The 13,791 km² Kerinci Seblat National Park has an elongated shape that spans four provinces (Jambi, Bengkulu, West Sumatra and South Sumatra). The evergreen rainforest of the park is abutted by a hard edge of mainly smallholder farmland supporting crops including coffee, cocoa and cinnamon, as well as industrial oil palm plantations in the south. Kerinci Seblat National Park is listed as a natural UNESCO World Heritage Site because of its rich biodiversity, which includes a globally significant population of tigers [18]. Unlike many other parts of Asia where tiger populations in human-dominated landscapes have been extirpated, Kerinci Seblat National Park not only contains a large tiger population, but also one that lives in close proximity to people [12, 19]. The national park operates six law enforcement patrol teams to deter poaching of tigers and their ungulate prey. These teams include members of local communities who are supported by a local informant network operating widely across the landscape to good effect [20, 21].

There are seven main ethnic groups (Kerincinese, Javanese, Minangkabau, Melayu, Rejang, Pekal and Sanat nomadic hunters) living in the Kerinci Seblat landscape. The three main ethnic groups interviewed in our study were the Kerincinese, Minangkabau and the Melayu Found mainly in the Kerinci District, Jambi Province, the Kerinci people live in an enclave surrounded by the national park. They are primarily a community of coffee, cinnamon and rice farmers, and predominantly follow Islam. However, shamanism and ancestral worship endure, as do rules and prohibitions for those wishing to enter the forest [22]. The Minangkabau, who are the largest matrilineal society in the world, are indigenous to the Minangkabau Highlands of West Sumatra Province. They are also predominantly Muslim and the basis of their customary law, known as Minangkabau adat, is strongly influenced by the concept that nature should be viewed as a teacher. Their most important natural law is growth, which in nature is nurtured by the sun and the moon and in a human context, by the mother, resulting in its matrilineal system [15]. The Melayu inhabit parts of North Sumatra, Riau, Jambi and South Sumatra provinces and share ancestry with the dominant ethnic group on the Malaysian peninsula. Early Malays were largely animistic with many elements remaining following the introduction of Islam [23].

Data collection was embedded within a larger questionnaire-based study which used information on 228 human-tiger incidents, reported by local people to the Kerinci Seblat National Park tiger response team, to stratify sampling across the landscape according to incident density (high, medium or low) [24]. Data were gathered from a systematic sample of male and female heads-or-households residing in 11 areas surrounding the national park (four high, three medium and four low incident density). The last seven questions in the questionnaire (Table 1) served as a filter triggering an invite for participation in this study which used semi-structured interviews to gain greater insight into people’s relationships with tigers from those identified as having particularly strong tiger-related beliefs. In addition, in order to maximise data collection from individuals perceived to have strong spiritual beliefs towards tigers, or tiger-related stories, snowball sampling, whereby enumerators asked village heads and questionnaire respondents if they knew anyone fulfilling such criteria, was also adopted.

The semi-structured interview contained closed-answer (yes/no) and open-ended questions designed to explore the associations between ethnicity and: (i) customary laws and Islamic beliefs associated with tigers; (ii) the existence of tiger shaman; (iii) responses to human-tiger conflict scenarios; and, (iv) different types of tiger perceived to be living in the study area. Socio-demographic data including sex, age, number of years of formal education and of residence in village, and main source of income were also recorded (S1 and S2 Files).

Closed-answer questions were used to investigate the prevalence of customary laws and religious teachings about tigers and whether or not respondents believed in tiger shaman. Where affirmative responses were given, interviewees were asked for further details. For example, with respect to tiger shaman, open-ended questions explored whether interviewees had a story about a tiger shaman, or if they knew of one in their village or elsewhere in Sumatra.

