Government Jawboning Doesn’t Turn Internet Services into State Actors–Doe v. Google

October 21, 2021 · by · in Content Regulation, Derivative Liability

The plaintiffs are “conservative content creators” (i.e., QAnon enthusiasts) who posted videos to YouTube. YouTube suspended their accounts. The plaintiffs sued for First Amendment violations (presumably a 1983 claim). The court previously denied a TRO. YouTube now gets the case dismissed with prejudice. It’s not a close call.

To show YouTube is a state actor, the plaintiffs tried four different unsuccessful arguments:

Public Function. This argument is foreclosed by Prager U. v. Google.

Compulsion. The court says:

The Court finds that the statements by federal lawmakers Plaintiffs point to are insufficient to plead that the government “commanded a particular result in, or otherwise participated in, [Plaintiffs’] specific case.” [cite to Daniels v. Alphabet] Plaintiffs point to generalized statements from lawmakers pertaining to “coronavirus-related misinformation,” “disinformation proliferating online,” “QAnon-related speech,” and “conspiracy theories.” None of the statements mention Plaintiffs’ names, their YouTube or Google accounts, their channels, or their videos. Plaintiffs argue that state actors “commanded a particular result” in their case because “Plaintiffs have alleged that Congress demanded that the unpopular speech dubbed ‘misinformation,’ and QAnon-related speech be limited and erased, which is precisely what Plaintiffs allege Defendants did.” The Court disagrees that broad lawmaker proclamations regarding “misinformation” or “QAnon-related speech,” for example, are sufficient to show that the government “commanded” the suspension of Plaintiffs’ accounts. Even if Defendants had complied with these lawmaker statements to the letter, they would still have had the ultimate discretion on what videos or accounts fit into buckets like “misinformation” or “QAnon-related speech.”

Plaintiffs claim that Defendants’ conduct is state action because it was in response to the threat of various government penalties—the repeal of CDA Section 230 protections, “show trials” in front of the U.S. Senate, and a DOJ antitrust suit against Google—allegedly linked to whether Defendants appropriately moderated certain types of content. The threats of penalties Plaintiffs point to are insufficient to convert private conduct into state action here. The Ninth Circuit has found that pleading “a private actor’s conduct is subject to penalties…is…insufficient to convert private action into that of the state.” Moreover, Plaintiffs fail to point to any penalties that necessarily or even likely would have followed if Defendants did not suspend their accounts….

Plaintiffs can point to no authority to support a compulsion theory of state action based on penalties, particularly “threats” as speculative as the ones they point to here.

To be clear, it’s offensive when our elected officials treat Section 230 like a political football, threatening to withdraw the immunity if services don’t bend to their censorial whims. When politicians do this, they are really saying they don’t care about good social policy that benefits their constituents. It’s all fun-and-games until an immunity gets broken. But the plaintiffs’ arguments–that the politicians’ jawboning make all subsequent content moderation decisions into state action–are even worse from a policy outcome. As they would say on Reddit, ESH.

Joint Action. The court says:

the Schiff-Wojnicki Twitter exchange Plaintiffs point to in support of their joint action claim clearly pertains to misinformation regarding COVID-19. As the Court discussed above, Plaintiffs have failed to allege any facts indicating that their posts pertained to COVID-19. Accordingly, it is unclear how this Twitter exchange supports a joint action theory pertaining to the suspension of Plaintiffs’ channels….Further, it is simply implausible to read a casual Twitter exchange between one member of Congress and YouTube’s CEO as joint action. Plaintiffs’ theory would effectively cause companies to cease communicating with their elected representatives for fear of liability, as Defendants compellingly argue.

Chilling companies’ abilities to petition the government is another censorial consequence of the plaintiffs’ arguments.

The court continues:

Based on Plaintiffs’ allegations, their content was removed through the following series of events: federal lawmakers publicly flagged general categories of content for Defendants to consider moderating and issued threats to compel Defendants to comply, Defendants independently chose what content fit into the lawmakers’ general categories, and Plaintiffs’ channels happened to be some of the content Defendants decided to remove. Courts have dismissed cases for lack of state action despite significantly more alleged cooperation between public and private actors compared to what Plaintiffs allege here

…far more is necessary to plead joint action than what they have alleged here…At most, Plaintiffs appear to allege that government officials identified categories of information Defendants should consider removing—there is no allegation that government officials were in the room or somehow directly involved in the decision to suspend Plaintiffs….there are no allegations that Defendants invoked state or federal procedure to bring about the suspension of Plaintiffs’ accounts. Defendants merely suspended Plaintiffs from Defendants own private platform.

Government Nexus. “Plaintiffs have failed to point to a single case in which governmental nexus was found.”

Implications. Unsurprisingly, this case wasn’t close. #MAGA plaintiffs are routinely overclaiming state action by Internet services as a short-sighted ploy to force the services to carry garbage content that harms the services’ audiences. This opinion is a clean and decisive ruling that standard government jawboning doesn’t automatically turn the entire Internet into a giant state actor. I can’t wait for this litigation fad to die out.

The court declined supplemental jurisdiction over the state law claims, which the plaintiffs could choose to refile in state court if they want to lose again. I think it’s more likely the plaintiffs will choose to appeal the case and give the Ninth Circuit a first-hand opportunity to explain why Prager U v. Google means they lose.

In contrast to many cases in this genre, the plaintiffs were represented by counsel. That counsel is Cris Armenta. In addition to being on the losing side of the Garcia v. Google case, she also lost the Daniels case, which this court repeatedly cited against her arguments in this case. 🏆🏆🏆

Case Citation: Doe v. Google LLC, 2021 WL 4864418 (N.D. Cal. Oct. 19, 2021)

Selected Related Posts About State Action Claims