My review of Edward Jay Epstein’s Snowden book

(originally posted 06.02.2017 on Google+)


I feel a bit ambivalent about the book. On the one side, it is really great. It is clearly the most complete and detailed description of the many, many inconsistencies in Snowden's own narrative, which was uncritically repeated and stated as fact by most mainstream media.
But on the other side, I'm a bit disappointed. First, I didn't really read anything new. Most of what he says is something I write since months or years. Second, Epstein obviously lacks the technical understanding (no offense intended, considering he is over 80), this leaded to some mistakes, that he exaggerates the technical capabilities of intelligence agencies (almost the same way Snowden and his fanboys do ...), and that he obviously didn't realize the many, many times Snowden lied about NSA programs and tools. And third, I find his snarky, sexist remarks toward Lindsay Mills and especially Sarah Harrison inappropriate.

Here more detailed these points …


Epstein's book is really great in describing the many, many points where the Snowden narrative simply doesn't add up, for example:

- Snowden introduced himself as a senior analyst, while in reality he was just a low-level technician; only the last 2 months he was "analyst-in-training".

- Why did Snowden first escape to Hong Kong (China), and not for example to Brazil?

- What did Snowden do while his first 11 days in Hong Kong? Not a single trace from him seems to exist for this time.

- Why did Snowden visit the Russian embassy in Hong Kong several times?

- Epstein compellingly explains that Snowden's passport was not revoked while he was in the airplane, but when he still was in Hong Kong.

- Epstein asks why the Russian airline Aeroflot allowed Snowden to fly with a revoked passport (but what is still unclear for me: Was Aeroflot able to do an online check about the validity of the passport? If not, why was Aeroflot -- remember Snowden’s alleged planned flight to Havana was with Aeroflot too -- suddenly able to check this in Moscow, just 12 hours later?)

- Why did nobody see Snowden at Sheremetyevo airport, where he was allegedly "trapped" for more than a month? Epstein was there and concludes that it is absurd.

- What happened with majority of documents that Snowden didn't hand over to journalists? Why did Snowden steal them in the first place? Documents, which had nothing to do with civil liberties or wrongdoings, but just contain military and other state secrets.

Epstein could have mentioned more strange inconsistencies in the Snowden narrative, for example:

- Snowden often claimed that Clapper’s “lie” (I do it in brackets because the question whether it was a lie or not is much more complex than Snowden and his supporters pretend, but that’s not the question here) before Congress was the “breaking point” in his decision to go public. But why didn’t he say this while his infamous interview with Greenwald in Hong Kong? Instead it took until January 2014 before he first publicly used this excuse. Sounds very implausible.

- Why doesn’t Snowden agree to be interviewed by a critical journalist? And even on Twitter, Snowden didn’t answer a single critical question so far. His hardest interview he ever had was with comedian John Oliver. That’s ridiculous.

Epstein describes and explains these many inconsistencies in the Snowden narrative very detailed and compelling. And his conclusion is that it cannot have happened like Snowden tries to make us believe. And then he asked the logical question how it likely really happened. Critics of Epstein often stress that he uses phrases like “presume”, “assume”, or “might have”, but what else should he do? We have the situation that the only known witness -- Snowden himself -- is absolutely unreliable and was caught lying several times, so it is of course a lot of speculation when someone tries to find out the truth. And Epstein clearly states that much is speculation.

And Epstein does not -- like many of his critics now say -- claim that Snowden was a Russian operation from the very beginning. Epstein just says that there is no serious doubt that Snowden handed over material to Russian intelligence (as further evidence he presents his own interview with Snowden’s lawyer Kucherna and the remarks from Klintsevich), so that it is only the question when Snowden’s collaboration with Russian intelligence began -- in Moscow, in Hong Kong, or earlier --, not if. And even if there is no 100% solid-proof evidence for this claim, Epstein is right. It is more than naive to trust in Snowden’s claims and believe that he irrevocably destroyed the documents before he headed to Russia.

Another evidence is that Epstein assumes (like I do since these stories emerged), that many leaked top secret documents (like Merkelphone, the ANT catalog. XKeyScore sources, Wikileaks intercepts or the drone papers -- all these documents are “Snowden-like” and none is dated after Snowden) that were not credited to Snowden but most likely come from the trove he removed anyway. These publications are a clear sign that there are more Snowden documents than the ones he handed over to journalists in Hong Kong, and that these documents are in possession of someone who is currently still using them.

