META PLATFORMS, INC., APPELLANT, v. DISTRICT OF COLUMBIA, APPELLEE.
No. 22-CV-0239
DISTRICT OF COLUMBIA COURT OF APPEALS
September 14, 2023
Argued January 31, 2023
Notice: This opinion is subject to formal revision before publication in the Atlantic and Maryland Reporters. Users are requested to notify the Clerk of the Court of any formal errors so that corrections may be made before the bound volumes go to press.
(Hon. Anthony C. Epstein, Trial Judge)
Ashwin P. Phatak, Principal Deputy Solicitor General, with whom Karl A. Racine, Attorney General for the District of Columbia at the time, Caroline S. Van Zile, Solicitor General, and Stacy L. Anderson, Senior Assistant Attorney General, were on the brief, for appellee.
Before BLACKBURNE-RIGSBY, Chief Judge, DEAHL, Associate Judge, and STEADMAN, Senior Judge.
Opinion for the court by Associate Judge DEAHL.
Concurring opinion by Associate Judge DEAHL at page 42.
DEAHL, Associate Judge: The District has subpoenaed Meta Platforms, the operator of the social media site Facebook, for documents related to Meta‘s enforcement of its COVID-19 misinformation policies. The District is investigating potential violations of the Consumer Protection Procedures Act, or CPPA,
Meta raises two arguments in support of its view that the District‘s subpoena is unenforceable. Its first argument concerns the Stored Communications Act, or SCA,
I.
Superior Court Proceedings
This case arises from an ongoing investigation by Attorney General for the District of Columbia into Meta‘s content moderation practices. Throughout the COVID-19 pandemic, Meta made various public statements about its efforts to police the spread of misinformation on its platform. In December 2020, for example, the company announced that it would be “remov[ing] false claims that COVID-19 vaccines contain microchips, or anything else that isn‘t on the official vaccine ingredient list.” Several months later, Meta unveiled an expansion of this policy, noting “a particular focus on pages, groups, and accounts that violate these rules.” By August 2021, Meta reported that these efforts had led to the removal of 20 million items of content and over 3,000 accounts, pages, and groups for repeat violations.
The District, perceiving a mismatch between these public statements and the widespread dissemination of vaccine misinformation on Facebook, is investigating Meta‘s potential violations of the CPPA. That statute, which prohibits unfair and deceptive trade practices, authorizes the District to conduct “investigation[s] to determine whether to seek relief under” its provisions, including by issuing subpoenas to “compel production of records, books, papers, contracts, and other documents.”
Documents sufficient to identify all Facebook groups, pages, and accounts that
have violated Facebook‘s COVID-19 misinformation policy with respect to content concerning vaccines, including the identi[t]y of any individuals or entities associated with the groups, pages, and accounts; the nature of the violation(s); and the consequences imposed by Facebook for the violation, including whether content was removed or banned from these sources.
This demand was eventually narrowed to only those documents related to public posts, or posts that were so widely accessible as to be functionally public.1
Meta refused to comply with the subpoena, and so the District brought an enforcement action in Superior Court. In that litigation, Meta principally argued that the government may compel the production of electronic communications only by procuring a warrant, citing to a provision of the SCA,
District is targeting only public posts, the SCA‘s “consent exception,”
Meta now appeals, pressing the same two arguments that it raised before the trial court. First, it argues that the SCA precludes the government from compelling disclosure of the targeted documents via subpoena, as the SCA requires it to instead procure a warrant. Second, it argues that the subpoena violates its and its users’ First Amendment rights of free speech and association. We address Meta‘s statutory argument concerning the proper interpretation of the SCA first, and then turn to its First Amendment challenges.
II.
The proper interpretation of the SCA is a question of law we review de novo. Facebook, Inc. v. Wint, 199 A.3d 625, 628 (D.C. 2019).
A. Background of the Stored Communications Act
Congress passed the SCA in 1986 to fill a perceived hole that technological advances had poked in the Fourth Amendment‘s protections of private communications and records. For most of our country‘s history, people typically kept their private communications and records in their homes or places of business, and the government generally needed a warrant supported by probable cause to seize those materials. See Coolidge v. New Hampshire, 403 U.S. 443, 455 (1971) (the Fourth Amendment‘s warrant requirement is “subject only to a few specifically established and well delineated exceptions“).
