Over the last five years, electoral politics in Kenya have been in a state of upheaval.
The east African country held elections in 2013, between Uhuru Kenyatta of the National Alliance Party and Raila Odinga of the Orange Democratic movement. Kenyatta was eventually declared the winner, with 50.07 percent of the vote, after a ruling from the country’s Supreme Court. Kenyatta then beat Odinga again in 2017, but the Supreme Court annulled the voted before the pair agreed to “resolve our differences” in March 2018.
Both of these elections were characterized by three factors: contested results, a hyper-partisan electoral environment often marked by violence, and the Kenyatta campaign’s use of the controversial firm Cambridge Analytica to help win the election.
Cambridge Analytica’s CEO, Alexander Nix, boasted to undercover reporters from Channel 4 that his firm had “rebranded [Kenyatta’s] entire party twice, written their manifesto, done two rounds of 50,000 surveys…Then we’d write all the speeches and we’d stage the whole thing. So just about every element of their campaign.”
The recent revelations that Cambridge Analytica used the Facebook data of 50 million Americans, taken without their permission, to help them micro-target voters during the 2016 presidential election, has set off a firestorm. But the roots of the problem extend far deeper than one country and one election. Cambridge Analytica honed its techniques in a host of countries with political institutions that are younger, more fragile, and far more vulnerable to interference.
“The problem that you have in Kenya, as in a lot of these countries, is that there are these very deep underlying socio-economic tensions and ethnic divisions,” Patrick Merloe, director of electoral programs at the National Democratic Institute, told ThinkProgress. “The kind of politics that were talking about is feeding on and reflecting those underlying social tensions.”
In Latvia, Cambridge Analytica’s parent company, Strategic Communication Laboratories (SCL), was accused of deliberately exploiting ethnic tensions during the 2006 national elections between Latvians and ethnic Russians (38 percent of residents in Latvia consider Russian as their native language).
“In essence, Russians were blamed for unemployment and other problems affecting the economy,” an SCL document seen by Bloomberg News said. While the candidate SCL was advising, Ainars Slesers, never joined the governing coalition despite his party winning eight parliamentary seats, it still shows how exploitation of ethnic tensions was a global modus operandi for Cambridge Analytica.
The impact of their strategy — spreading fake news, misinformation, and ads designed to stoke societal divides on social media — was clearly seen in the Kenyan election, which was marred by misinformation and hyper-partisanship. In one survey, 90 percent of Kenyans said that they’d come across some form of fake news, a highly significant statistic considering 88 percent of the population has access to the internet, and the country boasts broadband speeds faster than those in the U.S.
This connectivity, coupled with longstanding tensions between different tribal factions (which were often exploited by the British when they ruled the country) made Kenya an ideal environment for the type of campaign Cambridge Analytica bragged about running, one based on fear rather than facts.
Viral videos circulated on Kenyan social media claiming Odinga would revoke the constitution and trigger a health crisis that would leave women “giving birth in the streets” while “Al Shabaab launches attacks over the country.”
“There were definitely such ads depicting Odinga as the end of history and on the other side about [Kenyatta] having warehouses of premature ballots,” Merloe said. “There were lots of examples of negative targeting in order to motivate one side. More was done against Odinga.”
Merloe added that the negative online campaigns weren’t at all limited to Facebook, and that more were actually seen on WhatsApp. More broadly, however, he said it pointed to a world in which firms like Cambridge Analytica could influence the elections of developing countries from afar, with little chance of those countries being able to demand accountability from them.
To make matters even worse, security services and defense contractors have expressed a continued interest in the techniques used by Cambridge Analytica, raising the possibility that they could be honed and perfected to further exploit vulnerable democracies worldwide.
This relationship is seen most clearly in the case of Michael Kosinski, a Cambridge University researcher. Kosinski helped develop an app that measured personality traits through Facebook “likes.” That technology was then used by Cambridge Analytica to help microtarget specific voters based on their Facebook data.
According to The Guardian, Kosinksi’s PhD was funded by Boeing, a major defense contractor, while the U.S. government’s Defense Advanced Research Projects Agency (DARPA) is cited in two academic papers that supported Kosinski’s work. Kosinski’s CV also cites a talk he gave to the Ministry of Defense in Singapore on “technologies that may revolutionize psychological assessment.”
There’s even more evidence of the military applying the techniques honed by Cambridge Analytica in the links between the firm and governments in both Britain and the United States.
In the U.K., Cambridge Analytica was given “List X” status by the Ministry of Defense, giving it access to secret government documents. In the U.S., meanwhile, SCL has a contract with the State Department’s Global Engagement Center — which, ironically, was the part of the State Department tasked with combating foreign interference in elections.
Before whistleblower Christopher Wylie even started working for Cambridge Analytica, he boasted about wanting to create the “NSA’s wet dream,” again showing the potential military and intelligence application of the work he ended up doing for Alexander Nix.
The extent to which Cambridge Analytica’s techniques are actually effective in influencing elections has been disputed. Tom Dobber, a doctoral candidate studying political microtargeting at the University of Amsterdam, told The Verge that Cambridge Analytica was a “better marketing company than a targeting company.” Paul Manafort, the former Trump campaign chair who is no stranger to the political dark arts, was reportedly left unimpressed by Cambridge Analytica’s sales pitch.
However, the company’s track record of interfering in the elections of developing countries, coupled with the continued interest that the security and intelligence services have paid to the techniques developed by the firm, raises worrying questions about their possible use in future elections.
“They have been doing these [interference techniques] in a number of countries,” NDI’s Patrick Merloe said. “Now they’re taking them to other levels.”