From Casetext: Smarter Legal Research

Trump v. Twitter, Inc.

United States District Court, Northern District of California
May 6, 2022
21-cv-08378-JD (N.D. Cal. May. 6, 2022)




DONALD J. TRUMP, et al., Plaintiffs, v. TWITTER INC., et al., Defendants.


Re: Dkt. No. 138


Former President Donald J. Trump, the American Conservative Union, and five individuals have sued Twitter, Inc., and Jack Dorsey (together, Twitter), on behalf of themselves and a putative class of Twitter users who have been “de-platformed” and “censored by Defendants.” Dkt. No. 21 (AC) ¶¶ 8, 18. Plaintiffs alleged claims under the First Amendment and Florida state consumer and “social media” statutes, and seek a declaration that Section 230 of the Communications Decency Act, which states that online service providers like Twitter cannot be held responsible for content posted by others, is unconstitutional. Id. ¶¶ 168-233. Twitter has moved to dismiss under Federal Rule of Civil Procedure 12(b)(6). Dkt. No. 138. The amended complaint is dismissed.


As alleged in the amended complaint, which the Court accepts as true for Rule 12(b)(6) purposes, see In re Capacitors Antitrust Litigation, 106 F.Supp.3d 1051, 1060 (N.D. Cal. 2015), Twitter is the well-known “social networking service that allows its Users to post and interact with each other through short messages known as ‘tweets.'” AC ¶ 28. Dorsey co-founded Twitter in 2006, and the company today hosts more than 500 million tweets posted daily by approximately 340 million users worldwide. Id. ¶¶ 28-29, 36.

Plaintiff Trump opened a Twitter account in May 2009 and was an active user until January 7, 2021. Id. ¶¶ 43-49, 113. On January 8, 2021, Twitter stated that it had “permanently suspended” the account “due to the risk of further incitement of violence.” Id. ¶ 114.

The amended complaint alleges that the other named plaintiffs also had their Twitter accounts treated unfavorably. Linda Cuadros's account was “permanently banned” in 2020 “due to a post about vaccines.” Id. ¶ 124. Rafael Barboza's account was “indefinitely suspended” on January 8, 2021, “after retweeting President Trump and other conservatives on January 6, 2021.” Id. ¶ 137. Dominick Latella's account “was permanently removed from the Defendants' platform during the 2018 election cycle” after he “post[ed] positive messages about Republican candidates and President Trump, ” although Latella has a “second account [which] is still active” albeit “shadow banned.” Id. ¶¶ 142-46. Wayne Allyn Root was “banned permanently by Twitter” after “multiple occasions where the Defendants censored his account for messages he posted related to COVID-19 and the 2020 election results.” Id. ¶¶ 152, 155. Dr. Naomi Wolf's account was “suspended” for “vaccine misinformation.” Id. ¶¶ 159, 162. The American Conservative Union “started noticing a reduction in engagement in its content” in 2017, and alleges its “followers were purged, ” dropping from 99, 000 followers in June 2020 to 88, 000 by January 19, 2021. Id. ¶¶ 12829.

In plaintiffs' view, these account actions were the result of coercion by members of Congress affiliated with the Democratic Party. Id. ¶¶ 51-64. Plaintiffs quote Senator Mark Warner (D-VA) as saying on October 28, 2020, that “[w]e can and should have a conversation about Section 230 -- and the ways in which it has enabled platforms to turn a blind eye as their platforms are used to . . . enable domestic terrorist groups to organize violence in plain sight.” Id. ¶ 55. Section 230 of the Communications Decency Act is said to have “significantly encouraged defendants' censorship of the plaintiff and the putative class members, ” id. ¶¶ 65-77, and the amended complaint alleges that defendants “willful[ly] participat[ed] in joint activity with federal actors to censor plaintiff and the putative class members.” Id. ¶¶ 78-112.

Plaintiffs allege: (1) a violation of the First Amendment to the United States Constitution; (2) that Section 230 of the Communications Decency Act is unconstitutional; (3) deceptive and misleading practices in violation of the Florida Deceptive and Unfair Trade Practices Act (FDUTPA), Florida Statutes § 501.201 et seq.; and (4) a violation of the Stop Social Media Censorship Act (SSMCA), Florida Statutes § 501.2041. Id. ¶¶ 168-233. In the prayer for relief, plaintiffs seek, among other things, compensatory and punitive damages, and injunctive and declaratory relief, including an order for Twitter to “immediately reinstate the Twitter accounts of” plaintiffs. Id. at 56.

