Restricting spatial lives? The gendered implications of conservation in Cambodia’s protected wetlands

Understanding socio-environmental relationships, and utilising this knowledge to improve both human and environmental outcomes, is emerging as an issue of increasing urgency. Internationally conceived conservation programmes continue to support the proven effective protected areas (PAs) model of conservation (Rodrigues et al., 2004), but increasingly promote the incorporation of local agendas and needs as critical considerations within conservation frameworks (Borrini-Feyerabend et al., 2013). Extending and deepening research in this area, we seek to explore the specific interplay of gendered identities and PAs in developing countries, with an examination of a Ramsar-listed protected wetland¹ in rural Cambodia in Southeast Asia.

Our focus on identity–environment interactions furthers recent developments in feminist research and the work conducted under the feminist political ecology (FPE) framework. Building on this, understandings of multiple identities, extending to (for example) race, class and ethnicity, and the intersectional way in which these identities interact to impact lived experiences now underpin much FPE research (Nightingale, 2011; Valentine, 2007). Although gender nominally remains the primary focus, analysis and inclusion of alternative identities framed as ‘categories’ or ‘axes of difference’ is increasingly the norm (Cho et al., 2013). Significantly, feminist researchers utilising the FPE framework have positioned themselves at the forefront of gender–environment investigation, examining the ways in which gendered knowledge/s, gendered roles and responsibilities and gendered environmental politics and grassroots activism are critical to the success (or otherwise) of environmental conservation and governance initiatives (Elmhirst, 2011; Rocheleau et al., 1996). For recent overviews of FPE’s evolution, see Sundberg (2017) and Resurrección (2017) both of which articulate the important contribution to theory and praxis of a vast FPE scholarship vis-à-vis environmental change.

Paralleling this development, an emergent feminist legal geography (FLG) framework argues for further interrogation of ‘invisible’ legal forces operating on specific locales and identities (Gillespie and Perry, 2018 or Brickell and Cuomo, 2017). The ‘feminist’ aspect within an emergent FLG takes LG beyond a generic ‘people’/law/place remit to interrogate the specifics of gendered identities (in a similar vein, albeit in vastly different contexts, to Farries and Sturm, 2018 or the work of Brickell and Cuomo, 2017; Gillespie, 2017). We recognise that a feminist LG is in its infancy; however, emergent work under this rubric attempts to undertake gender analysis of LG. Such work adds depth and nuance to standard and evolving LG research (Delaney, 2017). In this paper, we want to use this approach to analyse a particular conservation regime. We seek to add to discussions concerning the need for context-specific environmental regulation which looks to incorporate the needs of all local people. Gender-differentiated spatial experiences emerge as influencing lived experiences of the conservation regulations.

Our research was conducted in Cambodia on the Tonle Sap, or ‘Great Lake’, which supports millions of Khmer people and is commonly known as the ‘beating heart of Cambodia’. The Tonle Sap is characterised by a biodiverse ecosystem and complex regulatory system. Due to the unique hydrological cycle, wherein the lake floods during the wet season inundating 1.25 million hectares of land (Bonheur and Lane, 2002), the lake and surrounding Mekong floodplain constitute an immensely fertile environment hosting a rich biodiversity (Campbell et al., 2006). This area has supported the Khmer people for thousands of years through enabling large-scale rice cultivation and fishing and is the primary source of protein for around 70% of the population (Van Zalinge et al., 2000). For the approximately 1.2 million people who live directly on the floodplain of the lake, the majority of livelihoods are enabled by and dependent upon the flood pulse (Nuorteva et al., 2010). However, the lake is subject to multiple and cumulative threats including those posed by overfishing (Hortle et al., 2004), upstream damming of the Mekong River (Kummu and Sarkkula, 2008), climate change (Arias et al., 2014) and population growth (Pech and Sunada, 2008). Thus, in an age of uncertainty regarding impacts on the lake (Kummu et al., 2008), the maintenance of the ecological value of this region is paramount to the sustainability of Cambodia’s rapidly growing population. Further, as a dynamic part of the Mekong river basin, the Tonle Sap and surrounding floodplain is of broader significance within both Southeast Asian and worldwide environmental networks (Campbell et al., 2006).

