Introduction
It was only mid-morning, and the heat was already strong. We were sitting in a shady spot under the house of Mai, a male ethnic Lao villager who has lived in Phluk village for all of his over 50 years of life and is a relative of the community’s female spirit medium, a 66-year-old farming-woman named On Niwan. Mai explained where the 14 different spirits channelled by the village spirit medium stayed. ‘Da Be Le (grandfather Be Le) used to live on Suan Island, but it was flooded when the dam [Lower Sesan 2 Dam, LS2] was built. We don’t know where Da Be Le went, or whether he is angry with us or not,’ commented Mai, with a serious look on his face. ‘Then there is Da Pa Kheuang, who lives downstream from the dam and still stays in the same old deep-water pool and rapids area.’ Mai stated that he had not heard of Da Pa Kheuang being angry, since he had not been displaced.
‘But we have heard that Kahomkor is angry,’ Mai suddenly blurted out, referring to the most important spirit that the village spirit medium channels. ‘He made the first deputy headman, Chan, sick. Chan had to make an offering to the spirit about 200 metres below the dam in a forested area,’ Mai continued. Crucially, Mai stated this as fact. He then explained that people in his village want Chan to establish a new spirit house (ta ho) for Kahomkor on the other side of the river from where its former spirit house was located before it was flooded by LS2’s reservoir. ‘Kahomkor hurt Chan because he did not arrange for a new spirit house for him quickly enough,’ asserted Mai.
Of particular interest, both Kahomkor and Da Be Le, but not Da Pa Kheuang, were said to be Khmer language-speaking spirits, while their wives, Nang Janlen and Nang Thale Thong, were identified as Lao speakers. Indeed, Kahomkor is a well-known spirit in Khmer-speaking parts of Cambodia, but Kahomkor has somehow been introduced into the spirit world of these ethnic Lao villagers. This seems to reflect the complex ethnic and linguistic circumstances in this part of northeastern Cambodia, with most of the communities along the lower Sesan River in Stung Treng province, northeastern Cambodia, being populated by ethnic Lao people, who mainly speak Lao. However, the area is located in the nation-state of Cambodia, and the Cambodian state is becoming increasingly economically, politically, and even culturally dominant there. Interestingly, Mai said that Kahomkor could understand Lao, but only ever spoke Khmer, mimicking the circumstances for some ethnic Khmer people who have moved into Lao villages in the area. In that these spirits are recognized locally as Khmer spirits who are perceived as having arrived more recently than the already existing Lao spirits, it seems that the two powerful male spirits represent the Cambodian state sending their ethnic Khmer representatives to the area, as has occurred many times in the past.Footnote 1 Mai mentioned that they were the spirits of powerful Cambodian officials.
It is certainly true that ethnic Lao people previously dominated this part of Cambodia, as this space was included in Laos until it was transferred to Cambodia in 1905.Footnote 2 Even after then, the Lao have often dominated, but since the 1960s, the government has frequently sent ethnic Khmer people to Stung Treng province to take up senior government positions, with the goal of gradually ‘Khmerizing’ the region.Footnote 3 In any case, Mai was certain that Kahomkor wanted his new spirit house to be as near to the old spirit house as possible. He was firm about that. Crucially, however, Mai emphasized that the spirit house could not be built just anywhere. Indeed, geography matters.Footnote 4
This initial vignette provides an indication of how ethnic Lao people living along the Sesan River in Stung Treng province see the relationship between spirits, dams, their own well-being, and space. Belief systems are crucial for interpreting the deep meanings of different circumstances, including those related to spirit mediums and contestation regarding a large hydropower dam, LS2. Here, we intend to contribute to the geographies of religion literature,Footnote 5 a field that has historically been quite Western dominated and has tended to focus on world religions, such as Christianity, Judaism, and Islam.Footnote 6
The focus is on spirit mediumship which, in the context of northeastern Cambodia, is intertwined with but believed by local people to be separate from Theravada Buddhism. This requires some explanation: on the one hand, almost all of the people who participate in spirit medium events self-identify as Buddhists, including the spirit mediums themselves. It is typical for locals to attend both Buddhist and spirit medium events organized in their communities. It is also certainly true that spirit mediums and Buddhism in this area have drawn from each other to a certain degree. For example, the sou khouan soul calling ceremony comes from animism, but chants done during spirit medium rituals are often versions of Buddhist chants. However, on the other hand, communities strictly separate Buddhist events from spirit medium events. Villagers attend both, but Buddhist monks do not attend spirit medium rituals, and spirit medium possession does not occur at Buddhist events. Thus, the two are both closely connected and also ontologically separated in the minds of the people.
LS2 is the largest hydropower dam built so far in Cambodia. Located on the lower Sesan River, and blocking both the Sesan and Srepok rivers in Sesan district in Stung Treng province, the dam has a 400 megawatt capacity.Footnote 7 The project was controversial from its conception,Footnote 8 but despite strong local and international opposition, the dam was ultimately built and completed in 2018 with Chinese government financing.Footnote 9 The project would not have been developed had it not been for financial support from the government of China,Footnote 10 as the LS2 was funded through the BRI, the Chinese government’s high-profile international development programme.