This case was originally filed by plaintiffs in the United States District Court for the Southern District of Florida, Dkt. No. 1, and transferred to this District on Twitter's motion, which was made on the basis of a forum selection clause in Twitter's Terms of Service. Dkt. No. 87. The amended complaint, Dkt. No. 21, is the operative complaint. Defendants ask to dismiss all four of the claims in the AC for failure to plausibly state a claim. Dkt. No. 138.



Plaintiffs' main claim is that defendants have “censor[ed]” plaintiffs' Twitter accounts in violation of their right to free speech under the First Amendment to the United States Constitution. AC ¶¶ 168-87. Plaintiffs are not starting from a position of strength. Twitter is a private company, and “the First Amendment applies only to governmental abridgements of speech, and not to alleged abridgements by private companies.” Williby v. Zuckerberg, No. 3:18-cv-06295-JD, Dkt. No. 19 at 1 (N.D. Cal. June 18, 2019), appeal dismissed as frivolous, No. 19-16306, 2019 WL 11662186 (9th Cir. Nov. 25, 2019); see also Manhattan Cmty. Access Corp. v. Halleck, 139 S.Ct. 1921, 1928 (2019) (“the Free Speech Clause prohibits only governmental abridgement of speech. The Free Speech Clause does not prohibit private abridgment of speech.”) (emphases in original).

Plaintiffs' only hope of stating a First Amendment claim is to plausibly allege that Twitter was in effect operating as the government under the “state-action doctrine.” This doctrine provides that, in some situations, “governmental authority may dominate an activity to such an extent that its participants must be deemed to act with the authority of the government and, as a result, be subject to constitutional constraints.” Edmonson v. Leesville Concrete Co., 500 U.S. 614, 620 (1991); see also Manhattan Cmty. Access, 139 S.Ct. at 1928. This is not an easy claim to make, for good reasons. Private entities are presumed to act as such, and maintaining the line “between the private sphere and the public sphere, with all its attendant constitutional obligations, ” is a matter of great importance, as “[o]ne great object of the Constitution is to permit citizens to structure their private relations as they choose subject only to the constraints of statutory or decisional law.” Edmonson, 500 U.S. at 619. “As a matter of substantive constitutional law the state-action requirement reflects judicial recognition of the fact that ‘most rights secured by the Constitution are protected only against infringement by governments.'” Lugar v. Edmondson Oil Co., Inc., 457 U.S. 922, 936 (1982) (citation omitted). “Careful adherence to the ‘state action' requirement preserves an area of individual freedom by limiting the reach of federal law and federal judicial power.” Id.

Plaintiffs say that the question of whether they have a First Amendment claim on the basis of the state action doctrine is a factual matter “ill-suited to a Rule 12(b)(6) motion.” Dkt. No. 145 at 7. Not so. It is certainly true that the ultimate determination of state action is a “necessarily fact-bound inquiry, ” Lugar, 457 U.S. at 939, but that does not relieve plaintiffs of their obligation under Rule 8 and Rule 12(b)(6) to provide in the complaint enough facts to plausibly allege a claim against Twitter on the basis of state action. See, e.g., Heineke v. Santa Clara Univ., 965 F.3d 1009, 1015 n.5 (9th Cir. 2020) (“Heineke's contention that it is inappropriate to dismiss his § 1983 constitutional claims at the motion to dismiss stage, is unpersuasive. We have accepted his allegations as true. Because he has failed to plead any allegations sufficient to support his argument that SCU acted under color of state law, however, his § 1983 claims must fail as a matter of law.”). To conclude otherwise, as plaintiffs urge, would fly in the face of the pleading requirements squarely stated in Bell Atlantic Corp. v. Twombly, 550 U.S. 544 (2007), and Ashcroft v. Iqbal, 556 U.S. 662 (2009).

Plaintiffs make the odd assertion that these pleading standards apply only in antitrust conspiracy actions. Dkt. No. 145 at 6 n.7. Twombly and Iqbal expressed no such limitation, and their standards have been applied to a myriad of Rule 12(b)(6) motions in non-antitrust actions in every federal district and circuit court. A scant minute of online research makes this abundantly clear. See, e.g., Mendoza v. Amalgamated Transit Union Int'l, 30 F.4th 879, 886 n.1 (9th Cir. 2022) (labor and employment case); Hoffman v. Preston, 26 F.4th 1059, 1063 (9th Cir. 2022) (Bivens claims).