Within the Indo-Burma biodiversity hotspot (Myers et al., 2000), the Tonle Sap is the subject of two voluntary international programmes, the Ramsar Convention and the UNESCO Man and Biosphere (MAB) Programme (UNESCO, 2007), both of which have been translated into national laws. The rural floating village study site of Prek Toal Village is subject to the full extent of these international environmental obligations. Situated only metres from a Ramsar wetland site (designated in December 2015) and within the buffer zone of the Tonle Sap Biosphere Reserve (TSBR), villagers are subjected to increasingly stringent regulations concerning the legality of certain types of fishing equipment, the restriction of fishing in certain areas at certain times and the types of natural resources they are allowed to collect (David, 2017; David and Dara, 2017). This paper focuses on the issues arising from the creation of PA boundaries directly adjacent to Prek Toal floating village. Prek Toal village lies on a small tributary river amongst the flooded forest at the Northern region of the lake. During the dry season when the waters have receded, each (floating) house in the village moves approximately 50–100 metres closer to the deep channel of the river, which remains inundated all year round. Overwhelmingly, the rural floating villages on the Tonle Sap are reliant on fish and rice as means of income (and nutrition) (Nuorteva et al., 2010).

Prior research on the conflict between human needs and conservation on the Tonle Sap further informed this research agenda (Gillespie, 2017, 2018; Gillespie and Perry, 2018). From previous study, located in an alternative Ramsar designated wetland (Boeung Tonle Chhmar) on the Tonle Sap, it emerged that local people had gender-differentiated concerns about conservation regulations (Gillespie and Perry, 2018). In summary, men were more concerned about the il/legality of fishing equipment, while women were more concerned about the specific sites/locations which were protected, and which restricted access to particular fishing spots. Based on this, it was hypothesised that the interplay between official laws and cultural norms may have implications for local understandings of dynamic gendered roles in Prek Toal. These findings are corroborated by other studies on inland fisheries in Asia, which are known to be sites of gender-differentiated roles and responsibilities (Eng et al., 2010; Kusakabe, 2003; Resurrección, 2006). Accordingly, in this paper we first look briefly at current understandings of the relationship between gender and environment. We then reaffirm the case for the conceptual marriage of legal geography and FPE, and we detail the methods used during fieldwork for this research. Following this, our findings around the (restricted) spatialised lives of local women are presented, supplemented by a discussion of ‘strong women’ who challenge these local norms. We argue that an explicit link can be made between the PA (spatial) restrictions and gendered identities. Further, we also argue that the nuance of a gendered people–place dynamic has been overlooked, potentially to the detriment of overarching conservation aims, in planning the conservation regime.

We are concerned with unpacking the human experience associated with PA conservation as it occurs in one floating village in rural Cambodia. We explore the factors shaping gendered spatial movement within Prek Toal village and question how these interact with the PA conservation initiative. A rich history of gender (and more recently, identity)–environment study, which has evolved significantly in recent decades, informs this praxis. FPE understandings of three primary facets of gender–environment relationships, first conceptualised by Rocheleau et al. (1996) in their seminal FPE text, understand socio-environmental interactions as gendered knowledges, gendered roles and responsibilities and gendered environmental politics and grass-roots activism. In line with the post-structural turn, understandings of identities and subjectivities as mutually constructed by environmental practices have led to complex analyses of the ways in which socio-environmental relationships form critical components of effective conservation (Lau and Scales, 2016). Essentially focusing on the constant, dynamic (re)formation of roles, through power, and discourse, conservation regulations introduce new constraints and opportunities within these roles (Sundberg, 2004). Consideration of these issues has seen FPE become a critical juncture between related disciplines including feminist geography and political ecology. We seek to build on the work of scholars such as Radel (2012), Badola and Hussain (2003) and Sundberg (2004) to demonstrate gender-differentiated experiences of conservation practices. Beyond this we are also interested in promoting the idea that spatial boundaries identified in law become a component of gendered identity. We argue that LG can contribute to FPE as the regulation of water/landscapes influences the way people act within place. In essence, we want to contribute an enviro-social-legal perspective to the FPE narrative on nuanced, complex people–place interactions.

