Year : 2003 | Volume
| Issue : 2 | Page : 223-246
The Role of Local Taboos in Conservation and Management of Species: The Radiated Tortoise in Southern Madagascar
Marlene Lingard1, Nivo Raharison2, Elisabeth Rabakonandrianina2, Jean-Aime Rakotoarisoa3, Thomas Elmqvist1
1 Department of Systems Ecology, Stockholm University, S-10691 Stockholm, Sweden
2 Department of Biology and Plant Ecology, University of Antananarivo, Antananarivo B.P. 101, Madagascar
3 Museum of Art and Archaeology, University of Antananarivo, BP 564 Antananarivo 101, Madagascar
Department of Systems Ecology, Stockholm University, S-10691 Stockholm
Source of Support: None, Conflict of Interest: None
|Date of Web Publication||20-Jul-2009|
| Abstract|| |
The radiated tortoise, Geochelone radiata, is endemic to the semi-arid region of southern Madagascar. Despite formal protection by law since 1960 and listing in CITES since 1975, tortoise populations have been reported to be in rapid decline, mainly due to illegal harvesting for food and commercial trade. The Tandroy people, inhabitants of the Androy region, which covers approximately half the tortoise distribution range, do not, however, exploit the species. The Tandroy prohibition against tortoise consumption is expressed as a taboo or fady. The aim of this study was to document the narratives, rules and enforcement mechanisms linked to the taboo, and to assess the potential role of the taboo for the protection and management of the radiated tortoise. Interviews revealed that the Tandroy perception of the animal as 'dirty' underlies the Tandroy taboo, although one informant suggested that the taboo once originated in notions of sacredness. Estimated tortoise abundances ranged from 20 tortoises per ha in an area with no harvesting to 0.6 per ha in an area where a significant proportion of residents were reported to violate the taboo. Infrastructure changes and increasing numbers of immigrants to the region are sources of new pressures on the tortoise. An official acknowledgement of the fady custom and the transformation of this institution for the purpose of conservation and sustainable management of the tortoise may considerably reduce the current high costs of enforcement by formal institutions. The tortoise may constitute an important economic source of revenue if local communities are granted rights to a regulated small-scale trade for the pet market based on locally controlled farming of tortoises. Such actions may provide economic incentives for further transforming and building effective institutions for sustainable management. However, a local institutional strategy also needs to be nested with institutions across scales, for example, at regional and national levels,assisting in controlling harvest and trade.
|How to cite this article:|
Lingard M, Raharison N, Rabakonandrianina E, Rakotoarisoa JA, Elmqvist T. The Role of Local Taboos in Conservation and Management of Species: The Radiated Tortoise in Southern Madagascar. Conservat Soc 2003;1:223-46
|How to cite this URL:|
Lingard M, Raharison N, Rabakonandrianina E, Rakotoarisoa JA, Elmqvist T. The Role of Local Taboos in Conservation and Management of Species: The Radiated Tortoise in Southern Madagascar. Conservat Soc [serial online] 2003 [cited 2020 Jan 21];1:223-46. Available from: http://www.conservationandsociety.org/text.asp?2003/1/2/223/49349
| Introduction|| |
THE ROLE OF local communities and local ecological knowledge in the management of natural resources and ecosystems has received increasing attention in recent years (for example, Berkes and Folke 1998, 2002; Berkes et al. 2000, 2003; Gadgil et al. 1993; Olsson and Folke 2001; Ostrom 1990). Local ecological knowledge refers to a cumulative body of knowledge, practice and belief of the relationships of living beings (including humans) with one another and with their environment, reflecting the dynamic way in which people organise perceptions of flora, fauna, ecosystem processes, culture beliefs and history (for example, Berkes and Folke 1998; Berkes et al. 2003; Gadgil et al. 1993). Such knowledge is often tacit, and its transmission and practical implementation is frequently accomplished through the prescriptions of social institutions, such as rituals and taboos (Colding and Folke 2001). The institution of taboos is a universal regulator of human behaviour, and a taboo can be described as a social prohibition of something that is regarded holy or unclean, and is often connected to ritual. Colding and Folke (2001) consider some taboos to be integral parts of 'invisible' systems of resource management and refer to nature-related social taboos as resource and habitat taboos [Table 1].
Colding and Folke argue that, although not necessarily perceived as instruments of resource management by the people who practise them, taboos nevertheless often show a functional similarity to the institutions of formal nature conservation. Sd fic-,lies taboos regulate the utilisation of particular species and are usually inclusive prohibitions, banning exploitation of a particular species at all times. The rationale behind the existence of specific-species taboos vary, ranging from notions of the species being toxic, being perceived as religious symbols, as well as being avoided due to their behavioural and physical appearance. In an analysis of seventy specific-species taboos, about 30 per cent, predominantly reptiles and mammals, were found to be involve species recognised as 'threatened' by the International Union for the Conservation of Nature (IUCN) (Colding and Folke 1997). While this may be an unintended consequence, the enforcement of taboos may have a direct impact on species conservation. Taboos and other forms of informal institutions have, however, seldom been incorporated in biological conservation schemes, partly due to narrow definitions of what constitutes conservation (Berkes et al. 2003; Colding et al. 2003).
The radiated tortoise, Geochelone radiata, is among the world's rarer tortoises and is one of four endemic tortoise species in Madagascar. It occurs in the semiarid region of the southern part of the island. Populations of the radiated tortoise are reported to be in decline throughout much of its range (Durrell et al. 1989; Juvik 1975; Nussbaum and Raxworthy 2000; O'Brien et al. 2003). Over the past twenty-five years the geographic range of the tortoise is estimated to have decreased by approximately 20 per cent (O'Brien et al. 2003), and the IUCN has listed radiated tortoise as 'vulnerable' (Hilton-Taylor 2000). Humans have harvested tortoises at least since the arrival of Europeans. The giant tortoise, Geochelone gigantea, most likely went extinct in Madagascar as a consequence of heavy exploitation and trade in the past (Bourn et al. 1999). During the eighteenth and nineteenth centuries great numbers of radiated tortoise were exported to the islands of Reunion and Mauritius where the species was considered a delicacy. Malagasy authorities initially protected the radiated tortoise in 1960, and in 1975 the species was listed in Appendix I of the Convention on International Trade in Endangered Species (CITES). Despite these initiatives, great numbers of radiated tortoise are killed for food and for the manufacture of tourist items, as well as exported as part of the exotic pet trade (Durrell et al. 1989; Juvik 1975; Nussbaum and Raxworthy 2000). Basing their proposition on three pieces of evidence, O'Brien et al. (2003) suggested that over-exploitation by humans is currently the most important driver of the decline of the radiated tortoise, that is: (a) commercial harvesters travel increasingly far, up to 200 km, to find sufficient densities of tortoises; (b) tortoises are either completely absent or present at very low abundance at sites subject to commercial harvesting, while in remote regions with no harvesting, tortoises persist at densities of up to 2,500 tortoises per sq. km ; and (c) tortoise abundance increases significantly with distance from urban centres of high demand for tortoise meat.
