Clapper’s Claim that FBI Cannot Count Back Door Searches for Technical Reasons Probably Bullshit

I wanted to explain why I think it’s such a big deal that James Clapper specifically highlighted the carve out for transparency reporting on FBI’s back door searches in Leahy’s version of Freedom Act’s in his letter supporting the bill.

As I described, the bill requires reporting on back door searches, but then exempts the FBI from that reporting.

But that’s not the part of the bill that disturbs me the most. It’s this language:


Subparagraphs (B)(iv), (B)(v), (D)(iii), (E)(iii), and (E)(iv) of paragraph (1) of subsection (b) shall not apply to information or records held by, or queries conducted by, the Federal Bureau of Investigation.

The language refers, in part,  to requirements that the government report to Congress:

(B) the total number of orders issued pursuant to section 702 and a good faith estimate of—

(iv) the number of search terms that included information concerning a United States person that were used to query any database of the contents of electronic communications or wire communications obtained through the use of an order issued pursuant to section 702; and

(v) the number of search queries initiated by an officer, employee, or agent of the United States whose search terms included information concerning a United States person in any database of noncontents information relating to electronic communications or wire communications that were obtained through the use of an order issued pursuant to section 702;

These are back door searches on US person identifiers of Section 702 collected data — both content (iv) and metadata (v).

In other words, after having required the government to report how many back door searches of US person data it conducts, the bill then exempts the FBI.

In his letter, Clapper says,

[W]e are comfortable with the transparency provisions in this bill because, among other things, they recognize the technical limitations on our ability to report certain types of information.

FBI back door searches are the most obvious limit on transparency guidelines, and FBI told PCLOB they couldn’t count them for technical reasons.

So effectively, Clapper is suggesting that Congress has recognized that FBI is incapable — for technical reasons — of counting how often it conducts back door searches.

That technical claim is almost certainly bullshit.

As a reminder, here’s what the government told PCLOB about FBI’s back door searches.

Because they are not identified as such in FBI systems, the FBI does not track the number of queries using U.S. person identifiers. The number of such queries, however, is substantial for two reasons.

First, the FBI stores electronic data obtained from traditional FISA electronic surveillance and physical searches, which often target U.S. persons, in the same repositories as the FBI stores Section 702–acquired data, which cannot be acquired through the intentional targeting of U.S. persons. As such, FBI agents and analysts who query data using the identifiers of their U.S. person traditional FISA targets will also simultaneously query Section 702–acquired data.

Second, whenever the FBI opens a new national security investigation or assessment, FBI personnel will query previously acquired information from a variety of sources, including Section 702, for information relevant to the investigation or assessment. With some frequency, FBI personnel will also query this data, including Section 702–acquired information, in the course of criminal investigations and assessments that are unrelated to national security efforts. In the case of an assessment, an assessment may be initiated “to detect, obtain information about, or prevent or protect against federal crimes or threats to the national security or to collect foreign intelligence information.”254 If the agent or analyst conducting these queries has had the training required for access to unminimized Section 702–acquired data, any results from the Section 702 data would be returned in these queries. If an agent or analyst does not have access to unminimized Section 702–acquired data — typically because this agent or analyst is assigned to non-national security criminal matters only — the agent or analyst would not be able to view the unminimized data, but would be notified that data responsive to the query exists and could request that an agent or analyst with the proper training and access to review the unminimized Section 702–acquired data.

Continue reading

Did Anthony Coppolino Fib about NSA’s New Architecture?

On Tuesday, EFF told the tale of yet another government freak-out over purportedly classified information. The DOJ lawyer litigating their multiple dragnet challenges, Anthony Coppolino, accidentally uttered classified information in a hearing in June. So the government tried to take the classified information out of the transcript without admitting they did so. After Judge Jeffrey White let EFF have a say about all this, the government ultimately decided the information wasn’t classified after all. So the Court finally released the transcript.

My wildarseguess is that this is the passage in question:

Judge Bates never ultimately held that the acquisition violated the Constitution. The problem in that case was the minimization procedures were not sufficient to protect the Fourth Amendment interests of the people of the United States.

And so he ordered that they be changed, and they were changed. And he approved them. And in addition, in the process of not only approving the minimization procedures, NSA implemented new system architecture that did a better job at assuring that those communications were minimized and ultimately destroyed, which is the goal here. It’s part of the statutory framework not to collect on U.S. citizens and when you’ve incidentally done it, destroy it. [my emphasis]

According to the John Bates opinions relating to this incident, the NSA implemented a new system of ingesting this data, marking it, checking it before it gets moved into the general repository of data, and purging it if it includes entirely domestic commuincations. But does that count as new architecture? I’m not sure.

Meanwhile, the NSA has been upgrading their architecture. We learned that (among other places) in the most recent Theresa Shea declaration on NSA systems in EFF’s Jewel case. It doesn’t mention new architecture pertaining to  upstream  702, though she does discuss a more general architecture upgrade and how it affects Section 215 specifically.

Then there’s this language, addressing the NSA’s inability to filter US person data reliably, from PCLOB.