The SCA sought to fill that potential gap by providing “a set of Fourth Amendment-like privacy protections by statute,” limiting “the ability of [service providers] to voluntarily disclose information about their customers and subscribers to the government.” Orin S. Kerr, A User‘s Guide to the Stored Communications Act, and a Legislator‘s Guide to Amending It, 72 Geo. Wash. L. Rev. 1208, 1212-13 (2004). Two of the SCA‘s provisions are particularly crucial to this appeal.
First is
Second is
content records,
B. Wint, Pepe, and the Parties’ Competing Readings of the SCA
We have interpreted these provisions twice before, and both cases are important here. We first addressed them in Facebook v. Wint, where we held that
We next addressed these provisions in Facebook v. Pepe, where unlike Wint, statutory exceptions did apply to permit the service provider to disclose the subpoenaed communications. 241 A.3d 248, 256 (D.C. 2020). Nonetheless, Facebook opted not to comply with the criminal defendant‘s subpoena in that case, highlighting statutory language providing only that it “may divulge” electronic communications when such an exception applies, rather than requiring it to do so. Id. at 257-58 (emphasis added). We disagreed and held that Facebook was required to comply with the subpoena where nothing in the SCA precluded it from doing so. Id. at 258. In short, because “the SCA did not authorize Facebook‘s refusal to comply with Mr. Pepe‘s subpoena,” Facebook was subject to “disclosure requirements imposed by other law.” Id. at 258.
The present case, like Pepe, involves communications that are exempted from the SCA‘s broad prohibition on disclosure. Meta does not dispute that the SCA permits it to comply with the District‘s subpoena, because the District seeks only publicly posted messages, which the parties agree fit within the SCA‘s consent exception to overcome the general bar on disclosure.
The parties offer competing theories about how §§ 2702 and 2703 should be read together and applied in this case. The District contends that Pepe‘s reasoning applies with full force here, and the same result—that Meta must comply with the subpoena—follows. Recall that Pepe held that so long as some § 2702(b) exception to
comply with the subpoena. The Superior Court adopted essentially this reasoning, positing that “[n]othing in the text of § 2702(b)(3) limits the consent exception to disclosure to non-governmental entities.”
Meta counters that Pepe is inapplicable because the SCA applies an entirely different set of restrictions when it is the government, rather than a private party, seeking to compel disclosures. Meta contends that Pepe and the § 2702 exceptions are inapposite in light of § 2703‘s directive that “[a] governmental entity may require the disclosure ... of the contents of a wire or electronic communication ... only pursuant to a warrant.”
We agree with the District‘s reading of the statute. The text and structure of the SCA support the District‘s interpretation, as we explain in Part II.C. The Act‘s legislative history also supports that interpretation, as we explain in Part II.D.
C. The SCA‘s Text and Structure
We begin with the text of the statute, which is “generally the best indication of the legislative intent.” In re B.B.P., 753 A.2d 1019, 1021 (D.C. 2000). As an initial matter, the District argues that the SCA does not apply to public posts at all, but instead applies only “to protect information that the communicator took steps to keep private.” Ehling v. Monmouth-Ocean Hosp. Serv. Corp., 961 F. Supp. 2d 659, 668 (D.N.J. 2013). It might be right about that.5 See infra at 44-47 (Deahl, J., concurring). But see Facebook, Inc. v. Superior Court (Hunter), 417 P.3d 725, 743-44 (Cal. 2018) (concluding that the SCA “initially prohibits
consent exception allows providers to disclose communications configured by the user to be public“). We ultimately bypass that question, though, because even assuming, as the trial court concluded, that the SCA applies to the public posts at issue here, we agree that the District‘s subpoena is enforceable.
As the trial court recognized, our holding in Pepe applies with equal force when it is the government, rather than a private party, seeking to compel disclosure of communications that fall within one of the § 2702(b) exceptions. Nothing in the text of
Meta counters that Congress enshrined an entirely different set of rules for government actors seeking to compel disclosure in
highlights the provision‘s directive that “[a] governmental entity may require the disclosure” of electronic communications “only pursuant to a warrant.”