As already suggested in the opening vignette, spirit mediums sometimes play important roles in the social and cultural lives and the spatialities of rural ethnic Lao communities in northeastern Cambodia. Along with a variety of other beliefs related to spirits and magic, broadly defined, spirit mediumship can be seen as part of a religious belief system that influences how people see themselves and their relationships with others and the natural world around them. Thus, when a large infrastructure project that has profound impacts on nature and associated sacred spaces and beliefs—as is the case for LS2—it becomes evident that social and cultural worlds and the natural world are intertwined in particular ways. It also becomes clear how important shadow infrastructure is, that is, infrastructure that is not typically considered to be infrastructure, such as the human body, but which plays an important role in facilitating the flows and movements of people.
In this article, we take a feminist political ecology approach, partly because women spirit mediums have contested LS2 since before its construction began and have been directly impacted by the project, but also because the natural world and human beliefs associated with it are always important. We do not deconstruct the idea of nature itself, but we do complicate our understanding of nature by recognizing how ideas about spirits are intertwined with ideas about nature. In particular, we argue that apart from having serious material impacts, large dam infrastructure such as LS2 can also alter nature-society relations through variously affecting spirit mediums and their practices, and that these relationships are associated with ethnic and other identities that affect understandings regarding possibilities associated with the roles of spirits, spirit mediums, the river, power, and geography.
The article is organized as follows: in the next section, we lay out our theoretical framework, which is based on feminist political ecology. We then provide some background information about spirit mediums, following the Lao tradition, before briefly reviewing the literature related to spirits, spirit mediums, and hydropower dams in the Mekong region. This is followed by some background information about the LS2. We then examine how LS2 has influenced nature-society relations through affecting spirit mediums and their practices. We discuss how changes are occurring before finally providing some concluding comments.
This article is based on fieldwork conducted by both authors. The first author has been investigating Lao spirit medium practices in southern Laos, northeastern Cambodia, and northeastern Thailand for over 30 years, first as a non-governmental organization worker in the 1990s, later as a graduate student in the 2000s, and finally as a university professor in the 2010s, including in northeastern Cambodia, where this study is focused. Most recently, he conducted village-level fieldwork in Stung Treng province in June 2022. He is a white male, but speaks Lao very fluently and has many years of experience living and working with Lao people.
The second author is a PhD candidate who is a Lao American. He started investigating the LS2 dam and spirit mediums in 2019 and 2020, and returned for follow-up research in February and March 2022, as Covid-19 travel restrictions started to ease. Villagers, generally, were vaccinated and showed very little concern for the coronavirus, thus making access to villages possible. Semi-structured interviews were conducted with spirit mediums and others from two ethnic Lao villages impacted by the Lower Sesan 2 Dam in Stung Treng province—one in the reservoir area of the dam and the other downstream from the dam.
All interviews by both authors were conducted in Lao language. The second author was also able to observe spirit medium rituals in the villages at the homes of the spirit mediums, while the first author has witnessed spirit mediums in various contexts in southern Laos, northeastern Cambodia, and northeastern Thailand over many years.
Feminist political ecology in an infrastructuralized world
Feminist political ecology comes out of political ecology in geography. Political ecology was originally intended to bring together political economy and a broad interest in ecology,Footnote 11 although insights from postmodernism and poststructuralism were integrated into political ecology in the 1990s and 2000s.Footnote 12 Feminist political ecology has, most recently, put particular emphasis on inequality of various types, intersectionality, embodiment, emotions, and relationality.Footnote 13 Feminist political ecology, like feminist geography more generally, is also particularly attentive to critically assessing research methods, praxis, positionality, and reflexivity.Footnote 14
An understanding that men and women experience the environment, access resources, and make meaning and respond to environmental change differently from one another was crucial to early feminist political ecology.Footnote 15 More recently, Siemiatycki et al.Footnote 16 have focused on how unequal gender dynamics have become key determinants of how various kinds of infrastructure are produced, thus combining feminist political ecology with an increasing interest in infrastructure studies.
In many parts of the developing world where labour is gendered, women bear the physical and emotional burden of a wide variety of domestic jobs, including water collection. This creates various kinds of material and immaterial connections. For example, Sultana argues that resource struggles and conflicts pose not just material challenges, but also emotional ones.Footnote 17 Demonstrating how the practices and processes of women’s water access in Bangladesh not only serve to better texturize understandings of resource access, use, and control, she also shows how practices and processes ‘produce different kinds of emotional geographies’.Footnote 18 The distances to retrieve safe water, being told off while fetching water, and drinking unsafe water are some of the factors that women have to consider in navigating daily interactions with water, or with nature. Understanding the emotional geographies of resource access, use, and control complicates—while also enriching—our understandings about power relations and how they are negotiated in different social, political, and ecological contexts.