The NSA’s acquisition of MCTs is a function of the collection devices it has designed. Based on government representations, the FISC has stated that the “NSA’s upstream Internet collection devices are generally incapable of distinguishing between transactions containing only a single discrete communication to, from, or about a tasked selector and transactions containing multiple discrete communications, not all of which are to, from, or about a tasked selector.”155 While some distinction between SCTs and MCTs can be made with respect to some communications in conducting acquisition, the government has not been able to design a filter that would acquire only the single discrete communications within transactions that contain a Section 702 selector. This is due to the constant changes in the protocols used by Internet service providers and the services provided.156 If time were frozen and the NSA built the perfect filter to acquire only single, discrete communications, that filter would be out-of-date as soon as time was restarted and a protocol changed, a new service or function was offered, or a user changed his or her settings to interact with the Internet in a different way. Conducting upstream Internet acquisition will therefore continue to result in the acquisition of some communications that are unrelated to the intended targets.

The fact that the NSA acquires Internet communications through the acquisition of Internet transactions, be they SCTs or MCTs, has implications for the technical measures, such as IP filters, that the NSA employs to prevent the intentional acquisition of wholly domestic communications. With respect to SCTs, wholly domestic communications that are routed via a foreign server for any reason are susceptible to Section 702 acquisition if the SCT contains a Section 702 tasked selector.157 With respect to MCTs, wholly domestic communications also may be embedded within Internet transactions that also contain foreign communications with a Section 702 target. The NSA’s technical means for filtering domestic communications cannot currently discover and prevent the acquisition of such MCTs.158 

The footnotes in this section all cite to John Bates’ 2011 opinion (including, probably, some language that remains redacted in the public copy, such as on page 47). So we might presume it is out of date.  Except that PCLOB has done independent work on these issues and the end of the first paragraph includes language not sourced at all.

That is, PCLOB seems to think there remain technical problems with sorting out US person data, the filtering problem cannot be solved. (Which makes the ridiculous John Bates more skeptical on this point than PCLOB.)

So do the data segregation techniques implemented in 2011 amount to new architecture? Does the larger architecture upgrade going on going to affect upstream collection in some more meaningful fashion?

I don’t know. One other reason I think this might be the language is because Coppolino was — as he frequently does — running his mouth. Bates did rule the US person data collected before 2011 violated the Fourth Amendment, even if the task before him was solely to judge whether the minimization procedures before him did. More importantly, Bates was quite clear that this US person collection was intentional, not incidental.

So Coppolino was making claims about one of the practices (the PRTT collection is another) that is most likely to help EFF win their suit, upstream collection, which actually does entail domestic wiretapping of US person content. He made a claim that suggested — with the fancy word “architecture” — that NSA had made technical fixes. But PCLOB, at least, doesn’t believe they’ve gotten to the real issue.

Who knows? It’s just a guess. What’s not a guess is that Coppolino seems to recognize upstream 702 presents a real problem in this suit.

USA Freedom Does Not Rein in the Spies

Honest. I started writing about this David Cole column asking, “Can Congress rein in the spies?” before John Brennan admitted that, contrary to his earlier assurances, his spooks actually had been spying on their Congressional overseers and also before President Obama announced that, nevertheless, he still has confidence in Brennan.

Cole’s column isn’t about the the Senate Intelligence Committee’s struggles to be able to document CIA torture, however. It’s about how Patrick Leahy introduced his version of USA Freedom Act “not a moment too soon.”

I don’t want to gripe with the column’s presentation of Leahy’s version of Freedom; with a few notable exceptions (one which I’ll get to), it accurately describes how Leahy’s bill improves on the bill the spies gutted in the House.

I first wanted to point to why Cole says Leahy’s bill comes not a moment too soon.

Leahy’s bill comes not a moment too soon. Two reports issued on Monday bring into full view the costs of a system that allows its government to conduct dragnet surveillance without specific suspicions of wrongdoing. In With Liberty to Monitor All, Human Rights Watch and the ACLU make a powerful case that mass surveillance has already had a devastating effect on journalists’ ability to monitor and report on national security measures, and on lawyers’ ability to represent victims of government overreaching. And the same day, the New America Foundation issued Surveillance Costs, a report noting the widespread economic harm to US tech companies that NSA surveillance has inflicted, as potential customers around the world take their business elsewhere.

Together, these reports make concrete the damaging effects of out-of-control surveillance, even to those with “nothing to hide.” Our democracy has long rested on a vibrant and vigorous press and open legal system. On matters of national security, journalists probably serve as a more important check on the executive than even the courts or Congress.


And, it turns out, tech companies also need to be able to promise confidentiality. Customers of Internet services or cloud computing storage programs, for example, expect and need to be certain that their messages and stored data will be private. Snowden’s revelations that the NSA has been collecting vast amounts of computer data, and has exploited vulnerabilities in corporate encryption programs, have caused many to lose confidence in the security of American tech companies in particular.

Cole describes the great costs out-of-control surveillance imposes on journalists, lawyers, and cloud providers, and implies we cannot wait to reverse those costs.

Then he embraces a bill that would not protect journalists’ conversations with whistleblowers (Leahy’s Freedom still permits the traditional access of metadata for counterintelligence purposes as well as the Internet dragnet conducted overseas) or alleged terrorists, would not protect lawyers’ discussions with their clients (the known attorney-client protected collections happened under traditional FISA, EO 12333, and possibly Section 702, none of which get changed in this bill), and would expose American companies’ clouds even further to assisted government access under the new Call Detail Record provision.

Cole does admit the bill does not address Section 702; he doesn’t mention EO 12333 at all, even though both the HRW and NAF reports did.

