Inventing traditional authority: Lhomwe chiefs in Malawi

Abstract

Chiefs in Malawi exercise significant authority based on legitimacy derived from the position's purportedly deep historical origins. But, does such legitimacy confer when a new chieftaincy is created from scratch? I address this question within the context of an ongoing cultural revival of the Lhomwe ethnic group in Malawi, which has included the appointment of many new Lhomwe chiefs and the creation of a Lhomwe paramount chieftaincy. Using a combination of in-depth interviews and original survey data, I explore the political drivers, mechanisms and implications of the this newly created Lhomwe chieftaincy. I find that its creation was driven by top-down, elite-led action rather than citizen demands, and that the newly appointed chiefs enjoy the most support among urban-based backers of the political party responsible for their appointments. This research contributes to historical work on the colonial-era creation of chieftaincies, but does so using a contemporary case of cultural engineering.

Chieftaincies in Malawi are nominally tied to discrete ethnic communities and derive at least some of their authority from claims of historical continuity. However, one ethnic community in Malawi – the Lhomwe – is distinct in that their chieftaincy is a very recent creation. Since 2008, there has been a concerted, elite-led effort to revive the cultural traditions of the Lhomwe ethnic community in Malawi (Kayira & Banda 2013; Kayira et al. 2019; Lora-Kayambazinthu 2019; Robinson 2023). One of the central achievements of this cultural revival has been the establishment and expansion of a Lhomwe chieftaincy. For a variety of historical reasons, discussed in depth herein, there were very few Lhomwe chiefs above the village level in Malawi prior to 2008. Since that time, many Lhomwe Traditional Authorities (TAs) and Senior Chiefs have been identified or appointed, and a new Lhomwe Paramount Chieftaincy was established in 2014. The goal of this paper is to describe the process through which the Lhomwe chieftaincy has been created within the context of the larger Lhomwe cultural revival, and to elucidate its reception by other chiefs and regular citizens.¹

To do so, I draw on a variety of data sources and types, most of which are original. First, I fielded two surveys targeted at members of the Lhomwe ethnic group. The first included face-to-face interviews with 1,254 Malawian citizens in three rural districts of Malawi in October and November 2016, 892 (71%) of whom reported their ethnicity as Lhomwe. The second survey captured responses from 1,087 Lhomwe citizens residing in three Malawian cities: Blantyre, Lilongwe and Zomba. Second, I collected thirty-eight in-depth oral histories from elderly members of the Lhomwe ethnic group, each from a different Lhomwe-majority rural locality. Third, I conducted seventeen interviews with members of the Mulhako wa Alhomwe leadership, politicians and activists within the Democratic Progressive Party (DPP) political party, and chiefs at all levels in the hierarchy of Malawian chiefs. Data were collected between 2016 and 2019 in collaboration with the Institute for Public Opinion Research (IPOR), a non-profit research organisation located in Zomba, Malawi.

These data reveal that the creation of the Lhomwe chieftaincy was largely a top-down process, led by political actors like the former president, his close advisors and other members of his party. The mechanisms through which Lhomwe chiefs emerged were twofold, including chiefs from other ethnic groups claiming a Lhomwe identity after the revival, and the rapid promotion of village-level Lhomwe chiefs. Qualitative data show that Lhomwe citizens were not demanding Lhomwe chiefs, and they characterise all chiefs, regardless of ethnicity, as legitimate and worthy of respect. Similarly, public opinion data from just before the start of the Lhomwe cultural revival suggest that Lhomwes in Malawi had no different perspectives on chiefs than members of other Malawian ethnic groups with existing chieftaincies. However, original survey data show that the newly created Lhomwe chieftaincy is valued by a subset of Lhomwes, namely those living in urban areas and those who support the DPP political party. These patterns suggest that this newly created chieftaincy's primary function is symbolic, representing the increased status of the Lhomwe ethnic group.

TRADITIONAL AUTHORITIES IN AFRICA

Traditional authorities in Africa – also referred to as traditional leaders, customary authorities or chiefs – govern sub-national communities based on claims to a customary, traditional or spiritual right to rule (Reid 2002; Logan 2009; Honig 2019; Neupert-Wentz & Muller-Crepon 2021). In the typology of authority developed by Weber (1958), the legitimacy of traditional authority derives from the fact that its rules and norms have been inherited from the past. As such, claims of continuity with the past – which Weber calls the ‘authority of the eternal yesterday’ – are crucial for traditional authorities to exercise legitimate authority.²

It is estimated that over 80% of the population of sub-Saharan Africa are governed by traditional institutions (Baldwin & Holzinger 2019). While there is certainly variation in the nature and scope of their authority across Africa (Logan 2013), traditional leaders and their affiliated institutions typically fulfil three functions. First, they are often the primary authorities relevant to day-to-day life at the local level, especially in contexts with weak formal government (Baldwin & Mvukiyehe 2015; Honig 2019; Tieleman & Uitermark 2019). In such cases, traditional leaders exercise their rights to collect taxes, allocate resources (especially land), enforce property rights, extract labour, co-ordinate community collective action and decision-making for the production of public goods, adjudicate local disputes, manage common pool resources, and maintain community peace and stability (Logan 2009, 2013; Acemoglu et al. 2014; Holzinger et al. 2016; Honig 2017, Wilfahrt & Letsa, 2023). Second, TAs act as intermediaries between local communities and other actors, including governments, political parties and civil society organisations, among others. Their local embeddedness, accessibility and legitimacy allow them to operate as both local gatekeepers and community advocates (Lund 2006; Koelble & Puma 2010; Baldwin 2013, 2014, 2016; Koter 2013; Logan 2013; Koter 2016). Third, TAs serve as symbolic or ceremonial leaders, and are viewed as being both custodians of cultural traditions and embodiments of group identity (Van Dijk & Van Rouveroy van Nieuwaal 1999; Williams 2010). In this capacity, they perform sacred rituals and ensure their preservation into the future (Logan 2009).

Traditional authorities are able to fulfil these functions largely because of popular support among their subjects and social norms that induce compliance with their rule (Baldwin 2013; Logan 2013; Chlouba 2022). While some have argued that Traditional authorities maintain influence primarily through their coercive control over valuable resources, their connection to powerful external actors, or the lack of any alternative authority (e.g. Mamdani 1996; Ntsebeza 2005; Oomen 2005; Koelble & Puma 2010), public opinion data nevertheless demonstrate that traditional leaders are widely trusted, are perceived to care about their subjects and are seen as less corrupt than elected leaders (Logan 2009, 2013). Still, the legitimacy of Traditional authorities is likely to be bolstered by the fact that it is recognised and reinforced by other institutions, especially the state (Guyer 1992; Lund 2006, de Kadt & Larreguy 2018). The degree to which these various functions are formally recognised and codified has varied significantly across states (Kyed & Buur 2007; Honig 2019) but has generally increased in the last thirty years (Englebert 2002; Ubink 2008; Logan 2009, 2013; Muriaas 2011).

