Positionality, ‘the field,’ and implications for knowledge production and research ethics in land change science

ABSTRACT

Fieldwork is often framed as an individualistic rite of passage, wherein the researcher is a ‘neutral’ observer, thereby obscuring gender, racial and citizenship hierarchies. Feminist scholars have pushed against this assumption, leading to insights on researchers’ roles in fieldwork relations and knowledge production. We combine feminist perspectives and land change science (LCS) approaches to examine interactions between researcher positoinality and knowledge production. While ‘the field’ is a critical component in LCS research, there is a dearth of critical work on positionality. We examine how our positionalities – as a black, Ghanaian graduate student and a white, tenured professor – produce differential fieldwork relations. We link our experiences and relations to methods, results and ethics, revealing important implications for LCS researchers. We conclude by suggesting validation of self-reflection on positionality, as well as disruption of white hetero-patriarchism, create inclusive spaces, in terms of social differences, epistemology and methodology.

KEYWORDS:

• Positionality
• land change science
• fieldwork
• knowledge production
• research ethics

In October 2008, I (Heidi) sat with research assistant, Jenaro Martínez, on a curb in rural Veracruz, Mexico, chatting with Lucia and Marcela,¹ two sisters making birthday tamales. Jenaro and I were well known by that afternoon, having spent months in the community conducting fieldwork.

Neoliberal reforms resulted in historically low prices paid to farmers for unprocessed coffee. I was wrapping-up 14 months of dissertation research that queried how rural people, like Lucia and Marcela, were coping with ‘the coffee crisis,’ in terms of livelihoods and land-use. While I expected extensive land-use changes as farmers sold or converted farms, triangulation of ethnographic, survey and remote sensing data painted a different picture: farmers were creatively maintaining coffee-forests² (Hausermann, 2014).

Jenaro excused himself while we chatted about the upcoming birthday party and ejido elections. With Jenaro out of sight, Lucia whispered, ‘You study las fincas [coffee farms], but you should ask about la reserva [forest reserve].’ She explained men from the family occasionally felled trees in a nearby forest reserve at night, a risky practice for its illegality and threat of injury. Trees were transported to cities as construction materials. Lucia was clearly concerned about this practice, saying several times how ‘very dangerous’ it was. I felt worried as she spoke, understanding Lucia cared for her husband, Luis. I also understood that Luis’s injury, death or arrest would create serious burdens for the family.

When Jenaro returned, the conversation quickly circled back to party and election talk. While Lucia and I spoke several times in the following weeks, we never again discussed timber harvesting. Three weeks later, when I visited the reserve to ‘ground truth’ cloud forests, I searched for selectively logged trees and found one.

Lucia’s admission certainly fell within the realm of my research interests: illegal logging is a method for coping with economic stress (Debela et al., 2012). However, I was more struck by the disclosure of this information: I was let in on a secret about gendered and illegal deforestation, one which Jenaro – for his maleness, citizenship status and/or employment at a regional university – was not privy to.

I understood how my positionality shaped research relationships and data collection in Mexico: as a white, educated American woman, participants rarely declined participation. Government officials, who were always men, attempted to flirt before turning over the requested information. I avoided state officials outside government offices (despite invitations) concerned that uninvited flirtatious remarks would precede other forms of harassment.³ In other sites, my positionality prompted care-taking behaviors. Single and childless, American and white, rural families fed and housed me, invited me to parties, and worried about my relationship and fertility status. The interaction with Lucia, however, illuminated how positionality impacts knowledge production. As field-based researchers, the stories and experiences we are told, and thus the information we gain, are shaped by entanglements that include race, gender, laughter, tamales, age, trees, and a male research assistant’s bathroom break.

As two geographers investigating land-use change in rural contexts, we examine how our positionalities – as a black, Ghanaian graduate student and white, tenured professor – produce differential fieldwork dynamics. We link our fieldwork experiences and relations to methods and results. We also discuss moments in academic spaces (conferences, job talks, etc.) where positionality played a role in attempts to define what ‘counts’ as land-change science (LCS) research.

