ROSLYN LA LIBERTE, Plaintiff-Appellant,
JOY REID, Defendant-Appellee.
We hold (for the first time) that California’s anti-SLAPP statute is inapplicable in federal court because it conflicts with Federal Rules of Civil Procedure 12 and 56. We also vacate the dismissal of the defamation claim under Rule 12(b)(6). As to one of the statements at issue, the court erroneously deemed La Liberte to be a limited purpose public figure (and accordingly dismissed for failure to plead actual malice); as to the other, the court mischaracterized it as nonactionable opinion. We affirm the district court’s conclusion that Reid does not qualify for immunity under section 230 of the Communications Decency Act.
G. TAYLOR WILSON (L. Lin Wood, Nicole Jennings Wade, on the brief), L. Lin Wood, P.C., Atlanta, GA, for Plaintiff-Appellant Roslyn La Liberte.
JOHN H. REICHMAN (Jason L. Libou, on the brief), Wachtel Missry LLP, New York, NY, for Defendant-Appellee Joy Reid.
The Reporters Committee for Freedom of the Press and 21 Media Organizations, Washington, DC, (Bruce D. Brown and Katie Townsend), filed a brief as Amici Curiae, in support of Defendant-Appellee.
Before: KEARSE, JACOBS, CABRANES, Circuit Judges.
JACOBS, Circuit Judge.
Plaintiff Roslyn La Liberte spoke at a 2018 city council meeting to oppose California’s sanctuary-state law; soon after, a social media activist posted a photo showing the plaintiff with open mouth in front of a minority teenager; the caption was that persons (unnamed) had yelled specific racist remarks at the young man in the photo. Defendant Joy Reid, a personality on cable television, retweeted that post, an act that is not alleged to be defamatory. The defamation claim is based on Reid’s two later posts: her June 29 post showed the photograph and attributed the specific racist remarks to La Liberte; her July 1 post, to the same effect, juxtaposed the photograph with the 1957 image of a white woman in Little Rock screaming execrations at a Black child trying to go to school.
The teenager who was photographed with La Liberte soon after publicly explained that La Liberte did not scream at him and that they were having a civil discussion. La Liberte sued Reid for defamation in the United States District Court for the Eastern District of New York.
The district court (Irizarry, Ch. J.) rejected Reid’s defense of immunity under section 230 of the Communications Decency Act, see 47 U.S.C. § 230(c)(1) (“Section 230”), but nevertheless dismissed La Liberte’s defamation claim as to both of Reid’s posts. The court deemed La Liberte to be a limited purpose public figure and held that she failed to allege actual malice as to the first post, and rejected the claim as to the second post on the ground that it was nonactionable opinion. Moreover, the court “struck” La Liberte’s defamation claim—and imposed attorneys’ fees (to be assessed)—under California’s Anti-Strategic Lawsuits Against Public Participation (“anti-SLAPP”) statute for failure to establish “a probability that the plaintiff will prevail.” Cal. Civ. Proc. Code § 425.16(b)(1),(c)(1). La Liberte appeals on the grounds that she was not a limited purpose public figure, that both posts were defamatory, and that California’s anti-SLAPP statute is inapplicable in federal court. Reid argues that the court erroneously denied Section 230 immunity as to her first post.
As a matter of first impression in this Circuit, we hold that California’s anti-SLAPP statute is inapplicable in federal court because it increases a plaintiff’s burden to overcome pretrial dismissal, and thus conflicts with Federal Rules of Civil Procedure 12 and 56 (Point I).
As to the merits, we agree with the district court that Reid cannot claim immunity under Section 230 of the Communications Decency Act (Point II). This lawsuit does not treat Reid as “the publisher or speaker of any information provided by another information content provider.” 47 U.S.C. § 230(c)(1) (emphasis added). To the contrary, she is the sole author of both allegedly defamatory posts.
We disagree with the rest of the district court’s analysis under Rule 12(b)(6). La Liberte was not a public figure on the matter in controversy, primarily because she lacked the regular and continuing media access that is a hallmark of public-figure status. (Point III). Accordingly, she was not required to allege that Reid acted with actual malice as to either post. Moreover, the court erred by characterizing Reid’s second post as nonactionable opinion (Point IV). That post could be interpreted as accusing La Liberte of engaging in specific racist conduct, which is a provable assertion of fact and therefore actionable.
The facts are plentiful but straightforward. Roslyn La Liberte is a California citizen who avows that she is “passionate about this country’s immigration policies.” (App. at 13.) She took a particular interest in California Senate Bill 54 (“SB 54”), a controversial 2017 law that limits cooperation between local law enforcement and federal immigration authorities. One provision is that state and local law enforcement officers are barred from disclosing (inter alia) an alien’s address and date of release from prison. Cal. Gov’t Code § 7284.6(a)(1). To register her opposition, La Liberte attended city council meetings in several cities, speaking out at some of them to urge resistance. On June 25th, 2018, La Liberte attended one such meeting in Simi Valley, California, along with hundreds of other people, where she spoke for about two minutes (the “Council Meeting”).
