REINHARDT, Circuit Judge:

BACKGROUND

This case arises from the notorious Palmyra trials and the publicity surrounding them. Palmyra is an uninhabited island located in the Pacific Ocean. During the summer of 1974, Stephanie Stearns and Buck Walker ailed to the island in an old sailboat. Once there, the couple discovered that the condition of their boat and the lack of adequate supplies prevented their return. Shortly after their arrival, Muff and Mac Graham arrived on a second sailboat. By the end of October that same year, the Grahams had disappeared, and Stearns and Walker had returned to Hawaii sailing the boat that once belonged to the Grahams. In 1981, the bones of Muff Graham were found washed up on Palmyra, and Stearns and Walker were indicted for her murder.

Earle Partington was appointed to represent Walker, while Stearns hired Vincent Bugliosi to represent her in a separate trial. Partington is a well-known criminal defense lawyer, although his "passive" handling of a I controversial murder case once caused the Hawaii Supreme Court to reverse his client's conviction sua sponte. Partington v. Bugliosi, 825 F. Supp. 906, 910 (D.Hawaii 1993) (describing the facts surrounding the Hawaii Supreme Court decision). Bugliosi is a noted lawyer and author who prosecuted Charles Manson and wrote the best-selling book Helter Skelter, but whose efforts to attain elected political office were rejected by the voters of California.

In the Palmyra Island murder cases, which took place in the federal district court in Honolulu, Partington's client was convicted and Bugliosi's acquitted. Following the trials, Bugliosi, along with Bruce Henderson, [the "Book Defendants" or "Bugliosi"] wrote And The Sea Will Tell, an account of his successful defense of Stearns. In 1991 CBS, in conjunction with a number of producers and the Epstein Productions company ["Movie Defendants"], produced a made-for-television movie based on Bugliosi's book.

Partington filed a damage claim against both the Book and Movie Defendants alleging defamation and false light claims. The action was filed in federal district court which had diversity jurisdiction. In Counts II and III, Partington contends that the Book Defendants defamed him and cast him in a false light by implying that he had not read the transcript of the state court theft trial (regarding the theft of the Grahams' boat) and that he was therefore an incompetent attorney. In Count IV, Partington alleged that a statement in the book criticizing him for taking an overly submissive stance toward the judge presiding over Walker's trial cast him in a false light. In Counts V and VI, Partington alleges that the Book Defendants defamed him and cast him in a false light by stating that he failed to introduce into evidence a diary indicating that Stearns and Walker had socialized with the Grahams, thereby implying that he was an incompetent attorney. In Count VII, Partington alleges that the book cast him in a false light when it criticized him for failing to call a particular witness at trial. Finally, Partington alleges that the Movie Defendants defamed him by portraying Bugliosi as telling Stearns that she would spend the rest of her life in prison if he defended her the way Partington defended Walker.

The district court granted the defendants' motion for summary judgment, concluding that Partington had failed to establish a claim for defamation or false light. Partington v. Bugliosi, 825 F. Supp. 906 (D.Hawaii 1993). Partington appeals.

We note preliminarily that there are several issues raised by one or the other of the parties that it is not necessary for us to reach. First, we need not decide whether the disputed statements can fairly be read to imply that Partington represented his client poorly and whether such an implication would be considered defamatory under state law. See Fernandes v. Tenbruggencate, 649 P.2d 1144, 1147 (Hawai`i 1982) (noting that, under Hawai`i law, a statement is defamatory if it "tends to `harm the reputation of another so as to lower him in the estimation of the community or to deter third persons from associating or dealing with him.'"); id. at 1147 n. 1 (noting that the statements must be reasonably capable of bearing the meaning ascribed to them in order to serve as the basis of a defamation claim). Because we hold that the First Amendment protects these statements regardless of what state law provides, see infra pp. 1153-1162, we assume arguendo that the statements do imply that Partington represented his client poorly during the Walker trial.

Next, we do not decide whether Partington was a limited purpose public figure or whether the passage of time would have had any effect upon that status since, whether or not Partington can allege malice, the statements he contests are not actionable. See infra pp. 1153-1162. In addition, we do not determine whether the fact that Partington's claim regarding passages in the book rests upon the implication arising from the statements, rather than upon their actual content, would affect the showing that Partington is required to make because, even if Bugliosi had stated directly what Partington contends he implied, his statements would be protected by the First Amendment. See infra pp. 1153-1160.