We understand that reflections on positionality – in scholarship, classrooms, and more intimate spaces – are challenging. However, such conversations help examine the conditions under which fieldwork and method are conducted, with important implications for knowledge production and research ethics. Respectful discourse on positionality can create more open and inclusive relations with students, research assistants, colleagues and others. We focus on the usefulness of positionality in LCS because, while the field is a critical component in LCS research (Brannstrom & Vadjunec, 2013; Perez et al., 2013; Turner & Robbins, 2008), there is a dearth of critical work on positionality.

The interdisciplinary field of LCS addresses land dynamics as foundational to global environmental change (Lambin & Geist, 2006; Turner et al., 2008; Turner & Robbins, 2008). Within LCS circles, land is often understood as ‘coupled socio-ecological systems’ and scholars address land-use change through integration of natural, social, and geographic information sciences, including remote sensing (Turner & Robbins, 2008, p. 298). The Global Land Programme (GLP), the largest international community of land scholars, encourages interdisciplinary and international collaboration, holding Open Science Meetings, providing financial support to researchers from the Global South and promoting the work of women and junior faculty (Brannstrom & Vadjunec, 2013). GLP focal themes reflect the current breadth of LCS scholarship and range from technical telecoupling and land trade-offs for ecosystem services to land governance and conflict (Global Land Programme (GLP), 2021).

While traditionally focused on modeling and geospatial techniques (Brannstrom & Vadjunec, 2013), LCS scholars employ mixed methods, including remote sensing, household surveys, ethnography, participatory techniques and policy analyses (Eakin et al., 2016; Kelley et al., 2020; Kinnebrew et al., 2021; Kleemann et al., 2017; McSweeney, 2005; Munroe et al., 2005). While studies have produced insightful studies of how gender, class and ethnicity impact land-use decisions and patterns (Casey et al., 2017; Green et al., 2020; Radel et al., 2017; De Sherbinin et al., 2008), how researcher positionality shapes knowledge production about land systems remains largely unexplored.⁴ In the next section, we review feminist scholarship on positionality to ground the reflections that follow.

Feminist scholarship on positionality and linkages to land

Donna Haraway’s theorization of ‘situated knowledges’ is foundational to feminist thinking on ‘the field’ and knowledge production. Pushing against taken-for-granted constructions of ‘objectivity’ in western science, Haraway argues scientists are not neutral and objective observers, ‘”Being” is much more problematic and contingent’ (D. Haraway, 1988, p. 583). Haraway deconstructs scientific objectivity through empirical observation in primate laboratories, detailing how researchers project western constructions of gender onto primate behavior (D.J. Haraway, 1989). Counter to performances and rhetoric of scientific objectivity, Haraway offers a ‘feminist objectivity’ termed ‘situated knowledges,’ perspectives and experiences that produce multiple, contingent and always partial ways of knowing (D. Haraway, 1988, p. 581).

Feminist scholars expanded Haraway’s work to examine relationships between researcher, participant, field sites, research assistants and knowledge production (Domosh, 2003; England, 1994; McDowell, 1999; Nast, 1994; Radcliffe, 1994; Sharp, 2005; Staeheli & Nagar, 2002). Gillian Rose urged reflexivity that ‘looks both “inward” to the identity of the researcher, and “outward” to research relations in “the wider world”’ (Rose, 1997, p. 309). And Sharp and Dowler (2011, p. 146) argue ‘the field’ is constituted through practice and interaction rather than existing as ‘a pre-existent and stable place awaiting discovery by the field researcher.’