At some point during the Council Meeting, La Liberte was photographed interacting with a fourteen-year-old teenager who appears to be (and is) Hispanic (the “Photograph”). (See App. at 265.) The Photograph showed La Liberte with her mouth open and her hand at her throat in a gagging gesture. On June 28th, a social media activist named Alan Vargas tweeted the Photograph along with the following caption:
“You are going to be the first deported” [and] “dirty Mexican” [w]ere some of the things they yelled they yelled [sic] at this 14 year old boy. He was defending immigrants at a rally and was shouted down. Spread this far and wide this woman needs to be put on blast.
(App. at 67.) The Photograph went viral. The next day, Joy Reid, a personality on the MSNBC cable station, retweeted (i.e., shared) the Vargas tweet to her approximately 1.24 million followers. (La Liberte is not alleging defamation by Reid as to that communication.)
Later that same day (June 29), Reid posted the Photograph on her Instagram with the following caption:
He showed up to a rally to defend immigrants . . . . She showed up too, in her MAGA hat, and screamed, “You are going to be the first deported” . . . “dirty Mexican!” He is 14 years old. She is an adult. Make the picture black and white and it could be the 1950s and the desegregation of a school. Hate is real, y’all. It hasn’t even really gone away.
(the “June 29 Post”) (App. at 84.) Meanwhile, the teenager in the Photograph stated during an interview with Fox 11 Los Angeles that La Liberte did not yell any racial slurs and that their discussion was “civil.” (App. at 38, 47.) Still, La Liberte began receiving hate mail, including threats of mutilation and recommendations that she commit suicide.
Two days later (July 1), Reid published another post about La Liberte, this time on Instagram and Facebook. This post juxtaposed the Photograph of La Liberte with the 1957 photograph showing one of the Little Rock Nine walking past a screaming white woman. Reid added the following caption:
It was inevitable that this [juxtaposition] would be made. It’s also easy to look at old black and white photos and think: I can’t believe that person screaming at a child, with their face twisted in rage, is real. By [sic] every one of them were. History sometimes repeats. And it is full of rage. Hat tip to @joseiswriting. #regram #history #chooselove
(the “July 1 Post”) (App. at 87.) La Liberte hired a lawyer, who contacted Reid on July 2 to demand that she take down the posts and apologize. Reid responded that evening by removing them from her accounts and issuing the following statement: “It appears I got this wrong. My apologies to Mrs. La Liberte and [the teenager].” (App. at 116.)
La Liberte sued Reid for defamation in the Eastern District of New York, claiming that Reid’s June 29 and July 1 posts falsely accused her of yelling racist slurs at the teenager in the Photograph. Applying California law by agreement of the parties, the district court dismissed La Liberte’s claim under Rule 12(b)(6) and struck it under California’s anti-SLAPP statute. And since that statute mandates fee shifting, the court “granted [Reid] leave to seek attorneys’ fees and costs.” (App. at 279.)
We review de novo a district court’s grant of a motion to dismiss, “constru[ing] the complaint liberally, accepting all factual allegations in the complaint as true, and drawing all reasonable inferences in the plaintiff’s favor.” Palin v. New York Times Co., 940 F.3d 804, 809 (2d Cir. 2019) (quoting Elias v. Rolling Stone LLC, 872 F.3d 97, 104 (2d Cir. 2017)). Likewise, whether California’s anti-SLAPP statute can be applied in federal court raises a question of law, which we review de novo. Knight v. State Univ. of N.Y. at Stony Brook, 880 F.3d 636, 640 (2d Cir. 2018); Adelson v. Harris, 774 F.3d 803, 807, 809 (2d Cir. 2014).
We begin with the procedural issue posed by the court’s decision to strike La Liberte’s defamation claim under California’s anti-SLAPP statute. For a category of cases related to a defendant’s speech, that statute subjects any claim to dismissal “unless the court determines that the plaintiff has established that there is a probability that the plaintiff will prevail on the claim.” Cal. Civ. Proc. Code § 425.16(b)(1). Many states have enacted “anti-SLAPP statutes” with the idea that they provide breathing space for free speech on contentious public issues. See Abbas v. Foreign Policy Grp., LLC, 783 F.3d 1328, 1332 (D.C. Cir. 2015). The aim is “to decrease the `chilling effect’ of certain kinds of libel litigation and other speech-restrictive litigation . . . by making it easier to dismiss defamation suits at an early stage of the litigation.” Id. (internal quotation marks omitted) (quoting Eugene Volokh, The First Amendment and Related Statutes 118 (5th ed. 2014)).
Specifically, California’s anti-SLAPP statute was enacted to provide “an efficient procedural mechanism for the early and inexpensive dismissal of nonmeritorious claims `arising from any act’ of the defendant `in furtherance of the person’s right of petition or free speech . . . in connection with a public issue.'” Annette F. v. Sharon S., 119 Cal. App. 4th 1146, 1159 (Cal. Ct. App. 2004) (quoting Cal. Civ. Proc. Code § 425.16(b)(1)). A defendant is afforded 60 days from service of the complaint to file a “special motion to strike,” which must be granted “unless the court determines that the plaintiff has established that there is a probability that the plaintiff will prevail on the claim.” Cal. Civ. Proc. Code § 425.16(b)(1),(f). Additionally, a prevailing defendant on this special motion “shall be entitled” to recover attorney’s fees and costs. Id. § 425.16(c)(1).