Until a few years ago, we drew a sharp, formalistic line between fact and opinion, holding that anything cast in the form of an opinion was absolutely protected by the First Amendment and could not serve as the basis for a defamation claim. See, e.g., Ault v. Hustler Magazine, Inc., 860 F.2d 877, 880-81 (9th Cir. 1988), cert. denied, 489 U.S. 1080, 109 S.Ct. 1532, 103 L.Ed.2d 837 (1989).

In Milkovich v. Lorain Journal Co., 497 U.S. 1, 110 S.Ct. 2695, 111 L.Ed.2d 1 (1990), however, the Supreme Court rejected the bright-line approach of this and other circuits. It found the opinion/fact dichotomy too simplistic. The Court stated that it had never intended "to create a wholesale defamation exemption for anything that might be labeled `opinion.'" Milkovich, 497 U.S. at 18, 110 S.Ct. at 2705. The Court reasoned that "[s]imply couching such statements in terms of opinion does not dispel [the false, defamatory] implications" because a speaker may still imply "a knowledge of facts which lead to the [defamatory] conclusion." Id. at 19, 110 S.Ct. at 2706. It therefore held that, while "pure" opinions are protected by the First Amendment, a statement that "may . . . imply a false assertion of fact" is actionable. Id. at 19, 110 S.Ct. at 2706.

We have had only one previous opportunity to interpret Milkovich. In Unelko Corp. v. Rooney, 912 F.2d 1049, 1053 (9th Cir. 1990), cert. denied, 499 U.S. 961, 111 S.Ct. 1586, 113 L.Ed.2d 650 (1991), we held that, in reviewing a defamation claim, a court must ask as a threshold matter "whether a reasonable factfinder could conclude that the contested statement `impl[ies] an assertion of objective fact.'" If the answer is no, the claim is foreclosed by the First Amendment. As a starting point for our analysis, we adopted a three-part test: (1) whether the general tenor of the entire work negates the impression that the defendant was asserting an objective fact, (2) whether the defendant used figurative or hyperbolic language that negates that impression, and (3) whether the statement in question is susceptible of being proved true or false. Id.

Under Unelko's basic framework, we examine the work as a whole, the specific context in which the statements were made, and the statements themselves to determine whether a reasonable factfinder could conclude that the statements imply a false assertion of objective fact and therefore fall outside of the protection of the First Amendment. Here, we flesh out the Unelko framework. Ultimately, we conclude that in this case the general and specific contexts in which the defendants' contested statements were made do not imply the assertion of an objective fact. We also conclude that those statements are not capable of being proved true or false. Accordingly, we hold that the district court correctly concluded that, under the standards outlined in Milkovich, Partington failed to state a defamation claim: while the defendants' descriptions of Partington's performance during Walker's trial may imply that he represented his client poorly, the statements are protected by the First Amendment and are therefore not actionable.

I.

The Supreme Court and other courts have emphasized that one must analyze a statement in its broad context to determine whether it implies the assertion of an objective fact. See, e.g., Masson v. New Yorker Magazine, 501 U.S. 496, 520, 111 S.Ct. 2419, 2435, 115 L.Ed.2d 447 (1991); Milkovich, 497 U.S. at 19, 110 S.Ct. at 2706; Phantom Touring, Inc. v. Affiliated Publications, 953 F.2d 724, 727 (1st Cir.), cert. denied, 504 U.S. 974, 112 S.Ct. 2942, 119 L.Ed.2d 567 (1992). With regard to the two statements in the book that Partington contends are defamatory, we find that the book's general tenor makes clear that Bugliosi's observations about Partington's trial strategies, and the implications that Partington contends arise from them, represent statements of personal viewpoint, not assertions of an objective fact. See Phantom Touring, Inc., 953 F.2d at 729 ("[Because] the sum effect of the format, tone and entire content of the articles is to make it unmistakably clear that [the author] was expressing a point of view only, . . . the challenged language is immune from liability.").

