EXOGENOUS SHOCKS

How do human groups cope with exogenous shocks to social arrangements? In the middle of a global pandemic of Covid-19 the question could not be more important. Social scientists tend to divide into two camps on the issue. One argues in favour of calculation. Risks are assessed and answers are translated into policy scenarios. Individuals and communities are advised to follow the recommended strategy on the grounds of rational self-interest. This assumes that humans are instructed by Nature. We need to ‘follow the science’.

A second camp takes its lead from the French social theorist Emile Durkheim (1858–1917). Durkheim stood the assumption that Nature instructs us on its head. He argued that for human minds first to recognise Nature they draw upon a prior schooling in social relations. Our understanding of who is whom in a group (beginning with cooperation between sisters and mothers and their children – evident in some other primates as well) enables humans to recognise patterns in Nature. Only then can we read the message, but it is a socialised message, and not Nature telling us directly what to think or do.

Durkheim also addressed the issue of how more abstract and generalised notions of patterning spread. His argument was that this was the work of ceremonies (Durkheim Reference Durkheim and Fields1995 [1912]; James Reference James2003). Ceremony creates a network in which social links become primed and weighted. The group acts and dances out who and what is important to it. This was the origin, Durkheim believed, of all scientific taxonomies and mathematical constructs (Durkheim & Mauss Reference Durkheim, Mauss and Needham1963 [1903]). An interest in documenting and understanding ritual gestures formed the basis of a distinctive Durkheimian approach to theory formation in the social sciences (Smith Reference Smith2020).

This approach, however, faces an obstacle to its advancement. Because humans have learnt the lessons of their social conditioning so well, they tend rather readily to overrate their analytical competencies. All too easily, it is assumed that an existing repertoire of concepts is adequate to all challenges. Exogenous shocks, however, offer unprecedented challenges. They require (as the phrase goes) ‘thinking outside the box’. What do we then do to generate new ways of thinking?

The Durkheimian answer is that we return to acting out our difficulties in a social register. We perform rituals. The ceremonial procedures we might then engage in are many and various and may seem to the worldly-wise to be at times utterly misguided – mere rearrangement of the deck chairs on the Titanic. But Durkheim argues that what we are doing is recalibrating our social resources for effective response. We are taking a second look at the existing weightings we apply to our networks of collaboration, to see, if by digging deep in the bedrock of social experience (for Durkheim, the foundation of our knowledge), we can come up with differently configured, and so we hope, more effective sets of responses.

The literature on the social value of ceremony is vast (Smith Reference Smith1987; Bell Reference Bell1992; James Reference James2003), and cannot be adequately reviewed here, save only to mention in passing two representative collections of essays on ritual dynamics across a range of cultural settings, by Seligman et al. (Reference Seligman, Weller, Puett and Simon2008) and Hausner (Reference Hausner2013), and important conceptual work on ritual dynamics both on the battlefield and in society at large by Collins (Reference Collins2004, Reference Collins2008). These will offer the sceptical reader some assurance that the Durkheimian approach should not be dismissed out of hand.

Our purpose in the present paper is limited. We examine an instance of an exogenous shock – the outbreak of Ebola Virus Disease in Sierra Leone in 2014 – and then provide a case study to raise questions about the part played by ceremonial competence in local response.

We define the ceremonial competence of a social actor as the ability to manage a successful social performance within a specific field of social relations such that compliance or cooperation by other partners is achieved. Our aim is limited to proof of concept, in application to epidemic shocks. We cannot yet fill out the larger picture – many more case studies will be needed – but we offer reasons and evidence for supposing such an approach might be valuable.

AFRICAN CHIEFS

The present paper on the ceremonial competence of community leaders faced with Ebola is part of a larger, collaborative project on public authority in rural Africa (Kirk & Allen Reference Kirk, Allen and Onyango2022). This collective initiative aims to describe how public authority works when viewed through an ethnographic lens of detailed local observation. A concern is to bring to light the importance of informal, and at times invisible, ways in which public authority is exercised, of which ceremonial competence is an important element. For lack of such an ethnographic lens, and the fine detail it supplies, prior assessments of governance in chaotic conditions dominated by drought, civil war and epidemic in Africa have often been wide off the mark. Beneath apparent chaos there is often a surprising degree of order, and a better understanding of local public authority helps explain this order-within-chaos.