But the far bigger textual problem with Meta‘s interpretation of
endeavor.” Grayson v. AT&T Corp., 15 A.3d 219, 238 (D.C. 2011) (en banc) (citation omitted). When viewed in its broader statutory context, it becomes clear that
the same external legal processes that private parties can avail themselves of when a § 2702(b) exception applies. Through
And there are further textual indications that the SCA grants the government a greater ability to compel protected communications than the average Joe, rather than less as Meta would have it. For instance, several of the § 2702(b) exceptions permit the disclosure of otherwise protected communications only to government entities. Service providers must disclose to the National Center for Missing and Exploited Children any communications that they become aware of which indicate a violation of various laws against child pornography.
“to a governmental entity” communications that trigger a good faith belief “that an emergency involving danger of death or serious physical injury” is afoot and requires “disclosure without delay” to avoid hazardous results.
When §§ 2702 and 2703 are read together, their import is clear:
257
(noting the “weighty and well-settled presumption against inferring that Congress silently intended to foreclose or restrict the availability of a core component of the judicial process such as the subpoena power“).Meta protests that
Meta next counters that our reading of the SCA‘s text would make us “the sole outlier” among courts to have considered this issue. That is a rhetorical sleight of hand. Meta points to just two decisions from trial courts that it suggests support its view, and only one of them even arguably does. In truth, Meta‘s position here is so novel that there are simply not any appellate court decisions addressing it, and the trial court decisions that Meta cites give us no pause.
Meta first points to FTC v. Netscape Communications Corp., 196 F.R.D. 559 (N.D. Cal. 2000). That case is inapposite because there was no suggestion in it that a § 2702(b) exception applied to the communication sought to be compelled via agency subpoena. Netscape thus stands for the unremarkable position that where no § 2702(b) exception applies, the government‘s sole recourse for compelling disclosure is to comply with § 2703‘s strictures. We agree with that—we held likewise in Wint, 199 A.3d at 629—but it is not the issue before us.
Meta‘s other authority is closer to the mark, but unpersuasive. See People v. Harris, 949 N.Y.S.2d 590 (N.Y. Crim. Ct. 2012). Harris involved a subpoena issued by a District Attorney‘s Office seeking tweets publicly posted from a Twitter account, allegedly operated by a criminal defendant, over the course of more than 100 days. Id. at 591. The trial court enforced that subpoena as to all but a single
day‘s tweets, because only that day‘s tweets were “less than 180 days old,” and therefore the court concluded they could be compelled only by a search warrant. Id. at 596, 598. The court did not address whether the tweets fell within any
D. The SCA‘s Legislative History
The legislative history supports our reading of the SCA as well. As previously explained, the SCA is roughly meant to extend Fourth Amendment protections to electronic communications and the like. It seeks to neutralize the incident of technology that things like emails are typically disclosed to third-party service providers—thereby calling the Fourth Amendment‘s protections into doubt—for purposes of transmission. There are two features of Meta‘s proposed interpretation of the SCA that do not square with this history: (1) it would extend Fourth Amendment-like protections to public disclosures, which would ordinarily receive slim-to-no Fourth Amendment protections; and (2) the protections it extends would not actually belong to the individual users themselves, but instead would belong to the service providers. We elaborate below on why neither feature aligns with the legislative history, then we respond to Meta‘s counterpoints to it, but before all of that, we detail the legislative history itself.
1. The Legislative History in Broad Strokes
The SCA was enacted as part of the Electronic Communications Privacy Act (ECPA) of 1986, Pub. L. No. 99-508, which predates the World Wide Web by several years. As one might expect, applying the SCA to modern technology is often like cramming a square peg into a round hole. See generally Orin Kerr, The Next Generation Communications Privacy Act, 162 U. Pa. L. Rev. 373, 378, 390-410 (2014) (detailing a variety of reasons “why the [ECPA] is based on outdated assumptions“).
As the Senate Report accompanying the legislation explained: “When the Framers of the Constitution acted to guard against the arbitrary use of Government power to maintain surveillance over citizens, there were limited methods of intrusion into the ‘houses, papers, and effects’ protected by the fourth amendment.” S. Rep. No. 99-541, at 1-2 (1986). Because of technological developments, however, Congress believed that the Constitution‘s protections had become “hopelessly out of date.” Id. at 2. Unlike one‘s physical property, electronic records and communications are frequently in the possession and control of third-party service providers, which arguably renders them “subject to no constitutional privacy protections.” Id. at 3 (citing Miller, 425 U.S. 435). “Thus, the information may be open to possible wrongful use and public disclosure by law enforcement authorities as well as unauthorized private parties.” Id. The SCA sought to fill this perceived gap in the Fourth Amendment‘s protections:
[T]he law must advance with the technology to ensure the continued vitality of the fourth amendment. Privacy cannot be left to depend solely on physical protection, or it will gradually erode as technology advances. Congress must act to protect the privacy of our citizens. If we do not, we will promote the gradual erosion of this precious right.