The Tandroy people, inhabitants of the Androy region [Figure 1], which covers approximately half of the tortoise distribution range, do not exploit the tortoise (sokake, or sokatra). The Tandroy prohibition of eating the sokake is expressed as a taboo, or fady, a regulating social institution likewise present among the Mahafaly people living to the west of Androy. Ruud (1960) comments on the omnipresence of fady in Malagasy everyday life and of the wealth of tales that explain the history behind animal and plant taboos. Many of these tales show a marked similarity and the most common form of decree communicated is the imperative 'you shall not' (Bloch 2001). The tortoise fady of the Androy and Mahafaly regions is one example of this kind of regulation. Although the radiated tortoise recently has received increased conservation attention (Nussbaum and Raxworthy 2000; O'Brien et al. 2003), the potential of the taboo as a viable local institution for the future protection and management of the species has so far not been studied and assessed.
The principal questions addressed in the present study were:
- What are the narratives, rules and enforcement mechanisms associated with the taboo?
- Is the taboo subject to erosion, and, if so, what are the main factors associated with a decrease in adherence to the taboo?
- To what extent is there a correlation between the local degree of adherence to the taboo and local tortoise abundance?
- What is the role, if any, of the taboo in future management of the tortoise?
| Studied Species|| |
Geochelone radiata is a large terrestrial tortoise; adults may weigh up to 13 kg and reach a carapace length of around 40 cm (Durrell et al. 1989; Glaw and Vences 1994). G. radiata is endemic to the extreme southern part of Madagascar and restricted to the xerophytic spiny forest of the Androy and Mahafaly regions [Figure 1]. It has an oval, dark brown-black carapace, with a distinctive yellow starshaped pattern radiating out from the centre of each scute [Figure 2]. The head, feet and legs of the tortoise are yellow, with a variably sized black patch on the top of its head. Young tortoises are not as distinctively marked as mature individuals. The radiated tortoise commonly lays one to five eggs per nest, and the incubation period is thought to be approximately ten months. G. radiata is regarded as the closest relative of another of Madagascar's endemic, and likewise threatened, tortoises, Geochelone yniphora (Caccone et al. 1999). The tortoise is diurnal and is most active during the rainy months of November and December. Due to the cooler weather conditions and aridity, with a subsequent shortage of food, the tortoises stay less active during the winter months of April to November. G. radiata feeds on grass, flowers and leaves of different plants, and the fruits of Opuntia spp. (Cactaceae).
The tortoise has been protected by Malagasy national law (Decree no. 60-26) since 1960 and is receiving further official protection in the nature reserves of Tsimanampetsotsa Reserve Naturelle Integrale, Cap Sainte Marie Reserve Sp&ciale, Beza-Mahafaly Reserve Sp6ciale and Andohahela Reserve Naturelle Integrale (Nussbaum and Raxworthy 2000).
| Study Sites in the Androy Region|| |
The Androy region is situated in the very southern part of Madagascar, roughly located between the rivers Menarandra and Mandrare. This area represents the central area of distribution for the tortoise [Figure 1]. The south is characterised by semi-arid climatic conditions with very irregular rainfall averaging less than 500 mm per year.
The annual rainfall declines from north to south and from north-east to southwest (Battistini and Richard-Vindard 1972), and the seasonal character of the precipitation is particularly pronounced in the south. There is a great irregularity of precipitation from year to year and long periods of drought may occur in this area. Violent rainstorms, which may bring almost half the total annual rainfall in one day, are not unusual (ibid.). The dry season usually lasts eight to nine months, but can extend over several years locally (Dewar and Wallis 1999; Richard et al. 2002).
The southern dry spiny forest is dominated by the endemic plant family Didiereaceae, and has the highest level of plant endemism in all of Madagascar at the generic (48 per cent) as well as the species level (95 per cent) (Koechlin 1972). Recently, the spiny forest was listed as one of the 200 most important ecological regions in the world (Olson and Dinerstein 1998). The range of the spiny forest has declined since the early 1970s, principally due to cattle herding, timber harvest and charcoal production (Sussman et al. 1994). However, in Androy there are still numerous small patches of forest, which for sacred reasons have remained relatively untouched, even in the most intensively used areas. These sacred forests are often sub-circular, 300m to 400m in diameter and often contain tombs (Clark et al. 1998). However, due to the fady and entering restrictions, very little is known about these forests.
This study was conducted during seven weeks of fieldwork in the Androy region between June and July 2001. Three principal locations were selected on the basis of differences in infrastructural and demographic characteristics, key factors that we assumed influenced Tandroy observance of customs and traditions, such as the directive of the tortoise fady. A fourth location, Cap Sainte Marie, a 1,750 ha Reserve Sp6ciale (administered by the Association Nationale pour la Gestion des Aires Protegees [ANGAP]) was selected as a control.
Lavanono Lavanono is a small coastal community with a population of approximately 500 permanent residents (personal communication of village elders). The capture and sale of fish and lobsters is one of the major sources of revenue for the people of this village. Lavanono is situated approximately 39 km south of Beloha. There is a small and very basic resort here, run by villagers as part of a development project. However, the roads that lead to the village are in poor condition, and Lavanono's out-of-the-way location is a factor that limits the number of visitors.
Lavanono study sites:
Beloha Beloha is a small town situated on National Route 10 between Tsiombe and Ampanihy. Due to size and location, Beloha has a slightly more heterogeneous population that Lavanono. The taxi brousse (bus) stops here on its way from Toliara (Tulear) to Tolagnaro (Fort Dauphin), and there is a lively but rather poorly stocked weekly market selling foodstuff and paraphernalia. The road leading to and from Beloha, National Route 10, is in very bad condition.
