Oppgrader til nyeste versjon av Internet eksplorer for best mulig visning av siden. Klikk her for for å skjule denne meldingen
Ikke pålogget
{{session.user.firstName}} {{session.user.lastName}}
Du har tilgang til Idunn gjennom , & {{sessionPartyGroup.name}}

Classifying, Domesticating and Extirpating the Zanzibar Leopard, a Transgressive Felid

Chief Editor, Polar Research, Norwegian Polar Institute, Fram Centre

Dr Helle V. Goldman carried out her anthropological fieldwork in rural Pemba, the northern island in the Zanzibar archipelago, in 1992-93. After receiving her doctorate from New York University, she then worked for a development and nature conservation project in Zanzibar. At the Norwegian Polar Institute in Tromsø since 1998, she has intermittently continued to research Zanzibari culture and wildlife for more than two decades. She recently returned to her original field site in Pemba to study family histories.

Wolfson College, University of Cambridge

Martin T. Walsh is a social anthropologist with extensive experience in international development and the assessment of research impact. He is an Adjunct Professor in the School of Business Studies and Humanities at the Nelson Mandela African Institution of Science and Technology (NM-AIST) in Arusha, Tanzania, and a Research Affiliate in the Sheffield Institute for International Development (SIID) in the University of Sheffield, UK. He is also a Senior Member of Wolfson College in the University of Cambridge and is currently the Book Reviews Editor of Tanzanian Affairs. His ongoing research focuses on the cultural and natural history of the Zanzibar archipelago and the wider Western Indian Ocean region.

A key aspect of the social construction of animals is how they are made to fit into categories such as wild versus domesticated, livestock versus pet, game versus vermin and edible versus non-edible. In this paper, we consider the case of the leopard on the island of Unguja, in the Zanzibar archipelago, Tanzania. It is thought that witches secretly feed, breed and deploy leopards for purposes that include terrorizing other people and supplying food to their owners. Perceived by the general population as a dangerous witches’ familiar, and by the government sometimes as vermin, at others as meriting official protection, the Zanzibar leopard is a transgressive, “out-of-place” animal that defies tidy classification. This illustrates how, once they are incorporated into human social worlds, animals are assigned to categories that shape how they are understood and acted upon. “Domestication” in this broader sense has consequences for their survival. In Zanzibar, the association of leopards with witchcraft has contributed to their probable extirpation through grassroots and government extermination efforts. Recently – but in all likelihood too late for the leopard – there are indications that Zanzibaris have begun to see a potential benefit in them.

Keywords: Anthrozoology, human-wildlife conflict, Zanzibar, leopards, domestication


[In] a certain Chinese encyclopedia … it is written that animals are divided into (a) those that belong to the Emperor, (b) embalmed ones, (c) those that are trained, (d) suckling pigs, (e) mermaids, (f) fabulous ones, (g) stray dogs, (h) those that are included in this classification, (i) those that tremble as if they were mad, (j) innumerable ones, (k) those drawn with a very fine camel’s hair brush, (l) others, (m) those that have just broken a flower vase, (n) those that resemble flies from a distance.

(Borges, 1964, p. 103)

At least one highly influential thinker (Foucault, 2005) and one well-known anthropologist (Sahlins, 1995) have held up the zoological classification system quoted above as an exemplar of the sometimes radically different ways non-Western cultures organize their perceptions compared to our own. It is, of course, a complete fiction (Windschuttle, 1997), the product of the fertile imagination of the Argentine writer Jorge Luis Borges, a founder of magical realism in literature. This makes the passage particularly suitable to introducing an essay on the shifting, overlapping, conflicting and fabulous classifications of the Zanzibar leopard, an animal that now probably survives only in our imaginations.

When people bring animals into human culture, endowing them with functional and symbolic meaning, they are assigned to categories that shape how they are perceived and acted upon (Douglas, 1957; Lévi-Strauss, 1966; Willis, 1974; 1990; Descola, 2013). An example of this is whether an animal is considered edible or non-edible, or may be understood to be edible under certain conditions or by certain human groups (e.g., Leach, 2015, p. 154). In addition to the edible/non-edible distinction, being classified in our society as a laboratory animal, pet, game animal, pest or livestock, among other categories, also has enormous consequences for the creature in question (DeMello, 2012). Likewise, so do science-based designations such as species and subspecies, and assessments like Least Concern, Vulnerable, Endangered and Extinct. The latter are categories applied by the International Union for Conservation of Nature (IUCN) as part of this global organization’s work guiding wildlife conservation action (IUCN, 2018). Such actions can involve large-scale economic and political processes, nationally and internationally.

The Zanzibar leopard – the primary focus of our anthrozoological research in Zanzibar for more than two decades – has been classified in a variety of cross-cutting ways by Zanzibaris and non-Zanzibaris, as we describe below. Among the most salient aspects of Zanzibari perceptions of the leopard is the understanding that whereas some leopards are wild, other leopards are partially domesticated animals belonging to witches. Beginning 8,000 years ago, if not further back in time, the domestication of animals has shaped human history and remains a key way that people relate to animals. Animals are usually domesticated when they are kept for a distinct purpose (such as transportation, food or protection); human-controlled breeding and mortality result in genetically-based behavioural and physical modifications, some of which compromise the animals’ ability to survive in the wild (DeMello, 2012). Some species have become only semi-domesticated, with humans having assumed only partial control over their movements, food supply, breeding and mortality, rendering them less altered by artificial selection than full domesticates. An example is the reindeer herded by the indigenous peoples of northern Fennoscandia and Russia.