Hereditary inheritance is the most common form of selection in traditional institutions, and individuals serving as traditional authorities tend to rule for life (Baldwin & Holzinger 2019; Honig 2019). But even among systems with hereditary succession rules, there is still variation across traditional institutions in terms of how pre-determined the selection process is, from strict rules based on gender and birth order to systems in which a leader can be selected from among many eligible individuals or families (e.g. Acemoglu et al. 2014; Baldwin & Mvukiyehe 2015; Nathan 2019). While such hereditary selection features have been characterised by some as anti-democratic (e.g., Ntsebeza 2005), others have shown that traditional institutions often have accountability mechanisms built in, including avenues for deposition and institutional checks provided by councils or assemblies (Logan 2013; Baldwin & Holzinger 2019; Honig 2022).

However, mechanisms for leadership selection – and the accountability they allow – tend to be undermined when traditional institutions are manipulated by external actors. Mamdani (1996) famously argued that colonial co-optation or replacement of traditional authorities, under the doctrine of indirect rule, gave rise to de-centralised despots who were no longer responsive to their constituents. Others have shown similar consequences of state intervention in traditional institutions in the post-colonial period (e.g. Santos 2006; Chiweza 2007). Perhaps the most extreme form of external manipulation is the wholesale creation of traditional authorities in societies that lacked such indigenous institutions, a surprisingly common occurrence during the colonial era (Fortes & Evans-Pritchard 1940; Hicks 1961; Afigbo 1972; Ranger 1983; Van Dijk & Van Rouveroy van Nieuwaal 1999; Honig 2022). Existing research shows that such ‘traditional’ institutions imposed from above resulted in elite capture, intractable succession disputes, worse leadership performance, weaker responsiveness and the loss of communal resources (Nathan 2019; Chlouba 2022).

This paper contributes to the study of traditional authorities in Africa by documenting a contemporary case of elite-led invention of a chieftaincy in Malawi, and the processes through which it was achieved. Through a combination of original interviews and oral histories, I identify two mechanisms for chiefly emergence that operate outside the normal practice of hereditary selection: ethnic switching by existing traditional authorities and the expeditious promotion of lower-ranking chiefs. Each of these mechanisms attempts to solve the fundamental challenge of inventing a traditional institution – that such an invention undermines the legitimacy that is normally conferred because of traditional continuity. The former mechanism seeks to overcome a lack of continuity by adjusting the cultural connection (ethnicity) of otherwise qualified traditional authorities, while the latter identifies individuals within the ethnic group (local leaders) who might have filled higher-ranked chiefly positions in the counterfactual world where a hierarchical chieftaincy was present.

CHIEFS IN MALAWI

Malawi is home to numerous ethnolinguistic groups, with the largest groups being the Chewa, Lhomwe, Yao, Ngoni, Tumbuka, Mang'anja/Nyanja, Sena, Tonga, Nkhonde and Lambya. Most of these groups had hierarchical structures of traditional authority in the pre-colonial era, although the strength and extent of these institutions fluctuated over time (Pike 1965). During the colonial era, traditional authorities were formally incorporated into the British colonial state through the practice of indirect rule (Chiweza 2007; Kayira & Banda 2018). Under this arrangement, native authorities were given formal training, an annual salary and authority to collect taxes and adjudicate civil cases. Traditional authorities were also formally recognised by the newly independent Malawian state through the 1967 Chiefs Act, but their degree of power fluctuated significantly over time in response to the political exigencies of the day (Chiweza 2007). With democratisation and de-centralisation in the 1990s, chiefs' de jure powers were curtailed while their de facto influence has persisted or even expanded (Chiweza 2007; Logan 2013). Most research suggests that traditional authorities in Malawi are respected, trusted, accessible and responsive to their communities (Logan 2013; Chiweza 2016; Muriaas et al. 2019).

In terms of structure, traditional authorities in Malawi are organised both territorially and hierarchically. Each of the 22,861 villages enumerated in the 2018 census has a local chief, referred to as a village head or village headman (VH). Villages in close geographic proximity are typically clustered into a group, with one of the VHs serving as the group village head (GVH) for the cluster.³ GVHs fall under the jurisdiction of Traditional Authorities (TA), which rule over a defined geographic unit, which is also referred to as a TA. As of the last census in 2018, there were 256 TAs in Malawi nested within the twenty-eight rural districts, with each TA overseeing around ninety-three GVHs (sd = 73) within their area.⁴ TAs can hold one of three ranks – sub-Chief (16% of all TAs in 2018), Chief (79%) or Senior Chief (4%) – based on seniority and performance. Finally, there are seven chiefs which hold the highest rank of Paramount, which is a position associated with a particular ethnic group rather than a formally demarcated territory (Chiweza 2007; Cammack et al. 2009).⁵ Traditional leaders at all ranks receive a modest remuneration from the state and chiefs at the TA rank are formally appointed (and can be removed) by the Ministry of Local Government and Rural Development or the president (Chiweza 2007; Cammack et al. 2009). However, in practice the ruling families and traditional power structures determine who serves in these roles under most circumstances (Chiweza 2007).

LHOMWE CULTURAL REVIVAL

The Lhomwe are concentrated in the southern region of Malawi, primarily in the so-called ‘Lhomwe belt’ districts of Mulanje, Thyolo and Phalombe. Members of the group originally migrated into present-day Malawi from the central part of what is now Mozambique in three waves between the mid-sixteenth and mid-twentieth centuries (Rashid 1978; Boeder 1984; White 1984). Migrants in all three waves settled in the southern region of present-day Malawi among the groups already residing in the area, mainly the Yao and the Mang'anja (White 1984). Their social status upon arrival was lower than the original inhabitants, as is clear from the derogatory terms sometimes used for the group: akapolo (slaves) because they were often required to work in the gardens of the existing chiefs (Bandawe 1971; Boeder 1984) or nguru, a derogatory term derived either from a specific dialect of Lhomwe (Boeder 1984) or from a Yao word meaning ‘speaker of an unintelligible language’ (Bandawe 1971). In addition to fleeing political instability and war in Portuguese East Africa, Lhomwes were also recruited by the Nyasaland colonial government to work on European estates in southern Malawi to grow tea, coffee, cotton and tobacco (Boeder 1984; White 1984). The colonial government, too, held relatively negative views on the Lhomwe group – whom they called Anguru – which were further exacerbated by the low-educational prospects afforded on commercial estates (Bandawe 1971; Boeder 1984). These multiple forces resulted in a set of stereotypes about Lhomwe, which have persisted to this day, namely that they are uneducated, heavily involved in witchcraft and snake-eaters (Kamwendo 2002).

