The Phone Dragnet Classified Appendix
The government has been releasing a bunch of documents under FOIA while we’re all out celebrating: a classification review of the two earlier Section 215 IG Reports, as well as NSA’s reports to the Intelligence Oversight Board (though thus far, NSA has mistakenly linked to 1Q 2012 rather than 2Q 2012, which should be one of the most important reports for reasons I’ll come back to).
In this post I just want to review the phone dragnet classified appendix included as part of the 2008 DOJ IG Report on the use of Section 215. We’ve known this appendix — one of two attached to this report (the other, which may be as long as 16 pages, remains classified) — dealt with the phone dragnet since the phone dragnet was revealed. One thing this report provides are clear dates (which I used to update the dates in my phone dragnet tracker), including exact (in case of the first addition) and rough updates for additional “agents of a foreign power” that may be chained on.
Here are details of interest:
The fourth redaction on the 2nd page of the appendix — in the sentence starting “The queries would attempt to identify…” — is rather interesting syntactically. The redaction should read something like “terrorist associates” or something similar. But in this context, it ties the contact chaining much more closely to the contact-chaining process. Somewhere there must be language purporting to make this case specifically, but the redaction here is remarkably short to do so.
The appendix notes in the first full paragraph on page 3 that the dragnet application promised the NSA Director would inform the Intelligence Committees (but not the Judiciary Committees) about the dragnet. That’s curious because we have every reason to believe the NSA did not inform the Intel Committees about the Internet dragnet until after PATRIOT reauthorization, as reflected by this April 27, 2005 briefing to SSCI. Presumably, the December 15, 2005 disclosure of the dragnet led the FISC to discover that Congress hadn’t been briefed.
The discussion of the additional terrorist group approved for contact chaining on page 4 seems heavily redacted. I wonder if NSA got Iran approved as early as 2006, with the later approvals being additional al Qaeda affiliates?
At least according to the changes noted in the dragnet orders, the only known addition in the second dragnet order was the pre-approval for FISA targets to be RAS seeds under the dragnet. I’m not sure whether the redaction here would refer to this change, but if it does, it is odd it remains redacted. But it’s also possible the government started collecting some other kind of telephony metadata in that order.
With the exception of the first order, it appears DOJ’s IG was working from the applications for the dragnet, not the orders. And the narrative of the dragnet appears to be silent on a number of changes, including the elimination of the compensation paragraph, the addition of spot checks (both in the November 15, 2006 order), and the exception of pre-authorized RAS approval for dockets 06-2081, 07-449, and PAA.
Most interesting still is the report’s silence on the change allowing NSA to put the BRFISA data in with other data for the purposes of analytical efficiency. That first shows up in the first dragnet order of 2008 — which the appendix helpfully clarifies was signed on January 10, 2008. It’s possible the IG Report doesn’t note it (or some of the other changes) because it was only supposed to treat Section 215 for 2006. Perhaps the other changes were done via amendment not shared with the IG (perhaps because of that scope issue). In any case, I find the timing of the order (which admittedly was dictated by the expiration date of the prior order). That would put the change — which I’ve speculated might relate to the roll-out of ICREACH — just days after Michael Mukasey signed the SPCMA order which allowed chaining on EO 12333 data on US persons. I increasingly believe all these things — ICREACH, SPCMA, and the insertion of FBI into the heart of the FISA process — were necessarily rolled out together.
One other silence of note: This appendix, at least, makes no mention of the 4- and 15-page October 31, 2006 opinions withheld from the EFF and ACLU FOIAs. That’s not surprising: if it had been central to the phone dragnet, the government probably would have had to release it. I wonder, though, if they pertain to the dragnet program discussed in the second, still unreleased appendix (and I wonder if that is the CIA money transfer program).