2. Meta Would Expand the SCA Far Beyond the Fourth Amendment
This Congressional intent—to eliminate an instance of legal arbitrage by applying the Fourth Amendment‘s protections to a new technology via statute—comports with our reading of the SCA‘s disclosure provisions. Communications blasted in public fora, for all to see or hear, generally are not protected by the Fourth Amendment, putting the nicety of third-party electronic transmitters of communications aside. “What a person knowingly exposes to the public, even in his own home or office, is not a subject of Fourth Amendment protection.” Katz v. United States, 389 U.S. 347, 351 (1967); accord Biles v. United States, 101 A.3d 1012, 1024 (D.C. 2014). So it would make little sense to extend the SCA‘s protections to such communications.
Meta‘s interpretation would do just that, despite the fact that publicly broadcast communications have no shelter in the Fourth Amendment itself. That would expand the SCA far beyond the Fourth Amendment protections that Congress sought to mimic. To the contrary, the House and Senate Reports affirmatively indicate that Congress did not intend for the SCA‘s protections to cover content that the user took no steps to keep private. For example, both reports include extended discussions of electronic bulletin board systems (BBS)—“early analogues to the social media platforms at issue here.” Hunter, 417 P.3d at 739. As one article describes this archaic technology, dialing into a BBS was akin to “visit[ing] the private residence of a fellow computer fan electronically. BBS hosts had converted a PC . . . into a digital playground for strangers’ amusement.” Benj Edwards, The Lost Civilization of Dial-Up Bulletin Board Systems, The Atlantic (Nov. 4, 2016). These early (now anachronistic) digital meeting spaces could be configured as either “public or semi-public in nature, depending on the degree of privacy sought by users.” S. Rep. 99-541, at 9. Only the latter, non-publicly accessible BBSs were intended to fall within the SCA‘s protections. As the Senate Report puts it, the SCA‘s protections do not apply where a BBS “does not require any special access code or warning to indicate that the information is private. To access a communication in such a public system is not a violation of the Act, since the general public has been ‘authorized’ to do so by the facility provider.” Id. at 36; see also Snow v. DirectTV, Inc., 450 F.3d 1314, 1321 (11th Cir. 2006) (“[T]he requirement that the electronic communication not be readily accessible by the general public is material and essential to” the SCA‘s scope of protections).
Meta counters that this discussion of BBSs relates only to the SCA‘s provisions prohibiting the unauthorized access and interception of electronic communications, or what is effectively the SCA‘s anti-hacking provision,
3. Meta Would Leave the SCA‘s Protections to Service Providers’ Discretion
There is another feature of Meta‘s statutory interpretation that is at odds with the SCA‘s legislative history. That history evinces Congress‘s intent to confer upon individual users of electronic services Fourth Amendment-like protections. But recall that Meta‘s view is not that its users have any right to prevent it from complying with the subpoena in this case—because a
It is true that when a
4. Meta‘s Counterpoints Are Unpersuasive
Meta offers two counterpoints from the legislative history, but neither is
Next, Meta argues that applying
* * *
In summary, the SCA does not authorize a service provider‘s refusal to comply with valid legal process seeking material that a
III.
