What Happened in Luvungi?

On rape and truth in Congo.

LUVUNGI, Congo — In August 2010, the respected Los Angeles-based aid group International Medical Corps (IMC) went public with a shocking account of horrific mass rapes perpetrated by rebel troops over a period of four days in and around Luvungi, a small town in the Democratic Republic of the Congo's embattled and lawless east. The media reports that followed described a "brutal spree of raping and looting," as the Associated Press put it. Some victims were very young children; one, according to the AP, was a "110-year-old great-great-grandmother." While United Nations sources told me that the initial count of victims IMC provided was around 60 or 70 women, by early September the organization had revised its figure upward, saying in a statement that it had provided medical care to more than 242 survivors of a mass incident of sexual violence.

A month later, the New York Times' Jeffrey Gettleman arrived in Luvungi, where he wrote a powerful front-page story describing the anguished cries of the town's women and faulting U.N. peacekeepers, based just 11 miles away, who failed to respond. Gettleman's reporting noted the rape tally as "at least 200 women." The horror of Luvungi seemed to confirm what Margot Wallström, then the U.N.'s special representative for sexual violence in conflict, had dubbed the country in April 2010: Congo, she said, was the "rape capital of the world."

Suddenly, that phrase was everywhere. Congo -- a poster-child failed state, the worst place on Earth to be a woman -- now had another horror added to its long rap sheet. A 2011 study published in the American Journal of Public Health found that more than 1,100 women are raped in Congo every day -- some 48 rapes per hour, as news outlets were quick to report. If anything, many analysts said, the true numbers could be much higher, as the pervasive stigma against rape victims in Congo likely suppressed the reporting of such crimes.

No one would claim that life in eastern Congo, embroiled for almost two decades in conflict, is anything but perilous -- especially for women. The country is justifiably infamous for its high incidence of rape by rebels, soldiers, and husbands alike. But statistics are notoriously hard to come by in Congo. The last census was conducted in 1984; there are few roads, paved or otherwise, in a country nearly the size of Western Europe. And even those areas that are physically accessible are often off-limits due to violence and insecurity. Human rights advocates and researchers in Congo have long had to use "baseline" estimates to determine the costs of conflict. The oft-cited death toll from Congo's decades of war, for example, now stands at more than 5 million. But this figure isn't a count of bodies piling up at morgues; it's an estimate of the difference between civilian mortality rates and the regional "baseline" historical average, last calculated in 2007. Likewise, the American Journal of Public Health study, the most authoritative report to date on rape in Congo, surveyed 3,436 Congolese women and extrapolated the findings across a population of more than 35 million women. The findings were horrific -- nearly one rape a minute, the authors estimated -- but the point is that it's hard to count anything there.

Even in Luvungi, ground zero of Congo's rape epidemic, things aren't exactly what they've been made out to be.

TUCKED AWAY IN the double-canopy rain forest of eastern Congo, Luvungi, a village of roughly 1,000, is a two-day excursion from almost anywhere. The journey from Walikale, the largest town in the area, is a bone-jarring two-hour trip by motorbike taxi. The rocky, rutted road winds up into the hills -- past pristine waterfalls, women coming from the terraced fields bearing hand-woven baskets piled with vegetables, and men weighed down by bundles of wood, machetes at their side. This is the turf of the notorious FDLR rebel group, whose leaders were involved in the 1994 genocide in neighboring Rwanda and have spent the past 19 years living in the bush of eastern Congo, preying on the population. On this visit in April 2011, one of three recent trips I made to Congo for this article, the motorbike driver pointed out places along the road where the FDLR tended to attack. From clearings in the thick jungle, the view to the horizon was all dense green hills, mist, and layers of purple mountains. Cell phones were useless.

Eventually, the narrow dirt road cut a straight line through Luvungi. Mud footpaths led off into the hills between clusters of houses and small garden plots. Simple structures made of wood and covered with loosely woven dried leaves served as the central market. A meager array of local vegetables and a few dozen pungent dried fish the size of one's palm, a local delicacy, was spread out on wobbly tables.


I had come to Luvungi with a Congolese colleague -- not to talk to rape survivors or extract more details about the widely covered attack the previous year, but to speak with village elders about how the community had fared since. Still, our hosts were insistent. A couple of elders led us to a small house and promptly brought three women, one on her own, two in a pair, to speak to us.

