I must have read, or at least skimmed, dozens of Kenya articles, papers or policy briefs that include, usually near the beginning, reference to the alleged circumstance of Kenya being “on the brink of civil war” at the time of February 2008 post election “peace deal” brokered by Kofi Annan between Mwai Kibaki and Raila Odinga. Invariably, this important assertion is without any type of citation or elaboration. It has become self-referential conventional wisdom.
In the case of political science papers on narrower topics–those along the lines of “What can ‘big data’ tell us about gender disparity in boda-boda fares in rural Kisii eighteen months after Kenya’s Post Election Violence?”–the “brink of civil war” reference is boilerplate contextual introduction. More significantly the “brink of civil war” phrase is standard in writings on issues of foreign policy, conflict avoidance and resolution, electoral violence specifically and the development of democracy more generally. In these writings, the validity of this relatively untested characterization matters a great deal.
I don’t say this to be critical–the “brink of civil war” line is found in the writings of personal friends and people for whom I have the utmost regard. Which in a way makes it all the more important to raise my concern that the terminology may unintentionally mislead those who don’t have personal knowledge of the ins-and-outs of what was happening in Kenya from December 27, 2007 to February 28, 2008 and may skew historical understanding.
There were several types of violence in various locations in the country triggered from the election failure. My contention is that none of them were close precursors to any likely civil war.
To put it directly, the incumbent administration seized the opportunity to stay in power through the up-marking of vote tallies at the Electoral Commission of Kenya and the immediate delivery of the contested certificate of election to State House for the quick secretly pre-arranged swearing in of Kibaki for his second term before his gathered supporters there. The incumbent President and Commander in Chief remained in effectively complete control of all of the instruments of state security–the Police Service and Administrative Police and General Service Unit paramilitary forces, along with the military forces and intelligence service–all of which were part of the unitary national executive.
Notably, the Administrative Police had been deployed pre-election to western areas of Kenya in aid of the President’s re-election effort as we in the International Republican Institute election observation were told in a briefing from the U.S. Embassy on December 24th and many Kenyans had seen on television news broadcasts. While this initially led to disturbing incidences of pre-election violence against individual AP officers, by election day the vote proceeded peacefully with voters cooperating with deployed state police at the polls.
A civil war scenario would thus have involved an insurrection against the State. I really do not think this was ever likely, most importantly because none of the major opposition leaders wanted it, nor a critical mass of the public without any pre-defined leadership.
While Kibaki’s official “victory” by roughly 200,000 votes rested on a reported 1.2m vote margin in Central Province, significant strongholds of the opposition were in parts of Nairobi and in the west overall, starting in the western/northern parts of the Rift Valley and including Western and Nyanza Provinces. The violence on the Coast was not broad and extreme and eastern Kenya was not destabilized in the way that it has been in recent times. The key ‘slum’ areas in Nairobi were fairly effectively sealed in on the eve of the vote as government security forces deployed in Nairobi. Violence in the slums was no threat to overthrow the government and never broadened to seriously threaten areas where the political class (of whichever party affiliation that year) lived.
Palpable fear of a mass scale conflict between opposition civilians and state security in Nairobi largely ended when Raila cancelled the planned ODM rally for January 3, 2008 as the GSU continued to surround Uhuru Park shoulder to shoulder. As best I could tell the EU at that point came around to support the U.S. position in favor of negotiated “power sharing” in lieu of a new election and/or recount or other remediation. Acts of terrible violence continued to ebb and flow in specific places but Kibaki’s hold on power was not threatened as far as I can see.
The only serious prospect for civil war in my view then would have been the idea of the opposition breaking off the western regions. Conceptually the opposition might have been able to rally popular support to challenge the State in this way and test the loyalty of the military if they were willing to bear extensive casualties by larely unarmed civilians in resisting the police and paramilitary forces. A basic understanding of Kenyan political economy and the practical politics and players in early 2008 readily explains why this was not on the table.
At that time, at least, all major political leaders in Kenya functionally lived in Nairobi, regardless of political constituency, just as nearly all governance was national. Even though it contained only about 8% of the population, Nairobi accounted for roughly 60% of Kenya’s GDP as estimated by Calestous Juma in a Wilson Center discussion that January (the same one in which Joel Barkan raised the suppressed exit poll). The roles played by Nairobi in Kenya would equate to that of Washington, New York, Los Angeles, Silicon Valley, Chicago, Houston, Detroit and most other major key urban centers for specific activities in the United States all rolled into one.
None of the ODM politicians I am confident had a desire to lend themselves to a fight to hive off the underdeveloped far west of the country and govern it without Nairobi.
It was my understanding that the assessment of CENTCOM, the cognizant U.S. military command at the time, was that Kibaki could in fact “tough it out” (I am not suggesting one way or the other whether they thought he “should” as opposed to simply “could”). This was consistent with the presentation to Deputy Secretary of State Negroponte in Washington made by Kalonzo Musyoka as Kibaki’s appointed Vice President on February 7, 2008 as reported in “The War for History” part nine that the main violence was over and there was no need for international mediation.
It is important to note that the organized part of the violence by Kalenjin militias to burn out Kikuyu farmers in portions of the Rift Valley–like the violence around the elections in those regions in 1992 and 1997–was directed at future voting within the existing polity rather than an effort to overthrow the government or split the country through a civil war.
All the while, the primary protagonists were spending much of their time “running around” Nairobi, not fighting themselves. See my post “When did Ruto and Uhuru fight? And why is the “Uhuruto” alliance allegedly so surprising?”.
The violence was horrific. To me, the belief that a civil war was never a likely prospect bolsters appreciation of how unnecessary and morally repugnant the entirety of the election crisis was. It also suggests that while the “international community” can rightly give itself credit for assisting Kenya in ultimately reaching the “peace deal”, we should not pat ourselves on the back for averting a “civil war” as a diversion from facing up to the failures in the underlying election assistance endeavor.
Likewise, going forward to 2013 election and beyond, I have been disappointed to see the significance of deficiencies in the billion dollar (!) election, with over $100m spent by the United States, downplayed on the excuse that Kenya is a “post conflict state.” In a neighborhood that includes Burundi, Rwanda, Uganda, Sudan and South Sudan, Ethiopia and Somalia, calling Kenya a “post conflict state” is to sell short the reasonable hopes and expectations of Kenya’s largely peaceable voters.