And the Sea Will Tell describes two controversial trials that resulted in dramatically different outcomes for defendants who were in some ways similarly situated. The purpose of the book is to offer the personal viewpoint of the author concerning the trials. Indeed, readers presumably purchased the book not to read a dry description of the facts but to learn of Bugliosi's personal perspective about the trials since he was both a key participant in the controversy and a well-known criminal defense lawyer. Because the book outlines Bugliosi's own version of what took place, a reader would expect him to set forth his personal theories about the facts of the trials and the conduct of those involved in them. Moreover, lawyers who write popular books, and particularly trial lawyers, are not known for their modesty; one would generally expect such authors to have a higher opinion of their own performance than of the professional abilities exhibited by other counsel. Cf. Phantom Touring, 953 F.2d at 729 (holding that statements are protected in part because they are found in "the type of article generally known to contain more opinionated writing than the typical news report"). Thus, Bugliosi's book is a forum in which a reader would be likely to recognize that the critiques of the judges, witnesses, and other participants in the two trials — and particularly of the other counsel — generally represent the highly subjective opinions of the author rather than assertions of verifiable, objective facts. Cf. Moldea v. New York Times Co., 22 F.3d 310, 313 (D.C. Cir.) ( Moldea II) (finding that the challenged statements were protected in part because they "were evaluations of a literary work which appeared in a forum in which readers expect to find such evaluations"), cert. denied, ___ U.S. ___, 115 S.Ct. 202, 130 L.Ed.2d 133 (1994); id. at 315 (emphasizing the importance of evaluating a critical analysis "with an eye toward readers' expectations and understandings [of it]").

Moreover, the subject of the book — the events that took place at Palmyra Island and the outcome of the two trials — is one about which there could easily be a number of varying rational interpretations. There is no question that the subject matter of the book, and the sources upon which Bugliosi relies in drawing his conclusions, are inherently ambiguous, and we believe that reasonable minds could differ as to how to interpret the events and actions described in it. Indeed, much of the public excitement surrounding the two trials stemmed from the fact that there is no clear answer as to what precisely occurred at Palmyra Island or as to why the two trials resulted in such different outcomes.

When, as here, an author writing about a controversial occurrence fairly describes the general events involved and offers his personal perspective about some of its ambiguities and disputed facts, his statements should generally be protected by the First Amendment. Otherwise, there would be no room for expressions of opinion by commentators, experts in a field, figures closely involved in a public controversy, or others whose perspectives might be of interest to the public. Instead, authors of every sort would be forced to provide only dry, colorless descriptions of facts, bereft of analysis or insight. There would be little difference between the editorial page and the front page, between commentary and reporting, and the robust debate among people with different viewpoints that is a vital part of our democracy would surely be hampered.

Our concern is shared by many other courts. For example, the Supreme Court has recently emphasized that the First Amendment guarantees authors "the interpretive license that is necessary when relying upon ambiguous sources." Masson v. New Yorker Magazine, 501 U.S. 496, 519, 111 S.Ct. 2419, 2434, 115 L.Ed.2d 447 (1991). The District of Columbia Circuit has concluded from a survey of Supreme Court doctrine that its cases clearly "establish that when a writer is evaluating or giving an account of inherently ambiguous materials or subject matter, the First Amendment requires that the courts allow latitude for interpretation." Moldea II, 22 F.3d at 315. Similarly, the Fourth Circuit has held that when an "answer [is] within the wide range of possibilities, [it] is precisely [when] we need and must permit a free press to ask the question." Chapin, 993 F.2d at 1096.

Thus, under the first part of the test we apply, we conclude that the general context in which the statements were made negates the impression that they imply a false assertion of fact. As we have already noted, readers would expect the book to contain the author's personal and subjective views about the trial and the conduct of the participants based on his experiences as a lawyer and his involvement in the case. Given the ambiguous nature of the subject matter from which Bugliosi draws his conclusions, we believe that the First Amendment requires us to give the author substantial latitude in describing and interpreting the events involved.