While feminist political ecology acknowledges and closely considers forms of difference that certain power structures sometimes render invisible, the subfield is continuing to push for more nuanced ways of understanding how gender intersects with other categories of difference, such as class, race, ethnicity, nationality, age, etc. For example, Mollett and Faria have called for more explicit consideration of race in feminist political ecology.Footnote 19 Examining the vulnerabilities that Indigenous Miskito women face in Honduras, they show how race and age further compound gendered oppressions. The example of an elderly Miskito dark-skinned woman is useful for making their point. Because men customarily clear land and women plant the seeds, the woman’s inability to clear and fence her land leaves it vulnerable to encroachment by Miskito and non-Miskito men and state intervention. Her dark skin subjected her to exclusion, as her involvement with an evangelical Christian group positioned her close to los moreño, an Afro-Indigenous community stereotyped as thieves and being ‘sneaky’.Footnote 20 Understanding how race and age further entangles analyses of gendered oppression is among the most important contributions of feminist political ecology.
Another contribution of feminist political ecology is the foregrounding of the roles of marginalized groups, including, but not limited to, women who engage in activist struggles related to environmental issues. While feminism has long been founded on ideas related to situated knowledges,Footnote 21 there is a growing recognition of the need to focus on particular knowledges and expressions of environmental concern in activism around growing global environmental movements.Footnote 22 Women’s involvement and leadership in environmental justice movements have historically been under-appreciated by scholars and environmental advocates. Sometimes they have been hidden from sight, either intentionally or due to various societal and structural impediments. Feminist political ecology works to rectify this by foregrounding the roles and voices of women and other marginalized groups, particularly when opposing large infrastructure projects that impinge on the socionatural forms of infrastructure that preceded them and supported local livelihoods. For example, Connie Campbell, in collaboration with the women’s group in Xapuri,Footnote 23 demonstrates the value of women’s work in the defence of forests in Xapuri, Acre, Brazil—forests that are important for their livelihoods. The authors demonstrate how women’s activities, previously undervalued in the economic life of their villages, were important in the extraction and processing of latex, in labour unions, and in public demonstrations in defence of forests against ‘the expansion of capital-intensive agricultural enterprises in the Amazon, most of which consisted of cattle ranches’.Footnote 24 Environmental movements such as this often mask the participation of historically underappreciated groups from public discourse, while others remain localized and couched in cultural-specific forms. Feminist political ecology is well suited for investigating these different forms, particularly where and when large infrastructure projects intrude on cultural understandings of place and environment. In Cambodia, Gillespie and Perry investigate the Tonlesap Lake and propose that it would be fruitful to combine feminist political ecology with legal geography.Footnote 25
In this article, we contribute to developing both feminist political ecology and infrastructure studies by revealing the roles of women in environmental struggles related to large-scale infrastructure and the geographies of religion, through taking an intersectional approach to analysing the role of spirit mediumship in contesting and otherwise engaging with the contested LS2. We also emphasize the politics of mobility, for both humans and spirits, in which we take historical mobilities and associated geographical understandings, and also new immobilities, seriously,Footnote 26 including the relationships between gender and mobility.Footnote 27 Crucially, we engage with types of Lao matrilineal feminist political ecologies that diverge from more common Global North forms of feminisms, and deserve much more attention than they have so far received.
Spirit mediums in the Lao tradition
Ethnic Lao people have complex belief systems associated with spirits, magic, and the natural world.Footnote 28 Spirit medium practices among ethnic Lao people have a long history and, critically, the vast majority of village spirit mediums in the Lao-speaking world (northeastern Thailand, Laos, and northeastern Cambodia) are women, typically older women. People to whom we have spoken do not have a clear explanation as to why these spirit mediums are mainly women. They typically just say that it has been the norm for as long as anyone can remember, but we suspect that it has a lot to do with the matrilineal nature of Lao society. In some cases, spirit mediums follow lineages of spirit mediums in their families, but not all do that. However, most of those who eventually become spirit mediums follow a path with similar key elements. First, a woman usually experiences chronic illness, the cause of which is uncertain and for which she is unable to find a cure. It is often believed that the person has become ill because she is resisting her fate to become a spirit medium. Thus, ill people typically dismiss such suggestions and refuse to comply. After some time, often many years, the ill person succumbs and agrees to accept her fate and becomes a spirit medium, at which time the illness suddenly goes away as the transformation to becoming a spirit medium occurs.Footnote 29 It appears that the illness that spirit mediums experience before eventually accepting their fate is what gives them some power.
Most rural communities have at least one key spirit medium, and sometimes that spirit medium is accompanied by a cult of other lower level spirit mediums who support the main spirit medium. This, for example, is the way it is done in parts of Chaiyaphum province, northeastern Thailand,Footnote 30 and in some other parts of Thailand.Footnote 31 However, in southern Laos and part of northeastern Cambodia, it is more typical for communities to each have a main spirit medium (nang thiem), along with two assistants, known as cha khoua (right assistant) and cha sai (left assistant). These two assistants do not channel spirits, but ensure that all the rituals are done correctly and generally support the spirit medium, who channels the spirit of an important person from the past, either a woman or, more likely, a man. That is why spirit mediums in southern Laos and northeastern Cambodia are known as nang thiem, or ‘fake women’, as they are women who channel the spirits of men.Footnote 32 They conduct their most important rituals at sacred places, where small altars known as ta ho are built to accommodate the rituals.Footnote 33
These spirits are thought of as moral beings, but it is unclear exactly why female spirit mediums typically channel the spirits of deceased men and our local informants have not been able to explain this either. However, it may be that these deceased men were understood to be the most powerful living men in their areas before passing away and so their spirits are also thought to be particularly powerful. That is, they are thought of as having more ‘natural potency’, to use Sarinda Singh’s term,Footnote 34 than other potential spirits of dead people who could be channelled by spirit mediums.
