Investigation of the potential for the utilisation of traditional ecological knowledge in the development of community-based resource management and conservation strategies in Guyanese Amerindian communities, with particular reference to human-animal interactions.

A proposal for pre-doctoral research by Thomas Henfrey, University of Kent at Canterbury, to be conducted under the APFT (L'Avenir des Peuples des Forêts Tropicales / Future of Tropical Forest Peoples) programme.


It is proposed to undertake a two-year programme of research which will investigate the potential for using the traditional ecological knowledge of Amerindian peoples in the design of local programmes for the conservation and management of natural resources. This will involve, firstly, the anthropological study of human perceptions and knowledge of the natural world, and human uses of natural resources, refering particularly to the ways in which humans exploit and otherwise interact ecologically with wild animals. The results of this phase of research will be used to inform a subsequent year-long programme of ecological research which will seek to address the impacts of current and proposed human activities within the reservation of the host community upon forest ecology in general, and populations of wild fauna in particular. It is hoped that this will produce results relevant to the design of a local plan for resource management addressing both socio-economic and conservation needs within a wholly participatory and bottom-up framework. The exact site for the research will be determined on the basis of extensive discussions with representatives of the community involved; it is hoped that a situation appropriate to the undertaking of such a research programme may exist among the South Rupununi Amerindian communities. The Researcher also hopes to spend some time in Orealla, and involve this community in the exercise.

1. Guyana

1.1 Amerindians in Guyana

Officially, Guyana is home to 9 Amerindian tribes, although some tribal designations refer to the descendants of several, formerly linguistically distinct, groups - notably the Wai-wai [Yde 1960: 84, Mentore 1995: 20] and Wapishana [Farabee 1918: 4]. The majority are of the Carib linguistic branch - true Caribs1, Akawaio, Patamona, Arecuna, Makushi and Wai-wai - coastal Arawaks (more accurately termed Lokono) and Wapishana speak Arawakan languages, and the Warrau are Guyana's sole representatives of the Warrau branch. The most recent census showed Amerindian people to number around 47,000 in Guyana, around 8% of the country's total population [Forte 1990a]. However, the concentration of the majority of the non-indigenous population on the coast means that Amerindians form a demographic majority in many parts of the interior. Despite this, industrial development in the interior has tended to by-pass Amerindian populations, and rarely been designed to cater for their needs [La Rose 1994]. Though this situation is being remedied with a greater focus on consultation with Amerindians in current development programmes [eg. Bishop 1996], the problem of ensuring their full participation in and benefit from the changes that are taking place in Guyana remains. The need for this is compelling - though no comprehensive economic surveys have been performed, conventional economic indicators suggest that Amerindian poverty is a continuing phenomenon [Forte 1993: 6-8]. Although the use of such indicators has been criticised on the grounds that the local economy in most Amerindian communities is largely subsistence-based [Fox & Danns 1993: 99-100], the truth is that Guyanese Amerindians have access to few economic opportunities. The current education system seems ill-equipped to provide a solution to this - still largely based on the European model and hardly designed with the realities of life in the interior in mind, it is more likely to alienate the most able students from their home villages than equip them to tackle the problems faced there [ibid.: 9]. The issue of land rights is a burning one, as it has been since the early days of colonial occupation of the interior. The Amerindian Act, chapter 29 of the laws of Guyana, guarantees all Amerindian communities usufruct rights to above-ground resources in reservations designated in their areas of occupation. In practice, however, there are a number of communities and settlements without land title, many communities are dissatisfied with the extent of their titled lands. In addition, only a single community, Orealla, has had the boundaries of its reservation demarcated, making the delineation of Amerindian lands ambiguous and the lands potentially vulnerable to encroachment by outsiders [Forte 1995a]. The recent involvement of Orealla with the Barama company, the product of which has been a local ecological and economic disaster, demonstrates a major weakness of the reservation system, which while providing a large degree of autonomy to Amerindian communities may also cut them off somewhat from central sources of advice and assistance, exposing them to such predatory incursions from outside parties [Henfrey 1995].

On the other hand, the situation of Guyanese Amerindians appears to be largely favourable relative to that of many of their neighbours, current circumstances and recent experiences appearing to better those of Amerindian peoples in Brazil and Venezuela [eg. Foster 1990, Cleary 1990, MacMillan 1995: 122-126]. They are also currently the subjects of an unprecedented volume of outside attention. Numerous foreign NGO's are currently operating projects in Amerindian communities, including Conservation International , Red Thread, Voluntary Service Overseas, Youth Challenge International and operation Raleigh. Guyana currently has no less than four separate Amerindian organisations - Guyana Organisation of Indigenous Peoples (GOIP), Amerindian People's Association (APA), the National Amerindian Council (NAC), and The Amerindian Action Movement of Guyana (TAAMOG). These organisation have arisen in recent years largely, apparently largely as a result of the growing sense of Amerindian identity in Guyana and throughout the Americas. Aside from the Wai-wai and a few scattered groups from other tribes, Guyanese Amerindians are conventionally thought to be highly acculturated. Though this concept, under its strict definition, refers to any process of cultural change resulting from prolonged contact between previously isolated cultures, it is more commonly used to refer to the assimilation by Western culture of tribal peoples, a process which tends to produce convergence among previously isolated societies [eg. Steward 1977]. The uncritical application of this latter sense is dubious, however, as it obscures the inherent and essential dynamism of cultural systems, as well as the fact that distinct societies tend to incorporate western culture in idiosyncratic ways2 . Despite having changed markedly over the last five centuries, Guyanese Amerindians seem to have largely retained their senses of tribal identity, as well as adopting a sense of unity as Amerindian peoples whose growing strength is reflected in the proliferation of interest groups referred to above.

1.2 Conservation in Guyana

Recent years have seen Guyana assume a position of huge importance in the global forest conservation debate. It is estimated that forest, much of it pristine, still covers 80% of the country [Ek 1996: 15]. Over 1000 tree species have been recorded, in a total flora in excess of 8000 species, 30% of which are estimated to be endemic, and 1063 species of vertebrates [Conservation International 1997]. A recent report highlighted Guyana as one of only 8 countries worldwide in which a realistic potential for large-scale sustainable management of forest resources still exists [World Resources Institute 1997].

Conservation interest in Guyana is a relatively new phenomenon, as it is only within the last decade that the country has been freely accessible to outsiders, but largely because there has, until recently, been very little pressure on the country's abundant natural resources. Economic stagnation under the Burnham regime led to the decline of the forestry and mining industries throughout the 1980's [Thomas 1993: 142, 146]. Coupled with the low population density in the interior, and consequent lack of agricultural pressure on land, this resulted in the country retaining almost all of its forest cover into the present decade. However, this situation has been changing since 1989, and the advent of economic liberalisation under a programme of structural adjustment promoted by the World Bank and IMF [Colchester 1994: 45]. A pressing need for new sources of revenue and foreign exchange has led attention to be turned to the vast natural resources of the interior. It is estimated that over 80% of previously unexploited state-held forests have been alienated in Timber Sales Agreements since 1985, mainly to foreign companies [Forte 1995b: 4]. International concern over the future of Guyana's forests was by 1995 sufficient to provoke the World Bank and Britain's Overseas Development Administration into expressing concern over the matter to the Guyanese government, and the ODA to make the granting of a loan for the strengthening of Guyana's forest department conditional upon a three-year moratorium on the handing-out of new logging concessions [Colchester 1997: 113-4]. Recently, however, several foreign timber companies have been granted 'exploratory leases', agreements based upon recently-introduced legislation according to which companies may prospect an area as a precursor to the possible future commencement of logging. The areas involved greatly extend the area of state forest (that over which the Guyana Forestry Commission has legal control), a matter of grave concern to environmental and human rights groups [eg. GHRA 1997], despite governmental assurances that environmental protection, sustainability and human rights will key factors in the operation of any concessions eventually given [Chandarpal 1997]. It has been suggested that, for a number of reasons, the economic benefits to Guyana from such rampant expansion of the timber industry are likely to be slight, and the potential environmental and social costs great [Colchester 1993]. Guyana's relatively late entrance into the global economy has provided a great opportunity to learn from the mistakes that have been made in other South American nations. For example, 10% of forests in Brazilian Amazonia have been cleared in the last 20 years, at a total investment cost of 15 billion dollars, but the overall effect on human welfare has been a decidedly negative one [Moran 1993: 159-160]. Evidence from this and other cases highlight the inappropriateness of the conventional development paradigm for the situation of countries such as Guyana; the need for attention to be turned to the exploration of alternatives which would allow the need for long-term conservation of natural resources, and their exploitation for much-needed economic development that would benefit the whole of Guyana's population, to be reconciled is pressing [Sizer 1996].

2.1 Guyanese and Related Literature Consulted

There is some ambiguity in trying to determine the appropriate geographical focus for a review of the Guyanese ethnographic literature. On the one hand, Guyana is only a part of the much broader Guiana shield area (hereafter referred to as 'Guiana'), which has been identified as forming a distinctive unit from the perspectives of culture [Steward 1946: 50, Riviere 1984: 2-4], biogeography and geology. On the other hand, the pattern of exogenous cultural influences upon the indigenous inhabitants of the area now forming the Guyanese state has been unique since the earliest stages of European contact. In the modern context, therefore, this area does, to a certain extent, form a meaningful unit of cultural analysis. However, the distributions of the majority of Guyana's Amerindian tribes cross national boundaries, and there exist extensive relationships of consanguinity, affinity, and commerce with neighbouring areas of Brazil, Venezuela and Suriname. I have therefore concentrated largely on the literature from Guyana itself, incorporating that from adjacent areas selectively, as relevant.

Several historical texts have provided useful background information. Vere Trevalyn Daly's [1975] textbook covers the entire period from first European contact to the time of publication, and has been the most useful general historical text. Texts focussing on particular periods include Spinner's [1984] work on the mid to late 20th. century. The early colonial period is covered by a historical work of Thompson [1987], who focuses mainly on colonial issues during the time of Dutch occupation, 1580-1803, but incorporates such historical information on Amerindians at the time as is available. Guyanese historian Mary Noel Menezes has worked extensively on the situation of Amerindians during the 19th. century, though her work is also largely concerned with their relationships with the colonial regime [Menezes 1979].

Contemporary accounts of some ethnographic value began to emerge from diverse sources during the period covered by Menezes, and provide a useful complement to her work. Among the earliest of these are Robert Schomburgk's accounts of his extensive exploratory journeys [Schomburgk 1922, 1923]. These include numerous anthropological observations, the majority on the Macusi people with whom he spent the most time but covering several tribes, along with a wealth of information on natural history, and are refreshingly devoid of the ethnocentric arrogance typical of the age. This is not the case with William Brett's account of his missionary labours amongst numerous tribes [Brett 1868], although he also recorded information of significant ethnographic value, nor with the work of reputed champion of Amerindian causes William Hillhouse [Hillhouse 1979]. Everard Im Thurn's account of his anthropological expeditions into the interior is the earliest piece of writing of which I am aware which is based on specifically anthropological research; the writer's botanical inclinations make this, like Schomburgk's work, particularly useful in providing early documentation of human exploitation of wild species [Im Thurn 1883]. None of these are ethnographies in the modern sense, and all are highly prone to make unfounded generalisations across, and even among, tribes. Nonetheless, they are invaluable and unique records of life among Guyanese Amerindian people whose lifestyles had not yet undergone radical change in response to European contact. More comprehensive and in-depth studies were undertaken in the early decades of the twentieth century - the tireless efforts of Walter Roth resulted in the production of an epic account of Amerindian material culture [Roth 1924, 1929], and a large anthropological expedition of several years duration was conducted by the University of Pennsylvania [Farabee 1918, 1924]. These lack the rigour of modern ethnography - they are pre-Malinowskian in method (Farabee apparently relied largely on interviews with a small number of informants) and include little in the way of quantitative ethnographic data - but nonetheless remain of relevance in the present day.

