It seems we can’t go a day without another bombshell in what we can now call the “Snowden Affair.” Many people are calling Edward Snowden a “whistleblower” for leaking two classified intelligence programs. But that term is usually reserved for someone who reveals government lies, law-breaking, or malfeasance. What Snowden revealed, though, are government activities that, based on what’s been reported, appear to be lawful, at least in the sense of being consistent with laws enacted by our elected representatives. One can legitimately argue about whether the two programs are effective, whether the benefits to security are worth the cost to privacy, and whether there is sufficient congressional and judicial oversight of such programs. One can also argue about whether such programs violate the Fourth Amendment. But there does not seem to be any reasonable doubt that the programs are consistent with the United States Code. Nor, really, should these revelations surprise anyone. So the member of Congress, reporter, or civil libertarian who professes shock that these programs exist hasn’t been paying close attention to developments in national security law the last eight years. Or, like Captain Renault in Casablanca, he’s just turned a blind eye to them.
Last week I discussed Snowden’s first leak, which involved the Hoovering of massive amounts of telephone “metadata” from US telecoms, including metadata about local calls in the United States. This program is being performed pursuant to Section 215 of the USA Patriot Act, which allows the Foreign Intelligence Surveillance Court (FISC) to issue “an order requiring the production of any tangible things…for an investigation to obtain foreign intelligence information not concerning a United States person or to protect against international terrorism or clandestine intelligence activities.” To obtain such an order, the government must show only “that there are reasonable grounds to believe that the tangible things sought are relevant to an authorized investigation.”
Today, I’ll focus on the second leak, involving a Top Secret government program known as “Prism.” According to press reports, the NSA and FBI are tapping into the databases of Microsoft, Google, Yahoo!, Facebook, PalTalk, YouTube, Skype, AOL, and Apple. The agencies are obtaining, inter alia, the content of e-mails, chats, videos, video conference calls, VoIP communications, photos, file transfers, and responses to “special requests.” Unlike the telephone metadata program, which involves only access to stored data, Prism appears to allow the government to wiretap communications during transmission.
Prism is based not on Section 215 of the Patriot Act, but on Section 702 of the FISA Amendments Act of 2008 (codified at 50 U.S.C. § 1881a). This provision authorizes US intelligence agencies to “target” for acquisition the communications of non-US persons who are “reasonably believed to be located outside the United States”—even if the surveillance is effected in, and one or more parties to the communication are located within, the United States. Section 702 allows the Attorney General and the Director of National Intelligence, without a court order, to direct an electronic communication service provider to “immediately provide the Government with all information, facilities, or assurance necessary to accomplish the acquisition.”
However, prior to such surveillance, the government must gain approval from the Foreign Intelligence Surveillance Court of its “targeting” and “minimization” procedures. The targeting procedures are supposed to ensure that surveillance is “targeted” only at non-US persons outside the United States. The minimization procedures are meant to limit the acquisition and retention, and prohibit the dissemination, of information collected against “unconsenting United States persons consistent with the need of the United States to obtain, produce, and disseminate foreign intelligence information.”
What does this mean in plain English? That the NSA and FBI can collect the communications of non-US persons located outside the US even if those communications are with a US person inside the US. They can do so without a court order. And they can retain and use any information collected about US persons as long as the government determines that such retention or use is needed for foreign intelligence purposes.
That last point has gone largely unnoticed by reporters and analysts. Much of the press seems to view the telephone metadata program as more worrisome because it appears to be collecting information about pretty much all calls that occur within, or cross through, the United States. But remember, that program does not collect the content of phone calls, or even the names of the persons on the call, but only information about what phone numbers are in communication with each other, the duration of the calls, and things like that. Prism, on the other hand, collects the content of a broad spectrum of Internet communications. Many people seem to regard Prism as a less serious privacy intrusion because the surveillance cannot be targeted at US persons (i.e., citizens and legal resident aliens). But the communications of US persons are being collected.
The government refers to the collection of US persons’ communications as “incidental,” since the US person may not be the “target.” This makes it sound like the collection is rare, inadvertent, and minor. But that’s not really accurate. One of the points of the collection is to determine whether foreign terrorists (or spies) are conspiring with someone in the United States, to do harm in America. Indeed, that is how the Administration has justified the program. Officials have cited, in particular, the use of Prism to disrupt a 2009 plot by an American, Najibullah Zazi, to bomb the New York subway system. They say the government learned about the plot because the NSA was monitoring the communications of an al-Qaeda bomb-maker, Rashid Rauf, and found he was in contact with Zazi, who was in Aurora, Colorado. Presumably the authorities then got a FISA warrant to eavesdrop on Zazi’s complete communications. Senator Dianne Feinstein (D-CA), Chair of the Senate Intelligence Committee, also said that Prism helped implicate American David Headley in the 2008 terrorist attacks in Mumbai.
