Nimeona kisutu, mwenye kisutu sijamwona

Friday, 8 June 2018


Our paper on 'Chasing imaginary leopards', originally published in 2012, was reprinted in 2017 in Iain Walker's edited volume on Contemporary Issues in Swahili Ethnography. Here's the title and abstract:

Chasing imaginary leopards: science, witchcraft and the politics of conservation in Zanzibar

Martin Walsh and Helle Goldman

Contemporary Issues in Swahili Ethnography (Hardback) book coverThe Zanzibar leopard (Panthera pardus adersi) is (was) a little-known subspecies endemic to Unguja island. Rapid population growth and the expansion of farming in the twentieth century destroyed leopard habitat and decimated their natural prey, bringing them into increasing conflict with people. Villagers explained the growing number of attacks on their children and livestock by supposing that the leopards responsible for them were owned by witches and sent by them to do harm. Following the Zanzibar Revolution in 1964, localised efforts to act on this theory culminated in an island-wide leopard eradication and witchfinding campaign, supported by the government. By the 1990s state-subsidised hunting had brought the leopard to the brink of extinction, and most zoologists now presume it to be extinct. However, many islanders believe that leopard keepers are still active in rural Unguja and sightings of leopards continue to be reported. Beguiled by such narratives, visiting researchers and local conservationists have continued to pursue these elusive felids. In this paper we describe and analyse a series of unsuccessful ‘‘kept leopard chases’’, including abortive calls by government officials for the capture and display of domesticated leopards. These quixotic efforts show no signs of abating, and the underlying conflicts of knowledge and practice remain unresolved, posing a challenge to the theory and practice of conservation not only in Zanzibar but also further afield.

The original journal paper can be viewed and downloaded here:


In November 2016 our long-awaited chapter on 'Cryptids and credulity' appeared in Samantha Hurn's edited volume Anthropology and Cryptozoology: Exploring Encounters with Mysterious Creatures. As publishers often do these days, the book was forward dated to 2017, confusing us all.

Here's an extract from the introduction to our chapter:

Cryptids and credulity: the Zanzibar leopard and other imaginary beings

Martin T. Walsh and Helle V. Goldman

There is nothing intrinsically unscientific about searching for new (undescribed) and old (extinct) species, and one cryptozoologist has defined the discipline concisely as ‘a targeted-search methodology for zoological discovery’, noting that it is only one of a number of possible means to achieving this end (Arment 2004: 9). But all too often cryptozoologists’ desire to find hidden species and identify the imaginary as real leads them to downplay the negative evidence that carries more weight with conventional zoologists and ethnozoologists (Simpson 1984: 12–14; Meurger 1988: 11–24). Although there is also a strong tradition of debunking fakes and false claims within cryptozoology, it has failed to establish itself as an academic discipline (Coleman 2002: xxxiii), while the professional association founded by Heuvelmans and colleagues – the International Society of Cryptozoology – has long since been defunct and its journal (Cryptozoology) extinct (Eberhart 2002: xxvii).

Our epigraph highlights a boundary problem that cryptozoologists have also struggled with: what kinds of phenomena or imaginary being fall within their remit? The subjects of cryptozoology are generally now referred to as cryptids, on one definition ‘the alleged animals that cryptozoologists study’ (Eberhart 2002: xxiii, also xlvii). Some restrict this to non-microscopic creatures they consider most likely to be discovered to be living species; others include historical and contemporary entities that are more obviously mythical (compare Greenwell 1985; Arment 2004: 11–12, 16–18). In his encyclopaedia Eberhart takes the broader view, and lists ten categories that most of the ‘mystery animals’ in his compilation fall into:

1. Distribution anomalies ...
2. Undescribed, unusual, or outsize variations of known species ...
3. Survivals of recently extinct species ...
4. Survivals of species known only from the fossil record into modern times ...
5. Survivals of species known only from the fossil record into historical times ...
6. Animals not known from the fossil record but related to known species ...
7. Animals not known from the fossil record or bearing a clear relationship to known species ...
8. Mythical animals with a zoological basis ...
9. Seemingly paranormal or supernatural entities with some animal-like characteristics ...
10. Known hoaxes or probable misidentifications ... (2002: xxiii–xxiv)

The subject of our paper is the analysis of narratives and statements about an animal, the Zanzibar leopard (Panthera pardus adersi), which might be included in Eberhart’s third category, given that it has been declared by some authorities to be extinct, though many Zanzibaris remain convinced of its continued existence. The classic example of a carnivore in this category is the Thylacine or Tasmanian tiger (Thylacinus cynocephalus), and in a later section we will explore some of the differences and significant parallels between the two cases. Unlike its marsupial analogue, the demise of the Zanzibar leopard has been so recent that zoologists cannot be sure that no individuals survive. In this respect its cryptozoological (and ontological) status remains uncertain.

However, as an anthropological analysis of indigenous and other narratives about the leopard shows, it can be described under more than one of Eberhart’s headings: as the undescribed (and unusual) variation of a known species, as a mythical animal with a zoological basis, and perhaps even as a supernatural entity with animal-like characteristics. Whereas popular cryptozoology typically reduces the investigation of imaginary beings to a single dimension (‘do they exist or not?’), we argue that only careful anthropological and ethnozoological research can unravel the complexity of cases like that of the Zanzibar leopard and other so-called cryptids. And, as this introduction has implied, cryptozoology itself can also be analysed anthropologically, a subject we will return to briefly in our conclusion.