California courts resolve these motions in two steps. “First, the court decides whether the defendant has made a threshold showing that the challenged cause of action is one arising from protected activity [i.e.,] . . . acts . . . taken `in furtherance of the defendant’s right of petition or free speech . . . .'” Equilon Enters. v. Consumer Cause, Inc., 29 Cal. 4th 53, 67 (Cal. 2002)(alteration omitted) (quoting Cal. Civ. Proc. Code § 425.16(b)(1)). If so, the court “determines whether the plaintiff has demonstrated a probability of prevailing on the claim.” Id. In doing so, “the court shall consider the pleadings, and supporting and opposing affidavits stating the facts upon which the liability or defense is based.” Cal. Civ. Proc. Code § 425.16(b)(2).
Our sister circuits split on whether federal courts may entertain the various state iterations of the anti-SLAPP special motion. The Fifth, Eleventh, and D.C. Circuits hold that they are inapplicable in federal court on the ground that they conflict with Federal Rules of Civil Procedure 12 and 56. See Klocke v. Watson, 936 F.3d 240, 242 (5th Cir. 2019) (Texas); Carbone v. Cable News Network, Inc., 910 F.3d 1345, 1350 (11th Cir. 2018) (Georgia); Abbas v. Foreign Policy Grp., LLC, 783 F.3d 1328, 1335 (D.C. Cir. 2015) (D.C.). The First and Ninth Circuits see no such conflict with the statutes of Maine and California, respectively. See Godin v. Schencks, 629 F.3d 79, 86-87 (1st Cir. 2010); United States ex rel. Newsham v. Lockheed Missiles & Space Co., 190 F.3d 963, 972 (9th Cir. 1999). We have decided some cases involving these special motions, but we have not yet decided the question of applicability.
The test is whether “a Federal Rule of Civil Procedure `answer[s] the same question’ as the [special motion to strike].” Abbas, 783 F.3d at 1333 (alteration in original) (quoting Shady Grove Orthopedic Assocs., P.A. v. Allstate Ins. Co., 559 U.S. 393, 398-99 (2010)). If so, the Federal Rule governs, unless it violates the Rules Enabling Act. Id. Applying that test, we first conclude that the special motion to strike in California’s anti-SLAPP statute answers the same question as Federal Rules 12 and 56.
The special motion to strike requires outright dismissal unless the plaintiff can “establish a probability that he or she will prevail on the claim.” Cal. Civ. Pro. Code § 425.16(b)(3). The statute thus “establishes the circumstances under which a court must dismiss a plaintiff’s claim before trial,” a question that is already answered (differently) by Federal Rules 12 and 56. Abbas, 783 F.3d at 1333-34. Under Rule 12(b)(6), the pleading burden is to allege “enough facts to state a claim to relief that is plausible on its face.” Bell Atl. Corp. v. Twombly, 550 U.S. 544, 570 (2007). This “does not impose a probability requirement at the pleading stage. . . . [A] well-pleaded complaint may proceed even if it strikes a savvy judge that actual proof of those facts is improbable.” Id. at 556. California’s anti-SLAPP statute, however, “abrogates that entitlement . . . by requiring the plaintiff to establish that success is not merely plausible but probable.” Carbone, 910 F.3d at 1353.
It also conflicts with Rule 56, which permits summary judgment only if “the movant shows that there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law.” Fed. R. Civ. P. 56(a). The Rule thus enables plaintiffs to proceed to trial by identifying any genuine dispute of material fact, whereas California’s anti-SLAPP statute “nullif[ies] that entitlement by requiring the plaintiff to prove that it is likely, and not merely possible, that a reasonable jury would find in his favor.” Carbone, 910 F.3d at 1353. Together, Rules 12 and 56 “express `with unmistakable clarity’ that proof of probability of success on the merits `is not required in federal courts’ to avoid pretrial dismissal.” Id. at 1351 (quoting Hanna v. Plumer, 380 U.S. 460, 470 (1965)). Therefore, California’s special motion requires the plaintiff to make a showing that the Federal Rules do not require.
Reid urges us to follow the Ninth Circuit, which holds that California’s anti-SLAPP statute and the Federal Rules “can exist side by side . . . without conflict.” Newsham, 190 F.3d at 972 (internal quotation marks omitted). We disagree—as do a number of Ninth Circuit judges.
