5 ms·
> I'm saying that I wouldn't be surprised if something happened while he was in Hawaii that inspired him to leave with a bang. This is a fair point, especially
by scintill76 12y ago
> I'm saying that I wouldn't be surprised if something happened while he was in Hawaii that inspired him to leave with a bang.
This is a fair point, especially in light of how his CIA stint ended. I started considering it myself, but didn't think much of it because it seems like the NSA & co. would be gleefully trumpeting such a thing if they knew it. For example, they said one of the first things he stole was a technical employment test with answers, and implied he used it to cheat (in my link from the last post.) Can you think of such an incident that the NSA wouldn't know about, or wouldn't reveal publicly?
And since you are arguing that he both a) leaked because he was disgruntled about something at work, and b) committed to leak months before changing jobs, the possible scenarios get even narrower, as they would have to carry over between jobs. Does he just hate the entire current US intelligence sector in general? Well, he's not keeping that a secret, so it shouldn't really be a mark against his credibility.
Now, just because I can't think of a reason he wanted to "leave with a bang" that we wouldn't know about, doesn't mean it can't be true, but it deflates that theory quite a bit IMO.
> If his intent was to blow the whistle on something, he would have just grabbed the few documents he needed and gone to press with them.
But his claim (or a popular simplification of it) is that the US is violating the rights of everyone, all the time, for any reason they feel like. He's not blowing the whistle on one office or something, it's a whole multi-billion dollar industry run by major branches of the US Government. If he just brought every document about PRISM, or just the Verizon metadata arrangement, or just personally hacking sysadmins at telecoms, or just tapping Google's private fiber lines, or just tapping Yahoo's private fiber lines, it would not be near as powerful as showing they are doing all of the above, and more! I don't know that he ever planned this, but there is also a practicality argument that it might be easier to "collect it all" first, flee, and sift through it for a few weeks full-time from the comfort of Hong Kong.
> What you left out from the paragraph that quote was pulled from
OK, and you left out where I introduced that quote with "On a different altercation, ... " Yes, I probably took too much license editing it, but it was meant as a quick allusion to his claims to have raised concerns through the "proper channels" at NSA and been ignored. In the full, more accurate context, it shows that on that occasion, he went to some lengths to work within the system to raise a concern (reading between the lines, it sounds like his boss eventually helped him to inject some Javascript tags into Snowden's personnel file, as a proof-of-concept for the vulnerability), and got nothing but trouble for trying to help in the way that was supposedly approved ("His immediate supervisor signed off on it" from your link).
> getting the banker drunk in Geneva to recruit him as a CIA asset
Yes, I am shocked the CIA didn't jump in and confirm or deny that this happened.</sarcasm> What proof would you expect? Granted, if it's not (dis)provable it's of little worth to us, but it's not really meant to be something we're persuaded to believe, just an anecdote about a formative experience. I don't know, is it inherently disingenuous to tell a story you can't prove, with the purpose of explaining your worldview?
> I find it hard to believe that he walked out the door with thousands upon thousands of documents, but forgot to bring a copy of his own e-mails that would corroborate his story.
First, that article says NBC sent a FOIA, but the single email shown was not received from FOIA. I'm not sure it matters, though.
How would anyone but the NSA and Snowden know with some certainty that any emails he did produce were legitimate? Would they admit to us if they were legitimate? (Tip: commit hashes of your proof to the Bitcoin blockchain early, so you can at least prove you were thinking about something before a certain date.) Hmm, I guess this goes for all the documents' authenticity, but I can imagine it's easier to confirm those than emails. Maybe he didn't really care about corroborating his story, or wasn't planning "a story" that far in advance. It's kind of funny that this is such a point of contention. Is anyone really saying the NSA would have stopped if they knew some sysadmin named Ed Snowden didn't like it? I guess alone it's pretty small, but it's another of several doubts about his credibility, that's why it's so hotly contested.
I agree that his claim about evaluating every document is suspect, and several of the other items you mentioned.