Here we consider a consider a case of the pseudo-domestication of the leopard, the largest predator on the island of Unguja, in the Zanzibar archipelago, which lies off the coast of mainland Tanzania. We will also examine how the leopard has been domesticated in the larger sense of being socially constructed and classified. The leopard emerges as a transgressive animal, crossing symbolic and physical boundaries, and is therefore “out of place” (Douglas, 1966), which has contributed to its extermination.

The scientific classification of the Zanzibar leopard

According to the system of binomial nomenclature devised by the Swedish biologist Karl Linnaeus in the mid-18th century and still in use by zoological taxonomists today, the leopard is known as the species Panthera pardus, species being the basic unit of taxonomy. For sexually reproducing animals, the term is often defined as the largest group of animals within which mating produces fertile offspring. (Although this is unsatisfactory for several reasons, it remains a useful thumbnail definition.) The leopard is a member of the Family Felidae, above that, the Order Carnivora and, at an even higher level, the Class Mammalia.

In 1932, British zoologist Reginald Pocock, on the basis of his examination of two skins and one skull in the British Museum, determined that the specimens that had been sent from Zanzibar (at that time a British Protectorate) were sufficiently distinctive in comparison to specimens from other parts of Africa to warrant a subspecies designation. Subspecies is the category ranking just below species. It designates a subpopulation of a species that is distinct from other subpopulations of the same species and that lives in a more or less bounded part of the species’ distribution range. Specifically, Pocock determined that Zanzibar leopards were smaller-bodied than other leopards and their coats were marked by unusually small, densely packed spots. Following the taxonomic convention for naming subspecies, he added a third name to the existing binomial: the new subspecies was called Panthera pardus adersi (Pocock, 1932).

Fig. 1

The mounted Zanzibar leopard specimen, shot in December 1927, on display at the Zanzibar Museum. (Photo by Martin T. Walsh.)

By 1964, 27 formally named subspecies of leopard had been published in the literature (Miththapala et al., 1996). Once found from westernmost Africa to eastern Asia, the leopard had the widest distribution of all the wild cat species and there was considerable regional variation for zoologists to account for. The development of genetic sequencing techniques has challenged this profusion of leopard subspecies, pruning them down to eight or nine (Uphyrkina, et al. 2001; CCTF, 2017). Zanzibar leopard tissue samples were not included in these studies, although they are potentially available from the six specimens of Zanzibar leopard held by museums in the UK, the USA and Zanzibar (Walsh & Goldman, 2008; Figs. 1, 2). It therefore remains unknown whether its subspecific status holds up against the genetic yardsticks that have replaced traditional morphological methods of grouping animals into scientifically meaningful units.

By 1996, when we began our research, the Zanzibar leopard was presumed “extirpated” by the Cat Specialist Group of the IUCN (Nowell & Jackson, 1996, 27). (The IUCN does not give labels such as Extinct and Endangered to subspecies. The current status of the leopard as a species is Vulnerable.) Nonetheless, that same year, the chairman of the Group – co-editor of the report that proclaimed the Zanzibar leopard’s extinction – was sufficiently open to the possibility of the leopard’s persistence in Zanzibar that he helped muster funds to pay for South African wildlife experts to search for solid evidence of leopards on the island. When this produced nothing (Stuart & Stuart, 1997), the international conservation establishment lost interest in the Zanzibar leopard. However, as stories of fresh leopard sightings and leopard attacks on livestock continued to come to the attention of the authorities in Zanzibar, the government did not give up on the leopard, as we will explain.

Fig. 2

Two Zanzibar leopard skins in the collection of the Harvard Museum of Comparative Zoology. Both were females killed in the 1930s. The skin on the left is from a leopard trapped by Zanzibaris, probably using a trap similar to that shown in Fig. 5. (Photo by Jon Winther-Hansen.)

Ethnozoological classifications of the Zanzibar leopard

Zanzibaris, who are Swahili-speakers, have their own words for leopards, of course. The generic Swahili term for the animal is chui. The leopard is understood to be a mnyama (“animal”/”mammal”), in contradistinction to birds (ndege), fish (samaki) and bugs (wadudu) (see Heine & Legère, 1995, 15–23, for an overview of Swahili ethnobiological classification). Informants, particularly people who hunt or who have some knowledge of wildlife for other reasons, also refer to a variety of named types of leopards. These are based – as in traditional Western zoology – on morphological characteristics, chiefly overall body size and build and the colouring and patterning of the fur (Table 1). There are short, robustly built leopard types and long, slender leopards; some leopard types are predominantly yellow and others are rufous; some have large markings and some have small markings; and so on. Ferocity is occasionally mentioned as a characteristic trait of a particular type.