It is also unclear why the spirits being channelled by the medium who is the focus of this study are associated with different parts of the river or other spaces, as we have not encountered this with other spirits and spirit mediums we have studied in the past. It is therefore unclear if these circumstances are exceptional or more normal than we and other researchers have recognized so far. We hope that raising this issue will inspire others to engage more with this important topic.
Lao society is historically matrilineal and matrilocal. That is, when couples marry, it is the man who typically moves into his bride’s family’s house. If there is more than one daughter in the family, the older one is expected to marry first and later establish a separate household. The youngest daughter and her husband and children are expected to look after their parents in their old age. Thus, most land and other assets are inherited by the youngest daughter and her family. Land and houses are expected to be inherited through the female line. However, the public faces of most Lao nuclear families are typically men. Men take most village and government leadership roles, as women are expected to fulfil many of the domestic responsibilities and family financial management duties. We contend that this discussion regarding matrilinearity and spirit mediums in Lao society is crucial for ensuring that feminist political ecology engages with culturally appropriate structures and ontological categories, including those that are typically not important in the Global North or in the geographies of religion literature.
The spirit medium tradition is crucial for giving women a voice in community governance. Although the village chiefs are typically men, at least twice a year, during the third day of the third lunar month (typically in February) and the sixth day of the sixth lunar month (typically in May), spirit mediums will conduct community-wide events to bring on good rains for the upcoming agricultural season. During these events, spirit mediums will typically channel a male spirit, drink whiskey, and then clearly inform the village chief of anything that the spirit believes needs to be rectified to keep the village in good condition. The catch is that spirit mediums cannot be blamed for anything they say when they are channelling a spirit. That means that they can complain harshly and not risk any sort of retribution later. This shields them and gives them the opportunity to speak more frankly than most. We believe that this spirit medium practice has long been crucial for giving women an important, albeit informal, role in community governance. They act as a vital layer of governance, checking and balancing other community leaders. It gives them a sort of immunity, which really is significant. Thus, women spirit mediums have more important roles to play than simply reproducing cultural rituals and practices, although they do that too; they are given the opportunity to influence important aspects of community governance, using an embodied and potentially emotional framework, one that is fundamentally relational and intersectional. Tanabe also emphasized the gendered power dynamics associated with spirit mediums.Footnote 35 Indeed, power can shift according to place and time, and can be manifested in various ways.
Spirit mediums channel spirits at the request of villagers, usually to help deal with health-related problems that someone is experiencing, to ask the spirit medium to suggest some spiritual remedy for the illness,Footnote 36 but sometimes for other reasons, including for environmental activist purposes,Footnote 37 as will be discussed later in the article. In this regard, we are speaking of village spirit mediums who only occasionally play this role and are not professional spirit mediums. They mainly live in rural areas, in villages, and they sometimes channel many spirits. In Phluk village, 14 spirits are channelled—seven men and each of their wives. Walter Irvine wrote about the decline of these more traditional, mainly rural-based spirit mediums, and how an upsurge of professional urban-based spirit mediums was occurring.Footnote 38 To be clear, this article relates to rural non-professional spirit mediums, who do not rely on spirit mediumship for their livelihood and tend to engage in spirit mediumship on an ad hoc basis, with the primary goal of following traditions and supporting community practices.
Spirits, spirit mediums, and hydropower dams in the Mekong River Basin
There has been a limited amount written about the entanglement of hydropower dam development and spirits, and what has been written has been presented differently than the case presented here. For example, Andrew Johnson’s work studying the downstream environmental and sociocultural impacts of a large hydropower dam in China along the mainstream Mekong River in Nong Khai province, northeastern Thailand, involves careful consideration of the role of spirits and spirit mediums in the changes that are occurring, and resistance to the development process that is causing. However, he does not consider how spirits are linked to particular parts of the river, nor how spirit mediums sometimes channel multiple spirits. He also does not consider the identities of the particular spirits channelled.Footnote 39 Baird and Shoemaker, on the other hand, write about how villagers in southern Laos circulated rumours about how an angry water spirit caused the death of five Chinese gold dredgers who were excavating gold from the Xekong River. This seemed to constitute a sort of ‘weapons of the weak’ when resisting destructive Chinese development efforts, but they too do not consider the particular spirits involved.Footnote 40 In yet another case, Baird and Shoemaker consider how ethnic Brao people in Attapeu province, southern Laos, expressed concern about how a planned hydropower dam on the Xexou River might impact on a particular powerful water spirit known as Arak Brawng, which means the spirit of the Xexou River in their language. Again, however, not much detail about this spirit is provided.Footnote 41 Sarinda Singh, in her book about the development of Nam Theun 2 Dam in central Laos, also deals with ideas about spirits and natural potency, but does not consider particular named spirits.Footnote 42 Most recently, Mahanty and her colleagues write about how ethnic Bunong people threatened with the loss of their village due to inundation caused by the creation of the reservoir for the Lower Sesan 2 Dam were particularly inspired to resist the project due to concerns about spirits taking revenge on villagers if the dam disrupted their special places of inhabitation. This example is the closest to ours, as it deals with a different community and a different ethnic group, but the same dam project that we consider. However, they too do not engage with the particular spirits involved and their relationships.Footnote 43
All this literature indicates that people in the Mekong River Basin often link spirits with hydropower dams and other developments, albeit in a variety of ways, including in relation to development initiatives such as the BRI. We contend that there is scope for conducting more in-depth research about various ways that hydropower development and spirits interact, and the links between particular spirits and parts of rivers. We do not believe that spirit practices are particularly changing due to the engagement of people with hydropower dams, but instead we want to show how present-day practices related to spirit mediums and spirits within a rural ethnic Lao village in northeastern Cambodia have manifested in relation to the BRI-supported Lower Sesan 2 Dam.