We now turn to Meta‘s argument that the District‘s subpoena impermissibly intrudes
We begin by laying some legal groundwork. A court will ordinarily enforce an investigative subpoena so long as it meets the three-prong test announced in United States v. Morton Salt Co., 338 U.S. 632, 652-53 (1950). Under that test, “[w]e consider only whether [1] ‘the inquiry is within the authority of the agency, [2] the demand is not too indefinite and [3] the information sought is reasonably relevant.‘” Resol. Tr. Corp. v. Thornton, 41 F.3d 1539, 1544 (D.C. Cir. 1994) (quoting Morton Salt, 338 U.S. at 652). There is no dispute, and we agree, that this test is satisfied here: (1) the District, through its Office of the Attorney General, is charged with enforcing the CPPA and may “subpoena witnesses” and “compel production of records” under its investigative authority to do so,
But Meta argues that Morton Salt does not apply here, because where compelled disclosures seriously implicate First Amendment interests, government subpoenas may face more exacting judicial scrutiny. The seminal case for this proposition is NAACP v. Alabama ex rel. Patterson, 357 U.S. 449 (1958), which involved an attempt by Alabama‘s attorney general to compel the disclosure of the NAACP‘s membership lists. Id. at 452. The NAACP refused to comply, and it was held in civil contempt and fined $100,000. Id. at 453-54. The Supreme Court of the United States reversed, reasoning that Alabama‘s investigation into the NAACP “entail[ed] the likelihood of a substantial restraint upon the exercise by petitioner‘s members of their right to freedom of association,” and that Alabama had failed to demonstrate an interest “which is sufficient to justify the deterrent effect.” Id. at 462-63. The Court more recently described this “exacting scrutiny” standard as requiring “a substantial relation between the disclosure requirement and a sufficiently important governmental interest.” Ams. for Prosperity Found. v. Bonta, 141 S. Ct. 2373, 2383 (2021) (”AFPF“) (quoting Doe v. Reed, 561 U.S. 186, 196 (2010)).9
But that more recent decision in AFPF did not suggest that all government subpoenas are doomed under the exacting scrutiny standard, regardless of the nature of the information sought. Rather, the Supreme Court indicated that compelled disclosures need only satisfy this standard when “First Amendment activity is chilled—even if indirectly.” Id. at 2384. When it is not, the far more deferential Morton Salt standard continues to govern our review. A party such as Meta claiming a First Amendment privilege bears the burden of “demonstrat[ing] that enforcement of the discovery requests will result in . . . consequences which objectively suggest an impact on, or ‘chilling’ of, the members’ associational rights.” Perry v. Schwarzenegger, 591 F.3d 1147, 1160 (9th Cir. 2010) (citation omitted). Only after this prima facie showing do we consider if a subpoena satisfies exacting scrutiny. Id. at 1161; accord In re Motor Fuel Temperature Sales Pracs. Litig., 641 F.3d 470, 488 (10th Cir. 2011) (“[T]he party claiming a privilege always bears the initial burden of establishing the factual predicate for the privilege.“).
For the reasons that follow, we conclude that Meta has not shown that the District‘s subpoena, which seeks information related to publicly accessible content generated by its users, will result in chilling Meta‘s free speech or associational rights. As to Meta‘s users, we assume the exacting scrutiny standard applies, but conclude that the District has demonstrated that its subpoena is “narrowly tailored to the government‘s asserted interest.” AFPF, 141 S. Ct. at 2383. We therefore hold that enforcing the District‘s subpoena does not violate the First Amendment.
A. Meta‘s Own First Amendment Rights
We begin with Meta‘s claim that the District‘s subpoena impermissibly intrudes upon its own First Amendment rights by “prob[ing] and penaliz[ing]” its ability to exercise editorial control over the content that is disseminated through its platform. The trial court disagreed, concluding that even if the First Amendment protects the ability of a private social media company to make unfettered content moderation decisions,10 enforcing the District‘s subpoena would not chill Meta from engaging in that activity, so that exacting scrutiny is unwarranted. We agree.
At its core, Meta‘s argument boils down to two assertions: that the District‘s investigation (1) is really just an attempt to “pressure Meta into changing how it exercises [its] protected editorial control over its platform“; and (2) that government scrutiny of its practices more generally will lead to a chilling of the company‘s speech.
On the first point, we disagree with Meta‘s characterization of the District‘s investigation. As the subpoena itself states, the District is investigating only whether Meta‘s “representations regarding efforts to prevent and remove vaccine misinformation from the Facebook platform” violate the District‘s consumer protection statute, the CPPA. There is no suggestion that the District is investigating whether Meta‘s moderation policies or efforts to police them were unlawful or insufficient in themselves (except to the extent that they belie Meta‘s representations). The District has disclaimed any interest in regulating Meta‘s editorial judgment when it comes to its content moderation, and Meta‘s reply brief expressly denies accusing the District of acting in bad faith. This was a prudent concession. While it is certainly possible for an otherwise valid government investigation to be launched on pretextual grounds, Meta points to no evidence that this is the case here. See Dep‘t of Com. v. New York, 139 S. Ct. 2551, 2573–74 (2019) (emphasizing that a “strong showing of bad faith or improper behavior” is required before
As to Meta‘s argument that the District‘s subpoena (even if issued as part of a legitimate investigation) nonetheless chills its speech, we again disagree. To reiterate, the only speech that is being targeted by the District‘s investigation are Meta‘s public statements regarding the company‘s content moderation practices, which the District alleges were deceptive and in violation of the CPPA. If those allegations are true, then an enforcement action under the CPPA would pose no constitutional problem at all, as the First Amendment “does not prohibit the State from insuring that the stream of commercial information flow cleanly as well as freely.” Va. State Bd. of Pharmacy v. Va. Citizens Consumer Council, Inc., 425 U.S. 748, 772 (1976). In other words, even if content moderation is itself protected speech, fraudulent misrepresentations regarding a company‘s moderation practices is not.