Table 1

Named types of leopards in Zanzibar and their descriptions, with temperamental characteristics noted below the table. Dashes indicate descriptions by different informants. (For a more detailed version of this table, see table 3.1, Walsh & Goldman, 2017. For an in-depth discussion of the terms, including their etymologies and information about the informants who were interviewed, see Walsh, 1997, 39–54.)

Swahili nameCharacteristics of body Characteristics of coat
bete short, robust
kibete small
bungala large, long glossy
chui asilia large yellowish
chui uwanda small rufous
futizi more gracile than kisutu, less so than konge
kariuki small or medium-sized rufous patches
keke large
kichigi b
  • short

  • small

  • small, short

  • rufous with no white

  • rufous

  • black and white, no yellow

koko small body
  • tall, slender

  • long

  • long body, big head

  • long, tall

  • red and black with little white

  • two kinds: black with big spots; and white with small spots

  • black and white

konga largelarger with more numerous spots than unyasi
  • long

  • larger than kisutu and unyasi

  • large

  • yellow with some white

  • the colour of sisal leaves (= mkonge)

  • long, slender

  • long

  • yellow with black spots

  • smaller markings than kisutu

  • richer yellow than kichigi and mkonge

shambi-shambi similar in appearance to the African civet
shwamba very long bodyvery black with very white spots, larger areas of white
kisutu c
  • same size as other types

  • large, robust

  • short

  • large

  • larger than chui uwanda

  • black and yellow spots

  • coloured like the African civet

  • yellow and black

  • with more white than konge

  • black stripes like a zebra, larger splotches/spots

  • yellow

  • yellow with largish black spots

  • large patches

  • large spots of every colour

  • small, long

  • smaller than kisutu

  • more yellow than konga, small spots

  • darker than kisutu, with no white

  • red with small white spots

  • rufous, like the colour of dry grass

a Fiercer than chui uwanda. b Fierce. c Fiercest type.

These descriptive names are analogous in some respects to the English words for the coat patterns of domestic cats and horses, and the terms used in different African languages to describe the colour combinations and hide patterns of cattle. They do not distinguish between male and female leopards or refer to distinct subpopulations that (concordant with the scientific concept of subspecies) live exclusively within certain geographical areas and breed within them. It is important to note, however, that relatively few Zanzibaris know and use these terms; we elicited them from (male) hunters and others with knowledge of leopards and other wildlife. Most informants distinguished between two or three types and different individuals sometimes described the same named type in contrasting ways. For example, kisutu is characterized variously as large, short or of average size. Konge, another type, has fur that is, according to one informant, black and white, whereas another person describes the pelage as black and red (rufous). Very few hunters produced the same classification or list of names, even when they hailed from the same village. Informants gave different type names for the small square of Zanzibar leopard pelt that one of our research assistants possessed (Fig. 3). The overall picture that emerges is one of considerable taxonomic and linguistic heterogeneity (Walsh, 1997, 2007; Walsh & Goldman, 2017). There seem to be three main reasons for this, as follows.

(1) The dialect geography of the island is notably complex. This is a result of the fact that Unguja has been subject to numerous external influences, including different episodes of immigration, as well as internal political shifts and local movements of population, in the course of its history. Although many features of local speech have been buried under the advance of the main Unguja dialect, a considerable amount of lexical variation has survived, notably in specialized terminologies such as that being considered here. Other aspects of ethnobiological classification in this region show a similar pattern of local fragmentation.

(2) A relatively small proportion of Unguja’s population have seen the Zanzibar leopard at close quarters, whether alive or dead in the form of carcases or skins. Rather fewer people have an active interest in differentiating one kind of leopard from another, the most obvious category of those who do being hunters. Younger hunters have had even less (if any) exposure to leopards than older generations, and there are likely no leopards left now for them to observe. In many areas, as wildlife is depleted, there are also fewer committed hunters than there were in the past, and cooperation, including sharing of knowledge, between hunters in different communities and at different levels of experience is not what it was.

(3) The lack of a shared descriptive discourse reflects in part people’s unwillingness to talk openly about leopards for fear of the consequences this may bring: demonstrating expertise regarding leopards may cause others to suspect one is a witch or may anger leopard-keeping witches (see the next section). This has generated a set of euphemisms for the leopard, such as bwana mkubwa (“big man”) and masharubu (“whiskers”). The distinction between euphemisms and descriptive names is not always clear, and some of the latter look as though they may have begun their life as the former.

At a broader level than these multifarious named types, rural Zanzibaris make a binary distinction between two kinds of leopard – wild ones and domesticated ones – to which we now turn.

Fig. 3

Fragment of a Zanzibar leopard pelt in the possession of the former Secretary of the Zanzibar National Hunters, photographed in 1996. (Photo by Helle V. Goldman.)

Wild and domestic leopards

Early in 1921, a leopard made its appearance in my district, and in a few nights accounted for eighteen goats. I went out and sat up all night for it in a tree over a kill, but it was of no use … [The Zanzibaris] said I might have saved myself the trouble, as an Mchawi had fuga-ed … the leopard and it would go where he told it.

(Ingrams, 1931, 471)

To understand this anecdote, related by a colonial officer in Zanzibar, it is necessary to know what mchawi and fuga mean. A mchawi is a witch. It is the understanding of many Zanzibaris that witches are everywhere and they cause all manner of misfortune. Even a person’s closest kin may be witches who bring terrible misfortune upon him (for more on understandings of witches in Zanzibar, see Goldman, 1996).