The Lower Sesan 2 Dam and its impacts
The Lower Sesan 2 Dam (LS2) is the largest hydropower dam project developed in Cambodia to date. The dam blocks both the Sesan and Srepok rivers, as it is situated just below where the Srepok flows into the Sesan (see Figure 1). Less than 20 kilometres downstream from the Lower Sesan 2 Dam, the Sesan River flows into the Sekong River, which flows into the Mekong River just a few kilometres downstream.Footnote 44 Although LS2 was originally envisioned through a master plan supported by the Asian Development Bank in 1998–1999,Footnote 45 LS2 did not re-emerge until 15 June 2007 when the government of Cambodia’s Ministry of Industry, Mines, and Energy granted permission for Electricité du Viet Nam (EVN) to conduct a detailed feasibility study for the project.Footnote 46 Then, in 2008 and 2009, Vietnamese and Cambodian companies were hired to conduct feasibility studies, including social and environmental impact assessments.Footnote 47
Figure 1. The Lower Sesan 2 Dam and the surrounding area in northeastern Cambodia. Source: Map owner, first author.
Later, however, EVN did not have the capital to develop the dam as planned. Initially, EVN brought in a Cambodian partner, the Royal Group of Cambodia, with the help of Prime Minister Hun Sen. Later still, however, EVN decided to withdraw from the project. It looked like the dam might not be built, since most international funders dared not get involved with it as it had been identified as being particularly destructive to important Mekong fisheries.Footnote 48 However, Cambodian Prime Minister Hun Sen appealed at a high political level for the People’s Republic of China to provide support for the project. Soon after, Hydrolancang, a Chinese public company, agreed to hold 51 per cent of the project, with the Royal Group taking 39 per cent of the shares and EVN retaining 10 per cent to cover their initial costs in developing the project.Footnote 49 China’s decision to support the controversial LS2 was made at a high political level and was clearly intended to make use of state BRI funding for political advances for China in Cambodia. This is one of the important aspects of the BRI that makes it different from private sector business investment, which tends to be more pragmatic and less political.
LS2 was immediately controversial, as villagers living along the Sesan, Srepok and Sekong rivers had already been experiencing the downstream impacts of the Yali Falls Dam, which began as early as 1996. However, the impacts were not documented until 2000 and 2001 when the first community-level studies were done about its impacts.Footnote 50
Because the Chinese took over as the main funders and owners of LS2, following a typical Build-Operate-and-Transfer (BOT) concession agreement, through which the dam was finally built and completed in 2018, LS2 is one of the many infrastructure projects supported by China and considered to be a part of the BRI. Indeed, the BRI is having a great deal of influence, although there are serious concerns in many countries that the sovereign debt that the BRI is facilitating is reducing economic options for governments and causing China to gain increasing amounts of influence through other nations owing them substantial amounts of money.Footnote 51 In Cambodia, the government has embraced the BRI, which has supported the development of various kinds of large-scale infrastructure projects, including the Lower Sesan 2 DamFootnote 52 and the Heng Fu sugarcane plantation concession and processing factory project in Preah Vihear province.Footnote 53 Concerns have also been raised about the BRI being used by the powerful elite in Cambodia to become even wealthier, while further marginalizing the poor.Footnote 54
LS2 has caused a number of serious impacts related to hydrology and water quality. In addition, it has blocked important migrations of different species of fish between the Mekong River and the Sesan and Srepok rivers.Footnote 55 Ziv et al. reported that LS2 would have the most destructive impact on wild-capture fisheries, estimating that the dam would lead to a 9.3 per cent reduction in biomass throughout the whole Mekong River Basin.Footnote 56
The spirit mediums, the Lower Sesan 2 Dam, and resistance
In February 2012, during the cold season, more than 500 villagers from communities along the Sesan and Srepok rivers, mainly ethnic Lao people from communities in Stung Treng province, organized a peaceful protest against LS2 near its construction site, the infamous Thada Rapids. These rapids were a sacred place for local people, as they were situated at the confluence of the Srepok and Sesan rivers.Footnote 57 Protesters arrived by long-tail boats and tractors, and carried banners reading, ‘We must preserve the river, which is the livelihood of the people’.Footnote 58
On Niwan, the spirit medium from Phluk village, just below the dam, discussed in the opening vignette, was the centre of attention during the 2012 protest. She channelled her main spirit, Kahomkor, surrounded by all the villagers, and once the spirit had taken over her body, she took out red clothes and slung them around her forehead and waist to make it clear that she had undergone the channelling transformation. Indeed, her body can be seen as crucial infrastructure, or an important conduit, for the spirits (see Figures 2, 3, and 4).Footnote 59 Through this lens, the human body can be seen as infrastructure for spirits, as it is the conduit for the spirits. It allows for spirit flows or movements across space. This helps to move us away from imagining that only built physical things are infrastructure. Similarly, rivers can be seen as the infrastructure that allows fish to move.