Senator Leahy’s bill is not a cure-all. It is primarily addressed to the collection of data within the United States, and does little to reform Section 702, the statute that authorizes the PRISM program and allows the government to collect the content of electronic communications of noncitizens abroad, even if they are communicating with US citizens here. And it says nothing about the NSA’s deeply troubling practice of inserting vulnerabilities into encryption programs that can be exploited by any hacker. It won’t, therefore, solve all the problems that the HRW and New American Foundation reports identify. But it would mark an important and consequential first step.

But he doesn’t admit the bill does little to address the specific sources of the costs identified in the two reports. It’s not a minute too soon to address these costs, he says, but then embraces a bill that doesn’t really address the actual sources of the costs identified in the reports.

That is mostly besides the point of whether Leahy’s bill is a fair apples-to-oranges trade-off with the status quo as to represent an improvement – an answer to which I can’t yet give, given some of the obvious unanswered questions about the bill. It is, however, a testament to how some of its supporters are overselling this bill and with it anyone’s ability to rein in the intelligence community.

But it’s one testament to that that bugs me most about Cole’s column. As I noted, he does mention Leahy’s failure to do anything about Section 702. Nowhere in his discussion of 702, however, does he mention that it permits warrantless access to Americans’ content, one which FBI uses when conducting mere assessments of Americans. Which of course means Cole doesn’t mention the most inexcusable part of the bill — its exemption on already soft reporting requirements to provide the numbers for how many Americans get exposed to these back door searches.

I’m not a fancy Georgetown lawyer, but I strongly believe the back door searches — conducted as they are with no notice to anyone ultimately prosecuted based off such information — are illegal, and probably unconstitutional. When retired DC Circuit Court judge Patricia Wald raised these problems with the practice, Director of National Intelligence Counsel Bob Litt simply said it would be “impracticable” to add greater oversight to back door searches. And in spite of the fact that both the President’s Review Group and PCLOB advised significant controls on this practice (which implicates the costs identified in both the HRW and NAF reports), the version of USA Freedom Act crafted by the head of the Senate Judiciary Committee — the Committee that’s supposed to ensure the government follows the law — not only doesn’t rein in the practice, but it exempts the most egregious part of the practice from the transparency applauded by people like Cole, thereby tacitly endorsing the worst part of the practice.

And all that’s before you consider that the IC also conducts back door searches of EO 12333 collected information — as first reported by me, but recently largely confirmed by John Napier Tye. And before you consider the IC’s explicit threat — issued during the passage of the Protect America Act — that if they don’t like any regulation Congress passes, they’ll just move the program to EO 12333.

The point is, Congress can’t rein in the IC, and that’s only partly because (what I expect drives the Senate’s unwillingness to deal with back door searches) many members of Congress choose not to. The have not asserted their authority over the IC, up to and including insisting that the protections for US persons under FISA Amendments Act actually get delivered.

In response to the news that Brennan’s spies had been spying on its Senate overseers, Patrick Leahy (who of course got targeted during the original PATRIOT debate with a terrorist anthrax attack) issued a statement insisting on the importance of Congressional oversight.

Congressional oversight of the executive branch, without fear of interference or intimidation, is fundamental to our Nation’s founding principle of the separation of powers.

Yet his bill — which is definitely an improvement over USA Freedumber but not clearly, in my opinion, an improvement on the status quo — tacitly endorses the notion that FBI can conduct warrantless searches on US person communications without even having real basis for an investigation.

That’s not reining in the spies. That’s blessing them.

A Good Idea that May Backfire: FISCR Fast Track

I’ve written several posts about Leahy’s USA Freedom already. To recap:

  • The bill is definitely an improvement off of USA Freedumber, though it retains “connection” chaining language I’m seriously concerned about
  • The bill permits the government to collect “bulky” collections in at least two ways: the use of IP addresses and non-individual persons (aka corporations)
  • The bill inexplicably exempts the FBI from reporting requirements on back door searches

My last new concern about the bill pertains to a measure that means well, but might backfire.

The bill includes language designed to provide for appeals of significant issues, first to the FISA Court of Review, and then to SCOTUS.

(j) REVIEW OF FISA COURT DECISIONS.—After issuing an order, a court established under subsection (a) shall certify for review to the court established under subsection (b) any question of law that the court determines warrants such review because of a need for uniformity or because consideration by the court established under subsection (b) would serve the interests of justice. Upon certification of a question of law under this paragraph, the court established under subsection (b) may give binding instructions or require the entire record to be sent up for decision of the entire matter in controversy.


(1) CERTIFICATION.—For any decision issued by the court of review established under subsection (b) approving, in whole or in part, an application by the Government under this Act, such court may certify at any time, including after a decision, a question of law to be reviewed by the Supreme Court of the United States.

(2) SPECIAL ADVOCATE BRIEFING.—Upon certification of an application under paragraph (1), the court of review established under subsection (b) may designate a special advocate to provide briefing as prescribed by the Supreme Court.

(3) REVIEW.—The Supreme Court may review any question of law certified under paragraph (1) by the court of review established under subsection (b) in the same manner as the Supreme Court reviews questions certified under section 1254(2) of title 28, United States Code.

That is, it provides a way for FISC to ask FISCR to review their work, and for FISCR to ask SCOTUS to review their work.

To some degree, the more eyes that look at these novel decisions, the better.

But neither the FISCR review nor the SCOTUS review requires even the Special Advocate. While FISCR has, in the past, permitted amici, they (and Yahoo, in the case where Yahoo appealed FISC’s 2007 recision on Protect America Act) were shooting in the dark. the new advocate, such as it exists, would be able to argue before FISCR if the court wanted it.