It was "systematic rape," one woman said, adding that the rebels had sealed off the town. "They came on Friday and left on Tuesday," she said. "They came to the doors of the houses at around 11 p.m., and they forced themselves in," another woman said. "They put their hands into my stomach." All three described in graphic detail what the armed men had done to them.

When the interviews were over and we were out of earshot, my colleague and I stood in confused silence. I had interviewed survivors of rape in eastern Congo before; a psychological element seemed to be missing in these interactions. Before I managed to articulate the uncomfortable feeling that we had just been lied to, my Congolese colleague spit it out: "Those women have been coached."

I asked a Congolese health-care provider working for the Congolese Ministry of Public Health near Luvungi about the source of the mass-rape numbers. He wore a crisp, white medical coat with the dark-blue insignia of International Medical Corps. The clinic where he works -- state-run but IMC-supported -- is the only one in the immediate area. We sat in a small office with the door closed. He confirmed that between July 30 and Aug. 2, 2010, rebels had been in the area, pillaging houses and harassing people. But during the incident and in the days immediately following, he said, he treated only six patients who had been raped -- not hundreds.

As rebels moved into Luvungi and surrounding towns toward the end of July 2010, the clinic started receiving patients for a variety of ailments, he said. Of the six rape victims, two indicated that their assailants were civilians, not the armed men occupying the village. But most of the 100 or so patients he saw between July 30 and Aug. 6, when IMC medical reinforcements arrived, needed treatment for maladies common to this region -- malaria and diarrhea -- or for injuries sustained while fleeing to the bush during the occupation.

The first outsiders to respond on the scene were members of the IMC team, who arrived four days after rebels left the area. The health-care provider said he was then reassigned to the pharmacy, leaving the treatment of patients to the IMC staff. Patients, men and women, began arriving in large numbers, he recalled, and the IMC registered every woman at the clinic as a victim of sexual violence, even those treated for other ailments; this, he said, included revising the log of the patients he had seen before IMC's arrival. Asked why the team would do this, he hesitated for a few moments, then quietly offered: "I guessed that they were trying to bring up a high number, but what could I do?"

The IMC disagreed with the health worker's account, saying in a statement to Foreign Policy that "no revisions were made to patient logs" and that the reason for the increase in the numbers was that "many reporting survivors did not come forward for weeks after the attack.... Up until that point, survivors were simply too frightened to walk the distances required to seek medical attention." In response to questions about these numbers, IMC's Los Angeles-based communications director, Margaret Aguirre, stood by the figures. "As a humanitarian, service-focused organization, IMC does not ever attempt to 'verify' reports of rape," she said in an email. "We reported on the number of people we assisted with medical services who reported being raped. Our policy is to provide assistance that self-reporting survivors seek, without subjecting them to inquiry." She added later: "We did not discuss internally or distort these figures in any way."


As IMC's numbers of reported victims grew and media pressure ratcheted up, the United Nations began its own investigation. Media accounts of the Luvungi mass rape gave the impression that the U.N.'s Indian peacekeepers based down the road from Luvungi were nowhere to be seen, despite their proximity. But a restricted report filed by the unit closest at the time and obtained by Foreign Policy describes an incident on Aug. 2, 2010, in which a U.N. patrol in Luvungi apprehended a member of the Mai Mai Sheka, a local militia, who was found to have an AK-47, two magazines, 49 rounds, and one handmade grenade. According to the report, the U.N. patrol that day stopped in two villages, and peacekeepers observed evidence of looting by the FDLR and Mai Mai Sheka -- two of three rebel groups that the subsequent U.N. report claimed were responsible for the mass rapes.

Why did the villagers of Luvungi not tell the peacekeepers that rebels had committed rape en masse? And if there was indeed such widespread sexual violence, why did the peacekeepers report observing other criminal activity, while apparently remaining oblivious to or ignoring evidence suggesting that sexual violence of this horrifying nature and scale was occurring in this one-road town?

In the days immediately after the attack, peacekeepers nearby conducted an initial assessment, tallying between 37 and 42 rapes, according to those involved in the inquiry. Then followed three waves of civilian U.N. investigators. The first arrived in the area of Luvungi on Aug. 13, nearly two weeks after the rebel attack. As international pressure grew, the second investigation concluded in an internal report that 154 people had been raped, citing IMC records. Building on those visits, the U.N.'s local human rights division then carried out what it called an "in-depth investigation" during two trips in the fall, interviewing civilians and compiling all available reports from 13 villages around Luvungi. The report's official tally, now cited as the definitive count, asserted that "at least 387 civilians were raped by these combatants."