Modern-day ethnographic study in Guyana began with Gillin's 6-month study of the Barama river Caribs [Gillin 1936], and continued with Iris Myers' wartime work on culture contact among the Macusi [Myers 1944], and later Audrey Butt's 2-year doctoral study of the Akawaio [Butt 1954]. The bias towards study of the Carib-speaking tribes continued in subsequent years, particularly after the establishment of missionaries in Wai-Wai country facilitated access by anthropologists to the last remaining Guyanese tribe living according to their traditional pattern. Two Wai-Wai studies quickly followed [Fock 1963, Yde 1965], and in the late seventies George Mentore conducted a comprehensive, largely structuralist analysis of Wai-Wai social organisation, producing several papers of interest in the present study [Mentore 1983-4, 1995]. The Wai-Wai remain the best-studied of Guyana's Amerindian tribes, but the country as a whole remained largely underattended, aside from Kathleen Adams' follow-up to Gillin's Carib research [Adams 1981]. This is reflected in the small amount of Guyanese data incorporated into several summary works which appeared over a short space of time. A compendium of research on Carib tribes included a little Guyana information [Basso 1977], as did a volume of Antropologica dedicated to Guianese Carib tribes, concluded and summarised by Butt-Colson [1984], and Peter Riviere's synthesis of the ethnography of the Guianas [Riviere 1984]. The Arawak tribes were even less well-attended, and remained barely studied throughout the Guianas as a whole. Two doctoral research projects focused on ethnicity and relations with broader Guyanese society among coastal Arawaks (Lokono) and 'true' Caribs on the Pomeroon river [Drummond 1977] and in the mixed Arawak-Warrau communities of the Correntyne [Sanders 1987]. These are of little direct relevance to the current project, particularly Sanders' monograph, a particularly poor example of Geertzian interpretivism, and condemned in the study community for its inaccurate portrayal of life there. More recently, Gloria Tang has produced a doctoral thesis concerning development in Region 9 of Guyana, based largely upon research conducted in the large Wapishana village of Aishalton [Tang 1995]. On the basis of extended interviews with Aishalton residents on their perception of and attitudes towards development, and case studies of two specific development initiatives in the area, she criticised the application of top-down development in the region. Despite the presence of a broad natural resource base, development projects in the area had floundered due to the lack of consultation with and participation of the local people, and deficiencies in the local infrastructure. She advocated the implementation of a bottom-up development strategy based on local natural resources, combining community-based initiatives with appropriate centrally-implemented schemes to improve infrastructure and enhance the functioning of the regional administration. The former point is a key feature of the present project, which aims to investigate the use of natural resources from an indigenous perspective with a view to assessing the potential for making this the basis of projects in community-based sustainable development.

The recent upsurge of academic interest in Guyana's Amerindians has occurred largely under the direction of the University of Guyana's Amerindian Research Unit, which began life as a purely linguistic project but now operates with a vastly broader remit. The ARU's own publications are the most important sources of recent information, most pertinent of which are a report on Wapishana material culture [ARU 1993], and background papers on various aspects of the Amerindian situation [Forte 1993, 1995a, 1995b]. Other relevant ARU material includes a survey of the status of Amerindian women and children [ARU 1993], a collection of papers concerning Amerindian use of forest resources [ARU 1995], and ongoing demographic studies in Amerindian communities throughout Guyana [Forte 1990a]. There has also been a relative proliferation of new publications concerning development issues in the country, including a summary of a government-sponsored investigation into how economic exploitation of forest resources may be reconciled with the need for conservation [Sizer 1996]. A recent book by Marcus Colchester is concerned with the issues surrounding development in Guyana's interior [Colchester 1997]; despite being less than objective in certain respects, this is of great use as the most up-to-date summary of the threats faced by the country's indigenous peoples. More balanced is a recent collection of papers, mostly by Guyanese writers, on development and land use in both coast and interior [Williams et al 1997], whose contrast in approach makes it complement Colchester's work well.

The literature on the Brazilian state of Roraima is very sparse, but includes a few items which are informative about the situation of Amerindians there, and of relevance to the present study. Most notable of these in the present context is Nancy Fried Foster's doctoral dissertation on Brazilian Wapishana identity [Foster 1990], which deals largely with relationships with Brazilian society, and includes more ethnographic information than the author gives herself credit for, although very little of this is quantitative. Peter Riviere's [1972] ethnography of Roraimense cattle ranchers includes some information on the position of acculturated Amerindians within the unique social system of the state; although the pace of change in Roraima over the thirty years since his study means it has dated rather rapidly, it is perhaps not unreasonable to suppose that many of the attitudes and perceptions governing interracial relationships still endure. A recent volume of papers on Roraima [Furley 1995] provides much-needed background information, some of which is relevant to the Rupununi; one chapter in this documents the history of Amerindian-settler relations, and updates Riviere's coverage of this somewhat {Hemming 1995].

2.1.1 General Models of Social Organisation

Basso, Butt-Colson and Riviere all use comparative methods in attempting to generalise about Carib or Guianese social organisation. Basso [1977: 17] identified several features common, with a small number of exceptions, to all the Carib tribes: cultivation of bitter manioc as a staple, cognatic evaluation of kinship relations and social categorisation, the practice of Shamanistic rituals and of communal commemorative ceremonies, the principal use of tobacco as a narcotic, a 'shadow' conception of the soul, and belief in menstrual pollution with which is associated with exclusion of pubescent girls (although the latter two are widespread in Amazonian groups). Riviere's focus is the Guiana culture area rather than the Carib linguistic group, but the limitations of the available ethnographic information restrict his focus to the Carib tribes. He identifies cognatic descent, two-line prescriptive relationship terminology, preference for either settlement endogamy and/or uxorilocal residence, the importance of co-residence as a factor in ordering relationships, and the small size and short duration of settlements as common features of their social organisation [Riviere 1984: 4]. Most sophisticated of these three analyses is that of Butt-Colson [1984], who describes a model of a three-generational cycle over which social formations are reproduced through the institutions of the extended family and the joint family, both products of the ideal of matrilocality and the attendant residential immobility of women. The paradigmatic extended family is a 3-generation line of female consanguines with both vertical and horizontal kinship relationships who, although not forming a corporate group in the strict sense of the word, act as an informal collaborative matriline about which men circulate. The ideal of cross-cousin marriage recreates in consecutive generations the brother-sister pairing, and in alternate generations the marital union, a couple who conform to this ideal having common grandparents. The system is thus cyclical, men marrying out of their natal matriline and their sons marrying back in. The joint family is based on the adherence of brother's-in-law via fraternal collaboration resulting from marrying in to the same matriline. This is exemplified by the ideal of marriage between two brother-sister pairs, resulting in a direct exchange of men between lineages, in which case the fraternal relationship is classificatory rather than real; overall there appears to be a great deal of conflation of consanguinity, affinity and co-residence in such systems, in the interests of achieving greater conformity to the stated ideal. This model is accepted as being representative of Guianese indigenous societies, and comprises a suite of features found widely throughout Amazonia [Viveiros de Castro 1996: 188].

2.1.2 Wapishana History and Ethnography

The Wapishana3 were first encountered by Europeans when Dutch traders ascended the Essequibo to its headwaters in the early 18th. century, at which time they were reported to occupy the entire savanna country south of the rivers Takutu and Uraracuera [Farabee 1918: 13]. The present-day population is scattered over a large area of savanna in south-west Guyana and the neighbouring Brazilian state of Roraima, which conforms to the mid-19th. century report that the Wapishana, Atorai and Macusi between them occupied an area between the Essequibo and the Rio Branco [Brett 1868: 341]. The Guyanese population occupies the south Rupununi area and has most recently been estimated to consist of around 6,000 people in ten main villages and numerous dispersed smaller settlements [Forte 1990a], although persistent cross border migration, as well as the passage of time, make this figure a somewhat unreliable indicator of present numbers. A single group, estimated at around 100 people, is reported to live in the forests of the Sierra Acaria, between the headwaters of the Essequibo and Takutu, and refuse all outside contact [Bahuchet n.d.: 111]. The modern-day Wapishana nation is thought to incorporate several tribes, absorbed over time by intermarriage - Farabee was informed that the Karapieu, Ilieu, Tarewinpidian, Saparas, Powisien, Inkerikub and Paravilhanas had all encountered this fate prior to his visit [1918: 14]. He also reported the ongoing assimilation of the Taruma and Atorai tribes, which existed as small but distinct groups at the time, a process which was almost complete by the late 1980's [Forte 1990a: 2]. The nature of the modern-day Wapishana language reflects this - pure Wapishana is spoken by only a handful of isolated Brazilian groups, while the remainder speak an amalgam of Wapishana and Atorai [ARU 1992: 2]. Preceding Farabee, Im Thurn additionally mentioned the Amaripa tribe (unlikely to be a synonym for Taruma, who are described as a separate tribe) which, along with the Atorai, he described as 'indistinguishable in habit' from the Wapishana, with whom they shared settlements. He also mentioned two other tribes resident on the savanna, to the south of Wapishana territory - Maopityans and Pianoghotto [Im Thurn 1883: 169-171], neither of which appears still to exist as a distinct group.

2.1.3. Social Organisation, Settlement and Subsistence

There has been no modern-day standard anthropological study of the Wapishana, and ethnographically the tribe remains one of the lesser-known in the Guiana region [Riviere 1984: 5]. Such reports as exist of Wapishana social organisation are somewhat contradictory - Farabee described it as patrilineal and patrilocal, marriage being exogamic - though the exogamic unit is not specified, it is presumably the village, as cross cousin marriage is reported to be the ideal [Farabee 1918: 93]. More recent studies of Brazilian Wapishana contradict this: post-marital residence was described by Lucila Herrmann as being matrilocal [cited in Riviere 1984: 39], while Foster found that there was no fixed rule for the inheritance of tribal identity, although there was some tendency to matrilineality [Foster 1990: 149]. The importance of matrilocality in the models described above make these differences crucial in terms of the conformity or otherwise of the Wapishana system. Since both are based upon studies of Carib tribes, the grounds for assuming Wapishana conformity without empirical support are highly suspect. There is a virtual absence of data on Lokono Arawaks - early reports suggest they were organised on the basis of numerous, individually named, exogamous, matrilineal groups, but this is based on observations of people already highly affected by European contact [Brett 1868: 98] - the capacity for inference about the Wapishana from this is slim. In addition, there is indirect evidence that the social system has changed over time as a result of contact with neighbouring tribes. Layrisse and Wilbert, for example, suggest modern-day kinship terminology to have been adapted from an earlier system similar to those of Hawaiians or Inuit in order to facilitate intermarriage with neighbouring Carib tribes [Riviere 1984: 44-5]. If this were true, it most likely reflects more profound changes resulting from intertribal contact, but in the absence of the information necessary to reconstruct either the historical situation or the pattern of change this remains speculation. Foster points out that Riviere's model also takes little account of the effects of non-Amerindian contact [Foster 1990: 75-80]. She herself found the degree of acculturation to be a major dimension of variation among the settlements she worked in, and as early as the last century the extent to which Wapishana people were affected by European contact was reported to be widely variable [ibid.: 113-4]. Foster observed this both between villages, depending on accessibility, and also within villages, residents of peripheral areas tending to retain traditional habits to a greater extent than those in the village centres [ibid.: 131]. One trait in which this is the case is in relationships with parents-in-law. The extent to which people have retained the traditional system of bride-service, facilitated by temporary uxorilocality, diminished in both frequency and in duration of service with degree of acculturation [ibid.: 191]. I would suggest that the effects of acculturation between Wapishana and other Amerindian tribes might have been similarly non-uniform, and that this is reflected in the contradictions between various reports mentioned above. This being the case, it would be inaccurate to postulate the existence of a unique pattern of Wapishana social organisation.

The modern settlement pattern tends to consist of clusters of related households, often separated from others within the same legally demarcated village by several miles [ARU 1992: 4, Foster 1990: 129]. Riviere observed shelters in the forest close to the farms sufficiently substantial to suggest some duration of settlement there [Riviere 1984: 24]. This is confirmed by the finding that Wapishana in both Brazil and Guyana usually have a second - or in some cases sole - dwelling near the farm [Foster 1990: 130, Tang 1995: 29]. Polygamy is reported to have been practiced [eg. Im Thurn 1883: 223], but at a low frequency compared with other tribes [Schomburgk 1923: 32]; it is unlikely the practice continues, as it is not mentioned in any of the more recent texts. Agriculture continues to be the major subsistence activity, and is concerned mainly with numerous varieties of cassava; yams, sweet potatoes, rice, eddoes, dasheen, sugar cane, corn, peppers, pumpkin, banana, watermelon, papaya, pineapple and mango are also grown [ARU 1992: 6-7]. Foster reports that much agricultural labour is performed collectively. Clearance of fields is performed by all-male work parties assembled by household heads via the provision of kari, a drink with apparently invigorating qualities produced by the fermentation of cassava, and usually a meal at the end of the working day. The process is repeated several weeks later following burning. Both men and women participate in subsequent tasks such as planting, which is finished by the start of the rains in late May/early June, and the farm is productive by September [Foster 1990: 158-161]. Hunting, fishing, gathering forest foods, and rearing livestock are the other important subsistence occupations [ARU 1992: 7-13]. Foster observed men going out to hunt or fish every 2-3 days, with variable success depending upon the local population density and proximity to non-Indian settlements. It is possible that the frequency of hunting was greater in the past - Schomburgk [1923: 32] makes mention of the abundance of hunting trophies, and of dogs, in the Wapishana settlements he visited. Schomburgk also identifies many of the species of plant foods collected by Amerindians in the South Rupununi. Some of his observations were of his Macusi guides, but it is unlikely that the food sources they used would be ignored by Wapishana. These include the use of the fruits of Eugenia cauliflora , Psidium turbiniflorum , and P. pomiferum for making drinks, and the consumption of the berries of a species of Malphigia [ibid: 7-12]. The fruits of Melicocca bijuga, Genipa marianae, and G. edulis were reported to be collected and eaten specifically by Wapishana [ibid: 36], along with those of Maximilia regia [ibid: 53]. Recent declines in these activities may have been enforced in some areas - the depletion of wild plant and animal resources by settler activity was condemned in the Brazilian villages of Baxio and Medio within the context of a clear ethic of conservation and sustainability [Foster 1990: 163-166]. It is hoped that the present study will provide a more substantial picture of the ecological relations of Wapishana subsistence, and some information on the social context in which they operate.