Amici curiae argue that “[a]s compared to Rules 12 and 56, the anti-SLAPP motion to strike `serves the entirely distinct function of protecting those specific defendants that have been targeted with litigation on the basis of their protected speech'” and that it “`supplements rather than conflicts with the Federal Rules.'” (Amici Br. at 22 (first quoting Godin, 629 F.3d at 89; then quoting Makaeff v. Trump University, LLC, 736 F.3d 1180, 1182 (9th Cir. 2013) (Wardlaw, J., concurring in the denial of rehearing en banc)).) The idea that the more stringent requirement of the anti-SLAPP standard is a beneficial “supplement” to the Federal Rules is a policy argument—and fatal, because the more permissive standards of the Federal Rules likewise reflect policy judgments as to what is sufficient. See Shady Grove, 559 U.S. at 401 (explaining that because “Rule 23 permits all class actions that meet its requirements, . . . a State cannot limit that permission by . . . impos[ing] additional requirements”). Finally, amici warn that refusal to apply the anti-SLAPP statute will “encourage forum shopping” and lead to “an increased burden on federal courts in this Circuit.” (Amici Br. at 11.) That may be so; but our answer to a legal question does not turn on our workload; and in any event, the incentive to forum-shop created by a circuit split can be fixed, though not here.
Since Rules 12 and 56 answer the same question as California’s special motion to strike, they “govern in diversity cases in federal court, unless Rules 12 and 56 violate the Rules Enabling Act.” Abbas, 783 F.3d at 1336. “So far, the Supreme Court has rejected every challenge to the Federal Rules that it has considered under the Rules Enabling Act.” Id. Neither Reid nor amici curiae invite us to deviate. Still, we briefly address the question for the sake of completeness. The test is “whether a rule really regulates procedure,—the judicial process for enforcing rights and duties recognized by substantive law and for justly administering remedy and redress for disregard or infraction of them.” Sibbach v. Wilson & Co., 312 U.S. 1, 14 (1941). Like the Eleventh Circuit, “[w]e have little difficulty concluding” that Rules 12 and 56 “comply with the Rules Enabling Act,” particularly because they “`affect only the process of enforcing litigants’ rights and not the rights themselves.'” Carbone, 910 F.3d at 1357 (second alteration in original) (quoting Burlington N. R.R. Co. v. Woods, 480 U.S. 1, 8, (1987)). Accordingly, federal courts must apply Rules 12 and 56 instead of California’s special motion to strike.
* * *
Finally, Reid and amici curiae contend that she is entitled to attorneys’ fees under the anti-SLAPP statute based on the district court’s separate Rule 12(b)(6) dismissal. We disagree. “The Act does not purport to make attorney’s fees available to parties who obtain dismissal by other means, such as under Federal Rule 12(b)(6).” Abbas, 783 F.3d at 1337 n.5; see also Klocke, 936 F.3d at 247 n.6. (“Suffice to say that because [Texas’s anti-SLAPP statute] does not apply in federal court, the district court erred by awarding fees and sanctions pursuant to it.”).
California’s anti-SLAPP statute likewise awards attorneys’ fees only to “a prevailing defendant on a special motion to strike.” Cal. Civ. Pro. Code § 425.16(c)(1) (emphasis added). So Reid cannot recover attorneys’ fees based on the district court’s Rule 12(b)(6) dismissal (which was nevertheless erroneous, as we explain below in Points III and IV). Nor may she recover them under the anti-SLAPP statute if she later prevails by other means.
We agree with the district court that Reid does not enjoy immunity under Section 230 of the Communications Decency Act. See 47 U.S.C. § 230(c)(1) (“No provider or user of an interactive computer service shall be treated as the publisher or speaker of any information provided by another information content provider.”). This immunity has three elements: “(1) [the defendant] is a provider or user of an interactive computer service, (2) the claim is based on information provided by another information content provider and (3) the claim would treat [the defendant] as the publisher or speaker of that information.” FTC v. LeadClick Media, LLC, 838 F.3d 158, 173 (2d Cir. 2016) (alteration in original) (internal quotation marks omitted) (quoting Jane Doe No. 1 v. Backpage.com, LLC, 817 F.3d 12, 19 (1st Cir. 2016)). An “information content provider” is “any person or entity that is responsible, in whole or in part, for the creation or development of information provided through the Internet or any other interactive computer service.” 47 U.S.C. § 230(f)(3).
The second element is dispositive here. La Liberte’s claim is based on posts of which Reid is the author, not on “information provided by another content provider.” Vargas had tweeted about vile remarks that “they yelled” at the meeting. (App. at 67 (emphasis added).) Vargas did not attribute the remarks to La Liberte. The following day, Reid authored and published her own Instagram post (the June 29 Post), which attributed to La Liberte (albeit not by name) what Vargas attributed only generally to the unnamed “they.” (See App. at 84 (“She . . . screamed, `You are going to be the first deported’ . . . `dirty Mexican!'” (emphasis added)).) The post also included Reid’s commentary on the conduct alleged: “Make the picture black and white and it could be the 1950s and the desegregation of a school. Hate is real, y’all. It hasn’t even really gone away.” (App. at 84.)