The United Nations says it stands by these final numbers, having cross-checked victims' accounts with "other sources." Asked how the U.N. investigative team confirmed the number of rape cases months after the incident transpired, U.N. spokeswoman Barbara Matasconi told Foreign Policy: "The verification of rape through medical sources is more pertinent for criminal investigators. For human rights investigators it is useful in terms of corroborating testimonies … but it is not a legal requirement." Asked why the number released by the United Nations was higher than all the others, Matasconi added, "We moved around with [a] team of experienced human rights officers, experts in protection issues and in interviewing victims of sexual assault/rape. These are all factors that make victims open up to us."

Clearly, the women of Luvungi opened up to the investigators. But how to reconcile the wildly disparate reports? Certainly, rapes were committed during the four-day occupation; the sources I spoke with all agreed on that. So why does it matter that the numbers may have been vastly inflated?

Because it distorts the nature of aid flows to Congo, say outside critics, as well as the concerned insiders I spoke with who were familiar with the various investigations. They worry that the numbers inevitably lead to a focus on the sensational, while ignoring the troubling underlying dynamics. While the military and civilian sides of the U.N. peacekeeping operation were internally at odds about the scale of the incident and the U.N. response to it, multiple sources on both sides, who asked for anonymity in order to speak candidly, said they were surprised by the number of rape cases cited in the final report. One peacekeeper deployed to the area at the time of the incident told me, "Ever since then I have held it in my heart. I want to understand what would motivate people to lie about this."

"MOST EMERGENCIES CENTER around one story and one category of victims," Congo researchers Maria Eriksson Baaz and Maria Stern wrote in an email to me. "In Liberia, it was the child soldiers, in Sierra Leone it was mainly the amputees, and in the DRC it was the raped women."

There's no question that the rape-as-a-weapon-of-war narrative has stuck. The prominence of this story about Congo's long conflict is in many ways a testament to the success of advocacy efforts, including the personal attention directed at the issue by high-profile figures like former U.S. Secretary of State Hillary Clinton and American playwright Eve Ensler, the creator of The Vagina Monologues. And the advocacy works: The month after IMC alerted the world to the Luvungi incident, the NGO announced it had received a $16 million grant from the U.S. government to provide services to sexual-violence survivors in eastern Congo, one of the largest-ever awards to an aid group working to combat sexual violence in the region. (IMC noted that the USAID grant was awarded in July, before the attacks in Luvungi, and was completely unrelated.)


No one suggests that giving millions of dollars to help this vulnerable, traumatized population isn't warranted. But many aid workers quietly say the strong focus on sexual violence, over all other issues and crimes, has created a perverse incentive structure. Simply put, organizations know that their programs are more likely to be funded if their beneficiaries are victims of sexual violence -- and women know that they will have a better chance of accessing medical care, school fees, microcredit, and housing if they report being a sexual-violence survivor.

In a 2012 report published by Wageningen University in the Netherlands, researchers Nynke Douma and Dorothea Hilhorst examined how funding to address the host of challenges Congo faces -- from its predatory army and police to its abysmal judicial system and the massive internal displacement of its people -- is apportioned. They found a disproportionate focus on rape. "[T]he sexual violence budget is nearly double the size of the budget for all security sector reform activities … and just under half the size of the entire peace building trust fund," Douma and Hilhorst wrote. Funding for internally displaced people -- there were an estimated 1.4 million in eastern Congo during the period they investigated -- is less than half the funding for sexual violence.

Today, according to Douma and Hilhorst, more than 300 organizations in eastern Congo's South Kivu province work on sexual violence; in 2002, fewer than 10 did.

FINDING THE SPACE for a quiet, candid conversation with a female resident of Luvungi, unorchestrated or unsupervised by village elders, is exceedingly difficult. In the flurry of attention following the four-day attack, citizens in this small town expected things to change. Perhaps a school would be built, the road paved, or maybe a tower for cell phones installed. None of that came to pass, leaving it now a place hostile to outsiders asking questions.