Im Thurn described Wapishana, Atorai and Amaripa as being highly active in inter-tribal trading - including salt collected from natural sources in the area and cassava graters manufactured by the Wai-Wai - and general intermediation. He accorded them further distinction as the great canoe-builders among the interior tribes, and also by the use of farine rather than cassava bread [Im Thurn 1883: 169, 263, 265, Farabee 1918: 21]. The Taruma were reported to be the best breeders of hunting dogs, hybrids of domestic dogs introduced by the Spanish and either or both of the indigenous species of canid [ibid.: 232].

2.1.4 Ecological Relations

Though mostly resident on the savanna, and commonly referred to as Guyana's savanna tribes, the Wapishana, like their northern Macusi neighbours, are perhaps more appropriately viewed as occupying the savanna-forest ecotone [eg. Potter 1993: 2], their dependence on forest resources being complete. In particular, the infertile soils of the savannas are unsuitable for swidden agriculture, and farm plots are located in nearby forests [ARU 1992: 3]. The majority of these are in the headwaters of the Essequibo to the east of the south Rupununi region, which support deciduous and semi-deciduous forests continuous with the main Amazonian hylaea [Myers 1936, Ducke & Black 1953, both in Eden 1986]. The villages of Shea, Maruranau, Aryan, Aishalton, Karaudanawa and Achiwuib are all located within a few kilometres of the forest-savanna boundary in this location. Sand Creek and Potarinau are located at the edge of the forested foothills of the Kanuku mountains, while residents of Sawariwau, the only village located remote from the forest boundary, rely on 'bush islands', isolated forest patches within the savanna [Eden 1986: 259-61]. Most hunting and some fishing also occur within the forests, although the latter activity is also performed on the savanna, in ephemeral wet-season pools or the Rupununi river and its tributaries [ARU 1992: 12]. The collection of timber and balata (Manilkara bidentata) for both domestic and commercial purposes were once major economic activities, though both have declined in volume considerably in recent years [ibid. : 42-3]. The trapping of live birds for the export market is nowadays a major commercial activity, involving men from every Wapishana village. Official figures show one-third of the 24,000 birds legally exported from Guyana in 1988 to have come from the Rupununi area, and the extent of the illegal - and therefore unregulated - trade is not known. In particular, the village of Sand Creek appeared at this time to have become something of a focal point for trappers migrating from more westerly villages remote from the forest habitats of valuable parrot and macaw species, prompting the toushau (village captain) to comment upon the need for the introduction of monitoring and improved regulation to prevent depletion of the populations of these birds [ibid. : 41-2]. This reflects a more general concern in the area with the management of forest resources - in a survey performed by the Carter Centre as a part of study on land use planning in Guyana, South Rupununi residents consulted at Aishalton expressed the need for the training of local people in the management and use of forest resources [Bishop 1996: 104-6]. A major objective of the present study is to attempt to identify ways in which this might be achieved within a framework largely constructed around the knowledge and ideas already existing within these communities, advancing the notion of participation to its fullest extent.

2.1.5 Outside Influences: Cattle, Mining and the Cash Economy

The Wapishana have been in sustained contact with the outside world for over a century, and their lives are affected by numerous outside influences. Foremost of these in both duration and magnitude is cattle ranching, pioneered by the Scotsman Harry Melville who settled in the area in 1891, marrying two daughters of an Atorai chief [ARU 1992: 14]. Melville played a crucial and possibly beneficial role in mediating relations between the inhabitants of the south savannas and the colonial regime, hindering the permanent establishment of traders and missionaries, and providing what Farabee described as a "wholesome influence" [Farabee 1918: 15]. Nonetheless, Walter Roth, on returning to the villages of Sand Creek and Wawanawa in 1925, a decade after his first visit, noticed a dramatic drop in living standards, manifest in the prevalence of drink, neglect of agriculture, adoption of European-style dress, and declines in standards of personal hygiene since 'patent preparations have replaced good old crab oil'. This he attributed to growing involvement in the cash economy, particularly the balata industry [Roth 1929: vii-viii]. By 1952, both Macusi and Wapishana were reported to have abandoned many of their traditional activities, although the native languages had been largely retained [Evans & Meggers 1960: 323, 339-40].

The negative consequences of involvement in the cash economy aside, the Rupununi ranching industry has since its establishment provided employment, and more recently a means of independent livelihood, for Wapishana people. Despite the massive decline of the industry following the Rupununi uprising in early 1969, it continues to exert a profound effect on their lives. Cattle are reported to be kept in every Wapishana village, and a few individuals have set up as independent ranchers. The most enduring influence has been the ongoing conflict with the Rupununi Development Company, whose massive land holdings in the south-central savannas literally fenced in several Wapishana communities. For many years, the latter expressed resentment concerning the inadequacy of their own titled lands in providing sufficient grazing for their herds [ARU 1992: 14-18]. The decline of the RDC herd eventually led it to submit to requests to hand over some of its land holdings to South Rupununi communities, although the same has not occurred in the northern savanna [Colchester 1997: 48-9].

As in many other areas of Guyana, mining is an issue which is increasingly affecting the Amerindian people of the south Rupununi. Many young Wapishana men leave their native area in order to work as divers in mining operations in other parts of the country. Others are involved in prospecting locally, particularly in the gold-rich areas of Marudi and Bat mountains, to the south of Aishalton. In 1990 there were reported to be extensive camps of miners in these areas, dominated numerically by coastal Guyanese but also including significant numbers of Amerindian people. This had visible detrimental effects upon the lives of local people, in terms of rising food prices, long-term absenteeism of male heads of households, and the consequent neglect of farming activities [ARU 1992: 29]. The situation is likely to have deteriorated since this study, as incursions by Brazilian garimpeiros into the Marudi and Kanuku areas is reported to have increased following clampdowns by the Roraimense authorities against illegal mining activities [Forte 1997: 73]. More recently, a company called Romanex (Guyana) Exploration has begun development of a 50 square mile concession in the area, setting up an operation they themselves compare to the notorious OMAI gold mine. This will entail the construction of a feeder road connecting the area with the recently completed Georgetown-Boa Vista highway at Lethem, the consequences of which are hard to predict [Colchester 1997: 89]. Generally, the impact of expansion in the mining sector has rarely been beneficial for local Amerindian communities. The best studied Guyanese example is the Upper Mazaruni, where the observed effects include increased economic stratification, extensive environmental damage including pollution of water supplies, nutritional decline resulting from the neglect of subsistence activities and the consumption patterns associated with mining culture, and social impacts as young men and women become increasingly involved in the mining business, including the breakdown of traditional family systems and an increase in schoolgirl pregnancy [Forte 1997: 77-81].

2.1.6 Wapishana and other Guyanese

The Wapishana have something of a reputation for xenophobia and distrust of outsiders, which is reflected in their relationships with other tribes [ARU 1992: 57]. Their relationship with their closest neighbours, the Macusi, has traditionally been one of suspicion and mistrust - according to Schomburgk, "The Wapishana consider the Macusi to be the most dangerous poison-compounders and kanaimas: every sickness, every feeling of indisposition is ascribed to their villainy." [Schomburgk 1923: 308]. He did come across a single joint Wapishana-Macusi settlement in the Haiowe valley, resulting from the former coming across and taking up residence in the settlement as they retreated from the slave trade in Brazil, but the two tribes lived separately and each had its own chief [ibid.: 229]. A marriage between a Wapishana man and Macusi woman was reported by Roth [1929: vi], and today there are a few mixed marriages in the Kanuku region [ARU 1992: 55]. The late 1980's population survey reported the existence of a single mixed settlement in this area, Jawarri, but no population figure was given [Forte 1990a: 15]. Relationships with the Wai-wai are somewhat better, there having been well-established trading links in which Wai-wai-manufactured and Surinamese goods were exchanged for dogs. A number of coastal Arawaks are settled in Wapishana communities, mainly schoolteachers from the Moruka area in northwest Guyana, whose relationships with the indigenous inhabitants also involve some degree of tension, though mixed marriages are reported nowadays to be better accepted [ARU 1992: 67]. The Wapishana relationship with non-Amerindian coastlanders is no better, the most positive view being that they are a necessary evil [ibid. : 57]. The relationship of the Rupununi area with the coast recently underwent a permanent change with the completion of the all-weather road between Georgetown and Lethem, which continues to Roraima's state capital, Boa Vista.

2.1.7 The Georgetown-Boa Vista Road

This road was the source of much controversy in Guyana [eg. Colchester 1993], but has been completed for too short a time to allow its long-term impact to be assessed. While its effects will undoubtedly extend to the Wapishana, the Macusi through whose territory it actually passes will be those most directly affected. A variety of views were expressed by Macusi in different areas prior to the completion of the road. Many in the large village of Annai welcomed the prospect of improved access to markets for their cash crops [Forte 1990b: 7], and actually participated in trail clearance along the route of the new road [ibid.: 3-4]. Members of other communities expressed less enthusiasm, however, at the prospect of a direct link with the coast [ibid.: 13]. A matter of possible greater concern is the improved accessibility of interior areas of Guyana from Brazil. The project was criticised from the Guyanese perspective on the grounds that its primary function was to serve Brazilian interests by providing access to a port in close proximity to the Caribbean, the consequences of which for Guyana had been given little consideration [Forte & Benjamin 1993: 1]. The most serious potential consequence envisaged was an increase in the scale of Brazilian encroachment onto Guyanese territory, and the prospect that a pattern of development similar to that characteristic of frontier areas of Brazil could ensue in the Rupununi [ibid.: 4]. Perhaps the most dramatic recent example is that of Roraima state itself, where highly organised gangs of garimpeiros and illegal loggers have made massive encroachments onto designated Amerindian lands. The Yanomamö situation is probably the best known of these, not least because of the well-publicised murder of 16 Yanomamö in 1993 - their reserves in Brazil and Venezuela are currently occupied by over 3,000 illegal intruders, who are accused of bringing with them disease, pollution and death [Christie 1997]. This is, however, only one aspect of the potential consequences of the road. Tang, writing before its completion, emphasised the problems caused by the poor transport infrastructure as among the most important facing region 9, and repeatedly called for road facilities within and beyond the region to be given high priority in its development [Tang 1995: 93, 97, 109-110, 119, 132].

2.1.8 Wapishana in Brazil

The political situation in Brazil has been hugely different from that in Guyana since the earliest days of European contact with the Wapishana, which is reflected in the differences between the situations of Wapishana people in the two countries. This is despite the extent of cross-border migration; Amerindian people are not legally committed to observe national boundaries where these cut across tribal lands, and there has historically been a great deal of movement, often in response to relative living standards [Hemming 1995: 45]. The Brazilian state of Roraima includes 23 nominally Wapishana villages, the populations of which all actually incorporate substantial non-Wapishana and mixed-race populations [Foster 1990: 138]. The ancestors of these people were the first of Roraima's indigenous population to encounter European explorers, who penetrated the area intermittently throughout the 18th. century. This was due to their being resident along the main rivers, which, along with their occupation of the area's prime cattle-ranching land, meant a great deal of exposure to colonial influences, to which their adaptation was rapid and relatively successful [Hemming 1995: 39-40]. Minimal government intervention in the area before the mid-20th. century left effective political control in the hands of cattle ranchers, and it was the ranching industry that largely determined the nature of indigenous-settler relations [ibid.: 55]. Riviere reported a system of socioeconomic stratification in which Indians who had been integrated into the dominant society - mainly Wapishana and Macusi, and known as caboclos - occupied a position subordinate to that of the Brazilian civilizado population. These classes were neither rigid nor racially defined, and some scope for upward mobility existed, usually by means of wealth, marriage or adoption, but always necessitating rejection of Amerindian tradition and custom for those of mainstream Brazil [Riviere 1972: 28-31]. By the 1970's, it was reported that 60% of Brazil's Wapishana and Atorai were 'integrated', speaking Portugese as their first language, and that the rest were largely bilingual and in permanent contact with the state [Migliazzi 1978: 9]. This process was by this point reinforced by the activities of the central government, whose tactics included alienation of Amerindian lands to ranchers and developers, manipulation of local customs, population displacement, and use of the protective agency FUNAI as an intermediary [ibid.: 14-5]. Foster's statement that the Brazilian Wapishana at the time of her study were 'impoverished in almost every conventional category of "culture"' [1990: 38], is informative about this situation despite being rather absurd if taken literally. There is evidence that acculturation, in this sense, has been more complete in Brazil than Guyana - Foster reports that the few informants able to give complete Wapishana kinship terminologies were all Guyana-born [ibid.: 38], and that traditional practices such as ritual cutting of adolescent boys, rubbing with plants, and the use by hunters of ant and nose binas, were more likely to have persisted among Guyanese Wapishana [ibid.: 197-8].

