Throughout the past year of covering the shoddy journalism orbiting the NSA story, it never ceases to amaze that so many award-winning journalists, some veterans, some not-so-much, are unable to adequately explain things. Decide for yourself whether this murky reporting is deliberate or symptomatic of the complexities of this issue. There’s no real way of knowing why. All we know is that even with access to a former NSA systems administrator, Snowden, one article after another is loaded with massive holes, which are only clarified 24-hours later (or more), after it’s too late.
Such is the case once again with the latest article from The Washington Post‘s Barton Gellman. Admittedly, Gellman has been one of the better, more responsible reporters with access to the Snowden files, but his latest article, co-written by Ashkan Soltani and Julie Tate, is a journalistic mess, to put it mildly.
Before we get into how much of a mess it is, let’s recap what we covered yesterday.
–Gellman reported that 160,000-plus intercepted communications between 2009 and 2012 contained 65,000 “minimized” (hidden or encrypted) incidental references to people who might be Americans not targeted by NSA. There were also 900 references to email addresses that were apparently not minimized, which Gellman and his team ascertained were “strongly linked” to U.S. persons.
–Meanwhile, the communications showed that NSA was doing its job. It had acquired “fresh revelations about a secret overseas nuclear project, double-dealing by an ostensible ally, a military calamity that befell an unfriendly power, and the identities of aggressive intruders into U.S. computer networks.” Additionally, the files showed that NSA provided information leading to the arrest of at least two terrorism suspects.
–Later in the article, Gellman wrote that some of the minimization wasn’t entirely discreet. He cited “minimized U.S. president-elect” from early 2009, and “minimized U.S. president” in communications from later years.
This article, as most of them do, incited the usual outrage addicts, who began flailing around the internet about the evil U.S. surveillance state. It’s what they do. It’s their brand. The incomplete nature of the reporting, whether from Gellman or Greenwald or the untold dozens of other reporters who somehow received copies of the Snowden documents, is irrelevant, as long as the over-arching hyperbole of the article jibes with their ongoing daily narrative.
How incomplete was the article?
1) Spying on the president? Speaking of narratives, let’s start with the last part first: the thing about the “minimized U.S. president,” and how one of the aforementioned outrage addicts initially reacted to the news. Here’s the former reporter for The Intercept, Marcy Wheeler:
WaPo story makes it clear Obama's communications were collected between his election and inauguration.http://t.co/dZShBk6xob
— emptywheel (@emptywheel) July 6, 2014
While I strongly disagree with Marcy’s take on a lot of things in the general Greenwald sphere of interest, she’s not a stupid person. Yet she got this totally wrong and had to retract it later. It’s safe to say that’s it’s partly because her agenda informed her reaction, but also considerable blame has to go to Gellman for incomplete, spotty reporting. It wasn’t until later in the day, via his Twitter feed that Gellman clarified this rather murky item.
Lotta questions on this. The 1200 references to “minimized US president” come when people talk about him in intercepted conversations.
— Barton Gellman (@bartongellman) July 6, 2014
Everybody: If NSA spied on Obama I'd say so in 1st half of the story 😉
MIN. US PRESIDENT = others saying his name. http://t.co/zi7vBmjIPR
— Barton Gellman (@bartongellman) July 7, 2014
So, no, there’s no evidence in the Snowden documents — at least the ones in this article — that NSA was intercepting the president’s emails. Instead, there were around 1,200 references to the president in emails written by other people. Imagine that. People talking about the president. Crazy.
2) 65,000 individual people, or 65,000 references? Another point of clarification offered by Gellman in his tweet about the president is that, no, there weren’t 65,000 individuals whose information was inadvertently collected then minimized. The fact that the president alone made up 1,200 of the 65,000 references shows that individuals could’ve appeared multiple times — who knows, perhaps hundreds or thousands of times. Gellman isn’t clear on that either. If NSA had been targeting me when I first met my wife, they’d find literally hundreds of references to her alone, from just one target: me. It’s clear now, following Gellman’s tweet, why the 65,000 number is deceptive.