Our own work on chiefs and Ebola in Sierra Leone attaches itself to a lively debate over more than two decades on the role of chieftaincy in rural post-colonial Africa (Mamdani Reference Mamdani1996; Boone Reference Boone2003; Baldwin Reference Baldwin2013; Acemoglu et al. Reference Acemoglu, Reed and Robinson2014; Voors & van der Windt Reference Voors and van der Windt2020). Part of this debate was driven by controversy over the surprising political decision of the British government to use money from the overseas aid budget to assist the government of Sierra Leone restore its chieftaincy system in the aftermath of a brutal civil war in the 1990s (Fanthorpe Reference Fanthorpe2006). The decision was surprising because failures of chieftaincy had been among causes of the war, according to some commentators (Richards Reference Richards2005; Thomson Reference Thomson2007).

Indirect rule by chiefs in the small West African protectorate of Sierra Leone was devised by the British in the first decade of the 20th century as a tool for maintaining order in an often-inaccessible rural interior at low cost (Crowder Reference Crowder1968). Chiefs and chiefdoms in Sierra Leone derived from rule by warlords in the unsettled period prior to the imposition of British control over the interior of the country in 1896. A number of these interior warlords rose against the British in a rebellion in 1898, and in the aftermath, large chiefdoms were broken up and new chiefs appointed, now as local agents of a hegemonic British system of colonial governance (Abraham Reference Abraham1978).

Pre-colonial Sierra Leone was on the frontier of agricultural settlement at the western end of the Upper Guinean forest; families claimed rights by virtue of having first cleared land to found communities (Richards Reference Richards1986). The land-owners would elect one of their number to be the chief of the community, who took office often only after severe admonishments that he or she ruled only with the consent of the elders. The powers of the chief were sharply constrained by the male and female sodalities, Poro and Sande. All elders belonged to one or other institution, and many key decisions were made in secret and were binding on the chiefs. Ceremony (notably dancing and masquerade) was a key register in which the sodalities conveyed their decisions to the wider community.

After the British conquest, chiefs continued to be elected by the elders of the land-owning families but were required to report to the British colonial authorities. Each chiefdom had a head chief, known in British colonial parlance as a paramount chief, presiding over a hierarchy of lesser village and section chiefs. The head chief was ‘paramount’ only in the sense that no other chief could interfere in the affairs of the chiefdom. Only the British overlord claimed that right.

The main task of a chief under the modified British system was to maintain the peace by resolving local disputes over land, marriage and debts, to collect tax, and to arrange local public works, such as keeping hammock trails and stick bridges in good condition.

The colonial government continued to recognise the established right of the leading families to present candidates for chieftaincy election, but sometimes imposed turn-taking arrangements (Reed & Robinson Reference Reed and Robinson2012), to lessen conflicts over disputed origins. Among the Mende-speaking people of the south and east of the country family heads would sometimes elect a woman as a paramount chief, if a family with a right to contest considered that she was their most able candidate.

At independence, this British-devised system was retained by successive national governments, since it was a proven way of maintaining local order at low cost. The national political economy was dominated by mineral exports, and rural development remained of lesser priority. This left in place a rural social system characterised by widespread poverty, but in which a large number of ordinary citizens had a stake by virtue of land rights derived from membership of one or other of the early settled clans.Footnote ¹

In 1991 the rebel Revolutionary United Front launched a bush war intended to address the inequities of the mineral economy (Richards Reference Richards1996). They targeted the chiefs as symptoms of national decay, but then found ordinary rural citizens were prepared to fight back to defend their land, via civilian militias, and with it, came a local demand to restore chiefly rule.

The civilian militias stemmed the rebel advance but did not end the threat. With the civil war at a stalemate in 1997 a newly elected government in Sierra Leone presented the British Prime Minister, Tony Blair, with a request for help in restoring Paramount Chiefs replaced by the war. The request was answered with a chiefdom restoration and reform project. When the war finally ended, assisted by a British military intervention in 2000, the restoration project kicked off (in 2002). It resettled chiefs in the countryside, carried out investigations into problems caused by autocratic chiefs, and arranged elections for vacancies unfilled during the war.

Chiefs were then required to sign up to a new code of conduct spelling out their duties and obligations and were also given training in these duties. These included an enhanced security role. Chiefs were explicitly required to report security threats to central government, including health threats such as spread of epidemic disease. How they performed that role under the shock of the Ebola epidemic is a question to which this paper now seeks an answer.

