Australian Mekong Resource Centre, School of Geosciences, University of Sydney, NSW 2006 Australia E-mail email@example.com
The Mekong River Basin is a highly diverse ichthyofaunal resource and a highly productive fishery in both subsistence and commercial terms, which has come under increasing stress. About this there is general agreement, but beyond these generalities the level of agreement rapidly dissipates. The politics of fisheries knowledge in the Mekong River Basin involves tensions along a number of lines: scientific and indigenous knowledge of fisheries; basic science and EIA-driven fisheries studies; culture and capture fisheries knowledge; taxonomic and livelihood-oriented fisheries research; NGO and governmental articulations of the causes of fisheries decline; and fisheries consumption and production statistics used by riparian countries and those produced by the Mekong River Commission. In this paper, I examine fisheries knowledge in the Mekong River Basin in the context of the politics of its production and ownership. The paper examines the tensions over fisheries knowledge in an attempt to direct attention to the circumstances of its production. The purpose of this approach is to highlight for fisheries managers and river basin managers more generally, the significance of understanding the politics of knowledge as a pre-requisite for using such knowledge as a management input. I argue that it is not sufficient to come up with expert knowledge that is privileged as "best estimate", particularly in the highly complex and politicised power/knowledge milieu that is prevalent in the Mekong River Basin. Rather we need to explore and develop approaches to knowledge that accommodate and go some way to resolving different epistemologies, through an inclusive and more culturally grounded research agenda.
Few doubt the significance of the fisheries of the Mekong River, its tributaries and its delta. The vastness and diversity of the fishery are well known, as is the fact that it is important for the livelihood of many, if not most, of the more than sixty million people who live in the Mekong River Basin. Most recognise that there are extractive and environmental pressures on the resource. Yet beyond these basic points of agreement, there is a lot of uncertainty, debate and tension over fisheries knowledge. Informed policy needs to be cognisant of the intricacies of knowledge production. While many of the tensions and points of disagreement are familiar to researchers and others who have worked in the area of Mekong fisheries, there has been relatively little structured investigation or discussion about the circumstances of knowledge production, appropriation and its significance for policy.
The politics of knowledge are significant in many circumstances and are certainly not the preserve of Mekong fisheries. Knowledge and power are intricately bound up and there is a continuing tension between positivist, objectified knowledge in the guise of science and a relativist, contextual epistemology in the guise of indigenous knowledge. Intermediate positions include the concept of situated knowledge, which does not negate the existence or significance of real-world facts, but at the same time points us to the circumstances of knowledge production and its "embodiment" in the sense of existing through the scientist or other producer of such knowledge (Haraway 1996).
In this paper, I briefly discuss the politics of environmental knowledge in the Mekong Basin and the significance of such politics for fisheries management, before identifying key lines of tension in fisheries knowledge production. The paper is not encyclopaedic about types of knowledge or forms of knowledge production, but it does try to cover key areas in which understanding the epistemology of fisheries is important for those working at policy and practical levels in the Mekong River Basin.
THE POLITICS OF ENVIRONMENTAL KNOWLEDGE IN THE MEKONG
Rapid economic development in the Mekong Region and more specifically in the Mekong River Basin has generated increasing environmental concern. The environment has become an arena of critique and debate over the most appropriate path of development for the countries and peoples of the Mekong, but sustainable development discourses have also allowed all to make claims to environmental concern. As different players have taken on the environmental mantle, environmental knowledge has become central to tensions between different interests.
The politics of environmental knowledge reflects the variegated interests within the Basin, delimited geographically, sectorally and socio-politically (Hirsch and Cheong 1996). Geographically, the division of the basin into six sovereign states means those different countries respective interests shape the way in which environmental knowledge is produced and treated. The division between China and the lower Mekong countries is sharpest here. A recent Chinese perspective on fisheries and other environmental implications of the development agenda in Yunnan (He Daming 2002) reflects the ways in which scientists rationale can be shaped by their countrys specific interest.