A version of the chapter can be viewed online here:


We've not posted for some time, so are publishing a series of catch-up posts to get the blog going again. Here's the first, an extract from the beginning of a paper that Martin presented to the European Swahili Workshop (No. 9) in Paris, on 18-19 April 2016. The theme of the meeting was “Working on/with archives and the written word in anthropology and literary studies: Perspectives on the Swahili world”:

The man-eater of Uroa and other collective fictions in Zanzibar

Martin Walsh, University of Cambridge

In the early hours of 30th June 1948 a small boy was seized and dragged from a field hut near Uroa on the east coast of Unguja island, Zanzibar, and was never seen again. Pugmarks were found nearby, and it was presumed that he had been taken and devoured by a leopard. Over the next seven weeks three more people were killed in similar circumstances in the same general area: a young girl, a middle-aged woman, and another boy. These events generated considerable alarm not only in the local communities involved, but also among officials in the British colonial administration, who ascribed the killings to a “man-eating leopard”, imagined in the mode of other “man-eaters” of African and Indian jungle lore. In an effort to
prevent further deaths and contain the panic, the authorities went to some lengths to trap and kill the supposed man-eater, and its demise was announced in the second week of September 1948.

We know these and other details from the contemporary documents that have survived. The archival record includes official correspondence – some of which reached up to the Secretary of State for the Colonies in London, district and other administrative reports, newspaper articles, and the personal notebook of Zanzibar‟s Senior Commissioner at the time, R.H.W. Pakenham. But though these written archives give us some insight into the knowledge and beliefs of officialdom, including Zanzibari Arab civil servants and local administrators, the voices of the villagers who experienced these events first hand are largely absent, though there are hints that they might have had something very different to say, not least their own interpretation of events. The few paragraphs that Pakenham wrote in his notebook make this very clear, though they were not intended for public consumption and did not enter the official record at the time.

Most of the people who witnessed or were close to the events of 1948 are now dead. In November 2011, having read the archives in Zanzibar some months before, I began to seek out and interview elderly people on the east coast who could recall something of what had happened more than 63 years earlier. Their accounts were very different from those preserved in the government archives, and not just richer and thick with local colour. Rather than a single man-eater, they described the depredations of multiple “kept leopards”, sent to do the bidding of the various witches who owned and controlled them. This reflected an understanding of leopard predation that persists to this day and that Helle Goldman and I have described in some detail in earlier publications (Goldman and Walsh 1997; Walsh and Goldman 2007; 2012). This paper is my first attempt to process material relating to the 1948 case, and to explore answers to some of the questions that it raises. Who and what should we believe? Can we reconcile the conflicting interpretations of Zanzibar‟s one-time rulers and the oral performances of their former subjects, discern unequivocal truth behind the competing fictions of colonial writing and postcolonial memory? What really happened?

The full paper is online here:

Thursday, 25 February 2016


Here's the abstract of 'Continental Island Formation and the Archaeology of Defaunation on Zanzibar, Eastern Africa' by Mary E. Prendergast, Hélène Rouby, Paramita Punnwong, Robert Marchant, Alison Crowther, Nikos Kourampas, Ceri Shipton, Martin Walsh, Kurt Lambeck, and Nicole L. Boivin, 'Continental Island Formation and the Archaeology of Defaunation on Zanzibar, Eastern Africa'. It's published by PLOS ONE and can be viewed online or downloaded via the links embedded in this paragraph.

With rising sea levels at the end of the Pleistocene, land-bridge or continental islands were formed around the world. Many of these islands have been extensively studied from a biogeographical perspective, particularly in terms of impacts of island creation on terrestrial vertebrates. However, a majority of studies rely on contemporary faunal distributions rather than fossil data. Here, we present archaeological findings from the island of Zanzibar (also known as Unguja) off the eastern African coast, to provide a temporal perspective on island biogeography. The site of Kuumbi Cave, excavated by multiple teams since 2005, has revealed the longest cultural and faunal record for any eastern African island. This record extends to the Late Pleistocene, when Zanzibar was part of the mainland, and attests to the extirpation of large mainland mammals in the millennia after the island became separated. We draw on modeling and sedimentary data to examine the process by which Zanzibar was most recently separated from the mainland, providing the first systematic insights into the nature and chronology of this process. We subsequently investigate the cultural and faunal record from Kuumbi Cave, which provides at least five key temporal windows into human activities and faunal presence: two at the end of the Last Glacial Maximum (LGM), one during the period of post-LGM rapid sea level rise and island formation, and two in the late Holocene (Middle Iron Age and Late Iron Age). This record demonstrates the presence of large mammals during the period of island formation, and their severe reduction or disappearance in the Kuumbi Cave sequence by the late Holocene. While various limitations, including discontinuity in the sequence, problematize attempts to clearly attribute defaunation to anthropogenic or island biogeographic processes, Kuumbi Cave offers an unprecedented opportunity to examine post-Pleistocene island formation and its long-term consequences for human and animal communities.