It is clear that Snowden fanboys don’t like this kind of analysis, but it’s crystal clear that the only question is when Snowden defected to Russia, and Epstein does a really great job here to figure out the most likely possibilities. Maybe some day we'll know for sure.


Now to the downsides of the book.

First thing is, that I didn’t read anything really new. While reading the book I never thought something like “oh, I didn’t know this” or “oh, I didn’t realize this”. It may sound arrogant, but almost everything Epstein wrote I said before, like in my “Snowden-FAQ”[1]. OK, Epstein interviewed Kucherna. And he went to Sheremetyevo airport to verify that it’s absurd Snowden was “trapped” in the transit zone for more than a month without anyone seeing him. Or he checked Aeroflot policies to verify that it’s absurd they allowed Snowden to fly with a revoked passport from Hong Kong to Moscow but not from Moscow to Havana. And he interviewed a lot of other people (often anonymous), but the answers are almost always only confirmations of facts that were already obvious. But I don’t really blame Epstein here, again his book is great in this regard, maybe it’s just me that I expected too much.

The second thing is that Epstein obviously lacks the necessary technical knowledge to really understand NSA’s -- or intelligence agencies in common -- abilities and limitations. Again, I don’t really want to blame Epstein here, given his age, and that it is very complex, even for technically very skilled people. But it’s a bit a pity, because this way his otherwise great book is missing something, as I will explain.

But first another related remark. I read that some critics of Epstein also stress that he got some technical points wrong. For example, Glenn Greenwald and Barton Gellman are two of these critics. This is so insane and ridiculous that I almost run out of words. The ones who wrote completely wrong articles about PRISM, claiming the NSA had a “direct access” to Google & Co. and that the companies were “partners” of the NSA. Both turned out to be completely imaginary, and it happened because Greenwald and Gellman misunderstood and misinterpreted the documents. Technical knowledge, haha. And additionally, Greenwald heavily misinterpreted the BOUNDLESSINFORMANT documents, resulting in his false claims that the NSA intercepts and collects data on large scale in and about European countries like Germany, France and The Netherlands. And not only this, the whole Snowden reporting was full of wrong and misleading “revelations”[2], almost always caused by the fact that journalists were unable to correctly read and interpret the documents. Compared to these heavy mistakes Epstein’s are negligible disaccords.

To illustrate what I mean here a quote from the book, where Epstein refers to the so called “702 collection”, or sometimes better known as PRISM and Upstream:

”Snowden did vastly more damage by revealing the PRISM program, also called "702" of FISA. Its effectiveness proceeded from the misplaced confidence that terrorist organizations [...] had in the encryption and other safeguards used by giant Internet companies, such as Apple, Google, Twitter, and WhatsApp. They evidently had not known that in 2007 the NSA found a way to intercept this data before it was encrypted. [...] So by 2013 the NSA was able to access 91 percent of the Internet before it was encrypted. This so-called upstream data included Google searches, tweets on Twitter, [...]”

This is wrong on so many layers, that it’s even hard to find a start. So let’s start with ”the NSA was able to access 91 percent of the Internet”. That’s plain wrong. The NSA is not able to access “91 percent of the Internet”. Where comes the number “91 percent” from? Most likely from the PCLOB 702 report[3], where you can find:

”As of mid-2011, 91 percent of the Internet communications that the NSA acquired each year were obtained through PRISM collection.”

So this means that 91% of NSA's internet collection (for the sake of simplicity, I don't differ here between "collection" and "acquisition") is from PRISM, not that the NSA is able to access "91 percent of the Internet".

Epstein's next misunderstanding is that this is not the "so-called upstream collection", but PRISM. He totally confuses Upstream with PRISM.

The next thing is that it is the wrong description of PRISM to say "to access before it was encrypted". PRISM is collection from stored data on servers, while Upstream is collection of raw internet data on the wire. It can be (and often is, but details are tricky and beyond this review) the case that data on servers is stored as cleartext while the same data is transmitted over the Internet encrypted, but that's not the key difference between PRISM and Upstream (see my technical description of the differences between PRISM and Upstream[4]).