- Ambohibovo, S25° 24.231', E044° 56.975': Characterised by Euphorbia stenoclada, Croton sp., Opuntia spp., Psiadia angustifolia and a few Al luaudia comosa.
- Anandimby, S25° 24.593', E044°57.411': Characterised by Alluaudia dumosa, Euphorbia fiherenensis and E. stenoclada.
- Betaimbolo, S25°20.330', E044°52.190', c. 11.8 km north-west of Lavanono: Characterised by Alluaudia dumosa, Euphorbia fiherenensis and Jatropha mahafaliensis, with a few A. comosa.
Beloha study sites:
Tsiombe The town of Tsiombe is situated approximately 55 km south-east of Beloha. The people of Tsiombe make up a more heterogeneous population than Beloha. The road east of Tsiombe, leading to Amboasary and Tolagnaro, is in better condition than the westward extension. Easier accessibility to larger trade centres is reflected in a slightly better stocked market than the one in Beloha. There are restaurants offering tortoise, if not on the menu then on request.
- Ambaro, S25° 04.491', E045° 01.605': Chracterised by Alluaudia procera, A. dumosa, Didierea trollii and Operculicarya decaryi.
- Andrambalo, S24° 59.172', E044°54.787': Characterised by Alluaudia procera, A. dumosa and Euphorbia fiherenensis.
- Andombiry, S25° 13.880', E045° 06.291': Characterised by Alluaudia procera, Euphorbia fiherenensis, Salvadora angustifolia and Ormocarpopsis parvifolia.
Tsiombe study sites:
Cap Sainte Marie Geochelone radiata is protected within the Cap Sainte Marie reserve, but there are reports of locals poaching tortoises from the reserve, and both zebu and goats graze within its limits. The tortoise is said to be abundant here despite these activities (Nussbaum and Raxworthy 2000).
- Analamena, S25° 21.316', E045° 31.672': Characterised by Alluaudia dumosa, Euphorbia plantagynea, E. fiherenensis, E. stenoclada, and a few Alluaudia humbertii and A. procera.
- Sakamasy, S25° 17.870', E045° 34.140': Characterised by Alluaudia dumosa, Euphorbia fiherenensis, E. stenoclada, Operculicarya decaryi, and a few Alluaudia humbertii.
- Evazy, S25° 17.215', E045° 22.009': Characterised by Alluaudia procera, Euphorbia sp., Mimosa delicatula, Ormocarpopsis parvifolia and Opuntia sp.
- Cap Sainte Marie Reserve, S25° 32.094', E045° 07.195': Characterised by Euphorbia stenoclada, E. leucodendron and Croton sp.
| Social Survey|| |
Interviews were held with members of the local Tandroy community to learn about the nature and strength of the tortoise fady, sanctions and enforcement mechanisms, and about if and to what extent the prohibition is thought to influence the survival of the tortoise population. To analyse the impact of tortoise harvesting, questions regarding exploitation, as well as pertaining to local knowledge of the tortoise and its habitat were put to key informants.
The social organisation of rural Madagascar is based on the fanjakana on the one hand, representing the formal institutions of Malagasy society, whereas traditions and customs constitute the more informal aspects of the institutional framework. Informal agreements combine the two sets of code, which together form the base for the management of all natural resources. Identification of the person(s) in control of management functions is crucial, and includes in addition to the fokontany (administrative level), the fokonolona, led by clan leaders. The informants of this study were key persons in each locality, such as representatives of fokontany, fokonolona leaders, mayors, local guides as well as ordinary people. Due to immigration, the population of Tsiombe is more heterogeneous than the populations of both Lavanono and Beloha. This variation in demography motivated interviews with representatives of the Tanosy community here, whose members openly admit to consuming tortoises. There was a conscious effort to choose informants from as many different sub-groups within the community as possible, including women and old people. Interviews were also held with public service officials in Beloha, Tsiombe and in the coastal town of Tolagnaro (Fort Dauphin), which is the region's administrative centre.
To shed light on the tortoise fady and on the culture and religious life of the Tandroy people from a different angle, interviews were held with the Christian clergy in Beloha and Tsiombe, two Catholic priests and two Protestant pastors. Three of these were immigrants to the region, albeit of long-standing residence, and could therefore contribute interesting insights from the 'outsiders' point of view. These interviews were also important for understanding the impact of Christianity, and whether this relatively recent influence on the Tandroy culture has affected the fady or the observance of this social regulation.
Most of the information was obtained through semi-structured and fairly informal conversations with informants, in line with the qualitative research interview approach described by Agar (1996) and Kvale (1996). The informants were all briefed on the subject and objectives of the research. The interviewers also communicated to the interviewees their interest in the Tandroy world-view and folklore. The interviewers invited informants to elaborate on subjects of religion, such as ancestors, burial traditions, etc. One of the two persons conducting the interviews was native Malagasy, and acted as an interpreter to the non-Malagasy interviewer throughout the surveys. Apart from the apparent dilemma of entering a region where most people speak no language other than the local Tandroy, a dialectal version of Malagasy, there were a great number of other practicalities that neededto be tackled in line with local customs. Local guides were also engaged, on recommendation of the president of the fokontany in Lavanono, and of the mayors of Beloha and Tsiombe. Local guides escorted the research team during all excursions to the countryside, and their presence constituted an unconditional requirement for obtaining the local research permit. The ANGAP guides of Cap Sainte Marie who accompanied the research team during the survey within the reserve were also interviewed. Most transportation within the region, and always when travelling to and from tortoise survey areas and villages, was by foot or by zebucart. Interviews with public servants and other representatives of government were conducted using a more structured and formal interview method than the one described earlier. The information attained through the interviews has been complemented with literature studies.