Saturday, 7 November 2015


Here's the abstract of our presentation to Baraza: Swahili Conference at SOAS, in the University of London on Saturday 31 October 2015:

Swahili scams: felids and fakes in Zanzibar

Martin Walsh, University of Cambridge
Helle Goldman, Norwegian Polar Institute

In addition to its usual cultural referents, the label ‘Swahili’ has become a byword for sharp practice (cf. ‘mswahili ... a crafty, tricky person’, TUKI 2001: 222). It is not unusual for economically successful groups to be stereotyped in this way, and Zanzibaris are apt to think similarly about the mainlanders of different ethnic origins who come to do business on their islands. At the same time it is not difficult to find real cases of deception that appear to confirm such stereotypes. In this presentation we will examine contemporary confidence tricks involving felids, focusing on cases that have apparently resulted in their illegal translocation from mainland Tanzania to Unguja (Zanzibar) island. These scams are made possible by and take advantage of a particular set of cultural ideas about the domestication of leopards and their use for nefarious purposes, ideas that we have already examined in detail elsewhere (e.g. Goldman and Walsh 1997; Walsh and Goldman 2012). These imagined practices are described by rural Zanzibaris as ‘Swahili’ in the traditional, cultural sense (Walsh and Goldman 2007: 1142), and we will consider the extent to which this belief in the reality of leopard-keeping is in turn founded upon and supported by intentional acts of deceit. This argument is premised on a position of scientific realism, and we will conclude by discussing its implications for relativistic approaches to understanding local knowledge and practice, as well as its application to the study of swindling and scams in the wider Swahili-speaking world and beyond.

Our presentation is online here:

Friday, 30 November 2012


Our paper 'Chasing imaginary leopards: science, witchcraft and the politics of conservation' has finally been published in the Journal of Eastern African Studies, along with other papers in a special collection on 'Contemporary issues in Swahili ethnography'. It's more than two years since we presented a first version of the paper at the VIII European Swahili Workshop in Oxford (see our earlier post on 'Chasing imaginary leopards', which gives the abstract that we submitted). The workshop was organised by Iain Walker, and we're very grateful to him for all the effort he put into this and then finding a home for revised versions of the papers, ours included.

Saturday, 7 May 2011


Over the Easter holiday we finished drafting a paper about the Zanzibar leopard for an edited volume provisionally entitled Animals Out of Place: Cryptozoology in Anthropological Perspective. This book gathers together case studies from around the world and is being edited by Samantha Hurn, who lectures in the School of Archaeology, History and Anthropology at the University of Wales Trinity Saint David. There are plans to present some of its chapters in a panel on 'Cryptozoology: animals out of place or time' at ASA11: Vital powers and politics: human interactions with living things, the annual conference of the Association of Social Anthropologists of the UK and Commonweath, which will be held at Lampeter in September. Here's an extract from our draft chapter, 'Cryptids and credulity: the Zanzibar leopard and other imaginary beings', which still has to be reviewed by colleagues. It follows sections on 'The Zanzibar leopard in scientific discourse', 'Ethnotaxonomies of the Zanzibar leopard', and 'Witchcraft and leopard-keeping narratives':


Unguja island
When we undertook our consultancy in July 1996 there appeared to be good evidence for the continuing presence of the Zanzibar leopard in the south and east of Unguja island. National Hunters’ reports indicated that leopards were still being killed (along with other ‘vermin’): the last kill on record was from a hunt in Jambiani on 17-18 April 1995 (Goldman and Walsh 1997: 31-36). We were also told of more recent kills by independent groups of hunters. The former secretary of the National Hunters, who worked as our research assistant, told us that ‘Omani Arab’ hunters from Mlandege in Zanzibar town were rumoured to have killed a leopard at Mtule, between Kitogani and Paje, in March 1996 (Goldman and Walsh 1997: 29). An interviewee in Dimani told us that three leopard cubs had been killed by young hunters in that area on or around 21 April 1996 (Goldman and Walsh 1997: 26). These and reports of earlier kills were supplemented by descriptions of recent leopard sightings and other evidence for their presence, including seven reported sightings in 1996 (Goldman and Walsh 1997: 3, 24-36; 2002: 19-22).

At the time we had no reason to disbelieve these claims, though some experienced hunters said that they had not seen evidence of leopards for a number of years. We knew that most kills by independent and local groups of hunters would be hidden from official view and would not be included in National Hunters’ statistics. And their former secretary – who showed us his own list of kills – told us that the National Hunters did not submit records of all of their leopard kills because they were concerned that the hunting of leopards might be stopped by the government. As it happens, when we began our joint research moves were afoot to do just this. The recently passed Forest Resources Management and Conservation Act of 1996 provided for the preparation of lists of protected wild animals and plants, and, in the interim, the use of lists that had been on the statutes since the colonial period but ignored since the Zanzibar Revolution (Sections 76-79, Part VIII, “Conservation of Wild Animals and Wild Plants”, in The Revolutionary Government of Zanzibar 1996: 40-42). The Zanzibar leopard was on the old schedule of protected wild animals (Walsh and Goldman 2007: 1140-1141, 1150), and was about to be included in the new one. As a result it had become illegal (again) to kill, injure, destroy, capture or collect leopards without a special permit, unless this was done “to defend against an attack or imminent threat of attack on human life” (Section 77 in The Revolutionary Government of Zanzibar 1996: 40).

This new legislation seems to have had an immediate effect on the practice of the National Hunters. From January 1996 onwards the vermin hunting statistics based on their reports no longer included leopard kills.[13] Since July 1996 we have not been told of any leopard kills by National Hunters. Nor have we heard of any certainly killed by local groups, individual hunters, or others. In August 1999 the then Head of Conservation in Zanzibar told Goldman that he had heard that a dead leopard had been found near Kinyasini about a year earlier (i.e. sometime in mid-late 1998), but he knew no more details, including whether this leopard had been killed or not. We do not know whether the lack of information about leopard kills in the past decade and a half reflects reluctance to talk about an activity that has been (re)defined as illegal, or the fact that very few and perhaps no leopards at all have been killed since 1996. It may be that there are very few or no leopards left to kill on Unguja. Otherwise we wonder whether we would have picked up on more stories of kills had we spent longer in the field or interviewed more hunters during subsequent research trips and other visits to Zanzibar.