As sole author of the June 29 Post, Reid alone was “responsible . . . for [its] creation or development,” which makes her the sole “information content provider.” 47 U.S.C. § 230(f)(3). Moreover, she went way beyond her earlier retweet of Vargas in ways that intensified and specified the vile conduct that she was attributing to La Liberte. She accordingly stands liable for any defamatory content. And she is similarly the sole “information content provider” for her July 1 Post, a point she does not contest.
Reid argues that “[t]he issue . . . is not whether a defendant posted or authored a publication, . . . but rather whether the publication is materially different from a prior internet publication.” (Appellee’s Br. at 42.) Thus Reid seeks Section 230 immunity on the ground that the June 29 Post “merely repeated what countless others had previously published before her, including Vargas and at least eight other individuals who specifically stated that La Liberte made racial slurs at the Council Meeting.” (Appellee’s Br. at 47.)
The contention is unsupported by fact or law. The June 29 Post did not “merely repeat” what Vargas had “previously published.” Among other salient differences, Reid’s post accused La Liberte of yelling racist insults at the teenager. Nor did Reid simply retweet or share a post that someone else authored. In effect, Reid is arguing that a plaintiff can sue only the first defamer. If that were so, a post by an obscure social media user with few followers, blogging in the recesses of the internet, would allow everyone else to pile on without consequence. No one’s reputation would be worth a thing.
Reid relies more persuasively on the “material contribution” test that we recognized in LeadClick. See Force v. Facebook, Inc., 934 F.3d 53, 68 (2d Cir. 2019) (“[W]e have recognized that a defendant will not be considered to have developed third-party content unless the defendant directly and `materially’ contributed to what made the content itself `unlawful.'” (quoting LeadClick, 838 F.3d at 174)). We apply this test to “draw the line at the crucial distinction between, on the one hand, taking actions . . . to . . . display . . . actionable content and, on the other hand, responsibility for what makes the displayed content [itself] illegal or actionable.” Id. (alterations in original) (internal quotation marks omitted) (quoting Kimzey v. Yelp! Inc., 836 F.3d 1263, 1269 n.4 (9th Cir. 2016)).
That test does not serve Reid because she did not pass along or edit “third-party content”; she authored both Posts at issue. To illustrate: in Force, victims of Hamas-organized terrorist attacks in Israel sought to hold Facebook responsible on the ground that “Hamas . . . used Facebook to post content that encouraged terrorist attacks in Israel.” 934 F.3d at 59. Facebook was immune under Section 230, as we held, because Facebook did not “`develop’ the content of the . . . postings by Hamas”; nor does Facebook “edit (or suggest edits) for the content that its users . . . publish.” Id. at 69-70. On the other hand, in LeadClick, the defendant “had `developed’ third parties’ content by giving specific instructions to those parties on how to edit `fake news’ that they were using in their ads.” Id. at 69 (summarizing LeadClick, 838 F.3d at 176).
Since Reid cannot claim immunity, we turn to the substance of the defamation claim.
The district court ruled that La Liberte was a limited purpose public figure on the California sanctuary-state controversy, and dismissed her claim as to the June 29 Post for failure to plead actual malice. See Ampex Corp. v. Cargle, 128 Cal. App. 4th 1569, 1577 (Cal. Ct. App. 2005) (noting the actual malice requirement for limited purpose public figures).
There are two kinds of public figures. “The all-purpose public figure . . . has achieved such pervasive fame or notoriety that he or she becomes a public figure for all purposes . . . . The limited purpose public figure . . . voluntarily injects him or herself or is drawn into a specific public controversy, thereby becoming a public figure on a limited range of issues.” Ampex Corp., 128 Cal. App. 4th at 1577. No one argues that La Liberte is an all-purpose public figure; the question is whether she became a limited purpose public figure with respect to California’s sanctuary-state law (SB 54), that is, did she “thrust [herself] to the forefront” of the controversy, “invite attention and comment[,] . . . [and] assume special prominence in [its] resolution.” Khawar v. Globe Int’l, Inc., 19 Cal. 4th 254, 263 (Cal. 1998) (internal quotation marks omitted) (quoting Gertz v. Robert Welch, Inc., 418 U.S. 323, 345, 351 (1974)). The district court answered affirmatively because La Liberte “attended and spoke about SB 54 at multiple city council meetings” and “appeared in a photograph in the Washington Post about the SB 54 controversy” one month before the Simi Valley Council Meeting. (App. at 273.)
That is not nearly enough. Thin as the findings are to begin with, the district court did not take into account the requirement that a limited purpose public figure maintain “regular and continuing access to the media.” Hutchinson v. Proxmire, 443 U.S. 111, 136 (1979). One reason for imposing the actual malice burden on public figures and limited purpose public figures is that “[t]hey have media access enabling them to effectively defend their reputations in the public arena.” Khawar, 19 Cal. 4th at 265 (citing Gertz, 418 U.S. at 344-45). We have therefore made “regular and continuing access to the media” an element in our four-part test for determining whether someone is a limited purpose public figure. Contemporary Mission, Inc. v. New York Times Co., 842 F.2d 612, 617 (2d Cir. 1988). The California cases cited by the district court similarly turn on media access.
