“Whether and how to classify people on the basis of identification is an enduring political question,” said Samantha Balaton-Chrimes of the School of Humanities and Social Sciences, Deakin University, Australia. “It’s obviously particularly relevant in countries like South Africa with a deeply divided past but it’s found across the globe.”
She pointed out that in her own country, Australia, “you are routinely asked if you self-identify as Indigenous. Uganda uses a list from 1926 to identify 65 ethnic groups entitled to citizenship, a list that has been contested many times. In Cameroon identity is bureaucratised by the state on the basis of religion not language or ethnicity. Denmark has three categories – Danish born, Danish by descent and immigrants – with a special classification created in 2020 for Muslims. In China people are asked to self-identify in the census into 56 groups based on classifications made in 1982. Sweden identifies five national minorities and languages, and often experiences controversies over immigrants and ancestry beyond those.”
But the formal classifications can’t be taken for granted. In her research on ethnic categories in Kenya, she explained “there are extra-textual things going on and text alone misses a big part of the story.” While it is common in political sociology to assume classifications reflect a desire, on the part of governments, for order, legibility and governability, this is not always the case, and “a desire for order is not the sole guiding principle. Classifications also reflect a desire to be inclusive, and a competing desire to be divisive and self-serving in some instances.” As a result, Balaton-Chrimes showed in her presentation how the identity classifications used in Kenya are not ordered, consistent or clear.
Balaton-Chrimes’ presentation looked at how minority and marginalised groups are classified in Kenya, a category entitled to special provisions under Kenya’s 2010 constitution. This work is part of a book project to understand how the Kenyan state classifies its citizens by ethnicity, and with what effects. She offered an analysis that ethnic categories in Kenya operate within what she called a ‘cultivated vagueness’ where precision and consistency are avoided. This, she showed, sometimes facilitates inclusion and recognition of small and marginalised groups, and sometimes facilitates manipulation of resources among those more powerful.
“There is potential in uncertainty” she argued, because “Identities are uncertain and uncertainty can sometimes be beneficial. But whether it is or not can only be worked out in the detail – not achieved by resolving some kind of finalised list of identities.”
Balaton-Chrimes described Kenya as a complicated plural ethnic landscape. The population stands at 47.5 million, 99% of whom identify, in the census, as being from specific ethnic groups. There is no single dominant majority – the largest is the Kikuyu, smaller than 20% of the population, and one of the so-called ‘Big 5’ ethnic groups – “a safari term colloquially used to evoke the fear and political competition that sometimes exists between the big groups”. There are, depending on which list you consult, anywhere between 38 and over 100 different ethnic groups in total, some of whom can be considered subgroups of others. Less than 1% of the population is made up of Asians/Indians, Arabs and Europeans – defined by race, not ethnicity.
The 2010 constitution was considered progressive with one of the driving goals being to rectify how the previous constitution had dealt (or rather not dealt) with ethnic divisions.
“Violence following the 2007 election was largely around land disputes, long-term neglect and poverty in some regions, as well as perceptions around ethnic state capture, so there was lots of expectation around the constitution-rewriting process,” explained Balaton-Chrimes.
The constitution offered two solutions – devolution into 47 counties with budgetary and legislative control giving more groups power and a place of their own; as well as affirmative action for ‘minorities’ and ‘marginalised communities’ in government, employment opportunities and resource allocation.
“But,” said Balaton-Chrimes, “it was vague on definitions, and doesn’t clarify who can access those benefits”.
Devil in the details
“We need to work through the details of how identity classifications have been put to work and the actual practices activated,” she said. She outlined some of the work of government, legal and civil-society bodies to operationalise codes for ethnic minority and/or marginalised status saying: “There are several ongoing efforts to define who belongs but these are not agreed upon – and probably never will be – it’s messy without systematic methods”.
Starting with the law, she pointed out that coding is not legislated through parliament and the cases that have come before the courts often sidestep detailed definitions and classifications. This is so even when international law is more precise, such as in the recent judgement by the African Court on Human and Peoples’ Rights that reparations must be made by the government to the Indigenous Ogiek people of the Mau forest.
She also pointed to three commissions.
The National Gender and Equality Commission established in 2011 draws on a broad and unjustified set of criteria to define 85 ethnic groups as minority and/or marginalised. This includes, among others, size, way of life, history, socio-economic status, and geography. “But although their reports appear rigourous with tables and graphs, their methods are weak and they provide no real justification for how the groups are chosen. Many civil-society and community organisations have rejected the findings.”