EBOLA IN SIERRA LEONEFootnote ²

Ebola is a virus causing an acute haemorrhagic fever that reaches humans through zoonotic spillover. Consumption of bush meat is a suspected source. The virus appeared in upper West Africa in December 2013, having never before been reported or detected on the margins of the Upper West African forests.Footnote ³ The first cases were reported in the Republic of Guinea, but infection quickly spread to Liberia (Lofa County) and Sierra Leone (Kailahun District). In Sierra Leone there were around 10,000 laboratory-diagnosed cases, with deaths exceeding 4000, before the outbreak was finally ended in December 2015 (Richards Reference Richards2016: 20). The capital territory – Freetown and its peri-urban fringe, accounting for about 30% of the national population – had a total of 41% of confirmed cases. The balance of cases occurred in provincial districts, where distribution was uneven. Out of 149 chiefdoms (the lowest level of local government), and counting only laboratory-confirmed cases, 14 chiefdoms had 60% of cases, 96 chiefdoms had 40% of cases and 39 chiefdoms had no cases at all (Richards et al. Reference Richards, Mokuwa, Vandi and Mayhew2020).Footnote ⁴

These 39 chiefdoms are of interest in this paper, since we wonder whether they were just lucky, or whether avoidance of infection can tell us something about local agency in protecting against new and unprecedented disease shocks. Our approach was to explore this question through an ethnographic case study.

Previously, one of us had made a detailed study of a chiefdom where there had been a large outbreak early in the epidemic, pointing to the role of the local public authorities in rapidly organising a chiefdom volunteer defence force largely effective in quickly terminating the outbreak (Richards Reference Richards2016: Ch. 6). The objective for the present paper was to provide a matching study of chiefdom-level response in a chiefdom with no cases. Kamajei chiefdom, located in a rice-farming zone at the geographic centre of the country, was selected for reasons explained below.

Ebola spreads through contact with the body fluids of an infected patient. This means that cases are clustered among the family members who care for the sick (a fact reflected in the Mende name for the disease, bonda wote – family turn around [or hold back]), or who prepare the corpse for burial (a majority of Sierra Leoneans are Muslims and washing the body before burial is a religious requirement).

These emergent clusters are then linked by long-distance transfers of infection (Richards et al. Reference Richards, Amara, Ferme, Kamara, Mokuwa, Sheriff, Suluku and Voors2015), resulting from an increasingly desperate search for treatment as the patient falls into a spiral of sickness, beginning with three days of acute fever and headaches (‘dry’ Ebola), and ending in organ collapse after a further three days (‘wet’ Ebola).

‘Dry’ Ebola resembles malaria, and worried patients search for malaria treatment among local herbalists and pharmacists, more trusted for their ability to cure malaria than clinics or hospitals. After the onset of ‘wet’ Ebola the patient cannot be safely moved and the pharmacist or herbalist also becomes infected, and infects other patients, with a new cluster of cases establishing itself, often at a considerable distance from the original cluster.

In this way cases jumped from the Guinea border in Kailahun District, to Kenema, and along the main road linking Kenema, Bo to the capital Freetown, with the virus establishing itself in busy market centres such as Waterloo, on the outskirts of Freetown, and from there spreading back up-country into centres in the north such as Port Loko, Makeni and Kambia (Richards et al. Reference Richards, Amara, Ferme, Kamara, Mokuwa, Sheriff, Suluku and Voors2015; Richards Reference Richards2016: 46).

Initial international estimates assumed that the initial clusters would be stemmed by the well-tried methods known to specialist agencies such as Médecins Sans Frontières (MSF) who had tackled previous African outbreaks. But this left out of account the regional road system, greatly improved by the European Union in the aftermath of the civil war, and the capacity of this road system to distribute case clusters rapidly to all parts of the country.

The first cases (in northern Kailahun District) were reported to WHO on 25 May 2014. There was then a major outbreak of nosocomial infection in Kenema government hospital in June, and case clusters began to appear along the main road to the Freetown in July as terrified people fled Kenema, seeking help from relatives, and herbalists and pharmacists, in Bo and Moyamba districts, and even Freetown itself (Richards et al. Reference Richards, Amara, Ferme, Kamara, Mokuwa, Sheriff, Suluku and Voors2015).

By August MSF reported it could no longer cope, and the WHO declared a Public Health Emergency of International Concern (PHEIC). This triggered a major international response, fully deployed by October 2014. An overall decline in cases began to be apparent by December. Follow-up studies (Richards et al. Reference Richards, Mokuwa, Vandi and Mayhew2020) establish that local response was an important factor in reducing cases in the earliest phases of the epidemic, but these local responses were later dwarfed and/or marginalised by later international responses (Walsh & Johnson Reference Walsh and Johnson2018).

As already mentioned, 39 chiefdoms (26%) had no cases (according to the national records of laboratory-confirmed cases). Most of these were in Bonthe (8/11 chiefdoms) Pujehun (6/12 chiefdoms) and Koinadugu Districts (8/11 chiefdoms), totalling 22 chiefdoms with no cases. These are dead-end districts in terms of road communications, and Ebola checkpoints restricted in-bound traffic. Koinadugu District, for example, quarantined itself by establishing a major checkpoint on the main road from Freetown, where supplies in and out were transferred to local vehicles and passenger numbers were highly limited. We discounted these readily isolated districts.

