This is the post I promised, in which I'll analyze what the timeline of the missing dates shows. As I said in that post, this exercise makes several assumptions, some of which clearly are not true:

  • It assumes all the missing emails have some tie to the Plame leak; we know this is not true because of the volume of email missing from offices uninvolved in the leak, and there is at least one period when no archive of OVP email exists for which I can think of no Plame leak correlation.
  • It assumes we're seeing all the missing emails; we're not. There's a bunch of dates on which there is a very small amount of email archived, and if we were to do this analysis properly, we'd need to know those dates, too.
  • It assumes the email archives were destroyed deliberately to hide legally dubious acts. While that might be a fair assumption with this administration, we don't know for sure that is true, so by trying to find correlations between missing emails and known events, we may end up imagining motivations on the part of the White House that didn't exist.

So understand that this is as much a thought experiment as useful analysis. It basically tries to answer the question, "Assuming most of the WH and OVP email gaps during this period relate to the Plame investigation, why might the WH have been deleting archives? What were they trying to hide?"

Also, consider some limits about the content of the email. We're assuming the email was dangerous enough to make it worthwhile to delete. Yet, given that Fitzgerald got at least 250 pages of the missing OVP emails (and presumably a similar amount of missing WH emails), one of the following must be true:

  • The emails were not damaging enough to support an indictment for anyone beyond Libby. Only one of these emails was ever even introduced at Libby's trial--and it was nowhere near the most incriminating piece of evidence. So the emails Fitzgerald received, at least, either contain no smoking gun or he chose not to pursue the smoking gun.
  • The truly damaging emails would not be included in any of the subpoenaed searches. I have already raised questions whether the draft search terms Addington proposed to search for emails pertaining to journalists would have returned all the responsive email, but it is also possible that, if they were trying to hide something, it was something outside of the general scope of the discovery, which was mostly limited to Wilson, Plame, Niger, and the journalists.

In other words, either those emails do not include a smoking gun, or Fitzgerald never found that smoking gun for a variety of reasons.

With all that said, it appears that almost all the periods for which OVP or WH were missing emails (the exceptions being September 12, 2003 and May 21-23, 2005) were periods during which they were responding to document requests or subpoenas. There is clear indication that OVP, at least, attempted to shield conversations with journalists outside of Novak, Phelps, and Royce (and given Libby's claim that he didn't speak to Novak the week of the leak even though his Novak's phone records showed he did, he appears to have tried to shield his conversation with Novak, as well) [h/t Jeff for the correction]. Thus, one possible explanation for the missing email archives is that OVP and WH were trying to hide email discussions about their attempts to hide the most incriminating discussions with journalists, notably with Judy Miller.

September 12, 2003 (no OVP emails): This one-day gap in the archives occurs on the Friday before two Sunday events: Joe Wilson writes an op-ed for the San Jose Merc-News, and Cheney appears on MTP and disavows ever knowing Joe. Wilson's op-ed was itself a response to a major Bush speech on terrorism. Thus, it certainly seems possible that emails from the 12th discussed the upcoming MTP appearance--strategizing a way for Cheney to rebut Wilson. I'm wondering if they had any way of knowing about the Wilson column beforehand, and if they placed Cheney on MTP for that reason?

Even if Fitzgerald found something to that effect at the trial, it would be unlikely he'd introduce it. We know Fitzgerald also has, as evidence, an annotated copy of Sy Hersh's Stovepipe article which (like any lead-up to MTP discussing Wilson) would be evidence of ongoing obsession with Wilson and his claims, presumably on the part of both Libby and Cheney, but Fitzgerald didn't introduce that article at trial. By the same logic, he probably wouldn't introduce this.

October 1-3, 5, 2003 (no OVP emails): Obviously, this four day gap occurs just after the investigation began. From trial evidence, we know that Libby and Cheney had two or three discussions about getting Scottie McClellan to publicly exonerate Libby, as he had Rove. We know there were email discussions among OVP staffers about how McClellan's comments about Libby--that is the subject of the one email entered at trial that hadn't been archived. So to some degree, we know that this gap covers a period when OVP was trying to get Libby exonerated.

The gap covers another significant process, as well: the four day gap between the time when DOJ told WH to save any relevant materials, and the day when Gonzales passed on the specific request. If any significant evidence destruction occurred, it likely occurred in this period.

At the very least, though, we know that OVP started to collect information responsive to the document request and Libby had at least two conversations with Cheney in which he told Cheney the (evolving) story he was going to tell the FBI. During this period, Libby devised a cover story (that he learned Plame's identity from Russert), then he revised that story when he found the note indicating that Cheney told him of Plame's identity. Would any of these discussions show up in email? The discussions between Cheney and Libby almost certainly took place in person--they were conveniently together in Jackson Hole just after this period. But it's possible there were email discussions about what evidence turned up in the search.