The Commission for Revenue Allocation established in 2010 has a mandate to determine how an Equalisation Fund should be spent, a fund designed to promote development in neglected areas. “It is notable both for using quantitative data and for defining marginalisation by area, not ethnic group,” said Balaton-Chrimes, “an outlier among all the efforts to make such classifications”. It does, she noted, also list four ethnic communities as in need of special project funding based on extreme levels of marginalisation but “this list was arrived at via a rushed consultancy, and the consultant himself noted many more groups could be added”. The list, though, has not attracted much attention because the Fund has not been disbursed.
The National Cohesion and Integration Commission established in 2008, also found itself making a list of minorities and marginalised groups. It did so in its audits of ethnic diversity in the public-service in 2016. “The reports make interesting reading,” said Balaton-Chrimes. “They determine ethnic categories in different and inconsistent ways, loosely based on the 2009 census, but with various weakly justified changes. In defining minorities or marginalised groups, they are vague and evasive on choices. But, it’s sensitive work so perhaps from their perspective, less precision is probably better.”
She also focused on the Katiba Institute, a legal civil-society organisation that has done ethnic community mapping. “They are the most frank,” she said, “openly acknowledging that ethnic communities want their own in power. They have done qualitative field research producing a thematic report but not a fixed list. Focusing on themes and how they arise with different groups means the Katiba work is probably the best reflection of the kinds of communities who should be considered minorities and marginalised. Avoiding a list, in this case, made for a much more sensitive outcome in terms of ‘mapping’.”
The messiness, inconsistencies and lack of clarity mean that nobody really knows who Kenya’s ethnic groups are, and who is minority or marginalised. This leaves communities wanting recognition to chart their own path, which many do by lobbying for a ‘code’. They typically imagine this will lead to material benefits as well as a sense of belonging. Two communities have been successful officially becoming the 43rd and 44th ‘tribes’ of Kenya (the first 42 are not listed anywhere), encouraging other groups to want this as well. “Groups still want the codes even though they are not linked to official policies, and don’t guarantee citizenship or affirmative action.”
She highlighted the Nubian community (originally brought to Kenya under British rule as soldiers) – the subject of her first book – who have undertaken decades of lobbying. In 2009 they were given a code by the statistics bureau but this wasn’t enough and now they are seeking another – from whom is unclear.
A fool’s errand?
“Overall there is no consistency in the lists. No lists of ethnic groups are draw up on the basis of quantitative data related to disadvantage, with no ethnically disaggregated data available, and many different ways of defining ethnic groups and minority or marginalised status. But does this even matter? Should such list making continue?” asked Balaton-Chrimes. Answers are not clear.
“For communities wanting codes it’s about inclusion and access, a signifier of inclusion when there is no other guarantee,” she said. Material measures and accessing benefits are the abiding concern for most of these communities, some of whom endure extreme poverty. “It makes sense they would seek recognition in this way, and anticipate development as a result.”
“But in practice, ethnic ‘coding’ can also be problematic” she explained. “It can be divisive, incentivising further splintering of groups to claim minority status, cultivating stereotypes, or allowing the manipulation of categories for personal gain and privilege, and diversion of resources from those who need them.” Not only that, but classification can never fully account for how people experience their ethnic identity. “Classification, by its nature, makes no allowance for hybrid identities”, she explained, and “some groups are always excluded.”
Lists are not accurate reflections of ethnic realities, but are rather results of “layers of bureaucratic and political sediment” that will forever be flawed.
Referencing Frances Nyamnjoh and others, she argued for an understanding of identity as inherently incomplete, as an abstraction and a state of becoming that is open minded and open ended. “This is in sharp contrast to fixed lists” she suggested. “If there is a way lists can be flexible, as identities are, then perhaps they can be worked with to cultivate democratic outcomes in an ethnically plural society”.
This might also, Balaton-Chrimes’ offered, help tackle the idea that ethnicity is all that matters in Kenya. “Ethnicity in Kenya does not always trump other identities but it does often when it comes to access to the state,” she said. “State capacity has been used for ethnic patronage consistently since 1963. Fixed lists often re-inscribe ethnicity as the identity of relevance. But it should be one among others.”
The question is not, then, whether or not classification should take place, but how. “Difference-blind state equality based on just being human is a fictional idea – it doesn’t exist and also implies there is something wrong with an African ethnic identity” she argued. So the hard work of being vigilant about the use of lists, calling for their flexibility, and making sure they are targeted to particular democratic projects is the only way to make them work in constructive ways.
Michelle Galloway: Part-time media officer at STIAS
Photograph: Ignus Dreyer