Figure 2. Villagers protest against the Lower Sesan 2 Dam, at the spirit house (ta ho) at the confluence between the Sesan and Srepok rivers, February 2012. The area has now been inundated by the dam’s reservoir. Source: Courtesy of Meach Mean.
Figure 3. The spirit medium from Phluk village, On Niwan, speaks with a villager after becoming the medium for the spirit, Kahomkor, at the spirit house at the river confluence, February 2012. Source: Courtesy of Meach Mean.
Figure 4. On Niwan, the spirit medium for Kahomkor at Phluk village, dances after becoming possessed at a protest rally against the Lower Sesan 2 Dam, February 2012. Source: Courtesy of Meach Mean.
Illustrative of the above, spirit mediums frequently play highly embodied roles, something to which feminist political ecologists have long been attentive. While dams tend to block flows, or movement, spirit medium bodies are often important for facilitating flows—not water flows but the flows or movement of spirits across spaces from place to place. This represents what we call ‘shadow infrastructure’, which we define as infrastructure that plays a crucial role, one that is not always evident or broadly understood by most people. Other villagers prayed to the local rapids spirit, asking them to protect the environment of the river.Footnote 60 According to Lao tradition, spirit mediums can drink any amount of alcohol without becoming drunk. Therefore, Lao villagers typically offer alcohol to spirit mediums once they start channelling spirits. In fact, they are often offered full bottles of whiskey, which the spirit mediums consume directly from the bottle. On the one hand, drinking a lot from the bottle is intended to demonstrate that the spirit medium is channelling the spirit of a male leader, typically imagined to be a heavy-drinking and powerful man, not a woman, but on the other hand, this drinking sometimes serves to remove some of their inhibitions, allowing the spirit mediums to speak boldly and dance for the community. Thus, the praxis of spirit mediums is highly embodied. They typically consume considerable amounts of alcohol, speak openly and loudly about things that they think are wrong and should be rectified, and dance, sometimes with ancient swords, in order to demonstrate their circumstances. Such embodied practice is also linked to evoking emotions related to protecting nature that is so important for other species, for their own livelihoods, and for spiritual and cultural life.
Although the spirit mediums supported activism from early on, parts of the two Lao villages of Sre Kor (Na Kor) 1 and 2, and part of the ethnic Bunong village of Kbal Romeas, located upriver in the reservoir area, later became the centre of resistance to LS2, even though the dam threatened the livelihoods of people throughout the Sekong, Sesan, and Srepok river basins and beyond.Footnote 61 The two villages of Sre Kor served as a single community and had one spirit medium, who reportedly initially supported those villagers in Sre Kor who were actively resisting LS2. Indeed, early on virtually all the people in Sre Kor opposed the building of the dam.Footnote 62 Later, however, she ended up in the project resettlement village north of the Sesan River, while the part of her village that more fervently resisted the dam, and refused to receive resettlement support from the dam developers, settled on the south side of the LS2 reservoir. Social and spatial divisions emerged in the community due to the influence of the dam and resettlement, since those who resisted the dam could not access their community’s spirit medium, both because of the physical distance between the two new village sites, but also due to community divisions that have arisen. The divisions that were created between the villagers are also believed by locals to have affected the vigour of the spirit mediums living near LS2.
The infamous Thada Rapids on the Sesan River was once home to the powerful male spirit discussed earlier—Kahomkor in Khmer or Tha Kho Deng in Lao, meaning ‘red neck’ in both languages. A ta ho (spirit house) was built on the sacred site that marked the Thada Rapids and the confluence where the Srepok River flows into the Sesan River. A Chinese Cambodian trader originally had the idea to build a spirit house: after his boat capsized and he was struggling for his life, he promised he would build the spirit house if he survived, which he did.Footnote 63 It seems possible that he introduced the idea of Kahomkor being associated with this spirit house, as he was a trader who travelled to many places, including ethnic Khmer communities who believe in Kahomkor. Before the dam was built, all boat operators who passed the rapids would stop along the way and visit the sacred place or ta ho of the rapids to make a small offering, or at least to briefly pray that the spirit would allow them to safely negotiate passing the difficult rapids. The first author remembers doing exactly that with a boatman when they passed the rapids by boat when he was studying the downstream impacts in Cambodia of the Yali Falls Dam in Vietnam in 2001, and when he was studying the potential impacts of LS2 on the river and the people who depend on it in 2009.