One night, on my second visit to Luvungi, after everyone had gone to bed, a middle-aged woman we had met that day slipped into the room where my interpreter and I slept. She pulled a small stool alongside the pallet where my interpreter was lying and quietly told her story. Not wanting to create a stir, I stayed across the room while they spoke in hushed Kiswahili. The next morning my interpreter explained: The woman said that the incident had been terrifying, worse than other bouts of fighting in the volatile area. People lost everything and were forced to hide in the forest. She said that a fighter caught her as she was trying to escape and raped her. But there weren't many like her, she said. And after the rebels left the village, elders decided that the community would say that many women had been raped to avoid ostracizing those who were. It was for the sake of community cohesion, she said. Once aid groups came it was important to protect this story so that everyone could benefit from the assistance that would surely flow.

Here was an explanation. But of course it was just one possibility.

More than two years on, a definitive account of what transpired in Luvungi will likely never exist. The legal process to bring justice to the alleged perpetrators has all but collapsed amid funding and security constraints; the only person ever arrested and charged in the mass-rape case, Lt. Col. Sadoke Kokunda Mayele of the Mai Mai Sheka, died in jail in August 2012, reportedly of malaria. But a new, wooden inpatient wing is under construction at the clinic. "We never have shortages of PEP [post-rape] kits, even though we sometimes run out of other medicines," said the Congolese health-care provider who treated patients during the four-day siege.

Last year, as I was on my way back to Luvungi for my third visit, the town was hit with a fresh wave of violence. Rebels had returned to the area, clashed with the Congolese army stationed there, and reoccupied several villages nearby. A rebel leader and Mayele's former boss, Ntabo Ntaberi Sheka, was angry that he had been accused in the U.N. report of orchestrating the mass rape. He had come back to this forsaken place to clear his name -- and perhaps exact his vengeance.

Sarah Elliott


The Fall and Rise of Raila Odinga

In Kenya's contested election, the tortured past of family dynasty is alive but not quite well.

NAIROBI — A third generation of leadership is emerging in post-colonial Africa, and with it a trend of sons being made to answer for their fathers. During Kenya's first-ever presidential debate, held three weeks ago in Nairobi, the moderator accused the two leading candidates of subjecting Kenya to a family rivalry that their fathers started a half-century ago and that the country needs to get past. The leading candidates are Raila Odinga, the prime minister, and Uhuru Kenyatta, the deputy prime minister. Their fathers were Jomo Kenyatta, the first president, and Oginga Odinga, his aide de camp and vice president -- before they came to detest one another.

Uhuru Kenyatta dismissed the moderator's charge breezily. As a politician, he knows to always keep to the future, and as a Kenyan politician he knows to avoid the past at all costs: Jomo Kenyatta, while an icon, also enriched his family and his loyalists beyond measure at the state's expense. (It's commonly said that the Kenyattas own a province's worth of land. That the claim can't be verified or disproven, such a mess are land titles in Kenya, only adds to its mythical quality.) Raila Odinga, however, couldn't resist the chance to deliver a history lesson. He reminded viewers that if not for his own father, his rival wouldn't be worth mentioning -- might not even exist. "If you go down the memory lane, my father spearheaded the trial for the release of Jomo Kenyatta from prison," he said, somehow managing to comingle (maybe a rhetorical first) nostalgia and peevishness in the phrase "memory lane."

It was a lesson no one needed. Kenyans are only too aware of the historical weightiness of this election. They have it shoved in their faces by their own press and foreign journalists every day. It's followed on by reminders of the last election, in 2007, which Odinga lost, probably fraudulently, leading to two months of chaos. In the Nairobi bar where I watched the debate, someone remarked: "Kenyans don't want to hear about the past anymore."

They don't have the choice. Odinga, who spent nearly a decade in prison for his political activism, is a walking reminder of Kenya's past. He seems to carry it in his bovine gait and breathless rasp of a voice. He's been a fixture of public life for 30 years, a member of parliament for 20. He's founded parties and disbanded them. The office of the prime minister was created for him, and it will be done away with when a new constitution, mostly of his making, is implemented. And while even his critics concede Odinga and his father must be remembered for what they've done for Kenya, his trip down memory lane came off as so much gloating. It also made him sound wistful, never a good note to strike on the campaign trail. He appeared to want to be alone with his reveries during most of the first debate, and a good deal of the second, this past Monday. Halfway through the latter he actually employed the analogy, "you cannot allow a hyena to protect your goats," while discussing the (hugely contentious) issue of land reform. Meanwhile, Kenyatta came off as swift and witty (his critics would say unctuous).

