383, 386-87, 415 A.2d 683 (1980), that one requirement to overcome the privilege was to show that the applicant had made all reasonable efforts to obtain the identity of the confidential source by other reasonable means, and that this requirement had not been satisfied by Gray. On appeal, Gray argues that the district court misread New Hampshire privilege law, which both sides now assume to govern the question, and that in any event it was error to find
that Gray had not satisfied the requirement.
New Hampshire law on the privilege in question, an issue for
de novo
review, is not a model of clarity; and, while the “failure to satisfy” finding would be reversed only for clear error or abuse of discretion, we have some initial sympathy for Gray’s claim that he did all he could to satisfy the requirement of exhausting other means. And if Gray were found to have exhausted all reasonable means of identifying the source and Trento still refused to reveal her source, Gray would have been entitled to a presumption that
no
source existed.
Dooming,
120 N.H. at 387, 415 A.2d 683. This could have helped Gray persuade the jury that Trento acted with actual malice in making the payoff charge.
The problem for Gray is that however the matter stood at the time of the privilege ruling, the jury returned a verdict as to statement (g) that rested on two alternative grounds: one was lack of actual malice but the other was Gray’s failure (in the jury’s view) to prove that the statement was false and defamatory. Defendants say that the verdict thus rested safely on a ground independent of the no-malice finding. Gray, who has not independently attacked the jury’s alternative ground, has not provided any very cogent answer to this claim, beyond saying rather tersely that if the privilege had been overridden, Trento had still baulked, and the jury been told that therefore it could presume that the source did not exist, then the jury might have thought the statement false.
It is hard to see why this is so. Obviously, there was far better direct evidence available — such as Gray’s own testimony— as to whether the payoff occurred; indeed, on this issue the source’s statement was inadmissible hearsay as to the truth of the charge. Nor is there all that much basis to doubt that some source did exist, whether reliable or otherwise; Trento produced redacted notes of her conversation with the source and Gray himself got some mileage out of an argument that this portion of Trento’s interview was not taped like the rest of the interview with the confidential source.
Gray also says that while Trento preserved the claim of privilege, St. Martin’s never asserted the privilege so the district court had to have erred in upholding the claim as it did. St. Martin’s agrees that it did not assert the privilege but argues, as it did in the district court, that it did not know the name of the source and therefore needed no privilege in order to withhold it. Gray responds that in previous court papers St. Martin’s never affirmatively stated that it had not been told the name of the confidential source. But Gray himself does not allege that St. Martin’s
does
actually know the identity of the source, and without evidence to contradict St. Martin’s assertion in its brief that it does
not
know the identity of the source, Gray’s argument does not undermine St. Martin’s commonsense position.
Gray’s final argument on appeal, properly placed last, is that the district court erred in refusing to grant Gray leave to amend his complaint to add 20 additional statements from the book now alleged to be defamatory. As earlier noted, the complaint was filed in June 1995; and the motion to amend was made three years later, after extensive proceedings including discovery. The district court found the motion untimely and unduly prejudicial.
We find no abuse of discretion.
Grant v. News Group Boston, Inc.,
55 F.3d 1, 5 (1st Cir.1995). It is enough to say that although leave to amend is to be “freely given,” Fed.R.Civ.P. 15(a), Gray had the book for six years before he moved to amend and at the outset of the litigation could have easily decided which charges he believed to be false. The district court amply explained its reasons for finding that under the circumstances there was undue delay and prejudice.
The judgment is
affirmed.
The conditional cross-appeal filed by St. Martin’s Press and Susan Trento is
dismissed
as moot.
ADDENDUM
a. “As others were cleaning out their desks, looking for jobs, briefing their successors, and preparing to leave the White House, Gray was busy dictating his memoirs to his White House secretary.”
The Power House,
p. 53.
b. “A senior Gray and Company executive insisted that Gray’s closeness to the President and others was often faked. ‘He completely faked his closeness with a number of senior administration officials.’ ”
The Power House,
p. 156.
c. “ T think there’s a degree of venality on the part of Bob and lack of integrity which always took me aback. A lot of it he would justify as being a businessman, but there was very little real basic principle and an awful lot, to me, of over charging.’ ”
The Power House,
p. 165.
d. “ ‘... at Gray and Company he [Mr. Gray] stage-managed impressive-sounding calls. A reporter would walk in and he would instruct his executive assistant to come in and announce that there was a call from . the White House. Totally fabricated. Absolutely. They would come in and they would say, “Mr. Gray, Mr. Meese is on the phone,” and he would pick up a dead line or a line that was set up by the executive assistant, carry on a conversation of four or five short rapid sentences as though he was in constant communication and hang up. And then, of course, the reporters, dazzled, would then report that a White House phone call came in,’ explained one Gray and Company executive.”
The Power House,
p. 167-8.
e. “And the Gray and Company employees in Spain were to be convinced that the office was used as a money laundering operation for the Reagan administration’s private intelligence network.”
The Power House,
p. 273.
f. “In the end, several Washington lobbyists feel that Gray and Company ultimately failed because it offered very little real substance.”
The Power House,
p. 323.
g. “One Gray and Company executive in a position to know said that Gray and Company was making payments to Zeller.”
The Power House,
p. 202.
h. “Robert Crowley believed that ‘Casey may have asked Gray to take on these controversial clients- — for the very purpose of spying on them.’ If that were so it would explain why Gray considered countries like Libya, and took clients like Angola.”
The Power House,
p. 260.