Witchcraft is ordinarily handled without directly confronting the suspected witch. Let’s say a child falls ill, an all too common occurrence in Zanzibar. The child is taken to the nearest rural clinic, or all the way to the hospital in the capital, but doesn't recover. (The clinics and hospitals are poorly equipped and staffed.) Back in the village, a diviner diagnoses that the child has been bewitched. More divination and rituals – from Koranic readings and prayers to elaborate events involving spirit possession – follow in an attempt to cope with the sickness. If the specialists figure it out, the identity of the witch who caused the sickness is typically kept under wraps or left vague. People speculate and have their theories, but the aim is rarely to expose the witch publicly and punish him or her, although there have been certain times in Zanzibari history when this has occurred, as we shall see. The general attitude is that witches are ubiquitous, quite possibly in one’s most intimate circle of family and other associates, and it behoves one to not antagonize them.

It is thought that some witches own leopards (Goldman & Walsh, 1997; Walsh & Goldman, 2007, 2012). A witch typically acquires a young leopard from another witch whose leopard has borne cubs. The witch stashes the cub in the forest, perhaps hidden in a hole or small cave in the rocky ground characteristic of Unguja, or keeps it secretly in his or her home, under the bed or in a room devoted to this purpose. The witch provides it with food and trains it to obey her or his commands. In due time, the witch puts the leopard to use. The witch may send the leopard to terrorize a person against whom the witch has a grudge, or to bring back food (often the meat of livestock) for the witch. The leopard can also serve as a guard for the witch's valuables. Using mystical methods, the witch dispatches the leopard here and there to carry out evil missions. Sometimes several witches, in different settlements, own a single leopard together. This has the advantage of making it easy to keep the leopard out of sight should there be trouble – for example, if government authorities are summoned to deal with a leopard that has been attacking people or livestock. The leopard is simply shifted to another owner some distance away until the brouhaha dies down.

Leopards that are kept in this way are left to look after themselves when their owners don’t need them. They are distinguished from purely wild leopards – chui (wa) mwitu(ni) (lit. “forest leopards”) – by being called chui wa kufugwa (lit. “tamed or kept leopards”). Kufugwa is the passive form of kufuga, the verb for the keeping of any domesticated animal. For example, kufuga ng’ombe means “to keep cattle.” When a person ritually cultivates relationships with spirits, this is kufuga mizimu, “to keep spirits.” To keep a leopard as a domestic animal, as described above, is kufuga chui. A cattle-keeper is mfugaji ng’ombe (“a keeper of cattle”); likewise, a leopard-keeper (i.e., a witch who keeps leopards) is mfugaji chui (for more detail on these and other terms used, see Goldman & Walsh, 1997; Walsh & Goldman, 2017).

Witches usually keep their leopards out of sight and the relationship between witches and leopards is neither publicly demonstrated nor openly acknowledged. People know about it by piecing together what they have heard from others and what they believe to have inadvertently or surreptitiously witnessed themselves. It is usually when people are in or near the bush that they see leopards. People go into the bush to collect firewood or to hunt wild animals, such as antelopes and bush pigs. They also hack out their fields from the bush in areas of shifting cultivation and sleep in small, roughly constructed shelters by their growing crops to protect them from animals. It may or may not be obvious whether a leopard that has been seen is a witch’s familiar or a wild animal. A leopard seen wearing clothing or jewellery definitely belongs to a witch. A leopard seen being carried on a person’s back, is also a kept leopard; a wild leopard could never be handled in this manner. When a leopard is seen within a settlement, there is little room for doubt about whether it belongs to a witch: the fact of its presence within an area of human activity is near-certain proof of its association with witches. It is thought that wild leopards are shy and generally avoid villages and people, whereas kept leopards are bolder in relation to humans. A leopard spotted slinking into, or out of, a house, typically through some concealed entryway, can only be a leopard visiting its owner. In these and other ways, kept leopards cross behavioural and physical boundaries that their wild counterparts are thought not to do.

There are sometimes no obvious outward signs that indicate whether a leopard that one has seen is wild or the property of a witch, but the after-effects of the sighting give it away. This is illustrated by the following incident that was told to us: One day a witch needed to send his leopard to a fellow witch – the leopard’s co-owner – in the next village. The man instructed his young granddaughter to walk to that other village on the pretext of an ordinary errand, giving her a small charm to carry in her pocket. As she walked along the path, the witch's leopard followed alongside her, out of sight in the bush, on the same side as the pocket with the magical charm in it. As the girl walked along, she shifted the charm to the other pocket, having no idea that she was being shadowed by a leopard that was following the charm. The leopard then crossed the path in front of her and the girl saw it and was afraid. She continued on her way, with the leopard walking parallel to her in the bush on the other side of the path. When she arrived at the village she told people that she had seen a leopard. She soon fell ill. This was her mystical punishment for having spoken about the leopard. This is one of the main ways that one knows one has seen a witch's leopard: sickness ensues. One of the symptoms is vomiting up leopard hairs. This doesn't happen following an encounter with a wild leopard.