Although the made-for-television movie represents a distinct type of forum, we conclude that the general tenor of the docu-drama also tends to negate the impression that the statements involved represented a false assertion of objective fact. As the Supreme Court has noted, statements made in "a so-called docudrama or historical fiction" should not be accepted unquestioningly. Masson, 501 U.S. at 512-13, 111 S.Ct. at 2430-31. Docudramas, as their names suggests, often rely heavily upon dramatic interpretations of events and dialogue filled with rhetorical flourishes in order to capture and maintain the interest of their audience. We believe that viewers in this case would be sufficiently familiar with this genre to avoid assuming that all statements within them represent assertions of verifiable facts. To the contrary, most of them are aware by now that parts of such programs are more fiction than fact. As the District of Columbia Circuit has observed, "it is in part the settings of the speech in question that makes their hyperbolic nature apparent, and which helps determine the way in which the intended audience will receive them." Moldea II, 22 F.3d at 314; cf. Ollman v. Evans, 750 F.2d 970, 984 (D.C. Cir. 1984) ("[C]ourts have long `considered the influence that . . . well-established genres of writing will have on the average reader.'"), cert. denied, 471 U.S. 1127, 105 S.Ct. 2662, 86 L.Ed.2d 278 (1985).

We now turn from the general to the specific. Even in contexts in which the general tenor of the work suggests that the author is expressing personal opinions, it is possible that a particular statement of opinion may imply a false assertion of objective fact and therefore fall outside the scope of the First Amendment's protection as limited by Milkovich. We do not intimate that the First Amendment shields from scrutiny every assertion in a book outlining a particular author's perspective on a public controversy or every statement made in a docudrama based upon such an event; indeed, Milkovich makes clear that authors of both types of publications must attempt to avoid creating the impression that they are asserting objective facts rather than merely stating subjective opinions. Accordingly, before reaching a final judgment, we must look to the specific context in which the statements were made and to the content of the statements themselves in order to determine whether they are protected by the First Amendment. We now consider the specifics of the defendants' statements.

II.

Partington's defamation claim rests on three statements by the defendants that are outlined in Counts II, V, and VIII of the complaint. The second and fifth counts are based on statements in the book. Count II is based on a statement found in a passage that discusses the cross-examination in the Walker trial of a witness who had had radio communications with one of the murder victims and had testified about those communications in two prior theft trials. Bugliosi reports that in the earlier trials the witness had never mentioned anything about a truce between the two couples that supposedly involved the gift of a cake, but that in the Walker trial the witness testified that such a truce had occurred and was never asked why he failed to tell the previous juries about the alleged incident. He then expresses incredulity that Partington and his cocounsel did not question the witness about the apparent conflict and suggests that defense counsel may have failed to do their homework. The statement reads:

From my review of the transcripts of the two theft trials, I know that [the witness], with every opportunity to do so, had never mentioned anything about a cake or a truce in his earlier testimony. Incredibly, Findlay, in his brief cross-examination, did not bring out this critical contradiction. Had Walker's defense lawyers not read the theft-trial transcripts? Our copy had ended up in a warehouse; perhaps theirs had, too.

And the Sea will Tell, supra, at 232-33. Partington claims that this passage is defamatory because it implies that he did not read the transcripts and that he therefore did not adequately represent his client.

The second statement upon which Partington bases his defamation claim, found in the fifth count, reads as follows:

Between July 6 and August 26, there were a total of twenty-three entries [in Jennifer's diary] concerning Jennifer's and Buck's contacts with the Grahams. I had Jennifer read each one to the jury.

"The contact, then, between and among the four of you was of a considerable nature? Is that correct?"

"Yes."

The Walker jury had never heard this fact.

And the Sea Will Tell, supra, at 408-09. Partington claims that this statement was defamatory because it suggests that he should have offered the diary entries into evidence at Walker's trial and, accordingly, that he represented his client poorly. We consider both counts together.

Reading each of the statements in context, we find that the statements themselves, as well as the implications that Partington attributes to them, do not represent assertions of objective fact. When one reads the first passage in context, it is clear that Bugliosi does not claim to know the reason for the defense lawyers' failure to bring out the existence of the contradiction; rather, he speculates on the basis of the limited facts available to him. The passage clearly represent Bugliosi's personal interpretation of the available information and not a verifiable factual assessment of Partington's conduct. As the Seventh Circuit has noted:

A statement of fact is not shielded from an action for defamation by being prefaced with the words "in my opinion," but if 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.

Haynes v. Alfred A. Knopf, Inc., 8 F.3d 1222, 1227 (7th Cir. 1993).

With regard to the second statement, Bugliosi merely outlines a set of facts, allowing the reader to draw his own conclusion about them. Even if we were to attribute to Bugliosi's statement the implication that Partington contends arises from it — that Partington represented his client poorly — Bugliosi can only be said to have expressed his own opinion after having outlined all of the facts that serve as the basis for his conclusion.