A small cluster of chiefdoms at the centre of the country attracted our further attention. These were all on or close to the Freetown highway along which many early cases had moved. Two of these chiefdoms (Gbo and Kowa) had between them only a handful of cases, and Dasse and Kamajei had no cases at all. Kamajei stood out, since it had no cases but is bisected by the road to Freetown, and also linked by bush paths, and inter-marriage, to two chiefdoms to the north and west (Bonkolenken and Kori) where there were substantial outbreaks of the disease from July (Kori) and from October (Bonkolenken).

CASE STUDY: KAMAJEI CHIEFDOM

Kamajei chiefdom is lightly populated. It has a total population of about 10,000, and is divided into two halves, north and south of the main cross-country road to Freetown. The headquarters is Senehun, a stopping point on the motor road. But the area to the north is centred on Gondama, a large village about 12 miles from Senehun, bordered by hills to the north-east and an extensive area of grassy wetlands to the west. It is the former headquarters of the Maje chiefdom, amalgamated by the British with an equally small chiefdom (Kamagai) centred on Senehun.

This amalgamation was undertaken for practical reasons and appears not yet fully accepted in the eyes of some people. Maje was a marcher state, controlled by a Kissi warlord, Sovula, from Kalaihun, described on his gravestone as ‘the terror of Maje’. He appears to have worked under the authorisation of leaders of a neighbouring, larger polity, Taiama (Kori chiefdom) and his task seems to have been to prevent Temne incursions from Bonkolenken chiefdom from the north, and perhaps also to maintain control among a somewhat restive local population of Banta-speaking background with family connections in areas to the south.

This complex terrain is not easily governed, and any paramount chief is required to tread carefully, bearing in mind the historical fractures dividing some of the leading families, as well as the demands of the leaders of the sodalities, of which the Kpa-Mende warlord association, Wunde, is an important presence.

Being paramount chief does not necessarily coincide with high rank in the secret associations, and this is a further potential constraint on the chief's power. Putting a foot wrong, not least in a dance (given that sodalities enact their will or judgement performatively), can soon lead to sanctions.

A first (highly symbolic) move for a new paramount chief is choice of residence. In such a complex and contested terrain this move is fraught with some risk. The present chief is the grandson of Madam Botei Sovula, herself said to be a granddaughter of the Kissi warlord, a noted woman chief who took charge of Kamajei in the 1960s from a modest house on the northern side of the main road in Senehun. Since she died in the 1980s the chieftaincy has rotated through two other ruling families (both of whom are based in Senehun) until returning to the present chief. He has chosen to reside in Gondama, even if necessarily spending much time on chiefdom affairs in meetings in Senehun. The chief justifies the choice of having his home in Gondama in terms of the need to reverse the under-development of the northern part of the chiefdom, from which he hails.

The south has long been connected by a feeder road to the former government railway station at Mano, and the employment and marketing hub of Njala University's main campus at Mokonde. Gondama, by contrast, was connected to the rest of the chiefdom only by a tenuous footpath crossing several flooding rivers. Dubious stick bridges were plentiful; drowning was a real risk to the traveller. But slowly, over the years, the track has been made partially motorable, and now a final road bridge over the Kati river is approaching completion. The road will then go further and join Gondama to Yele in the northern province. The government has funded the road and bridge, but with constant pressure from the chief. The symbolic presence of a paramount chief living across a river cut off from the rest of the country during the rains was a powerful symbolic driver rewarded by lasting investment.

A local capacity that paramount chiefs have in great abundance is knowledge of and involvement in the domestic concerns of their subjects, fortified by the expectation that they will participate in the rituals accompanying these transitions. This suggests, in turn, that chiefs exercise beneficial influence through command of protocols and rights, rather than through executive power. A ceremonially competent chief is said to ‘know how to talk to people’. This was an important factor in being able to persuade families to accept the reality of Ebola and to adopt temporary but major changes in social conventions regarding hospitality, sick visiting and death.

So, treading carefully, and dancing adroitly, are key requirements for a chief. The rewards are often meagre (the tax base is low and government too cash-strapped to help). The charge a chief is doing little to help the chiefdom is all too easily voiced. Offence at slights – imagined or real – lie in wait at every turn for a chief who is inexpert in the tangled history of her or his complex terrain. But an asset is that he or she lives with the people, and thus builds up an enviable detailed knowledge of a micro-sociological landscape.