| Tortoise Survey|| |
The radiated tortoise is a relatively large and slow animal, and population surveys are usually based on counting along transects. We used a combination of a circular plot ard x 1 line transect survey (see Buckland et al. 1993; Sutherland 2000) to estimate the relative abundances of tortoises. We considered this method as most suitable to counting tortoises during the dry and relatively cool winter months of June to August, a time when the tortoises are less active. A further advantage of this combined method was that it permitted a larger sampling of variation in habitats compared to a sampling scheme based on a few large plots. Plots could in this way be considered independent sampling units, which permitted statistical comparisons between different habitat types. During these surveys we actively searched for animals since during dry periods they frequently hide in dense vegetation and may be difficult to spot. Five 400m long transects were walked once at each study site. The starting point of the first transect was identified after consultations with local guides, and located to permit a corridor (approximately 2 km 1 km), that is, an extended stretch of tortoise habitat, free from any intersecting agricultural areas, roads, settlements, etc. Five separate stops were made along each transect, counting the starting point as number one, with four subsequent stops at intervals of 100 m. A limited time (five to eight minutes depending on type and density of vegetation) was spent at each stop in active search of tortoises. Each search was carried out by three persons, of whom two participated in the entire study, and covered a circular area with a radius of 20 m, making the total search area at each stop approximately 1.250 sq. m and 0.625 ha at each transect. A handheld GPS unit was employed to navigate and to determine the positions of the transect stops. The length of the transect lines were approximated by pacing in combination with the GPS odometer function. Each encountered tortoise was classified according to sex and size; carapace length (CL) determined the three size categories: small CL<15 cm, medium 15>CL<30 cm and large CL>30 cm). An approximate area of 3.1 ha was actively searched at each study site, making the total for each location approximately 9.3 ha, with the exception of Cap Sainte Marie, where the search was limited to an area of 3.1 ha.
A multitude of factors may have generated biases in the survey, such as time of day of search (tortoises may be active during a certain time of day); large tortoises being more easily detected than small; tortoises being more difficult to detect in some habitat types than others, etc. Efforts were made to perform surveys at the same time of day, but this was not always possible. Furthermore, it was difficult to adjust for differences in habitat structure. Due to thorny vegetation some of the sites visited were extremely difficult to walk through, whereas others were surveyed at leisure. To account for this and to reduce biases the length of search time at each stop was adjusted depending on type and density of vegetation. Given these uncertainties, the reported estimates here should only be interpreted as indices of abundances.
The different vegetation types of the circular plots were classified into grasslands (G), shrub (S) and forest (F) to allow comparisons to be made between different habitats.
G = Grasses dominate (Cenchrus ciliaris, Aristida congesta, Panicum voeltzkowii and Eragrostis sp.) with sparse stands of shrub or cacti (Opuntia ficus indica); mean height<1.5 m.
S = Ground covered by stands of shrubby plants (several members of Euphorbiaceae, Acanthaceae, Rubiaceae, Anacardiaceae and Combretaceae, and/ or cacti (Opuntia ficus-indica); mean height >1.5 m <4 m. F = Trees (predominantly Allaudia dumosa, A. comosa, A. procera, Euphorbia stenoclada) making up the dominant (>80 per cent) vegetation cover; mean height >4 m.
| Results|| |
Lavanono Informants in and around Lavanono (n = 30) maintain a strict adherence to the tortoise taboo. There are few immigrants living permanently in the area, and although there is both a Catholic and a Protestant church, with their respective influences of Christian ideals and of practices and customs of external origin, traditional values and institutions still seem to be predominant regulators of everyday life. All Lavanono informants said that the Tandroy regard the radiated tortoise as 'dirty' (maloto) and some persons stated that they would never even consider touching the animal. As the animal was thought of as 'dirty', none of the informants considered the tortoise habitats as in any way special or sacred. The estimates of the informants indicate that approximately 99 per cent of the people living in and around Lavanono adhere to the fady [Table 2]. Some informants related the background story of the tortoise fady (see sokake fady 1 in the appendix) to try to explain why the tortoise is so thoroughly despised, while others considered the story to be too embarrassing to be told.
According to one informant, an elderly man who has lived all his life in Lavanono, tortoises have always been abundant in the area, and are so still. However, large numbers of tortoises are harvested by outsiders visiting the area. The fish and lobster collectors who visit Lavanono twice weekly are stated to comprise the greatest threat to tortoise survival in the Lavanono vicinity. According to an informant with a central position in the trade with the lobster collectors, this harvesting of tortoises may be as high as 100 individuals a week.
Beloha The Catholic and Protestant churches have a strong representation in Beloha. The majority of people questioned here (n = 40) said that the fady was grounded in superstitions, and that they did not really believe that breaking it would lead to any repercussions in the form of ancestral punishment. Some people said that the ancestors were poorly educated as they had not been to school, and that this would explain why they believed in the tortoise fady. They did, however, state that they followed the traditional principles out of respect for family and relatives. All respondents swore that they had never tasted sokake. Paradoxically, most people thought that there are Tandroy in Beloha who do eat the tortoise and key informants estimated that about 15 per cent of the population violate the taboo [Table 2]. In the villages outside Beloha informants showed a notable disgust when asked about the sokake. Many informants said that people from outside come and harvest sokake in great numbers. Lorries on the way to coastal towns stop on their way and pick up every animal in sight. During the rainy season great numbers of tortoise come out on the roads to drink, at which time they are an easy catch for anyone who happens to pass by. There is also a more active search for tortoises by people systematically rummaging through shrub land and forest areas. Tortoise numbers were reported to have declined over the past years.
Tsiombe Exploitation seems to have had a greater impact on the tortoise popula - tion around Tsiombe than in the area of Beloha. Tsiombe is situated closer to the centres of Amboasary and Tolagnaro (Fort Dauphin), and the road leading here from the east coast is in better condition than the one extending westwards. The people of Tsiombe are a more heterogeneous population than the one in Beloha, and many Tanosy (people from the east) live in and around the town. Most inform ants (n = 40) said that sokake had been abundant in the area until very recently . Now, they said, there are hardly any tortoises to be found in the vicinity of Tsiombe.
Many expressed satisfaction at being rid of the animal, because 'sokake eat our food', referring to the animal's preference for grazing in cultivated fields. There were informants from the Tanosy community who proudly showed their captured tortoises. Overcrowded pens housed live tortoises in wait of household consumption or transport to the town centre for sale. One informant estimated that on average he and his family consumed one medium-sized individual per day. To them the meat of the animal was considered a delicacy and the restaurants of Tsiombe served sokake. As in Beloha, most Tandroy say that they never touch sokake themselves, but that many others do. The six key informants estimated that 20 per cent of the population violate the taboo [Table 2].