Foster's presentation of Wapishana as bereft of "culture" is of course misleading, and although the modifying influences of Guyanese, and especially Brazilian society have been huge, even in Brazil some aspects of Wapishana tradition remain. Although Portugese kinship referents are used, for example, their meanings are distorted so that 'aunt' and 'mother-in-law' are synonymous, reflecting the ideal of cross-cousin marriage [Foster 1990: 146-7]. Foster also reports that the division of labour and responsibility according to sex and gender is retained in theory, and largely in practise, although it is not carried over into the wage labour market [ibid.: 150-4]. Belief in kanaima remains widespread, and indeed a survey as recently as 1986 found over half of all deaths were still attributed to this agency [ibid.: 172]. As is normally the case, kanaima killings are attributed to outsiders - interestingly, these included Guyanese Wapishana as well as members of different tribes, suggesting nationality to have become an important aspect of identity [ibid.: 139].

The major general difference between the situations of Guyanese and Brazilian Wapishana is with respect to land rights. As in Guyana, Amerindian villages - known there as malocas - are state-designated administrative units. However, there is nothing corresponding to the reservation system and villages are isolated from each other, located on small pockets of land separated by settler-owned territories from which Amerindian people are generally excluded. The villages are not demarcated, and Amerindian protection is regulated by the highly ambiguous Indian Statute [Foster 1990: 127-129], which provides little scope for combating illegal encroachment by ranchers onto village lands [ibid.: 164-5]. The extent of the latter problem on occasion borders on to the ludicrous - such as the occupation of the very centre of the village of Baixo by the ranch of a squatter, tolerated by FUNAI despite its illegality - and there is great support for land demarcation, despite the fixing of boundaries that this would entail [ibid.: 132-3]. The absence of available land prevents the establishment of new settlements, thus eliminating the traditional means of resolving disputes - irreconcilable conflicts generally result in one party moving to another maloca or to Boa Vista [ibid.: 135-8]. Foster views social relationships in Roraima as having crossed a threshold sometime around the 1950's, beyond which interactions between non co-resident Amerindians came about largely as an incidental consequence of relationships between Amerindians and Brazilians, such as employment relationships [ibid.: 86, 122-4].

2.1.9 Differences between Brazilian and Guyanese Wapishana

While the situation of Guyanese Wapishana is better than that of the majority of their Brazilian counterparts, it is still far from idyllic. Taken as a whole, Amerindians are the poorest segment of Guyanese society, and this poverty has numerous attendant problems, particularly in terms of health and education [Forte 1993]. The Rupununi region is one of those whose Amerindian population is particularly afflicted by these problems, as a UNICEF-sponsored study performed by the ARU demonstrated [ARU 1993: 100-120]. This highlighted the low socioeconomic level and lack of income-generating opportunities, the sparsity of village infrastructure, lack of reliable supplies of potable water and the consequent risk of water-borne disease, the inadequacy of health care facilities and high incidence of diseases - particularly malaria , and the lack of easy access to post-primary education and vocational training. While, in accordance with the Amerindian Act, most communities have title to the land they occupy, the South Rupununi communities of Rupunau, Katoonarib and Parikwarunau, along with the Macusi village of Rewa in the north, do not [Forte 1995a: 2]. Furthermore, like many communities elsewhere in the country, those of the south Rupununi have expressed dissatisfaction with the extent of their titled lands, requesting the release of land occupied by the Rupununi Development Company to allow the expansion of village grazing lands [ibid.: 4], a request with which the company has complied following the decline of its own herds [Colchester 1997: 48-9].

Tang has criticised the top-down approach to development in region 9, asserting that the absence of Amerindian involvement in the planning and decision-making process observed in her specific case study reflected general weaknesses typical of development projects, both in Guyana and generally [Tang 1995: 29, 42-44].

2.2 Main non-Guyanese Anthropological Literature Consulted

The major theoretical perspectives within which this project has been conceived are those of ethnobiology and human ecology. The theoretical texts I have found most useful in ethnobiology have mostly been written by researchers working in South-east Asia, a geographical bias reflected in the works reported here. A huge body of literature exists on the human ecology of Amazonia, which I consider in its own section below. A few non-Amazonian studies have also proved of value in preparing the present study. Rappaport performed a classic study of the ecological function of ritual in the society of the Tsembaga Maring of New Guinea, and the debates raised in the second edition of his ethnography are still pertinent despite the relative antiquity of the study [Rappaport 1984]. Dwyer's study of the ecological relations of the Etolo, also of New Guinea, is close in spirit to the current study, and includes an important quantitative ethnographic study of hunting ecology and behaviour [Dwyer 1983, 1990]. Both these studies examined human ecology within an ecosystem framework; much current research taking this approach is reviewed in the two editions of a volume edited by Emilio Moran [1984, 1990]. These studies tend to be largely concerned with energy and nutrient flows through the system, rather than the population-level synecological approach taken of the present study, but are still of great relevance.

2.2.1 Ethnobiological Literature

There is a large and sprawling literature on ethnobiology in its various forms, selected areas of which have had direct influence upon this work. In terms of theory, Hal Conklin's classic Ph.D. dissertation based upon his research on the ethnobotany of the Hanunoo of the Phillipines remains the baseline text. This work was crucial to the formation of modern academic ethnobiology, being the first to address in an anthropological framework the relationship between perception and classification of, and interaction with the natural world [Conklin 1954]. Hyndman's [1984] research on New Guinea hunters addressed human-animal interactions within a similar ethnoclassificatory framework. The most recent summary text on ethnoclassification is Brent Berlin's theoretical treatise [Berlin 1992]. This is a fairly comprehensive and essential review of the literature, which is weakened by Berlin's obsessive adherence to a theoretical framework which is clearly not supported by the data presented. His analysis can at best be described as naive and misguided, at worst ludicrous. A much more eclectic and realistic theoretical framework is provided by Roy Ellen's analysis of his research on ethnozoological classification amongst the Nuaulu of south Seram. This emphasises the flexibility, variability and non-uniformity of folk classificatory systems within the context of dynamic systems of social and ecological relations [Ellen 1993], and better accommodates much of the data erroneously presented by Berlin as supporting his own theory.

Ellen's work has also provided important methodological information. Ellen emphasises the importance of not allowing one's research methods to constrain the results which might be obtained in studies of ethnoclassification [Ellen 1986: 87-89]. Berlin's methods are a classic example of how this occurs [eg. Berlin et al 1974: 56-8], and the quality of his above-mentioned analysis testimony to Ellen's criticism of such an approach. Also of great use is Gary Martin's book dedicated exclusively to ethnobiological methodology [Martin 1995], the only one of its kind, which is appropriately broad in scope and down-to-earth, if at times a little idealistic. Methods of investigating perceptions of ecological zonation are derived from research into classifications of soil types [Johnson 1974]. The only study of which I am aware which deals with indigenous perceptions of the ecology of particular animal species is a piece of ethnoprimatological research. Simple questions on the behavioural ecology of local primate species were put to a single Colombian Indian collaborator and compared with the published literature; the two were found to be in almost total agreement [Townsend 1995].

2.2.2 Ethnobiology and Conservation

Probably the most important influence on the design of this project has been a paper by Darrell Posey and coworkers, which emphasised the great potential of ethnobiology to make scientific contributions to development and conservation in Amazonia [Posey et al 1984]. This observation has since been frequently echoed but rarely acted upon. The current fashion for emphasising the role indigenous peoples have to play in development and conservation is enough for it to have become a cliche; Jason Clay provided an early summary of the types of initiatives that had arisen by the late eighties [Clay 1988]. Much of the data that has emerged in support of Posey's argument has come from Amazonia, including edited volumes on the traditional resource management strategies of Amazonian Amerindians [Posey & Balée 1989] and a more directly activist volume addressing their role in Amazonian development [Sponsel 1995]. These and other publications in the same vein strongly promote the view that interaction of Amerindian peoples with the natural world, and the systems of knowledge underlying this, are crucial components of the historical ecology of Amazonia, and still have a vital contribution to make despite the speed and extent of current changes.

2.2.3 Amazonian Human Ecology

The study of human ecology is the investigation of the ways in which culture effects the adaptation of human populations to their environment, ecological and social, by the study of the articulation of ecological factors with all three components of the cultural system - infrastructure, structure, and superstructure [Sponsel 1986: 68]. Early studies of the human ecology of Amazonia concentrated on the ecological factors limiting the size, permanence, and hence cultural complexity of Amerindian societies. Betty Meggers suggested agricultural potential to be foremost among these, proposing a broad-scale categorisation of Amazonian environments along this dimension which she suggested mapped closely onto a 4-fold categorisation of Amazonian culture areas devised by Julian Steward [Meggers 1954]. Her argument - which has since been substantially revised [Meggers 1996] - was based on a rather unilineal model of cultural evolution and vast understatement of the potential for culture to exact environmental modifications, and was undermined by a series of papers by Carneiro. His research on Kuikuru agriculture demonstrated swidden cultivation of manioc to have the potential to support permanent settlements vastly larger than those typical of Amazonian forest societies. He suggested settlement size and duration to be limited by two factors - protein availability, and circumscription, the ability to respond to pressure on resources by moving to new areas rather than via technological innovation [Hames & Vickers 1983a: 10] The former point was taken up by Gross [1975], who suggested that given the low faunal standing crop that had been reported for Amazonian forests, the scarcity of non-meat sources of dietary protein, and the absence of domesticated game, protein availability must have been a limiting factor for many Amazonian societies, and their traditional social organisation an adaptation to this. Beckerman [1979] pointed out that this argument was based on numerous misconceptions, including underestimations of the availability of fish, of the pre-contact population of Amazonia, and of the amount of protein available from non-meat sources (such as reptiles, invertebrates, palm fruits, wild nuts and fermented foods), as well as the use of figures for animal densities from areas vastly different ecologically from those inhabited by people. In particular, there is evidence that the secondary growth associated with abandoned swiddens functions to attract game animals, the local biomass of which is artificially raised in a process which provides a functional substitute for domestication of animals [Linares 1976]. The protein question has been subject to much debate since then, but this has largely been based upon analysis of site-specific data which is of little value in illuminating the general picture [Moran 1993: 19]. There is clearly a need for a large amount of further study - in particular, the development of a sufficient number of databases based on anthropological, biological and nutritional research at the same sites to allow comparative study to be undertaken - before the nature of environmental limitations on the development of human societies is understood. As Johnson [1982] has pointed out, if models can be usefully constructed within a paradigm that considers the adaptive nature of human society to be largely determined by responses to scarcity, this scarcity is likely to be manifest along several ecological dimensions, as well as being mediated and modified by structural elements which may not themselves have any ecological basis.

A second current has attempted to employ models derived from behavioural ecology, notably optimal foraging theory (OFT), to the study of human ecology in Amazonia. Models derived from OFT have been used fairly successfully in trying to account for data on, amongst others, the hunting behaviour of the Siona-Secoya of Ecuador and Yanomamö and Ye'kawana of Venezuela [Hames & Vickers 1982], and the hunting and gathering of the Aché of eastern Paraguay [Hawkes et al 1982]. Both of these studies concentrate only on energetic costs and benefits and as such could be accused of exhibiting the 'calorific obsession' common to many areas of ecological anthropology. In a similar vein was Behren's [1981] application of a linear programming model to data on time allocation to hunting and fishing among the Shipibo of Peru, an improvement on the aforementioned studies in that it also took protein intake into account. Sponsel [1986: 83] points out the need to consider the full range of nutritional requirements in such studies - a point of which the above researchers are undoubtedly aware but which is beyond the logistical means of majority of anthropological studies - in continuing the general trend towards improved theoretical sophistication in studies of human ecology. In addition, it is worth noting that subsistence is not the only important component of human survival - the Machiguenga of the Peruvian Amazon, for example, are reported to abandon settlements in response to a scarcity of firewood [Johnson 1982: 415-6]. Furthermore, evolutionary models are not concerned with survival but with lifetime reproductive success - to return to the Aché, it is reported that successful hunters are more frequently involved in extra-marital affairs, a factor which in the evolutionary context probably outweighs purely nutritional considerations [Hawkes et al 1982]. The evolutionary naivete of the application of these models to human situations is more fundamental, however. OFT assumes that the foraging behaviour of the subject of investigation is genetically based and has evolved so as to maximise its inclusive fitness [Hamilton 1964] within existing constraints. In the absence of any adequate heuristic theory of cultural evolution and consequently of the evolution of human behaviour, the application of OFT to human situations has no theoretical basis. Connections between the quantities being measured and their evolutionary consequences are not adequately explored, even to the extent of specifying whether the entity being reproduced is the individual, a cultural trait or traits, or the social group. Adaptation takes different, possibly contradictory, forms at each of these levels - Carneiro [1982] observed that many Amazonian people will express a 'need' for meat even if their physiological requirements for protein are abundantly satisfied. This ignores the possibility that the need reflects a deficiency in some other nutrient supplied by meat, but the point is that cultural factors unrelated to physiology may affect the dietary needs of humans, and hence the nature of their adaptation to their environment. The future development of OFT approaches to human ecology needs to be closely allied to the search for evolutionary theories of culture, and the construction of models of foraging behaviour integrated with the development of models of cultural evolution such as that of Durham [1982]. The other major flaw in these studies is their tendency to be componential, focusing on a single subsistence activity in isolation, an approach which can lead to analytical problems. Bari hunting and fishing behaviour, for example, can be seen to fit the predictions of OFT models only if the two are considered simultaneously, as complementary parts of an integrated foraging strategy. So long as progress continues to be made in overcoming these current limitations, the potential contribution of OFT to the study of human ecology may be vast.