Sectoral tensions reflect the different charges that fisheries, hydropower, environmental and other agencies may have, showing that the politics of a shared river basin are not delineated only along national lines. Within government as well, there can be quite different perspectives between local and national government, with local authorities sometimes - but not always - better aware of and more responsive to local realities and concerns. Socio-politically, the power divides between small-scale farmers and fishers, government agencies, corporate interests in the resource base and the scientific community are reflected quite openly in some countries (notably Thailand) and remain well below the surface in others (notably Laos and Viet Nam). Nevertheless, the politics of knowledge remain relevant in all cases, perhaps nowhere more so than in fisheries.
KEY LINES OF TENSION IN FISHERIES KNOWLEDGE PRODUCTION
Key lines of tension mark out the current state of knowledge of fisheries in the Mekong River Basin and the ways in which this knowledge affects policy formulation and implementation. A summary with illustrative examples is presented for each of several "axes of tension" below. Two caveats are in order. First, there are many shades of grey and these "axes" are drawn to illustrate lines of tension rather than to suggest absolute polarisation or dichotomies. Second, there is a good deal of overlap and resonance between the axes, for example between scientific and local knowledge on the one hand and government and NGO tensions on the other.
SCIENTIFIC AND INDIGENOUS KNOWLEDGE: WHOSE KNOWLEDGE COUNTS?
Research on fisheries in the Mekong and its tributaries has focused on taxonomy, migrations, trends in fish stocks, fish ecology and fishing-based livelihoods. Direct sampling has been difficult for logistical reasons, for reasons of access by scientists until recently (including the years of conflict) and due to the very limited or non-existent local scientific research infrastructure in Laos and Cambodia. The very diversity and highly migratory nature of the Mekong fishery also pose challenges to sampling-based research.
Of course, there is an enormous repository of knowledge about the Mekong fishery in the communities who have long been dependent on the resource and who daily observe the spawning, migratory and feeding habits of different fish species. There has been increasing recognition over the past two decades that local knowledge "counts" (Chambers 1983). International symposia and publications on the significance of indigenous knowledge have moved well beyond the anthropological and ethno-ecological arena in which the subject has been developed since the 1950s (Conklin 1954). Yet, indigenous knowledge has continued to be ignored or treated by many resource managers and developers as unscientific, hence of little interest or value. The low status of indigenous knowledge and the pejorative connotation of what is sometimes termed "anecdotal evidence" continue to devalue such knowledge in a research context. This has immediate implications where, for example, the only evidence in decline of a fishery is based on such knowledge, such as in the case of the Se San River where local reports of fisheries impacts of upstream hydropower development at Yali Falls in Viet Nam are not accepted because there is no scientific data to back them up, compounded in this case by the lack of baseline information in the EIA.
There have been recent accommodations between scientific research and indigenous knowledge. In particular, the Assessment of Mekong Fisheries Component (AMFC) of the Mekong River Commissions (MRC) Fisheries Program has employed and developed techniques for using local ecological knowledge (LEK) to develop an understanding of fish migrations on the Mekong mainstream and tributaries (Poulsen and Valbo-Jorgensen 2001). This is a breakthrough in the application of fishers own knowledge for large-scale fisheries management at a basin level. Nevertheless, while natural and social scientists have come to recognise the significance and value of indigenous knowledge, there remain deeper epistemological questions of ownership. Research that "mines" LEK for publication in scientific papers or for development of national or basin-wide policy formulation is inherently different to a participatory research framework that works with local fishers and their knowledge as the basis for local management (Baird 1999; WWF 2002). Adaptive management is a promising research/management approach that incorporates this type of participation and local iteration between research and practice (Garaway et al. 2002). Another mode in which fishers knowledge is used is for articulation of those fishers interests vis-a-vis developments that actually or potentially threaten the fishery (Searin 2002).
Earlier tensions between measurement based on catch per unit effort (CPUE), on the one hand (Warren, Chapman and Singhanouvong 1998) and understandings of the local fishery through intimate knowledge of community practices and familiarity with villagers on the other (Baird 1996), have prompted a syncretic approach whereby methods such as hydro-acoustic sampling of deep pools is being combined with CPUE measurements to assess the effectiveness of management through establishment of fish sanctuaries that are based on local knowledge and community-based processes.