Meta tries to take this argument a step further, claiming an investigation into its statements about its content moderation practices might indirectly chill those practices themselves. “[J]ust as a subpoena demanding notes from an editorial board meeting would risk chilling a newspaper‘s editorial rights,” Meta argues, so too does the subpoena here threaten its “exercise of editorial control.” The problem with that analogy is that Meta not only made its content moderation policies publicly available, it then widely touted the actions that were supposedly taken pursuant to those policies; indeed, those public statements were the basis for the District‘s investigation. To piggyback on the editorial board analogy, if the newspaper itself had published an account of its editorial policies and decisions, and it turned out to be potentially fraudulent in some way, it would not chill the newspaper‘s exercise of editorial control to investigate whether the newspaper‘s public statements on that topic were false. Meta offers no theory for how a subpoena targeting documents that tangentially relate to this entirely public information risks any chilling of its speech, and we likewise discern none.
B. Meta‘s Users’ First Amendment Rights
Meta also argues that enforcing the District‘s subpoena would chill the First Amendment rights of its users. In essence, its theory is that forcing Meta to identify the users whose posts were removed under the company‘s COVID-19 misinformation policy “associate[s]” those users with “speech that [the District] views as undesirable.” That association, Meta argues, risks deterring these users from engaging in future online discussions of controversial topics. See Talley v. California, 362 U.S. 60, 65 (1960) (“[I]dentification and fear of reprisal might deter perfectly peaceful discussions of public matters of importance.“). We seriously doubt that. The District seeks disclosures related to public posts, and the users who made those posts have already openly associated themselves with their espoused views by publicly posting them to Facebook. While we doubt exacting scrutiny should apply here, we will assume that it does for the sake of argument, and conclude that the District‘s subpoena nonetheless passes constitutional muster.
Recall that exacting scrutiny examines the fit between the importance of the government‘s interest and the means used to realize that interest. “To withstand this scrutiny, the strength of the governmental interest must reflect the seriousness of the actual burden on First Amendment rights.” AFPF, 141 S. Ct. at 2383 (quoting Doe, 561 U.S. at 196). More
The District‘s subpoena satisfies both of these requirements. The CPPA “establishes an enforceable right to truthful information from merchants about consumer goods and services that are or would be purchased, leased, or received in the District of Columbia.”
Meta acknowledges that the District has a “legitimate interest in consumer protection in general.” It nonetheless argues that such an interest is not implicated here, where Meta‘s public statements about content moderation were mere “puffery” and therefore non-actionable under the CPPA. But commercial puffery is non-actionable because it consists of statements whose “truth or falsity . . . cannot be precisely determined,” such as a sign in a storefront window promising “Satisfaction Guaranteed.” Pearson v. Chung, 961 A.2d 1067, 1076 (D.C. 2008) (quoting Tietsworth v. Harley-Davidson, Inc., 677 N.W.2d 233, 245 (Wis. 2004)). This sort of general assertion, incapable of measurement, is unlikely to lead reasonable consumers astray and therefore cannot be the basis for a CPPA violation. Id. But that does not describe Meta‘s public statements about its COVID-19 misinformation policy. Meta claimed that it removed 20 million items of content and over 3,000 user accounts as a result of enforcing that policy. These are not the “[l]ofty but vague” statements that can be chalked up to puffery. See Prager Univ. v. Google LLC, 951 F.3d 991, 1000 (9th Cir. 2020). They are instead quite detailed, quantifiable, and capable of verification.
