My understanding is that properly cleared representatives and their staff can absolutely get information on what they need and speak candidly about all the sensitive details in a SCIF.
This recurring narrative that intelligence agencies are holding back is not good, because it makes it seem like representatives are being misled and are tricked into voting for these authorities, deflecting accountability, which is far more convenient than the truth of them being well-informed and still choosing to support it.
If a senior Senator from the Select Committee on Intelligence says he has been unable to get even preliminary, basic information on the impact of this program from the military, I for one believe him.
The recurring narrative sounds completely correct & shocking & egregious. It sounds like the military is operating completely out of control, with no oversight, because the Senators in charge of oversight say they can't get even basic preliminary information.
Your statement seems dangerous and misleading @willstrafach. If the Senate Select Committee on Intelligence gets nothing, not even basics, something is deeply awry. Who are the cleared representatives? Who has gone and asked serious questions? What does Wyden, who cares deeply, need to do to get whatever clearances you wishfully suggest as magic remedy here? This is fradulent bad misdirection @willstrafachs. It's all suggestion, with not a shred of backing to it.
It's an unacceptably bad and insubstantial reply to one of the most serious, can-kicked-down-the-road the longest issues, a terrible spectre still haunting us from Bush II years. This issue seems never to have been at all handled at all properly, since it's inception, and things are still fast & loose, the basic measures of oversight & responsibility seem deeply obstructed. That's a threat to Democracy, and it constantly throws into peril American trustworthiness under essential programs like EU Privacy Shield.
The purpose of the SSCI and HPSCI is intelligence oversight, so if there was "no oversight" it would be a much bigger deal. You are assuming that as the default, which is equally as dangerous and misleading without substantiation. It is a huge accusation.
I can understand that you don't like ambiguity in my reply, that is fully understandable, but it is not very reasonable to discard it wholesale. I am happy to elaborate.
First, "preliminary, basic information on the impact of this program" is your interpretation, and is a generalization. There are very specific and valid questions Wyden has. Not only that, but he might know the answers to some, as you can see that some in the linked source are "The public doesn't know" whereas others are "No one knows" the answer. For the record, I do agree that the public (beyond cleared reps) should have a whole lot more information to review on these matters for the sake of transparency, but that is not relevant to this current claim of cleared representatives allegedly not getting information.
Now the precise issues at hand:
> Who can be targeted by 702 spying?
> A: The public doesn’t know. If you listen to intelligence agency leaders, you would think only foreign terrorists could be targeted for spying under Section 702. But when Sen. Mike Lee, R-Utah, and I asked Director of National Intelligence Dan Coats what broad categories of people could be targeted by this program, he refused to answer the question publicly.
The Snowden leaks additionally showed that beyond counter-terrorism, 702 is used for counter-proliferation, counter-intelligence and tracing cyber attacks as well. Additionally, due to the nature of how every intel agency in every country in the world operates, one can reasonably assume it is probably also used to assess the intentions of foreign government/military leaders, as that is a very basic reason for nation-state level spying.
His use of "the public" implies there is a classified answer given in closed session, in the same manner I described in my previous comment. I agree with him, there is probably little reason to keep this list classified. I am also personally curious if it has changed since the Snowden leaks (Worth noting, if it didn't, the leaks did not de-classify this list so answer would still be "classified" thus the uncertainty).
> How many emails and other communications from Americans are collected by the government under Section 702?
> A: No one knows, because the government refuses to count.
I think it would be great to have the answer to this one as well, but I cannot think of any non-invasive manner this could be achieved technically.
I will use X-Keyscore an example here. According to the Snowden leaks, X-Keyscore is a unified search interface for a 5 day rolling content buffer (30 days for metadata) containing traffic from different points of accesses around the world (Whether it be a dish pointed at a satellite used as a data link used for embassy comms, "CNE"/hacked router in a middle eastern ISP, a tapped cable connecting traffic between countries, or otherwise). Lots of traffic is going to be "collected" and is subsequently purged. Based on my understanding of how traditional firewalls work, I don't see how any sort of deep packet inspection could be used on the subnets/IPs "selected" for collection without maintaining a large internal buffer or choking on all the intake, so that is likely why there is no decision logic at this point and thus there will be incidental collection of US Person Information if you are communicating via e-mail with someone in a target country, for example.
The fix for this now is prohibition of queries on US Person selectors. This is not hard to impose for US IP addresses and US phone numbers. However, e-mails are murkier, as the originator's location is not going to be clear based on the e-mail address alone. In order to try to identify a communicant, they would need to read the message content in order to try making a determination on this.
I am aware that, again according to the Snowden leaks, this can happen when pursuing legit targets, requiring US identities to be "masked" when found and I believe all such incidents are reported. However, I am personally far less comfortable with the idea of that determination being done at scale in order to get a proper count of US Person communications which have been collected on. Right now, the majority of incidentally collected communications will be discarded. If the IC was required to produce this count though, it means each communication would actually need to be analyzed to make the USPI determination, thereby ensuring US Person communications would actually be read. Again, to me, this seems far more invasive.
I don't see a better way to solve this issue. I am looking forward to your feedback on this aspect, because perhaps there is a non-invasive way to determine which comms are US/non-US that I have not thought about.
Based on the above logic, I think it would be better for him to push the IC to publicly release counts of incidentally collected US Person communications which have actually been viewed, because (again to my understanding - based on Snowden leaks), the USP determination has to be made anyway when analyzing traffic as they wade through to write their reports and such. So such counts must informally exist in some manner due to the need to make this foregin-ness determination, there should be no technical or personnel issues in producing an answer to this question, nor should there be classification issues due to the fact that it is just a number.
> How many times does the government search for Americans’ communications without warrants?
> A: No one knows, because only some government agencies count. The National Security Agency and the CIA both conduct warrantless backdoor searches of communications collected under Section 702. There are thousands of such searches for the content of communications and tens of thousands for communications records. But the FBI searches Americans’ information so frequently, it does not bother to count.
Not keeping a count is a logical answer for the question, and is different from a refusal to answer.
That said, at least in my view, the count is not relevant. It does not make sense for a domestic law enforcement agency such as the FBI to have any sort of access to American data without a warrant, full-stop. Even if the specific details of their access to 702 data made it technically in compliance with the law, the optics of the blurred foreign/domestic separation are very problematic.
> Can the government use Section 702 to collect communications it knows are entirely domestic?
> A: The public doesn’t know the answer to this fundamental question. When I asked the DNI at a hearing, he first replied no, which was reassuring. Later, he said he was answering a different question. Then he said it was classified.
I'm glad he gave Wyden a classified answer to this, and I would very much be interested in that answer being declassified. However, again, the point in question was keeping cleared representatives in the dark (The non-cleared public being kept in the dark is a very different discussion, one which I assume we would probably both be more agreeable).
One more thing to address:
> If the Senate Select Committee on Intelligence gets nothing
What do you think they discuss, then, if the allegation is that no information is given to them? As shown with the above "public" versus "no one" phrasing, is it your belief that Wyden was lying about receiving classified answers of which he would like the public to know?
My understanding is that properly cleared representatives and their staff can absolutely get information on what they need and speak candidly about all the sensitive details in a SCIF.
This recurring narrative that intelligence agencies are holding back is not good, because it makes it seem like representatives are being misled and are tricked into voting for these authorities, deflecting accountability, which is far more convenient than the truth of them being well-informed and still choosing to support it.