Studies recently undertaken to monitor the fisheries benefits against the costs (in foregone revenue from electricity production) of keeping open the gates of Pak Mun Dam provide another interesting example of tensions between academic science and research based on local knowledge. The Thai government commissioned a team led by Dr Kanokwan Phankasem of Ubonratchathani University to carry out a range of studies, notably fisheries research, in order to monitor the benefits of keeping the dam gates open. The researchers came up with four options based on this research, which found that there were indeed significant livelihood and biodiversity impacts of closing the dam gates and they recommended adoption of the fourth of these - keeping the dam gates open for five years to further monitor the impacts while power demand in Thailand remained well below the countrys combined generating capacity. However, the government opted for a "compromise" option, closing the dam for 8 months and opening the gates for 4 months during the wet season. In contrast, villagers along the Mun River organised their own research program under the title Tai Baan Research (Settrachau 2002). As Poh Dam, one of the villagers involved in this program stated:
If we have researchers here, we fear that they cannot get the information straight, or they cannot do it entirely correctly. Since they only live in the town, how can they know where the fish live, or where they get together in a large number or what they eat? They will end up having to ask the villagers. We think we ought to collate the information ourselves, as outsiders will not understand our way of life. We are the ones affected by the project and our resources have been destroyed (Assembly of the Poor 2002).
The politics of science versus indigenous research is thus based not only on the quality or reliability of information, but also on questions of ownership and the uses to which different types of fisheries knowledge are put. In the accommodation of the two types of knowledge we see a bi-directional process: on the one hand, recognition by the science community of the value and significance of local knowledge and interest in developing tools to make better and systematic use of that knowledge and on the other hand local communities empowerment by putting their own knowledge base to use in a more systematic and legible way:
The significance of the Tai Baan Research lies in empowering communities, equipping them with the tools to turn their local wisdom into the form of written documents, collectively produced and owned by the communities producing them (Traisawasdichai 2002).
SCIENCE AND CONSULTANCY: HYDROPOWER AS A KNOWLEDGE DRIVER
The politics of funding are an important element of knowledge production in any setting. Studies of fisheries in the Mekong are funded from four main sources, each with quite specific implications for the type of knowledge produced, its ownership and its input into the policy process. First, scientific research is funded by project grants such as the work of Tyson Roberts supported by the Smithsonian Institution, or the work of eminent fish taxonomists such as Walter Rainboth and Maurice Kottelat with North American and European science grants - although some of this work was also supported by FAO. Second, development and management-oriented research initiatives are funded largely through the MRC, most notably the fisheries program supported by DANIDA. Other, much smaller sources of funding in this category come from development research agencies such as IDRC (Canada) and ACIAR (Australia) and involve collaborations between foreign and local researchers, usually on a specific applied management problem but also, in the early phase of IDRC in quite basic research, for example, involvement in migration studies at Khone Falls. Third, smaller scale fish studies have been supported by non-governmental organizations such as Searin (Southeast Asian Rivers Network), involving local communities documenting their own fisheries practices, knowledge and livelihood dependence in the context of threats to the fishery from projects such as hydropower and blasting of rapids for navigation (SEARIN 2002). Finally and in many cases dominating the knowledge production process, is the swathe of consultancies associated with large scale resource projects (notably dams and mines) that require environmental impact assessments (EIAs) including fisheries impact studies.
A major problem in determining the impacts of a number of recently and not so recently completed dams on Mekong tributaries has been the paucity of baseline studies. Yali Falls, Nam Song and Theun-Hinboun dams, all of which appear to have had major fisheries impacts, were built with quite minimal fisheries research as part of their EIAs (Hirsch 2001). Where fisheries studies are carried out, for example in the case of the Sepon gold and copper project, the EIA is based on data collected over a very short period, almost always during a single season. Yet the natural seasonal and annual variability demands much longer research for a baseline against which impacts can be measured reliably.