Although gender is nominally the primary focus of our study, we acknowledge that incorporating theories of intersectionality in qualitative human geography is fast becoming the norm (Harris et al., 2017; Lau and Scales, 2016; Nightingale, 2011; Valentine, 2007). Across feminist theory, intersectionality is understood to be the way an individual’s lived experience is shaped by and shapes their multiple, intersecting identities often such as race, gender, ethnicity and class (Crenshaw, 1991). Far from assuming a ‘base’ identity and adding compounding factors on top, intersectionality explores complex relationships between dynamic identities, subjectivities and lived experience (Anthias and Yuval-Davis, 2005). However, researching intersectionality presents a challenging task. Some intersectional studies accept and utilise certain categories of difference, defined as using an ‘intercategorical complexity’ lens, which allows a simpler approach (Cho et al., 2013; McCall, 2005). Critics argue that categories are a simplistic rendering, employing essentialist categories antithetical to feminist praxis (Pratt and Hanson, 1994). Similarly, Valentine (2007: 14) argues that entering the field with clearly defined categories of analysis may in fact ‘…unwittingly shape and limit the narratives we are told’. Despite this, we argue that for practical purposes a (perhaps arbitrary) line must be drawn between theorising and the need for praxis, and categories remain a useful tool when carefully employed. Nonetheless, in the context of our study location, where base-line data about demographics and any quantitative statistics from census data are absent or severely limited, we have adopted a binarised gendered approach in this study, despite the limitations (for a full discussion, see Harris et al., 2017).

Current critical legal geography is characterised by diverse areas of investigation which counteract the ‘taken-for-granted’ attitude that law and geography previously held in relation to one another. In keeping with post-structural thinking, legal geographers pay attention to the way in which laws mutually constitute spaces, investigating all areas where matters of law, space/environment and people interact (Benson, 2012; Delaney, 2015). Conceptually, legal geographers work to make the ‘invisible visible’ (Braverman, 2011) through tethering abstract legal concepts to lived realities in space and place. Recognising the spatiality of the law as inextricably intertwined with human experience unlocks extensive areas of potential investigation (Delaney, 2003). Through seeking to illuminate the unseen, and in investigating purportedly ‘objective’ and ‘value-less’ hegemonic perspectives, this scholarship shows unanticipated impacts of laws (Acker et al., 1993; Gillespie and Perry, 2018; Robinson, 2013), in an anti-essentialist and challenging scholarship.

Within legal geography, particular attention is directed towards various types of boundaries (Blandy and Sibley, 2010). The legal delineation of boundaries creates new socio-spatial meanings and materialities for all stakeholders, and, critically for our later discussion of PA boundaries, creates areas of identity-based inclusion and exclusion. A legal geography perspective thus can give us a gendered spatiality to legal landscapes. Additionally, boundaries in reality are rarely as distinct as the law (through maps) portray them (Gillespie, 2013). Rather, when translated into daily life, boundaries may adapt, become permeable or take on new meanings relative to the place and context where they are imposed (Taylor, 2010). Thus, initially distinct legal meanings may become blurred, and it is the task of legal geographers to ‘make the invisible visible’ through interrogating the hidden and co-constitutive interactions which law may have with social realities.

Delaney (2017) has recently developed calls for clear future research directions of critical legal geographies, including flagging the pressing need for expansion of research into non-Western settings and consideration of non-human and law interaction. Overwhelmingly, previous research in the field has investigated Western and developed world phenomenon and has thus been concerned with Western models of law-making and implementation. Thus far, this call has been addressed in case studies based in non-Western locations, including two related papers on the Boeung Tonle Chhmar Ramsar wetland lawscape vis-à-vis the local inhabitants (Gillespie, 2018; Gillespie and Perry, 2018). In particular, Gillespie (2018) argues for the need for a better understanding of conservation situations where ‘legal layering’ has resulted in complex networks of laws, lore and regulations. She builds on Rusca et al.’s (2015) conception of ‘bricolage’ for her work on another core PA on the Tonle Sap, Boeung Tonle Chhmar, located on the southern-eastern boundary approximately 80 kilometres from case study used in this paper (Gillespie, 2018). Gillespie’s findings indicated a tendency for regulations to reinforce pre-existing power disparities – a form of ‘elite capture’ (withheld for peer review). In a similar vein, we wish to answer the call for legal geography scholarship to expand into new spatial (Global South) and methodological (gendered) contexts (Delaney, 2017).