In international fieldwork contexts, scholars have examined the relational and emotional contradictions that constitute research, including vulnerability to harassment (Berry et al., 2017; Clancy et al., 2014; LaRocca et al., 2019; Mahmood, 2008). Smith (2014) explores how intimate conversations in Ladakh, India shape angst in other career moments (e.g. copywriting knowledge). Sultana (2007), raised in urban Bangladesh, unpacks how intersecting identities mediate relations with rural people during dissertation research. And Berry et al. (2017, p. 540) produce insightful accounts of how – as black, brown, indigenous, mestiza and/or queer cisgender women – ‘the field travels with and within our bodies,’ arguing for a politically-engaged ‘fugitive anthropology’ that grapples with physical, emotional and spiritual well-being.

This article builds from this work to examine the relationships between researcher positionality and knowledge production on land systems and land-use. Reflection on positionality promotes better understandings of power dynamics in myriad contexts; it can also, we argue, lead to more thoughtful and ethical methods and data interpretation on land. We utilize mixed methods in our work including ethnography, remote sensing, household surveys and environmental sampling and analysis . We also understand land as infinitely complex and, for Indigenous people, constituted through interspecies kinship relations and linguistic, artistic, ceremonial, and philosophical traditions (Whyte, 2020; also Deloria & Deloria, 2006). Simpson (2017, p. 43) explains, ‘Indigenous bodies don’t relate to land by possessing or owning it or having control over it. We relate to land through connection – generative, affirmative, complex, overlapping, and nonlinear relationship.’ Land-use is always situated in cultural, political-economic, and biophysical contexts, including diverse ontologies.

Reflections on positionality do not dismantle existing power relations or research extractivism, nor does such analysis bring about dramatic changes for the struggling populations with whom we work (Chattopadhyay, 2012; Mbembe, 2021; Mitchell, 2013). However, ‘the alternative of not heeding such issues is even more problematic’ (Sultana, 2007, p. 383). Arguing that the discussion of positionality falls beyond disciplinary responsibilities reproduces legacies of white heteropatriarchy (Berry et al., 2017).

Clearly, epistemology is also at stake. Like Lucia’s disclosure of illegal timber harvesting, research relations and positionalities produce situated scientific knowledges about land. Neely and Nguse (2015, p. 142) explain, ‘Quite simply, knowledge is shaped by the people and places involved in its creation … ’ What and how we know is fundamentally developed through co-constitutions between method, positionality and research relations.

In the following sections, we discuss how our specific positionalities and identities shape method and knowledge production. We begin with brief reflections on positionality, while acknowledging identities shift over space and time. It is important to note that Heidi is Janet’s dissertation advisor. Janet is a doctoral candidate at Rutgers University and did two fieldwork sessions totaling 8 months. Heidi completed her dissertation in 2010 and thus has more fieldwork/research experience. While we differ in terms of race, household income, academic rank, and other important factors, we take positionality seriously in our mixed-methods, field-based research on land-use change.

Heidi: ‘Any white person coming all this way to ask about mining has power.’

I identify as a white woman, American and Swiss citizen and associate professor of geography. I grew up in a middle-class, single-parent household in Oregon, where I explored – by bike and foot – the agrarian landscapes that extended from the urban boundaries of my hometown. The summers I was 12 and 13 and too young for work permit, and my mother dropped me off at berry farms, where I worked alongside Mexican migrants.⁵ I mention these experiences because rural landscapes, from Oregon’s Willamette Valley to the Swiss alpine cheese-making huts operated by my dad’s family helped constitute my childhood, shaping scholarly interests in rural land-use, race and livelihood.

My masters and dissertation fieldwork with Veracruz coffee famers totaled 18 months between 2003 and 2008. The post-doc that followed introduced me to fieldwork and compelling land-use and justice questions in Ghana, where I have led projects since 2010, including those on gold mining, disease and dam development. I was single and struggling with untreated mental health issues in graduate school. Research in Ghana has coincided with a heterosexual marriage, the birth of two children and (ongoing) management of anxiety. I speak English, Spanish and Twi (basic). Research described here has been conducted with the help of research assistants, all of whom I pay; most are co-authors on articles.⁶

Within academia, I identify as a political ecologist and feminist scholar engaged with LCS questions and debates. I always employ mixed-methods and collect primary data through ethnography (interviews, participant observation), household surveys, ground truthing, environmental sampling and document analysis. These methods are time and labor intensive, requiring frequent trips abroad; fieldwork has become more difficult with children. I also use remote sensing techniques to detect land-use change through time-series analyses. Triangulation of diverse datasets enables comprehensive understandings of how and why landscapes change.