If contact with a witch's leopard has made a person ill, a specialist may recommend the performance of a ritual involving the recitation of a curse. The subsequent death of someone in the village or in a nearby settlement is a strong indication of the guilt of a local witch. If no one dies nearby, it is understood that the witch must live further away. What happened to one man, when he was a little boy, illustrates how this works: One night when he was perhaps five or six years of age, he was dragged out of the small hut in which he was sleeping by a leopard. The leopard released him and bounded off when people raised a hue and cry. The child’s wounds on his head and his leg – the scars were still visible when we interviewed him as an old man – were treated at the hospital and they healed. But the boy began vomiting up leopard hairs. A specialist was consulted and a curse recited. In another village, some distance away, the boy’s biological father (not his social father) died. It was understood then that this was the witch who had sent the leopard. The rationale was that the man was bitter about having no rights to the boy and his revenge was to terrorize the child with his leopard. (For more incidents that show how leopard attacks are explained in relation to various kinds of human conflicts, see Goldman & Walsh, 1997; Walsh & Goldman, 2007; Walsh, 2016.)

Hunters are said to sometimes shoot leopards accidentally, particularly at night, mistaking the dark silhouette of the creature, or its eyeshine, for an antelope or other animal. Fearing that the leopard belongs to a witch, the hunter buries the carcase, mentions the incident to no one, and hopes he doesn’t start regurgitating leopard hairs.

From vermin to protected species, and back and forth again

[Leopards] do a great service to the farmer by living mainly on the pig, monkeys, and duiker that raid his fields. About the middle of the year, however, a leopard which turned man-eater and killed a woman and three children in the course of two months spread terror over a large area of country on the east coast of Zanzibar Island and seriously interfered with local food production because peasants feared to sleep in their fields, as is customary, to keep away marauding pig, with the result that much of the crops was lost.

(Zanzibar Protectorate 1949, p. 24)

The leopard to which R. H. W. Pakenham, a high-ranking official in the administration of the Zanzibar Protectorate, referred, was trapped and killed. Or a leopard was killed: the fatal leopard attack on another child three months later undermined the administration’s theory that the responsible man-eating leopard, imagined in the mode of other “man-eaters” of colonial African and Indian jungle lore, had been dealt with. (For details on the killings of 1948, and their contrasting interpretations by Zanzibaris and British administrators, see Walsh, 2016.)

Fig. 4

Undated colonial-era photograph of the fresh carcase of a Zanzibar leopard. (Photo in the possession of Martin T. Walsh, kindly donated by Jono Jackson.)

Fig. 5

Undated colonial-era photograph of a leopard trap of local design and construction, Zanzibar. (Photo by Helle V. Goldman of image on display at Zanzibar Museum.)

The government’s views on leopards were mixed. On the one hand, some colonial officials wanted to prevent people (and livestock) being harmed by leopards or leopard panics to interfere with subsistence activities. In this view, leopards were socially troublesome and economically deleterious animals – vermin – that should be exterminated (Figs. 4, 5). On the other hand, Pakenham maintained that leopards should not be treated as vermin because they benefited farmers by preying on bush pigs, antelopes and other wild animals that ate crops. Pakenham’s high rank ensured that his assessment prevailed, and in 1950 the Zanzibar leopard was put on a list of protected species whose killing required a special permit from the Senior Commissioner.

Leopards were already under a degree of protection on account of people’s fear of the witches who supposedly owned them. Nonetheless, in the late 1950s and early 1960s, there were several locally organized efforts directed against leopard-keepers in particular communities in the countryside. These involved public oathing ceremonies and other collective ritual measures. Whatever the local effects of these initiatives were, they paled in comparison to the scale of a campaign spear-headed by a man named Kitanzi. This took off in the late 1960s with the support of the independent nation’s first president, the despotic Abeid Karume, who was increasingly concerned about witchcraft in the rural areas. In the absence of any kind of documentation, we know few specifics about this campaign – for example, the methods Kitanzi and his team used to expose leopard-keepers and other alleged witches, how many people were arrested and detained, or the number of leopards that were killed. This was a murky period in Zanzibar’s history, and it is easy to imagine how such a crusade might have been used as an instrument of harassment.

It is generally understood that Kitanzi and his associates employed various magical means to help them capture witches and leopards. Leopard parts, such as part of the throat and the claws, were used in charms to protect leopard hunters, and hunting dogs were fed leopard meat to make them fiercer. One striking innovation was that Kitanzi’s hunters cooked and ate the meat of the leopards they killed and even distributed cooked leopard meat among children for their consumption. Eating leopard meat had hitherto been taboo, not least because Islam prohibits eating animals with claws and fangs. Such flesh is harami (“prohibited”), in contrast to meat that is halali (“lawful”). Nefarious practices and transgressive creatures, it seems, justified equally transgressive measures in response. They also served to bind Kitanzi and his fellow hunters into a tight band with their own rituals and secrets.