So to a significant extent that would result in the same people (the government and the Court’s permanent staff, on one side, and the unproven advocate on the other) arguing the same issue over and over. with the courts themselves choosing to have their own decisions certified by the higher courts.

With the potential result that you’d have appellate decisions or even a SCOTUS instruction without ever giving a real adversary a shot at the issue. If FISC responded to the phone dragnet question before the way they have since Snowden leaked details of it, they would have gotten it certified to confirm their authority.

One addition to Leahy’s bill could exacerbate that. His bill requires the FISC to consult with PCLOB on appointees as  Advocates. With today’s PCLOB, that’d be a good thing. But if Republicans win back the Senate — especially if Mitch McConnell retains his seat — you’d see another PCLOB member the likes of Elisabeth Collins Cook and Rachel Brand. Both are really smart. But both were architects of the surveillance regime while serving as DOJ Policy AAGs. Add a third of that ilk, and PCLOB could load up the Advocates corp with people like Steven Bradbury.

Moreover, for the foreseeable future, Justice John Roberts will be handpicking these judges, which doesn’t give me a lot of confidence.

I just think the Advocate system is unproven right now. It may work out, it may be gamed to reinforce the dysfunction of the court. And the record of the FISCR — especially Laurence Silberman’s efforts to rule FISA illegal in 2002 — give me no confidence this kind of self-appeal would do anything but sanction bad decisions.

Mind you, the Leahy bill also permits the government to go on denying aggrieved people of review of Section 215 collection, so it’s not clearly anyone else will get standing to challenge this program in particular.

But it seems like the FISC system is so dysfunctional, there’s no reason to pre-empt the possibility of real adversarial court function.

Update: Orin Kerr thinks this is unconstitutional.

NSA Only Finds 59% of Its Targeting of US Persons

This will be a minor point, but one that should be made.

The Privacies and Civil Liberties Oversight Board report on Section 702 included this little detail:

In 2013, the DOJ undertook a review designed to assess how often the foreignness determinations that the NSA made under the targeting procedures as described above turned out to be wrong — i.e., how often the NSA tasked a selector and subsequently realized after receiving collection from the provider that a user of the tasked selector was either a U.S. person or was located in the United States. The DOJ reviewed one year of data and determined that 0.4% of NSA’s targeting decisions resulted in the tasking of a selector that, as of the date of tasking, had a user in the United States or who was a U.S. person. As is discussed in further detail below, data from such taskings in most instances must be purged. The purpose of the review was to identify how often the NSA’s foreignness determinations proved to be incorrect. Therefore, the DOJ’s percentage does not include instances where the NSA correctly determined that a target was located outside the United States, but post-tasking, the target subsequently traveled to the United States.

0.4% of NSA’s targeting decisions falsely determine someone is a foreigner who is in fact a US person.

That’s a pretty low amount. Though based on ODNI’s number — showing 89,138 people were targeted in 2013 — that means 356 US persons get wrongly targeted each year. Again, still not a huge number, but it compares rather interestingly with the 1,144 people targeted under FISA each year. Those wrongly targeted under Section 702 actually make up 24% of those targeted in a year.

Just as interesting is comparing the NSA’s internal audit (see page 6)  with DOJ’s results. For a period presumably covering some of the same time period, NSA discovered 20 US persons tasked (for some reason there was a big increase in this number for the last quarter of the report) and 191 incidences of “other inadvertent” tasking violations, which are described as, “situations where targets were believed to be foreign but who later turn out to be U.S. persons and other incidents that do not fit into the previously identified categories” (my emphasis). Not all of those 191 incidents should be counted as wrongly targeted US persons — the description includes other inadvertent targeting. But even counting them all as such, that means NSA only found 211 of the potential wrongly targeted US persons in a year, while DOJ found 356.

Again, in a country of 310 million people, these numbers are small, particularly as compared to the collection of US person communications under upstream collection, which is thousands of times higher.

But it does say that NSA’s internal reviews don’t find all the Americans who get wrongly targeted.

Correction: I originally mistranscribed DOJ’s number as .o4%–though I had calculated using .4%.

WaPo and PCLOB Agree: NSA Does Not Comply with Its Minimization Procedures

There are a number of issues with Marc Ambinder’s interpretation of the WaPo’s analysis of the content of NSA’s 702 collections as a “bust.” Ambinder:

  • Overstates the specificity of the certifications, particularly in light of the general “foreign government” one recently revealed by WaPo
  • Makes the same email rather than overwhelmingly IM mistake Stewart Baker made
  • Doesn’t deal with the fact that the bulk of US identifiers that got minimized — the largest category, constituting over 57,000 instances — is IP address, which presents different privacy concerns than what he addresses
  • Suggests this collection includes traditional FISA warrants; WaPo suggests it is all 702 collection, which ought to mean it includes less US person content (but apparently doesn’t)
  • Ignores how readily the NSA provides unaudited access to raw data for tech personnel and SIGDEV, and therefore how (in)secure we should expect this data to be in practice

But the most troublesome problem with it is Ambinder’s treatment of the NSA’s minimization obligations and practices. Here are some statements Ambinder makes about NSA’s minimization requirements.

Ok, so: having run the data through an automatic minimization system of some sort, the NSA analysts are required to minimize every U.S.-person communication that they see. Minimize does not “to get rid of.” It means to anonymize the U.S.-based non-target source.