At the same time, reports of leopard sightings and other evidence for their presence continue to reach us. Every time we visit Zanzibar and meet with former colleagues working in the Department of Commercial Crops, Fruits and Forestry (DCCFF) we are titillated with tales of sightings and other incidents, some of them involving kept leopards and their alleged keepers, others more prosaic, and so to us more believable. During his latest trip in October 2007, Walsh was told at the headquarters of Jozani-Chwaka Bay National Park that about three months earlier a ranger on night patrol had reported seeing a tree shaking and then a half-eaten Blue monkey (Cercopithecus mitis) – presumed to be a leopard’s meal – lodged in its branches. The details of this, it was said, had been recorded by the Chief Park Warden (who was one of our research assistants in 1996), and in a subsequent interview the latter recounted that the ranger in question was a 40-year-old man from Ukongoroni, and that what he had actually seen was not a monkey, but a leopard in a fork of a tree about four metres off the ground. This was not at night, but around ten o’clock in the morning at a place called Kiwandani. The Chief Park Warden was not sure of the date: he had maybe interviewed the ranger in April, and perhaps the incident had taken place in January 2007. The ranger was very excited when he came to tell him what he had seen. He described making eye contact with the leopard, whereupon he began to step slowly backwards, thinking that the leopard’s keeper might be somewhere near. He also related how he had watched the leopard stretching its way down the trunk of the tree after leaving the fork in which it was resting. He had not seen a leopard for many years, but claimed not to have been afraid by this unexpected encounter.

It is instructive here to note the discrepancies between the second- and third-hand reports of this sighting: the ranger involved would presumably have given a different account again. In the same interview the Chief Park Warden also mentioned that leopards had been much in evidence in Makunduchi in 2007: they were reported to have preyed on chickens, ducks and goats, and villagers had responded by reading the halbadiri curse [...] against the leopard-keepers, whoever they were. Other reports of leopard sightings have come to us by email. In February 2009 the Zanzibari owner of Zala Park, a small private zoo in Muungoni, wrote to say that a leopard had been heard at Mtule on the Kitogani-Paje road, and that the watchman at a brick-makers claimed to have seen it twice, in December 2008 and January 2009. In March 2009 we were copied into correspondence by the Chief Park Warden reporting that since the start of the year four rangers had independently seen a leopard at a single location in the forest in Jozani-Chwaka Bay National Park. The last sighting was on 1 March. We were unable, however, to obtain further details about these sightings. It is difficult to elicit information on cases like this at a distance, though our former assistant does make an effort to record particular incidents. Indeed he was also involved in the following case, which provides further illustration of the difficulty that we have in verifying reports and reconciling the accounts of different informants.

The Wangwani site
 In April 2002 this officer told Walsh that in August 2001, while undertaking a survey of the mangrove forest north-west of Jozani, he, other staff, and the local villagers who were also members of the surveying team, had come across leopard tracks at Wangwani. They followed the spoor until they encountered the remains of a male Suni antelope (Neotragus moschatus), which they assumed to have been killed by the leopard. He collected one of its horns and also leopard faeces from the site, and took them back to Forestry Commission headquarters at Maruhubi, where they were kept in bottles. However, when Walsh asked to see this material, it could not be found, and he was led to understand that the bottles had probably been thrown away by cleaners or other staff, perhaps afraid of the presence of leopard-related relics in their workplace. Walsh followed this up by interviewing a Jozani Forest Guard who had been present when the finds had been made at Wangwani. He provided a detailed account that differed in a number of ways from the first one. He agreed, however, that they had collected leopard faeces, adding that the men on the team (eight in all) were quite apprehensive about this, afraid that it was a kept leopard and knowing that keeping such objects was a dangerous thing to do. The first officer (our former assistant) later said that he had kept this material on his desk for some time, but that it had indeed been thrown out by the office cleaners. It emerged that the same fate had also befallen presumed leopard scat collected by Goldman and handed over to the office in 1997.[14] In January 2003 Goldman followed up on the Wangwani case. Two of the other men who had been on the original survey team took her to the site of the antelope kill, and provided accounts that were at variance with both of those given to Walsh. They could not cast any light on the loss of the material taken to Maruhubi, but like others in the party were clearly steeped in leopard-keeping lore and not entirely happy with the collection of leopard faeces. Further efforts by Goldman to find out more information produced only more discrepancies, but did at least result in the finding of the original data record of the mangrove survey (Walsh and Goldman 2010: 13-14).

