I never thought of myself becoming a “whistleblower” in relation to my “democracy support” work on the failed 2007 Kenyan election as resident director for the International Republican Institute.  I worked internally to press for the release of the USAID-funded exit poll contradicting the “results” of the election announced on Sunday December 30, 2007 by the Electoral Commission of Kenya and worked internally to try to uphold what I saw as required for the integrity of the IRI Election Observation Mission, also funded by USAID as a separate program.

From mid-December 2007 I was actively resisting what I understood to be, and described to my superiors in IRI as “some agenda” by the U.S. Ambassador in relation to the election itself, with the understanding that we were in complete agreement within IRI of the need for such resistance to attempted interference with our independence.

My Contracting Technical Officer at USAID was caught in the middle between me (and IRI) and the Ambassador.  While she was directly answerable to USAID in Washington as I was to IRI in Washington, and the funding agreements for the programs were issued in Washington, as a practical matter, the Ambassador controlled the process.  The Election Observation was initiated by the Ambassador specifically contrary to the prior planning of USAID (which was changed to accommodate him).  The exit poll was added on to our polling program–contractually and as confirmed in our explicit conversations, as a check on potential election fraud–but really as she told me by phone on the afternoon of election day, as “early intelligence” for the Ambassador as to who was winning.  I know she agreed with some of my concerns and it was certainly my impression from my interactions with her in the aftermath of the election that she felt as badly about what happened as she could allow herself to show in the context of doing her job.  On balance I see her primarily as more a victim of rather than a willing participant in whatever the shenanigans were.

I complained internally about interference from the Ambassador by writing a long e-mail missive to the USAID CTO on Tuesday, December 18, 2007 following a phone conference with the senior IRI leadership in Washington in the wee hours of the morning Nairobi time.  I do not have a copy of that e-mail and USAID did not produce it, or any of the other e-mail correspondence regarding the agreements in response to my FOIA request.

The IRI leadership had called me that Monday afternoon (their time) to follow up on my e-mail report on my private meeting with the Ambassador at his residence that Saturday, December 15.  This was the e-mail noting the “some agenda” of the Ambassador and reporting that he had said “people were saying” that opposition candidate Odinga might, implausibly to my assessment, lose his own Langata parliamentary constituency and thus be disqualified from taking the presidency regardless of the outcome of the national vote, and the Ambassador’s desire to take our lead Election Observation delegate Connie Newman to meet with Stanley Murage, “President Kibaki’s Karl Rove,” on the day before the election to be followed by observing the election with the Ambassador and his staff rather than with our IRI delegation.  I had gone to the Ambassador’s residence based on a phone call that Friday afternoon from an unidentified caller who “worked for the Ambassador” having been told by IRI’s president at the time, Lorne Craner, from Thailand, that the Ambassador wanted to talk to talk with me.  As I have written, Craner had called Assistant Secretary of State Jendayi Frazer on his way to the airport, as he related it, to “get her Ambassador under control”, then followed up with a call to Ambassador Ranneberger upon arriving in Thailand, after the Ambassador had twisted my arm hard on Thursday to get his predecessor Ambassador Bellamy removed as an Election Observation delegate.  My instructions from Mr. Craner were explicit: accept “no more b.s.” from the Ambassador.

I had been in a quiet “push-pull” on behalf of IRI with the Ambassador and his staff and USAID for some period of time over the independence of our Election Observation Mission before things came to a head with the issue of removing Bellamy, the proposed Murage meeting, etc., leading to my complaint to USAID.

As I have written, after Ranneberger’s meeting with myself and my boss and the late Amb. Rich Williamson in August in which Ranneberger again expressed his desire to have IRI observe the election, USAID told me they would “move heaven and earth” to make the observation happen and they came up with $235,000 of “Economic Support Funds” at the end of the fiscal year in September for the mission.

Ranneberger wanted, as I was told later by the CTO, to select all of IRI’s Observation delegates.  She said that she explained to the Ambassador that this was not doable, but promised him as a minimum the approval of the “lead delegate”.

