Government Releases “Primary Order” for Telephony Metadata Under FISA Section 215

Readers will recall that when Glenn Greenwald released the FISC order indicating that the NSA was getting all domestic telephony metadata, that order was designated the “secondary order.” We have since heard that there was a “primary order” covering what the government is allowed to do with the data obtained under the “secondary order.” The DNI has today released a somewhat redacted version of the primary order: You can see it here.

A few thoughts on the “primary order”:

1) Although billed as imposing minimization procedures, the “primary order” strikes me as more like its own surveillance statute: Starting on page 4, it lays out in detail what the government can and can’t do and what procedures they must follow.

2) As has been widely reported, the basic standard for querying the database is reasonable suspicion. More specifically, on page 7, we learn that there are 22 people at the NSA that are empowered to make the call as to whether there is reasonable suspicion to query the database with a “seed” query. The NSA General Counsel’s Office also gets involved if the selection term is reasonably believed to be used by a U.S. person, in order to block queries based on 1st Amendment activities.

3) Specifically, the standard to be applied according to the primary order is whether

based on the factual and practical considerations of everyday life on which reasonable and prudent persons act, there are facts giving rise to a reasonable, articulable suspicion that the selection term to be queried is associated with [redacted].

(p.7) I assume the redacted part here is something like Al Qaeda and its associates, but who knows. If you’re wondering where they’re getting this standard, it’s from two lines of Fourth Amendment cases. The first part is from an oft-repeated Supreme Court standard for the methodology of how to evaluate probable cause first announced in Brinegar v. United States, 338 U.S. 160, 175 (1949) (emphasis added):

In dealing with probable cause, however, as the very name implies, we deal with probabilities. These are not technical; they are the factual and practical considerations of everyday life on which reasonable and prudent men, not legal technicians, act.

The second part is the Fourth Amendment Terry standard for making a temporary stop, except that the standard here is reasonable suspicion that the selection term is associated with [redacted] rather than reasonable suspicion that criminal activity is afoot.

4) There are some interesting details about how the minimization order works after a query term is approved. Ordinarily, search terms that are approved are valid for 180 days, although if a term is thought to have a specific shorter time frame of relevance, the approver of the query can give it a shorter period of approval (p.10). Data is deleted no more than 5 years after its collection. (p.14) The FISC also gets a monthly report that includes a “discussion” of the NSA’s implementation of the reasonable suspicion standard, although we don’t know how detailed a discussion that is. (p.16)

5) I’m still not sure where the FISC is coming up with the idea of the reasonable suspicion standard for querying the database. This document is just the order, not a legal opinion, so that’s not too surprising. But we still don’t know if that standard is just the FISC’s imposed minimization procedure imposed to satisfy the statutory minimization standard or if it was designed to ensure compliance with some kind of Fourth Amendment principle.