Chiefs acquire much of this local knowledge through their involvement in the day-to-day business of dispute resolution. A government reform in the post-colonial period took away the Paramount Chief's role in presiding over local courts, administering customary law relating to family, land and debts. Today, these courts are presided over by a Court Chairman. Official chiefdom courts were (and are), however, only the tip of the iceberg. Informal moots carry the main burden of local dispute resolution (Gibbs Reference Gibbs1963). Chiefs and elders continue to resolve a vast majority of cases informally from their verandas (estimates reported in Mokuwa et al. (Reference Mokuwa, Voors, Bulte and Richards2011) suggest more than 90% of all village disputes are so settled). Most of these cases are intra-household matters, resolved by heads of families. Where the dispute is between families, town and section chiefs become involved. Anything that cannot be resolved at the village level ends up with the paramount chief for advice and arbitration. Much of this dispute-management activity is largely invisible from outside, but it places paramount chiefs at the nerve centre of their communities.

Each chief's store of local social knowledge is amplified by the requirement that life events (births, marriages and deaths) must be reported to the chief. This was significant during the time of Ebola because safe burial was enormously important in infection control. Where responders did not have the confidence or support of the local chief, infections were liable to rise (Voors & van der Windt Reference Voors and van der Windt2020). Chiefs had a choice – whether to side with their citizens and turn a blind eye to breaches of burial rules, or to use their powers to persuade people to comply, even while expressing the deepest sympathy and concern, in order to keep case numbers under control.

Early in the epidemic, while the international response was still forming, chiefs in Kailahun seized the initiative, in some case training and equipping their own burial teams and developing chiefdom byelaws that then became the national standard from August 2014 (Richards Reference Richards2016: Ch. 6). In other areas, awareness of an approaching acute dilemma caused some chiefs to think pre-emptively. Their role in managing death among their subjects would be easier if there were no Ebola cases to begin with. That meant mobilising the chiefdom to keep the threat at bay.

EBOLA RISKS IN KAMAJEI

Risk of nosocomial infection was a major driver of Ebola spread, especially the outbreak at Kenema government hospital in June 2014. Persons exposed to the disease, then falling sick with malaria-like symptoms, tried by all possible means to avoid detention in makeshift and ill-equipped holding facilities, such as the Lassa fever ward at Kenema hospital, where risk of cross-infection with Ebola was high. The option was to seek assistance from families and local herbal practitioners in home villages. If it then turned out the patient had Ebola, a local case cluster would emerge, as family and friends tried desperately but ineffectively to help a patient with ‘wet’ symptoms and became infected themselves.

As we have already shown, it might be reasonable to assume that many of the 39 chiefdoms without Ebola cases in Sierra Leone in 2014–15 were protected mainly by distance and isolation. That proviso, however, does not apply to Kamajei chiefdom, with its headquarters on a busy main road and Ebola cases nearby. A substantial outbreak in an outlying village of Kori chiefdom was seeded in July 2014 by an infected patient from Kenema, seeking medical assistance from his sister, a noted local herbalist. In October, an outbreak in Bonkolenken chiefdom was started when a sick patient was treated at a small private charity hospital in Yele. Kori and Bonkolenken chiefdoms share extensive borders with northern Kamajei chiefdom, and these are crossed by multiple footpaths leading to Gondama, along which itinerant traders and family visitors pass daily, often ferried by motorcycle taxis.

The government posted army and police to both active case clusters as soon as cases were confirmed by laboratory tests, but numbers were never sufficient to prevent all inter-village movements. There was always a risk that infected villagers might seek help from health centres in neighbouring communities.

In Gondama the village health centre, in 2014, was temporarily located in a house in the village, pending completion of a new centre on land in the village outskirts. The paramount chief immediately built a lean-to shelter against a wall of the temporary health centre in which to quarantine sick patients, pending testing. When we asked him why he had taken this pre-emptive action he explained that he anticipated that unannounced arrivals at the health centre were a major infection threat. He had worked in public health before assuming the chieftaincy and found out about Ebola as soon as cases were announced in Guinea (December 2013). His brother had been about to take a job in South Sudan when his plans were cancelled by the WHO announcement that Guinea had Ebola and neighbouring countries were at risk. This impelled Chief Sovula to find out for himself about this new and dangerous disease, even before it was officially declared in Sierra Leone (25 May 2014).

In neighbouring Kori chiefdom, a stick bridge on a track leading from the infected village to a health centre in a neighbouring village was hacked down by two youth leaders from the latter village, against the wishes of families whose farmland bordered the infected settlement. The Paramount Chief of Kori chiefdom dismissed the complaints of the land holders on the grounds that in order to come to the chiefdom headquarters to inform him they had broken the government's newly announced bye-laws on Ebola control, restricting all unnecessary inter-village movement, and were accordingly liable to pay a fine. Whether the fine was paid was unclear, but often fines serve as ceremonial threats to induce compliance and are lifted by negotiation among the conflicted parties. It is within the ceremonial competence of chiefdom authorities both to impose and to remit fines, as an inducement to public good order.