In addition, a government official, in describing Prism, said that
“[i]f the N.S.A. comes across information about an American citizen during the search, it turns over that material to the F.B.I. for an assessment.”
That hardly sounds like “incidental” collection the way ordinary people would understand that term.
Once the government has reason to focus surveillance on—i.e., “target”—a particular US person, it would have to get a regular FISA surveillance order, or a criminal wiretap order. One would expect that the government’s targeting and minimization procedures address this.
Another issue concerns how Prism works. The leaked slides describing Prism say that NSA collects the information “directly from the servers” of the participating communication providers. Several providers deny that the government has direct access to their servers, and say that they respond only to targeted requests for information. But this seems to be a semantic quibble. According to another leaked document, Prism allows “collection managers [to send] content tasking instructions directly to equipment installed at company-controlled locations.” And this process is probably automated. So it sounds like the government attaches hardware to the companies’ networks; it sends “tasking” instructions to that hardware instructing it to collect the communications of certain targets; and the hardware then extracts the communications and sends them to NSA. So is the government collecting the information directly from the companies’ servers, or is it collecting from the hardware it attached to the servers? Does it matter?
After you cut through all the semantics, the only important questions are whether the program is effective, whether it is constitutional, and whether the benefit to security is worth the cost to privacy. As to its effectiveness, the government claims that Prism is a vital tool and has been instrumental in stopping terrorist attacks. We’ll probably hear about cases other than the Zazi and Headley matters in the coming days to demonstrate this. And according to the NSA slides, Prism data has been cited in the President’s Daily Intelligence Brief more than any other source, and its data has resulted in nearly one in seven intelligence reports. If this information is not just puffery, it suggests that Prism is pretty darn effective.
Is it legal? Well, all three branches of government seem to think so. The Executive Branch obviously does, as it is implementing the program, and President Obama has given a full-throated defense of it. The Legislative Branch does, too, since Prism appears completely consistent with Section 702 of the FISA Amendments Act. And at least part of the Judicial Branch does, since the FISC hasn’t chosen to reject the procedures under which Prism operates and thereby stop the program.
But can there be more meaningful judicial review? After all, the FISC has only a limited role, and is often accused (fairly or not) of being a rubber stamp for the intelligence agencies. Its opinions are also almost never made public.
Unfortunately, the likelihood of more meaningful review is quite small. Just this March, the US Supreme Court, in Clapper v. Amnesty International, turned aside a constitutional challenge to Section 702 on the ground that the plaintiffs—a group of civil liberties lawyers, human rights advocates, and others—lacked “standing” to bring suit. To establish standing, a plaintiff must show some concrete “injury in fact.” If the injury is only something that might take place in the future, then that threatened injury must be “certainly impending.”
The plaintiffs in Clapper claimed standing based on their fear that the US government would intercept their communications with non-US persons overseas without a warrant, and on the costs they had incurred to protect the confidentiality of their communications. But the Court found that these claimed injuries did not pass muster, since they “relie[d] on a highly attenuated chain of possibilities.” The Court said that even if plaintiffs could demonstrate that their foreign contacts were being monitored (which they hadn’t), they could not show that the US government was monitoring those persons under Section 702, as there are “numerous other methods of conducting surveillance,” or that spying on such persons would also lead to the interception of the plaintiffs’ communications. Thus, the plaintiffs’ argument was based on what the Court implied was “a paranoid fear.”
That fear doesn’t seem so paranoid in light of last weeks’ revelations. But the reality is that the Court’s decision makes it difficult for anyone to challenge the law. The only real chance is if someone is prosecuted based on information gathered pursuant to Section 702, and that person moves to suppress the evidence at trial. But that is fairly unlikely, since the government would almost certainly use at trial only wiretap evidence that it collected pursuant to a regular FISA or criminal wiretap order, or documentary or testimonial evidence. Moreover, a challenge would have to overcome the state secrets privilege, which has proven to be an all but insurmountable obstacle to challenges of government intelligence programs, including the Bush Administration’s warrantless wiretapping program.
Finally, there’s the question of whether the security benefits of the program are worth the cost to privacy. At bottom that’s a subjective question, and people will make their own judgments based largely on their personal notions of privacy. In a democracy, the societal judgment about such things is supposed to be made by our elected representatives. No President, whose first responsibility is to protect the security of the nation, is realistically going to decline to use a legal authority that has been granted to him. So that leaves Congress. And as I’ve said, Congress already enacted Section 702. If Congress now collectively determines that it didn’t realize what it was authorizing back in 2008, or thinks the intelligence community has gone too far, it can a) eliminate or cut back on the program, or b) exercise more rigorous oversight to make sure the program stays within bounds. Will it do either of these things? We’ll have to wait and see. But I wouldn’t hold my breath.