Despite the discrepancies, and the allusions to leopard-keeping, it is easy to imagine a core of truth in these accounts, and to hypothesise that they were based on real sightings and/or signs of a leopard’s presence – or another animal mistaken for a leopard. But other claims stretch credibility. We have written at length elsewhere (Walsh and Goldman 2010) about the ‘kept leopard chases’ that a number of researchers and visitors to Zanzibar have taken part in and sometimes paid money for, lured by the promise of being shown a tame leopard. These quixotic quests for imaginary animals have always proved fruitless, much to the chagrin of their pursuers. In recent years Zanzibari conservationists have joined the pursuit along with non-Zanzibaris beguiled by leopard-keeping narratives. In January 2003 DCCFF staff told Goldman about an ongoing kept leopard chase that had involved a number of them. According to one official in the department, it began three months earlier with reports of leopard predation on livestock. Another official in the same department denied that there had been reports of leopard attacks on livestock. Instead he claimed that the case had been brought to the attention of the DCCFF because a leopard had been seen entering and exiting a house in Marumbi, and the person who had observed this had been bewitched and struck mute. DCCFF employees were sent to follow up on these reports, and one of them saw the house of the supposed leopard keeper together with a peculiar opening at the back that would allow a leopard to come and go. The visitors met with the deputy Sheha (local administrator) to talk about events, but were spooked into silence when they realised that the alleged leopard keeper was lurking nearby. A DCCFF team returned again in January 2003, and their leader asked the deputy Sheha to collaborate in an attempt to get a photo of the kept leopard. The deputy was very reluctant to agree to this and clearly afraid, but was told that as a government employee he had no option but to cooperate. Eventually he yielded, but it was agreed that when the researchers returned with a camera they would have to pretend that they were doing something else, like surveying monkeys or birds. Back in Zanzibar town, the team leader approached Goldman, who was then photo-trapping in Jozani forest, hoping that she would carry out the plan, for which transport for a team of DCCFF staff as well as money for accommodation and a payment to the Sheha were all required. While she considered the limited time and resources at her disposal – and the implications of the discrepancies in different accounts, together with a rumour that the leopard in question had been shifted to another location – the team leader promised to write a letter of introduction that would smooth the way with the local administrator. This letter was never delivered and there were other indications that the DCCFF had dropped their plan to follow up on this case. That was the last we heard of the Marumbi leopard.

Whereas foreign researchers have sometimes been beguiled into believing that leopard-keeping is really practised on Unguja, some Zanzibari researchers and others have sought to reconcile their own witchcraft beliefs with conservation science by proposing that kept leopards be displayed to the public and tourists in particular. The idea of a zoo or holding pen for leopards was suggested by one of the American student researchers referred to earlier, Benjamin Selkow (1995: 12); it has since been taken up enthusiastically by Zanzibaris believing that leopard keepers might make good use of their leopards in this way. One of our research assistants in 1996, the former secretary of the National Hunters, asked a number of our interviewees whether it would be feasible to persuade leopard keepers to display their leopards to the public and fee-paying tourists. The same idea also came up in the discussion that followed our end-of-fieldwork presentation to the Jozani-Chwaka Bay Conservation Project (JCBCP) and other government staff (Walsh 1996). We were careful in that meeting not to overtly criticise the beliefs of the many people in the audience who believed in leopard-keeping, and we did not question the display proposal as directly as we might have done otherwise. In October 1996 Goldman and a colleague in the Commission for Natural Resources were asked to investigate just such a proposition. They met with four men in Kizimkazi who wanted to capture and display leopards: they claimed to have seen a leopard in the area in recent months, and thought it possible that this and indeed all leopards were kept. But they had no clear idea of how they would obtain and care for the animals, other than making a range of suggestions (for details see Walsh and Goldman 2010: 15-16). The proposal did not receive official approval and Goldman (who was then working for JCBCP) heard nothing more of this scheme.

The leopard display idea clearly did not wither and die after the dissemination of our final report, in which we made it clear that we thought leopard-keeping to be wholly imaginary (Goldman and Walsh 1997: iii, 1, 13-15). Indeed it surfaced in a quite unexpected place, in a debate in the Zanzibar House of Representatives in April 2003, when the Deputy Minister for Agriculture, Natural Resources, Environment and Co-operatives declared that his ministry would be happy to buy leopards to display them to tourists (Walsh and Goldman 2010: 16-17). As far as we are aware, no one has yet come forward to sell a live Zanzibar leopard to the government. Needless to say, proposals for the display of leopards continue to be made by independent entrepreneurs such as the owner of Zala Park. In February 2009 he wrote to Goldman to tell her about his plan to build a leopard enclosure at a site in Pete. He asked for any advice she could give on keeping leopards for the purposes of conservation, education and research. In a subsequent communication he declared that he had been forced to put his plans for the new facility on hold because he had been denied permission by the authorities. Expansion of the existing zoo at Muungoni would also require official approval if he was to display animals protected under the Forest Resources Management and Conservation Act of 1996.

Zanzibar servaline genet
These contemporary notions about the possible social and economic benefits of displaying captive leopards are indicative of the continuing strength of the leopard-keeping narratives that underpin them. Meanwhile, we can hypothesise that the cultural salience of leopards on Unguja, the widespread belief in the existence of leopard-keeping, and the consequent apprehension that many people feel about this form of witchcraft, result in many more imagined sightings and claims regarding leopards than would otherwise be the case. Hunters themselves note that some of their colleagues and fellow villagers are liable to mistake other, smaller carnivores on the island for leopards.[15] These include the spotted and banded African civet (Civettictis civetta), which is explicitly compared to the Zanzibar leopard (see Table 1), and the Zanzibar servaline genet (Genetta servalina archeri), known in some parts of Unguja as uchui (Goldman et al. 2004: 6), literally ‘the slender leopard’.[16] The tracks of different local carnivores (and sometimes other animals) can also be readily confused, for example when they have been made in sand or on dusty ground. During their 1997 survey, the Stuarts, who had authored a field guide to tracks and signs (1994), were disappointed by a number of such misidentifications. They cast doubt on the identity of the alleged leopard pugmarks illustrated in our printed report (Goldman and Walsh 1997: 56), and concluded that many islanders erroneously attribute the tracks of the African civet to the leopard (1997a: 4; 1997b: 1; 1998: 37).[17]