Lhomwe cultural traditions – including dances, initiation rites and language – have slowly eroded over time (White 1984; Kayira & Banda 2013). In addition to the shame-induced suppression of Lhomwe cultural traits due to the negative stereotypes discussed above (Matiki 1996; Kamwendo 2002; Kayira & Banda 2013), the loss of cultural traditions was also a product of assimilation into the ethnic communities among whom the initial Lhomwe migrants settled (White 1984) and the general suppression of languages other than Chichewa during the first thirty years of Malawian independence (Forster 1994; Matiki 1996; Chirwa 1998).

However, since 2008, there has been a concerted effort to revive the Lhomwe culture, starting with the creation of an active cultural association called Mulhako wa Alhomwe (Ntata 2012; Kayira & Banda 2013; Lora-Kayambazinthu 2019, Robinson 2023). ‘Mulhako’ means ‘door’ in Chilomwe, and thus the name of the association refers to itself as the gateway to Lhomwe culture. The Mulhako wa Alhomwe association is headquartered on a large plot of land in Chonde, Mulanje district, in the heart of the Lhomwe belt. The motto of the group, prominently displayed on a billboard outside their headquarters, is ‘Alhomwe, Alhomwe, noophiya!’, which in Chilomwe means ‘Lhomwes, Lhomwes, we are here!’ The group hosts annual gatherings at their headquarters every October, in order to showcase and celebrate Lhomwe cultural heritage. In addition, Mulhako wa Alhomwe has established a small cultural museum on the premises and has plans for both Chilomwe classes and a Chilomwe radio station (Zeze 2015).

While Mulhako wa Alhomwe adamantly claims that it is non-political, many see a political aim in its creation (Kayira & Banda 2013). This is due, at least in part, to the fact that the founder and first patron of Mulhako wa Alhomwe was then-president Bingu wa Mutharika. Mutharika was handpicked to succeed the first president elected under multiparty democracy, Baikili Muluzi of the United Democratic Front (UDF), after Muluzi lost his bid for a third term (Dulani 2011). However, after winning the presidency on the UDF ticket in 2004, Mutharika left the party in 2005 and created his own party, the DPP. In 2009, Mutharika easily won re-election with 66% of the national vote – the largest presidential vote share in Malawi's democratic history (Ferree & Horowitz 2010). In his second term, however, Mutharika was increasingly seen as favouring the Lhomwe at the expense of other groups, including both the disproportionate appointment of Lhomwes to his cabinet (Nyasa Times 2009) and the channelling of development to the Lhomwe belt (Kayira & Banda 2013).

Bingu wa Mutharika died in office in 2012, and Joyce Banda, his estranged vice president, replaced him, despite efforts to install Mutharika's younger brother instead (Dionne & Dulani 2013). In 2014 and 2019, that younger brother, Peter Mutharika, was elected president, although the 2019 election was ultimately annulled and Mutharika lost the ‘fresh’ election in 2020. All three of these elections showed a return to strong regional voting patterns, reminiscent of the first three elections in Malawi's multiparty era (Dulani & Dionne 2014). While Peter Mutharika was perceived to be less invested in Mulhako wa Alhomwe and the Lhomwe cultural revival than his elder brother, even he was regularly accused of favouring the Lhomwe, a particular manifestation of tribalism referred to colloquially as ‘Mulhakolism’.

Elsewhere, I explore the political logic of this investment in ethnocultural revival by the Mutharikas, the DPP political party and other prominent Lhomwe politicians (Robinson 2023). There, I argue that political elites have invested in this revival because doing so bears political returns via the increased ethnic visibility of group members. Ethnopolitical leaders benefit from having the identity of their group members easily visible to others, because such visibility ties those individuals' fate to that of the larger group, engendering a dependable ethnic base of support. In this article, however, I focus exclusively on a single aspect of the Lhomwe cultural revival: the creation of a Lhomwe chieftaincy.

INVENTING A LHOMWE CHIEFTAINCY

As part of the larger Lhomwe cultural revival, Lhomwe political elites – including the president, his advisors and members of the DPP political party – embarked on the creation of a robust Lhomwe chieftaincy. This section draws primarily on original interviews to better understand why there were few Lhomwe chiefs prior to 2008, the process through which the number of Lhomwe chiefs grew, and the methods of creating a set of ‘traditions’ around this new chieftaincy.

Why so few Lhomwe chiefs?

There was general agreement across respondents that, prior to 2008, there were very, very few Lhomwe chiefs at the GVH level and virtually none at the TA level and beyond. For example, one chief told me, ‘We Lhomwes did not have a paramount chief, there were no senior chiefs, no TAs, not even sub TAs who were Lhomwe’ (Group village head 20/10/16). Another said, ‘all the time back, Lhomwe chiefs only used to hold the position of village heads and group village heads, but the TA who was overseeing was either a Mang'anja or a Yao. You will see that the village is largely consisting of the Lhomwes, but Lhomwe chiefs were failing to become the TA’ (Senior Chief 20/10/16).

Across the interviews, I heard two separate narratives to explain why there were so few Lhomwe chiefs. Lower-ranking chiefs and elderly Lhomwe citizens typically attributed the dearth of Lhomwe chiefs to the nature and timing of Lhomwe migration into present-day Malawi. For example, a GVH pointed to fact that most Lhomwes emigrated from present-day Mozambique into Malawi under duress, and did not move together as an organised group. He said, ‘when we [the Lhomwes] were coming from Mozambique, we were just coming as a family while the people from other tribes were coming as a group’ (Group village head 20/10/16). He went on to explain that ‘back then there was war in the Portuguese area (Mozambique)…so when they were coming from Mozambique, they were running from the war as just a family…so when they came here [to Malawi] they were just few. But [other groups] came in large numbers and when they find those few [Lhomwe] people, they would just select themselves to be chiefs because they were large in numbers’ (Group village head 20/10/16).

Relatedly, members of ethnic communities who preceded the Lhomwe in settling in Malawi were seen as having a more legitimate claim to rule. One respondent said, ‘there is no Lhomwe chief here, the chiefs are the Mang'anjas, the original inhabitants’ (Oral history 30/10/16, emphasis added). Others said, ‘Mang'anjas are the owners of the land’ (Oral history 25/10/16), ‘Mang'anjas were considered landowners, [because] they were one of the first groups to settle into Nyasaland and the rest of the groups found them [already] settled’ (Oral history 27/10/16), and ‘the other groups came earlier than us, that is the simple reason’ (Oral history 1/11/16). This temporal advantage has persisted because chiefly successions in Malawi are hereditary. As one respondent explained, ‘there are a few [Lhomwe chiefs] but the Mang'anjs have a lot. Chiefs in the past used to predict about their death so they would choose their successor from their clan so that the chosen successor should be elevated to the position of the chief. This process is quite repetitive, thus why we have a lot of Mang'anja TAs’ (Oral history 26/10/19). This temporal advantage is compounded by the material advantage that other chieftaincies have in their control of land – a chief's ability to allocate and control land is an important source of their authority.

