Wikimedia Foundation v. National Security Agency

DAVIS, Senior Circuit Judge,

concurring in part and dissenting in part:

I agree with the holding that Wikimedia has standing to challenge the NSA’s surveillance of its internet communications. However, because I would find that the non-Wikimedia Plaintiffs also have standing, I respectfully dissent in part.

I.

In order to explain my disagreement with the majority, I briefly recount the relevant allegations in this case, taken as true, of course, at this stage of the proceedings. Plaintiffs make essentially two sets of factual allegations: the first explaining how international internet communications function and the second describing how the NSA surveils international internet communications as they enter and exit the United States.

First, Plaintiffs allege that internet communications are governed by certain technical rules as they travel from sender to recipient. The majority of international internet communications that move through the United States are transmitted through forty-nine submarine cables and a limited number of terrestrial cables. These cables (combined with the cables and networks that transmit domestic internet communications) are known as the internet backbone, and the different physical entry and exit points into the United States are known as backbone links. The junctions where these cables meet are chokepoints through which nearly all international internet traffic passes. Internet communications do not flow along the backbone as discrete and intact entities but instead are broken into smaller packets of information. *218The packets that make up a single internet communication travel to their common destination independently from one another— in the process becoming intermingled with packets from unrelated communications— and are reassembled only once they reach their destination. Each packet reaches its destination following a different and wholly unpredictable path, which is determined by rapidly changing factors such as network conditions. Because packets travel along independent and dynamic paths, communications sent between two individuals in “real-time” can traverse different backbone links “even though the end points are the same.” J.A. 50. Similarly, a single individual’s communications sent at different times can traverse different backbone links.

Second, based on the government’s disclosures and media reports, Plaintiffs allege that the NSA is surveilling internet communications as they travel along the internet backbone, a practice known as Upstream surveillance. The NSA accomplishes this by installing surveillance devices at backbone links, which allow the agency to copy the internet communications traversing these links. The NSA searches the copied communications for selectors. Selectors are “specific communications facilities]” (e.g. email address, telephone numbers, and IP addresses) associated with the NSA’s foreign surveillance targets. PCLOB Report 32. The NSA retains communications sent to or from a selector as well as communications containing a selector in their content, which are known as “about communications.” About communications are not necessarily sent to or from a foreign surveillance target. According to the government’s disclosures, surveillance of about communications is necessary because the NSA seeks to “comprehensively acquire communications that are sent to or from its targets.” Id. at 10. With respect to the scope of Upstream surveillance, the New York Times reported that, through the use of this form of surveillance, the NSA is copying “what is apparently most emails and other text-based communications that cross the border.” J.A. 51. Plaintiffs also quote an NSA document that states the “NSA has expended a significant amount of resources to create collection/processing capabilities at many of the chokepoints operated by U.S. providers through which international communications enter and leave the United States.” Id.

II.

I agree with the majority’s analysis concluding that Clapper v. Amnesty International USA, 568 U.S. 398, 133 S.Ct. 1138, 185 L.Ed.2d 264 (2013), does not control this case and that—accepted as true, as they must be—Plaintiffs’ allegations satisfy the three elements of standing. The majority also correctly finds that the factual allegations necessary to establish Wik-imedia’s standing are plausible. However, the majority errs, both by reaching out to decide the issue of the non-Wikimedia Plaintiffs’ standing1 and, as well, in the *219answer it gives to the question it need not even reach in holding that the non-Wik-imedia Plaintiffs’ lack standing because the pertinent allegations are not plausible.

In order to find that WiMmedia has standing in this action, the majority credits as true three factual allegations. First, because Wikimedia sends and receives so many international internet communications, its communications travel across every internet backbone link. Second, based on the government’s disclosures, the NSA is surveilling at least one backbone link. Third, the NSA intercepts and copies every packet that passes through the backbone link(s) being surveilled (what the majority calls the Wikimedia Allegation). The third allegation is not based on Plaintiffs’ knowledge of the NSA’s surveillance techniques. Instead, the majority finds this factual allegation is plausible because it is based on Upstream surveillance’s stated purpose and the technical rules that govern internet communications. The logical chain is as follows: The NSA has acknowledged that it uses Upstream surveillance to target “about communications,” which contain a selector in the content of the communication. Before it can search the contents of an internet communication that has been broken up into discrete packets while in transit, the NSA must copy and reassemble all of the packets that make up the communication. However, packets from targeted communications cannot be segregated from the packets of unrelated communications. Thus, in order to “reliably” intercept targeted communications, the NSA must copy all of the packets that flow across a backbone link so that the government can be assured that it has captured all of the packets that make up the targeted communication (and in the process capturing unrelated packets). J.A. 48-U9.

Conversely, under the majority’s “crabbed plausibility analysis,” see Woods v. City of Greensboro, 855 F.3d 639, 642 (4th Cir. 2017), the non-Wikimedia Plaintiffs are denied standing because, in the majority’s view, those Plaintiffs rely on an implausible guess regarding Upstream surveillance’s operational scope. For the *220non-Wikimedia Plaintiffs to have standing, according to the majority, Plaintiffs must plausibly allege an additional fact beyond those discussed with respect to Wikimedia: the NSA is surveilling most backbone links (what the majority calls the Dragnet Allegation). Just as with the Wikimedia Allegation, Plaintiffs base this factual allegation on Upstream surveillance’s stated purpose and the technical rules governing internet communications.2 However, the majority finds this allegation implausible because it believes that “neither theory nor purpose says anything about what the NSA is doing from an operational standpoint.” Op. at 214. This misapprehends the full scope of Plaintiffs’ allegations.