The courts of appeals that have considered defamation claims after Milkovich have consistently held that when a speaker outlines the factual basis for his conclusion, his statement is protected by the First Amendment. As the Fourth Circuit noted, "[b]ecause the bases for the . . . conclusion are fully disclosed, no reasonable reader would consider the term anything but the opinion of the author drawn from the circumstances related." Chapin, 993 F.2d at 1087. Similarly, the District of Columbia Circuit has noted that "`[b]ecause the readers understand that such supported opinions represent the writer's interpretation of the facts presented, and because the reader is free to draw his or her own conclusions based upon those facts, this type of statement is not actionable in defamation.'" Moldea II, 22 F.3d at 317 (citing Moldea v. New York Times Co., 15 F.3d 1137 (D.C. Cir. 1994), Moldea I). Finally, the First Circuit has held that, as long as the author presents the factual basis for his statement, if can only be read as his "personal conclusion about the information presented, not as a statement of fact." Phantom Touring, Inc., 953 F.2d at 730 (emphasis added). Thus, even if Bugliosi had explicitly written what Partington contends his statements imply, the statements would be protected since, read in context, they are not statements implying the assertion of objective facts but are instead interpretations of the facts available to both the writer and the reader. Thus, we join with the other courts of appeals in concluding that when an author outlines the facts available to him, thus making it clear that the challenged statements represent his own interpretation of those facts and leaving the reader free to draw his own conclusions, those statements are generally protected by the First Amendment.

Further, at least with regard to the first passage, the rhetorical device used by Bugliosi negates the impression that his statement implied a false assertion of fact. Bugliosi's use of a question mark serves two purpose: it makes clear his lack of definitive knowledge about the issue and invites the reader to consider the possibility of other justifications for the defendants' actions. As the Fourth Circuit noted:

A question can conceivably be defamatory, though it must reasonably be read as an assertion of a false fact; inquiry itself, however embarrassing or unpleasant to its subject, is not an accusation. The language cannot be tortured to "make that certain which is in fact uncertain."

Chapin, 993 F.2d at 1094 (citation omitted); see also Beverly Hills Foodland, Inc. v. United Food Commercial Workers Union, 39 F.3d 191, 195-96 (8th Cir. 1994). Indeed, the First Circuit has explicitly distinguished a question like Bugliosi's from the statements found actionable in Milkovich: "[W]hile the author's readers implicitly were invited to draw their own conclusions from the mixed information provided, the Milkovich readers implicitly were told that only one conclusion was possible." Phantom Touring, Inc., 953 F.2d at 731; see also Beverly Hills Foodland, 39 F.3d at 196.

Finally, Partington bases his third defamation claim, in Count VIII, on a statement that did not appear in the book. In the made-for-television movie, Bugliosi is portrayed as saying to his client, "If I defend you the way Partington is defending Walker, you'll spend the rest of your life in prison." Partington claims that the statement implies that he had adopted the wrong defense strategy and therefore provided inadequate representation to his client.

In contrast to the two statements above, the speaker does not outline the factual basis for his conclusion. We nevertheless believe that the context in which the statement was made negates the impression that it implied the assertion of an objective fact. The defendant's use of hyperbolic language strongly suggests that the movie character was not making an objective statement of fact. Unelko, 912 F.2d at 1053. In this case, as the district court noted, Partington, 825 F. Supp. at 925, the statement appears to be a rhetorical device used by an attorney to gain his client's confidence — and, at a broader level, a dramatic passage of dialogue designed to maintain the viewer's attention — rather than an objective statement of fact. As the Supreme Court emphasized in Milkovich, the First Amendment protects the "rhetorical hyperbole" and "imaginative expression" that enlivens writers' prose. Milkovich, 497 U.S. at 20, 110 S.Ct. at 2706. Accordingly, we conclude that the statement cannot be read to imply the assertion of an objective fact. Id. at 17, 110 S.Ct. at 2705.

III.

While we conclude that the general and specific contexts in which the contested statements were made negate the impression that they implied the assertion of an objective fact, we do not rest our decision solely upon that analysis. Instead, continuing to follow the Unelko framework, we turn to the third part of the test our court applies: whether the statements at issue are capable of being proved true or false. Here, too, the answer to our inquiry compels the conclusion that Partington has failed to state an actionable defamation claim. Partington contends that all of the contested statements imply that he should have adopted different trial strategies and therefore that he provided inadequate representation to his client. We conclude that negative statements concerning a lawyer's performance during trial, even if made explicitly, are generally not actionable since they are not ordinarily "susceptible of being proved true or false." Milkovich, 497 U.S. at 21, 110 S.Ct. at 2707.