The higher-ranking chiefs we interviewed, however, instead attributed the lack of Lhomwe chiefs to colonial discrimination. One respondent said it was ‘because the azungu (white men) government was not recognizing the importance of [Lhomwe] chiefs’ (Traditional Authority 7/7/17). Another said ‘the white people could not allow a Lhomwe person in their yard, let alone give them the position to be the caretaker of the yard’ (Senior Chief 20/10/16). Consistent with the secondary literature outlined above, one chief linked colonial discrimination to one of the earliest and most successful uprisings by colonial subjects in Nyasaland:

Back then, we had few Lhomwe chiefs because of John Chilembwe. When Chilembwe came he was against the barbaric acts of the colonial masters, so he started mobilizing the people and sensitizing them to the cruelty of the colonial masters, despite the fact that this was not their country. He mostly used the Lhomwes from Chiradzulu district, most Lhomwes started following him. That's when the colonialists started burning down his church, that's when they realised that he meant war. [Those who] went ahead and beheaded Livingstone, those were Lhomwes.⁶ After they beheaded him they brought his head and his wife, they wanted to taste (rape) his wife. So after they raped the white woman that time, the white men came and picked her and she explained to them what had happened …That time they did not know that the Lhomwes were the ones responsible for it, they just knew it was black people who did it. So they started enquiring as to who was responsible for this, and when they asked, the Yaos they declined and said that is was the Lhomwes who did it. The colonial masters started suspecting John Chilembwe and from that time they started hating us. They had huge farms, but they did not want to employ Lhomwes but the Yaos. So, because the farms were large and when they started choosing chiefs, they would pick the foreman who was a Yao and they were becoming TAs together with the Mang'anja people. That is why there are few Lhomwes chiefs, because we were not liked by those people. Because the white people hated us, that is why Lhomwes were not given any position as the senior chiefs. (Senior Chief 20/10/16)

Emergence of Lhomwe chiefs

There was also general agreement from everyone I spoke to that the number of Lhomwe chiefs had increased markedly in recent years. For example, one GVH said ‘back then there were fewer Lhomwe chiefs right? But right now, Lhomwe chiefs have increased in number’ (Group village head 20/10/16). A Senior Chief similarly stated, ‘yes Lhomwe chiefs are now being encouraged’ (Senior Chief 20/10/16).

In theory, chieftaincies in Malawi are an ‘institutionalised form of traditional rule with the cardinal characteristics of prescribed kinship and lineage succession to office’ (Chiweza 2007: 53), with legitimacy rooted in continuity with the pre-colonial period. But, given the low supply of Lhomwe chiefs for the reasons outlined above, the rapid creation of a Lhomwe chieftaincy required shortcuts. Interviews revealed two mechanisms, both linked to the larger cultural revival, through which Lhomwe chiefs emerged: non-Lhomwe chiefs claiming Lhomwe heritage and the rapid promotion of chiefs outside of normal protocol.

The first mechanism focused on ‘hidden’ Lhomwe chiefs. For example, when asked why there were so few Lhomwe chiefs in the past, one TA explained that ‘it was because people were not open…they were shy’ (Traditional authority 7/7/17), while another said more directly ‘it's because we have come out from where we have been hiding’ (Group village head 20/10/16). Another traditional leader echoed the same explanation, saying, ‘most of these [Lhomwe] chiefs were there, but they chose to be silent and hide themselves…It is because they were so ashamed, and now we are free to realise that our tribe is well known, and everyone has realised the goodness and success of the tribe. Lhomwe are many and some still are hiding their identity’ (Traditional authority 20/10/16).

This emergence from hiding by ‘Lhomwe’ chiefs was attributed directly to Mulhako wa Alhomwe, its patron Bingu wa Mutharika and the cultural revival they spurred. A TA explained, ‘most of them were hiding but now they have come out. Muhlkho wants to differentiate the Lhomwe and other tribes from their cultures. Like here most of the chiefs were thought of as Mang'anja, while they were really Lhomwe’ (Traditional authority 20/10/16). A Lhomwe VH confirmed, ‘there were few Lhomwe chiefs back then [because] us Lhomwes were looked down upon, but now we are free because of the late Bingu, that's why we are in the open now. If we can count the number of Lhomwe chiefs, now we are many’ (Village head 20/10/16).

One of the TAs we interviewed claimed to be a ‘hidden Lhomwe’. While my understanding prior to the interview was that the chieftaincy was held by a Yao clan, upon arrival at the chief's home, we found her dressed from head to toe in Mulhako wa Alhomwe cloth, only worn by Lhomwes. When I asked about this, the chief replied, ‘yes, this is a Yao chieftaincy from my mother's side. But I am well known by my father's tribe, the Lhomwe’ (Traditional authority 20/10/16). Later, the chief again referred to her own situation in the discussion of ethnic switching, saying:

Even me, some chiefs were saying that I am Yao and not Lhomwe. They talk about this in my face. I remember the former president, Joyce Banda, once shouted at me saying asking why I stay among the Lhomwe when I am a Yao. She asked me why I interact with Nguru [Lhomwe] people when am not one of them. I just asked her ‘Have you forgotten who [TA's father's name] is?’ But she did not reply. My father, [TA's father's name] is Lhomwe, so I took after him as a Lhomwe. (Traditional authority 20/10/16)

This was a particularly interesting claim, given that both the Lhomwe and the Yao practice matrilineal kinship, meaning that a mother's ethnicity rather than a father's is passed on to children in mixed marriages (Dulani et al. 2021; Robinson & Gottlieb 2021).

When I mentioned this nominally Yao chief's claim to Lhomwe identity, another of my respondents was surprised. But, they followed up by saying, ‘you see, you can't just automatically dispute that, it may well be that there is still a little Lhomwe heritage back somewhere’ (Mulhako leader 10/29/16). This expansion of the Lhomwe chieftaincy through the unveiling of ‘hidden Lhomwes’ is consistent with a larger strategy of expanding the definition of who belongs to the group. The Lhomwe ethnic association, Mulhako wa Alhomwe, has formally stated that they consider anyone with any Lhomwe heritage to be Lhomwe, not just those with maternal Lhomwe heritage (Robinson 2023).