International recognition of the value of the Tonle Sap’s biodiversity has resulted in several PA classifications. Four Ramsar wetlands under the Ramsar Convention have been identified around the lake, in addition to the designation of the Tonle Sap and Surrounding Floodplain as a Multiple-Use Biosphere Reserve under the UNESCO (2007) MAB Programme. The Royal Government of Cambodia is responsible for the domestic translation of these international commitments. To this end, several national laws have been implemented including the Law on Nature Protected Areas 2008² and the Fisheries law 2006³ (Mensher, 2006). The regulations are enforced by various actors in a ‘fragmented institutional patchwork’ (Gillespie, 2018), and management and conservation responsibility has been shunted between government departments (Sithirith, 2017). Jurisdiction over restricted activities and areas on the lake overlap and are complicated by provincial boundaries and differing levels of law enforcement between regions (Sithirith, 2017).

Prek Toal village is itself subject to a complex interpretation of these laws, due largely to the existence of a large migratory bird colony of significant natural value directly adjacent to the village. As mentioned, the colony and surrounding area (depicted in Figure 1) was nominated as a core area of the TSBR (UNESCO, 2007), and in 2015 the same approximate area was designated a Ramsar Wetland of International Importance.⁴ The pre-existing village of Prek Toal, although not within the bounded PA, is located only metres away and is situated within the buffer zone of the TSBR (Figure 1). Under the conservation-inspired national laws, local people are thus subject to significant legally enforceable restrictions on their use of particular types of fishing equipment and where/when they can fish. In this paper we are concerned to focus on the spatial extent of these laws – the bounded space of the PA and the implications the boundaries have for lived experiences for breaching these regulations about access to the resources (fishing grounds) of the PA can result in the imposition of fines. Fines can have significant consequences for poor households. During fieldwork it became evident that there was much confusion within the village about the specifics of these restrictions.⁵

At this point, it is necessary to briefly characterise natural resource management practices in Cambodia. We do not wish to detract from our primary focus of gendered identity in relation to conservation boundaries and practices; however, given that neo-patrimonial systems of regulatory enforcement are deeply embedded in Cambodia political culture (Sithirith, 2014), it is impossible to write about conservation without (at least briefly) examining the impact of this in situ. Milne (2015) characterises the Cambodian state in the vein of Paul Gellert’s (2010), ‘extractive regime’, whereby the continuation of state power is inherently reliant on illegal and ‘quasi-illegal’ practices of natural resource exploitation through the support of exterior powers. While people in Prek Toal were confused about the exact nature of the regulations, they were all but unanimous in their view that they were only enforced against people who could not afford to pay to be exempt. For example, one woman expressed her discontent stating that:

I am not happy [with the conservation] because the friends and family of the leaders are allowed into the conservation area, so conservation doesn’t work. It’s not fair. If it was closed for everybody, the conservation would be good. But this won’t happen because of the money, and the people who live off the money from letting people in to the area. [Proverb] The chicken who lives in the rice field will eat the rice. No-one is honest. (Interview 8, 1st trip)

Thus, when we talk about the impact of socio-economic status in later sections, we are referring to this unequal system of access which applies to both men and women. In a similar vein Harris (2006) made observations about power networks becoming reinforced through water management institutions which act to ‘effectively widen the gulf between those with access to power and resources and those without’ (196). In our case study those people (male or female) with economic power were not perceived to be disadvantaged by the imposition of new rules limiting access to fishing grounds.

A total of 52 interviews were collected via ‘opportunistic or emergent’ sampling (Patton, 1990). More purposeful sampling was not possible given the lack of available means of prior communication with the village. This was addressed through the collection of a larger sample across two visits (the first constituting five days of intensive interviewing, followed by a three-day repeat/confirmation visit approximately one week later) which together were considered sufficient to achieve ‘data saturation’ (Suri, 2011). Fieldwork notes were produced alongside the interview answers to aid in analysis. Based on our intensive reading and re-reading answers and in conjunction with our Research Assistant we generated themes based upon changing lake conditions sorted according to ‘Flooding’, ‘Forests’ and ‘Fishes’. Each theme was based on how people used the waterscape, and this enabled us to build a narrative about lake use for different demographic groups (particularly in relation to the gendered male/female binary). During fieldwork, we lived in the village with our field assistant and were facilitated by a local NGO. In meeting with the Commune Chief to establish the terms of our research in the village, we held a quick, informal discussion which resulted mainly in the obtaining of current demographics of Prek Toal.