La Liberte plainly lacked such media access. The earlier photograph, which showed her conversing, was in a Washington Post photo spread of attendees at an SB 54 protest. The article did not name La Liberte, let alone mention her views. The single caption described everyone depicted as “[s]upporters and opponents of [SB 54] rally[ing] and debat[ing] outside Los Alamitos City Hall.” (App. at 140-41.) Such incidental and anonymous treatment hardly bespeaks “regular and continuing access to the media.”
Nor does La Liberte’s participation at city council meetings. La Liberte is said to have “testif[ied] eight times around the state” (Appellee’s Br. at 26 (citing App. at 102-05)); but Reid does not identify instances in which the media singled out La Liberte’s participation as newsworthy. Nor does speech, even a lot of it, make a citizen (or non-citizen) fair game for attack. Imposition of the actual malice requirement on people who speak out at government meetings would chill public participation in politics and community dialogue.
True, La Liberte received media attention. Reid emphasizes that La Liberte appeared for a television interview after Vargas published his tweet but before Reid’s posts were published. However, media access that becomes available only “after and in response to” damaging publicity does not make someone a public figure. Khawar, 19 Cal. 4th at 266. By the time of the interview, the Photograph had gone viral, along with accusations that La Liberte had screamed vile racist remarks at a child. The interview was “only the media access that would likely be available to any private individual who found himself the subject of sensational and defamatory accusations.” Id. “If such access were sufficient . . ., any member of the media . . . could confer public figure status simply by publishing sensational defamatory accusations against any private individual.” Id.
It makes little sense to deem La Liberte a limited purpose public figure when she stepped forward solely to defend her reputation. People become limited purpose public figures only when they “voluntarily invite comment and criticism” by “injecting themselves into public controversies.” Id. at 265 (emphasis added) (citing Gertz, 418 U.S. at 344-45). La Liberte, however, did not use the interview to inject herself to the forefront of the sanctuary-state controversy; she was pulled into a spotlight. Her experience suggests why the Supreme Court has only hypothetically recognized the notion of an involuntary public figure.
Since La Liberte was not a limited purpose public figure, the district court erred by requiring her to allege actual malice, and her claim as to the June 29 Post should not have been dismissed for failing to do so. On remand, the district court may assess whether La Liberte adequately alleged that Reid acted negligently with respect to that post, the standard for private-figure plaintiffs. See Khawar, 19 Cal. 4th at 274; Brown v. Kelly Broadcasting Co., 48 Cal. 3d 711, 742 (Cal. 1989) (“[A] private person need prove only negligence (rather than malice) to recover for defamation.”).
The district court dismissed La Liberte’s claim as to the July 1 Post on the ground that it “express[ed] nonactionable statements of opinion.” (App. at 275.) We disagree. A reader could have understood the July 1 Post as equating La Liberte’s conduct with archetypal racist conduct, which is a provable assertion of fact, and therefore actionable.
Whether a statement is nonactionable opinion “is a question of law to be decided by the court.” Baker v. L.A. Herald Examiner, 42 Cal. 3d 254, 260 (Cal. 1986). The test is “whether a reasonable fact finder could conclude the published statement declares or implies a provably false assertion of fact.” Franklin v. Dynamic Details, Inc., 116 Cal. App. 4th 375, 385 (Cal. Ct. App. 2004) (citing Milkovich v. Lorain Journal Co., 497 U.S. 1, 19 (1990)); see also Overhill Farms, Inc. v. Lopez, 190 Cal. App. 4th 1248, 1260 (Cal. Ct. App. 2010) (“Statements of opinion which imply a false assertion of fact are actionable.”).
Relatedly, “accusation[s] of concrete, wrongful conduct” are actionable while “general statements charging a person with being racist, unfair, or unjust” are not. Overhill Farms, 190 Cal. App. 4th at 1262. Overhill Farms is instructive: a press release and leaflets discussing a company’s termination of immigrant workers were actionable because they did not “merely accuse [the company] of being `racist’ in some abstract sense.” Id. Rather, the press release “contain[ed] language which expressly accuse[d] [the company] of engaging in racist firings,” and the leaflets “refer[red] to [the company’s] conduct as `racist and discriminatory abuse against Latina women immigrants.'” Id.
A reader could interpret the juxtaposition of the Photograph with the 1957 Little Rock image to mean that La Liberte likewise screamed at a child out of racial animus—particularly in light of Reid’s comment that “[h]istory sometimes repeats.” (App. at 87.) That interpretation is bolstered by Reid’s description of the white woman in the Little Rock photograph as a “person screaming at a child, with [her] face twisted in rage” and Reid’s comment that it was “inevitable” that the photos would be juxtaposed. Reid thus portrayed La Liberte as a latter-day counterpart of the white woman in 1957 who verbally assaulted a minority child. Like the defendants in Overhill Farms, Reid “did not merely accuse [La Liberte] of being `racist’ in some abstract sense.” 190 Cal. App. 4th at 1262. Rather, her July 1 Post could be understood as an “accusation of concrete, wrongful conduct,” which can be proved to be either true or false. Id. That makes it potentially defamatory.
