9th Circuit cuts public out of free speech case

The 9th U.S. Circuit Court of Appeals has announced that it will consider a significant free speech issue bearing on the rights of anonymous internet users with sealed briefs, sealed record, and no amici.

The case arises from a federal grand jury subpoena served on San Francisco-based Glassdoor.com, which allows employees to rate employers anonymously; the subpoena demanded identifying information about several critics of a company that is under federal criminal investigation. Glassdoor moved in Arizona federal court to quash the subpoena (docket no. 2:17-mc-00036-DJH), invoking its users' First Amendment right to speak anonymously (McIntyre v. Ohio Elections Commission, 514 U.S. 334, 342 (1995)), which Glassdoor contended created a privilege against production of the information. Glassdoor cited a line of cases, beginning with Dendrite v. Doe (775 A.2d 756 (N.J. App. Div. 2001); "Litigating Civil Subpoenas to Identify Anonymous Internet Speakers," 37 Litigation No. 3 (Spring 2011)), under which the First Amendment right to speak anonymously supports procedural and substantive obstacles to civil subpoenas seeking to identify online speakers so that they can be served with process and sued for wrongful speech. Glassdoor also invoked Bursey v. United States, 466 F.2d 1059 (9th Cir. 1972), where the 9th Circuit quashed a grand jury subpoena directed at members of the Black Panther Party.

The entire subpoena litigation was conducted under seal, but we know some details because the trial court granted Glassdoor's motion to partially unseal the briefs to enable amicus participation on appeal.

The trial judge refused to apply Bursey believing that it only protected against grand jury subpoenas to dissenters opposing the government, and she refused to take seriously the First Amendment rights of Glassdoor's users, albeit for reasons that showed a misunderstanding of the First Amendment rights at issue. On the one hand, she thought that the First Amendment privilege being asserted could extend only to political speech (which is plainly wrong), and on the other hand she seems to have suggested the First Amendment cannot block a grand jury subpoena because newspapers generally have no First Amendment rights to stop grand jury intrusion into their sources absent a showing of bad faith on the part of the government. The judge deemed Glassdoor indistinguishable from a journalistic enterprise.

Public Citizen looked forward to dissecting this analysis from a consumer and employee perspective; other groups were also preparing amicus briefs. Such efforts were cut short when the 9th Circuit denied Glassdoor's motion to unseal the appellate record in part and stated, "the Court sees no need for participation by amicus curiae at this stage of the litigation."

It is hard to see what possible justifications support this order. Given that the briefs below are largely unsealed, no legitimate secrecy interests are preserved by appellate sealing. And without real benefits from secrecy, concealing the arguments that persuade the Court of Appeals to resolve this significant First Amendment case will result in a serious loss of the sort of public accountability that the First Amendment and common law right of access to judicial records is intended to preserve.

Equally distressing is the peremptory decision not to accept amicus curiae submissions. This is not an easy appeal. Glassdoor's analogies to Bursey and Dendrite are imperfect. First Amendment privileges are qualified ones that depend, in part, on how great the need for disclosure may be in any given circumstance. Dendrite and other cases test the extent of the discovering party's need, but grand jury subpoenas carry a presumption of regularity; hence the rules that govern the enforcement of those subpoenas cannot be the same as in the civil context. And unlike Bursey, the government here does not appear to be targeting the Glassdoor reviewers out of antagonism toward their protected speech; if anything, the government is pursuing the company the reviewers criticized. But might a different result be appropriate if, for example, the subpoena were directed to identifying former employees of the Trump Organization who criticized that company on Glassdoor?

On the other hand, the trial court's reliance on Branzburg v. Hayes, 408 U.S. 665 (1972), as the key to the case is plainly wrong: Branzburg is addressed to the First Amendment rights of journalists, which is inapposite here. The 9th Circuit has expressly acknowledged that the First Amendment can support a privilege against identifying anonymous online speakers. In re Anonymous Online Speakers, 661 F.3d 1168 (9th Cir. 2011). And at the same time, the government's contention below that speech in employee reviews enjoys something less than full First Amendment protection, because it is not "political speech," is also plainly wrong: consumer and employee criticism of companies enjoys full First Amendment protection. See Bose Corp. v. Consumers Union, 466 U.S. 485 (1984); Letter Carriers v. Austin, 418 U.S. 264 (1974). See also Edward J. DeBartolo Corp. v. NLRB, 463 U.S. 147 (1983).

With the seal in place, it is hard to know how such considerations might apply here. But the proper disposition of the case might well lie someplace between the positions staked out by the government and Glassdoor. The court should be willing to hear from amici offering alternate approaches.

Moreover, it is not at all clear how the panel could be so sure, without having reviewed argument from potential amici, that nonparty submissions cannot be useful in deciding the case and, equally important, in explaining the decision. Even assuming that the panel believes, based on the sealed Glassdoor appellate brief, that they already know the proper disposition of this case, one of the most important roles that amicus briefs play is in helping judges look around the corner at other related cases and, possibly, avoid needlessly prejudicing the outcome of future cases with strong dictum or alternative holdings.

Too many opinions by 7th Circuit judge Richard Posner exude the sentiment that he is too smart to need the thoughts of amici to inform his judgment. It would be disappointing to see that attitude spread to the 9th Circuit.