The salient question under the state action doctrine is whether “the conduct allegedly causing the deprivation of a federal right” is “fairly attributable to the State.” Id. at 937; see also Belgau v. Inslee, 975 F.4th 940, 946 (9th Cir. 2020) (“The state action inquiry boils down to this: is the challenged conduct that caused the alleged constitutional deprivation ‘fairly attributable' to the state?”). The answer is determined by a “two-part approach, ” which requires that “the deprivation must be caused by the exercise of some right or privilege created by the State or by a rule of conduct imposed by the state or by a person for whom the State is responsible”; and that “the party charged with the deprivation must be a person who may fairly be said to be a state actor.” Lugar, 457 U.S. at 937; see also Sutton v. Providence St. Joseph Medical Center, 192 F.3d 826, 835 (9th Cir. 1999). These factors “are not the same, ” and they “diverge when the constitutional claim is directed . . . against a private party.” Lugar, 457 U.S. at 937.

As the parties noted, different formulations of the factors appear in the case law. See, e.g., Manhattan Cmty. Access, 139 S.Ct. at 1928; Lugar, 457 U.S. at 939; Dkt. No. 145 at 7-21; Dkt. No. 147 at 1-6. But “[w]hether these different tests are actually different in operation or simply different ways of characterizing the necessarily fact-bound inquiry, ” Lugar, 457 U.S. at 939, is a question that the Court need not resolve for present purposes. That is because there is “no specific formula for defining state action.” Sutton, 192 F.3d at 836 (quotations and citation omitted). What matters is whether plaintiffs have plausibly alleged facts to “show that there is a sufficiently close nexus between the State and the challenged action of” the private defendants, such that “the action of the latter may be fairly treated as that of the State itself.” Id. (quotations and citations omitted). The specific question the Court must answer here is: have plaintiffs plausibly alleged that Twitter was behaving as a state actor pursuant to “a governmental policy” when it closed their accounts? Id. at 835.

This inquiry “must be determined based on the circumstances of each case, ” id. at 836, and the facts alleged in the amended complaint are not nearly enough for plaintiffs to proceed on a state action theory. To start, the amended complaint does not plausibly show that plaintiffs' ostensible First Amendment injury was caused by “a rule of conduct imposed by the government.” id. at 835 (cleaned up); see also Mathis v. Pacific Gas and Elec. Co., 891 F.2d 1429, 1432 (9th Cir. 1989) (“no state, or federal, action unless” a private entity's decision is “made on the basis of some rule of decision for which the State is responsible.”) (quotations and citation omitted). The amended complaint merely offers a grab-bag of allegations to the effect that some Democratic members of Congress wanted Mr. Trump, and “the views he espoused, ” to be banned from Twitter because such “content and views” were “contrary to those legislators' preferred points of view.” See, e.g., AC ¶¶ 53, 55, 60, 61. But the comments of a handful of elected officials are a far cry from a “rule of decision for which the State is responsible.” Legislators are perfectly free to express opinions without being deemed the official voice of “the State.” Government in our republic of elected representatives would be impossible otherwise. It is also not plausible to conclude that Twitter or any other listener could discern a clear state rule in such remarks, or even determine what a legislator's “preferred views” might be.

The weakness of the state action theory in the amended complaint is further demonstrated by plaintiffs' own explanation of why their accounts were closed. Twitter is said to have closed Mr. Trump's account because of “the risk of further incitement of violence” and “threats to physical safety.” Id. ¶¶ 114-15. Twitter closed plaintiff Cuadros's account “due to a post about vaccines, ” id. ¶ 124, and Dr. Wolf's account for “vaccine misinformation, ” id. ¶ 162. Plaintiff Barboza's account was closed “after retweeting President Trump and other conservatives on January 6, 2021, ” id. ¶ 137; plaintiff Latella after he “post[ed] positive messages about Republican candidates and President Trump, ” id. ¶ 142; and plaintiff Root for “messages he posted related to COVID-19 and the 2020 election results, ” id. ¶ 152.

If anything, these explanations indicate that Twitter acted in response to factors specific to each account, and not pursuant to a state rule of decision. These circumstances are not at all comparable to those in Bantam Books, Inc. v. Sullivan, 372 U.S. 58 (1963), as plaintiffs urge. In that case, which is discussed infra in more detail, a state commission was empowered to compel a private book distributor from selling or supplying certain books. The amended complaint does not allege anything like this type of state dictate to Twitter.

The amended complaint also does not plausibly allege that Twitter could fairly be deemed to be a state actor. Plaintiffs say they have done so by cataloguing “coercive statements” in paragraph 55 of the amended complaint, and statements made during a March 22, 2021, House Committee on Energy and Commerce hearing on the topic of “Disinformation Nation: Social Media's Role in Promoting Extremism and Misinformation.” See Dkt. No. 145 at 3-4 (quoting from AC ¶ 55, and Dkt. No. 145-1 (RJN) ¶¶ 2-4). These statements are said to have compelled Twitter to act as a government entity.