Maybe I could be a customer service representative from the pizza place that got his order wrong, and I’m e-mailing him to apologize for it. The NSA and the FBI are required by statute to minimize the communication if they determine it has no intelligence value. (And why would the NSA waste time reading a conversation about pizza anyway?)


The analyst’s judgment can be subjective. On the first instance, the analyst has to figure out whether the communication is relevant to a foreign intelligence purpose.

First he states that minimization does not mean “get rid of,” then states NSA is required by statute to get rid of communications that have no intelligence value, then notes an analyst has to determine whether a communication has foreign intelligence value. Overall, though, Ambinder suggests that NSA does get rid of communications involving US persons without foreign intelligence value.

Ambinder is absolutely right the law requires the government to get rid of US person data that has no foreign intelligence value.

Here’s what one version of the minimization requirements say:

(1) specific procedures, which shall be adopted by the Attorney General, that are reasonably designed in light of the purpose and technique of the particular surveillance, to minimize the acquisition and retention, and prohibit the dissemination, of nonpublicly available information concerning unconsenting United States persons consistent with the need of the United States to obtain, produce, and disseminate foreign intelligence information;

(2) procedures that require that nonpublicly available information, which is not foreign intelligence information, as defined in subsection (e)(1) of this section, shall not be disseminated in a manner that identifies any United States person, without such person’s consent, unless such person’s identity is necessary to understand foreign intelligence information or assess its importance;

(3) notwithstanding paragraphs (1) and (2), procedures that allow for the retention and dissemination of information that is evidence of a crime which has been, is being, or is about to be committed and that is to be retained or disseminated for law enforcement purposes; and

(4) notwithstanding paragraphs (1), (2), and (3), with respect to any electronic surveillance approved pursuant to section 1802 (a) of this title, procedures that require that no contents of any communication to which a United States person is a party shall be disclosed, disseminated, or used for any purpose or retained for longer than 72 hours unless a court order under section 1805 of this title is obtained or unless the Attorney General determines that the information indicates a threat of death or serious bodily harm to any person.

And here’s how that translates into the minimization procedures approved in 2011.

Personnel will exercise reasonable judgment in determining whether information acquired must be minimized and will destroy inadvertently acquired communications of or concerning a United States person at the earliest practicable point in the processing cycle at which such communication can be identified either: as clearly not relevant to the authorized purpose of the acquisition (e.g., the communication does not contain foreign intelligence information); or, as not containing evidence of a crime which may be disseminated under these procedures. Except as provided for in subsection 3(c)(2) below, such inadvertently acquired communications of or concerning a United States person may be retained no longer than five years from the expiration date of the certification authorizing the collection in any event.

Both the law and the minimization procedures approved by the FISC require NSA to get rid of US person communications that have no foreign intelligence purpose.

But here’s what the WaPo reveals about what NSA analysts do when they determine collection has no foreign intelligence value (note, however, these passages do not specify how many of these conversations include US person communications, though almost half of these communications involve US person identifiers).

Many other files, described as useless by the analysts but nonetheless retained, have a startlingly intimate, even voyeuristic quality. They tell stories of love and heartbreak, illicit sexual liaisons, mental-health crises, political and religious conversions, financial anxieties and disappointed hopes. The daily lives of more than 10,000 account holders who were not targeted are catalogued and recorded nevertheless.


“None of the hits that were received were relevant,” two Navy cryptologic technicians write in one of many summaries of nonproductive surveillance. “No additional information,” writes a civilian analyst. [my emphasis]

While these passages are not quantifiable — both because WaPo didn’t say how many files NSA had determined to be “useless” and because WaPo didn’t identify how many of those include US persons — they do suggest that NSA is not complying with the legal requirement that they destroy communications involving US persons that don’t have foreign intelligence value. Not even for communications they describe as “useless” or “not relevant.”

That’s not surprising. As I noted the other day, PCLOB found that NSA “rarely” complies with this requirement and CIA and FBI never do.

[A]lthough a communication must be “destroyed upon recognition” when an NSA analyst recognizes that it involves a U.S. person and determines that it clearly is not relevant to foreign intelligence or evidence of a crime,531 in reality this rarely happens. Nor does such purging occur at the FBI or CIA: although their minimization procedures contain age-off requirements, those procedures do not require the purging of communications upon recognition that they involve U.S. persons but contain no foreign intelligence information.

Ambinder is absolutely right that WaPo’s sample shows that NSA is pretty good, but not perfect, at masking US person identities in their data.

But both WaPo’s detailed analysis and PCLOB’s general review show that NSA does not comply with another key part of its legally required minimization obligations, to destroy communications involving US persons that have no foreign intelligence value. US person identifiers may be masked, but many of them shouldn’t be in the NSA’s databases at all. That needs to be acknowledged in any discussion of the NSA’s minimization procedures. The law requires them to get rid of US person communications with no intelligence value. But they don’t.

That’s why the sheer volume of very personal information in this sample is of concern (aside from the concern we should have for foreigners’ privacy; though again, WaPo doesn’t say how much of the US person data includes that personal information). Because the NSA and FBI and CIA can access this data without needing any suspicion of wrongdoing.

NYT Mischaracterizes PCLOB Report While Transcribing NSA Pushback to WaPo

The NYT has a story transcribing Administration efforts to “play down new disclosures” from the WaPo showing that the bulk of people whose communications were collected in a sample provided by Edward Snowden were not targets. The key claim NYT transcribes is that NSA “filters out” US person communications.