Recent work has noted the significance of ‘making visible’ the impact of research assistants in cross-cultural research (Caretta, 2015). Edwards (1998) writes about in-country research assistants as being a form of key informant, wherein although they are not considered experts on the subject being investigated, through explicitly understanding the research assistants’ own understanding of and thoughts on the research topic, their influence on the research process is made visible. In this vein, we would like to make an explicit acknowledgement of the contribution and positionality of our assistant, an educated Khmer woman with prior experience conducting qualitative research in this and other areas of Cambodia. She provided the in-field translations, and discussions with her were pivotal to gaining a more comprehensive appreciation of the cultural phenomena being researched. During interviewing we were ferried from house to house by a local ranger, and he would commonly sit in on the interviews. Both our research assistant and the ranger himself took care to inform participants that he was not there in an official capacity, and conversations with the field assistant following interviews revealed that she did not believe the presences of the rangers (in both trips) influenced people’s ability to speak freely. However, we cannot ignore the reality that some responses were influenced by their presence. Both women and men were interviewed, as it was difficult in the circumstances to request an audience with one or the other.

Though roles of Khmer women in society are constantly evolving (Derks, 2008; Lilja, 2016), it is understood that broadly, men and women today retain different household roles. Women are expected to defer to their husbands, move around the house quietly and be fiscally responsible, while men are expected to provide the financial underlay (Brickell, 2011). Additionally, domestic work is ingrained into custom as the domain of women (Brickell, 2011). Given these factors, it was expected that differentiated natural resource use would primarily result from this ‘female domesticity’ and ‘male livelihood’ (fishing) binary. In this broad context we were interested in understanding whether the imposition of the PA, particularly the nearness of the boundary, would have differential gendered impacts. Our approach links to recent work examining and exposing complexity in gender and fishing practices in community-based management of marine PAs in the Philippines (Kleiber et al., 2018). We found intertwining gendered factors created differentiated spatial experiences: two of which related explicitly to the imposition of the PA.

First, many women indicated that they were not comfortable going fishing alone far from the village due to ‘security’ or personal safety concerns, underpinned by a fear of sexual assault. For example, one woman stated ‘I go with my children, less scared. [What are you afraid of?] I’m afraid of people stealing, sexual violence, mostly from the Vietnamese.⁶ Women will always go [fishing] with their families’ (Interview 6, 2nd trip).

Another women spoke of stopping going to the forest after her husband died, saying ‘Since my husband died, I no longer use the forest. [What are you worried about?] I am worried about security, I am afraid of humans, men, ghosts [laughs] and sexual violence’ (Interview 1, 1st trip).

The close proximity of the PA boundaries to the village means the distance women travel when fishing alone are restricted. The PA boundaries nominally prohibit everyone from fishing within the conservation zone; however, the gendered impact of the boundary is important, as the nearness of the village to the boundary pushes people to seek fishing spots further from their village. Women are less able or willing to extend their fishing range. Similarly, Nightingale (2011) also documents the gendered implications of a nexus between proximity to particular resources (species/products) and restrictions/access. Arguably, women re-create and reinforce their own vulnerability through their ‘coping strategies’ (Valentine, 1989), which include going fishing in groups and with their children and avoiding the need to go fishing out of the village all together. The narrative of vulnerability, recounted above, is thus perpetuated by their actions. This creates clear gendered differences in the spaces occupied and used in the village and creates a subtle form of gendered exclusion (Brickell, 2011; Koskela, 1997). The proximity of the PA boundary asks us to question whether conservation initiatives through in situ boundary-making that restrict livelihood access actually reinforce these gendered dynamics. The spatiality of lived conservation experiences demonstrated here highlights the broader legal geography (regulation through bounding) and FPE (gendered resource access) linkages.