Plaintiffs have plausibly alleged that the NSA surveils most backbone links because—based on the technical rules governing internet ' communications—the agency cannot know which link the communications it targets will traverse when they enter or leave the United States. The path that packets take along the internet backbone is determined dynamically based on unpredictable conditions. Thus, a communication sent by a surveillance target can enter the United States through one backbone link, but an immediate response returned to the surveillance target can traverse a different backbone link. Similarly, communications sent by a surveillance target at different times or locations can traverse different backbone links. Given this technical limitation, the government’s disclosure that the NSA seeks to “comprehensively acquire communications that are sent to or from its targets,” J.A. 49, renders Plaintiffs’ allegation plausible. If the NSA cannot know which backbone link its targets’ internet communications will traverse, then the only way it can comprehensively acquire its targets’ communications is by surveilling virtually every backbone link. .

This allegation is essentially a logical extension of Plaintiffs’ earlier allegation that the NSA must copy every communication that flows across a backbone link it surveils. Just as it is plausible that the government must copy all of the packets that flow through a backbone link in order to “reliably” capture the packets that make up a targeted internet communication, because the government does not know across which backbone link a communication will travel, it is also plausible that the government must monitor virtually every link in order to “comprehensively” capture its targets’ communications. Given that we review here a motion to dismiss and not a motion for summary judgment, the non-Wikimedia Plaintiffs have provided enough factual support to their allegation to survive dismissal.

III.

For the reasons set forth, while I discern no need whatsoever to review the district court’s legal determination of the non-Wikimedia Plaintiffs’ standing, I respectfully dissent from the majority opinion’s unnecessary resolution of that issue.

. See Horne v. Flores, 557 U.S. 433, 446, 129 S.Ct. 2579, 174 L.Ed.2d 406 (2009) ("Because the Superintendent clearly has standing to challenge the lower courts’ decisions, we need not consider whether the legislators also have standing to do so.”); Vill. of Arlington Heights v. Metro. Hous. Dev. Corp., 429 U.S. 252, 264 and n.9, 97 S.Ct. 555, 50 L.Ed.2d 450 (1977) (holding because “one individual plaintiff ... has demonstrated standing,” the Court “need not consider whether the other individual and corporate plaintiffs have standing to maintain the suit”). The majority’s “same relief” gloss on Horne and Arlington Heights completely reads out of Justice Alito's opinion in Home the following sentence: "[I]n all standing inquiries, the critical question is whether at least one petitioner has alleged *219such a personal stake in the outcome of the controversy as to warrant his invocation of federal-court jurisdiction.” Horne, 557 U.S. at 445, 129 S.Ct. 2579 (citations and internal quotation marks omitted). In any event, this case actually fits within the majority’s "same relief” paradigm because all plaintiffs seek declaratory and injunctive relief intended to shut down the government’s Upstream surveillance program. The mere fact that a "purging order” of the sort contemplated by the majority would operate only to "purge” seized communications of a particular plaintiff is a thin reed indeed on which to base the majority's unnecessary door-closing result.

It is not clear to me why the majority elects to ignore the Chief Justice’s sage admonition: ”[I]f it is not necessary to decide more, it is necessary not to decide more.” PDK Labs., Inc. v. Drug Enforcement Admin., 362 F.3d 786, 799 (D.C. Cir. 2004) (Roberts, J., concurring in part and concurring in the judgment). The majority's assertion to the contrary notwithstanding, I think I know dicta when I see it, and here I see dicta. If, in fact, the Wikime-dia Plaintiffs go on to prove their claims in this case, i.e., establish a violation of the Fourth Amendment as to themselves, it is beyond my capacity to conjure a rational basis on which the non-Wikimedia Plaintiffs would not be entitled to similar relief from seizures effected pursuant to the Upstream program and of course, the dismissal here of the non-Wikimedia Plaintiffs will be without prejudice. S. Walk at Broadlands Homeowner’s Ass’n v. OpenBand at Broadlands, LLC, 713 F.3d 175, 185 (4th Cir. 2013).

In sum, the day cannot be far off when defendants in a broad array of multi-plaintiff cases will point to the majority’s holding in this case as authority requiring already shorthanded and overworked federal district judges to separately assess the standing of each and every plaintiff in complex, impact litigation. Needless to say, we should avoid imposing such a requirement in the absence of the absolute necessity that we do so.

. Plaintiffs provide additional support for this allegation by corroborating it with a N.Y. Times report, which stated that the NSA is surveilling "most e-mails and other text-based communications that cross the border.” J.A. 51. The majority finds that this report is entitled to "little significance” because it "is effectively a recitation of" Plaintiffs’ allegation. Op. at 214. The N.Y. Times report predates the complaint, however; thus, the allegation is a "recitation” of the factual news report, not the other way around. Moreover, the fact that Plaintiffs based their allegation on factual news reporting rather than their own conjecture means the allegation is entitled to more weight not less.