The second mechanism through which the Lhomwe chieftaincy has emerged is via the rapid appointment and promotion of Lhomwe chiefs, as well as the creation of new TA jurisdictions. In Malawi, the official appointment of chiefs at the rank of TA or above formally falls under the Directorate of Chiefs Administration within the Ministry of Local Government and Rural Development, although the president can also directly influence appointments (Cammack et al. 2009). Typically, government involvement is simply a formality, with actual selection following local customs and traditions (Chiweza 2007). However, the government has legal authority to appoint or remove chiefs, promote or demote their rank and to divide existing TAs in order to create new ones, a right that was exercised under DPP rule in order to create the Lhomwe chieftaincy. A GVH explained that ‘when Mapwiya Dr Bingu wa Mutharika came into power, he started elevating the Lhomwe chiefs to become sub-TA and now we have a paramount Lhomwe chief’ (Group village head 20/10/16). An elderly Lhomwe respondent similarly noted that ‘the trend these days is that Lhomwe chiefs are being anointed of late’ (Oral history 10/26/19). Someone who was involved in decision-making around chiefs at the time noted, ‘in fact, the Lhomwe chiefs that have risen have gotten those positions largely because of the manipulations of people like myself and Bingu’ (Mulhako leader 10/29/16).

One striking example of the expansion of Lhomwe chiefs is the creation of several new chieftaincies, given to Lhomwes, in Machinga district, which is outside the so-called ‘Lhomwe belt’. Starting in 2004, the Malawian government initiated a controversial World Bank-funded re-settlement programme that aimed to move up to 15,000 citizens from densely populated districts to less-dense districts, ostensibly to ease pressures associated with land scarcity (Sharp et al. 2019).⁷ However, given that most beneficiaries were members of the Lhomwe ethnic group from districts in the Lhomwe belt (e.g. Thyolo and Mulanje districts), and were given land in Yao-majority areas (Machinga and Mangochi districts), many viewed the programme as a form of ethnically targeted political patronage (Chinsinga 2011). Perhaps in response to this influx of Lhomwe citizens, a number of Lhomwe chiefs have been appointed in Machinga recently. One respondent brought up this example in discussing the expansion of the Lhomwe chieftaincy, saying ‘I'll give the example of Machinga. There was a time when the land there was, you know about it [referencing the resettlement scheme]. They were transferring to Machinga, right and chiefs there, some Lhomwe chiefs there, they are now being made as TAs because they are from that side, but they are now in a Yao area. There was a big fight there, so that's the influence now of the Lhomwes trying to penetrate in the Yao areas’ (Politician 4/5/18).

Creating a Lhomwe Paramount chief

In addition to the appointment of chiefs at the village, group village, TA and senior chief levels, the Lhomwe cultural revival has also managed to have a Lhomwe paramount chieftaincy created. When asked why it was so important to create a paramount chief for the Lhomwe, one respondent who had been part of the creation of the chieftaincy told me:

That was part of the movement, to make the Lhomwes feel that they had a family there. And in order to for them to feel that they are a family, one family, there had to be a head of that family. And also, looking at other tribes, there were paramount chiefs. So, if there is a paramount chief for the Mang'anjas, such as Lundu, and the paramount chief of the Tumbukas and whatever. Why then, not a paramount chief for the Lhomwe, especially at a time when the country is led by a Lhomwe president? So it was very urgent that there be a [Lhomwe] paramount chief. It was very much an important part of the movement to unite the Lhomwes, to make them feel that there was somebody who would be a figurehead for this movement of the self-assertion of the Lhomwes. (Mulhako leader 10/29/16)

Bingu wa Mutharika installed the first Lhomwe paramount, Chief Mkhumba, on 25 October 2008 (Kayira & Banda 2013). At the 2014 Lhomwe festival, Peter Mutharika installed a new Lhomwe TA, TA Ngolingoliwa, who was very active in the leadership of Mulhako wa Alhomwe. Ngolingoliwa was ultimately appointed as Paramount Chief of the Lhomwes in 2017, after the death of Mkhumba in 2014. These two Paramount chiefs exemplify well the two mechanisms of chiefly expansion outlined above – ethnic switching and rapid advancement – both of which fall outside the tradition of hereditary succession. These mechanisms were necessary because, as outlined above, there were very few Lhomwe chiefs in the pipeline.

Mkhumba, the first Lhomwe paramount chief, is widely known to have not been a member of the Lhomwe ethnic group, though he ruled in a Lhomwe-majority area of Phalombe. One TA recalled, ‘we just heard about his elevation. I remember one time he told me that he is a Mang'anja himself. I asked him why he was found among the Lhomwe chiefs and he just said “mmm these things” …He meant that he was favored by politicians but actually he told me himself that he was Mang'anja’ (Traditional authority 20/10/19). One respondent who was involved in the decision to appoint Mkhumba explained, ‘the business of making [Mkumba] the paramount chief of the Lhomwes was indeed bad. He was not Lhomwe. But you see, there was a shortage of Lhomwe chiefs’ (Mulhako leader 29/10/16).

Most other chiefs accepted that a non-Lhomwe would act as the Lhomwe paramount chief as a sort of placeholder until there were Lhomwe chiefs to choose from. One TA admitted, ‘we discovered that he was not Lhomwe, we chose him just to cover us as a leader’ (Traditional authority 7/7/17), while another explained that ‘the late Mkhumba was a Chewa (Mang'anja), but he was just asked to be the acting chief according to our culture, and he agreed, saying that when everything is settled amongst the Lhomwe tribe, he will hand over the seat to the rightful owner’ (Senior chief 20/10/16). A lower-ranking chief explained the acceptance this way: ‘Mkhumba was not a Lhomwe, but we welcomed that development because Professor Bingu wa Mutharika knew what he was doing. He knew that for this thing to be established, he needed an acting chief so that when he finds a Lhomwe chief he can finally take over’ (Group village head 20/10/16).

In addition to this placeholder justification, the history of the Lhomwe and the ethnic mixing in the southern region meant that there is always the possibility of at least some Lhomwe heritage. One respondent referenced this explicitly, saying ‘oh, I think some people knew that Mkhumba was not Lhomwe, but decided, well, we'll just respect him. He was already a prominent chief there [in the Lhomwe belt]. The idea of having him as paramount, you know, from the Phalombe area was more important in the thinking of ordinary people than whether he was genuinely Lhomwe. And of course, you must always again go back to the idea of all this mixing. He probably has some Lhomwe, at least a little’ (Mulhako leader 29/10/16).

The second Lhomwe Paramount Chief's heritage was never in question: ‘while the late ParamountChief Mkhumba wasn't a pure Lhomwe, he was an Mang'anja, Ngolingoliwa is the first pure Lhomwe paramount chief’ (DPP operative 1/5/18). As a result, he was viewed as a more legitimate leader of the Lhomwe from a cultural perspective, as explained by a TA involved in his appointment. We were told, ‘I was one of those who helped to choose him. They asked us about who can be our great leader, and we chose him because others do not really know the real Lhomwe culture and its roots. But Chief Ngolingoliwa follows everything about our culture and he speaks [the Lhomwe language] fluently, so we chose him’ (Traditional authority 20/10/16).