Privilege, method and implication for knowledge production

A significant way my positionality has impacted knowledge production on land systems is through largely unencumbered access to ‘research subjects.’ Since 2003, I have recruited nearly 1,500 participants for surveys and int in Mexico and Ghana. These folks include rural smallholders, gold miners, traditional authorities, government officials in various state agencies, health practitioners, international development experts and NGO employees. In nearly two decades of leading research abroad, only three people have declined to participate in my projects. By contrast, when I recently carried out interviews on hydraulic fracturing in Colorado, half the folks contacted for interviews did not respond or declined participation.

Cultural and racialized postcolonial conditions, as well as people’s graciousness, are entangled with my race, class, gender, education and citizenship privileges as a foreign researcher. While people voluntarily participate in projects for their own reasons – curiosity, boredom, generosity, hope – I am acutely aware of the power-laden nature of these interactions and how my career benefits, thereby reproducing neocolonial relations. While I believe there are valuable and transformative practices forged through my research,⁷ I am also deeply conflicted about doing research abroad as a white, well-educated, American woman.

Wide participant engagement in research – across diverse settings – has implications for scientific knowledge production. First, various perspectives can be analyzed for trends and tensions. In both Mexico and Ghana, I documented disconnects between the narratives and practices of state officials and lived experiences of rural smallholders, demonstrating how land-use changes – driven by corruption, gold markets, coffee deregulation – are not only unjust, but often ‘illegal’ (Hausermann, 2012, 2018; Hausermann et al., 2020; Hausermann & Ferring, 2018; Hausermann et al., 2018). Recently, other LCS scholars have produced excellent work on the ways illegal activities, including narcotrafficking, drive land-cover change (Bonilla-Moheno & Aide, 2020; Devine et al., 2021; Tellman et al., 2020; Yang et al., 2021).

Ghanaian government officials, for instance, perform ‘regulatory fictions’ wherein mining processes and land-use changes (according to them) play out according to law and policy (Hausermann & Ferring, 2021). Through fieldwork, however, I observe firsthand where laws and policies are broken, which is also confirmed through remote sensing analyses. According to law and state officials, mining should not occur within 300 meters of a water body. Yet, mining pits flank rivers for miles. Streams and rivers are dredge-mined and sometimes rerouted to wash sediment (Hausermann et al., 2018). Farmers and traditional authorities have led me to former lands transformed (illegally, and often without land user permission) into mining pits.

Walking landscapes extensively with rural people helps categorize pixels in remote sensing analyses and reveals disconnects between land-use ‘categories’ and existing ecologies and local knowledge (also Robbins, 2001). In a time-series analysis of gold mining expansion between 2008 and 2013 polygons were manually drawn to classify abandoned mining pits based on expert knowledge and field experience. And in Mexico, working alongside farmers helped differentiate coffee, cloud forest and other land-uses in Landsat data. Long-term commitment to research sites, combined with largely unencumbered access to sites and people (due to privileges described above) verify and classify land-uses in complex landscape matrixes using remotely sensed data.

People’s willingness to participate in research also ensures solid sampling techniques. In spatially stratified household surveys, the general rule of thumb is to survey 30% of the community (Isreal et al., 2013). Access to participants has ensured this goal is met or exceeded in foreign fieldwork contexts. In Mexico, I easily hit goals of surveying 40% of the households in two case study communities. And in Ghana, I surveyed between 40 and 100% of households, depending on community size. Survey questions queried land-use, land acquisition processes, farming practice, and disease incidence and vulnerability, providing correlative understandings of land-use changes and implications. Again, survey data are triangulated with land-use, geospatial and ethnographic data to provide robust analyses of land-use change.