There is a large literature on the psychology of eyewitness testimony (e.g. Kapardis 2003: 21-125), and a smaller literature on the evaluation of claimed animal sightings, some of it written by cryptozoologists (e.g. Rabbit 2002). It is widely acknowledged that preconceptions can affect perceptions, and there is little doubt that this has happened in the case of the Zanzibar leopard, generating both false sightings and the misinterpretation of tracks and perhaps also other signs. Kept leopards and leopard keepers are imaginary beings, and belief in them has certainly contributed to imagined sightings, mistaken identifications, and the proliferation of gossip and rumour relating to these (cf. Stewart and Strathern 2004). But some false attributions would probably be made anyway, for example when other animals or indirect evidence for their presence are confused with the leopard and its presumed signs. It is also possible that some misidentifications are not involuntary, but fabricated or otherwise elaborated by informants eager to please researchers, as the Stuarts argued (1997a: 4). We have seen examples of this ourselves when asking people to identify photographs of animals, a procedure which suffers from a number of pitfalls (cf. Barley 1983: 96-97; Diamond 1989).

This leads us to ask whether all recent reports of leopard sightings and signs can be explained away, as the Stuarts suggested, although even they allowed that a few individual leopards might survive (1997a: 3; 1997b: 1). As we have seen, some reports are more convincing than others. Leopards elsewhere are known to be largely nocturnal and secretive animals, capable of roaming through city suburbs as well as through the African countryside without being observed (Guggisberg 1975: 228; Seidensticker 1991: 107; Nowell and Jackson 1996: 28; Sunquist and Sunquist 2002: 321). With this in mind, we cannot entirely rule out the possibility that the Zanzibar leopard does persist, as we stated in an earlier paper (Goldman and Walsh 2002). The same is suggested by the relatively recent discovery of two nocturnal carnivores on Unguja: the Zanzibar servaline genet in 1995 (Van Rompaey and Colyn 1998; Goldman and Winther-Hansen 2003a; 2003b), and a local population of the African palm civet (Nandinia binotata) in 1998-99 (Perkin 2004). Both of these were previously unrecorded, though they were known to villagers (Goldman et al. 2004).


The title of Jared Diamond’s ‘Extant unless proven extinct? Or, extinct unless proven extant?’ (1987) summed up a question that continues to be debated by conservation biologists (e.g. Brussard 1986; King 1988; Mace and Collar 1995; Reed 1996; Kéry 2002; Butchart et al. 2006; Roberts et al. 2009; Vogel et al. 2009). How is extinction in the recent past to be recognised or inferred? As Diamond and others have observed, detecting extinction is often easier said than done. The various qualitative and quantitative answers given to this question have influenced successive editions of the IUCN Red List of Threatened Species.[18] The current Guidelines for Using the IUCN Red List Categories and Criteria (Version 8.1) state that:
“A taxon is Extinct when there is no reasonable doubt that the last individual has died. A taxon is presumed Extinct when exhaustive surveys in known and/or expected habitat, at appropriate times (diurnal, seasonal, annual), throughout its historic range have failed to record an individual. Surveys should be over a time frame appropriate to the taxon’s life cycles and life form.” (IUCN 2010: 9).
Operationalising this definition, however, and drawing a sharp dividing line between the IUCN categories of Critically Endangered and Extinct, has proved problematic. One problem is that premature declarations of demise can lead to what has been called the ‘Romeo Error’, whereby “any protective measures and funding are removed from threatened species in the mistaken belief that they are already extinct” (IUCN 2010: 67, citing Collar 1998). In response to discussions of this problem (see Butchart et al. 2006), the IUCN has introduced a qualifying tag so that taxa of indeterminate status can be described as ‘Critically Endangered (possibly extinct)’:
“Critically Endangered (possibly extinct) taxa are those that are, on the balance of evidence, likely to be extinct, but for which there is a small chance that they may be extant. Hence they should not be listed as Extinct until adequate surveys have failed to record the species and local or unconfirmed reports have been investigated and discounted.” (IUCN 2010: 67)
If we had to choose a category (and tag) for the Zanzibar leopard it would probably be this one.[19] But we are also aware that a single corroborated sighting (for example supported by photographic evidence), or a physical specimen (and genetic profile) with well-documented provenance, would be sufficient to remove the ‘possibly extinct’ tag, at least in the short term.