Next, eleven human-tiger conflict scenarios were presented to respondents as open-ended questions. The scenarios ranged from the benign (e.g. a tiger passes through your village), to the more severe (e.g. a tiger kills your livestock) and the severest (e.g. a tiger kills your friend whilst he is out in the forest hunting for non-tiger prey). These questions aimed to gather local perspectives on human-tiger conflicts that occur in the human domain (village) and the tiger domain (forest); and under conditions that would likely be associated with either a zoological tiger (such as livestock depredation), or a spirit tiger (e.g. a village member is killed by a tiger for committing adultery). Six main themes to these open-ended questions were identified and coded as: (i) do nothing; (ii) kill the tiger; (iii) report the incident to the authorities; (iv) relocate the tiger to another forest; (v) permanently remove the tiger from the wild; and, (vi) other.

Finally, given the strong belief systems that have previously been identified from the Kerinci Seblat landscape [16], respondents were asked to describe the different types of tiger that they believed to live in the forests near their village. This included information on: the types of tiger (zoological or spirit) villagers recognise as living in their forest and their physical characteristics; stories and anecdotes concerning incidents involving any type of tiger; tiger folk-lore, including references to its role in the forest, connection with the village and human-tiger associations; and, any different tiger-related perspectives within and between local ethnic groups. These qualitative data were coded in order to estimate the prevalence of different types of tiger and to identify quotes illustrative of typical examples and experiences.

Data were collected between October 2014 and July 2016 after piloting the survey. Interviews were conducted in the Indonesian language with one or two interviewers present. Interviews lasted between two and three hours. No financial incentives were offered to interviewees. However, they were offered snacks and cigarettes during the interview as this is considered a friendly and polite gesture that is culturally appropriate in the study landscape and it helps creates a relaxed atmosphere. To ensure anonymity, we did not record the names or addresses of interviewees. With permission, interviews were recorded using a Dictaphone or transcribed directly.

Data from open-ended questions were transcribed in full (Microsoft Word) at the soonest opportunity; coding was conducted in Excel. Descriptive statistics of demographic and fixed-answer questions were generated Microsoft Excel. Pearson’s chi-squared test, performed in R version 3.2.0, were used to investigate differences between ethnicity and the prevalence of tiger-specific customary laws and Islamic beliefs, understanding of Khalifa and belief in, and presence of tiger shaman.

Ethical approval was granted by the School of Anthropology and Conservation Research Ethics Advisory Group, University of Kent. Free prior informed consent was obtained verbally from all participants. The University of Kent partner, FFI, collected the field data for this project. It operates in Indonesia under an institutional memorandum of understanding with the Ministry of Environment and Forestry.

Focussing on the three main ethnic groups interviewed, most Minangkabau (73.5%) cited that their ethnic group had customary laws towards the tiger (Table 2). A significantly lower proportion of Melayu (52.0%) and Kerincinese (44.1%) reported the existence of tiger-specific customary laws (X² 8.882, p = 0.031). Overall, there was little support for the belief that there was a connection between Islam and tigers, although significantly more Minagkabau (22.4%) stated that there was (X² 8.725, p = 0.033). There was no significant difference between ethnic groups and their level of understanding of the term Khalifa, which defines the role of humans as guardians (Khalifah) of God’s creations (X² 2.202, p = 0.531). Amongst those who understood the term Khalifa (n = 78), most reported that there was no connection to tigers; this believe was most prevalent amongst Kerincinese (78.6%), followed by the Minangkabau (68.0%) and Melayu (60.7%).

Regardless of ethnicity, most interviewees believed in tiger shaman (61.7%), but few knew of a tiger shaman living in their village (10.4%), or believed there were tiger shamans elsewhere in Sumatra (16.2%) (Table 3). There were no significant differences between ethnic groups and whether they believed in tiger shamans (X² 0.196, p = 0.978), or whether they believed there were shamans elsewhere in Sumatra (X² 6.879, p = 0.075). However, Kerincinese respondents were significantly more likely to report there being a tiger shaman in their village (X² 17.518, p = 0.001). Typical Kerincinese narratives concerning tiger shaman are captured in this quote, “When I was young there was a tiger shaman in this village. He helped our village capture a [zoological] tiger that came to the village. He captured the tiger and killed it, after it came to a cage he made. He captured this tiger because it had killed many livestock in the village.”