Perhaps the most significant political aspect of consultant-generated knowledge is its ownership by the client who commissions it. This lends such knowledge to direct or indirect manipulation - direct in the selective use or delayed release of consultancy reports by project holders when there are potentially embarrassing or costly findings, indirect in the holding back by consultants themselves of findings, or recommendations based on such findings, that might prejudice future contracts. Only in some cases (e.g. Warren 1999) do consultants have the courage to publish findings in a wider forum and more often than not they are contractually bound to keep findings confidential. Consultant-generated research is not normally subject to peer review, although recent moves at MRC have been a positive step in this direction. In some cases, post-impoundment studies of fisheries and other livelihood impacts of hydropower projects have been carried out. For example, ADB commissioned a post-impoundment impact study of Nam Song Dam after it became apparent that the earlier EIA work had not done a proper fish study and that certain recommendations (e.g. for aquaculture ponds) had not been acted upon. The consultants report found uncompensated losses valued at approximately US$2 million - largely due to impacts on fish catches - among 13 affected communities, but this report has yet to be released publicly more than 24 months after its completion and more than seven years after the dam was finished.
Publicly funded research at MRC presents an interesting intermediate case between consultancy and academic science modes of knowledge production. The non-riparian scientists employed through the fisheries program are on consultancy contracts, but there has been a healthy increase in level of research involvement by riparian nationals seconded from their respective ministries and departments. The sense of ownership of MRC knowledge and data and the level of independence of the research carried out has moved in a positive direction since 1995. The Fisheries Program newsletter, Catch and Culture, produces articles that demonstrate a considerable degree of openness and independence in an institution whose riparian member states tend to give fisheries quite a low priority in development planning - with the relative exception of Cambodia. As a number of publications quoted in this paper attest and indeed in the holding of the LARS2 conference for which an earlier version of this paper was presented, there are positive moves toward a more accountable and open fisheries knowledge production process. Nevertheless, there remains room for further "indigenization" of research - not a single one of the papers presented over three days at LARS2 was delivered by a riparian national.
CULTURE AND CAPTURE FISHERIES: DEVELOPMENTALISM AND ITS IMPLICATIONS FOR KNOWLEDGE BIAS
Approximately 90 percent of the fish caught and consumed from the Mekong River and its tributaries are wild. Yet the significance of the capture fishery is not matched in relative terms by the resources that Departments of Fisheries put into enhancing knowledge. Aquaculture receives much greater attention as a development program and most of the traditional "research" effort has been at fisheries stations and on-farm ponds where exotic species (notably Chinese and Indian carp and Tilapia spp), but more recently indigenous pangasids in cage culture in the Mekong Delta (Trong, Hao and Griffiths 2002), are raised. The MRCS Fisheries Program component Aquaculture of Indigenous Mekong Species (AIMS) aims at domesticating other Mekong species. Meanwhile, significant research effort is now in the private sector, notably CPs sex-reversed Tilapia and with privatisation of knowledge production the aquaculture bias can be expected to intensify given that corporate profits can only be made in this sector.
Part of the explanation for the neglect of capture fisheries relative to their significance is the developmentalist notion that wild fisheries are essentially an undeveloped use of nature in a hunting and gathering mode, whereas modernity demands cultivation, sedentarisation, capitalisation, improved species and so on. A more fundamental reason is that most aquaculture work increases production, while most capture fisheries work is aimed at maintaining, or halting the decline, in natural fisheries. The former is seen as development, the latter is given less priority because it is not measurably making something better. The relative sizes of the capture and culture fisheries in the Mekong (estimated at approximately ten to one, respectively) is lost on most decision-makers. Seemingly dramatic - and, more importantly, readily measurable - increases in aquacultural production achieve a higher profile than the much larger, but less dramatic and less verifiable arresting of capture fisheries decline. These perceptions affect national governments and donor agencies alike, both of which have a hard time understanding the relativities between maintenance of a hugely important resource versus cultivation, sedentarisation, capitalisation, improved species and so on of a much smaller resource.
