Printer-friendly versionSend by emailPDF version
Tom Maruko

Concerned by government statistics on Kenya's ethnic composition, Samuel Abonyo laments the enduring difficulty presented by the concept of tribe in Kenyan society. First developed under the auspices of colonial-era governance, the practice of tribal geography, writes Abonyo in this week's Pambazuka News, is still going on in today's Kenya. But what really is a tribe? What is the government actually counting? Figures on tribes, Abonyo concludes, remain at best inaccurate and at worst highly damaging in depicting false, divisive categorisations.

'Tribalism will live for at least another 50 years', Daniel arap Moi said in 1957, historian Keith Kyle tells us in 'The Politics of the Independence of Kenya'. Moi’s prophesy has been fulfilled, and his contribution to its fulfilment is huge. In the 1950s, his construction of the Kalenjin tribe had begun in earnest, and by the 1990s, Kalenjinisation was an established word in Kenya. Yet the existence of the Kelenjin tribe is still being contested.

But what is tribe?


A tribe, according to Oxford Advanced Learner’s Dictionary, is a group of people of the same race, and with the same customs, language, religion, etc., living in a particular area and often led by a chief. Webster’s Dictionary says that a tribe is any aggregate of people united by ties of descent from a common ancestor, community of customs and traditions and adherence to the same leaders. Evidently, these definitions are not of much help, as, according to them, any group of people can conceivably be a tribe.

Peter Skalnik, an anthropologist, believed that tribes were politically defined units having dimensions such as culture, language and territory. To that strange belief, he added the weird opinion that basic tribal identities are ancient, powerful and closed to amelioration, with the result that hostility and tensions break out when members of different tribes come into contact. Skalnik’s definition of tribe is definitely an exercise in pure futility.

In 'Ethnic Groups and Boundaries', social anthropologist Fredrik Barth says that, as they are understood in social anthropology, tribal groupings 'are categories of ascription and identification by the actors themselves' that structure interaction between people. In the opinion of social anthropologists, tribe has other attributes in addition to that basic one. A tribe, they believe, is largely biologically self-perpetuating, shares fundamental cultural values, makes up a field of communication, and has a membership which identifies itself and is identified by others as constituting a group different from other groups of the same order.

I have applied that meaning of tribe to my own tribe, the Luo, and it has not worked.

A tribe is a label. A tribe is a logo. A tribe is a categorical identity which classifies you in terms of the biological background assumed to form your ancestry. A tribe is a socially defined biological master status others – who are excluded from it – use to recognise the difference between you and them and which you use to distinguish yourself from them. The other has its own socially defined biological master status. A tribe is a socially defined master status from which, because it is strictly enforced by sanctions of all sorts and the many mechanisms of social control that are the cages in which our lives are kept, those it includes and those it excludes can escape only at the price of achieving the status of social deviants. As we know, however, most people conform to the rigidities that are their lives, so that the tribe’s stranglehold on us is immensely powerful indeed.

Once fully constructed, tribes tend to stick like leeches.

But they are not concrete, they cannot be seen, they cannot be touched and they cannot be counted. They are not real. But they count. And they have real and palpable consequences.


We are members of our tribes. But tribal membership does not constitute tribalism. The existence of tribes is not a necessary and sufficient condition for tribalism to occur. For tribalism to arise, a tribe in itself must be transformed into a tribe for itself. In pre-colonial Kenya, for example, there was no tribalism, even though we had tribes. But tribes were then not tribes for themselves. Tribalism was at the time not a reality, let alone the paramount reality it is now.

It goes without saying that we fell from tribe to tribalism because of colonialism. The colonialists exploited our cultural pluralism to create tribalism. The colonialists brought with them Western nationalist discourse and ideology. Because of the discourse and ideology of nationalism, and the Western criteria for success and achievement the colonialists transplanted into Kenya, tribesmen began referring to their lots as better than their neighbours, or more advanced or superior in some way. That was tribalism.

To institutionalise tribalism, the colonialists established administrative units that were almost the same as tribal ones. The practice of tribal geography, an effective means of maintaining tribalism, is still going on in Kenya.

Colonialists lived in dread of African unity and fought hard to prevent it from arising. As late as the 1930s for example, colonial administrators sought to control the activities of the Roho Musanda, lest the members of that movement should proselytise among non-Luo communities. The 'religion' of Odongo Mango, the founder of the Roho Musanda, shows his theology was for Africans. As Cynthia Hoehler-Fatton reports in 'Women of Fire and Spirit: History, Faith and Gender in Roho Religion in Western Kenya', it was the colonial administration that turned the movement into a Luo thing.

Once the colonialists had institutionalised tribalism, it now determined the life chances of individuals. Tribes were now politically significant. They now had leaders or spokesmen. They could now be represented as acting agents. They had now gone past beginning to call themselves Luo and Nandi, etc., to borrow a phrase from the nobleman in George Bernard Shaw’s 'Saint Joan'.

But still – and this is one of the simplest but most effective tricks used to maintain politically important collective identities, and it is to keep our tribal identities alive – the Luo, the Nandi and so on have to be continually invoked and pronounced by 'authorised' persons like writers, priests, prophets, politicians, journalists, administrators and tribal statisticians.

And that is how tribal statistics participate in the maintenance of tribalism.

But it is not only the role of the statistics in tribalism that is the trouble with them, they are also of poor quality.


It must be allowed that, if tribal statistics were up to standard, they would be useful. We know there are ethnic inequalities in Kenya. Those who want to reduce the inequalities may use the statistics to establish their causes so that they design appropriate policies. Even just confirming what we already know would in itself be good enough, since we would be confident that our policies are based on fact.

The problem with counting tribes and their members is that we do not know what we are counting. Nobody really knows what a tribe is. Even the government does not know. And the way tribal membership is defined may also vary from tribe to tribe. For example, is it tribe at birth that is your tribe? Or is it acquired tribe? And if Kenyans were allowed to state more than one tribe, that is, if the question on tribe were open-ended, then we would have cases of dual or multiple tribal identities, even though tribe is a categorical identity.

Further, we inaccurately count what we do not know. And we incorrectly aggregate the figures.

The practices of tribal statistics discriminate on grounds of ethnicity. The statistics do not recognise the identities of tribes like Ogiek, and that is ethnic discrimination. The statistics amalgamate diverse tribes into fictional political identities such as the Kalenjin 'tribe'. The statistics have, for example, divided the Luo tribe into the Suba and Luo tribes.

The statistics paint a misleading picture of the ethnic composition of the country.

And there is no proper reason to collect the data. The government has not documented its claims that the statistics are used in planning. It has merely asserted falsehood after falsehood.


* Samuel Abonyo is studying for a PhD in Sociology at the University of Oslo.
* Please send comments to [email protected] or comment online at Pambazuka News.