Both the limiting factor and optimal foraging approaches have been criticised by William Balée for ignoring the dialectical aspects of the relationship between culture and environment, the analytical distinction between which should be rejected if Amazonian cultural ecology is to be properly understood [Balée 1989: 2]. He presents evidence to suggest a minimum of 11.8% of 'natural' Amazonian forests to have been modified by prior human activity, which has had the effect of increasing the abundance of economically important plants such as Brazil nut (Bertholletia excelsa) and numerous species of palms. A case study of two Amerindian groups resident in an area containing a high proportion of such anthropogenic forests, the Araweté and Asurini of the Upper Xingu, shows them to rely extensively on anthropically enriched terra preta soils and on numerous tree species found most abundantly in forests disturbed by prior human activity [ibid.: 6-14]. Also in the Xingu basin, the foraging Guajá tribe are highly dependent on plant resources produced largely as an incidental by-product of the activities of the ancestors of their agriculturalist neighbours (and one-time mortal enemies), the Ka'apor [Balée 1993]. The Kayapó of the campo cerrado of central Brazil extensively manipulate their scrub-savanna habitat by inducing the formation of bush islands (apêtê) , almost the entire floristic composition of which consist of plants for which they have some practical application [Anderson & Posey 1989]. I have already made reference to the increase in game abundance that occurs in the vicinity of 'old' (and of actively cultivated) swiddens. In one of the few fully quantitative, contemporaneous studies of this, it was found that the amount of the fresh game entering a Ka'apor village that had been caught in or near swiddens was disproportionate relative to the amount of time spent hunting there [Balée & Gély 1989: 137]. Phillipe Descola, coming from a vastly different, less empiricist perspective than Balée, also argues for the rejection of the culture/environment dichotomy; his study of the Achuar of the Ecuadorian Amazon portrays their interactions with their environment as a process of socialisation of nature [Descola 1994]. His work is also a reminder of the need to consider human ecology in a much broader context than the techniques used to study the ecology of non-human species can provide. One major determinant of differences in ecological relations, and numerous other features of society, that existed among the different Achuar groups and the other Jivaro peoples in the area, for example, was the duration and nature of missionary contact, which is a demonstration of how different societies occupying ecologically identical habitats can manifest radical cultural and social differences as a result of historical factors without any relation whatsoever to the local ecology [ibid.: 20-35]. Audrey Butt's comparison of three tropical forest peoples in the Guianas makes a similar point. The differences in social organisation and ecological relations between the Waiyana Indians and the Boni, both resident on the Upper Maroni river, she attributes to historical factors, specifically the retention by the latter of the systems of land use employed by their ancestors in the forests of West Africa, adapted to reflect the differences between the two habitats [Butt 1970].

The nineties have seen an increasing focus on the practical and applied dimensions of Amazonian human ecology, and in particular an increased interest in the potential for the development of extractive industries. The sustainable harvesting of non-timber forest products such as fruits and nuts [Clement 1993], palm products [Kahn 1993] can form part of a multi-faceted production system which reconciles use of the forest as an economic resource with the long-term maintenance of biodiversity and ecological functions [Lescure & Pinton 1993, Clement 1993: 140]. Interest in this approach also stems from the fact that such industries would necessarily be based largely upon indigenous knowledge of the properties of the local biota [eg. Posey et al 1984] and, indeed, often depend upon the very floral resources now recognised to be the products of past human activities [Balée 1989]. The economic importance of extractivism is already vast - it was estimated in 1982 to employ 1.5 million people in the Brazilian Amazon alone, producing an annual income of around 100 million dollars [Anderson & Ioris 1992: 178]. The island of Combu, in the Guamá river near Belém, supports over 600 people at a population density of 43 per km2, 92% of which rely on extractive industries for the majority of their income [ibid.: 184-5]. The economic viability of extractivism compared to alternative land uses - specifically permanent agriculture and livestock raising - has been investigated by Susanna Hecht. She found that, although the alternatives were more lucrative over a single short-term cycle of 10-15 years, if the cost of land degradation was incorporated into her models extractivism was the only economically viable option. In other words, the potential for sustainable practice makes extractivism a more economically sound long-term management option than the alternatives [Hecht 1992]. The concept of the extractive reserve has its social and political roots in the rubber-tapping industry of the Brazilian Amazon; the concept has proved politically acceptable there, where numerous such reserves now exist, and is being widely adopted in conservation and development circles elsewhere. The Siona and Secoya of Ecuador are successfully adapting to the economic realities of modern times, along lines firmly based in their own culture and traditional activities, by the use of a system of resource exploitation in which extractivism is a major and vital part [Vickers 1993]. Maya people resident in the Uaxactún-Carmelita multiple-use reserve in the Petén of northern Guatemala benefit financially from an management regime based upon the extraction of palm leaves (Chamadorea oblongata and C. elegans), allspice (Pimenta dioica and P. officinalis) , and chicle gum, in combination with nature/archaeology tourism [Nations 1992]. The use of faunal resources has been largely underattended in the literature, despite its nutritional importance in subsistence economies [Redford 1993: 227]. One collection of papers which deals with this [Robinson & Redford 1991] includes several of relevance to Guyanese fauna. The incorporation of hunting and trapping into sustainable forestry programmes is not straightforward however, requiring a great deal of background data on the population dynamics and carrying capacities of the species concerned [Feer 1993: 692-699], and careful attention to the social and ecological effects [Redford 1993: 243-4]. The present study will provide a much-needed addition to this small body of literature, and will furthermore be the first to consider human use of forest fauna within a framework which links ethnobiology and human ecology in a fashion explicitly geared towards the investigation of resource management issues.

2.3. Wildlife Trade

The major institution affecting the international trade in wild-caught animals is CITES, the Convention on International Trade in Endangered Species. The purpose of this treaty is to regulate international trade in those species for which the continued removal of individuals from their natural habitat is a threat - actual or potential - to the continued survival of wild populations, either on a global or a national level [Favre 1989: xvii]. I am aware of only a single evaluation of the success of the treaty, that of Trexler [1990], who expresses a great deal of scepticism about its merits as a conservation initiative. According to Trexler, CITES is 'often labelled the most successful international wildlife conservation instrument' [ibid.: 30], but the grounds for this are shaky in the absence of any indicators of success - implementation costs are not borne by developed countries, implementation is itself often counterproductive to conservation (such as when the impounding of products encourages a rise in capture rate to meet demand), the relationship between trade flows and the status of wild populations is generally obscure, and there is in any case no evidence that overexploitation of any species has been halted by the convention [ibid.: 98-99]. The real successes of CITES, he continues, are not directly linked to its putative conservation mission, but in enhancing awareness and understanding of the wildlife trade and the associated problems, especially in animal welfare, and encouraging conservation-related research [ibid.: 100]. His recommendation is that CITES could be made more effective if it combined focussing implementation on the minority of species actually affected by the trade, and expanding its mandate to include a much broader range of policy goals [ibid.: 125-6].

The detailed reasons given by Trexler for CITES' lack of effectiveness as an item of conservation policy are relevant here in so far as they illuminate the Guyana situation. He points out that since the year 1600, the importance of live trade as a cause of extinction in all four classes of tetrapod has been minimal. Furthermore, the numbers of individuals involved in international trade are generally small compared with those removed from the wild for subsistence or domestic live trade, and that the effects of these factors are difficult to separate. The conservation impact of the international wildlife trade is small compared with that of habitat loss and degradation - in fact only a small proportion of it affects endangered species, and is therefore of conservation relevance - but its glamour and the possibility of control have led to it being the focus of an inordinate amount of attention [Trexler 1990: 4-15]. Although there is some truth in these arguments, they are also naive in many respects. The current wave of anthropogenic extinction lacks any precedent, making inference from historical trends of dubious relevance. In addition, there are interactive effects between species depopulation and habitat degradation. Firstly, the depletion of local populations of species with important roles in the ecosystem, such as in dispersal or pollination of plant species, can itself be a component of habitat degradation, and it is therefore incorrect to assume that this is of conservation importance only when it directly threatens the wild populations of harvested species [Mulliken et al 1992: 9-11]. Second, extensive loss or degradation of a species' habitat can increase the vulnerability of the remaining population to significant depletion by live capture, as is becoming true for many species of psittacines in the neotropics [Collar & Juniper 1992: 12-13].

Literature on the wildlife trade in Guyana is minimal - a brief report on the trade in live birds was published by TRAFFIC international in 1992 [Edwards 1992], but there has been no independent study of the situation since, and it is unlikely to have been unaffected by the radical economic and social changes the country has undergone since that time. Guyana became a CITES signatory in 1977 [ibid.: 77]; legislation concerning the wildlife trade currently takes the form of the third revision of the 1987 wildlife regulations, dated June 10 1995, and concerned with both the conservation of wild species and the welfare of individual animals involved in the trade. The most significant element of this from the conservation perspective is the quota system - since 1987 national export quotas have been set for all species which can be legally exported from Guyana. The levels of these quotas were set in the absence of any baseline population data at conservative levels assumed to be below those at which wild populations might begin to be threatened; many of these have since been revised downwards under advice from the CITES secretariat [ibid.: 80]. It has been suggested that the scientific naivete of these figures might mean they exceed sustainable harvesting levels [Thomsen & Mulliken 1992: 229]. On the other hand, it is reported that exporters were generally of the opinion that the size of wild populations is sufficient that the export quotas were in fact unnecessary, especially since only around 10% of the available habitat is exploited in this way and a vast reservoir of habitat therefore available for repopulation [Edwards 1992: 79]; however local depopulation is obviously a possibility. Although it is not unlikely that the legal international trade is within the limits of sustainability on a national level, it is quite possible that there may nevertheless be declines in local populations. Furthermore, the regulated trade may be only a small proportion of the total catch of live animals; the magnitude of the domestic trade is unknown, (but assumed to be substantial due to the popularity of keeping pets, especially birds and monkeys) [ibid.: 84], as is that of illegal export across the land borders to Brazil, Venezuela and Suriname.

The economic importance of the wildlife trade in Guyana in 1992 was estimated by Mrs. Reece of the Guyana Wildlife Exporters Association. She calculated that the industry involved 16 exporters employing a total of 430 people, and around 7540 trappers and domestic traders, meaning that around 50,000 people (including the families of all those involved in the trade) receive some form of direct financial benefit from the trade, along with an estimated 2,000 who benefit financially from the provision of goods or services to those directly involved. Amerindians comprise around 75% of these people, and the vast majority of trappers are Amerindian men, who may operate individually or as part of a village cooperative [Edwards 1992: 82-83]. It is notable that this is an economic activity based upon exploitation of natural resources via the application of local knowledge of the biology and behaviour of the animals involved. The most conspicuous component of the trade involves birds, mostly of the order Psittaciformes (parrots, macaws and relatives), but also including several species of toucans and other taxa. In terms of quota sizes, birds are actually outnumbered by reptiles, several species of which are exported in numbers of 10,000 or more per annum. A small number of species of amphibian are exported, and small quotas exist for several species of mammal [P.E. Williams, pers. comm. 1997]. The export of numerous species is totally banned, including all native CITES-listed species [IUCN 1996], and it seems unlikely that any native Guyanese species is threatened with extinction by capture for the live export market.

3. Research Methods to be used

The basic approach combines elements of anthropological and biological methodologies. Anthropological methods will be used to investigate attitudes, perceptions, usage and knowledge of natural resources. This will be a precursor to assessing their compatibility with, and potential contribution to, management plans which seek to situate them within a framework derived from the conventional scientific approach to conservation biology.

3.1 Background and Justification to Methodological Approach

The key question this research will attempt to address is the extent to which 'ethnobiology' - in the sense of understanding and knowledge that the people resident in a particular place have of the properties of the local biota - can provide the basis for a local plan for the conservation and sustainable use of forest resources. The full breadth of this question is too great for a project of this scale to attempt to tackle; the field of enquiry will therefore be narrowed down by focusing on human exploitation of wild fauna, and direct and indirect effects of human activity on animal species of particular economic, conservation, or other importance. It is worth noting that this makes sense only as a methodological and analytical device and does not alter the need for a holistic approach to conservation, to which a focus on particular species or groups may nonetheless be an appropriate starting point.