Ngolingoliwa's appointment as paramount chief also happened outside of normal procedure, via the second mechanism: an unusually rapid advancement through the rankings. Ngolingoliwa served as a VH and then GVH from the early 1980s until 2014. In 2014, he was appointed as a TA and only one year later, in 2015, as a senior chief. Such rapid advancement from a GVH to a senior chief was basically unheard of, and then Ngolingoliwa was further promoted to the very highest rank of paramount chief in 2017. The chief himself attributed his advancement to his cultural competence, explaining, ‘I have come a long way, so because of the skills I have, my zeal and living well with people, that's when I was elevated to become the TA. So because I became the TA, this is also associated with the Lhomwe culture, so in that sense I am also like the senior person there, so I am responsible for [Lhomwe] rituals like sacrifices… That's also why I was elevated to become the senior chief’ (Senior chief 20/10/16).

Not everyone saw this rapid advancement as legitimate. Some pointed to the political nature of Ngolingoliwa's rise over (the few) existing more senior Lhomwe TAs. One chief told us:

First, we have Nazombe and Nkulambe [two Lhomwe chiefs in Phalombe]. Now we are being told that the true Lhomwe is Ngolingoliwa? Right. We are just wondering. We expect that position to be held by one from here in Phalombe, because this is where the real Lhomwe came from. More Lhomwes are here, Mang'anjas are very few. We are just accepting this because the current president says he is Lhomwe, so he is favoring his side, we have to be honest here. (Traditional authority 7/7/18)

By ‘favoring his side’, the chief is referring to the fact that Ngolingoliwa hails from Thyolo district, the same as the Mutharika presidents. A political operative made a similar point, noting, ‘let's take an example like Phalombe, you see we had a paramount chief from there, Mkumba. Now when he died and they took that chieftainship to Thyolo to Ngolingoliwa, people in Phalombe were not happy because they thought that's their own chieftaincy’ (Political operative 4/5/18). A former leader of Mulhako wa Alhomwe concurred, arguing that ‘the natural person to take over the position of paramount chief after Mkhumba should have been TA Nkhulambe [of Phalombe] in my view’ (Mulhako leader 29/10/16).⁸

Inventing chiefly traditions

The leaders of the Lhomwe cultural revival had a problem when it came to the newly created and expanded Lhomwe chieftaincy: there were no real existing traditions associated specifically with Lhomwe chiefs. In establishing the appropriate traditional practices to revive, leaders of the revival sometimes drew from practices within present-day Mozambique, where Lhomwe culture is perceived to have survived more intact. However, connections between the Lhomwe across borders are quite weak. They also drew on older people within Lhomwe communities in Malawi. For example, a Mulhako leader explained, ‘We have people in the village, even up to now, our grandfathers and grandmothers. Those people, they know [the traditions]. So we go to them, we've been asking them like, “how do you conduct a wedding?” or other ceremonies. After finding out from maybe one or two people, we take it back to the chiefs and tell them how others are doing it so they should also learn and then maybe teach even our kids’ (Mulhako leader 15/8/16).

However, in many cases, traditions are simply being invented, sometimes by borrowing from other ethnic groups. For example, a GVH told us, ‘we are still learning of the ways of how to respect our chiefs, so we are borrowing ideas from people of the other tribes because we were lost and now we are back and we are still trying…Our friends like the Chewa and Ngoni, their respect toward their chief is one hundred per cent and everyone knows that the person is a senior chief, but us here we still lagging behind on that’ (Group village head 20/10/16). A VH similarly described the input from other groups during the initiation of the Lhomwe paramount: ‘When the paramount was anointed, two paramount [chiefs] from other groups came and they taught us on how to respect him because he is now on a different level’ (Village head 20/10/16).

One of the best examples of the ‘invention of tradition’ (Hobsbawm & Ranger 1983) around the Lhomwe chieftaincy was a training for more than forty traditional leaders (VHs and GVHs) at the Senior Chief Ngolingoliwa's headquarters in early August 2016 (Chief's councillor 18/8/2016). The format was similar to trainings on public health hosted by the government, civil society organisations and non-governmental organisations, with a classroom-style format and attendee allowances. The purpose of the training was to ‘sensitise’ Lhomwe traditional leaders on the proper ways to show respect to higher-ranking chiefs. The chief's councillor explained, in a very matter-of-fact way, that the Lhomwe did not have such traditions, so the chief's council had chosen behaviours they appreciated from other groups in Malawi with clearer traditions, including the Ngoni, the Chewa and the Nkhonde (Chief's councillor 18/8/2016).

Several different types of behaviours were explained during the training. For example, ‘we were told not to touch [the paramount], if he wishes he is the one who should touch us’ (Village head 20/10/16) and ‘you cannot just go and meet the Mwene wa Mwamwene (chief of chiefs), you firstly meet a boy and then he goes and tells him that there are such and such people who want to meet you, so when he accepts to meet you, you will go and meet him’ (Group village head 20/10/16). Other proscribed behaviours included lower-ranking chiefs not going near the senior chief nor speaking directly to him, and never showing higher-ranking chiefs their backs.

It is fascinating that these ‘traditions’ were being actively invented by elites, and then disseminated to lower-ranking chiefs through the modern mode of a workshop or training. As one respondent put it, ‘we are getting there because back then there were no such things’ (Group village head 20/10/16). Despite these very modern origins, perhaps in fifty or a hundred years, such behaviours will be regarded as primordial, a reflection of deep and perennial Lhomwe traditions.

CITIZEN SUPPORT FOR THE LHOMWE CHIEFTAINCY

The sections above report on the creation of a Lhomwe chieftaincy, and related cultural traditions, by elite actors, including leaders of the Mulhako wa Alhomwe ethnic association, DPP politicians and party leaders and TAs. But how have such actions been viewed by Lhomwe citizens? How important is it to members of the Lhomwe ethnic group to have Lhomwe chiefs? Do they see the newly created Lhomwe chieftaincy as legitimate? What explains variation in such support? This section uses data from the oral histories, Afrobarometer data from Malawi, and two original surveys with Lhomwe citizens to interrogate these questions.

As described in the sections above, the justification for expanding – and, in some cases, creating – the Lhomwe chieftaincy was primarily that Lhomwe citizens should not be under the leadership of non-Lhomwe chiefs. For example, a Mulhako wa Alhomwe leader said ‘a lot of the Lhomwes find themselves living under the chiefs of other tribes, so we had to quickly promote all the chiefs to these positions’ (Mulhako leader 29/10/16). A TA similarly noted that ‘it's because the government has discovered that chiefs are important, like here there are more Lhomwes [citizens] but few [Lhomwe] chiefs, so they decided to have more chiefs to lead their people’ (Traditional authority 7/7/17). A political operative gave the same justification, saying ‘even the Mang'anja chiefs or tribe were the people who were ruling the Lhomwes in the area, because they were not Lhomwes and were ruling Lhomwes they would not want the Lhomwes to rise up to know that they are the owners of the land. But, since the inception of Mulhako, the Lhomwes have managed to say this belongs to us, we need to have a Lhomwe king and chiefs’ (DPP operative 8/5/18).