3) What about the 900? This might be the murkiest of the murky information in the article. Again, Gellman wrote: “The Post found nearly 900 additional e-mail addresses, unmasked in the files, that could be strongly linked to U.S. citizens or U.S. residents.” How are they strongly linked? What indicators led him to conclude this? Later, again on Twitter, Gellman admitted that the records pertaining to the 900 email addresses were “imperfect.”
.@AntonSirius 1. More nontargets than targets 2. Nearly half of files have NSA-ID’d USPs 3. Records say 900 more USPs but records imperfect.
— Barton Gellman (@bartongellman) July 7, 2014
Okay, what the hell does that mean? Imperfect? And why are we only hearing about this on Twitter? (We’ll come back to this point in the conclusion.) Just by way of a review, here’s what the Privacy and Civil Liberties Oversight Board (PCLOB) reported about PRISM and the FISA Amendment Act’s Section 702, through which Gellman’s data was gathered:
While the scope of targeting under Section 702 is broad, that targeting cannot include U.S. persons or people located in the United States. As a result, this program does not allow the government to gain comprehensive access to any U.S. person’s communications: the government will not be able to hear every telephone call a U.S. person makes, for instance, or collect every email sent or received by that person.
PCLOB also reported:
In 2013, the vast majority of the intelligence reports disseminated by the NSA that were based on intelligence derived from Section 702 contained no reference to any U.S. person.
Also, why weren’t these 900 references minimized? Does Gellman know? Is there a legal justification for it? Is there any context whatsoever here? Gellman seems to just toss information into his article without elaboration, only to later offer passing tweets, maybe.
The Post article says that MOST of those communications were minimized, according to law. But 9,000 or so identifiable U.S. person terms/selectors/account holders were not. This means that, in the above example, maybe the NSA minimized my e-mail address but not my telephone number. Why might the NSA not have minimized this? Because doing so would require an analyst to look at the content of every communication and run every selector (telephone number and e-mail address) through a database to figure out if it should be minimized. This would be inordinately time-intensive, so the NSA relies on automation, and THEN, when the analyst IS looking at specific communication, the analyst is required to minimize any un-minimized selector that makes it through.
4) What are these documents anyway? Gellman reported in the body of his article that the documents are “a large cache of intercepted conversations.” But a graphical illustration linked from the main article labels the documents as being “surveillance reports” and not raw emails and text messages. Once again, which is it? We simply don’t know. Furthermore, the law disallows the distribution of incidentally collected communications. In other words, the inclusion of these untargeted or incidental communications in a series of “surveillance reports” is illegal. That means either NSA broke the law, a violation which is required to be reported to Congress, or we’re simply not being given the whole context about the documents stolen by Snowden and reported by Gellman. Regarding the latter, it could be that all of the communications included in the reports were collected from targets. Such context, detailed clearly in the article, would completely dull the impact of the story.
In the final analysis, it’s shocking to see yet another article in a major publication that’s loaded with so many holes and with very little by way of context. Given the extent of Snowden’s documents and given Gellman’s access to Snowden and the intelligence community, not to mention a legion of editors, this kind of shoddy reporting should never occur.
What’s even more shocking is that clarifications are only going out via Twitter and oftentimes days after the article has circulated through the internet, with its “facts” seeping into popular wisdom. How many people who read the article are also following Gellman’s Twitter account? And should this be even necessary?
Given how Snowden and his supporters, both his reporters and readers alike, have continuously preached about the importance of this story, it continues to confound reason why the reporting has been so awkward and slapdash. You’d think the goal would be to nail this information and its context in the most concise and clear way possible. Instead, it’s been exactly the opposite. Articles, with their hyperbolic headlines and ledes, have been rushed to press and, nearly every time, the 24-hour rule has changed significant chunks of the reporting. Why? It wouldn’t be unreasonable to conclude that either the reporters are incompetent, or they don’t fully understand the topic, or they’re being deliberately coy. Any of these options are unforgivable.