Table 4 presents responses to the human-tiger conflict scenarios broken down by ethnicity. For each scenario, one of three responses, which we discuss further, was typically given: (i) Do nothing; (ii) Kill tiger; and (iii) Report to the authorities. A general pattern observed in the data was that for nearly all of the scenarios presented, most respondents, irrespective of ethnic group, suggested doing nothing to the tiger. Where there was no overall consensus, the conflict scenario related to an unexplained loss. For example, if a tiger killed livestock, either belonging to the respondent or a village member, the Minangkabau were more likely to suggest that the tiger be killed (#2 46.7% and #3 47.4%) than an alternative management option, whereas the Kerincinese were least likely to do this (#2 11.5% and #3 9.1%), being more inclined to report the incident to the authorities (#2 38.5%, #3 36.4%); the Melayu were split between the three responses of do nothing, kill the tiger, or report it to the authorities (#2 46.7%, 31.1% 20.0% and #3 43.8%, 28.1%, 28.1%). This type of response pattern between ethnic groups was also found for the two scenarios where a person was killed for an unprovoked attacked (#4 and 5). However, if a village member or a family member was killed by a tiger whilst out hunting in the forest, then the typical response for all ethnic groups was to do nothing (#6 and #7), with the exception of the Kerincinese; 45.5% would report the killing of their brother by a tiger to the authorities. The response of ‘do nothing’ became more pronounced, particularly for the Minangkabau and Melayu, when a village member or family member who had committed adultery was killed by a tiger (#8–11), with the Kerincinese again being split between doing nothing and reporting to the authorities.

Three main types of tiger were mentioned by respondents. Most believed in the existence of a zoological tiger (90.9%) and spirit tiger (53.9%), whilst some (31.2%) mentioned the existence of weretiger (Fig 1). In describing these different types of tiger, several distinctive and reoccurring adjectives or phrases were identified. A weretiger would typically be the reincarnation of a person who had committed bad deeds in their former life and then begged God to allow them to return to earth to avoid further torture and punishment in the afterlife. Weretiger behaviour ranged from providing protection to its ancestors or their property, killing village livestock, or killing people who had a disease locally referred to as darah buruk (bad blood). Weretiger activities were expressed such: “A man passed away, after 45 days he transformed into a weretiger… because during his life, he did bad things. This brought God’s wrath upon him, but it was too painful, so the person asked God to return to the world again and God returned him as a weretiger. When transformed as a tiger, this person eats people’s poultry and scares the villagers”; and, “The weretiger doesn’t disturb people. It takes care of his/her descendants. For example, it takes care of their grandchildren’s farm from wild boar or other pests”.

The term “spirit tiger” was found to encompass three core elements: embodiments of ancestors; protectors of the village; and enforcers of moral codes. For example, spirit tigers may embody the soul of a villager’s ancestor, help people and guard the village: “They [the spirit tigers] were the villager’s ancestors. They passed away then transformed to become tigers. Based on the elders’ [stories], one day the tigers came to their grandchildren’s dream and told them that they transformed to be tigers and will guard the village. …they will show the villagers the way home when they get lost in the forest, or they will guard us when there is a fight between villages.” Acting as an enforcer of moral rule, spirit tigers alert villagers to transgressions “If there is someone having sex without being married, villagers believe that a spirit tiger will leave a [pug] mark in the village”. The role of the spirit tiger as village protector and enforcer of moral rule can be combined “There is a tiger that protects this village. If something bad comes to this village, like a disaster, crime or having sex without being married, it [the spirit tiger] will let the villagers know by roaring or show itself near forest edge.”