Institutionalization of Ethnic Categories

Key Dates


Key Dates

  • 1891: Nyasaland established as British protectorate
  • 1964: Malawi gains independence
  • 1994: First elections under multi-party democracy

back to top



As in other British colonies in southern African, race was institutionalized under colonial rule, but clearly to a lesser degree than in the more extreme cases (i.e. South Africa, Southern Rhodesia). The colonial state implemented its rule within a system that made distinctions between “natives” and “non-natives,” and with a distinction for Asians or Indians. Such institutions mostly fell away at independence or soon thereafter.

For example, a race question was asked on all censuses between 1911 and 1987, but was not included on the 1998 and 2009 censuses.

The ‘Thangata’ system under colonial rule was essentially a system of forced labor for Africans (Page 1978). It is not clear whether the system made explicit distinctions between Africans and Europeans or if this was simply how it functioned in practice. Power (1993: 578) points out that unlike in Rhodesia and South Africa, Africans were allowed to hold many skilled jobs. Nonetheless, trading licenses were all categorized in terms of key racial categories such as European and Indian, with separate distinctions for Africans (Power 1993: 589).

As in most other colonial systems, racial categories were used as fundamental distinctions within legal and trade policies. For example, Lee (2005: 456-7), in describing some ambiguities associated with the status of individuals of mixed race origin in colonial Nyasaland, also mentions the critical legal distinctions between “natives” and “non-natives.”

The Asian category has been fairly well institutionalized, and to a degree this has persisted into the period of independence. Asians have only been allowed to own property in certain places in Malawi: Lilongwe, Blantyre, Zomba and Mzuzu (Carver 1990: 60). The state also enacted separate marriage laws for people of Asian descent, which allowed people to celebrate marriages in accordance with Hindu or Islamic traditions (Kapindu 2009). While the Asiatics Act was repealed in 1967 (Roberts 1968), as far as we can tell, the other institutions remain in place.

Neither Nysasaland nor independent Malawi have ever been highly diverse in terms of traits recognized as “racial,” and we are not aware of any time during which the “European” or “Asian” population groups ever passed the one percent population threshold.

back to top


Although ethno-linguistic and related (often referred to as tribal) categories are generally thought to be important in Malawi, we find only a weak institutionalization of such categories (Chewa, Tumbuka, Yao, etc.), because available evidence reveals little consistency in the use of particular labels or in the stability of the institutions themselves.

Language was enumerated periodically throughout Malawi’s history. It was recorded for the first time in 1966, when the census asked respondents about their home language and comprehension of other languages. Language questions were removed for the 1972 and 1987 censuses, but reappear in 1998 and 2008. However, even the language question did not provide opportunities to identify all possible languages, asking instead, can (name) “read and write the following languages – English, Chichewa, Other,” providing opportunities for Yes/No responses on each, a practice which we do not classify as an instance of institutionalizing ethnic categories.

From 1968 to 1994, Chewa, the language spoken by a majority of the population, was declared the only national language, while all other African languages were banned from government, schools, and the press (Britannica Online Encyclopedia, Kaspin 1995: 607). While certainly discriminatory, using our framework, we treat this as an example of non-institutionalization of ethnic categories.

A regional quota system, in place until 1993, managed enrollment in secondary schools. This policy favored students from the Central Region by requiring higher grades from northerners and southerners to qualify for places in secondary schools (Vail 1989: 183). A similar quota system was reintroduced in 2010, affecting enrollment in secondary school and university (University World News 2010). While it is true that regional borders typically coincide with ethnic boundaries, we do not classify this as a case of institutionalizing ethnic categories because relevant categories were not explicitly used, and the regions are not named in terms of ethnic categories.

An effort to address perceived imbalances in power and resources along ethnic lines generated a proposal for institutionalization of ethnic categories. Article 68 states that a nominating committee should be responsible for the composition of the Senate and that, “The Nominations Committee shall endeavour to ensure, when considering nominations, that the Senate is proportionally representative of the various groups in Malawian society and therefore shall seek to ensure, so far as it is possible, that one-half of the members of the Senate are women. “ However, no categories are specifically identified, and more important, the senate has never been called into session (Mutharika 1996; BBC 2001). The Constitution also dictates that, “any boundaries determining the territorial jurisdiction of any local government authority shall be geographical only, without reference to race, colour, tribe or ethnic origin of the inhabitants of the area”.

Kaspin (2008: 598) points out that ethnic maps of Malawi in 1950 and 1965 depicted very different groupings. Although this evidence is not a direct reflection of institutionalization, it reflects the resulting ambiguities concerning which categories were socio-politically salient.

back to top


We find minimal evidence of institutionalized racial categories. Religion was enumerated on all censuses between 1911 and 1945, as well as on the 1966, 1998, and 2009 censuses.

As mentioned earlier, the 1994 Constitution provides for a Senate with balanced religious and tribal representation; however, a Senate has not been created (Mutharika 1996; BBC 2001).

back to top


None found

back to top


None found

back to top

Other ethnic, including tribe

In the post-War era, tribal affiliation was only enumerated in 1945 and 2008. Although we have not been able to confirm exactly what was asked on earlier censuses, Marjomaa (2003, fn66-7) mentions the size of tribal groups based on the 1926 census, and a government report on the census (Census Report 1926: xxv), including the categories Yao, Nyanja, Chikunda, Nguru, Ngoni, Chewa, Tonga, Nkonde, Tumbuka, and Wemba.

Under colonial rule, military recruitment policies favored the Yao – a category treated by the British as a “martial” race (Marjomaa 2003). While Marjomaa (2003: 430) identifies intelligence reports making explicit distinctions among the qualities of different “tribal” groups, it is not clear whether such distinctions were made explicitly. In any case, the institution was dissolved around the type of the end of the Second World War (Majormaa 2003: 430).

back to top


back to top


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s