The proffered statements appear in part only in a request for judicial notice, Dkt. No. 145-1, and not in the amended complaint. Twitter did not object to the judicial notice request, and the Court has taken those materials into account.

They are again not enough for pleading purposes. Paragraph 55 is said to offer “examples of Democrat legislators threatening new regulations, antitrust breakup, and removal of Section 230 immunity for Defendants and other social media platforms if Twitter did not censor views and content with which these Members of Congress disagreed.” AC ¶ 55. The actual quotes do not live up to that billing. The statements attributed to “Bruce Reed, Biden's Top Tech Advisor, ” and Michelle Obama are of no moment because Reed and Obama were not legislators. Id. at 5th and 12th bullet points. Other statements in Paragraph 55 pertain only to Facebook, and not Twitter. Id. at 8th, 9th, and 15th bullet points (Senator Markey's question and Mark Zuckerberg's answer regarding Facebook's algorithms and policies; Rep. Adam Schiff's Tweet that “Facebook must ban” Trump). Then-Senator Kamala Harris is quoted three times for calling for “Trump's Twitter account [to be] suspended” and calling on Dorsey to “do something about this Tweet” from Trump, but conspicuously missing is any threatening remark directed to Twitter. Id. at 1st, 6th, and 7th bullet points. Five statements are nothing more than general comments about Section 230 (e.g., “We can and should have a conversation about Section 230”) untethered to any substance that might have conveyed any threat or punishment tied to any specific action by Twitter. Id. at 2nd, 3rd, 4th, 11th, and 13th bullet points. The remaining two statements, at the 10th and 14th bullet points, express general concerns and criticisms that Twitter has become a “terrifying tool[] of persuasion and manipulation, ” and that “[i]ndustry self-regulation has failed.”

The statements attributed to the “Disinformation Nation” congressional hearing may have been more heated, but they are still not enough to satisfy plaintiffs' pleading obligation. Committee Chairman Frank Pallone, Jr., is quoted as saying, “it is time for Congress and this Committee to legislate and realign these companies' incentives to effectively deal with disinformation and extremism....The time for self-regulation is over. It is time we legislate to hold you accountable.” Dkt. No. 145 at 4. Representative Mike Doyle said, “Your companies need to be held accountable . . . and we will legislate to stop this.” Id. Representative Janice D. Schakowsky said, “What our witnesses need to take away from this hearing is that self-regulation has come to the end of its road, and that this democratically elected body is prepared to move forward with legislation and regulation. Misinformation regarding the election dropped by 73% across social media platforms after Twitter permanently suspended Trump ....The question is, what took so long?” Id.

Even giving plaintiffs every benefit of the doubt, these comments fall short of the mark. Plaintiffs' own case citations show why. See Dkt. No. 145 at 7-17. Strictly speaking, not all of plaintiffs' cases involve the state action doctrine, as the ensuing discussion makes clear. Nevertheless, plaintiffs argued the cases to that end, and the Court will take those arguments on their own terms.

In one of plaintiffs' main citations, Bantam Books, 372 U.S. 58, the Rhode Island Legislature established a “Rhode Island Commission to Encourage Morality in Youth” that was composed of members appointed by the state Governor. Id. at 59-60 & n.1. The Commission was empowered “to educate the public concerning any book . . . manifestly tending to the corruption of the youth, ” and “to investigate and recommend the prosecution” of all violations of the relevant laws. Id. at 59-60. Put more simply, it was tasked with formulating and enforcing a blacklist of proscribed books. Id. at 68. The Commission sent letters “on official Commission stationery” to a book distributor listing “certain designated books or magazines distributed by him” that had been declared by the Commission as “objectionable for sale, distribution or display” to minors. Id. at 61. The notice thanked the distributor in advance “for his ‘cooperation' with the Commission, usually reminding [him] of the Commission's duty to recommend to the Attorney General prosecution of purveyors of obscenity.” Id. at 62. “Copies of the lists of ‘objectionable' publications were circulated to the local police departments, and [the distributor] was so informed in the notices.” Id. at 62-63. “A local police officer usually visited [the distributor] shortly after [his] receipt of a notice to learn what action he had taken.” Id. at 63.

In a lawsuit brought by the publishers of books supplied to the distributor, the Supreme Court had no trouble concluding that the acts of the Commission “were performed under color of state law, ” and that the distributor's “compliance with the Commission's directives was not voluntary.” Id. at 68. As the Court stated, “[p]eople do not lightly disregard public officers' thinly veiled threats to institute criminal proceedings against them if they do not come around, and [the distributor's] reaction . . . was no exception to this general rule.” Id. “The Commission's notices, phrased virtually as orders, reasonably understood to be such by the distributor, invariably followed up by police visitations, in fact stopped the circulation of the listed publications ex proprio vigore.” Id. The Court held that the “procedures of the Commission” were constitutionally deficient, and that “the system of informal censorship disclosed by this record violates the Fourteenth Amendment.” Id. at 71.