Demand for Lhomwe chiefs

During the collection of the oral histories, we interviewed elderly members of the Lhomwe ethnic group who had spent most of their lives under non-Lhomwe chiefs. These individuals were asked about whether they would prefer to live under a Lhomwe chief, and whether Lhomwe chiefs garnered greater support than non-Lhomwe chiefs in Lhomwe majority areas.

While two respondents out of thirty-six said they preferred a co-ethnic chief because of shared language – ‘Chilhomwe is both our language’ (Oral history 1/10/16) – and fear of discrimination by non-co-ethnic chiefs – ‘a chief of another tribe might discriminate you while a chief of your own tribe can not discriminate his fellow tribemen’ (Oral history 25/10/16) – the vast majority of respondents did not express a desire for Lhomwe chiefs. While somewhat repetitive, the following comments emphasise the widely held view that the ethnicity of the chief is irrelevant to many Lhomwe citizens:

All chiefs receive equal respect as chiefs, whether Chewa or Manganja. (Oral history 2/11/16)

Any chief can lead…the respect is the same. A chief is a chief and all deserve a kind of respect due a chief. (Oral history 4/11/16)

Chiefs are respected and this applies across all the tribes. (Oral history 7/11/16)

A chief is a chief – Lhomwe, Chewa, Mang'anja – a chief is a chief. (Oral history 8/11/16)

Anyone [can be a good chief], regardless of tribe, depending on one's character. (Oral history 10/21/16)

It doesn't matter the tribe of the chief, whoever is chief, he is our leader. (Oral history 1/11/16)

I personally think it doesn't matter having say a Yao or a Lhomwe chief, there is no reason for people not to agree. (Oral history 16/11/16)

The chiefs receive respect, since it comes from the people, it doesn't matter if the chief is a Lhomwe or a Yao, they are all given respect by the people. (Oral history 16/11/16)

Be it a Lhomwe, Yao or Mang'anja chief, they are all chiefs. (Oral history 27/10/16)

These excerpts suggest that the justification for the expansion of the Lhomwe chieftaincy is not being driven by strong demands from Lhomwe citizens, at least not those in rural parts of the country in 2016. This suggests that the demand for a Lhomwe chieftaincy was not widespread, and perhaps driven from above by elite interests. For example, the promotion of Lhomwe chiefs may have allowed powerful actors to access additional land, or to rule in such actors' favour in matters under adjudication. It is also possible that political actors anticipated that such chiefs would be instrumental in voter mobilisation (Kayira et al. 2019). While the present research cannot address the question of elite interests, the lack of broad citizen demand suggests that this would be an important avenue for future research.

Attitudes towards traditional leaders

In order to get a broader understanding of attitudes about chiefs among Lhomwes, both over time and relative to other groups, I turn to the Afrobarometer data from Malawi. Afrobarometer has collected systematic public opinion data across many African countries every few years since 1999. Here, I utilise five rounds of data collection in Malawi (R3 2005, R4 2008, R5 2012, R6 2014 and R7 2018) (Afrobarometer 2018).

I first use round 4 data on attitudes towards traditional leaders. Round 4 is useful because it was collected in 2008, just prior to the start of the Lhomwe cultural revival and therefore before Lhomwe chiefs had been appointed in large numbers. It also includes a wider array of questions on traditional leaders than prior or subsequent rounds. Figure 1 summarises attitudes towards traditional leaders by ethnic group, including trust, frequency of contact, expectations of being heard, impressions of influence and preferences for increased influence. While Lhomwe respondents were the most likely to say that they do not trust their traditional leaders at all, they are very closely followed by the Chewa and Ngoni, both of which have very strong ethnic chieftaincies, and most Lhomwes reported trusting their traditional leaders a lot. Lhomwes are also not markedly different in terms of contacting their traditional leaders or expecting that those leaders will listen to them. Finally, Lhomwes do not differ from other groups in their assessment of the influence of traditional leaders, or in their preferences over increasing or decreasing chiefly powers. The patterns in Figure 1 thus suggest that, despite being much less likely to have co-ethnic chiefs, Lhomwe respondents did not typically have worse views of their traditional leaders.

Figure 1. Attitudes towards traditional leaders by ethnic group in 2008, prior to the Lhomwe cultural revival.

Note: Data come from Afrobarometer round 4, 2008. n = 1,128–1,158.

It is possible that the Lhomwe were no different than other groups prior to the cultural revival, but that the revival has reshaped their opinions on the chieftaincy after the dearth of Lhomwe chiefs became salient. To assess this possibility, I look at over time trends in trust and contact with traditional leaders. I focus on trust in chiefs and frequency of contacting chiefs because these questions were asked in almost all rounds of Afrobarometer data collection, and in a largely consistent way. Figure 2 shows that there has been a general decline in trust of traditional leaders over time. While the Lhomwe are consistently lower than other groups, the difference is small. More importantly, there does not seem to be any change in the trend for the Lhomwes with after the expansion of the Lhomwe chieftaincy between 2008 and 2018. Figure 3 shows that rates of contacting traditional leaders is increasing overall, but the Lhomwe are again quite similar to other groups.

Figure 2. Trust in TAs over time by ethnic group.

Note: Proportion who reported trusting their TA ‘a little bit’, ‘somewhat’ or ‘a lot’.

Figure 3. Frequency of contacting TAs over time by ethnic group.

Note: Proportion who reported ever contacting their TA in the past 12 months.

Variation in support for chiefs among Lhomwe citizens

The Afrobarometer data reported above suggest that the Lhomwe are not systematically different in their assessments of chiefs than other groups in Malawi, despite the overt and extensive expansion of a Lhomwe chieftaincy since 2008. In this section, I use original survey data to explore whether there is greater support for this new Lhomwe chieftaincy among certain subsets of Lhomwe citizens than others.

The survey data were collected in two waves. The first wave (rural sample), collected in October and November 2016, included face-to-face interviews with 1,254 Malawian citizens, 892 of whom reported their ethnicity as Lhomwe (71%). Respondents were sampled from thirty-six randomly selected Lhomwe-majority localities across three Malawian districts: two in the Lhomwe heartland, Mulanje and Thyolo districts (75 and 76% Lhomwe, respectively), and one with a sizable Lhomwe population outside the homeland, Machinga district (28% Lhomwe). The second wave (urban sample) was collected in April and May 2019, and included face-to-face interviews with 1,087 Lhomwe residents of three Malawian cities: Blantyre, Lilongwe and Zomba. While the rural survey comprised a random sample of citizens within a randomly selected set of Lhomwe-majority enumeration areas in the three districts, the urban survey was necessarily conducted with a non-random sample through respondent referrals. Both rounds of survey data used the same instrument, and were collected by trained enumerators.