In part, also, there is an established idea of fisheries research as a technical, natural science issue, whereas livelihood-relevant researches into capture fisheries necessarily involves social aspects of fisheries management. Another, somewhat speculative reason that government fisheries agencies have stayed clear of wild capture fishery research is that findings on the significance of the fishery could quickly enter sensitive territory when such findings highlight the destructive aspects of projects in politically powerful areas of government, notably ministries of energy and industry. Indeed, aquaculture and reservoir fisheries are commonly put forward as mitigation for such projects.
More recent awareness of the size of the capture fishery, combined with interest in aquaculture using indigenous species, has somewhat blurred the lines between capture and culture fisheries research. Additionally, areas of overlap such as rice-field and enhanced back-swamp fisheries also provide examples of "greyer" areas between aquaculture and wild fish stocks. There has been less attention to more critical issues of incompatibility or competition between the two, for example where aquaculture is supported by the feeding of domesticated fish with fishmeal made from wild fish, or where the capture of wild juveniles and fry for aquaculture represents a threat to wild fish populations and hence to capture fisheries (van Zalinge 2002).
TAXONOMY, HABITATS AND ECONOMY: BIODIVERSITY AND LIVELIHOOD ORIENTED ENVIRONMENTALISM
The ichthyofaunal biodiversity of the Mekong River system is second only to that of the Amazon. More than 1 200 species are known and informed estimates suggest there may be up to 1 700 species in total. Not surprisingly, species identification has been a significant part of building fisheries knowledge in the Mekong (e.g. Kottelat 1998). A number of well-known fish experts have produced useful directories (Rainboth 1996; Kottelat and Whitten 1996). Most of the species that await discovery are likely to be endemic to remote montane tributary reaches in Laos, in relative terms having more scientific than livelihood importance. Furthermore, useful research on specific types of habitat (Poulsen et al. 2002) and on key wetlands (Daconto 2001) are starting to build up an increasingly sophisticated understanding of the ecosystem functions relevant to fisheries and their vulnerability to hydrological changes.
In a highly biodiverse, until recently little studied, scientifically important yet income-poor river basin whose fishery faces a range of extractive and environmental pressures, there is in principle an allocation of resources in fisheries research between taxonomy and scientific ecology, on the one hand, versus livelihood-oriented questions, on the other. This reflects a wider tension in global environmental concern between biodiversity as a "good" in itself, versus environment as livelihood. The different environmental concerns that prompt research interest in these alternative areas arise from quite different aesthetics and value sets, as suggested by Guha and Martinez-Alier (1997) in their book "Varieties of Environmentalism". In the Mekong, far greater resources go into livelihood studies, for the good reason that donor agencies and national governments alike prioritise peoples livelihoods over pure research for knowledge sake alone. In this sense, there is not so much a tension as a point of difference in conservation objectives that lie behind the research carried out in each case. On occasion, however, the biodiversity and livelihood knowledge can come into conflict, for example in the case of Probarbus jullieni (Sauvage 1880): as a CITES endangered species, research on this valuable table fish is partly hampered by the ban on taking specimens from the wild, even though it is in fact part of an established fishery in southern Laos.
On the other hand, concern with particular species can also have, through political means, a fisheries livelihood benefit. This is where spectacular species, particularly of larger fish species such as the giant catfish, serve as "flagship species" in support of greater attention to ecosystem preservation (Mattson et al. 2002).
NGOS AND GOVERNMENT: ENVIRONMENTAL AND EXTRACTIVE CAUSES OF FISHERIES DECLINE
One of the sharpest lines of tension in debates over development futures for the Mekong has emerged between advocacy-oriented and rural livelihood-focused non-governmental organizations, on the one hand and state agencies concerned with infrastructure development on the other. Fisheries have rapidly risen to prominence as a cause taken up by NGOs, primarily because of their importance to livelihoods of millions of people, in part because of their significance as environmental indicators of river health and in part due to their vulnerability as a livelihood resource to many destructive aspects of development occurring in or planned for the region.