Enough has now been said, we think, to suggest that the risk of Ebola entering Kamajei chiefdom was immediate and real, and that public authority, led by the paramount chief, was alert to the threat posed by uncontrolled local movement of persons of unknown health status. The claim by Voors & van der Windt (Reference Voors and van der Windt2020) that a paramount chief is an autocrat is harder to substantiate, however. Each chief manoeuvres between government and the people, but also needs to take account of the myriad and historically complex fault lines potentially dividing various interests in his or her chiefdom. The competing ambitions of the leading families must be balanced, and messages from the powerful village sodalities require attention. For this task, material resources are pitifully meagre. The people are poor, and often hungry, and tax revenues tiny. Thus, the executive capacity of a chief is strictly limited. The main tool within a chief's grasp is to propose byelaws to be enacted by the chiefdom council of which the chief is chair. These must be implemented deftly. They need to make sense to the majority and administered in ways that appear fair. Making them stick is a major test of chiefly ceremonial competence.

From early August 2014, with the international declaration of a PHEIC over Ebola the government adopted a standard set of chiefdom Ebola byelaws already field tested by the Kailahun chiefs who developed them (Richards Reference Richards2016: 131). Each chiefdom now had a framework within which to develop its Ebola response. However, no chief had the public authority or resources to deliver Ebola byelaws by force. Each chiefdom has a handful of court bailiffs (chiefdom police) and a small lock-up. But mass civil disobedience would overwhelm these arrangements. Justice is largely self-administered, so must be delivered in a way that makes sense to a majority. The bottom line is a fine paid eventually, but not without bargaining and pleading. As noted, some villagers allege that chiefs who levied arbitrary or unjust fines were a factor in the descent towards civil war (Richards Reference Richards2005; Mokuwa et al. Reference Mokuwa, Voors, Bulte and Richards2011). Arriving at socially acceptable outcomes for offences against local byelaws is a major test of the ceremonial competence (and determination) of a chief.

THE KAMAJEI EBOLA TASK FORCE COMMITTEE

Ebola prevention at chiefdom level hinged on keeping out visitors who might have been exposed to the disease elsewhere. Prior to the national application of Ebola byelaws in August 2014 any stranger lodging in a village was supposed to be reported to the chief. From this point onwards, no visitors were to be allowed. Any exceptions would be met with a steep fine.

But more than this was required. People had to acquire a common appreciation of how Ebola might enter and then to develop the shared resolve to act to minimise the risk. Concrete action would be required – e.g. roadblocks imposed and manned by volunteers; communities would need to come together to feed the volunteers.

Coordination required a chiefdom Ebola task force committee. This was formed by the Paramount Chief after a briefing by the district authorities. Below, we draw on evidence from the minutes of the seven meetings of this committee to bring out issues of ceremonial competence as key concerns in mobilising an Ebola response. An initial symbolic balancing act (given the merged origins of the chiefdom – see above) is evident in the constitution of the meetings. The paramount chief assumed the chair but the task of recording the minutes was given to Pastor Augustine Margao from Pelewahun, a village at the southern extremity of the chiefdom.Footnote ⁵

The debates reported in the minutes are quite various, and at times might appear to an outsider to be quite irrelevant to Ebola control. We argue such a conclusion would be unwarranted and show instead that the overriding aim of the committee was to bring about sufficient unity of purpose to bind a diverse chiefdom together to achieve a common goal.

However, we do not want to argue that ceremonial competence is merely a makeweight for missing material resources, but rather that it has its own independent capacity to drive infection control. To make our argument we turn to terminology suggested by the philosopher J.L. Austin (Reference Austin, Sbisà and Urmson1975 [1962]). Ceremonially competent speech has what Austin termed ‘illocutionary’ force. Pronouncements are more than empty words; they cause something to be done, or done in a certain way, and not in another, as when the committee issues verdicts on apparently irrelevant issues such as an order of precedence in arranging meetings, or who speaks when.

Structural silence also has illocutionary force in rural communities in Sierra Leone. The voice of the paramount chief is heard little in the minutes. This is not a sign of lack of leadership but its opposite – an arrangement whereby the chief listens to the debate and then decides. Speaking is the job of one of his key officials, called (logically enough) the chiefdom speaker. The speaker is often an unsuccessful rival, which is to say he or she represents a clan unsuccessful in capturing election to the chieftaincy. The job of the speaker is to articulate the debate in such a way that options, pitfalls and clash points are exposed to and weighted by the chief in making a final decision.