In Lombard v. State of Louisiana, 373 U.S. 267 (1963), the Supreme Court reversed the state criminal mischief convictions of three black and one white college students who had participated in a “sit-in demonstration” at a restaurant in New Orleans to promote racial desegregation. Although it was a private restaurant that had refused service and called the police, the Court held that “these convictions, commanded as they were by the voice of the State directing segregated service at the restaurant, cannot stand.” Id. at 274. Prior to the demonstration, the Superintendent of Police had stated that “such actions are not in the community interest, ” and “we want everyone to fully understand that the police department and its personnel is ready and able to enforce the laws of the city of New Orleans and the state of Louisiana.” Id. at 270. In addition, “four days before petitioners' arrest, the Mayor of New Orleans issued an unequivocal statement condemning such conduct and demanding its cessation.” Id. at 271. The Mayor said that “I have today directed the superintendent of police that no additional sit-in demonstrations will be permitted regardless of the avowed purpose or intent of the participants, ” and “[i]t is my determination that the community interest, the public safety, and the economic welfare of this city require that such demonstrations cease and that henceforth they be prohibited by the police department.” Id. Both statements were well publicized, and there was evidence in the case “to indicate that the restaurant manager asked petitioners to leave in obedience to the directive of the city officials.” Id. at 272.

In Carlin Communications, Inc. v. The Mountain States Telephone and Telegraph Company, 827 F.2d 1291 (9th Cir. 1987), the defendant, a private regional telephone company, “refuse[d] to carry smut on its dial-a-message network.” Id. at 1292. This happened after a “deputy attorney of Maricopa County, Arizona, wrote to Mountain Bell threatening to prosecute if the company continued to provide [dial-a-message network service] to Carlin. The letter stated that Carlin's 976 service violated an Arizona statute prohibiting the distribution of sexually explicit material to minors.” Id. at 1293. “Mountain Bell immediately sent Carlin a notice that its service would be terminated in five days.” Id. The Ninth Circuit concluded that the termination decision amounted to state action because the “county attorney's threat of prosecution provided the requisite ‘nexus' between the state and the challenged action.” Id. at 1295. In effect, “Arizona ‘exercised coercive power' over Mountain Bell and thereby converted its otherwise private conduct into state action.” Id.

In Mathis v. Pacific Gas and Electric Company, 891 F.2d 1429 (9th Cir. 1989), defendant PG&E had barred plaintiff Mathis from entering the Diablo Canyon Nuclear Power Plant for work because “he was suspected of illegal drug use or sales.” Id. at 1430. Mathis argued that this was state action because “his denial of access by PG&E was directed or encouraged by the [Nuclear Regulatory Commission (NRC)] pursuant to its Fitness for Duty program.” Id. at 1433. At the time, the Fitness for Duty program was only a proposed rule, and the NRC eventually withdrew it “to encourage the initiatives concerning fitness for duty being taken by the nuclear power industry, ” so long as “the industry programs produce the desired results.” Id. at 1433. The NRC added that “[i]t is Commission policy that the sale, use, or possession of alcoholic beverages or illegal drugs within protected areas at nuclear plant sites is unacceptable, ” and that the “decision to use discretion in enforcement to recognize industry initiatives in no way changes the NRC's ability to issue orders, call enforcement meetings, or suspend licenses should a significant safety problem be found.” Id. The NRC also stated that “[a]n acceptable fitness for duty program should at a minimum include the following essential elements: (1) A provision that the sale, use, or possession of illegal drugs within the protected area will result in immediate revocation of access to vital areas and discharge from nuclear power plant activities . . .; [and] (2) A provision that any other sale, possession, or use of illegal drugs will result in immediate revocation of access to vital areas, mandatory rehabilitation prior to reinstatement of access, and possible discharge from nuclear power plant activities.” Id. In these circumstances, the circuit court concluded that Mathis had plausibly made out a state action claim because “[t]he minimum standard [wa]s stated by the NRC, as a minimum acceptable industry plan, backed up by threats of enforcement or of formal rulemaking.” Id. at 1434.