To begin, assessments of a lawyer's trial performance are inherently subjective and therefore not susceptible of being proved true or false. Opinions vary significantly concerning what skills make a good trial lawyer and whether a particular individual possesses them. There is no objective standard by which one can measure an advocate's abilities with any certitude or determine conclusively the truth or falsity of statements made regarding the quality of his or her performance. Moreover, as the Supreme Court has noted, there is a wide variation in opinion concerning the appropriate trial strategy that should be pursued in a given circumstance; in the words of the Court, "[t]here are countless ways to provide effective assistance in any given case. Even the best criminal defense attorneys would not defend a particular client the same way." Strickland v. Washington, 466 U.S. 668, 689, 104 S.Ct. 2052, 2065, 80 L.Ed.2d 674 (1984). Critiques of a lawyer's performance in a particular case generally cannot be proved true or false and, consequently, cannot ordinarily serve as the basis of a defamation claim. Reasonable minds can and do differ as to what strategy should be adopted in a trial, particularly in a trial before a jury. Indeed, what may be a good strategy before one jury may be a disastrous one before another. Thus, we decline to chill Bugliosi's right to express his opinion regarding what constituted good and bad trial strategy in the Palmyra Island cases.

Our conclusion that statements about a lawyer's performance constitute "subjective" statements and are not susceptible of being proved true or false is well supported by existing caselaw. For example, the Seventh Circuit has held that criticisms of an attorney's performance "are more in the nature of opinions on performance rather than statements of fact." Quilici v. Second Amendment Foundation, 769 F.2d 414, 418 (7th Cir. 1985), cert. denied, 475 U.S. 1013, 106 S.Ct. 1192, 89 L.Ed.2d 307 (1986). Indeed, our court has gone so far as to hold that "[t]he inference that [the plaintiff's] . . . legal abilities were doubtful . . . is a broad, unfocused, wholly subjective comment . . . ." Lewis v. Time, Inc., 710 F.2d 549, 554 (9th Cir. 1983) (emphasis added). Here, we conclude only that an evaluation of a lawyer's performance in a specific trial is not actionable; the case before us is in the Quilici rather than the Lewis mold.

The courts of appeals that have recently addressed claims based upon comments that are in the nature of personal assessments or criticisms have continued to hold that the First Amendment protects subjective evaluations because they are not susceptible of being verified as true or false. For example, the First Circuit has held that a statement alleging that a production was "fake" and "phony" is "unprovable, since those adjectives admit of numerous interpretations." Phantom Touring, Inc., 953 F.2d at 728. Similarly, the Fourth Circuit has held that the phrase "hefty mark-up" is too subjective a phrase to be verifiable, Chapin, 993 F.2d at 1093, and the Eighth Circuit has reached a similar conclusion concerning the term "unfair." Beverly Hills Foodland, Inc., 39 F.3d at 196.

We find the District of Columbia Circuit's recent decision in Moldea v. New York Times Co., 22 F.3d 310 (D.C. Cir. 1994) ( Moldea II), to be especially relevant to our decision. The District of Columbia Circuit, in a reversal of its earlier opinion in Moldea I, recently rejected a claim similar to Partington's. In Moldea II, the plaintiff alleged that statements made by an author implied that he was an incompetent journalist. The court held that criticisms of a journalist's "sloppy journalism" and unprofessional techniques are not actionable under Milkovich because "[r]easonable minds can and do differ as to how to interpret a literary work." Id. at 316.

The District of Columbia Circuit emphasized that courts should be reluctant to hold comments concerning the professional abilities of an individual actionable, noting that "it is highly debatable whether [a statement regarding the plaintiff's "sloppy journalism"] is sufficiently verifiable to be actionable in defamation." Moldea II, 22 F.3d at 317. Indeed, the District of Columbia Circuit noted that in Moldea I it had "failed adequately to heed the counsel of both the Supreme Court and our own precedents that `[w]here the question of truth or falsity is a close one, a court should err on the side of nonactionability.'" Moldea II, 22 F.3d at 317 (quoting Liberty Lobby, Inc. v. Dow Jones Co., 838 F.2d 1287, 1292 (D.C. Cir.), cert. denied, 488 U.S. 825, 109 S.Ct. 75, 102 L.Ed.2d 51 (1988)).