Compare the Stuarts’ conclusion following their 1997 survey:
“We encountered absolutely no sign of leopards during the survey and we believe that this cat is now extinct on the island, or at best present in such low numbers that there is little, or no, hope of doing anything to save it in the wild state.” (Stuart and Stuart 1997b: 1)
Although they allowed for the survival of leopards, this pessimistic statement, as we have seen, discouraged further research and efforts to develop a leopard conservation programme.[20] The Stuarts’ search for leopards and leopard sign, which included camera-trapping, together with a later photo-trapping survey conducted in Jozani forest by Goldman and Winther-Hansen (2003a), have been the only systematic attempts to detect the Zanzibar leopard to date. These do not constitute the “exhaustive surveys in known and/or expected habitat” without any record of an individual that are required if a taxon is to satisfy IUCN criteria for Extinction. The Stuarts were sceptical about the evidence for recent leopard kills and sightings that we recorded in our 1997 report, and discounted other reports that they heard in the field without thoroughly investigating these. Nor, it must be said, have we had the opportunity to follow up many of the accounts that we have been given since 1997 that might be counted as “reasonably convincing recent local reports or unconfirmed sightings” and that would support the categorisation of the Zanzibar leopard as ‘Critically Endangered (possibly extinct)’ (IUCN 2010: 68). Needless to say, it would be difficult to apply probabilistic methods for inferring extinction (e.g. Roberts et al. 2009; Vogel et al. 2009) to these reports, given questions about their reliability, and the fact that many of them are not first-hand. Likewise any attempt to use the statistics of leopard kills for the same purpose would run into difficulties because they represent kills by only one group – the National Hunters – and because they appear not to have been kept once it became illegal again to kill leopards. Even if these methods were applied, it would only take a single corroborated observation, sample or specimen to falsify a hypothesis of extinction. Argument then shifts to determining what exactly counts as corroboration and so adequate evidence for falsification (cf. Roberts et al. 2009), and brings us back to assessing the reliability of Zanzibaris’ claims to have seen leopards and other signs of their presence.

Cryptozoology flourishes in this epistemological abyss. [the discussion continues]
ENDNOTES [numbered as in the draft]
13. In addition to the statistics for the years 1983-95 analysed in our earlier work (Goldman and Walsh 1997; 2002), we have since obtained a complete record for 1995, and for the years 1996-99 and 2001-02.

14. Our interest in locating this material was prompted by the possibility that it might be used in genetic analysis, but nothing kept in the Maruhubi office appears to have survived, including the photographs of leopard pelt and presumed pugmarks that were included in the original printed version of our 1997 report (Goldman and Walsh 1997: 55-56, Figures 7-12: these particular images are therefore absent from the more widely distributed pdf version of the report).

15. On the other hand, there are a number of reports of a leopard being accidentally shot by a hunter who has caught no more than a fleeting glimpse of it in the dark, and assumed that it was a duiker or other animal.

16. The local dialect name uchui (earlier *luchui) is derived from a combination of the Swahili class 11 noun prefix u- (*lu), which typically signifies length and/or thinness, and the noun root -chui, ‘leopard’ (cf. Nurse and Hinnebusch 1993: 349-351, 639).

17. Ordinarily, pawprints left by African civets and leopards are readily distinguishable because the former include claw marks whereas the latter do not: leopards keep their claws retracted when walking (Stuart and Stuart 1994: 17, 19, 26, 40). The Stuarts argued that the clawless pugmarks that are sometimes observed on paths in the Zanzibari forest and bush are signs of large African civets whose claws have been worn down by the rocky outcroppings prevalent on the coral rag of Unguja. Zanzibaris who are knowledgeable about wildlife deny this; according to them, African civet claws are never abraded to the extent that they leave no trace in pugmarks.

18. For the IUCN Red List and its background see (last accessed on 30 April 2011).

19. Our Zanzibari informants, however, have offered a wide variety of opinions that would translate into the full range of IUCN categories and tags: Not Evaluated, Data Deficient, Least Concern, Near Threatened, Vulnerable, Endangered, Critically Endangered, Critically Endangered (possibly extinct in the wild), Critically Endangered (possibly extinct), Extinct in the Wild, and Extinct (IUCN 2010: 7-9, 67-70)!

20. This example suggests that applying the ‘possibly extinct’ tag might not always avert the Romeo Error: declaring an animal extinct or almost certain to become so (“Going or gone” in the words of the title of Butchart et al. 2006) can kill off hope and funding for its conservation as effectively as pronouncing its definite extirpation.


Barley, Nigel. 1983. The Innocent Anthropologist: Notes from a Mud Hut. London: Penguin Books.

Brussard, Peter F. 1986. The likelihood of persistence of small populations of large animals and its implications for cryptozoology. Cryptozoology 5: 38-46.

Butchart, S. H. M., A. J. Stattersfield and T. M. Brooks 2006. Going or gone: defining ‘Possibly Extinct’ species to give a truer picture of recent extinctions. Bulletin of the British Ornithologists’ Club 126A: 7-24

Collar, N. J. 1998. Extinction by assumption; or, the Romeo Error on Cebu. Oryx 32 (4): 239-244.

Diamond, Jared M. 1987. Extant unless proven extinct? Or, extinct unless proven extant? Conservation Biology 1 (1): 77-79.

Diamond, Jared M. 1989. The ethnobiologist’s dilemma. Natural History 6: 26-30.

Eberhart, George M. 2002. Mysterious Creatures: A Guide to Cryptozoology. Santa Barbara: ABC-CLIO, Inc.

Goldman, Helle and Martin Walsh. 1997. A Leopard in Jeopardy: An Anthropological Survey of Practices and Beliefs which Threaten the Survival of the Zanzibar Leopard (Panthera pardus adersi). Zanzibar Forestry Technical Paper No. 63. Jozani–Chwaka Bay Conservation Project, Commission for Natural Resources, Zanzibar.

Goldman, Helle and Martin Walsh. 2002. Is the Zanzibar leopard (Panthera pardus adersi) extinct? Journal of East African Natural History 91 (1/2): 15-25. Map printed in the 2003 issue: JEANH, 92: (1/2): 4.

Goldman, Helle, Martin Walsh and Jon Winther-Hansen 2004. Zanzibar’s recently discovered servaline genet. Nature East Africa 34: 5–7.