Men further contribute to this narrative and discursively perform this gendered fear (Mehta and Bondi, 1999), as can be interpreted from the comments of men in the interviews. For example, a man who mentioned that he used to invite women to stay in his boat when he was younger simultaneously suggests female vulnerability and male protector status. Other men who talked about women being right to be afraid contributed to this – when they agree that women need to be careful, they agree that women are inherently in need of protection. One man stated

Women can’t go to the forest for security reasons, but men can. [Does this mean it’s more difficult for women?] Yes it’s more difficult for women because they can’t go, so they work more with water hyacinth, for example cutting it, to giving it to the ducks, and things that are closer to the village. Women also go and collect the plants to make the houses float – anything that’s closer to the village. (Interview 12, 1st trip)

The limited spatial range of women brought about with the PA boundary, as described by both men and women, appears to reinforce a male/female binary in this context. Arguably these attitudes would exist regardless of the PA boundary, but it is possible to argue that the PA boundaries may reaffirm such gender environment divides.

Second, many participants indicated that women were physically weaker than men and unable to perform the arduous physical task of fishing, limiting them to the collection of resources closer to the village such as through fish farming, using smaller fishing gear and collecting water hyacinth. One woman stated ‘Normally, my husband fishes further and I stay closer to the village because it is too far for me to row’ (Interview 3, 1st trip). In a different interview, one man reported ‘Because fishing is difficult, women are doing other things like collecting the floating plants for the house, cutting fish heads’ (Interview 13, 1st trip).

Notably, although it was a constant that some types of work were too hard for women, there were differences between the work that different households considered to fall into this category. Mehta and Bondi (1999) suggest that gendered physical differences are simultaneously discursively produced and situated and cannot be removed from their cultural context. Thus, the view that women are unable to perform the harder fishing tasks may be seen as a culturally constructed and performed aspect of identity, only tenuously based on physiological difference. Nonetheless, again the near proximity of the PA boundary to the village means everyone must travel further from the village to fish. Physical in/capacity to practice fishing is not a consideration in the siting of PA boundaries. However, the need to understand all stakeholder perspectives and localised fishing practices before PA designation must consider gender roles (see also Kleiber et al., 2018 in relation to the siting and regulation of marine PAs). The designation of PA sites and boundaries typically relies upon scientific assessment of both the local ecological value of the area and the sites’ representative contribution to broader systematic conservation networks (Margules and Pressey, 2000). To achieve a Ramsar wetland listing, sites must contain ‘representative, rare or unique wetland types’ and/or fulfil biodiversity conservation criteria based on local species numbers, types and migratory characteristics.⁷ Although we are aware that there are nominal provisions within local conservation laws allowing subsistence ‘family’ fishing within the PA,⁸ our research demonstrated that this has not translated into adequate concessions for local people in terms of access to regular fishing locations. We argue that considerations that take better account of the nuanced dynamics of the human–environment nexus – in particular, locations need to be considered in PA boundary-making.

Two clear factors contributed to the limited spatial range of women in and around the village and adjacent PA. We found that many women are concerned about issues of security and that women are considered to be physically ‘weaker’ than men and are thus unable to complete arduous natural resource collection tasks. We draw on Massey (1994) in this context to highlight that gender both reflects and constructs spatial practices. It is the entangled cultural expectations, mutually constituted through agency, discourse and tradition, which help inform the embodied experiences of some Prek Toal women. This, in turn, has significant implications for the imposition of spatial conservation regimes.

Returning to our theoretical concerns, we find that drawing on both FLG and FPE has enabled a deeper investigation into the gender-differentiated impact of conservation laws on Prek Toal. Gender-differentiated spatial ranges are linked to local norms around female vulnerability and strength, which are understood to be dynamic and performed (Sundberg, 2004). Similarly, in Kleiber et al.’s (2015) review work identifying the importance of capturing gender roles in small-scale fishery research we also see that gendered fishing practices can relate directly to the limited ability of women to travel longer distances for the catch ‘… which may narrow women’s range for fishing …’ (553). Gendered roles and responsibilities which prescribe fishing far from the village as predominantly the domain of men limit the ability of women to work around spatial conservation restrictions. Thus, in legal geographic terms, the previously ‘invisible’ implications of a bounded, physical conservation area in can now be seen as contributing to the introduction of new constraints and opportunities within these local roles. In light of these observations we argue that it is essential that understandings of complex interactions of this nature be incorporated when considering the suitability of different forms of conservation regulatory regimes. We suggest that planning a conservation area that restricts pre-existing access to places potentially exacerbates gendered imbalances. PA planning needs to take better account of spatialised gendered dynamics and must recognise that this is an essential ingredient for balancing human–environment PA relationships.