What the minutes reveal is the ritual processes through which speech is converted into agreement on collective action. At the level of the committee, the actions themselves are often highly symbolic. Good examples are the deliberations of the committee about how to distribute a donation of Veronica Buckets, to facilitate hand washing with chlorinated water in public places, or which groups and individuals, in which locations, should be given the Ebola Task Force tee-shirts. Addressing such issues in the midst of a huge health crisis might seem trivial in the face of potential disaster. But these were far from inconsequential discussions. Both spoke to motivations beyond the meeting – whether villages were being treated fairly, according to their size and historical importance, for example. Getting an agreed allocation would improve chances of effective inter-village cooperation over, for example, roadblocks or reporting suspicious movements by strangers. In this light a successful ritual can be seen as a rehearsal for practical collaboration to come.

By and large, epidemiologists have tended to ignore or down-play the significance of ritual gesture in infection control (for a welcome exception see Greenhalgh Reference Greenhalgh2020). But as a result of global experience with Covid-19 there is now a wider appreciation that washing hands or wearing a mask is a kind of ritual engagement triggering a greater awareness on the part of those acting in these ways that they need to become more generally alert to infection risks (Tett Reference Tett2021). As Durkheim would have insisted, ceremonial competence has its rewards.

The work of Goffman (Reference Goffman and Goffman1967) establishes the role of apparently minor quotidian rituals in sustaining much wider social or political commitments. Former British prime minister Harold Wilson spent considerable effort in deciding on the seating plan for cabinet meetings (6, Perri 2016), appreciating that the success or failure of larger arguments might flow from a judicious positioning of debaters within an order of ritual interaction (Collins Reference Collins2004). That which is well-ordered in the microcosm is a basis for achieving order in larger social contexts, as Mary Douglas convincingly argues in relation to food and hospitality (Douglas Reference Douglas1982).

According to the minutes, discussing the proper way to hold meetings occupied a significant amount of time, reflecting a subtle, and not so subtle, jockeying for position among chiefdom numu waisia (big people). An example of this is a substantial discussion devoted to debating whether the councillor and parliamentarian were politicians. The minutes record that only the parliamentarian was to be so considered, despite both being elected by secret ballot under national electoral law. Evidently, there was more behind this issue than appears on paper. The councillor was a regular attender at meetings. The parliamentarian made promises but was thought by some to deliver less. In effect, the minutes encompass a gentle rebuke that as a class, politicians are too busy, with priorities elsewhere, to grapple efficiently with local challenges, such as acquiring resources for basic hygiene. Politicians talk; others do.

The location of meetings was also a point of discussion freighted with ceremonial significance. The first meeting was held in Senehun, headquarters of the ‘amalgamated’ chiefdom, and located on a main road down which Ebola ambulances ran daily, with sirens screaming. The final meeting (of a sequence of seven) was held in Gondama, the much quieter off-road former headquarters of Maje chiefdom, and home to the current paramount chief. This last meeting may have perhaps been intended to demonstrate to others on the committee the appalling state of the access track.

In between, there was a meeting in Mogbuama, a large rice-farming village halfway between Senehun and Gondama. Mogbuama clans do not participate in competition for the chieftaincy, which rotates between Senehun and Gondama. But the village is home to a large Wunde society enclosure, an indicator of former sublimated political influence. Richards (Reference Richards1986) was told that deceased members of one of the land-owning clans in Mogbuama are not buried but transform into the elephants they once hunted. To be successful, hunters address their prey in the Banta language, an apparent reference to a history no longer openly told. The choice of location was perhaps a rare public nod towards the occluded politics of a marginalised group of ancestors with a continued stake in the welfare of the chiefdom.

Debates over points of principle are frequently recorded in the minutes. Someone raised the important question ‘where were the youth?’. The committee, the questioner observed, was ‘full of old people’, whereas quarantining villages through effective monitoring of incursions by strangers required energetic youth to staff checkpoints. The seriousness of the issue was underlined by a report that youth in Senehun had begun to take action into their own hands. The town's main road location made it hard to prevent visitors arriving by any of the frequent vehicles stopping in the town. Asserting that there were frequent breaches in Ebola security, the youth of the town had formed their own quarantine task force. This alarmed the committee. Members doubtless recalled wartime experiences when civil defence volunteers from Senehun controlled a notorious road-block on the Freetown road, taxing travellers at gun point. Getting the youth back on board was urgently necessary. It was agreed forthwith that the chiefdom youth leader should be invited to sit on the committee.

This brought a further important fact to light; some villages had no youth leaders. The Paramount Chief, as chair, was asked to discipline the sub-chiefs at whose door blame for this oversight was laid, and also to ensure youth leaders were now promptly appointed. In confronting Ebola, the chief-in-council had begun to address some of its own weaknesses, thereby enhancing its authority.