When she wrote up the Request For Proposals (“RFP”) for a Cooperative Agreement to conduct the Election Observation it was de facto directed to IRI, in accordance with the Ambassador’s previously expressed desire.  The RFP was issued on a non-competitive basis to the CEPPS (“Coalition for Political Parties and Process Strengthening”) comprised of IRI, NDI and IFES, thus eliminating the Carter Center and Democracy International.  Based on language in the RFP, NDI was in effect eliminated by their work with the competing political parties and IFES was eliminated by their role as “embedded” with the Electoral Commission of Kenya.  (After my return to the States I found that USAID had paid a consulting firm, MSI, in the spring of 2006 to study and advise on USAID’s preparations for the 2007 election.  After extensive interviews in Nairobi, including staff of all three CEPPS entities working on the USAID programs at the time, they recommended that USAID plan for and fund an election observation and that the Carter Center was the most appropriate entity to conduct it.)  In the RFP the CTO included descriptions of the credentials matching without naming the specific people that Ranneberger wanted as “lead delegates”, former Assistant Secretaries for Africa for whom Ranneberger had worked, Connie Newman and Chester Crocker.  The “lead delegate” was to be formally approved as USAID’s “substantial involvement” in the program.   For the rest of the delegates that Ranneberger had specified to me that he wanted IRI to invite, the RFP listed matching credential descriptions, but as examples without a contractual right of approval.

As I have written, IRI went along with inviting Newman and Crocker (Crocker declined as unavailable) while refusing to submit Newman’s name for formal approval as being an impermissible intrusion on IRI’s independence in conducting an international Election Observation Mission.  Of the other potential delegates that Ranneberger wanted IRI to invite as per his after hours cell phone calls to me, Joel Barkan was the only one included in the EOM as he had already been identified separately by IRI.  None of the others, of which well known former diplomat Frank Wisner, then at insurer AIG, stands out in my recollection, were invited by IRI.

The Ambassador took a keen interest in the lodging arrangements, in particular wanting Ms. Newman to stay at the embassy residence, or alternatively at the Serena hotel (near State House as well as closer to his residence and others in exclusive Muthaiga) rather than at the Holiday Inn Mayfair which we had selected for the delegation. We internally insisted on planning for Connie to stay with the rest of the delegation, even before the alarm bells went off from the Ambassador’s December 15 expression of desire to take her to meet with Stanley Murage the day before the vote.  Likewise, I nixed having our delegation travel in State Department cars with State Department drivers (I did go along with having interpreters for many of our teams).  I also declined to merge our observation headquarters operation into the Ambassador’s diplomatic command post at the Embassy in Gigiri, keeping our operation separate at the Mayfair, with a staff liaison to the Embassy and to the EU observation headquarters.

During that wee hours December 18 phone call from Washington (I was awoken at home) following my report on meeting with the Ambassador, I was given the opportunity by IRI’s number two official (filing in since Mr. Craner was in Thailand) to cancel the election observation on my say so based on the Ambassador’s interference.  This is one of the crucial things that has always made me believe, in accordance with what I was told directly, that everyone on the IRI staff was in accord that we were committed to “playing it straight” on the election itself and that all the “agenda” issues came from or through the Ambassador and not from within IRI.

Unfortunately, having to make a judgment call on the spot, in the context of our detailed discussion of our plans and logistics, I made the decision that we could go forward.  Mea culpa.  If I had to do it over again, with more foresight into what would come, of course I would have said we have to cancel.

In fairness, I have to say that my decision was based on counting on the fact that it was agreed that Connie Newman would be accompanied by and briefed by the other senior IRI officer on the call (who would be the senior official on the ground for the election observation) as to the interference problems and the need for Connie to keep her distance from the Ambassador.  I was given explicit assurance that Connie could be expected to understand and cooperate.  I simply did not appreciate the possibility that this agreed approach would either be abandoned without notice or explanation to me or simply fail through refusal by Connie to cooperate.