Reid argues that readers could not understand the July 1 Post as defamatory absent familiarity with Reid’s June 29 Post or the Little Rock image, and that La Liberte was therefore required to plead defamation by implication and special damages. We disagree.
To be sure, California defamation law (which governs) recognizes two categories of libel, one of which requires proof of special damages. A publication is libelous “per se” when ” a reader would perceive a defamatory meaning without extrinsic aid beyond his or her own intelligence and common sense.” Bartholomew v. YouTube, LLC, 17 Cal. App. 5th 1217, 1226 (Cal. Ct. App. 2017) (quoting Barnes-Hind, Inc. v. Superior Court, 181 Cal. App. 3d 377, 386-87 (Cal. Ct. App. 1986)). A publication is libelous “per quod” if a “reader would be able to recognize a defamatory meaning only by virtue of his or her knowledge of specific facts and circumstances, extrinsic to the publication, which are not matters of common knowledge rationally attributable to all reasonable persons.” Id. at 1226-27 (quoting Barnes-Hind, Inc., 181 Cal. App. 3d at 386-87). And unlike libel per se, libel per quod “is not actionable unless the plaintiff alleges and proves that he has suffered special damage as a proximate result thereof.” Cal. Civ. Code § 45a.
Reid’s argument confuses libel per quod, which imposes the special damages requirement, with libel by implication, which can be libel per se nevertheless. “A statement can also be libelous per se if it contains a charge by implication from the language employed by the speaker and a listener could understand the defamatory meaning without the necessity of knowing extrinsic explanatory matter.” McGarry v. Univ. of San Diego, 154 Cal. App. 4th 97, 112 (Cal. Ct. App. 2007).
Readers who were unfamiliar with the June 29 Post could still interpret the July 1 Post to mean that Liberte engaged in racist conduct. The Little Rock encounter is a “matter of common knowledge rationally attributable to all reasonable persons.” Bartholomew, 17 Cal. App. 5th at 1226-27. Far from an obscure episode, it is a landmark event in one of the most vital historic developments of twentieth-century America, and the 1957 photograph is an indelible image of it. Presumably, that is why and how Reid used it.
Moreover, even those with an impoverished frame of reference could interpret the post as accusing La Liberte of engaging in racist conduct. There was no need for “extrinsic aid beyond [a reader’s] own intelligence and common sense.” Id. at 1226. The 1957 photograph shows a white woman “screaming at a [Black] child, with [her] face twisted in rage.” (App. at 87.) When viewing that image next to La Liberte’s Photograph and reading Reid’s comment that “[h]istory sometimes repeats,” a reader could believe that La Liberte had likewise engaged in racist conduct. And Reid “is liable for what is insinuated, as well as for what is stated explicitly.” Bartholomew, 17 Cal. App. 5th at 1227. Because that accusation is capable of being proven or disproven, the district court erred by characterizing the July 1 Post as nonactionable opinion.
* * *
Since the district court concluded that La Liberte adequately alleged malice with respect to the July 1 Post, it follows that La Liberte adequately alleged negligence, the standard for private-figure plaintiffs. Her claim as to this post should proceed to discovery.
For the reasons stated above, the district court’s judgment is VACATED and REMANDED for further proceedings consistent with this opinion.
 For our purposes, these anti-SLAPP statutes are analogous to California’s. Each raises the bar for plaintiffs to overcome a pretrial dismissal motion. See Carbone, 910 F.3d at 1348 (addressing Georgia’s anti-SLAPP statute, which imposes the identical requirement: plaintiffs must “establish that there is a probability that the nonmoving party will prevail on the claim”); Abbas, 783 F.3d at 1332 (addressing D.C.’s anti-SLAPP statute, which similarly requires plaintiffs to demonstrate that “the claim is likely to succeed on the merits”); see also Klocke, 936 F.3d at 244 (addressing Texas’s anti-SLAPP statute, which requires the non-movant to “establish by clear and specific evidence a prima facie case for each element of the claim in question”).
 Maine’s special motion requires dismissal “unless the party against whom the special motion is made shows that the moving party’s exercise of its right of petition was devoid of any reasonable factual support or any arguable basis in law and that the moving party’s acts caused actual injury to the responding party.” Godin, 629 F.3d at 82.
 Reid argues that we already held (or implied) that these special motions can be applied in federal court. Not so. She relies on Liberty Synergistics Inc. v. Microflo Ltd., 718 F.3d 138, 157 (2d Cir. 2013) (“Liberty I”), which vacated the denial of a California special motion to strike. But that case turned on choice-of-law principles under the Rules of Decision Act. Id. at 142, 157.The question on appeal was whether the district court could entertain California’s special motion to strike notwithstanding that it was applying New York substantive law. (The case originated in California state court. It was removed to federal court and then transferred to the Eastern District of New York.) Id. at 143. Liberty I therefore did not consider whether the special motion conflicts with any Federal Rule, and a later decision in that case clarified as much, noting that Liberty I “expressly declined to reach” the issue of “whether California’s anti-SLAPP statute is applicable in federal court.” Liberty Synergistics Inc. v. Microflo Ltd., 637 F. App’x 33, 34 n.1 (2d Cir. 2016) (summary order) (alteration omitted).