Finally, consider the fact that, at least given the evidence that appeared at trial, OVP did not turn over materials clearly responsive to the initial document request that happened to mention one of the journalists not, originally, requested. For example, this hard copy of the Martin-Cooper email was not produced until after October 21 (Martin had an FBI interview on October 22), under Martin's certification. But the actual email did not come up on an email search until February 11. Yet the email clearly was a document "that relate[s] in any way to a contact with any member or representative of the news media about Joseph C. Wilson [and] his trip to Niger in February 2002," one of the first three document requests. Cooper asks, "Who in the vice president's office communicated to the CIA their interest in the Niger allegation?" While the email never mentions Wilson by name, Wilson's central assertion was that the CIA sent him in response to interest from Cheney. (Note, all this is true for Rove's Hadley email, as well--it was clearly responsive to the first document request.) OVP clearly interpreted the document request very narrowly. Was there a discussion on email to that effect? We know that Jenny Mayfield, at least, knew to leave all of her and Libby's email discovery to a centralized search (even as she did not turn over other clearly-responsive materials, annotated documents in her Niger/Uranium file), so at the very least, there must be some discussion that such searches would occur.

December 17, 20, 21, 2003 (no WH emails): Two significant events happened during this period. First, Jim Comey officially assumed the role of Deputy Attorney General on December 11. Prior to that date, Ashcroft had responded to demands that he recuse himself from the Plame investigation that he could not do so until there was a DAG. That suggests discussions about the recusal may have started almost immediately after Comey took over. If there were emails at the WH pertaining to that recusal, they would not have come up under any known subpoena.

It's also possible that the White House received a parallel subpoena to the one OVP received on December 16, asking for "any records relating to" either version of the INR memo, "including without limitation any records relating to the dissemination of such document." If it did, then it would mean there is no WH email for three of the four days between the date the subpoena was issued and its due date. We know the INR memo was on Air Force One the week of the leak in at least one form (from Powell; Howard Fineman once reported that a copy was also in Condi's briefing book), so a request for information about it would have implicated Dan Bartlett, as well as others aboard Air Force One (up to and including Bush), and it may have implicated those back at the White House corresponding with Air Force One (like Rove). If there were emails pertaining to the INR memo, they would have come up under a known subpoena only if they mentioned Wilson or Plame (which seems likely, though given the narrow response the White House made in October 2003, who knows if they would have turned it over?).

January 9-11, 2004 (no WH emails): These are the three days leading up to the day that Fitzgerald would arrange an interview with Novak, with waivers from Rove and Armitage in hand. Certainly, the WH was contemplating whether their various stories would hold up once Fitzgerald started talking to journalists about their specific contacts with Administration staffers. And given that we know Libby reached out to Russert, at least, in the period after signing a waiver, it's possible WH staffers reached out to other journalists. Such discussions may not have come under a known subpoena unless they mentioned Wilson or Plame.

January 29, February 7-8, 2004 (no WH or OVP emails); January 30-31, 2004 (no OVP emails), February 1-3, 2004 (no WH emails): This is the period during which WH and OVP were responding to a variety of subpoenas, including:

  • Documents relating to the Air Force Two July 12 trip
  • Air Force One phone records for leak week
  • July 12, 2003 gaggle transcript
  • Gerald Ford party guest list
  • WHIG documents for July 2003
  • Wilson-Plame-Niger documents since the start of the investigation
  • Contacts with a slew of reporters

In other words, this is the period when it would become clear to the WH and OVP that Fitzgerald was interested in leaks to journalists besides Novak, Phelps, and Royce, WHIG activities after the leak, and contacts with journalists from Air Force One and Two. This was the period when it would have become clear that the very narrow response the WH and OVP took in the fall was not going to work and that Libby's and (presumably) Rove's lies might be exposed as such. It was also the period when people like David Addington were drafting narrow email search instructions that would have shielded contacts pertaining to Judy Miller, Andrea Mitchell, Matt Cooper (and, curiously, Glenn Kessler).

If there were discussions at OVP and WH--and between the two--about strategies for shielding some of this material, I can imagine they wouldn't want them to be easily discoverable. Since this batch of subpoenas asked for materials through the present (that is, January 23), any emails about these topics after January 23 would not be discoverable through any known subpoena.

February 15-17, 2005 (no OVP emails): I have to brag a little. When I saw these dates in Waxman's list, I guessed that it might correlate with the Appeals Court ruling that Judy and Cooper would have to reveal their sources. And boy did it, the day of the opinion and the two immediately thereafter.

I would imagine that OVP was in a full panic at this point: they knew that if Judy testified honestly, it would become fairly clear that Cheney had ordered Libby to leak Plame's identity to her, possibly with the foreknowledge of Bush. Would they be so stupid as to panic on email or leave some indication they were doing so? Might they have emailed Judy? I don't know. But if they did, these emails would not be discoverable through any known subpoena.

May 21-23, 2005 (no OVP emails): Unless Scalia gave his duck hunting buddy a heads up that SCOTUS was not going to review Judy and Cooper's appeal more than a month before SCOTUS announced this publicly, I can think of no Plame-related leak coordinating with this date.