We do not wish, however, to present an overly simplified narrative. Returning here to feminist concerns about empowerment, representation and authenticity, it is necessary to adopt a reflexive approach to the impact of presenting the Prek Toal women as vulnerable and afraid. Jacobsen notes that gendered research in Cambodia conceiving power solely in narrow Western terms risks erasing alternative and culturally significant forms of power (Jacobsen, 2008). ‘Fear’ must not be studied without examining those who challenge, and without appreciating that there is much to be gained from prioritising the study of female agency (Koskela, 1997). This is because fear is only one dimension of more complex motivations underlying an individual’s decisions about which socio-spatial situations they are willing to place themselves in, and it is therefore necessary to delve deeper. Thus, within this section we wish to create a fuller picture of the complexities underpinning gender-differentiated spatial lives.

In this context we want to draw attention to the counter-narrative of one respondent. In Prek Toal, one woman emerged as refusing to accept the standard set for her by dominant social practices and discourse through referring to herself as a strong woman, who was not afraid to drive boats and go fishing alone far from the village:

“80% of the family responsibility is on me, because my husband had an operation and he can’t do heavy jobs. I do everything – finance, business, fishing, kids, housework…” [(later) Are you worried about security when you go fishing?] “I’m not worried/afraid about security concerns, I also respect the other fishing people. Sometimes I go alone – I am a strong woman.” (Interview 8, 2nd trip)

Another respondent, a male, also described women who do this. Women who self-identify as a ‘strong woman’ actively participate daily in challenging and re-shaping the ways in which it is possible to be a woman in the village (Koskela, 1997). In contrast to Nightingale’s findings around the ways in which women may perform their femininities to their own advantage in certain situations (Nightingale, 2006), these women challenge dominant ideals of femininity in Cambodia. This, we argue, is partially the reason that several people, men and women, so strongly emphasised that both men and women have the right to do everything. Though ‘common wisdom’ from the villagers suggests that it is generally regarded as sensible for women to be afraid, it is recognised that not all women are like this. This and other examples (e.g. Harris, 2006) throw into relief the constructed nature of gender differences even when there is a discourse of physical differences.

Narratives about empowerment become further complicated when considering that it is women of a lower socio-economic status who are most unable to refuse to go fishing alone. This idea is supported by the claims of villagers who stated that money is more important than gender, particularly one man who stated that ‘Happiness is money – if you have money, you have happiness’. During our time in the village, we observed that the woman who referred to herself as a ‘strong woman’ and said that she had no fear of going fishing by herself was both rich and powerful in the village. Thus, building on our earlier contextualisation of the resource management paradigm within Cambodia, this contributes to observations on socio-economic status as forming a further entanglement. We state this acknowledging that this research presents only limited instances.

The imposition of the PA boundary adjacent to the community has altered the lived experiences of women. From having (relatively) open access to fishing grounds close to the village women are now forced to travel further or fish in company with others or to assert as a ‘strong’ woman in a counter-narrative to prevailing attitudes about female roles in this floating village. This lived experience is different from what it was before or what it may be elsewhere in the flooded forest. This paper identifies that the way an inhabited flooded forest is bounded and spatially regulated through formal regulations or laws, imposed originally from global environmental conservation commitments, impacts different women in different ways. The importance of the physical environment is essential to our observations. In this way Sultana’s (2009) work identifying the importance of ecological, social and spatial relations links directly with our observations. But, in our context, it is the particular legal spatiality of bounding and restricting daily activities that becomes an important part of a complex intersectional identity (in the same sense that wealth, race, age etcetera also impacts identity): the boundary of the PA exerts an influence on the lived experience of women and thus becomes part of their identity. The legal spatiality enables and disables identity. Understanding the role of legal spatiality in gendered relations reflects the value in linking FPE (or feminist geography more broadly) with legal geographical practice.