Administration officials said the agency routinely filters out the communications of Americans and information that is clearly of no intelligence value.

In addition, the NYT claims that PCLOB had no problems with the way the government minimized all this data.

Just days before the Post article, an independent federal privacy board had largely endorsed the N.S.A.’s execution of the program. The Privacy and Civil Liberties Oversight Board concluded last week that the “minimizing” of that data was largely successful, at least under the current law, which Congress passed six years ago.

Um, no.

I hope to explain this at more length, but the WaPo suggests that the government did not comply with targeting and minimization requirements in two ways: first, because the standards for foreignness were not as stringent as witnesses have claimed for a year (something which NYT’s sources apparently don’t even try to rebut). But also, WaPo showed the NSA was not destroying communications that — at least from their own and even some of the analysts’ own descriptions of it — had no foreign intelligence value. Here are some analysts judging the data collected irrelevant.

“None of the hits that were received were relevant,” two Navy cryptologic technicians write in one of many summaries of nonproductive surveillance. “No additional information,” writes a civilian analyst.

It’s this second detail NYT’s sources attempt to rebut.

But NYT’s claim that PCLOB concluded minimization “was largely successful” ignores a number of concerns they raised about it, a number of which pertain to back door searches and upstream collection.

In addition to those concerns (which about four of PCLOB’s recommendations address), PCLOB raised this issue:

Therefore, although a communication must be “destroyed upon recognition” when an NSA analyst recognizes that it involves a U.S. person and determines that it clearly is not relevant to foreign intelligence or evidence of a crime,531 in reality this rarely happens. Nor does such purging occur at the FBI or CIA: although their minimization procedures contain age-off requirements, those procedures do not require the purging of communications upon recognition that they involve U.S. persons but contain no foreign intelligence information.

A communication must be destroyed upon recognition if it’s a US person communication with no intelligence value — PCLOB restates the standard that NYT’s sources claim is actually used. But after laying out that standard, PCLOB immediately says meeting that requirement “rarely happens.”

NYT’s sources say it routinely happens. PCLOB says it rarely happens at NSA, and not at all at CIA and FBI.

PCLOB, incidentally, recommends addressing this issue by having FISC review what tasking standards are actually used and then reviewing a subset of the data returned — precisely what the WaPo just did, though we have no way of knowing if WaPo had a representative sample.

But the story here should have been, “Administration’s rebuttal has already been refuted by PCLOB’s independent review.”

PCLOB and WaPo disagree about the tasking — PCLOB sides with past Administration witnesses on the assiduousness of NSA’s targeting.

But PCLOB entirely backs WaPo on how many worthless communications NSA is keeping and documenting.

What Happened to Obama’s Ordered Restrictions on Back Door Searches?

In the wake of yesterday’s PCLOB Report, Presidential Review Board Member Geoffrey Stone reminded that Obama’s hand-picked group recommended requiring warrants before accessing US person data collected via Section 702.

In effect, the Review Group recommended that backdoor searches for communications involving American citizens should be prohibited unless the government has probable cause and a warrant. This is essentially what the recently enacted House amendment endorsed.

The Review Group concluded that the situation under section 702 is distinguishable from the situation when the government lawfully intercepts a communication when it has probable cause and a warrant. This is so because, in the section 702 situation, the government is not required to have either probable cause or a warrant to intercept the communication. Because section 702 was not intended to enable the government to intercept the communications of American citizens, because our recommended reform would leave the government free to use section 702 to obtain the types of information it was designed and intended to acquire—the communications of non-U.S. citizens, and because the recommended reform would substantially reduce the temptation the government might otherwise have to use section 702 impermissibly in an effort intentionally to intercept the communications of American citizens, we concluded that this reform was both wise and essential.

But there’s a forgotten detail from ancient history of greater interest. Even the President ordered up changes for back door searches in criminal contexts.

Specifically, I am asking the Attorney General and DNI to institute reforms that place additional restrictions on government’s ability to retain, search, and use in criminal cases, communications between Americans and foreign citizens incidentally collected under Section 702.

Yet in spite of the fact the President asked the Attorney General and DNI to place additional restrictions on the government’s ability to keep, search, and use Section 702 collected information in criminal cases, here’s what we learned yesterday.

[A]lthough a communication must be “destroyed upon recognition” when an NSA analyst recognizes that it involves a U.S. person and determines that it clearly is not relevant to foreign intelligence or evidence of a crime,531 in reality this rarely happens. Nor does such purging occur at the FBI or CIA: although their minimization procedures contain age-off requirements, those procedures do not require the purging of communications upon recognition that they involve U.S. persons but contain no foreign intelligence information.


FBI requires that metadata queries, like content queries, be reasonably designed to return foreign intelligence or evidence of a crime. As noted above, however, the FBI does not separately track which of its queries involve U.S. person identifiers, and so the number of such metadata queries is not known.

As illustrated above, rules and oversight mechanisms are in place to prevent U.S. person queries from being abused for reasons other than searching for foreign intelligence or, in the FBI’s case, for evidence of a crime. In pursuit of the agencies’ legitimate missions, however, government analysts may use queries to digitally compile the entire body of communications that have been incidentally collected under Section 702 that involve a particular U.S. person’s email address, telephone number, or other identifier, with the exception that Internet communications acquired through upstream collection may not be queried using U.S. person identifiers.540 In addition, the manner in which the FBI is employing U.S. person queries, while subject to genuine efforts at executive branch oversight, is difficult to evaluate, as is the CIA’s use of metadata queries.