A key factor in my decision was that it seemed clear that abruptly cancelling the election observation days before the vote–without explaining why (or most especially if we did explain, which of course was totally unrealistic)–would be a disruptive factor in the last days leading up to the election, and potentially something of an “international incident”.  We were the only international non-governmental organization scheduled to observe and the observation had already been announced and publicized in Washington and Nairobi.  No one was publicly predicting violence or major problems and there was no obvious reason why we would suddenly just cancel.

Again, in a key sign that people on staff at IRI in Washington were trying to do the right thing, I got permission to do a last minute poll of Raila’s Langata constituency in response to my meeting with the Ambassador.  It seemed to me a clear way to telegraph that we would be “observing” seriously and were not going to go along with an obviously bogus result from Lanagata when, as confirmed by the poll, the race there was in no way remotely in doubt.  I told the Ambassador’s top aide on Christmas Eve that we had done the poll and conveyed the results to the Ambassador in person that evening as requested.

As it turned out, Connie and the Ambassador were obviously close and quite well coordinated.  When she visited Nairobi in 2009 he introduced her at the residence as “his great friend and mentor” and during the pre-election in 2007, even though she formally remained lodged at the Mayfair, she stayed behind at the embassy residence after our pre-election gathering there with the Ambassador when the rest of us boarded the bus to leave.  She told me she would be driven back to the Mayfair later, but I was told that the other delegates took notice of the fact that she didn’t end up returning.  I have no idea whether she ended up meeting or talking to Stanley Murage with the Ambassdor or not, one way or the other.  The issue was never mentioned after our internal agreement that it “must not” happen and I hope it didn’t.

On the evening of the vote, I learned from our liaison to the EU observation mission security team that the Ambassador had called his State Department observers in to Nairobi from “the field” that night due to concerns of violence, but no one else told me, including our liaison at the U.S. Embassy observation headquarters.  Our IRI teams stayed out as did the EU’s.

On the morning after the election, when Connie and I and my two IRI superiors from Washington convened as planned ahead of the vote to draft an IRI Preliminary Observation Statement, Connie and I took opposite angles–she steered to make the statement as positive as possible, I steered to keep it as reserved and as cognizant of obvious issues as possible, given that we did not really know much yet.  Through the Freedom of Information Act I learned several years later that the Ambassador had reflected in his cable to Washington that day that IRI was expected to release a “largely positive” statement that same day.  In the afternoon Connie presented the final “Preliminary Statement” to the media in a solo press conference with IRI staff and such other of our observers as were back from the field by that afternoon in the audience.

When the three senior IRI staff (myself included) and Connie met with the leadership of the EU delegation the next day, December 29, at the Serena Hotel, I learned that we were significantly criticized for releasing our Preliminary Statement before any of the other observation missions and while the vote tally was ongoing.  During the formal discussion between the two delegations, Connie asserted as an example of the positives from the vote the notion that the election officials had done a good job of consistently handling assistance to voters who needed it.  I spoke up and said that I had observed otherwise since Connie was obviously pulling that notion out of thin air. In fact, our Preliminary Statement itself the day before had said “As happens in many elections around the world, the ECK must address the issue of polling stations opening late, voting materials being delivered in a timely manner, and appropriately providing assistance to voters”.  I was sitting right beside Connie chatting along the back wall of the polling station when I took the photograph below of a voter beseiged by would be “assistants”:

Subsequently, I made the mistake of pressing for release of the exit poll results indicating an opposition win over Kibaki to my bosses from Washington in front of Connie.  Connie immediately spoke up to object to any release of these results.  My regional director, my immediate superior from DC, pulled me aside and pointed out that I had made a mistake raising the topic in front of Connie as it was not her place to be involved.  I acknowledged my error, but the bell was rung at that point as Connie was an IRI board member and the rest of the senior staff as career employees were not going to openly resist once she preemptively staked out her ground to quash the poll.  (And to be clear, there was no discussion or any claim whatsoever on Connie’s part at that time–or ever in my presence–of any confusion about the “validity” of the poll based on a misunderstanding about the performance of the polling firm, or the “methodology” or any other grounds offered from Washington in later weeks as scrutiny came to bear.)

To be continued . . .