Lastly, some brief reflexivity around the responses to questions about gender is required. Both men and women were apt to laugh and slide sidelong glances at each other when the conversation turned to gender. As in most interviews issues of conservation were approached first, it became clear that many did not think that the different roles of men and women were important at all – their focus was on the household as a unit. Asking explicitly gendered questions first did not change this reaction. This is important for two primary reasons. First, it indicated that although the villagers were cognisant of gender equality and gender issues, they were not regarded as important within Prek Toal – these types of issues are subsumed by larger and more immediate concerns of family survival, obscuring or even masking the complexities and nuance of household power dynamics (Agarwal, 1997). Second, this has troubling implications for research conducted in the feminist vein, which, as explored, is expressly political and charged with performing a service for and based on the needs of the people with whom it is conducted (Kobayashi and Peake, 1994).

We have attempted to make visible the gendered spatial dynamics of Prek Toal, in order to demonstrate the significance of such spatial controls in creating effective and context-appropriate conservation regulations. While acknowledging that this picture is by no mean comprehensive or all encompassing, our research demonstrates that women in Prek Toal are subject to spatial controls, propagated by cultural narratives around female weakness and vulnerability, which in the rural, isolated setting limits their movements.

Boundaries created by the PA hold distinct legal definitions and influence lived experiences. Women who refuse to fish alone contribute to and perpetuate cultural norms and the PA placement reinforces this situation. A counter-narrative of the ‘strong women’ challenges this narrative, but a lone voice is far from becoming a dominant influence in this setting. Moreover, wealthier women are more readily able to choose their course of action – if they adopt the status of a ‘bold’ woman (Koskela, 1997), then they challenge traditional gender norms and are empowered by their decision. Further, these women have easy access to fishing in the PA, enhancing pre-existing inequality. However, for lower socio-economic status women who fish alone, the empowering aspects of the ‘choice’ to do so become muddied. Based on the insights gained through the application of these frameworks, we posit that it is possible to distil some straightforward observations. Through the boundaries of the PA being in such close proximity to the village, some women are cut off from a significant resource which would, culturally, be ideal for their needs. This exclusion occurs for both men and women; however, men are more readily able to travel to other locations and stay overnight if necessary to guard their nets. Women may travel further if they are accompanied by their husbands, children or friends; however for those without those options (e.g. female-headed households) the choice is complicated by the cultural narrative of vulnerability of women in remote fishing areas.

We do not especially think of shifting conservation boundaries and associated legal terrains in PA management as playing a role in re-shaping gendered identities; in this case study we reveal that it does. While there are countless examples where law restricts/enables gendered lived experiences – in our example we demonstrate that the siting/location of a PA boundary adjacent to a floating village in a remote corner of Southeast Asia has unforeseen impacts on the mobility of women in their waterscapes.

Re-situating this research within overarching themes, we wish to re-consider the suitability of internationally formulated conservation regimes. Environmental protection regimes are inherently bound to the physicality of place. Whether designed to preserve physical, ‘natural’ ecological systems in their entirety or only particular species, such laws will always have a tangible and immediate effect on daily lives. This effect is enhanced in areas wherein people are directly reliant on natural resources for their livelihoods, such as the floating villages of Cambodia. Further, and inevitably, places are subject to previously established (though dynamic) local norms or ‘lore’ which regulate human activity. When a disconnect between the generality of global PA regimes and the specificity of place-based norms occurs, there is a significant risk that both conservation and human outcomes will be compromised through unexpected interactions. Thus, in creating conservation regulations for specific places, with unique environmental/biodiversity values and well-established human–environment relationships, it is essential to incorporate an understanding of these features and the interactions between them. Invoking feminist intersectionality concerns, this paper demonstrates these interactions between identity-based norms and imposed laws, which are unacknowledged in current regulatory frameworks in Prek Toal. It argues that given the fundamental role law plays in shaping everyday lives, more appropriate conceptualisations of legal conservation regimes may be formulated through consideration of the emergent gender-based norms.

We would also like to acknowledge the valued contribution of anonymous reviewers and the editorial staff.