Among the debates between NGOs and other actors is the cause of actual or perceived fisheries decline. That there is a perceived decline in fisheries is clear. The nature of this decline is more ambiguous, with both local actors (including fishers) and wider players often failing to distinguish between decline in the overall fish stock and decline in catches (Coates 2002). It is quite conceivable overall and in more specific local circumstances that there may be declines in stocks and associated CPUE, while the level of catch may be stable or increasing (through more "effort"), possibly but not necessarily to unsustainable levels. Even less clear are the reasons for fish decline, but it is here that "knowledge" tends to be declared with greatest certainty. Thus, small fishers tend to be blamed - through destructive practices, lack of management and so on - by state actors and by those with vested interests. Dams and other development-induced impacts tend to get the blame from affected fishers themselves and NGOs who help articulate their concerns and interests. This tension between environmental and extractive causes of fishery "decline" does not always follow the NGO-state actor axis, but it tends to serve the wider discursive interests of those with quite different interests in the rivers resources. Pak Mun is, again, a case in point, where EGAT has blamed fishers directly below the dam for extractive destruction of the fishery, while the blasting of rapids and blockage of the Mun River is the main point of contention by local fishers and NGOs working with them.
The International Rivers Network has been one of the more active international NGOs documenting fisheries impacts of dams. Seminal in this process was a study of the Theun-Hinboun Dam carried out by Bruce Shoemaker (Shoemaker 1998), which ADB tried to refute by sending in its own consultants. Since this time, a rather more measured approach has been taken and the Theun-Hinboun Power Company has accepted a considerable proportion of the responsibility for disrupted livelihoods (Theun-Hinboun Power Company 2000). Without the politicisation engendered by the NGO study, slammed at the time by the company and by the public agency that helped fund the dam, the knowledge of impacts would have remained firmly in the minds and communities of the impoverished fishers.
More recently, the series of studies supported through the MRC Fisheries Program has come out more firmly in identifying the anthropogenic environmental threats to fish abundance and diversity in the Mekong system. For example, a recent status review states quite firmly that the palliative notion of "mitigation" or "amelioration" of migratory impacts of dams by constructing fishways is simply not valid for larger projects (Sverdup-Jensen 2002). A political challenge will be to incorporate such conclusions into the basin development planning process.
RIPARIAN LINE DEPARTMENTS AND MEKONG RIVER COMMISSION: HOW MANY FISH ARE EATEN?
One of the most dramatic but also controversial "bits" of fisheries knowledge pertaining to the Mekong is the size of the fishery. The MRC Annual Report for 1996 stated that while the official statistics reported that only 360 000 tonnes of fish were caught in the Mekong Basin, data from MRC executed projects indicated that production may be as high as one million tonnes (MRC Annual Report 1996). In 2001, this was revised upward to a yield of two million tonnes as a "conservative estimate" (Sverdup-Jensen 2002). Further upward revisions are likely. These volumes make the Mekong by far the largest freshwater fishery of any river basin in the world and suggest a raw value of approximately US$1.4 billion (i.e. not taking into account the value generated through processing, transport and other multipliers, retailing and so on).
The figures for the total catch (production) are based on a surrogate measure, fish consumption. The bases for the MRC figures are 15 consumption surveys, five of which were conducted by MRC and ten by other agencies (Sverdrup-Jensen 2002). These were extrapolated to cover the entire LMB using population data from provinces whose characteristics most closely matched those of the survey sites. The basin-wide estimates are based on extrapolations from findings in aqua-ecologically representative areas.
A difficulty with the MRC figures is that they meet with considerable scepticism by the line agencies in each riparian country (although Coates (2002) reports that, since 1999, the Cambodian government has been using figures very similar to the MRC estimates for the country). The skepticism is based on a number of factors related to politics of fisheries knowledge. In part, there is an issue of sense of ownership. In part it is an issue of lack of consensus over methodology, with perceived biases in sampling and in interpretation of the data. In part it is due to the challenge to historically collected statistics that, if the MRCS studies are to be believed, grossly and systematically under-enumerated the fishery. It is perhaps in this context that most attention needs to be paid to the circumstances of knowledge production.