Intimacy, disclosures and empirical insights

Of the many things required of academics (writing, public speaking, teaching, analysis, project leadership, collegiality and so on) those involving social interaction and relating as humans come most naturally to me, i.e. teaching, mentoring and engaging ‘research subjects.’⁸ I always attempt to develop rapport with people, emphasize things we have in common, listen attentively, make jokes (also Smith, 2014). While these performances are, in part, what constitutes the research itself, they rarely feel inauthentic, and I enjoy being decentered as an ‘expert’ and listening to others explain experience and insights.

Across study sites, I am continuously surprised by people’s openness and ability to discuss intimate and difficult matters with me. I am told secrets – about love affairs, suspicious deaths, family dramas, land dispossession, stigmatized illnesses, and illegal logging. While many disclosures occur after months or years of interaction, some are disclosed quickly as part of wider narratives about communities and land. My ‘outsider’ positionality may help people feel safe in disclosing information: I am not engaged in family or local matters and may appear relatively neutral vis-à-vis stigmatized or scandalous issues. My temporary presence in communities may also put people at ease.

People’s openness, however, burgeons from more than insider/outsider binaries. As D. Haraway (1988, p. 588) puts it, ‘Feminist accountability requires a knowledge turned to resonance, not to dichotomy.’ At times, it has seemed like people needed to talk, and I was there, ready to listen. In Mexico, a woman joked after an emotional discussion about her husband’s migration to the US, ‘I’ve never been to psychotherapy … is this what it is like?’ While these moments feel genuine, intimacy can muddle the uneven power dynamics that shape research relations. As Smith (2014, p. 3) writes, ‘the generosity of research “subjects” can obscure questions of knowledge production. If it feels easy and natural to chat with someone about their life, it makes it easier to forget our place within the spatial structures of power that render one person the interviewer and one the interviewee.’

Many ‘secrets’ have led to new empirical insights about land-use. A Ghanaian friend working for Chinese miners explained he secretly visits farmers before his Chinese bosses, advising them on what to say and do to receive the most money for their land. Women in central Ghana explained the local chief sold land to miners without their permission, but always sought permission from men. In Mexico, farmers confessed they received coffee subsidies for land converted to other crops. These disclosures, combined with other datasets, led to new scholarly insights. Yet, the tensions between intimacy and knowledge production, including how my career and livelihood have profoundly benefited, create anxiety and ambivalence for me around research practice (also Smith, 2014).

Positionality and experiences in academic spaces

Knowledge and positionality play out differently across time and space. While in fieldwork experiences described above, I have embodied racial, class, citizenship, educational and other privileges, I have been situated differently in academic sites based on my gender, age, and academic rank. I include the following experiences because, like relationality in the field, they also shape and mediate knowledge production, including attempts to control what ‘counts’ as ‘good’ LCS scholarship.

On a conference panel years ago, a senior scholar consistently interrupted my results to share his experience of Latin American land-use change. During my allotted 15 minutes, he interjected without invitation, practices the session moderator allowed. In another conference setting, a senior scholar pressed me relentlessly and aggressively to solve the ongoing conundrum of linking pixels and environmental practice (Liverman & Cuesta, 2008; Rindfuss et al., 2004).

I have had similar experiences during job interviews. Following a job talk, I was critiqued for not publishing the research in a remote sensing journal, thereby making the work have a higher impact. In another interview, a senior faculty member argued, ‘ethnography is a waste of time,’ and I should stick to geospatial methods, while at a liberal arts college two anthropology professors asked how I was ‘maintain neutrality’ in ‘the field’ and rolled their eyes at each other as I answered from a feminist science studies perspective.