Dead Sokake Encountered Empty carapaces were encountered during the surveys along transects and along the roads, sometimes in great numbers (twenty to thirty carapaces). These were the remains of tortoises said to have died for one of several reasons: lobster collectors stopping on their way from the coast and collecting the meat of sokake to boost their earnings; lobster collectors feasting on tortoises during their stay in the area; tortoises dying in the flames of the annual 'slash-and-burn' fires, which are a part of the customary cultivation practices; dogs killing young tortoises; children throwing stones on tortoises.
The respective areas showed significant differences in the estimates of relative abundances of tortoises (ANOVA, F3,234 =30.40,P<0.000). In Lavanono a total of 102 tortoises were encountered with a mean of 10.8 per ha, in Beloha, fiftynine tortoises were observed with a mean of 7.7 per ha, and in Tsiombe, a total of six individuals were recorded with a mean of 0.6 per ha [Table 3]. In Cap Sainte Marie fifty-eight individuals were recorded with an estimated density of 20 individuals per ha (based on one site only).
An analysis of the proportion of large tortoises (CL>30 cm) in relation to distances from major roads (that is, roads used by motor vehicles) revealed a significant positive correlation (Spearman r=0.706, P<0.05,n=7) [Figure 3].
No significant deviations in sex ratios from an expected 1:1 were observed at any of the sites (Chi-square test, df=1, P>0.05 in all tests). The total numbers observed in Tsiombe were too low to permit statistical analyses of sex ratio deviations [Table 3].
An average of more than twice as many tortoise individuals per ha were found in shrub and grassland habitats compared to the forests, a difference that was significant (ANOVA,F2,235 =11.4, P<0.0001) [Figure 4]. There was no significant difference between the average proportions of habitats in the samples: forest 39 per cent, shrub 31 per cent and grassland 30 per cent (Chi-square test, df=2, P>0.05).
Sampling and analyses of tortoise droppings, and personal communication with local guides, revealed that the dominant part of the tortoise's diet during the dry months of June and July seems to consist of grasses (for example, Cenchrus ciliaris, Aristida congesta and Eragrostis sp.) and the fruits of Opuntia spp. Interviews indicated that several other species are consumed by sokake at other times of the year [Table 4].
| Discussion|| |
The most common narrative associated with the fady is tied to myths relating incidents between ancestors and the tortoise and was told in similar versions throughout the study area (sokake fady 1 in the appendix). The narrative does not contain any clear element of the tortoise carrying a spiritual value, and today most Tandroy consider the tortoise 'dirty', an animal of no use. However, one key informant referred to a narrative of potential significance. This narrative (the second one in the appendix), a story claimed to be largely forgotten by the Tandroy today, describes the fady as originating in the animal's once-sacred status. Further research should look into the origin and significance of this story (cf. Juvik 1975).
The Androy region was previously a rather isolated part of Madagascar, and customs and traditions were long lived and in all probability mostly followed (Parker Pearson 1997). Wherever the traditional Tandroy social practices and customs were followed, the fady with its 'no touch' principle has most likely offered the tortoise effective protection (Juvik 1975). Nussbaum and Raxworthy (2000) even expressed the opinion that the fady has prevented the sokake from going extinct. According to the findings of our interviews, violations of the fady still appear to be rare among Tandroy living in rural areas, the common sanction being community exclusion. In Lavanono and in the rural areas of Beloha informants reported of one or two cases of offenders who, as a result of the fady, had been driven away from their families and ultimately forced to move to areas where there was no knowledge of the offence. The fady also appears to carry a strong element of self-enforcement since many Tandroy informants, especially inhabitants of the more remote tracts, explained that not even in critical situations, such as during occasional famines caused by drought or locust outbreaks (for example, during 1996 and 1997), would they consider eating the tortoise (but see O'Brien et al. 2003). A different attitude was observed among the urban Tandroy population of Beloha and Tsiombe, where a common view was that the fady was grounded in superstitions and that breaking it would not lead to any repercussions in the form of ancestral punishment. However, all informants stated that they did follow the traditional principles out of respect for family and relatives. Intensifiedcontacts with the surrounding world bring new impulses to the people who live in the region. One consequence of this scenario is erosion of the institution of the fady with increased rates of both local subsistence harvest and harvest related to commercial trade of the tortoise. Immigration is clearly one factor influencing the erosion of the fady. In contrast to the Tandroy, Tanosy immigrants in Beloha and Tsiombe openly admitted that they harvest and consume large quantities of tortoise, and in both Beloha and Tsiombe key informants reported that a substantial portion of the population do in fact violate the taboo.
All people questioned appeared to be aware of a recent decline in numbers of tortoises, and according to the majority of informants, illegal harvesting constitutes the major driver of this decrease. The significant difference between Lavanono and Tsiombe [Table 3] in estimates of tortoise abundance may thus be interpreted as reflecting different harvesting pressures in areas with different degrees of adherence to the taboo within the human population. However, O'Brien et al. (2003) correctly pointed out that without careful experimental and comparative studies it is difficult to identify with certainty which factor is causing the decline, that is, whether it is human harvesting or some other factor influencing population dynamics. Nevertheless, they suggested that all available evidence points to human harvest as the primary driver of tortoise decline during the last twenty-five years. In our study a harvesting pressure was evident from the numerous empty carapaces observed along the roads, but was also indicated by the observation that the proportion of large tortoises (CL>30 cm) decreased significantly with proximity to major roads. Tortoises are on average smaller in harvested areas, since harvesters actively search out and harvest the large adult tortoises (O'Brien 2002).
Although there are some inevitable biases connected to our method for estimating tortoise abundances, other studies in the region, using different survey methods, have produced similar relative abundance estimates. O'Brien (ibid.) made a study west of our area during the wet season and estimates ranged from 15.5 tortoises per ha in an area with no harvesting to 4 to 5 tortoises per ha in areas with some commercial and subsistence harvesting of the tortoise. Lewis (1995) reported densities of up to 10.7 tortoises per ha in south-western Madagascar, whereas Leuteritz (2002) found a mean density of 25.2 tortoises per ha across seven sites in southern Madagascar. O'Brien et al. (2003) reported that in remote regions with no harvesting, tortoises may persist at densities of up to 25 tortoises per ha. In our study the highest relative abundance, 20 tortoises per ha, was found in the preserve Cap Sainte Marie where, at least officially, no harvesting occurs.