Let's go on: "that in 2007 the NSA found a way to intercept this data before it was encrypted" is just another inaccuracy. 2007 was not a year in which the NSA discovered something magic. 2007 was the year in which Congress passed the "Protect America Act", the legal base with which the U.S. government could compel American internet companies to (without an individual warrant) hand over user data if the target was reasonably believed to be located outside of the U.S. (its successor, the "FISA Amendment Act of 2008", is still in force). And related: "Its effectiveness proceeded from the misplaced confidence that terrorist organizations [...] had in the encryption and other safeguards used by giant Internet companies, such as Apple, Google, Twitter, and WhatsApp" is a very far-fetched claim. Which "encryption and other safeguards" should this be, especially when we speak about pre-Snowden? Back then, Google was the only major American email provider that even supported STARTTLS. And it was known even before 2007 that for example the FBI could access Gmail data when they have an individual warrant. And section 702 was worded very clearly (contrary to section 215), there was no doubt that NSA can demand data from Google & Co. The only thing that actually happened with Snowden's "revelations" was that terrorists and other bad guys now could know that the NSA actually did it. But that terrorists before had "confidence in the encryption and other safeguards" is really far, far-fetched, if not plain wrong.

All these inaccuracies show that Epstein doesn't really understand PRISM and Upstream. This itself isn't that bad, it is complex, and a lot of others, including Snowden reporters like Greenwald and Gellman, and Snowden himself, don't understand it either. But regarding Epstein's book, it's really a pity, because his misunderstanding leads to two flaws: First he is misled with a few assumptions, and second the book is incomplete.

Epstein is misled because he obviously believes in the various media reports that depict the NSA as an almighty agency, that is able -- at least technically -- to monitor everyone they want, to hack everyone they want, to collect any information they want. You can see this in the part of the book were Epstein simply assumes that people like Appelbaum were successfully monitored by Russian intelligence. Epstein seems to think that when the NSA is almighty, then Russian intelligence is no less. But this is simply not true. In reality hacking is very hard and risky (you can get an idea of it when you read my explanation why Hillary Clinton's email server was most likely not hacked[5]), and it is very unlikely that someone with the technical skills and paranoia like Appelbaum was successfully monitored by any intelligence agency.

The second thing is that Epstein's book is unfortunately incomplete. He correctly described that Snowden's narrative is contradictory and is most likely almost completely lied. And he points out that the NSA was incorrectly depicted as a rogue agency out-of-control, contrary that in reality the NSA is a law-abiding, bureaucratic agency under strict oversight. But Epstein misses another, equally important point: That the "revelations" were also technically wrong and misleading, that Snowden didn't only lie about the circumstances of his crime and the NSA's legal restraints -- Snowden did lie about the NSA's tools and programs as well.


My last critical observation are the snarky and sexist remarks against Lindsay Mills and Sarah Harrison. I really felt uncomfortable reading this parts. This was completely unnecessary and almost insane. He should have gone without this. And it showed that Epstein has a basic misconception about privacy (ironically almost the same that a lot of Snowden fans have ...).

I mean, it is a book about Snowden and his lies and shady motives. So it is absolutely ok when Epstein stresses that Snowden liked to make and publish revealing pictures of himself. This may hint to Snowden's narcissistic character, and may help to understand why he did what he did. But why did Epstein do the same with Mills? She is completely unrelated here. Epstein says one reason, and this is telling: He asks how someone who, like Snowden, publishes "sexy" pictures of himself, and encourages his girlfriend to do even sexier pictures, how can such people be so concerned about privacy? But that's easy: Even pornstars likely don't want that their phones are tapped or their emails read. Every adult person has the right to decide which parts of his private life can go public and which has to stay private. So here Snowden isn't contradictory (as I said, a lot of Snowden fans have a quite similar misconception, I often read from them that "NSA-fans" like me would have no right for privacy, because I would support that the NSA violates other people's privacy).

Epstein's attacks against Harrison were even worse. He almost depicts her as a hussy. Again, that's not my cup of tea, and isn't helpful in any way to understand the Snowden operation. The book would have been much better without this.




[1] https://plus.google.com/+RolfWeber/posts/Ry26xSVRVo7 
[2] https://plus.google.com/+RolfWeber/posts/jmKrp4F75sg 
[3] https://www.pclob.gov/library/702-Report.pdf 
[4] https://plus.google.com/+RolfWeber/posts/bkvXuB9DfXJ 
[5] https://plus.google.com/+RolfWeber/posts/JejVLNzvoDX