By the early 1970s, the Kitanzi campaign had fizzled out. Fortified with anti-leopard magic, an organization of government-supported hunters called the National Hunters (Wasasi wa Kitaifa), took over part of the duties of Kitanzi and killed leopards during their hunts for the next two decades. With the motto Shime tuwapige vita wanyama waharibifu (“Come, let’s wage war on vermin”), the National Hunters’ raison d'être was to reduce vermin (wanyama waharibifu) in the countryside. According to the National Hunt records that we saw, nearly all wild mammals were considered to be vermin. An exception was the Zanzibar red colobus, a high-profile endemic species that attracted tourism and conservation dollars. If the somewhat spotty records of the National Hunters are to be believed, this group killed between 92 and 115 leopards between 1985 and 1995 (Goldman & Walsh, 2002).

The results of our initial research became known in 1996, rousing the interest of both the government and general populace. Our report (Goldman & Walsh, 1997) detailed numerous recent sightings of leopards and their tracks, as well as other signs of leopards, particularly in the eastern and southern parts of the island. We could not confirm the veracity of these sightings, nor the accuracy of National Hunt reports, but there seemed to be a chance that some leopards had survived on Unguja. Whether the Zanzibar government would continue to treat them as vermin was another question. As it happened, our investigation coincided with a sea change in official attitudes toward wildlife. This was probably not unconnected to the burgeoning tourism industry, which came late to Zanzibar in comparison to mainland Tanzania and neighbouring Kenya. Legislation passed in 1996 laid the groundwork for the declaration of Zanzibar’s first national park (Zanzibar Government, 1997a), and many wild animals were listed as protected species that could not be “killed, injured, destroyed, captured, collected or otherwise taken” (Zanzibar Government, 1997b).

The government’s new positive attitude toward leopards was not necessarily echoed in the countryside, where fear of leopards was (and remains) widespread. Yet even in rural areas there were indications that people saw a benefit in leopards and in the 1990s we heard of villagers’ plans to display leopards to paying tourists. Though these came to nothing, the government was beginning to see potential in this general idea too. In the House of Representatives, in 2003, the Deputy Minister for Agriculture, Natural Resources, Environment and Cooperatives declared that if Zanzibar “had these animals [leopards] they would increase government income, and he asked for citizens to sell them to the state” (Ame, 2003; our translation). He was reported in the newspaper to have continued, “There’s a fear that if anyone appears with a leopard, then he’ll feel the noose around his neck. Get rid of that fear: my ministry is ready to buy leopards” (Ame, 2003; our translation). To date, no one has taken up this offer, and though compelling stories of new leopard sightings keep coming to us and to Forestry Department officials, there has been no incontrovertible evidence for the survival of leopards on Unguja for decades.

The transgressive leopard

Animals are considered problematic when they resist neat categorization on account of their appearance or behaviour or where they exist (Douglas, 1966; Hurn, 2012). We have seen how the Zanzibar leopard has been classified by different groups of people through recent history as a domestic animal, a wild animal, vermin, a protected animal, edible, non-edible, and a subspecies in the scientific sense (though, through its omission from recent genetic studies, this status has effectively been revoked). It has also been subdivided ethnotaxonomically into numerous types. Some leopards have been labelled “man-eaters.” More recently, they have been seen as potential money-makers.

Animals are also problematic when people cannot control them (Hurn, 2012). This, along with the perceived benefits or harm they are thought to confer upon humans, is what makes vermin bad and domesticated animals good (Arluke & Sanders, 1996). The Zanzibar leopard is an interesting case because it has been perceived to be vermin and at the same time to be a domesticate. The animal’s supposed domestication is not seen as a redeeming quality, for reasons we have explained here and in more detail elsewhere (e.g., Walsh & Goldman, 2007).

Zanzibaris consider some leopards to be wholly wild. They are dangerous and frightening, but, like most other wild animals, they avoid encounters with people. Other leopards – which may look just like wild leopards – are under the control of witches. These leopards are transgressive, out-of-place, animals – attacking humans, appearing in or near settlements, responding to mystically transmitted human commands and sometimes even wearing human adornments. Similar ideas linking witchcraft to physically dangerous or culturally ambiguous animals are widespread in Africa (e.g., Lindskog, 1954) and were historically common in our own society. Some of the actions attributed to leopards in Zanzibar can be chalked up to their natural behaviours, especially in the context of the dramatic reduction of Unguja’s wild habitat as the human population has grown over the last 100 years from some 114,000 to a figure on the order of one million. As human expansion shrinks and fragments wilderness areas in Africa and elsewhere, clashes between large carnivores and people proliferate. In these situations, predators are regarded as troublemakers that behave anomalously or trespass into spaces that they should stay out of (Hurn, 2012, 79).

The Zanzibar leopard illustrates the social construction of animals – that is, how people classify animals, how they project human qualities onto them, how animals’ bodies and actions are imbued with symbolic meaning, and how animals unwittingly play roles in the construction of human identities as well as in social conflicts at small and large scales. In colonial and post-colonial Zanzibar, leopards have helped explain misfortune in the context of local inter-personal conflicts. They have also been drawn into larger political and religious discourses. In all these respects, they have been domesticated in the sense of being incorporated into human affairs, though with unfortunate consequences.