These cases, which are the centerpieces of plaintiffs' state action argument, are strikingly different from the allegations in the amended complaint. In each of the cases, a concrete and specific government action, or threatened action, was identified. Here, plaintiffs offer only ambiguous and open-ended statements to the effect that “we may legislate” something unfavorable to Twitter or the social media sector. This is a world away from: (1) a state commission sending local police officers for drop-in visits and threatening prosecution by the state attorney general (Bantam Books); (2) a city mayor and police superintendent threatening law enforcement action to crack down on sit-in demonstrations (Lombard); (3) a deputy county attorney threatening prosecution against a private company under a specific law (Carlin); and (4) a federal administrative commission threatening the suspension of licenses or formal rulemaking if its specified elements for an anti-drug program were not followed voluntarily (Mathis).

The fact that enacting a bill is rarely fast or easy further attenuates the plausibility of the legislative threat plaintiffs speak of. As the process was described in another context, “[d]issatisfaction . . . is often the cost of legislative compromise. And negotiations surrounding enactment of this bill tell a typical story of legislative battle among interest groups, Congress, and the President. Indeed, this legislation failed to ease tensions among many of the interested parties. Its delicate crafting reflected a compromise amidst highly interested parties attempting to pull the provisions in different directions.... The deals brokered during a Committee markup, on the floor of the two Houses, during a joint House and Senate Conference, or in negotiations with the President are not for us to judge or second-guess.” Barnhart v. Sigmon Coal Co., Inc., 534 U.S. 438, 461 (2002) (internal citations omitted). There is no way to allege with any degree of plausibility when, if ever, the comments voiced by a handful of members of Congress might become a law, or what changes such a law might impose on social media companies like Twitter.

Plaintiffs also overlook Congress's role as an investigatory body, and the fact that “each House has power ‘to secure needed information' in order to legislate.” Trump v. Mazars USA, LLP, 140 S.Ct. 2019, 2031 (2020) (citation omitted). This power “encompasses inquiries into the administration of existing laws, studies of proposed laws, and ‘surveys of defects in our social, economic or political system for the purpose of enabling the Congress to remedy them.'” Id. (quotations and citation omitted).

Much of what plaintiffs challenge fits within the normal boundaries of a congressional investigation, as opposed to threats of punitive state action. Plaintiffs' own submissions indicate that the House Committee was making inquiries and surveying possible problems “for the purpose of enabling the Congress to remedy them.” In this respect, the allegations in the amended complaint are much more comparable to the cases plaintiffs cited in which no state action was found. See Hammerhead Enterprises, Inc. v. Brezenoff, 707 F.2d 33 (2d Cir. 1983) (no state action where Administrator of the Human Resources Administration of New York City wrote letters “urg[ing]” department stores, as a gesture of “public service, ” not to carry a board game the Administrator found problematic; state action absent where “comments of a government official” cannot “reasonably be interpreted as intimating that some form of punishment or adverse regulatory action will follow the failure to accede to the official's request”); see also American Family Ass'n, Inc. v. City and County of San Francisco, 177 F.3d 1114 (9th Cir. 2002) (no First Amendment violation found where City and County of San Francisco passed resolution “urg[ing] ‘local television stations not to broadcast advertising campaigns aimed at “converting” homosexuals'”; holding that “public officials may criticize practices that they would have no constitutional ability to regulate, so long as there is no actual or threatened imposition of government power or sanction”).

Overall, the amended complaint does not plausibly allege that Twitter acted as a government entity when it closed plaintiffs' accounts. This resolves the main thrust of plaintiffs' state action theory. Plaintiffs' cursory mention of state “encouragement” (through Section 230(c)) and “joint action” are minor variations on the state action theme, and are unavailing for the same reasons. Dkt. No. 145 at 17-21. In addition, “compliance with generally applicable laws” is not “sufficient to convert private conduct into state action.” Heineke, 965 F.3d at 1013. The government cannot plausibly be said to have compelled Twitter's action through Section 230, which in any event imposed no affirmative obligations on Twitter to act in any particular way.

Consequently, the amended complaint does not plausibly allege a First Amendment claim against Twitter. Plaintiffs' first claim is dismissed.


Plaintiffs' claim for a declaratory judgment that Section 230 is unconstitutional is dismissed for lack of standing. As Article III of the United States Constitution states, federal courts have the “power to decide legal questions only in the presence of an actual ‘Cas[e]' or ‘Controvers[y].'” Wittman v. Personhuballah, 578 U.S. 539, 543 (2016). The party invoking a federal court's jurisdiction must demonstrate standing by showing that she has “suffered an ‘injury in fact,' that the injury is ‘fairly traceable' to the conduct being challenged, and that the injury will likely be ‘redressed' by a favorable decision.” Id. (quoting Lujan v. Defenders of Wildlife, 504 U.S. 555, 560-61 (1992)). Standing is an ongoing inquiry, and “[t]he need to satisfy these three [Article III standing] requirements persists throughout the life of the lawsuit.” Id. (citation omitted). The Court has an independent duty to be vigilant about standing.