CONCLUSION: KNOWLEDGE, POLICY AND RELEVANCE
If knowledge is power, then knowledge is political and knowledge about fisheries is no exception. The multi-dimensional politicisation of fisheries knowledge in the Mekong along the lines of tension indicated above suggests some fundamental rifts, which need to be understood - if not necessarily bridged - if understanding of the circumstances of knowledge production is to better inform policy that seeks equitable and sustainable management of the fishery.
Ultimately, policy is concerned with relevance. Yet, what is relevant knowledge to the policy maker may seem different to that which is relevant to the scientist or the fisher, as indicated for example in the tension between biodiversity- and livelihood-oriented fisheries knowledge. Here there is an important distinction to be made between science and scientism, the latter being a mindset - a particular culture of knowledge - that rejects that which has been produced other than under very specific circumstances. Yet, becoming less scientistic does not necessarily mean becoming less scientific, so long as caveats are drawn. In many circumstances, the best available knowledge of the fishery on which to base policy is that of the fishers themselves.
The tension between science and other means of knowledge production is further complicated by questions of ownership. The realpolitik of connections between knowledge and policy determines that decision makers are most likely to act on understandings in which they have some stake. Externally produced knowledge, if paid insufficient attention to the circumstances, agencies and actors involved in its production, will carry much less weight than data generated by a more inclusive, participatory and almost certainly slower and more patient means.
Knowledge is power also through the pecuniary incentives to apply particular types of knowledge to decision making. The relationship between fisheries knowledge and large-scale resource development remains a fragile one and it is probably no coincidence that the consultancy-generated EIA knowledge remains the least open and accountable, particularly in the context of privatisation of large projects that puts significant areas of information under "commercial in confidence" wraps.
For policy makers, scientists and those who depend on the fishery, the public good is likely to best be served by transparent, inclusive and accountable knowledge production. This includes access by fishers and those working with them to scientific resources, as well as further development of syncretic methods to bring local knowledge into science.
The author wishes to acknowledge the helpful comments of an anonymous referee.
Assembly of the Poor, 2002. The return of pladaek: The struggle and changes after the opening of Pak Moon Dam, Villagers at Mae Moon Manyuen village, Assembly of the Poor.
Baird I.G., 1999. Towards Sustainable co-management of Mekong River inland aquatic resources, including fisheries, in southern Lao PDR. Pakse, Lao PDR, CESVI Cooperazione e Sviluppo, Agriculture and Forestry Division of Champassak Province.
Baird I.G, 1996. Inland community fisheries in southern Laos, Naga, The ICLARM Quarterly, January 1996: 13-15
Chambers R. 1983. Rural development: Putting the last first. London, Longman.
Coates D. 2002. Inland capture fishery statistics of Southeast Asia: Current status and information needs. Bangkok, Asia-Pacific Fishery Commission, Food and Agriculture Organisation of the United Nations, Regional Office for Asia and the Pacific.
Conklin H.C. 1954. An ethnoecological approach to shifting agriculture. Transactions of the New York Academy of Sciences, 172: 133-142.
Daconto G. ed. 2001. Siphandone wetlands. Bergamo, CESVI.
Garaway C.J., Arthur R. & Lorenzen K. 2002. Adaptive learning approaches to fisheries enhancements. final report to the DFID fisheries management science programme. London, MRAG. 54 pp
Guha R. &. Martinez-Alierin J. 1997. Varieties of environmentalism: Essays north and south. London, Earthscan.
Haraway D.J. 1989. Primate visions: Gender, race and nature in the world of modern science. New York, Routledge.
He Daming & Chen Linhui. 2002, The impact of hydropower cascade development in the Lancang-Mekong Basin, Yunnan. Mekong Update and Dialogue, 53: 2-4.