Today, the ta ho of Kahomkor on the sacred rapids can no longer be visited, nor does the area pose the navigational danger it once did. The sacred place is now deep underwater due to the creation of LS2’s large reservoir. Following the completion of the dam, offerings once made to Kahomkor at the Thada Rapids site ceased to be given. While local people believe that Kahomkor is no longer located at the Thada Rapids, which are underwater, villagers in Phluk and Sre Kor expressed differing opinions as to where Kahomkor currently resides. One thing is certain, however: spirits like Kahomkor go where they are fed. The fact that the builders of the dam are Chinese is important, as stereotypes about Chinese people, who have lived in the area for over a hundred years, feed into Lao understandings of the circumstances. Indeed, some villagers even believe that Kahomkor has gone to live with the Chinese workers at the dam site, where the Chinese who work for LS2 regularly fed him roast pig,Footnote 64 a luxury that local Lao villagers are rarely able to offer the spirit and also falls outside of the type of food that they regularly eat. Moreover, it is imagined that Chinese people eat more pork than they do, both because they can afford it and like it, but also because it is thought to be part of their culinary tradition.
This is a powerful allegory.Footnote 65 In particular, it seems to demonstrate the idea that China has more resources than Cambodia, and thus is able to gradually gain more control and power through its wealth. As the spirit medium put it, ‘Kahomkor did not like the Chinese, but they gave him the head of a pig every week at the dam site, so it ended up there.’ The villagers did not have the power of the resources that the Chinese have.Footnote 66 The central role of the Chinese in this BRI project is thus an important factor.
Shades of resistance, however, can also be found. For example, Phluk’s spirit medium, On Niwan, told the second author that ‘Kahomkor would not go to the dam site, and neither would any other Lao spirits.’Footnote 67 A kind of spirit boycott was evoked. This also attests to the importance of geography, or the politics of place and mobility, when it comes to environmental struggles involving spirit mediums. The mobility of spirits also relates to the beginning of this article, when the Phluk villager, Mai, explained how the displacement of spirits over space could lead to dangerous retribution to the community of individuals, something that locals obviously want to avoid.
LS2 has shifted the material ways in which Lao villagers understand their changing relationship to the river and has negatively affected their ability to access resources such as fish and other aquatic animals. The disappearance of the rapids, previously an obstacle to mobility, but also an important site for the making of social and spiritual coherence, impacted on spirit mediumship and its practice in the area. Most ethnic Lao villages in northeastern Cambodia have their own spirit mediums, but it is also generally understood that some spirit mediums are more powerful than others, especially in their ability to call powerful spirits to dwell in and speak through them. Prior to the inundation of the rapids, On Niwan, the spirit medium from Phluk village, was widely recognized as a powerful spirit medium capable of calling the powerful spirit, Kahomkor, along with other spirits. According to Phluk’s spirit medium, Kahomkor resides with her in Phluk and visits other places. According to a villager at Phluk, ‘Kahomkor goes to other villages, but he does not stay. It’s like when we go visit other villages…it’s only for a short while.’Footnote 68
On Niwan is in her mid-sixties and has been a spirit medium for well over a decade. A soft-spoken woman, she spoke with certainty when asked about Kahomkor’s residence in Phluk. Contrary to this, Sre Kor’s spirit medium, aged 72, was convinced that, following the loss of the spirit house on the Thada Rapids, Kahomkor resided with her.Footnote 69
The changing ecology of the Sesan River can be described as disorienting and disruptive to the social and environmental rhythms of life within and near the river, including those of animals, of people, and also of spirits.Footnote 70 Here is another way that the feminist political ecology perspective, one designed and intentional about revealing the less visible, can contribute to how we think about infrastructure and its impacts, including on changing spatio-temporal rhythms, both ecological and social, that are being fundamentally transformed by the expansion of new built-infrastructure. Threats to life and livelihoods that accompany unpredictable hydrological flows, loss of familiar waterscapes, and declining or disappearing fish species and aquatic resources have led spirit mediums and villagers to rationalize the spirits’ inability or unwillingness to grant requests to stop the dam’s construction. This differs from 2012, when dam construction had not yet begun and the possibilities for stopping the dam from being constructed seemed more viable, including to the spirit mediums.
Given that both spirit mediums and villagers largely agree that spirits, including Kahomkor, have a great deal of agency, it is plausible, in their view, that Kahomkor allowed the dam to be constructed. Additionally—and crucially—Kahomkor is mobile, having been known to travel south to Phnom Penh, and is willing to go where he is fed.Footnote 71 In speaking with villagers in both Phluk and the new resettled villages of Sre Kor, the second author was not able to discern any resentment towards Kahomkor for potentially allowing the dam to be built. However, the temporalities of impacts of infrastructure projects, such as LS2, are still unfolding in uncertain ways.Footnote 72 While spirit mediums and spirits like Kahomkor in both Phluk and Sre Kor maintain similar roles and potency as they had before the dam’s construction and operation, the future of spirit mediums and their practices remain uncertain.