Such ramifications for leopards and other animals in Zanzibar and elsewhere raise the issue of advocacy. Many anthropologists have found it morally incumbent to speak on behalf of, and work to improve the welfare of, the people they have studied. In step with the current “animal turn” in anthropology and other disciplines, some are now asking why this compunction should not cross the species divide (Hurn, 2012). However, when an animal is extinct, it makes no sense to advocate on its behalf, except to use it as a warning and negative example to support the conservation of other species and habitats. This is now the case with the Zanzibar leopard (and our own writing about it), although earlier in our research, when there were indications that there were still leopards to be found on Unguja, we advocated for its conservation.

The Zanzibar leopard’s perceived transgression has been explained in terms of another social construction: the transgressive behaviour of alleged witches. This contributed to determined efforts not just to exterminate leopards, but also, during some periods, to rid Zanzibar of their alleged keepers. In these ways, the fate of the Zanzibar leopard, like that of other animals in Africa that harm people or are perceived to do so, has at times rebounded onto humans themselves.

The Neolithic domestication of plants and animals has traditionally been seen as a watershed development in human history in which people conquered the wilderness and took a decisive step forward in their advance toward civilization, a view that has been subjected to considerable post-modern critique (see, e.g., Russell, 2007; Cassidy, 2007; Hurn, 2012). For analytical purposes, human societies are still put into categories – such as hunter-gatherers, pastoralists and agriculturalists – that are largely defined in terms of characteristic relationship between people and animals. In considering how the concept of domestication has been applied to animals in anthropology, Cassidy argues that domestication is, like culture, a “slippery and imprecise” concept (2007, p. 3). Domestication potentially encompasses extremely diverse human-animal relationships that challenge a unified definition of the concept.

The relationship between people and leopards in Zanzibar is one such challenging case, exemplifying an “unruly relationship” (Cassidy, 2007, p. 18) of domestication. For both Cassidy (2007) and Russell (2007), the difficulty in the concept of domestication as it is often used lies partly in the problematic nature of the “wild” versus “domestic” dichotomy and other simplistic classifications. Russell draws fruitful comparisons and interconnections between kinship and domestication – both of which involve classification, a fundamental and longstanding concern of anthropology (e.g., Needham, 1963) – and proposes that “all domestic animals complicate the boundaries between humans and animals, nature and culture” (Russell, 2007, p. 35).

The case of the Zanzibar leopard is challenging for a number of reasons that add to the complications of domestication already discussed in the literature. When Zanzibaris describe leopard-keeping, they are not referring to domestication as it is generally understood and argued over by natural and social scientists, though superficially this may seem to be so. Rather, they are referring to a set of practices that are embedded in local understandings of a world in which leopards can be controlled by spiritual forces in the way that other beings, including spirits themselves, can also be manipulated by those in possession of the requisite skills and means. While there is no place in zoological science for imaginary or pseudo-domestication of this kind, it has played a determining role in the relationship between people and leopards in Zanzibar, and has contributed significantly to their extermination. While many Zanzibaris believe the Zanzibar leopard to be extant, most zoologists consider that it is now extinct, and that recent claimed sightings are as imaginary as its domestication – as imaginary, indeed, as Borges’s Chinese encyclopedia.


Ame, M. (2003). Serikali iko tayari kununua chui. Government is ready to buy leopards. Zanzibar Leo, April 13.

Arluke, A. & Sanders C. R. (1996). Regarding animals. Philadelphia: Temple University Press.

Borges, J. L. (1964). Other inquisitions, 1937–1952. Austin: University of Texas Press.

CCTF (Cat Classification Task Force) (2017). A revised taxonomy of the felidae. Cat News, Special Issue, 11. Cat Specialist Group, Species Survival Commission, International Union for Conservation of Nature.

DeMello, M. (2012). Animals and human society: an introduction to human–animal studies. New York: Columbia University Press.

Descola, P. (2013). Beyond nature and culture. Chicago: University of Chicago Press.

Douglas, M. (1957). Animals in Lele religious symbolism. Africa 27(1), 46–58. https://doi.org/10.2307/1156365

Douglas, M. (1966). Purity and danger: an analysis of concepts of pollution and taboo. London: Routledge.

Foucault, M. (2005). The order of things: an archaeology of the human sciences. London: Routledge. (First published in English in 1970.)

Goldman, H. V. (1996). A comparative study of Swahili in two rural communities in Pemba, Zanzibar, Tanzania. Ph.D. dissertation, New York University.

Goldman, H. V. & Walsh, M. T. (1997). A leopard in jeopardy: an anthropological survey of perceptions and practices threatening the survival of the Zanzibar leopard (Panthera pardus adersi). Zanzibar Forestry Technical Paper 63. Zanzibar: Commission for Natural Resources.

Goldman, H. V. & Walsh, M. T. (2002). Is the Zanzibar leopard (Panthera pardus adersi) extinct? Journal of East African Natural History 91, 15–25. https://doi.org/10.2982/0012-8317(2002)91[15:itzlpp]2.0.co;2

Heine, B. & Legère, K. (1995). Swahili plants: an ethnobotanical survey. Köln: Rüdiger Köppe Verlag.

Hurn, S. (2012). Humans and other animals: cross-cultural perspectives on human–animal interactions. London: Pluto Press.