As in the civil war, the mobilisation of youth was a crucial issue. It was this group that would provide the backbone of labour to man the checkpoints necessary to the prevention of inter-village movements, to ensure that Ebola would not spread. Issues of feeding, or provision of light, were essential to the activities and motivation of volunteers.

The village chief of Mogbuama (Madam Mamawa Tarawali) described to us in some detail the trouble she had taken over supervising the roadblock into her village, and in checking the youth volunteers were up to their task. She took particular care over preparing and delivering meals, to avoid a scenario in which hungry volunteers abandoned their post to seek food. She also described how she sat with the volunteers to show them by example how to do their work. In the course of one such session the chief turned back, politely but implacably, some of her own relatives from Freetown.Footnote ⁶ All outsiders, without exception, she asserted, were potential infection risks. Forcing members of her own family immediately to return to where they had come from was an especially impressive demonstration of determination to keep the disease out of Mogbuama, and the news of this incident spread far and wide in the chiefdom.

The minutes describe a planned visit by the committee to Joya, one of the larger villages in the east of the chiefdom, in order to help the village chief solve a dispute with her people. This dispute had nothing to do with Ebola, but committee members realised that a public quarrel of this sort would undermine the community cooperation needed to keep Ebola at bay. A concern for good governance more generally was essential for building up the social cohesion needed to fight Ebola.

The committee also asked for regular updates from the Community Health Officer (CHO) on case numbers. Here, it is interesting to note the evident pride the committee members took in Kamajei's status, as one of only a handful of chiefdoms in a heavily infected sector to have kept the disease at bay. In fact, there was an outcry on the committee over a misdiagnosed patient, who was somehow erroneously included in the statistics for Kamajei chiefdom. The person should be sent back home, the committee suggested, since even though he was no longer a contact at risk of developing the disease, his test status threatened the chiefdom's ‘clean’ record. In short, the Ebola status of both individuals and communities had taken on a competitive aspect, and the Paramount Chief invested considerable effort in ensuring the records of the District Ebola Response Committee (DERC) in Moyamba were corrected.

This makes clear that ‘dashboards’ covering an epidemic are more than simply tools for public information. They become rituals that actively engage public response aimed at reducing or preventing cases. Dashboard symbolism is a topic to which responders in Sierra Leone might have paid more careful attention. Red signified a district with active cases, and green a (so-called) silent district or chiefdom. In Sierra Leone, red and green are also the colours of the two main political parties. This was unfortunate, since some citizens in ‘green’ voting districts were adamant that neglect by a ‘red’ government had made the epidemic worse in their areas.

The tendency for epidemics to take on political colouring is apparent also from the more recent pandemic of Covid-19, as seen at its worst in the partisan rows in the USA over masks. Coherently managing the symbolism of epidemic response to ensure that polarising outcomes are avoided would be a practical application of the arguments concerning ceremonial competence advanced in this paper.

Kamajei chiefdom was widely praised, nationally, for its efforts in keeping Ebola at zero. However, this paradox also flagged a contradiction. The Paramount Chief explained that his requests for external assistance were largely ignored because of the success of his chiefdom in preventing infection. He gave us an example based on his experience in building the above-mentioned booths to house sick visitors. Typically, testing could take up to five days or more. The sample would need to be conveyed by a motorcycle taxi rider to Bo or Moyamba and the results couriered back by the same means. During this time, undiagnosed patients needed to be kept outside the health post, to allow vital maternity activities to continue.

So, the booths needed proper roofing, in the heaviest part of the rainy season, if they were to be effective. Requests to an international aid agency tasked with Ebola response in the district for a tarpaulin to waterproof the shelter were turned down on the grounds that Kamajei chiefdom had no cases, so needed no assistance. The chief wondered (ironically) whether his Ebola task force should deliberately allow in a few cases, to put the chiefdom on the aid map. He might not have been surprised to learn that in some districts where infection chains were prolonged, attempts were made to transfer cases over chiefdom, constituency, or even district, borders in the hope of attracting financial inputs from the well-funded international response. Walsh & Johnson (Reference Walsh and Johnson2018: 297) remark that there were ‘problems with particular local leaders trying to profit from the [Ebola] response’.

The chiefdom Ebola task force committee in Kamajei received some limited financial support from the DERC, but no more than barely sufficient to allow it to operate as a coordinating body for the local voluntary action. Yet despite the paucity of funds the reputation of the community Ebola response remains high today. In subsequent interviews in Gondama over Covid-19 in 2020 villagers repeatedly mentioned the success of civilian efforts to avoid Ebola transmission, led by the paramount chief and his council, and how this earlier success provided reasons for believing the new infection could also be kept at bay. ‘We know the rules, and now we need to stick to them’, was how one informant put it (Kamara et al. Reference Kamara, Mokuwa and Richards2020).