Crucially, as well as displacing spirits, the dam is materially affecting local livelihoods. The overall logic of the practices related to spirits has not significantly changed, but the impacts of displacement on the spirits, and the wrath that the people face due to this displacement along with material losses, helps explain what has happened, as the explanation is now related to both the material and cultural loss associated with the impacts on the spirits.
Changes in nature-society relations
The circumstances associated with LS2 and ethnic Lao spirit mediums living near the dam are revealing. First, spirit mediums and their sacred spaces have been historically linked to particular ecologies and geographies. The Thada Rapids was passable by boat, but not without considerable difficulties, particularly during the dry season. Therefore, the geographical and ecological challenges that people faced due to the rapids justified establishing a sacred space associated with spirit mediumship—a place that, if utilized appropriately, could help people pass through the area unscathed.
Since the rapids and the river were understood by locals to be connected to spirit mediums, it should come as no surprise that spirit mediums initially played significant roles in supporting community efforts to oppose the project. This was particularly evident during the 2012 protests, which bought local people opposed to the dam together with a spirit medium at the site of the ta ho on the sacred space adjacent to the rapids. The spirit of Kahomkor was clearly important for mediating human interactions with nature, as it has affected how people understand nature and interact with it.
However, as time passed, it became clear to everyone, including the spirit mediums, that there was little they could do to stop the dam’s construction. In response, they moderated their positions and came up with new explanations as to why they could not resist the project, and what happened to the spirit ta ho and the rapids it was linked to, which had been flooded by LS2’s reservoir.
Interestingly, as opposition to the dam became more difficult, attempts appear to have been made to safeguard the reputation of the spirit and the spirit medium by denying that they were necessarily defeated. Their agency was emphasized, including to potentially allow the dam to be built, as well as the mobilities of spirits. These changes may have made it easier for some to make sense of the new infrastructure in place and integrate it into the new infrastructural landscape of the area. Ultimately, nature-society relations are being challenged and changed, not only due to the construction of the LS2 dam and all the ecological and social changes that resulted, but also because the dam has affected spirit mediums and their sacred spaces. This has led to new understandings of the relationships between spirits and nature, along with the influence of Chinese dam developers linked to BRI.
Conclusion
In this article, we have emphasized the importance of spirit mediums in the spiritual lives of ethnic Lao people in Stung Treng province. Through spirit mediums, ethnic Lao people have sought the guidance, protection, and intervention of spirits in dealing with disruptive changes in their lives. The Thada Rapids was once a powerful place where villagers along the Sesan River once gathered to seek spiritual intervention from spirit mediums. According to locals, it was built by an ethnically Han Chinese trader, but someone who lived in Cambodia and was a citizen of Cambodia, something that differentiates the way he is thought of by villagers, and also the way that the Chinese dam-builders (citizens of China only temporarily in Cambodia) are understood. Indeed, Chinese Cambodians and Chinese nationals are viewed differently, at least to some extent.
What happens when sacred spaces disappear on a changing river? Villagers would say that the spirits merely move, travelling to other rapids or islands on other rivers and tributaries, to live and eat with people of other ethnicities, at the dam site or in Phnom Penh. It is unclear whether Phluk’s spirit medium has—in fact—lost the power to call Kahomkor exclusively, as Sre Kor’s spirit medium also claims the spirit stays with her. Also unclear is the spirit’s affinity towards Lao spirit mediums and villagers. As villagers suggest, Kahomkor is a Khmer spirit who speaks Khmer and was presumably connected to the Cambodia state, which is important for understanding the broader villager-state relations that exist today, and the power dynamic in Stung Treng province. What is clear is that the inundation of the sacred space at the Thada Rapids has opened up the possibility for other sacred spaces and new orientations away from the river. The extent to which spirit mediums and their practices will play a role in the future, shaping large infrastructure projects like LS2, remains to be seen. Will Da Be Le, another spirit who has been displaced by the dam, end up taking out its wrath on one or more of the villagers? Only time will tell.
Acknowledgements
The research upon which this article is based was partially supported by the Tracking Changes Project at the University of Alberta, Edmonton, Canada, which received funding from the Social Sciences and Humanities Research Council of Canada (SSHRC PG 895-2015-1024 Parlee). Thanks to all the Lao people in southern Laos and northeastern Cambodia who have discussed spirit mediums with us over the years. Thanks to the Cartography lab at the Department of Geography at the University of Wisconsin-Madison for helping to prepare Figure 1, and to Meach Mean for providing the photos included as Figures 2–4. Thanks also for the thoughtful comments provided about the article during the Sacred Dimensions of the BRI’s ‘Infrastructural Commons’ workshop organized online by the University of Hong Kong on 11–12 June 2022, with the support of the Research Grants Council of the Hong Kong SAR, grant no. RFS2021-7H04 including by Emily Hertzman, Andrew Johnson, David Palmer, Galen Murton, Jos Estevez, Laur Kilk, and Orlando Woods. Thanks, also, for detailed comments on an earlier version of this article provided by David Palmer and Orlando Woods, and the reviews of two anonymous reviewers and the editors of Modern Asian Studies.
Competing interests
The authors declare none.
Open access