And the best estimate we’ve been given for how many of these FBI queries take places is a “substantial” amount.

It has been 6 months since the President ordered changes. And the FBI still can’t even count its US person queries, much less quantify them. PCLOB calls it “difficult to evaluate.”

Um, did James Clapper and Eric Holder just blow off the President’s order in January? Because it sure looks like FBI’s back door searches remain a relatively unregulated mess.

In Advance of USA Freedom and CISA Fights, PCLOB Pretends Section 702 Doesn’t Have a Cyber Function

In a piece for Salon, I note some of the weird silences in yesterday’s PCLOB report, from things like the failure to give defendants notice (which I discussed yesterday) to the false claim that Targeting Procedures haven’t been released (they have been — by Edward Snowden). One of the most troubling silences, however, pertains to cybersecurity.

That’s especially true in one area where PCLOB inexplicably remained entirely silent. PCLOB noted in its report that, because Congress limited its mandate to counterterrorism programs, it focused primarily on those uses of Section 702. That meant a number of PCLOB’s discussions — particularly regarding “incidental collections” of Americans sucked up under Section 702 — minimized the degree to which Americans who corresponded with completely innocent foreigners could be in a government database. That said, PCLOB did admit there were other uses, and it discussed the government’s use of Section 702 to pursue weapons proliferators.

Yet PCLOB remained silent about a use of Section 702 that both Director of National Intelligence James Clapper’s office, in its very first information sheet on Section 702 released in June 2013, and multiple government witnesses at PCLOB’s own hearing on this topic in March, discussed: cybersecurity. Not only should that have been discussed because Congress is preparing to debate cybersecurity legislation that would be modeled on Section 702. But the use of Section 702 for cybersecurity presents a number of unique, and potentially more significant, privacy concerns.

And PCLOB just dodged that issue entirely, even though Section 702′s use for cybersecurity is unclassified.

In the transcript of the March PCLOB hearing on Section 702 uses, the word “cyber” shows up 12 times. Four of those references come from DOJ’s Deputy Assistant Attorney General Brad Wiegmann’s description of the kinds of foreign intelligence uses targeted under Section 702. (The other references came from Information Technology Industry Council President Dean Garfield.)

MR. WIEGMANN: You task a selector. So you’re identifying, that’s when you take that selector to the company and say this one’s been approved. You’ve concluded that it is, does belong to a non-U.S. person overseas, a terrorist, or a proliferator, or a cyber person, right, whoever it is, and then we go to the company and get the information.


It’s aimed at only those people who are foreign intelligence targets and you have reason to believe that going up on that account that I mentioned, bad guy at Google.com is going to give you back information, information that is foreign intelligence, like on cyber threats, on terrorists, on proliferation, whatever it might be.


So in other words, if I need to, if it’s Joe Smith and his name is necessary if I’m passing it to that foreign government and it’s key that they understand that it’s Joe Smith because that’s relevant to understanding what the threat is, or what the information is, let’s say he’s a cyber, malicious cyber hacker or whatever, and it was key to know the information, then you might pass Joe Smith’s name.

Yesterday’s report, however, doesn’t mention “cyber” a single time. Indeed, it seems to go out of its way to avoid mentioning it.

As discussed elsewhere in this Report, the Board believes that the Section 702 program significantly aids the government’s efforts to prevent terrorism, as well as to combat weapons proliferation and gather foreign intelligence for other purposes.


The Section 702 program, for instance, is also used for surveillance aimed at countering the efforts of proliferators of weapons of mass destruction.473 Given that these other foreign intelligence purposes of the program are not strictly within the Board’s mandate, we have not scrutinized the effectiveness of Section 702 in contributing to those other purposes with the same rigor that we have applied in assessing the program’s contribution to counterterrorism. Nevertheless, we have come to learn how the program is used for these other purposes, including, for example, specific ways in which it has been used to combat weapons proliferation and the degree to which the program supports the government’s efforts to gather foreign intelligence for the benefit of policymakers.

It’s footnote to that last section cites DOJ’s 2012 report to SSCI on the uses of Section 702 (which doesn’t mention cyber) rather than the information sheet released in June 2013, which does.

I find PCLOB’s silence about the use of Section 702 to pursue cyber targets particularly interesting for several reasons.

First, because cyber targets pose unique privacy threats — in part because cyberattackers are more likely to hide their location and exploit the communications of entirely innocent people, meaning Section 702′s claimed targeting limits offer no protection to Americans. Additionally, targeting (as Wiegmann describes it) a “malicious cyber hacker” goes beyond any traditional definition of foreign agent; it is telling he didn’t use a Chinese military hacker as his example instead! Indeed, while proliferation (along with foreign governments, the other presumed certification) is solidly within FISA Amendment Act’s definition of foreign intelligence, cybersecurity is not. In its discussion of back door searches, PCLOB admits there are concerns raised by back door searches that are heightened (or perhaps more sensitive, because they involve affluent white people) outside the counterterrorism context, that’s especially true for cybersecurity targeting.

Consider, too, the likelihood that cyber collection is among the categories of about collection that PCLOB obliquely mentions but doesn’t describe due to classification.