Goldman, Helle and Jon Winther-Hansen. 2003a. The Small Carnivores of Unguja: Results of a Photo-Trapping Survey in Jozani Forest Reserve, Zanzibar, Tanzania. Tromsø: privately printed.

Goldman, Helle and Jon Winther-Hansen. 2003b. First photographs of the Zanzibar servaline genet Genetta servalina archeri and other endemic subspecies on the island of Unguja, Tanzania. Small Carnivore Conservation 29: 1–4.

Guggisberg, Charles A.W. 1975. Wild Cats of the World. London: David & Charles.

IUCN (International Union for Conservation of Nature and Natural Resources) 2010. Guidelines for Using the IUCN Red List Categories and Criteria: Version 8.1 (August 2010), prepared by the IUCN Standards and Petitions Subcommittee. Online at (last accessed on 30 April 2011).

Kapardis, Andreas 2003. Psychology and Law: A Critical Introduction (second edition). Cambridge: University of Cambridge Press.

Kéry, Marc 2002. Inferring the absence of a species – a case study of snakes. Journal of Wildlife Management 66 (2): 330-338.

King, F. Wayne 1988. Extant unless proven extinct: the international legal precedent. Conservation Biology 2 (4): 395-397.

Mace, Georgina M. and N. J. Collar 1995. Extinction risk assessment for birds through quantitative criteria. Ibis 137: S240-246.

Nowell, Kristin, and Peter Jackson (eds.) 1996. Wild Cats: Status Survey and Conservation Action Plan. Gland, Switzerland: IUCN.

Nurse, Derek and Thomas J. Hinnebusch 1993. Swahili and Sabaki: A Linguistic History. Berkeley: University of California Press.

Perkin, Andrew 2004. A new range record for the African palm civet Nandinia binotata (Carnivora, Viverridae) from Unguja Island, Zanzibar. African Journal of Ecology 42: 232-234.

Rabbit, Jack 2002. Native and western eyewitness testimony in cryptozoology. In Eberhart 2002: xxxv-xliii.

Reed, J. Michael 1996. Using statistical probability to increase confidence of inferring species extinction. Conservation Biology 10 (4): 1283-1285.

Roberts, David L., Chris S. Elphick and J. Michael Reed 2009. Identifying anomalous reports of putatively extinct species and why it matters. Conservation Biology 24 (1): 189-196.

Seidensticker, John. 1991. Leopards. In John Seidensticker and Susan Lumpkin (eds.) Great Cats: Majestic Creatures of the Wild. Emmaus, PA: Rodale Press. 107-114.

Selkow, Benjamin 1995. A survey of villager perceptions of the Zanzibar leopard. Student paper, SIT Study Abroad, Zanzibar.

Stewart, Pamela J. and Andrew Strathern. 2004. Witchcraft, Sorcery, Rumors, and Gossip. Cambridge: Cambridge University Press.

Stuart, Chris and Tilde Stuart 1994. A Field Guide to the Tracks and Signs of Southern and East African Wildlife. Cape Town: Southern Book Publishers.

Stuart, Chris and Tilde Stuart 1997a. A Preliminary Faunal Survey of South-eastern Unguja (Zanzibar) with Special Emphasis on the Leopard Panthera pardus adersi. Unpublished report, African-Arabian Wildlife Research Centre, Loxton, South Africa.

Stuart, Chris and Tilde Stuart 1997b. Zanzibar leopard – myth or reality? The Arc (Newsletter of the African-Arabian Wildlife Research Centre) 3:1-2.

Stuart, Chris and Tilde Stuart 1998. Unguja[,] Zanzibar’s island of mystery. Africa –Environment and Wildlife 6 (5): 32-38. [with addendum in Vol. 7 (3): 11]

Sunquist, Melvin E. and Fiona Sunquist 2002. Wild Cats of the World. Chicago: The University of Chicago Press.

The Revolutionary Government of Zanzibar 1996. The Forest Resources Management and Conservation Act, 1966. Zanzibar: Government Printer.

Van Rompaey, Harry and Marc Colyn 1998. A new servaline genet (Carnivora, Viverridae) from Zanzibar island. South African Journal of Zoology 33 (1): 42-46.

Vogel, Richard M., Jonathan R. M. Hosking, Chris S. Elphick, David L. Roberts and J. Michael Reed 2009. Goodness of fit of probability distributions for sightings as species approach extinction. Bulletin of Mathematical Biology 71 (3): 701-719.

Walsh, Martin 1996. The Zanzibar Leopard: An Anthropological Survey. End of Fieldwork Summary. Report to Jozani–Chwaka Bay Conservation Project, CARE Tanzania, and Commission for Natural Resources, Zanzibar.

Walsh, Martin and Helle Goldman 2007. Killing the king: the demonization and extermination of the Zanzibar leopard / Tuer le roi: la diabolisation et l’extermination du leopard de Zanzibar. In E. Dounias, E. Motte-Florac and M. Dunham (eds.) Le symbolisme des animaux: l’animal clef-de-voûte dans la tradition orale et les interactions homme–nature / Animal Symbolism: The ‘Keystone’ Animal in Oral Tradition and Interactions between Humans and Nature. Paris: IRD. 1133-1182.

Walsh, Martin and Helle Goldman 2010. Chasing imaginary leopards: science, witchcraft and the politics of conservation in Zanzibar. Paper presented to Contemporary Issues in Swahili Ethnography, VIII European Swahili Workshop, University of Oxford, 19-21 September 2010. [subsequently revised and submitted for publication]