Odinga usually has an abstracted air about him. Still, it would have been indulging in denial not to come away with the suspicion that he has, like John McCain or Yuri Andropov, wanted to be head of state for so long he believes Kenya owes him the position, despite signs of his decline. (Sixty eight as of January, he is 17 years Kenyatta's senior.) This happens to be precisely the point on which he should be distinguishing himself from his rival. Everyone knows Kenyatta thinks the presidency is his birthright; he first ran for it as a feckless 41 year-old, having been inserted into parliament because of his name a year before. But not long ago, Odinga was thought to be above this kind of entitlement-chasing. Arguably no one has done more to reform the government than Odinga, as much a living martyr to the cause of Kenyan democracy as the country has.

So how did this old king -- whom the political analyst and former anti-corruption czar John Githongo described to me as "the gravitational center of Kenyan politics" and the "alpha male lion" of Kenyan public life -- get to fumbling on this lonely last-act heath? Destiny, first off. As with all statesmen of serious consequence, and tragic heroes, much of Odinga's story is not his own.

It would be difficult to overestimate how much Jomo Kenyatta's and Oginga Odinga's feud, one of the great forgotten power struggles of the Cold War in Africa, shapes Kenya to this day. While Kenyatta won the favor of the West by creating a market economy with small but highly productive land-holding and merchant classes, Odinga, supported by Moscow, held out hope for socialism. He couldn't deny Kenyatta's success in avoiding the growing pains of other African republics -- within a decade of independence, Kenya was among the most prosperous countries on the continent -- but their visions could do nothing but collide. After Odinga served as Kenyatta's vice president, in 1966 he ran against him for the presidency. No sentimentalist, Kenyatta in turn barred Odinga's party from meeting and the media from covering it, and won. In 1969, the president travelled to Kisumu, the Odingas' home on Lake Victoria, to address a gathering and affect a rapprochement with his old friend. Odinga loyalists jeered Kenyatta. He got incensed. "Those creeping insects of yours are to be crushed like flour," he said, looking at Odinga. "I have left you free for a long time because you are my friend. Were it not so, you yourself know what I would have done." A riot ensued, Kenyatta's agents shot into the crowd, and the president barely escaped. He had Odinga placed in detention for close to two years.

As Kenyatta declined into doddering tyrrany, Odinga became a folk hero, the man who would be king but never was. At home, however, he took after his rival, according to Babafemi Badejo's Raila Odinga: An Enigma in Kenyan Politics. Badejo, who describes the elder Odinga as a "slave-driver," recounts that once, after the young Raila refused to do some yard work, Odinga "floored him and started jumping on him with his gumboots." Unsurprising, then, that after flirting with the family politics -- he studied in East Germany, where he opened up an office of his father's opposition party, and named his first son Fidel -- Raila bolted the nest.

"He had his own charisma from the beginning. People think he began speaking up because of his father, but it's not the case," said Willy Mutunga, the Kenya Supreme Court Chief Justice, who taught at the University of Nairobi at the same time as Raila. "He was always his own man."

So much so, in fact, that in 1982, Raila helped plan a coup against Jomo Kenyatta's successor, Daniel arap Moi. The hideously brutal and cartoonishly corrupt Moi deserved no better, though he warranted a better effort. The plot was wrapped up easily and Odinga imprisoned. The torture began at once. He was beaten with a table leg by a police inspector, according to Badejo. Then he was placed in a room in an ankle-high pool of cold water and made to stand in it through the night. Then he was beaten again. That was the first week. He spent the rest of the 1980s in prison, much of it in solitary confinement, repeatedly tortured, but never tried for anything. He didn't see his children for six years. (Uhuru Kenyatta was attending Amherst College at the time.) He was interrogated in the bowels of Nyayo House, Moi's answer to Moscow's Lubyanka, the infamous KGB pen. "Detainees in Nyayo House were normally naked during interrogation," Badejo writes. "The detainee was usually under a flood-light while the interrogators sat in partial darkness. The detainee could see fresh blood on the floor ... as if a battle had just ended."

Willy Mutunga was also imprisoned by Moi. "No one came out apologetic," he told me. "There was no chance of you joining the dictatorship. Those years hardened [Odinga]." Prison also made him more ambitious, according to Kenyans I spoke with. "He's more hungry than his father was," a former government official who didn't want to be named said. According to a current official and scholar of Kenyan politics, who likewise did not want to be named, Odinga's involvement in the coup didn't endear him to all of his countrymen. Moi was hated, but on the whole Kenyans are politically conservative -- they'd rather see a despot ousted in elections, or die, than get deposed by a cabal. "Up to now, that's what's made him not become the president of Kenya," the official said. "Because there are a lot of people who fear him, who don't understand him. He's an enigma. What kind of man are we talking about? How far can he go?"