Hirsch P. 2001. Globalisation, regionalisation and local voices: The Asian Development Bank and rescaled politics of environment in the Mekong region. Singapore Journal of Tropical Geography, 223: 237-251.
Hirsch P. & Gerard Cheong. 1996. Natural resource management in the Mekong River Basin: Perspectives for Australian Development Cooperation. Canberra, AusAID.
Johnston R.J., Gregory D., Pratt G. & Watts M. 2000. The dictionary of human geography. Oxford, UK, Blackwell.
Kottelat M. 1998. Fishes of the Nam Theun & Xe Bang Fai basins, Laos, with diagnoses of twenty-two new species Teleostei: Cyprinidae, Balitoridae, Cobitidae, Coiidae and Odontobutidae, Ichthyological Exploration of Freshwaters, 91: 1-128.
Kottelat M. & Whitten A.J. 1996. Freshwater biodiversity in Asia with special reference to fish. Washington, DC, World Bank Technical Paper No. 343.
Mattson N.S., Buakhamvongsa K., Sukamasavin N., Nguyen Tuan & Ouk Vibol. 2002. Mekong giant fish species: On their management and biology. Phnom Penh, Cambodia, Mekong River Commission.
MRC Secretariat. 1997. MRC Program for fisheries management and development cooperation. Phnom Penh, Cambodia, Mekong River Commission.
Poulsen A. & Valbo-Jorgensen J. 2001. Local knowledge in the study of river fish biology: Experiences from the Mekong, MRC.
Poulsen A., Ouch Poeu, Viravong S., Suntornratana U. & Nguyen Thanh Tung. 2002. Deep pools as dry season fish habitats in the Mekong River Basin. Phnom Penh, Cambodia, Mekong River Commission.
Rainboth W.J., 1996. Field guide to the fishes of the Cambodian Mekong. Rome, Food and Agricultural Organisation of the United Nations.
SEARIN. 2002. Mekong rapids under fire. Bangkok Project For Rivers And Communities and Love Chiang Khong Group, Southeast Asia Rivers International Network & OXFAM America.
Settrachau C. ed. 2002. Mae Mun: Kaan klab maa khong khon haa plaa Mun River: The return of fishers. Bangkok, Southeast Asia Rivers International Network.
Shoemaker B. 1998. Trouble on the Theun-Hinboun: A field report on the socio-economic and environmental effects of the Nam Theun-Hinboun hydropower project in Laos. Berkeley, CA, USA, International Rivers Network. (Available at www.irn.org)
Sverdup-Jensen S. 2002. Fisheries in the Lower Mekong Basin: Status and perspectives. Phnom Penh, Cambodia, Mekong River Commission.
Theun-Hinboun Power Company. 2000. The Theun- Hinboun Power Companys mitigation and compensation program. Vientiane, Theun- Hinboun Power Company. (Available at www.adb.org)
Traisawasdichai M. 2002. The politics of research. Bangkok Post, 12 November 2002, p. 3.
Trong T.Q., Hao N.H. & Griffiths D. 2002. Status of Pangasiid aquaculture in Viet Nam. Phnom Penh, Cambodia, Mekong River Commission.
Van Zalinge N., Sopha L., Ngoe Peng Dun, Heng Kong & Valbo-Jorgensen J. 2002. Status of the Mekong Pangasianodon hypophthalamus resources, with special reference to the stock shared between Cambodia and Viet Nam. Phnom Penh, Cambodia, Mekong River Commission.
Warren T.J. 1999. A monitoring study to assess the localized impacts created by the Theun-Hinboun Hydro-Scheme on fisheries and fish populations: Final report to Theun-Hinboun Power Company, (Available at www.irn.org)
Warren T.J., Chapman G.C. & Singhanouvong D. 1998. The upstream dry-season migrations of some important fish species in the Lower Mekong River in Laos. Asian Fisheries Science, 11: 239-251.
Worldwide Fund for Nature WWF. 2002. Community based indicators for sustainable fisheries management in the Lower Mekong Basin of Lao PDR. Vientiane, WWF Lao Program.