The above question is broad and vague, and in order to design a methodology that can answer it, it must be deconstructed into simpler, specific questions :

1) a) What are the most important conservation issues nationally ?
b) What are the most important conservation issues locally ?
c) How do the answers to these questions differ among different interest groups ?

These will be addressed by means of structured interviews with people from the following groups :

A) Academics, civil servants, staff of NGOs, politicians, and any other appropriate figures within Georgetown. These will take place as and when the opportunity arises during any time the researcher should spend in the city.

B) Key figures locally in a number of villages - especially captains, council members, teachers, health workers. These will take place as part of the initial surveys in which a suitable study site will be sought.

C) As large and representative a sample as is possible from the host village. The exact interview schedule is likely to differ between each of these groups, reflecting the different levels of knowledge of local- and national-level conservation. However, enough correspondence between them will be retained to allow some direct comparison to be made.

2) What are the local perceptions of the importance of different forest species, in economic, conservational, symbolic and aesthetic terms ?

This question will be addressed by extension of the interviews to be performed with groups B and C in (1), to include questions such as 'what kinds of plant and animal are most important to you, and why ?' and 'what kinds of plant and animal are most important to the forest, and why ?'.

3) What information is conventionally regarded as necessary for the management and sustainable exploitation of populations of these species, according to ecological theory ?

Reference to published literature - eg. Robinson & Redford [1991] and others cited above as appropriate, IUCN publications re. protected area management, protected area management plans concerning these species or their ecologically similar, close relatives.

4) a)How much biological data is available on these species ?

b)To what extent does the available data, in theory, satisfy the demands identified in (2) ?

Review of literature, direct contact with researchers at relevant institutions.

5) a)What is the extent of local knowledge regarding the behaviour and ecology of these species ?

b)To what extent can this satisfy the demands identified in (2) ?

Prompted elicitation of ethnoecological knowledge with key collaborators, both in village and field.

6) a)To what extent are the bodies of data used in (4) and (5) consistent, and to what extent contradictory ?

b)Assuming contradictions exist, are any patterns identifiable either within or between species ?

Direct comparison of 2 data sets.
It may be possible to use computer modelling, or database operations, to create indices of correspondence or some other quantitative means of comparison.

7) If the contradictions identified in (6) are amenable to empirical investigation -
How are the contradictions between the two bodies of data resolved by empirical research ?

Empirical ecological research with Amerindian researchers, concentrating on areas of contradiction relevant to local resource management needs.

3.2 Details of methods to be used during first stage of research.

The following outlines the methodologies to be used during the first stage of research, which it is envisaged will last for around 9-10 months. These methods aim to address questions 1, 2 and 5 above. The major goal of this stage of research will be to lay the foundations for the second stage - a programme of ecological research performed in collaboration between the researcher and the host community and aiming to tackle local conservation/resource management problems. This first stage will therefore involve the assessment of local needs and wishes, and of the capacity of local ecological knowledge to form the basis of a management plan which would address these. It is envisaged that the researcher will follow this by leaving the field for around two months in order to plan in detail the second stage, which is therefore not well-defined at this juncture.

3.2.1 Survey of Several Wapishana Villages

In order to assess the general interest in the proposed type of study, and to identify where it would best suit the activities, interests and concerns of local people, the first stage of the study will comprise visits to the major centres of Wapishana settlement. Of these, Sand Creek, Shea, Maruranau, Awariwanau, Aishalton, Karaudanawa and Achiwuib may all be possible study sites. The proposal to perform a study specifically in the Sawariwau area (see section 8, below) means that a visit to this site at this stage is desirable, but not essential if time constraints do not allow it. The geographical location of Potarinau makes it an unlikely candidate for long-term study, and may add considerably to the time-scale, but a visit to this area would be desirable at some stage, if it should practicable within the other constraints.

The exact schedule of work at this stage should be determined in discussion with relevant institutions in Guyana, particularly the Ministry of Amerindian Affairs, in order to obtain information suitable to their activities and needs. Provisionally, it is hoped to survey opinion and knowledge about conservation issues locally, nationally, and globally, and perceived needs and wishes in the use and management of natural resources. A combination of interviews with key figures in the communities - toushau, councillors, schoolteachers, community health workers, and any other individuals the above might suggest - and open fora will be used to achieve this. A participatory method along these lines for introducing proposals for new projects has been developed specifically for use in Guyanese Amerindian communities [Shook 1995], many of whose methods have relevance to the current situation. It is also hoped that interviews with a certain degree of overlap in content with the above might be performed with members of various groups in Georgetown and Lethem - including academics, politicians, civil servants and NGO staff - as opportunity might arise on occasions when the researcher is in these places.

Given the vagaries of travel in the Rupununi, the exact duration of this is hard to predict - 6 to 8 weeks would probably be a realistic period of time to allocate. It would be appropriate to contact the relevant toushaus by post in advance, as well as the regional administration in Lethem, in order to inform them of these plans and request permission to visit the communities.

3.2.2 Brief Ethnographic Surveys within Host Community

While it is not the purpose of this study to fulfil the need for a comprehensive ethnographic study of the Wapishana, any ethnobiological study must be situated in a general ethnographic context [Ellen 1993]. The scarcity of ethnographic information on the Wapishana makes the collection of basic ethnographic data in the host community an essential first stage. It will also provide a means for the investigator to become quickly oriented within the community, to begin collecting useful data soon after arriving in the field, to get to know and be known by the people there, and become familiar with local idioms so that appropriately phrased question frames may be devised for interviews in later stages of the study.

The research goals and methodologies for this phase of study can be determined by reference to a number of criteria. Most important from the Guyanese perspective is the work of the ARU on village sizes, composition and demography - by consultation with them it should be feasible to modify the questionnaire design so as to produce information that will be useful to this ongoing research programme. Also relevant is a study sponsored by Demerara Timbers Limited of the situation of the community of great falls [Fox & Danns 1993], in which basic information sheets largely compatible with the aims of the present study were designed and used. Tang conducted interviews with Aishalton residents on attitudes towards development [Tang 1995: 68-70]; replacing the term 'development' with 'conservation' would make the questions she asked fairly appropriate to the present study. APFT's own research guidelines also provide suggestions as to the areas that this phase of the study should focus upon [Kocher-Schmidt et al , n.d.].

It is proposed to conduct semi-structured interviews as widely as possible within the host village to obtain information on the following : Population, demographic structure, kinship ties, residence patterns, educational and employment background, linguistic competence, income, relationships with other communities (Amerindian & non-Amerindian).

Importance of forest resources - economic and otherwise.

Perception of conservation - awareness of issues, knowledge and understanding of terms. Assessment of local conservation situation, knowledge of national and international situation and their relationships to the local. This will overlap to a large extent with the interviews conducted in section 3.2.1, in order to provide comparable data sets.

Knowledge and perception of relevant organisations, institutions and programmes. Although the exact period of time dedicated to these activities depends on the size of the survey population, the enthusiasm with which people participate, and the time dedicated to other activities, it may be expected to be the major activity for 8 to 10 weeks. Further collection of data by observation and questioning will occur throughout the study period, as and when opportunities arise.

3.2.3 Ethnozoological Inventories & Initial Ethnoecological Inquiry

Although a comprehensive investigation of Wapishana ethnozoology is beyond the immediate aims of this study, collection of a certain amount of basic nomenclatural and classificatory data is an obvious necessary early stage. It will initially be attempted to derive a basic picture of relevant areas of the local zoological lexicon and classificatory scheme, such as is an essential precursor to studies of local knowledge of animal ecology. The potential academic value of continuing this line of enquiry is sufficient to justify spending more time than is strictly necessary for this. The interrelatedness of people's apprehension of, and interaction with, the natural world means that fairly extensive investigation may be illuminating with respect to the central goals of the project. Additionally, the patterning of different areas of ethnobiological knowledge along numerous dimensions, particularly gender and occupation [eg. Ellen 1993: 135-8], makes it desirable to conduct an investigation that is as comprehensive as possible with respect to coverage of the different groups that exist. Methods will largely follow those in Ellen [1993], and will mainly consist of direct questioning; some sorting tests using pictures and photographs, if available, will also be performed.

The ethnoecological studies will have two main dimensions - local perceptions of ecological zonation and spatial differentiation of biotopes and local knowledge of the autoecology of native species, focussing on animal species of economic (commercial or subsistence) importance. The former will use methods based upon Johnson [1974], who investigated categorisation and investigation of soil types using questions that may be simply modified for the present purposes. Maps will also be used at this stage: both asking collaborators to draw their own maps of the local area, and to annotate whatever maps of the area may be available, should be informative at this stage. The latter component of this will involve direct, open-ended questioning on the habits, diet and behaviour of the focal species [following Townsend 1995]. One problem with this approach is that the structure of the questions used inevitably reflects preconceived notions of the investigator stemming from his education in the scientific tradition, and there is a danger of forcing the results into the theoretical framework underlying the questions. A way round this might be to ask literate collaborators to write their knowledge down ad libitum prior to engaging in the formal interview situation. Potential collaborators whose activities mean they have extensive contact with forest wildlife, will be identified in the initial village-wide ethnographic surveys, and asked if they would be interested in participating in these further studies. The focal species with which this stage of the study will be mostly concerned will be those identified by interviewees as important to them, and of which they themselves profess to knowledge. It is hoped that more extensive field-based studies will be conducted with the same collaborators at later stages in the project.

3.2.4 Accompanying Trapping & Hunting Trips

Having identified the individuals most active in hunting and trapping, who are willing to participate in the study, the investigator will arrange to accompany these people when they journey into the forest to trap or hunt. There will be several goals to this stage of research. First will be to document, via direct observation, the techniques used by trappers and hunters, and if possible to obtain some quantitative data on how often and in what circumstances different methods are used, and the respective success rates of these. Second will be to document the spatial extent of forest use - the frequency with which different ecological zones, both emically and etically recognised, are used by people for these activities. This will be recorded both in the form of time allocation data and the production of maps. The use of Global Positioning Equipment (GPS) could be advantageous at this stage, if conditions permit. Thirdly, this will be the first circumstance under which it will be attempted to conduct ethnoecological enquiry in the field, building upon previous interviews conducted in the village and acting as an introduction to more extensive field interviews to be conducted later in the study.

These field trips will commence as soon as is possible, given the need for a certain amount of prior research and the dependence on the itineraries of the local participants. They should continue certainly throughout the first stage of research, and probably into the second, in order that any seasonal variation be documented. The evident danger in this is that the presence of the investigator will distort the normal pattern of activities, if participants should modify their schedule in order to accommodate his wishes or obtain renumeration for their assistance. It is hoped to be as unobtrusive as possible and not compromise people's prior plans in the early stages - in this situation substantial renumeration will not be appropriate, and it is anticipated that there will be little problem in communicating a desire to observe a 'natural' pattern. As time passes, it is anticipated that collaborators will develop sufficient appreciation of the goals of the investigation to allow explicit steps to be taken to avoid this problem in discussion with them. By this stage it should prove feasible to combine study of hunting and trapping with extensive field-based ethnoecological interviews using phenomena directly observed as stimuli to discussion. Prior to this, opportunities to record such data will be taken opportunistically, but it is expected that in a field situation these should arise with a reasonable frequency.

It is also hoped to find the time to follow the fate of trapped animals that enter the live trade, to obtain some information on their destinations and the social and commercial networks within which this trade operates. Consultation with the relevant institutions in Georgetown - the Wildlife Department and Ministry of Agriculture - will provide essential background information on the wildlife trade, and assist in placing the local trade within its national and institutional context.

3.2.5 Investigation of Hunting Rate and Returns

The time investment and rate of return involved in hunting will be determined using methods based on those of Dwyer and Vickers. Dwyer managed to obtain accurate records of over 90% of the game animals killed over the course of a year within his study community by making his willingness to purchase skulls of game known to the people. On purchase of the skulls, he was able to make enquiries concerning the place, time and circumstances of capture; from measurements of the skulls he was able to derive estimates of the weight of the individual animals, based upon linear regression models constructed in the field [Dwyer 1983: 148]. The main problem here is in setting a suitable level of renumeration - sufficiently high as to provide incentive for people to participate, but not so high as to artificially inflate the level of hunting success or to encourage deception among participants [Dwyer 1990: 103-119]. Dwyer's methods for measuring hunting effort were prone to severe inaccuracy [Dwyer 1983: 147-8,154]. Vickers [1991] recorded the yields from several thousand man-hours of hunting, and the spatial usage of ecological zones, in an Amazonian community by a combination of direct observation and questionnaires administered by local assistants. The volume of direct observation required to replicate this is not practicable within the present study. However, given the higher educational standards and literacy rate that is likely to occur in a Guyanese community, it may be possible to modify and combine these two approaches by asking participants to keep diaries of hunting activities in conjunction with skull collections. The self-recording of data by participants has its own potential pitfalls, but given the highly dispersed nature of Wapishana settlements this may prove the only practicable means of obtaining broad-scale data on hunting effort within the context of this project - in addition, the direct involvement of hunters in this area of research may be a useful means of demystification and engagement.