As to the fit between that government‘s interest and the scope of the District‘s investigation, we likewise conclude that the subpoena—now that it has been limited to documents relating to publicly accessible posts—is sufficiently tailored. Though Meta claims that the District‘s subpoena could have pursued “less intrusive alternatives,” such as aggregated or anonymized data,11 some of the statements
IV.
For the foregoing reasons, the judgment of the Superior Court is affirmed.
So ordered.
DEAHL, Associate Judge, concurring: I am in full agreement with the court‘s opinion and write separately to address two issues it rightly bypasses. First is whether
First, on the question of consent, I adhere to a general rule of thumb when trying to figure out if somebody consents to something: You ask them. Here, the users whose posts are targeted by the District‘s subpoena have not been asked whether they consent to disclosure to the government, and so I find it artificial to say they have consented to such disclosure. The trial court reasoned to the contrary, that “when a user posts content on Facebook that is generally accessible to the public, the user implicitly consents to disclosure.” While that might be a fair inference if the posts remained public, the posts at issue here have all been removed, so I see no reason to conclude that any consent to disclosure endures. People are generally free to withdraw consent and might do so by, for instance, removing or restricting access to a once-public post. See Ford v. United States, 245 A.3d 977, 984-85 (D.C. 2021) (recognizing ability to withdraw or revoke consent to a search). The fact that the posts at issue here are no longer public would preclude me from inferring any present consent to disclosure. Nonetheless, Meta does not challenge this aspect of the trial court‘s ruling, and so I agree with the opinion for the court that this point has been conceded. Supra at 11-13.
To be sure, a person who publicly posts something opens themselves up to the risk—really, the high likelihood when it comes to popular social media sites—that some third party will save their post for posterity and render any attempt to delete it from public viewing futile. But that is just to say that third parties generally may do what they will with publicly disclosed communications, which is quite different from saying the user consents to whatever they do. And a service provider is not free to do what they want with the communication if the SCA‘s protections apply to it (an important caveat discussed next): they are constrained by the statute, and where the statute requires the user‘s consent to disclosure, I do not think that eternal consent
Second, I agree with the District that the SCA does not apply to public posts in the first place,1 so my above concern with the consent exception‘s application is an entirely academic point here. The SCA was enacted to “protect electronic communications that are configured to be private.” Konop v. Hawaiian Airlines, Inc., 302 F.3d 868, 875 (9th Cir. 2002). When read as a whole, and in light of the legislative history discussed extensively in the court‘s opinion, the SCA‘s apparent “purpose is to protect information that the communicator took steps to keep private.” Ehling v. Monmouth-Ocean Hosp. Serv. Corp., 961 F. Supp. 2d 659, 668 (D.N.J. 2013); Snow v. DirecTV, Inc., 450 F.3d 1314, 1321 (11th Cir. 2006) (“[T]he requirement that the electronic communication not be readily accessible by the general public is material and essential to” the scope of the SCA‘s protections.).
When a person publicly posts a message for the world to see, it falls outside of the SCA‘s protections altogether. In that case, the service provider is best seen as providing a public platform for a user to broadcast a message, rather than acting as an “electronic communication service,” a phrase the statute seems to use to refer to a third-party transmitter of otherwise private communications.
This reading of the SCA makes sense. The SCA was meant to effectively neutralize the undesired but necessary disclosure of private communications to third-party service providers; it is not as if the user wants to share their communications with service providers, so much as they are necessary conduits for relaying messages to their intended recipients. The SCA steps into that relationship to dictate that the disclosure to a third-party service provider merely for the purposes of transmitting the message is a non-event, and should not affect the user‘s privacy interests in their communications that might otherwise be deemed private. But when a user blasts a message for the world to see, the service provider does not act merely as a necessary transmitter of that communication, but can itself be seen as a recipient of it (just like everybody else). The third-party transmittal problem that is the SCA‘s raison d‘être no longer exists. In that situation there is no Fourth Amendment gap for the SCA to fill, so it makes little sense to extend the SCA‘s protections to it.
My view admittedly faces a textual hurdle, which is that nothing in the statutory definitions of “electronic communication” or “electronic communication service” expressly says that the communication at issue must be a private one. See
The SCA is antiquated and could no doubt use a legislative update, but in the meantime courts should read its provisions in a way that makes sense of the entire statutory scheme, while cognizant of just how much has changed in the nearly-four decades since it was passed. Doing that leads me to conclude that the SCA‘s protections do not extend to public posts, and the court should say so if a more appropriate occasion ever arises.