We agree with the District of Columbia Circuit that statements like the ones before us are not actionable. Authors should have "breathing space" in order to criticize and interpret the actions and decisions of those involved in a public controversy. If they are not granted leeway in interpreting ambiguous events and actions, the public dialogue that is so important to the survival of our democracy will be stifled. We must not force writers to confine themselves to dry, factual recitations or to abstract expressions of opinion wholly divorced from real events. Within the limits imposed by the law, we must allow, even encourage, them to express their opinions concerning public controversies and those who become involved in them.

We find these principles to be especially relevant with regard to controversial trials like the one before us. While we recognize that there is no wholesale protection extended to commentary upon trials, see Time, Inc. v. Firestone, 424 U.S. 448, 457, 96 S.Ct. 958, 966, 47 L.Ed.2d 154 (1976), we believe it important that the public be fully informed about what transpires inside our courtrooms. This is especially true when, as here, one of the attorneys has publicly accused the presiding judge of biased decisionmaking and a government witness of lying. Thus, while the divorce of a socialite does not in itself constitutes a matter of public controversy, Time, Inc., 424 U.S. at 457, 96 S.Ct. at 966, controversial trials that raise questions concerning the fairness of the justice system clearly do. Indeed, in cases like the one before us, we believe that openness, debate, and the free exchange of ideas are necessary to maintain the legitimacy of the court in the eyes of the public.

We recognize that there is a risk in exposing the judicial process to full and open examination and an even greater risk in expanding the boundaries of permissible criticism that may be leveled at judicial officers and lawyers. Substantial harm occurs when over a period of time the public views highly publicized but unrepresentative proceedings that significantly mislead it regarding what transpires in the normal course of trials. Similarly, permitting highly subjective criticism of judges and opposing counsel involves a true cost. In the short run, respect for the justice system may be lessened by any or all of these events. Nevertheless, the Constitution requires that we permit the people to be fully informed about the operations of government, including the operation of the judicial branch. It also requires that we tolerate individual expressions of opinion, hostile or otherwise, regarding the performance of those who carry out all aspects of our governmental functions.

For the reasons set forth above, we conclude that the contested statements, and the implications Partington contends arise from them, are absolutely protected by the First Amendment and cannot serve as the basis for a defamation claim. Thus, we affirm the district court's decision to dismiss Counts II, V, and VIII.

IV.

Partington also contends that he is placed in a false light by two statements we have discussed in Parts I-III, as well as by two other statements that we have not previously discussed. The first of the two additional statements is that Partington (and his cocounsel Findlay) "stuck with their submissive stance [during the Walker murder trial] not unlike steers being led to the slaughterhouse." See supra note 3. The second statement relates to Partington's failure to call a prisoner by the name of J.W. Williams as a witness. See supra note 5.

We reject Partington's false light claims regarding the two contested statements discussed in Parts I-III of our opinion for the same reason that we rejected his defamation claims based on those statements: both statements are protected by the First Amendment, regardless of the form of tort alleged. See Moldea II, 22 F.3d at 319. Thus, we affirm the district court's dismissal of Counts III and VI of Partington's complaint.

We also reject Partington's two additional false light claims. The statement describing Partington and his cocounsel as "steers being led to the slaughterhouse" is protected for precisely the same reasons as the statements we have previously discussed. The general context in which the steers-to-the-slaughterhouse statement was made negates any implication that it constitutes a false assertion of fact. See supra pp. 14-19. So, too, does the content of the statement; it is cast in colorful, indeed hyperbolic, terms — precisely the type of rhetorical flourish that the First Amendment protects. See supra pp. 23-24. Moreover, it is, beyond question, not susceptible of being proved true or false: it solely concerns the quality of an attorney's trial performance in a particular case. See supra pp. 23-28.

Partington's remaining false light claim essentially involves three contentions. The first — that the passage casts him in a false light by implying that he represented Walker poorly — cannot give rise to an actionable claim for the reasons we have already explained. See id.