The investigator will attempt to set up a system of the sort described above at as early a stage of research as proves possible. The initial community-wide surveys should prove instructive as to how this can best be achieved in practice.

3.2.6 Extended Field Trips with Key Collaborators

From accompanying key collaborators on their own trips into the bush, it is hoped to proceed ultimately to organising dedicated field research trips. These trips will combine biological research with extended in-depth ethnoecological interviews in a field context. The biological component will be mainly determined by local interests, but is expected to consist largely of collection of specimens of plant species mentioned in ethnoecological studies - for use as voucher specimens, and if possible incorporation into local and national herbaria - and ecological sampling for the assessment of ethnoecological zonation. It is hoped that collaboration with botanists resident in Guyana will be possible for the scientific identification of plant specimens. Contact will be made with researchers on the Iwokrama programme and the Smithsonian Institution's Biological Diversity of the Guianas project in order to establish if collaboration is possible.

It is difficult to assess how soon it will be possible to conduct research of this sort. However, it is anticipated that it will commence in the latter half of the first stage of research, becoming the major research activity at this stage, and merge into the more extensive field activities of the second stage, during which time ethnoecological enquiry will continue as appropriate.

3.2.7 Investigation of Sawariwau Bush Islands

The presence of 'bush islands' throughout the South Rupununi, and their use in human agriculture, poses the intriguing question of whether their origin might be anthropogenic. Eden [1986] comments that their irregular shape is suggestive of anthropic origins, but to date this matter has not been investigated, and the possible importance of this matter to local conservation and resource use issues makes this a research priority. It is proposed to spend a period of time in and around the villages of Sawariwau, Wichakai and Dadanawa to investigate the possible human agency in the ecology of these areas. Such was demonstrated in the case of forest islands in the savanna scrub habitat of the Gorotire Kayapo, central Brazil, by Anderson & Posey [1989]. Their methods involved in situ and ex situ interviews regarding the identity, plantability and utilitarian value of all large plants found in these islands; it is hoped to replicate them here. This will be a brief study of around four weeks, with the major aim of assessing whether further, more comprehensive investigation would be desirable.

3.2.8 Work in Orealla

The people of Orealla have recently begun to attempt to organise local initiatives for the development of sustainable livelihoods. These include the documentation of local ethnopharmacological knowledge ('bush medicine'), and the development of a nature tourism project based at a proposed conservation area within the reservation. It is within this broader context that I will initiate work in Orealla, some with direct relevance to my own research programme, some without. With regard to the former, it is hoped to work extensively with two individuals.

Stanley Herman is a young Arawak man who for many years has worked as a macaw trapper. It is proposed to employ him to work as research assistant, to conduct unsupervised research on the ecology of all four macaw species present at Orealla, for the duration of the project. He has a keen interest in wildlife and has expressed an interest in building upon this in a professional capacity, if the opportunity were available. It is proposed to employ him to spend several days each month engaged in observational research upon the ecology and behaviour of macaws, and any other species that might take his interest. Initially, the design of this research will be left entirely up to him, but at later stages some training in conventional biological research methods will be given, and incorporated into a framework largely of his own design. It is hoped that this might lead to the collection of data relevant to such local management programmes as might develop in the future, as well as helping him to develop skills which will be useful to the community in its efforts to begin to make more prudent use of local natural resources. From the research perspective, there is also the possibility that this exercise could prove to be an interesting experiment, abliet one that, with a sample size of one, is somewhat limited in scope. The major anthropological interest will be in the ways in which the research goals and methods develop under the assistant's initiative according to his own experience and insights, and how the conduct of research affects his own outlook on and perception of the subject matter of this research. These will be investigated by interviews - an initial, extensive ethnoecological interview, to be repeated at such times as the investigator is able to either visit Orealla or arrange for such interviews to be conducted.

William Herman (no relation to the above) is a septagenarian whose skills as a 'bushman' are legendary in Orealla. He appears to possess an extraordinary amount of knowledge of local wildlife, and has expressed a concern that, since his sons show little interest in this subject, this knowledge will be lost on his death. Clearly, he would be a model ethnobiological informant - to attempt to record at least some of his knowledge, and ideally to allow it to be passed on to younger generations, would be a great service to the people of Orealla. Time will permit the investigator only a superficial study of his knowledge, but such a study could be a starting point for more thorough documentation by community members themselves. Such an exercise may be in the interests of conservation and development NGO's - Conservation International , for example, run a 'Shaman's Apprentice' programme in Suriname which is similar in outlook,and have recently made moves to extend it to Guyana - who may be persuaded to collaborate. If funds are available, assistants could be employed within the village to collect background data along similar lines to those in the Wapishana study, but modified as local needs dictate.

In addition to the time spent with the above individuals, the investigator will also attempt to offer such advice and guidance as is possible on the ethnobotany and ecotourism projects, especially by providing informative literature on similar projects and suggesting appropriate organisations to contact. The accessibility of Orealla from the coast - a round trip from Georgetown takes as little a 2-3 days - makes it feasible for occasional short visits to be made by the investigator over the course of the research programme, following an initial stay of around 4 weeks.

4. Literature Cited

Adams, Kathleen 1981. 'The narrative of a Barama River Carib youth'. Archaeology and Anthropology 4(1,2): 39-50.

Anderson, Anthony A. & D.A. Posey 1989. 'Management of a tropical scrub savanna by the Gorotire Kayapó of Brazil'. Pp. 159-173 in Posey & Balée (eds.).

Anderson, Anthony A. & E.M. Ioris 1992. 'The logic of extraction: resource management and income generation by extractive producers in the Amazon estuary.' Ch. 9 (pp. 175-199) in Redford & Padoch (eds.).

ARU 1992.'The material culture of the Wapishana of the South Rupununi savannas in 1989'. Amerindian Research Unit, University of Guyana.

ARU 1993.'Status of women and children in the Guyana Amazon'. Report complied for UNICEF by the Amerindian Research Unit, University of Guyana.

ARU 1995.'Situation analysis of indigenous use of the forest with emphasis on region 1'. Amerindian Research Unit, University of Guyana.

Bahuchet, Serge (ed.) n.d. 'The situation of indigenous people in tropical forests . APFT preliminary research report for EU DG XI.

Balée, William & A. Gély 1989. 'Managed forest succession in Amazonia: the Ka'apor case'. Pp. 129-158 in Posey & Balée (eds.).

Balée, William 1989. 'The Culture of Amazonian Forests'. Pp. 1-21 in Posey & Balée (eds.).

Balée, William 1993. 'Indigenous transformation of Amazonian forests: an example from Maranhão, Brazil'. L'Homme 126-128, XXXIII(2-4): 235-258.

Basso, Ellen (ed.) 1977. 'Carib-speaking Indians. Culture, society and language.' University of Arizona Press, Tucson, Arizona.

Basso, Ellen 1977. 'The status of Carib ethnography.' Pp. 9-22 in Basso (ed.).

Beckerman, Stephen 1979. 'The abundance of protein in Amazonia: a reply to Gross'. American Anthropologist 81: 533-560.

Behrens, Clifford A. 1981. 'Time allocation and meat procurement among the Shipibo Indians of eastern Peru'. Human Ecology 9(2): 189-221.

Beissinger, Steven R. & N.F. Snyder (eds.) 1992. 'New World parrots in crisis. Solutions from conservation biology' . Smithsonian Institution Press, Washington and London.

Berlin, Brent, D. Breedlove & P. Raven 1974. 'Principles of Tzeltal plant classification: an introduction to the botanical ethnography of a Mayan-speaking people of the highland Chiapas' . Academic Press, New York.

Berlin, Brent 1992. 'Ethnobiological Classification' . Princeton University Press, Oxford.

Bishop, Andrew 1996. 'Baseline Document on Land Use in Guyana'. Report prepared on behalf of the Government of Guyana.

Brett, William H. 1868. 'The Indian tribes of Guiana: their conditions and habits' . Bell & Daldy, London.

Butt, Audrey J. 1954. 'Systems of belief in relation to social structure and organisation with refence to the Carib-speaking tribes of the Guianas'. Ph.D. thesis, University of Oxford.

Butt, Audrey J. 1970. 'Land use and social organisation of tropical forest peoples of the Guianas'. Pp. 1-17 in Garlick & Keay (eds.)

Butt-Colson, Audrey J. 1984. 'Conclusion'. Antropologica 62: 359-383.

Carneiro, Robert L. 1982. Comment on Johnson. Current Anthropology 23(4) : 418-9.

Chandarpal, Navin 1997. 'Sustainable development and exploratory forest leases in Guyana'. Statement from Presidential Adviser on Science, Technology and Environment, Office of the President, Georgetown. Available from World-wide web: URL

Christie, Michael 1997. 'Amazon Indians demand development rights' . Reuters, 31 Aug 1997.

Clay, J.W. 1988. 'Indigenous Peoples and Tropical Forests'. Cultural Survival, Cambridge, MA.

Cleary, David 1990. 'Anatomy of the Amazon gold rush' . MacMillan Press Ltd., London.

Clement, Charles R. 1993. 'Native Amazonian fruits and nuts: composition, production and potential use for sustainable development'. Ch. 11 (pp. 139-152) in Hladik. et al (eds.)

Colchester, Marcus 1993. 'Who's Who in Guyana's Forests'. World Rainforest Movement.

Colchester, Marcus 1994. 'The New Sultans'. The Ecologist 24(2): 45-52.

Colchester, Marcus 1997. 'Guyana - Fragile frontier. Loggers, miners and forest peoples'. Latin American Bureau, London.

Collar, N.J. & A.T. Juniper 1992. 'Dimensions and causes of the parrot conservation crisis'. Pp. 1-24 in Beissinger & Snyder (eds.)

Conklin, Harold C.1954. 'The relation of Hanunoo culture to the plant world'. Ph.D. dissertation, Yale University, New Haven.

Conservation International 1997. Report on Guyana Workshop. Available from World-wide web : URL

Daly, Vere T. 1975. 'A short history of the Guyanese people'. MacMillan, London.

Descola, Phillippe 1994. 'In the society of nature. A native ecology in Amazonia'. English translation by Nora Scott. Cambridge University Press, Cambridge.

Drummond, Lee 1977. 'The outskirts of the earth: a study of Amerindian ethnicity on the Pomeroon river, Guyana.' Ph.D. thesis, University of Chicago.

Durham, W. H. 1982. 'Interactions of genetic and cultural evolution: models and examples'. Human Ecology 10: 289-323.

Dwyer, Peter D. 1983. 'Etolo hunting performance and energetics'. Human Ecology 11: 143-72.

Dwyer, Peter D. 1990. 'The pigs that ate the garden'. University of Michigan Press, Ann Arbor.

Eden, Michael J. & J.T. Parry (eds.) 1986. 'Remote Sensing and Tropical Land Management'. John Wiley & sons Ltd.

Eden, Michael J. 1986. 'Monitoring indigenous shifting cultivation in forest areas of southwest Guyana using aerial photography and landsat'. Pp. 255-277 in Eden & Parry (eds.).

Edwards, Stephen R. 1992. 'Wild bird trade: Perceptions and management in the Cooperative Republic of Guyana'. Pp. 77-85 in Thomsen et al (eds).

Ek, Renske 1996. 'Botanical richness of a part of the Guyana shield: the Guianas'. BOS Nieuwsletter 15(2), no. 34: 13-17.

Ellen, Roy F. 1986. 'Ethnobiology, cognition and the structure of prehension: some general theoretical notes'. Journal of ethnobiology 6 (1): 83-98.

Ellen, Roy F. 1993. 'The cultural relations of classification'. Cambridge University Press, Cambridge.

Evans, Clifford & Betty J. Meggers 1960. 'Archeological investigations in British Guiana'. Smithsonian Institution Bureau of American Ethnology, bulletin 177. Government printing office, Washington DC.

Farabee, William C. 1918. 'The Central Arawaks'. University of Pennysylvania Museum, Anthropological Publications vol. 9.

Farabee, William C. 1924. 'The Central Caribs'. University of Pennysylvania Museum, Anthropological Publications.

Favre, David S. 1989. 'International trade in endangered species: a guide to CITES'. Martinus Nijhoff, Dordrecht, Netherlands.

Feer, Fran¨ois 1993. 'The potential for sustainable hunting and rearing of game in tropical forests'. Ch. 59 (pp. 691-708) in Hladik et al (eds.).

Fock, N. 1963. 'Waiwai. Religion and society of an Amazonian tribe.' National museum, Copenhagen.

Forte, Janette 1990a. 'The Populations of Guyanese Amerindian Settlements in the 1980s'. Amerindian Research Unit, University of Guyana.

Forte, Janette 1990b. 'A preliminary assessment of the road to Brazil'. Amerindian Research Unit, University of Guyana.

Forte, Janette 1993. 'Amerindians and Poverty'. Background paper. Amerindian Research Unit, University of Guyana.

Forte, Janette 1995a. 'Amerindian land issues'. Background paper. Amerindian Research Unit, University of Guyana.