- csandreasen 12y ago> it seems like the NSA & co. would be gleefully trumpeting such a thing if they knew it. I don't think so for two reasons: 1) they don't talk much about anything unless they absolutely have to for classification and vetting reasons (this gets into topics like overclassification and whether or not things are really necessary for national security - but that's a different debate). For example, the op-ed that Michael Hayden wrote[1] in the USA Today in response to the 'NSA audit reveals thousands of violations' articles was apparently originally going to be written by the NSA itself, but the declassification and vetting process prevented a timely, point-by-point response[2]. I'm surprised that the e-mail was released so quickly. Overall, the NSA's response has been bumbling, and I think that has a lot to do with existing public relations policy that doesn't fit well with a fast-moving media. 2) There's a pending court case, and it's usually good policy to stay mum about anything regarding any pending cases until afterwards. I don't know that it's necessarily good policy now, as Snowden is probably never going to come back and face trial, but he is still indicted. > Does he just hate the entire current US intelligence sector in general? Well, he's not keeping that a secret, so it shouldn't really be a mark against his credibility. No, it's an indication that we should treat the leaks may have a skewed perspective, and we should keep that in mind when evaluating them. The credibility issue comes into place when he gets on national television and tells the American people “I reported that there were real problems with the way the NSA was interpreting its legal authorities. And I went even further in this, to say that they could be unconstitutional, that they were sort of abrogating our model of government in a way that empowered presidents to override our statutory laws. And this was made very clear. And the response was, more or less, in bureaucratic language, was, ‘You should stop asking questions.’"[3] The e-mail he sent did in fact ask a question about legal authorities, but nothing like what he described to the public; there were no concerns raised at all. Even his description of the response was a lie: instead of 'you should stop asking questions' the guy said 'Please give me a call if you'd like to discuss further'. On top of that, the e-mail was sent months after he first contacted Greenwald and Poitras - he was almost on his way out the door to Hong Kong at that point. That's the kind of stuff that speaks to his credibility. When you know that he hates the NSA and will make bald-faced lies on television should raise some flags when evaluating the leaks. This is getting away from the Snowden credibility issue and more to issues with the reporting, but ... > If he just brought every document about PRISM, ... for which all of the reporting turned out to be wrong. They blew a massive intelligence source over perceived mass civil liberties violations when the truth ended up being 'We can get content on specific accounts from Google/Yahoo/etc. when we present them with a court order.' > or just the Verizon metadata arrangement, ... I probably wouldn't have had so much of a problem if he released that one and stopped there. Even so, the details from initial story still didn't match reality. On top of that, the program has debated at length in public and in Congress since it was initially disclosed back in 2005[4]. > or just personally hacking sysadmins at telecoms, 1) I don't care that GCHQ hacked Belgacom - it does not affect me at all; 2) we still lack a lot of context behind this - Why did GCHQ break in? What information did they obtain? Where they using this access to spy on legitimate intelligence targets, or something we would object to? The story was presented simply as 'they hacked a Belgian telecommunications company - be outraged!' > or just tapping Google's private fiber lines, or just tapping Yahoo's private fiber lines, I'm pretty sure Google and Yahoo didn't lay their own transcontinental undersea fiber optic lines. That's the nature of the internet - at some point your communications pass over lines owned by someone else. I'm a lot more angry at Google for not bothering to encrypt their transnational data than I am at GCHQ for intercepting it - that means the intelligence services of EVERY country their traffic passed through had access to it. > it would not be near as powerful as showing they are doing all of the above, and more! I agree with you there! [1] http://www.usatoday.com/story/opinion/2013/08/18/nsa-privacy-surveillance-editorials-debates/2669479/ http://www.usatoday.com/story/opinion/2013/08/18/nsa-privacy... [2] http://www.worldaffairsjournal.org/article/beyond-snowden-nsa-reality-check http://www.worldaffairsjournal.org/article/beyond-snowden-ns... (search for 'interagency') [3] http://abcnews.go.com/Blotter/nsa-releases-snowden-email-nbc-truth/story?id=23918598&singlePage=true http://abcnews.go.com/Blotter/nsa-releases-snowden-email-nbc... [4] http://www.nytimes.com/2005/12/16/politics/16program.html?pagewanted=all&_r=0 http://www.nytimes.com/2005/12/16/politics/16program.html?pa...