To establish an injury in fact, a plaintiff must show that he or she suffered “an invasion of a legally protected interest which is (a) concrete and particularized, and (b) actual or imminent, not conjectural or hypothetical.” Lujan, 504 U.S. at 560 (cleaned up). These facts are entirely absent from the amended complaint with respect to Section 230. Plaintiffs offer only the vague and speculative allegation that “[u]pon information and belief, defendants would not have de-platformed the plaintiff or similarly situated putative class members but for the immunity purportedly offered by Section 230(c).” AC ¶ 190. Why this might be plausible is left unsaid. The Court declines to accept such speculative and conclusory allegations as grounds for a declaratory judgment claim. See In re Gilead Sciences Sec. Litig., 536 F.3d 1049, 1055 (9th Cir. 2008).


Twitter says that plaintiffs' third claim under the Florida Deceptive and Unfair Trade Practices Act should be dismissed because plaintiffs have agreed, pursuant to the Twitter Terms of Service (TOS), that California law will govern all disputes that arise between Twitter and its users. Dkt. No. 138 at 14. Plaintiffs do not dispute that the TOS is a valid contract between the parties, or that it includes an express choice of California law. Dkt. No. 145 at 22-23.

Neither side identified the choice-of-law rules that should apply, but the basic rule is that, “[i]n determining what state law to apply, a federal court applies the choice-of-law rules of the state in which it sits.” Unified Western Grocers, Inc. v. Twin City Fire Ins. Co., 457 F.3d 1106, 1111 (9th Cir. 2006) (quotations and citation omitted). An exception may arise for federal courts sitting in diversity when a case has been transferred from another jurisdiction, as is the situation here. Atlantic Marine Const. Co., Inc. v. U.S. Dist. Ct. for W. Dist. of Tex., 571 U.S. 49, 65 (2013). In that case, the Court will apply the choice-of-law rules of the transferor court, unless the transfer was made on the basis of a valid forum-selection clause. Id.; see also In re Facebook Biometric Information Privacy Litig., 185 F.Supp.3d 1155, 1168 (N.D. Cal. 2016). The TOS specifies venue in this District, and the case was transferred here from the Southern District of Florida for that contractual reason. See Dkt. No. 87. Consequently, the Court will apply the choice-of-law rules of its home state, California.

Under California law, to determine whether a choice-of-law clause is enforceable, courts consider “(1) whether the chosen state has a substantial relationship to the parties or their transaction, or (2) whether there is any other reasonable basis for the parties' choice of law.” Washington Mutual Bank, FA v. Superior Court, 24 Cal.4th 906, 916 (2001). If either test is met, the parties' choice of law “generally will be enforced unless the other side can establish that the chosen law is contrary to a fundamental policy of the state law alternative to the contractual choice, and that the other state has a materially greater interest in the determination of the matter.” In re Facebook, 185 F.Supp.3d at 1168 (internal quotations omitted).

There is no reasonable doubt that California has a substantial relationship to this case. Twitter is a leading social media company with hundreds of millions of users, and has its principal place of business in California. Plaintiffs do not dispute this point. They contend only that Florida law should be applied on the basis of a “fundamental policy” conflict with California law. Dkt. No. 145 at 22-23.

This is plaintiffs' burden to establish, In re Facebook, 185 F.Supp.3d at 1168, and they have not succeeded. Plaintiffs contend, in the space of less than half a page, that a fundamental conflict exists because the FDUPTA allows claims for injunctive relief without financial injury, and the UCL supposedly does not. Dkt. No. 145 at 23. The point is not well taken. To start, plaintiffs substantially undercut their position by saying that they have, in effect, already pleaded “a UCL claim for injunctive relief” on the same facts alleged for the FDUPTA claim. Dkt. No. 145 at 23 n.18. That goes a long way to neutralizing the suggestion that the two state statutes are fundamentally at odds. In addition, plaintiffs did not account for the California Supreme Court's conclusions that:

Section 17203 makes injunctive relief “the primary form of relief available under the UCL, ” while restitution is merely “ancillary.” (In re Tobacco II Cases (2009) 46 Cal.4th 298, 319, 93 Cal.Rptr.3d 559, 207 P.3d 20.) Nothing in the statute's language conditions a court's authority to order injunctive relief on the need in a given case to also order restitution. Accordingly, the right to seek injunctive relief under section 17203 is not dependent on the right to seek restitution; the two are wholly independent remedies. (See ABC Internat. Traders, Inc. v. Matsushita Electric Corp. (1997) 14 Cal.4th 1247, 1268, 61 Cal.Rptr.2d 112, 931 P.2d 290 [§ 17203 “contains ... no language of condition linking injunctive
and restitutionary relief”]; Prata v. Superior Court (2001) 91 Cal.App.4th 1128, 1139, 111 Cal.Rptr.2d 296 [plaintiff could pursue injunctive relief even though restitution was unavailable].)
Clayworth v. Pfizer, Inc., 49 C.4th 758, 790 (2010). Plaintiffs also did not demonstrate that Florida “has a materially greater interest in the determination of the matter.” In re Facebook, 185 F.Supp.3d at 1168. Consequently, plaintiffs have not presented a good reason to disregard the choice of California law in the TOS in favor of the FDUTPA.

Although this is enough to dismiss the third claim, some additional observations are useful. A good argument can be made that plaintiffs did not plausibly allege deceptive conduct by Twitter for purposes of either the FDUTPA or the UCL. The TOS expressly states that Twitter may suspend or terminate an account “at any time for any or no reason.” Dkt. No. 138-7 at 8. It also states that Twitter may remove or refuse to distribute any content. Id. at 5. There is nothing cagey or misleading about these provisions, and plaintiffs' suggestion that Twitter may have applied them inconsistently, see AC ¶¶ 212-16, or at the government's behest, does not change that. The TOS gave Twitter contractual permission to act as it saw fit with respect to any account or content for any or no reason, which makes its ostensible motives irrelevant for a deceptive practices claim.


Plaintiffs' fourth claim under Florida's Stop Social Media Censorship Act is also due for dismissal. Twitter did not make a choice-of-law argument for this claim, see Dkt. No. 138 at 1419, and so the Court addresses the SSMCA claim on its own terms.

An initial problem for plaintiffs is that only one named plaintiff (Dominick Latella) was a Florida resident with any active Twitter account at the time the statute took effect on July 1, 2021, AC ¶¶ 138, 146, and so he is the only plaintiff who might conceivably have a SSMCA claim. See Fla. Stat. § 501.2041(1)(h) (“user” is “a person who resides or is domiciled in [Florida] and who has an account on a social media platform.”). The amended complaint alleges that all of the other plaintiffs were domiciled outside of Florida, or had their Twitter accounts closed prior to July 1, 2021.

Another problem is that plaintiffs say they are challenging only conduct that occurred after the SSMCA effective date. Dkt. No. 145 at 24. But the amended complaint focuses on actions affecting plaintiffs' accounts prior to July 1, 2021. See AC ¶¶ 113, 124, 128-29, 137, 143, 150, 158-59, 226. Consequently, it is unclear what plaintiffs allege to be the potential application of the statute to their case.

There is also a major concern about the enforceability of the SSMCA. Florida government officials were enjoined from enforcing the SSMCA on June 30, 2021, the day before the law was to take effect, in a well-reasoned decision issued by the Northern District of Florida. NetChoice, LLC v. Moody, 546 F.Supp.3d 1082 (N.D. Fla. 2021), appeal pending sub nom, NetChoice LLC v. Attorney Gen., State of Fla., No. 21-12355 (11th Cir.). The court concluded that the statute violated the First Amendment and was preempted by 47 U.S.C. § 230; it also expressed strong concerns that the statute was impermissibly vague. The Court declines plaintiffs' invitation to disregard this decision, particularly while an appeal is pending, and dismisses the SSMCA claim without prejudice.


The amended complaint is dismissed in its entirety. Plaintiffs will have an opportunity to amend their complaint. See Lopez v. Smith, 203 F.3d 1122, 1130 (9th Cir. 2000). Plaintiffs may file an amended complaint that is consistent with this order by May 27, 2022. The amended complaint may not add any new claims or defendants without express prior leave of Court. Plaintiffs are advised that further opportunities to amend are not likely to be granted.


Summaries of

Trump v. Twitter, Inc.

United States District Court, Northern District of California
May 6, 2022
21-cv-08378-JD (N.D. Cal. May. 6, 2022)
Case details for

Trump v. Twitter, Inc.

Case Details

Full title:DONALD J. TRUMP, et al., Plaintiffs, v. TWITTER INC., et al., Defendants.

Court:United States District Court, Northern District of California

Date published: May 6, 2022


21-cv-08378-JD (N.D. Cal. May. 6, 2022)

Citing Cases

Grocers v. Vilsack

“Standing is an ongoing inquiry, and ‘[t]he need to satisfy these three [Article III standing] requirements…