Although we cannot discuss the details in an unclassified public report, the moniker “about” collection describes a number of distinct scenarios, which the government has in the past characterized as different “categories” of “about” collection. These categories are not predetermined limits that confine what the government acquires; rather, they are merely ways of describing the different forms of communications that are neither to nor from a tasked selector but nevertheless are collected because they contain the selector somewhere within them.

At the beginning of the report, PCLOB repeated the government’s claim this is primarily about emails; here in the guts of it, it obliquely references other categories of collection, without really considering whether these categories present different privacy concerns.

Remember, too, that the original, good version of USA Freedom Act remains before the Senate Judiciary Committee. That bill would disallow the use of upstream 702 for any use but counterterrorism and counterproliferation. Did PCLOB ignore this use of Section 702 just to avoid alerting Senators who haven’t been briefed on it that it exists?

Finally, I also find PCLOB’s silence about NSA’s admitted use of Section 702 to pursue cyberattackers curious given that, after Congress largely ditched ideas to involve PCLOB in various NSA oversight — such as providing it a role in the FISA Advocate position — Dianne Feinstein’s Cyber Information Sharing Act all of a sudden has found a use for PCLOB again (serving a function, I should add, that arguably replaces FISC review).

(1) BIENNIAL REPORT FROM PRIVACY AND CIVIL LIBERTIES OVERSIGHT BOARD.—Not later than 1 year after the date of the enactment of this Act and not less frequently than once every 2 years thereafter, the Privacy and Civil Liberties Oversight Board shall submit to Congress and the President a report providing—

(A) an assessment of the privacy and civil liberties impact of the type of activities carried out under this Act; and

(B) an assessment of the sufficiency of the policies, procedures, and guidelines established pursuant to section 5 in addressing privacy and civil liberties concerns.

Feinstein introduced this bill on June 17, several weeks after PCLOB briefed her staffers on their report (they briefed Congressional committee aides on June 2, and the White House on June 17 — see just after 9:00).

A renewed openness to expanding PCLOB’s role may be entirely unmotivated, or it may stem from PCLOB’s chastened analysis of the legal issues surrounding Section 702.

But I do find it interesting that PCLOB uttered, literally, not one word about the topic that, if DiFi’s bill passes, would expand their mandate.

PCLOB Ignores Glaring Section 702 Non-Compliance: Notice to Defendants

I will have far more to say about PCLOB once I finish my working thread. But there’s one glaring flaw in the report’s claim that the government complies with the statute.

Based on the information that the Board has reviewed, the government’s PRISM collection complies with the structural requirements of the statute.

But here’s the report’s discussion of what happens with aggrieved persons — those prosecuted based in information derived from Section 702 information.

Further, FISA provides special protections in connection with legal proceedings, under which an aggrieved person — a term that includes non-U.S. persons — is required to be notified prior to the disclosure or use of any Section 702–related information in any federal or state court.447 The aggrieved person may then move to suppress the evidence on the grounds that it was unlawfully acquired and/or was not in conformity with the authorizing Section 702 certification.448 Determinations regarding whether the Section 702 acquisition was lawful and authorized are made by a United States District Court, which has the authority to suppress any evidence that was unlawfully obtained or derived.449 

But for 5 years after the passage of the law, the government never once gave defendants notice they were aggrieved under Section 702. It lied to the Supreme Court about not having done so. And even while it has since given a limited number of defendants — like Mohamed Osman Mohamud — notice, there are others — David Headley, Najibullah Zazi and Adis Medunjanin, and Khalid Ouazzani — who are known to be aggrieved under Section 702 who have never received notice. Finally, there is the case of the Qazi brothers, which seems to be a case where the government is parallel constructing right in the face of the magistrate.

PCLOB said that the government is generally in compliance with the statute. And yet, it made no mention of known, fairly egregious violations of the statute.

That suggests the report as a whole may be flawed.

Emptywheel Twitterverse
emptywheel RT @JustADCohen: Story merits more coverage: Energy Dept. inspector blocked from probing dismissal of Hanford engineer http://t.co/p521yIKi
emptywheel @9Joe9 Harder to track TV than YouTube. Tho NSA will have fun once everyone's on Internet TVs.
emptywheel @yaelwrites Yup. Cause torturing Muslims is more socially acceptable than torturing minks.
JimWhiteGNV RT @cherylrofer: So now that the nurse in New Jersey has tested free of Ebola, she's free to go, right? I guess not. Science, people! http:…
emptywheel @9Joe9 Nah. Just piggybacking off their comments about hatchet man.
emptywheel @armandodkos Will you watch for me in case we get close to 17?
emptywheel If NYPD is going to start tracking everyone who watched lots of ISIL beheadings, does that mean all CNN viewers under surveillance?
JimWhiteGNV RT @DrJudyStone: Must read:"I have been quarantined in NJ...The US must treat returning #HCW w dignity and humanity,” not as criminals http…
JimWhiteGNV @libdemwasjailed And then there was Friday near Seattle.... #FuckTheNRA
JimWhiteGNV RT @lrozen: Ron Klain should read this. 3 hours passed. No one in charge. http://t.co/Pbg7bDaVJy
JimWhiteGNV RT @lrozen: "Eight police cars escorted me to the University Hospital in Newark. Sirens blared, lights flashed." http://t.co/Pbg7bDaVJy #in
bmaz @SarebrumSarah Trust me, the dog does not lead an overly taxing life....
October 2014
« Sep