The interviews did not indicate that the fady incorporates the tortoise habitat. This may hypothetically have been the case if the tortoise represented strong spiritual values. However, there are numerous small sacred forest patches in Androy that are informally protected (Clark et al. 1998). Some of these ala faly may be upto 5 ha large and subject to complete entry restrictions (Elmqvist, personal obser vation). Tortoise populations found within the limits of these sanctuaries are probably effectively protected, and the role of these sacred forests in maintaining biodiversity throughout the Androy region deserves more detailed studies.
The tortoise occurs within the geographic range of the dry spiny forest; however, it is not clear to what extent decreases in forest cover (Sussman et al. 1994) contributes to tortoise decline. The results of our study suggest that declining forest cover may have relatively minor effects on tortoise viability. The tortoises appear to prefer the open spaces of low shrub and grassland, probably because these environments offer a better food supply than densely forested area. Juvik (1975) also points out that tortoises prefer low vegetation, such as shrub and grassland. Both O'Brien et al. (2003) and Lewis (1995) found that the radiated tortoise persisted at low abundance in relatively pristine forest habitats when exposed to harvest pressure, and at high densities in some regions characterised as heavily degraded open habitats.
It is yet unknown if the extinction of the tortoise would result in changed dynamics for plant species consumed or potentially dispersed. Durrell et al. (1989) reported that radiated tortoises have been observed to feed on at least eighteen plant species, and in our study we recorded at least seven species. The large herbivore Geochelone gigantea, once present in Madagascar, is known to have dramatic impacts on the vegetation dynamics on the Aldadra atoll, where it is still present (Bourn et al. 1999; Merton et al. 1976). G. radiata, which may reach very high densities locally (Leuteriz 2002; O'Brien et al. 2003), is now the only representative in southern Madagascar of a once larger group of reptile herbivores (Arnold 1979; Burleigh and Arnold 1986). To what extent the radiated tortoise, at least partially, maintains crucial functions as a browser, grazer or seed disperser in ecosystems remains to be studied.
At the time of our study, the tortoise's diet consisted almost entirely of the fruits of the cacti Opuntia (O. ficus-indica and O. monacantha). To what extent these widespread introduced species (cover exceeding 20 per cent in some areas in southern Androy [Elmqvist, personal observation]) have increased the capacity of the environment to support tortoise populations deserves further research, par ticularly since the cacti produce an abundance of fruits at a time of the year when there is very little else to feed on.
A large proportion of harvested tortoises probably end up in the illegal markets of the coastal towns of Toliara and Tolagnaro. O'Brien et al. (2003) estimated that annually up to 45,000 tortoises are collected from Toliara alone. The tortoise trade is virtually unchecked, and in Tolagnaro the price varies between 15,000 and 25,000 Malagasy francs (personal communication with informants in Tolagnaro), which equals approximately US$ 2.50 to 4.50. Prices for tortoises reaching the illegal pet market may, however, be considerably higher (Nussbaum and Raxworthy 2000). Despite current official national legislation and international trade regulation, the high exploitation rate of the radiated tortoise is predicted to continue due to very high costs of enforcement. In addition, improvements in infrastructure, such as the repair of existing roads and the construction of new ones, will without doubt lead to intensified tortoise exploitation in presently isolated areas. The fady in its present form and in isolation from other measures will only to a minor degree slow this process. It is important to stress that the institution of the fady does not in itself contain any principle of discouragement other than the regulation of tortoise exploitation by the Tandroy themselves. The fady does not protect the tortoise from exploitation by outsiders, nor from subsistence utilisation by other inhabitants than the Tandroy, that is, there are no rights of exclusion (see Ostrom 1990). This represents a weak institutional framework for the protection and management of species, likely to be vulnerable to even small social and demographic changes. Our study, although limited, suggests that even small increases in the proportion of the human population violating the fady will result in a sharp decline in tortoise abundance.
Measures to prevent a future extinction of the radiated tortoise must nonetheless include the active involvement of local representatives of the Tandroy (and Mahafaly) communities, also incorporating an official recognition of the fady custom as a legitimate institution, for example, through a conversion into a formal bylaw (see Johannes 2002). A number of studies have emphasised the crucial importance of local institutions for improving management of natural resources and ecosystems (Alcorn and Toledo 1998; Berks and Folke 1998; Ostrom 1990; Turner et al. 2000), including reinforcement of specific - species taboos (Horowitz 1998). The rationale for this is simple: environmental regulations that are based on traditional customs and sanctioned by local institutions are more likely to be respected than those imposed by external administrative agencies, even where the regulations themselves are very similar (Berkes and Folke 1998). A transformation of the fady for the purpose of conservation and sustainable management of the tortoise may thus considerably reduce transaction costs, that is, costs for protecting rights and enforcing agreements (Baland and Platteau 1996; Colding and Folke 2001). It has been pointed out that local actors may detect environmental changes faster than external administrative agencies due to close proximity to the resource base and ecosystems, as well as performing such monitoring at significantly lower costs (Baland and Platteau 1996).
Johannes (2002) reports on several cases in Oceania where traditional marine resource management practices such as closed areas, closed seasons and specific species taboos have gore through a renaissance during the past two decades. Factors oontriaitirg to this press irrlu b a gruairp ice of scarcity, a seamed marine tenure system by means of legal recognition, and effective assistance and advice from regional and national governments and NGOs. The process was further reinforced by increased economic incentives due to improved fish stocks and income from tourist-related activities. Ruttan and Borgerhoff Mulder (1999) used a game theoretical approach to analyse conditions under which sound management of common property resources is likely to evolve. They demonstrated that under certain conditions conservation can be the result of narrow self-interest and economically motivated decisions may have the fortuitous but important consequence of conserving resources. Presently, the Tandroy do not consider the tortoise to represent any significant spiritual or economic value. It may, however, constitute an important economic source of revenue if local communities are granted rights to a regulated small-scale tortoise trade, further providing economic incentives for transforming and building effective institutions for sustainable management. In building local incentives for management it is, however, crucial that an institutional transformation also grants local communities the right to exclude outside harvesters in order to avoid the current partial open-access situation. Johannes (2002) reported of one case in Samoa where a strong incentive for local conservation and re-establishment of traditional marine management practices emerged only after rights were given to village councils to exclude outsiders from their fishing grounds. Previously, with a weak or non-existent authority to exclude outsiders, local incentives to conserve marine resources were weak since outsiders could always expropriate the benefits.
