June 10, 2009 / by emptywheel


Why Doesn’t the CIA’s Vaughn Index Match the CIA’s Vaughn Index?

Bear with me, because this is going to be weedy, even for me.

The CIA has produced two different Vaughn Index descriptions of four cables recording Abu Zubaydah’s interrogation: the cables from August 1, 2002 (page 1-2; page 24), August 7, 2002 (page 21; page 25), August 11, 2002 (page 12 or 14; page 26), and August 16, 2002 (page 23; page 27).

Basically, what happened is that the government produced a Vaughn Index for the first half of August 2002 back on May 1, but then got ordered to produce a Vaughn Index that covered a wider range of dates, which was released two days ago. The two Vaughn Indices both include these four dates (as well as the interrogation log dated August 4, 2002), which means we’ve got two versions of the index descriptions of the cable for those dates. To avoid confusing dates of cables with dates of indices, I’m going to call the first Vaughn Index–dated May 1, 2009–Vaughn A, and the second Vaughn Index–dated June 8, 2009–Vaughn B.

The series are worth comparing generally, but by comparing these same-day descriptions, we learn a few things.

Somebody (the FBI?) Left the Interrogation Site after August 6, 2002

Yesterday, I pointed out that Leon Panetta admitted that there were contractors on site whose identities could not be revealed because it’d be an unwarranted invasion of their privacy. That manifests itself as an Exemption b(6) for every single cable in Vaughn B.

Vaughn A makes different Exemption claims for the persons present. The Vaughn A August 1, 2002 description includes the same Exemption b(6) claimed in the Vaughn B August 1 description. But it also includes another person-based exemption:

Exemption b(7)(C) – This document contains the names and other personal information of law enforcement officials acting in their official duties. The disclosure of this information could reasonably be expected to constitute an unwarranted invasion of personal privacy and for which there is no public interest in the disclosure. Therefore the information is protected from disclosure by Exemption b(7)(C). [my emphasis; note this strikes me as a really bogus use of this exemption]

In other words, Vaughn A claims there were personnel involved in counterterrorism operations and claims there were law enforcement personnel on site on August 1, 2002.

But Vaughn A stops making both those exemption claims after August 6. It appears someone left the interrogation site after August 6. Given that the Index claims a law enforcement exemption, I wonder if this was an FBI agent, perhaps Ali Soufan’s partner (who was supposed to have left in June)?

Vaughn B never makes a law enforcement exemption claim–but it keeps making its personnel involved in counterterrorism exemption claim, even after August 6.

Field and HQ Were Doing a Lot More Deliberation Than They Now Want to Admit

Yesterday, I noted that none of the earlier cables included policy deliberations. Given reports that the interrogators in the field were getting near-daily approvals for techniques from Alberto Gonzales, I wondered whether the CIA just cherry-picked the cables that showed no such deliberation. 

Nope. At least for the August cables, they’re just not telling us about whether or not deliberations took place. 

The Vaughn A August 1 cable claims an Exemption b(5) for "information relating to intra-agency predecisional deliberations, including preliminary evaluations, opinions, and recommendations of CIA personnel"–precisely the kind of thing we’ve heard was getting cabled back to Alberto Gonzales on a nearly daily basis. The Vaughn B August 1 cable, however, makes no such exemption claim. 

The same is true of the August 7 cables: Vaughn A claims an Exemption b(5), but Vaughn B does not.

We can’t be sure which of the August 11 cables appears in Vaughn B, but one of the two possibilities from Vaughn A claims an Exemption b(5).

(Neither version of the August 16 cable claims a deliberation exemption.)

In other words, on May 1, the CIA was trying to exempt perhaps a quarter to a third of all cables using a deliberative exemption. Now, it is making no such claim, even for some of the very same cables. We can’t say whether the same frequency of deliberation occurred in May and June as it did in August, but it’s possible. The CIA is simply not calling things deliberative that they did label deliberative just over a month ago.

CIA Is Hiding Its Raw Intelligence

There’s one more significant difference between Vaughn A and Vaughn B: Vaughn B doesn’t reveal whether a cable includes raw intelligence.

Compare these descriptions for the August 16, 2002 cable.

Vaughn A

This is an eight-page cable from the Field to CIA Headquarters. The cable includes information concerning the strategies for interrogation sessions; the use of interrogation techniques to elicit information on terrorist operations against the U.S.; reactions to the interrogation techniques, raw intelligence; and a status of threat information. The cable also includes CIA organizational information, CIA filing information, locations of CIA facilities, and the names and/or identifying information of personnel engaged in counterterrorism operations. [my emphasis]

Vaughn B

This document is a eight-page cable from the Field to CIA Headquarters. The cable includes information concerning the interrogation of Abu Zubaydah, atmospheric, and behavioral comments. The cable also includes CIA organizational information, CIA filing information, locations of CIA facilities, and the names and/or identifying information of personnel engaged in counterterrorism operations.

In general, Vaughn A admits more freely that this cable strategizes about interrogation (though remember, this is not a day on which the CIA claims deliberation took place). But by far the biggest difference in these two descriptions is that Vaughn A mentions that the cable includes raw intelligence.

And note that this is an eight-page cable.

In fact, every single longer cable in Vaughn A (every cable longer than 4 pages, and all but one 4-page cable) includes raw intelligence. Yet in Vaughn B, it appears, raw intelligence is never marked.

Now, the possibility that cable length correlates with the amount of intelligence collected has a very important implication. All the longest cables come from the earliest period of Abu Zubaydah’s interrogation, including what is by far the longest cable, 28 pages, on May 6, 2002. So if the pattern shown in the Vaughn A August cables were to hold up (that is, that there is a correlation between page length and the collection of raw intelligence), then it would show that the CIA was collecting more information when the FBI interrogators were present–and more information before they started the worst kinds of torture with Abu Zubaydah.

All of this certainly raises the question of whether the patterns shown in the August cables exist in the earlier cables. Just as importantly, it raises the question of why the CIA, describing the very same cables, has given significantly different descriptions of why it can’t turn over those cables.

Copyright © 2009 emptywheel. All rights reserved.
Originally Posted @ https://www.emptywheel.net/2009/06/10/why-doesnt-the-cias-vaughn-index-match-the-cias-vaughn-index/