Odinga was released in 1989, then imprisoned again; then released again, then imprisoned again. In 1991, he fled Kenya, disguised as a priest. The next year, when Moi opened up the elections to new parties -- and evicted and harassed whole swaths of the population, initiating a cycle of tribal-political violence that continues to this day -- he returned. The Cold War was ending and the Soviet Union, the Odingas' political patron and touchstone, was about to disappear. "He was very much exploring a different way," recalled Al Eastham, a political officer in the United States embassy in Nairobi at the time. "He saw the world was changing. He was trying to find a place more in the mainstream." Eastham informally advised Odinga on how to expand his political activities without getting thrown back in prison. "He was very smart, very adaptable," Eastham said, "and very ambitious." He added that in the three years he served in Kenya, Odinga was the only person who invited him to his home for dinner.

Oginga Odinga died in 1994, after three decades as the living embodiment of Kenya's opposition movement. Before he did, he stunned his supporters by reconciling with Moi. "Perhaps because the majority of us in Kenya have never lived as adults under multiparty rule," he told Time, "many of us have not yet fully grasped the idea that the role of the opposition is not to engage in confrontational stand-offs with the government every day of the week."

Raila took the advice to heart. In 1997, he merged his party with Moi's. But then he ran against Moi for the presidency. It was a bizarre turn of events, and one result was the worst election violence Kenya had seen in its history. For Odinga, there was a more enduring effect: looked upon as the spiritual inheritor of Kenya's principled opposition, the additional perception now arose that power was more important to him than principle. "Many people said wow, this guy, he can go to bed with anybody, so long as there's a chance to win political office," the current official said.

"Raila has always said that he is a social democrat. Without a doubt he was always a nationalist and patriot," a prominent Kenyan jurist told me. "But when it comes to social democracy, he vacillates. You get the impression that sometimes he is on the left, other times on the right."

By the 2000s, Odinga had become something more -- the savior of the Luo ethnic group, which is based in Nyanza Province, in Kenya's west. Like many of the rest of Kenya's 40 or so tribes, Luos often believe they've been excluded from power and property by Kikuyus. The country's largest tribe, Kikuyus have, since the days of Jomo Kenyatta, made up a good deal of the country's land-holding, business, and political elite. (The Kenyattas are Kikuyu, as is Mwai Kibaki, the current president. Moi is Kalenjin.) Oginga Odinga had been not just the symbol of opposition, but of that exclusion.

His son, in turn, took on messianic expectations. Luos began calling Raila "Agwambo", the Mysterious One. (Bullish of build and indefatigable, he was also known as "Tinga," or Tractor). "To us, Raila embodied our collective struggles, aspirations and dreams," writes Miguna Miguna, a former aide. "It was as if Raila had convinced an entire community that he was both invincible and indispensable." The sentiment isn't confined to Luos. In a society raw with feelings of dispossession, Odinga represents the promise of a future liberated from tribal domination. He symbolizes a kind of rebirth of the independence movement. (Conveniently for his rally speeches, Kenya marks 50 years of independence this year.)

Some Kenyans worry he'd do anything to pander to his following.  "He's very populist. That's another thing people fear about him. He doesn't want to lose Luo support, and he may make wrong decisions in order not to," the official said. In his campaign, Odinga has been promising to carry out a nationwide investigation of land titles, and to evict and resettle Kenyans based on it. An impracticable promise -- untold numbers of specious titles have been issued over the years, and mass displacements, a tactic employed most famously by Moi, would certainly lead to violence -- it's nonetheless music to the ears of Luos and others who believe Kikuyus have stolen their land. At the same time, Odinga has seemed to grow more and more fickle, forming parties and then abandoning them, forging alliances and breaking them.

In 2002, Odinga took over Kibaki's campaign when the candidate was injured in a car accident, and brought in the Luo vote. In return, Kibaki gave Odinga a series of ministries, and another charge accrued to him: corruption. Hardly a month goes by in Kenya without some official being charged with it, of course, but Odinga has become so associated with presumptions of venality that the moderator in the second debate could, in total sincerity, introduce a question on the subject with this preamble: "Mr. Odinga, you've been in public service for a while, and so you are associated with several scandals..."