Harvest of large adult tortoises (the ones preferred in the meat market) on a sustainable basis is not realistic (O'Brien 2002). Due to the life history characteristics of the tortoise, including slow growth rate and very long-lived individuals, harvest of large adults may easily lead to over-harvesting and rapid population decline (ibid.). On the other hand, it is possible to develop a sustainable locally controlled farming of tortoises where juveniles are sold on the pet market. The economic incentives for entering the pet trade are considerable, since a single juvenile may currently be valued up to US$ 5,000 (Nussbaum and Raxworthy 2000). The prospects for the development of a legal and strongly regulated commercial pet trade based on local farming and body size limits should be thoroughly investigated, including its institutional requirements. This course of action would require an amendment of the radiated tortoise CITES listing from Appendix I to Appendix II as suggested by Nussbaum and Raxworthy (ibid.).
However, there are reasons to believe that local institutions cannot carry out all enforcement and regulation alone, and therefore are in need of assistance from institutions existing at other hierarchical levels and across scales, for example, at the regional and national levels (Johannes 2002). Such cross-scale linkages are referred to as nested institutions. The simplest kind connects local-level management with governmental-level management in partnerships, for example, comanagement. Gadgil et al. (2003) report on several encouraging instances of successful management of ecosystems and species based on local institutions nested with other institutions both vertically (regional and national) and horizontally (other local institutions). In Androy other local institutions, such as fady customs related to lemurs (Propithecus verreauxi and Lemur catta) and forest habitats (ala faly), may likewise be incorporated in nested institutions for sustainable landscape management. There is clearly a large and untapped potential in many regions of similarly trying to incorporate and strengthen local institutions, which may not necessarily have been instruments of resource management in the past. However, based on culturally accepted norms and values they may now be used to fulfil new purposes in conservation and sustainable management of species and ecosystems (Berkes et al. 2000, 2003; Colding and Folke 2000, 2001; Colding et al. 2003; Johannes 2002).
Acknowledgements : We thank Randriamoeliarivony Alfred, Samisoa, Mon Pθre Alonso of the Catholic Mission in Beloha, Zakasoa Dauphiι, Mark Fenn, Jakob Lundberg, and all the helpful and friendly people of the Androy region for contributing to the study. We thank Susan O'Brien for generously sharing her knowledge about the sokake. We also thank Carl Folke, Johan Colding and Per Olsson for valuable comments on an earlier draft. The research was supported by a grant from the Swedish International Development Agency (Sarec/Sida).
| References|| |
|1.||Agar, M.H. (1996), The Professional Stranger. San Diego: Academic Press. |
|2.||Alcorn, J.B. and V.M. Toledo (1998), 'Resilient Resource Management in Mexico's Forest Eco- systems: The Contribution of Property Rights', in F. Berkes and C. Folke (eds), Linking Social and Ecological Systems. Management Practices and Social Mechanisms for Building Resilience, pp. 216-49. Cambridge: Cambridge University Press. |
|3.||Arnold, E.N. (1979), 'Indian Ocean Giant Tortoises: Their Systematics and Island Adaptations', Philosophical Transactions of the Royal Society of London, Series B, 286: 127-45. |
|4.||Baland, J.M. and J.P. Platteau (1996), Halting Degradation of Natural Resources: Is there a Role for Rural Communities? Oxford: FAO and Oxford University Press. |
|5.||Battistini, R. and G. Richard-Vindard (1972), Biogeography and Ecology in Madagascar. The Hague: Dr W. Junk B.V., Publishers. |
|6.||Berkes, F. and C. Folke (1998), Linking Social and Ecological Systems. Cambridge: Cambridge University Press. |
|7.||(2002), 'Back to the Future: Ecosystem Dynamics and Local Knowledge', in L. Gunderson and C.S. Holling (eds), Panarchy: Understanding Transformations in Human and Natural Systems, pp. 121-46. Washington: Island Press. |
|8.||Berkes, F., J. Colding and C. Folke (2000), 'Rediscovery of Traditional Ecological Knowledge as Adaptive Management', Ecological Applications, 10: 1251-62. |
|9.||(2003), Navigating Social-Ecological Systems: Building Resilience for Complexity and Change. Cambridge: Cambridge University Press. |
|10.||Bloch, M. (2001), 'The Ethnohistory of Madagascar', Ethnohistory, 48: 1-2. |
|11.||Bourn, D., C. Gibson, D. Augeri, C.J. Wilson, J. Church and S.I. Hay (1999), 'The Rise and Fall of the Aldabran Giant Tortoise Population', Proceedings of the Royal Society: Biological Sciences, 266: 1091-100. |
|12.||Buckland, S.T., D.R. Anderson, K.P. Burnham and J.L. Laake (1993), Distance Sampling: Estimating Abundance of Biological Populations. London: Chapman and Hall. |
|13.||Burleigh, R. and E.N. Arnold (1986), 'Age and Dietary Differences of Recently Extinct Indian Ocean Tortoises (Geochelone sensu lato) Revealed by Carbon Isotope Analysis', Proceedings of the Royal Society of London, Series B, 227: 137-44. |
|14.||Caccone, A., G. Amato, O.C. Gratry, J. Behler and J.R. Powell (1999), 'A Molecular Phylogeny of Four Endangered Madagascar Tortoises Based on MtDNA Sequences', Molecular Phylogenetics and Evolution, 1: 1-9. |