Forte, Janette 1995b. 'Forestry and Amerindians'. Background paper. Amerindian Research Unit, University of Guyana.

Forte, Janette 1997. 'Mining in Guyana and its impact on Amerindian communities in the Upper Mazaruni District'. Pp. 68-83 in Williams et al (eds.).

Forte, Janette & Anne Benjamin 1993. 'The road from Roraima state'. Manuscript, originally serialised in the Sunday Stabroek newspaper, 14/11/93-19/12/93.

Foster, Nancy Fried 1990. 'Contexts of Guianan Amerindian Identity'. Ph.D. dissertation, Columbia University.

Fox, Desrey C. & George Danns 1993. 'The indigenous condition in Guyana'. University of Guyana, Turkeyen, Georgetown.

Furley, P.A. (ed.) 1995. 'The forest frontier. Settlement and change in Brazilian Roraima'. Routledge, London.

Garlick, James P. & R. W. J. Keay (eds.) 1970. 'Human ecology in the tropics'.

GHRA 1997. 'Legality of new Timber deals open to question'. Guyana Human Rights Association Executive Committee news release, 10 April 1997. Available from world-wide web:URL gopher://

Gillin, John 1936. 'The Barama River Caribs of British Guiana'. Papers of the Peabody Museum of American archaeology and ethnology, Harvard University , vol. 14(2).

Gross, Daniel R. 1975. 'Protein capture and cultural development in the Amazon basin'. American Anthropologist 77: 526-549.

Hames, Raymond B. & William T. Vickers 1982. 'Optimal diet breadth theory as a model to explain variability in Amazon hunting'. American Ethnologist 9(2): 358-378.

Hames, Raymond B. & William T. Vickers (eds.) 1983. 'Adaptive responses of native Amazonians'. Academic Press Inc., London.

Hames, Raymond B. & William T. Vickers 1983a. Introduction (pp.1-26) in Hames & Vickers (eds.).

Hamilton, W. D. 1964. 'The Genetical Evolution of Social Behaviour'. Journal of Theoretical Biology 7, 1-52.

Hawkes, Kristen, K. Hill & J.F. O'Connell 1982. 'Why hunters gather: optimal foraging and the Aché of Eastern Paraguay'. American Ethnologist 9(2): 379-398.

Hecht, Susanna B. 1992. 'Valuing land uses in Amazonia: colonist agriculture, cattle, and petty extraction in comparative perspective'. Ch.19 (pp. 379-399) in Redford & Padoch (eds.).

Hemming, John 1995. 'Indians, cattle and settlers.' Ch.2 (pp. 39-67) in Furley (ed.).

Henfrey, Colin V.F. 1995. 'Democracy and the sustainability of livelihoods and natural resources: the issues currently underlying Amerindian development options'. Pp. 72-75 in ARU [1995].

Hillhouse, William 1979. 'Indian notices'. National commision for research materials on Guyana, Georgetown.

Hladik, C.M., A. Hladik, O.F. Linares, H. Pagezy, A. Semple & M. Hadley (eds.) 1993. 'Tropical forests, people and food. Biocultural interactions and applications to development'. Man and the Biosphere series, UNESCO, Paris.

Hyndman, David C. 1984. 'Hunting and the classification of game animals among the Wopkaimin'. Oceania 54(4): 289-309.

Im Thurn, Everard F. 1883. 'Among the Indians of Guiana'. Kegan Paul, Trench & Co., London.

IUCN 1996. 'Red list of threatened animals'. International Union for the Conservation of Nature, Gland, Switzerland. Updated version available on world-wide web: URL

Johnson, A. 1974. 'Ethnoecology and planting practices in a swidden agricultural system'. American Ethnologist 1: 87-101.

Johnson, Allen. 1982. 'Reductionism in cultural ecology: the Amazon case'. Current Anthropology 23 (4) : 413-428.

Kahn, Francis 1993. 'Amazonian palms: food resources for the management of forest ecosystems'. Ch. 12 (pp. 153-162) in Hladik et al (eds.).

Kocher-Schmidt, Christin, O. Kortendick & L. Rival n.d. 'Research guidelines for APFT researchers'. Unpublished manuscript.

La Rose, Jean 1994. 'The Price of Progress'. Guyana Review, June 1994: 18-19.

Lescure, Jean Paul & F. Pinton 1993. 'Extractivism: a controversial use of the tropical ecosystem'. Ch. 65 (pp. 767-774) in Hladik et al (eds.).

Linares, Olga F. 1976. 'Garden hunting in the neotropics'. Human Ecology 4(4): 331-351.

MacMillan, Gordon 1995. 'At the end of the rainbow ? Gold, land and people in the Brazilian Amazon'. Earthscan publications Ltd., London.

Martin, Gary 1995. 'Ethnobotany - a methods manual'. Chapman & Hall, London.

Meggers, Betty 1954. 'Environmental limitation on the development of culture'. American Anthropologist 56: 801-824.

Meggers, Betty 1996. 'Amazonia. Man and nature in a counterfeit paradise', revised edition. Smithonian Institution Press, Washington and London.

Menezes, Mary N. 1979. 'The Amerindians in Guyana 1803-73: a documentary history.' Frank Cass, London.

Mentore, George 1983-4. 'Wai-wai labour relations in the production of cassava'. Antropologica 59-62: 199-221.

Mentore, George 1995. 'Peccary meat and power among the Waiwai Indians of Guyana'. Journal of Archaeology and Anthropology 10: 19-33.

Migliazzi, E.C. 1978. 'The integration of the indigenous peoples of the territory of Roraima, Brazil'. International Working Group for Indigenous Affairs document 32, IWGIA, Copenhagen.

Moran, Emilio F. (ed.) 1984. 'The ecosystem concept in anthropology'. Westview Press, Boulder, CO.

Moran, Emilio F. (ed.) 1990. 'The ecosystem approach in anthropology : from concept to practice'. University of Michigan Press, Ann Arbor.

Moran, Emilio F. 1993. 'Through Amazonian eyes. The human ecology of Amazonian populations'. University of Iowa Press, Iowa City.

Mulliken, Teresa A., S.P. Broad & J.B. Thomsen 1992. 'The wild bird trade - an overview'. Pp. 1-41 in Beissinger & Snyder (eds.).

Myers, Iris 1944-6. 'The Makushi of British Guiana'. Reprinted from Timehri, no. 26: 66-77 (part 1), no. 27: 16-38 (part 2).

Nations, James D. 1992. 'Xateros, Chicleros, and Pimenteros: harvesting renewable tropical forest resources in the Guatemalan Petén'. Ch. 10 (pp. 208-219) in Redford & Padoch (eds.).

Posey, Darrell A. et al 1984. 'Ethnoecology as applied anthropology in Amazonian development'. Human Organisation 43: 95-107.

Posey, Darrell A. & W. Balée (eds.) 1989. 'Resource Management in Amazonia: Indigenous and Folk Strategies - Advances in Economic Botany 7'. New York Botanical Garden.

Potter, Lesley 1993. 'The Amerindians of Guyana and their environment'. History Gazette 52, 2-19.

Rappaport, Roy A. 1984. 'Pigs for the ancestors. Ritual in the ecology of a New Guinea people'. 2nd. ed., enlarged. Yale University Press, New Haven, CT.

Redford, Kent H. & C. Padoch 1992. 'Conservation of Neotropical forests. Working from traditional resource use'. Columbia University Press, New York.

Redford, Kent H. 1993. 'Hunting in neotropical forests - a subsidy from nature'. Ch. 19 (pp. 227-246) in Hladik et al (eds.).

Riviere, Peter 1972. 'The forgotten frontier - ranchers of northern Brazil'. Holt, Rinehart & Winston inc., New York.

Riviere, Peter 1984. 'Individual and society in Guiana - a comparative study of Amerindian social organisation'. Cambridge University Press, Cambridge.

Robinson, John G. & K.H. Redford (eds.) 1991. 'Neotropical wildlife use and conservation'. University of Chicago Press, Chicago & London.

Roth, Walter E. 1924. 'An inquiry into the arts, crafts and customs of the Guiana Indians'. Smithsonian Institution Bureau of American Ethnology, annual report 38, pp 25-745. Government printing office, Washington DC.

Roth, Walter E. 1929. 'Additional studies of the arts, crafts and customs of the Guiana Indians, with special reference to those of southern British Guiana'. Smithsonian Institution Bureau of American Ethnology, bulletin 91. Government printing office, Washington DC.

Sanders, A. 1987. 'The powerless people'. Macmillan, London & Basingstoke.

Schomburgk, Richard 1922. 'Travels in British Guiana 1840-1844, vol. 1'. Translated by W. E. Roth. The Daily Chronicle Ltd., Georgetown.

Schomburgk, Richard 1923. 'Travels in British Guiana 1840-1844, vol. 2'. Translated by W. E. Roth. The Daily Chronicle Ltd., Georgetown.

Shook, Josh 1995. 'A participatory approach to research and development, as applied in the Amerindian communities of Guyana'. University of Guelph, Ontario. Available from World-wide web: URL

Sizer, Nigel 1996. 'Profit without plunder: reaping revenue from Guyana's tropical forests without destroying them'. World Resources Institute, Washington. Available from World-wide web: URL

Spinner, Thomas J. Jr. 1984. 'A political and social history of Guyana, 1945-1983'. Westview Press, Boulder, Colorado.

Sponsel, Leslie E. 1986. 'Amazon ecology and adaptation'. Annual Review of Anthropology 15: 67-97.

Sponsel, Leslie.E. (ed.) 1995. 'Indigenous Peoples and the Future of Amazonia'. University of Arizona Press.

Steward, Jane C. & R.F. Murphy (eds.) 1977. 'Evolution and ecology. Essays on social transformation by Julian H. Steward'. University of Illinois Press, Urbana.

Steward, Julian H. (ed.) 1946. 'Handbook of South American Indians'. Smithsonian Institute, Washington.

Steward, Julian H. 1977. 'Tappers and trappers: parallel processes in acculturation'. Ch. 7 (pp.151-179) in Steward & Murphy (eds.).

Tang, Gloria T. 1995. 'Towards a regional plan for the Rupununi district, Region 9 of Guyana'. Ph.D. dissertation, Walden University.

Thomas, Clive 1993. 'Lessons from experience: structural adjustment and poverty in Guyana'. Social and Economic Studies 42(4), 133-184.

Thompson, Alvin 1987. 'Colonialism and underdevelopment in Guyana 1580-1803.' Carib Research & Publications Ltd., Bridgetown, Barbados.

Thomsen, Jorgen B. & T.A. Mulliken 1992. 'Trade in Neotropical psittacines and its conservation implications'. Pp. 221-239 in Beissinger & Snyder (eds.).

Thomsen, Jorgen B., S.R. Edwards & T.A. Muliken (eds.) 1992. 'Dimensions, conservation and management of wild birds in trade'. TRAFFIC International.

Townsend, W.R. 1995. 'Cultural teachings as an ecological data base : Murui (Witoto) knowledge about primates'. Latinamericanist 31(1): 1-7.

Trexler, Mark C. 1990. 'The convention on international trade in endangered species - political or conservation success'. Ph.D. dissertation, University of Harvard.

Vickers, William T. 1991. 'Hunting yields and game composition over ten years in an Amazonian Indian Territory'. Pp. 53-81 in Redford & Robinson (eds.).

Vickers, William T. 1993. 'Changing tropical forest resource management strategies among the Siona and Secoya Indians'. Ch. 38 (pp. 463-478) in Hladik et al (eds.).

Viveiros de Castro, Eduardo 1996. 'Images of nature and society in Amazonian ethnology'. Annual Review of Anthropology 25: 179-200.

Williams, Patrick E., J.T. Parry & M.J. Eden (eds.) 1997. 'Land use, land degradation and land management in Guyana'. Commonwealth Geographical Bureau.

World Resources Institute News Release, 4th. March 1997. WRI Publications, Baltimore, MD.

Yde, Jens 1960. 'Agriculture and division of work among the Waiwai'. Folk 2, 83-98.

Yde, Jens 1965. 'Material culture of the Waiwai'. National museum, Copenhagen.


1. No satisfactory nomenclatural consensus for distinguishing the Carib tribe from the Carib linguistic group of which it is a part. Here I follow the practice of using 'true Carib' to refer to the former, and 'Carib' alone or 'Carib tribes' for the latter, derived from Riviere [1984].

2. The concepts of 'acculturation', in both senses, do remain useful in social science so long as their limitations are recognised, and are employed, with this reservation, on several occasions in this text.

3. Several legitimate spellings of this word exist, and modern literature includes all the alternatives Wapisana, Wapisiana, Wapishiana and Wapixiana. In addition, Farabee [1918: 13 fn.1] lists several altenative spellings, most of which seem to be obsolete: Wapityans, Wapsichana, Wabijanas, Mapisianas, Uabixanas, Uapixianas, Uapichanas, Vapeschanas. The usage of 'Wapishana' in the present document is based upon what appears to be standard practice in the recent Guyanese literature (eg. Forte [1990], ARU [1992]).