Here, I focus on two main indicators of support for the Lhomwe chieftaincy. The first asked respondents about their level of trust in the highest ranking Lhomwe chief, Chief Ngolingoliwa.⁹ Across both rounds of the survey, 32% did not trust Ngolingoliwa at all, 11% somewhat and 57% a lot. However, a full 31% of the sample refused to answer the question or said they did not know, perhaps due to social desirability. So, in analyses below, I also consider a version of this outcome that re-codes all refusals and ‘don't know’ responses to ‘not at all’. The second question asked whether the respondent believed that the highest ranking Lhomwe chief should have authority over all Lhomwes, which captures perceptions of the chief's legitimacy.¹⁰ Among the sample who knew who the highest ranking Lhomwe chief was, 91% felt that he should have authority over all Lhomwes. However, similar to above, 15% said they did not know who the chief was or refused to answer, perhaps due to social desirability, so I also construct a version of this variable that re-codes those non-responses to ‘no’.

In terms of correlates of trust in the chief and judging the chief to be legitimate, I evaluate gender (50% male), age (mean = 36, sd = 13), years of education (mean = 7, sd = 4), employment status (57% employed), urban vs. rural (55% urban) and support for the DPP political party (65%). Model 1 of Table I reports an ordered logistic model, which regresses degree of trust in the chief on the variables described above. It shows that men, urban residents and supporters of the DPP party are all more likely to trust the chief. The correlations for both urban and DPP support are stronger when refusals are re-coded to the least socially desirable answer (model 2 of Table I).

Table I. Correlates of trust in the Lhomwe Paramount Chief.

Note: The dependent variable is a three-level measure of trust in Ngolingoliwa, including ‘not at all’, ‘somewhat’ and ‘a lot’. Model 1 includes all observations with non-missing data for the dependent variable, while model 2 re-codes refusals to answer as ‘not at all’. Both models are ordered logistic mixed-effects models with community (enumeration area in rural areas and neighbourhoods in urban areas) random effects. Standard errors are reported in parentheses.

*p < 0.10, **p < 0.05, ***p < 0.01.

We see similar patterns for logistic regressions of the legitimacy of the highest-ranking chief in Table II. Urban residence and DPP support is again consistently positively correlated with seeing the chief as legitimate in both versions of the variable. In addition, years of education is positive correlated with judgements of legitimacy, and employment is negatively so, but only when refusals are re-coded.

Table II. Correlates of seeing the Lhomwe chieftaincy as legitimate.

Note: The dependent variable is a dichotomous indicator of whether the respondent believes that the paramount chief should have authority over all Lhomwes in Malawi. The question was only asked of respondents who could identify the highest ranking Lhomwe chief. Model 1 includes all observations with non-missing data for the dependent variable, while model 2 re-codes refusals to answer as ‘no’. Both models are logistic mixed-effects models with community (enumeration area in rural areas and neighbourhoods in urban areas) random effects. Standard errors are reported in parentheses.

*p < 0.10, **p < 0.05, ***p < 0.01.

The strong positive correlation between support for the party and trust in the chief is consistent with the political underpinnings of the chief's rise to power. The positive effect of urban residence, however, is at first glance more puzzling. It represents a reversal of general trends in Malawi, with urban residents typically trusting chiefs less than rural residents across ethnic groups and rounds (Afrobarometer 2018), a pattern echoed across the broader African continent (Logan 2009). The general pattern of stronger chiefly support in rural areas is most likely due to rural residents' greater reliance on chiefs' authority in their day-to-day lives than their urban counterparts. However, chiefs also serve as the physical embodiment of cultural and ethnic identity (Van Dijk & Van Rouveroy van Nieuwaal 1999; Williams 2010), and their value in this capacity may be especially prominent in the city, consistent with a broader literature on the ethnicisation of political and social life in urban Africa (Melson & Wolpe 1970; Bates 1983; Posner 2005). Indeed, urban respondents in my surveys report stronger ethnic identification than rural respondents and were more likely than rural residents to report having benefited non-materially – in terms of increased pride, ethnic group unity or cultural knowledge – from the Lhomwe cultural revival.¹¹

CONCLUSION

This paper described the creation of a Lhomwe traditional authority structure which was largely elite-led. Based on a variety of data sources, both pre-existing and original, the paper reports three major findings. First, there were two main ways that individuals were chosen to serve in the newly created Lhomwe chieftaincy: having existing non-Lhomwe chiefs claim Lhomwe heritage and rapidly promoting lower-ranking Lhomwe chiefs through the ranks. Second, I find no evidence that the creation of the Lhomwe chieftaincy was in response to demands by Lhomwe citizens. Instead, the data suggest that Lhomwe citizens did not hold their non-co-ethnic chiefs in less regard than Malawians from those chiefs' own ethnic groups. Third, survey data show that some Lhomwes do, however, value the newly appointed Lhomwe chiefs. In particular, DPP supporters and individuals residing in cities are much more supportive of these new chiefs.

Given that the evidence in this paper is based on a single ethnic group in one country, it is worth considering which features of the Lhomwe, and the Malawian context within which their cultural revival has unfolded, are unique. Two aspects of the case seem particularly relevant. First, the nature and recency of the group's migration into Malawi, and their settlement among other ethnolinguistic groups, resulted in significant ethnic intermarriage and cultural assimilation. In fact, Malawi has one of the highest rates of interethnic marriage in Africa (Bandyopadhyay & Green 2021; Dulani et al. 2021). Such ethnic mixing, especially in southern Malawi where the Lhomwe are concentrated, allows for ethnic ambiguity and flexibility of the kind that allowed chiefs from other ethnic groups to make plausible claims to Lhomwe identity. Second, compared to other African countries, traditional authorities in Malawi enjoy more influence, garner more respect and hold more responsibilities than similar leaders in other countries. Logan (2013) reports that Malawian citizens' reliance on traditional authorities for solving local disputes and allocating land is well above average, and that their influence on local governance is the strongest across eighteen African countries. In many ways, this represents an ideal context into which a new chieftaincy can be created, as the new institution can rely on existing norms and practices of deference to support it. Thus, we might expect similarly successful inventions of traditional authorities in other contexts where chiefs are highly regarded but not all groups have them.

The findings in this paper also suggest avenues for future research on chieftaincies in Malawi and beyond. First, what are the mechanisms through which social and political leaders benefit from the support of traditional authorities? Are those benefits primarily material (e.g. access to land) or political (e.g. mobilising residents to vote or contribute to public goods)? Second, how does the recency of a chief's reign, and his or her ability to claim continuity with the past, affect the nature of governance and service delivery? While this question is important in its own right, it would also contribute to more fundamental questions about the nature of chiefly authority and legitimacy. Finally, how are the links between ethnic communities and chieftaincy structures perceived by co-ethnic and non-co-ethnic constituents? It is imperative to better understand how rooting traditional authority within ethnic claims may have differential effects across citizens and domains.