|15.||Clark, C.D., S.M. Garrod and M. Parker Pearson (1998), 'Landscape Archaeology and Remote Sensing in Southern Madgascar', Journal of Remote Sensing, 19: 1461-77. |
|16.||Colding, J. and C. Folke (1997), 'The Relations among Threatened Species: Their Protection, and Taboos', Conservation Ecology (online), 1: 6 (http://www.consecol.org/vol1/iss1/art6). |
|17.||(2000), 'The Taboo System: Lessons about Informal Institutions for Nature Management', Georgetown International Environmental Law Review, 12: 413-45. |
|18.||(2001), 'Social Taboos: "Invisible" Systems of Local Resource Management and Biological Conservation', Ecological A,zplications, 11: 584-600. |
|19.|| Colding, J., C. Folke and T. Elmqvist (2003), 'Social Institutions in Ecosystem Management and Biodiversity Conservation', Tropical Ecology (in press). |
|20.||Dewar, R. and J. Wallis (1999), 'Geographical Patterning of Interannual Rainfall Variability in the Tropics and Near Tropics: An L-moments Approach', Journal of Climate, 12: 3457-66. |
|21.||Durrell, L., B. Groombridge, S. Tonge and Q. Bloxam (1989), 'Geochelone radiata-Radiated Tortoise, Sokake', Occasional Papers of the IUCN Species Survival Commission (SSC), 5: 96-98. |
|22.||Gadgil, M., C. Folke and F. Berkes (1993), 'Indigenous Knowledge for Biodiversity Conservation', Ambio, 22: 151-56. |
|23.||Gadgil, M., P. Olsson, F. Berkes and C. Folke (2003), 'Exploring the Role of Local Ecological Knowledge in Ecosystem Management: Three Case Studies', in F. Berkes, J. Colding and C. Folke (eds), Navigating Social-Ecological Systems: Building Resilience for Complexity and Change, pp. 189-209. Cambridge: Cambridge University Press. |
|24.||Glaw, F. and M. Vences (1994), A Field Guide to the Amphibians and Reptiles of Madagascar. Koln, Germany: M. Vences and F. Glaw Verlags GbR. |
|25.||Hilton-Taylor, C. (comp.) (2000), 2000 IUCN Red List of Threatened Species. Gland, Switzerland, and Cambridge, UK: IUCN. |
|26.||Horowitz, L. S. (1998), 'Integrating Indigenous Resource Management with Wildlife Conservation: A Case Study of Batang Ai National Park, Sarawak, Malaysia', Human Ecology, 26: 371-404. Johannes, R.E. (2002), 'The Renaissance of Community-based Marine Resource Management in Oceania', Annual Review of Ecology and Systematics, 33: 317-40. |
|27.||Juvik,J.O.(1975),'The Radiated Tortoise of Madagascar', Oryx, 13:145-48. |
|28.||Koechlin,J.(1972),'Flora and Vegetation of Madagascar', in R. Battistini and G. Richard-Vindard (eds), Biogeography and Ecology in Madagascar. The Hague: Dr W. Junk B.V., Publishers. Kvale. J. (1996), Interviews: An Introduction to Qualitative Research Interviewing. Sage Publications . |
|29.|| Leuteritz, T. E. (2002), 'Distribution, Status, and Reproductive Biology of the Radiated Tortoise, Geocheloneradiata (Shaw, 1802) in South west Madagascar', unpublished Ph. D. Thesis. Fairfax, Virginia: George Mason University. |
|30.||Lewis, R. (1995), 'Status of the Radiated Tortoise (Geochelone radiata)', unpublished report. Madagascar: WWF. |
|31.||Merton, L. F. H., D. M. Bournand R. J. Hnatiuk (1976), 'Giant Tortoise and Vegetation Interactions of Aldabra Atoll: PartI-Inland', Biological Conservation, 9:293-317. |
|32.||Nussbaum, R.A. and C.J. Raxworthy (2000), 'Commentary on Conservation of "Sokake", the Radiated Tortoise (Geochelone radiata) of Madagascar', Amphibian and Reptile Conservation, 2: 6-14. |
|33.||O'Brien, S. (2002), 'Population Dynamics and Exploitation of the Radiated Tortoise Geochelone radiata in Madagascar', unpublished Ph.D. thesis. Cambridge: Darwin College, University of Cambridge. |
|34.||O'Brien, S.E.R. Emahalala, V. Beard, R.M. Rakotondrainy, A. Reid, V. Raharisoa and T. Coulson (2003), 'Decline of the Madagascar Radiated Tortoise (Geochelone radiata) due to Overexploitation', Oryx, 37: 338-43. |
|35.|| Olson, D.M. and E. Dinerstein (1998), 'The Global 200: A Representation Approach to Conserving the Earth's Distinctive Eco-regions', Washington, DC: Conservation Science Programme, WWF. |
|36.||Olsson, P. and C. Folke (2001), 'Local Ecological Knowledge and Institutional Dynamics for Ecosystem Management: A Study of Lake Racken Watershed, Sweden', Ecosystems, 4: 85-104. |
|37.|| Ostrom, E. (1990), Governing the Commons: The Evolution of Institutions for Collective Action. Cambridge: Cambridge University Press. |
|38.||Parker Pearson, M. (1997), 'Close Encounters of the Worst Kind: Malagasy Resistance and Colonial Disasters in Southern Madagascar', World Archaeology, 28: 393-417. |
|39.|| Richard, A.F., R.E. Dewar, M. Schwartz and J. Ratsirarson (2002), 'Life in the Slow Lane? Demography and Life Histories of Male and Female Sifaka (Propithecus verreauxi verreauxi)', Journal of Zoology, 256: 421-36. |
|40.|| Ruttan, L.M. and M. Borgerhoff Mulder (1999), 'Are East African Pastoralists Truly Conservationists?', Current Anthropology, 40: 621-53. |
|41.|| Ruud, J. (1960), Taboo: A Study of Malagasy Customs and Beliefs. Oslo: Oslo University Press. |
|42.||Sussman, R.W., G.M. Green and L.K. Sussman (1994), 'Satellite Imagery, Human Ecology, Anthropology and Deforestation in Madagascar', Human Ecology, 22: 333-54. |
|43.|| Sutherland, W.J. (2000), The Conservation Handbook: Research Management and Policy. Oxford: Blackwell Science. |
|44.||Turner, N.J., M. Boelscher Ignace and R. Ignace (2000), 'Traditional Ecological Knowledge and Wisdom of Aboriginal Peoples in British Columbia', Ecological Applications, 10: 1275-87. |
[Figure 1], [Figure 2], [Figure 3], [Figure 4]
[Table 1], [Table 2], [Table 3], [Table 4]