Ingrams, W. H. (1931). Zanzibar: its history and its people. London: Frank Cass and Co.

IUCN (International Union for Conservation of Nature) (2019). The IUCN Red List of Threatened Species. Accessed at https://www.iucnredlist.org/ on 1 February.

Leach, E. (2015). Anthropological aspects of language: animal categories and verbal abuse. Anthrozoös 2(3), 151–165. https://doi.org/10.1037/e685262012-044

Lévi-Strauss, C. (1966). The savage mind. London: Weidenfeld & Nicolson.

Lindskog B. (1954). African leopard men. Uppsala: Almquist & Wiksells.

Miththapala, S., Seidensticker, J. & O'Brien, S. J. (1996). Phylogeographic subspecies recognition in leopards (Panthera pardus): molecular genetic variation. Conservation Biology 10, 1115–1132. https://doi.org/10.1046/j.1523-1739.1996.10041115.x

Needham, R. (1963). Introduction. In R. Needham (Ed.): Primitive classification (pp. vii–xlviii). Chicago: University of Chicago Press.

Nowell, K. & Jackson, P. (Eds.) (1996). Wild cats. Status survey and conservation action plan. Gland, Switzerland: Cat Specialist Group, Species Survival Commission, IUCN.

Pocock, R. I. (1932). The leopards of Africa. Proceedings of the Zoological Society of London II, 543–591. https://doi.org/10.1111/j.1096-3642.1932.tb01085.x

Sahlins, M. (1995). How “natives” think: about Captain Cook, for example. Chicago: University of Chicago Press.

Stuart, C. & Stuart, T. (1997). A preliminary faunal survey of south-eastern Unguja (Zanzibar) with special emphasis on the leopard Panthera pardus adersi. Loxton, South Africa: African–Arabian Wildlife Reserve Centre.

Uphyrkina, O., Johnson, W. E., Quigley, H., Miquelle, D., Marker, L., Bush, M. & O’Brien, S. J. (2001). Phylogenetics, genome diversity and origin of modern leopard, Panthera pardus. Molecular Ecology 10, 2617–2633. https://doi.org/10.1046/j.0962-1083.2001.01350.x

Walsh, M. T. (1997). Appendix C: leopard classification and linguistics. In H. V. Goldman & M. T. Walsh: A leopard in jeopardy: an anthropological survey of practices and beliefs which threaten the survival of the Zanzibar leopard (Panthera pardus adersi). Forestry Technical Paper no. 63 (pp. 37– 54). Zanzibar: Commission for Natural Resources.

Walsh, M. T. (2000). Why does the Zanzibar leopard have so many names? Lexical variation and its interpretation on Unguja island. Presentation at the international colloquium on Kiswahili in 2000, Institute of Kiswahili Research, University of Dar es Salaam, 20–23 March.

Walsh, M. T. (2016). The man-eater of Uroa and other collective fictions in Zanzibar. Presentation at Working on/with Archives and the Written Word in Anthropology and Literary Studies: Perspectives on the Swahili World, IX European Swahili Workshop, École des Hautes Études en Sciences Sociales, Paris, 18–19 April.

Walsh, M. T. & Goldman, H. V. (2007). Killing the king: the demonization and extermination of the Zanzibar leopard. In E. Dounias (Ed.): Le symbolisme des animaux: l’animal, clef de voûte de la relation entre l’homme et la nature? (Animal symbolism: animals, keystone in the relationship between man and nature?) (pp. 1133–1182). Paris: IRD Éditions.

Walsh, M. T. & Goldman, H. V. (2008). Updating the inventory of Zanzibar leopard specimens. Cat News 49, 4–6.

Walsh, M. T. & Goldman, H. V. (2012). Chasing imaginary leopards: science, witchcraft and the politics of conservation in Zanzibar. Journal of Eastern African Studies 6, 727–746. https://doi.org/10.1080/17531055.2012.729778

Walsh, M. T. & Goldman, H. V. (2017). Cryptids and credulity: the Zanzibar leopard and other imaginary beings. In S. Hurn (Ed.): Anthropology and cryptozoology: exploring encounters with mysterious creatures (pp. 54–90). Milton Park: Routledge.

Willis, R. (1974). Man and beast. London: Hart-Davis, MacGibbon.

Willis, R. (Ed.) (1990). Signifying animals: human meaning in the natural world. London: Routledge.

Windschuttle, K. (1997). Absolutely relative. National Review, 49(17), 28–30.

Zanzibar Government 1997a. The Environmental Management for Sustainable Development Act, 1996. Legal Supplement (Part I) to the Zanzibar Government Gazette. Vol. no. CVI No. 5743 of 31st May, 1997. Zanzibar: Government Printer.

Zanzibar Government 1997b. Forest Resources Management and Conservation Act. no. 10 of 1996. Legal Supplement (Part I) to the Zanzibar Government Gazette. Vol. no. 5769 of 6th December, 1997. Zanzibar: Government Printer.

Zanzibar Protectorate (1949). Annual Report of the Provincial Administration for the Year 1948. Zanzibar: Government Printer.

Idunn bruker informasjonskapsler (cookies). Ved å fortsette å bruke nettsiden godtar du dette. Klikk her for mer informasjon