Second, Partington contends that, in contrast to other parts of the book which he concedes correctly state the facts, the passage at issue here misrepresents what actually transpired. However, the factual misstatements if any are, in the circumstances of this case, of minor importance. We have assumed in our previous discussion that Bugliosi's book portrays Partington as having represented Walker poorly. The additional charge adds little if anything to the gist of Bugliosi's criticisms. It merely sets forth a comparatively minor incident that tends to support the basic protected characterization. Courts have consistently rejected attempts to base damage claims upon minor factual errors when the gist of the work, taken as a whole, cannot serve as the basis for a defamation or false light claim. See e.g., Moldea II, 22 F.3d at 318-19; White, 909 F.2d at 520. As the Seventh Circuit noted, "in determining the `gist' or `sting' of a newspaper article [to assess whether it is actionable, a court] must look at the highlight of the article, the pertinent angle of it, and not to items of secondary importance . . . ." Vachet v. Central Newspapers, 816 F.2d 313, 317 (7th Cir. 1987). We must evaluate Partington's final claim in light of the general picture which exists independent of the statement on which the claim is based. If, as Partington himself contends, all of the other contested statements — which we hold protected — would lead a reader to believe that he represented his client poorly, the additional item (even assuming it includes factual errors) will not in any way affect the reader's view of Partington. The light in which this remaining claim puts Partington is precisely the light in which he has already been put by material that is protected by the First Amendment. Under these circumstances, the additional example of which Partington complains cannot support a false light action.

Finally, Partington contends that by stating that his version of the facts is not correct, the passage implies that he is a "liar" and therefore casts him in a false light. We reject this contention because the inference that Partington draws is not tenable. Bugliosi has not accused him of being a "liar" or a "perjurer." He has instead asserted that Partington's explanation of one incident to a prospective author is not consistent with the facts.

It is true that Bugliosi has suggested that at some time during the lengthy and detailed interviews that Partington voluntary granted to the authors of And the Sea Will Tell he deliberately misrepresented what transpired in connection with the decision not to call a particular witness. However, such a charge is not sufficient to support a false light claim. There is a significant difference between suggesting that someone misrepresented the facts regarding a single tactical decision in a criminal trial and accusing that individual of being a "liar" or a "perjurer." The disputed statement, read in context, simply is not of the same order as the example of actionable defamation set forth in Milkovich — the statement that "Mayor Jones is a liar," Milkovich, 497 U.S. at 20, 110 S.Ct. at 2706. Moreover, Milkovich's holding does not extend nearly so far as to cover an allegation that a person engaged in the type of minor misrepresentation of which Bugliosi accuses Partington. Id. Reporters and historians routinely dispute the accuracy or truthfulness of the statements of their sources when those statements conflict with the facts as the authors perceive them. We would severely limit the ability of such writers to explain fully many of the ramifications of crucial issues of public importance were we to allow them to be sued every time they suggested that one of their sources was being less than truthful in describing an incident that is discussed in the published work. Finally, we note that unlike the obligation of a witness testifying in court, a person being interviewed by an author-for-profit is under no duty of full and accurate disclosure. A comparatively minor misrepresentation made to such a writer clearly does not render the interviewee a liar or a perjurer.

For the above reasons, we affirm the district court's dismissal of Counts IV and VII of Partington's complaint.

V.

Finally, we conclude that the district court did not err in denying Partington's request for leave to amend his complaint. Under Rule 15(a) of the Federal Rules of Civil Procedure, leave to amend "shall be freely given when justice so requires." Although, as Partington points out, there is a policy that favors allowing parties to amend their pleadings, Howey v. United States, 481 F.2d 1187, 1190 (9th Cir. 1973), a district court may properly deny such a motion if it would be futile to do so. DCD Programs, Ltd. v. Leighton, 833 F.2d 183, 186 (9th Cir. 1987). Here, the district court correctly concluded that, because Partington had failed to state a claim for defamation or false light, it would be futile to allow him to amend his complaint to plead malice or special damages. Partington, 825 F. Supp. at 925. Because it is clear that the deficiency in Partington's complaint could not have been overcome by amendment, the district court did not abuse its discretion in this case. DCD Programs, Ltd., 833 F.2d at 186.

CONCLUSION

For the above reasons, the district court's decisions denying Partington leave to amend and dismissing his complaint with prejudice are

AFFIRMED.