Reid also relies on Adelson v. Harris, 774 F.3d 803, 809 (2d Cir. 2014), which approved certain aspects of Nevada’s anti-SLAPP statute. That case is inapposite because Nevada’s statute is quite different. As the district court in Adelson recognized, “[t]he Nevada statute does not establish a `reasonable probability of success’ standard that must be met without discovery, like the California Anti-SLAPP law.” Adelson v. Harris, 973 F. Supp. 2d 467, 493 n.21 (S.D.N.Y. 2013). Instead, “the Nevada statute immunizes `good faith communication[s]’—defined as communications that are `truthful or . . . made without knowledge of . . . falsity’—thereby effectively raising the substantive standard that applies to a defamation claim.” Id. Accordingly, “even if the procedural elements of certain Anti-SLAPP statutes present [conflicts with the Federal Rules of Procedure], those problems [were] not presented in [Adelson], where the effects of the [Nevada] Anti-SLAPP law . . . are substantive.” Id. (citation omitted).
 See Makaeff v. Trump Univ., LLC, 736 F.3d 1180, 1188 (9th Cir. 2013) (Watford, J., joined by Kozinski Ch. J., Paez J., and Bea, J., dissenting from denial of rehearing en banc). Judge Watford explained that Newsham is no longer good law (to the extent it ever was) in light of the Supreme Court’s intervening decision in Shady Grove. See id. at 1189 (“Just as the New York statute in Shady Grove impermissibly barred class actions when Rule 23 would permit them, so too California’s anti-SLAPP statute bars claims at the pleading stage when Rule 12 would allow them to proceed.”); see also id. (“The anti-SLAPP statute eviscerates Rule 56 by requiring the plaintiff to prove that she will probably prevail if the case proceeds to trial—a showing considerably more stringent than identifying material factual disputes that a jury could reasonably resolve in the plaintiff’s favor.”).
 Amici moved for leave to file their brief on February 19, 2020. That motion is now granted.
 The California Legislature presumably could have awarded attorneys’ fees to the prevailing party in any defamation action, but it chose not to do so. See Abbas, 783 F.3d at 1335 (“Had the D.C. Council simply wanted to permit courts to award attorney’s fees to prevailing defendants in these kinds of defamation cases, it easily could have done so.”).
 By only addressing the June 29 Post, Reid concedes that her July 1 Post do es not qualify for Section 230 immunity.
 La Liberte’s initial complaint included Reid’s retweet of the Vargas tweet; but since La Liberte later dropped that claim, we need not decide whether a retweet qualifies for Section 230 immunity. Nor are we called to decide whether Section 230 protects a social media user who copies verbatim (and without attribution) another user’s post, a question that may be complicated by issues as to malice and status as a public figure.
 In Nadel v. Regents of the University of California, 28 Cal. App. 4th 1251 (Cal. Ct. App. 1994), one plaintiff was deemed to be a limited purpose public figure because he had “spoke[n] to print media reporters who included his comments in articles about the [controversy]” and had “been on 6 radio stations,” underscoring his “`more realistic opportunity to counteract false statements than private individuals normally enjoy.'” Nadel, 28 Cal. App. 4th at 1269-70 (quoting Gertz, 418 U.S. at 344); see also Rudnick v. McMillan, 25 Cal. App. 4th 1183, 1190 (Cal. Ct. App. 1994) (reasoning that a cattle rancher became a limited purpose public figure by successfully prompting two newspapers to write articles about his dispute with California’s Bureau of Land Management).
 The Court acknowledged the possibility of becoming an involuntary public figure but cautioned that “the instances of truly involuntary public figures must be exceedingly rare.” Gertz, 418 U.S. at 345. As the Supreme Court of California elaborated, “assuming a person may ever be accurately characterized as an involuntary public figure, we infer . . . that the [Supreme Court] would reserve this characterization for an individual who, despite never having voluntarily engaged the public’s attention in an attempt to influence the outcome of a public controversy, nonetheless has acquired such public prominence in relation to the controversy as to permit media access sufficient to effectively counter media-published defamatory statements.” Khawar, 19 Cal. 4th at 265 (last emphasis added). La Liberte is clearly not that exception. In any event, since Gertz, the Supreme Court has “emphasized the voluntary nature of the public-figure status.” Sack on Defamation: Libel, Slander, and Related Problems at 5-68 (5th ed. 2017).
 “`Special damages’ means all damages that plaintiff alleges and proves that he she has suffered in respect to his or her property, business, trade, or profession, or occupation, including the amounts of money the plaintiff alleges and proves he or she has expended as a result of the alleged libel, and no other.” Cal. Civ. Code § 48a(d)(2).
 The court reasoned that “it is plausible that [Reid] learned about the falsity of the content of the July 1 Post before publication,” including by way of emails from La Liberte’s son. (App. at 275.)