Others say that if he's not corrupt, Odinga is so negligent about his duties that he may as well be. Recently I was in Kibera, Kenya's (some say the world's) largest slum. In parliament, Odinga represents Kibera, where it's easy to find residents who say he's done nothing to improve their lives. They can get quite heated about it. "He's fucking with our brains," one man told me. "He can't even get us toilets here."

This is the chief complaint made by the man who's become Odinga's most outspoken critic -- not Kenyatta, but his former aide, Miguna Miguna. He was Odinga's political strategist and speechwriter, before Odinga fired him, in 2011. In a memoir of their relationship published last year, Peeling Back the Mask, Miguna calls Odinga "the quintessential opportunist." Odinga is surrounded by corrupt lackeys, Miguna told me when I met him recently. He added that his former boss is duplicitous, greedy, nepotistic, cruel, and "morally bankrupt." He claimed he's been attacked and beaten on Odinga's orders. He also claimed his maid was out to poison him, so it was hard to take all of his claims seriously. Miguna's main detraction, however, is his most sober: namely, that Odinga is exactly what he and his supporters accuse Kenyatta of being -- fundamentally unserious. An entitled amateur. "I thought constitutionalism and rule of law were the most important things to this man," he said. But Odinga combined "a lack of depth and absence of interest, a combination of which is dangerous for a leader." 

At a certain point, none of this matters. An air of inevitability now surrounds Odinga. It's not just in how he regards himself, but in how he's regarded, by supporters and opponents alike. "Kenyan politics is now perceived in the light of whether you're pro-Raila or anti-Raila," said Fred Amayo, a businessman who is running to take over Odinga's assembly seat. John Githongo called him "the big beast on the plain of our politics" This air of inevitability only grew thicker as Odinga broke with Kibaki over efforts to write the new constitution. After the fallout, he ran against his old boss in 2007. (There's a fair argument to be made that Odinga would have found any reason to run.) In the resulting chaos, both men allowed their proxies to kill each other. But Odinga came out the moral victor. When a new constitution close to his own template was approved in 2010, his future as head of state seemed all but assured.

Two and half years later, it is anything but. After the debates, Kenyatta took a lead in many polls. This could not be more galling to Odinga's supporters, and to many scholars of Kenyan history, who appear to overwhelmingly hope for an Odinga victory (as do Western governments, thanks to the International Criminal Court's indictment of Kenyatta). They tend to see an Odinga presidency as a long-needed redemption. Kenya should elect him to finally confront and come to terms with its past, the logic goes. "Kenya needs to get him out of its system," is how a friend put it.

Kenyatta's supporters see no such need. If they don't feel that Odinga has invented his personal history of adversity in order to gain power, they are simply bored by it. As the man in the bar said, they're sick of hearing about history. To them, the younger, swifter, more arrogant Kenyatta is the way to get past the past.

Whoever wins, large parts of the other side will claim they cheated, and violence will almost certainly erupt somewhere. There is little question that if Odinga loses, the violence will be worse. His most avid supporters see his victory as foregone, a historical necessity. A friend who studies tribal politics interviews a lot of Luos. Their conviction that Odinga will -- must -- be the president has the ring of prophecy to it. "It's not possible," they say, baffled at the question, when she asks what they'll do if Kenyatta wins.

Towards the end of the second debate, the questioning arrived at the Kenyatta family's land holdings. The discussion had covered unemployment and corruption, and the moderators had dug into the candidates with brio. However, this was the moment every Kenyan, or at least every Kenyan who isn't Kikuyu, had tuned in to witness. Uhuru Kenyatta was finally going to be held to account for his family's outsized wealth. (Forbes estimates he personally owns 500,000 acres of Kenya.) The scion was finally going to have to answer for his father's sins.

But before it could get started, the grilling was doused -- by Odinga, of all people. To everyone's surprise, he jumped into defend Kenyatta. "He was just an innocent inheritor; he didn't commit original sin," Odinga said. He then launched into another history lesson, tracing Kenya's land disputes back to the days before independence. "The issue is there was a betrayal of the freedom fighters," he said. "Those who sacrificed most were completely abandoned by the leadership."

He was right. But it didn't matter. By that point, people had stopped listening.

Mackenzie Knowles-Coursin