These experiences occurred when I was a graduate student, post-doc or very early on the tenure-track. All but one of these encounters were with white, tenured male professors. While I have had many more supportive experiences with colleagues of diverse identities, these moments left me feeling dismissed and unfairly criticized. I internalized these critiques, fueling my anxiety. Scholars have addressed white, heteromasculinism at conferences (Domosh, 2014), in research (Faria & Mollett, 2016; Gilmore, 2002; Mott & Roberts, 2014) everyday academic spaces (Mahtani, 2014; Peake & Kobayashi, 2002; Sanders, 2006) and citation politics (Mott & Cockayne, 2017). Mansfield et al. (2019) explain of harassment in geography, ‘Women who successfully run the gauntlet enter their professional lives with an extra burden of defensiveness and battered intellectual confidence.’ In reflection, I am particularly surprised by others’ silence in these moments, especially those that occurred when I was a graduate student. These performances, including passive observation, reproduce and reinforce geographic patriarchy. Moreover, dismissing methods, results and journal choices, commentors sought to define what types of knowledge and method ‘count’ in LCS rather than entering the dynamic with an openness for mixed methods, publication choices, and the messiness of research relations.

Janet: From ‘less endowed school’ to visa lottery, Rutgers PhD and ‘the field’

I [Janet] write as a first-generation student and woman of color from a family background dominated by farming and petty trading livelihoods. I was born in a small village adjacent to Lake Bosomtwe, south of Kumasi, Ghana.⁹ When I was fourteen, my fourth-grade teacher transferred to a larger township and took me with him to attend junior high school. He perceived I was a bright student and was close to my mother.

My sister put me through secondary and tertiary education. In Ghana, such support among siblings is common. My sister, who is ten years older, did not attend high school because my mother, a single mom, could not afford it. She perceived secondary school was a privilege afforded to wealthier children, particularly boys. Denied of this privilege, she worked hard running a drinking spot (bar) so I could have an education.

I entered Kwame Nkrumah University of Science and Technology (KNUST) as a student from a ‘less endowed school,’ an identification given to disadvantaged students. My sister continued to help with school fees. In return, I worked at the drinking spot during vacations. I did bachelor’s and master’s degree in geography at KNUST. My master’s thesis focused on Kumasi’s urban expansion and livelihood impacts, including in the peri-urban community where I lived as a child.

In 2013, I won the diversity visa lottery and immigrated to the United States. My initial experience in the US was like many African immigrants. I worked various jobs to support myself and the family I left behind. I served coffee, performed caregiving and clerked in an African grocery store. While this was a difficult time, I recognize how my privileged status as an educated woman opened doors for employment and entrance to Rutgers.

I entered Rutgers’s doctoral program in geography in 2016 and married and had a child the following year. My husband was in Canada working most of this time. Managing childcare and work on a graduate student stipend has been challenging. In graduate seminars, most material I encountered was rooted in Anglo-European political philosophies. I felt overwhelmed and alienated at times, although faculty members were overwhelmingly supportive. The absence of Indigenous ontologies in coursework resulted in my dissertation project, which focuses on cultural practices in Ghana’s small-scale mining industry.

In 2019, I returned to Ghana for dissertation fieldwork with my husband and one-year-old daughter. Doing research without my child was not an option for various reasons, including the undesirability of long-term separation and our inability to afford childcare (also Jenkins, 2020). I conducted dissertation research in two mining communities, Odumasi and Tongo, using semi-structured interviews, informal group discussions and participant observation.

‘Insider’ as a contradictory identity

Ghanaian small-scale mining has gained political and media attention, largely framed as ‘illegal’ and dangerous. Many miners fear arrest. Within this political atmosphere, miners initially viewed me with suspicion. I was positioned as an ‘outsider,’ and my motives deemed suspicious, even in the country of my birth. In Odumasi, I navigated this process through marriage. My brother-in-law helped introduce me to local miners. As we approached the site, a team leader started making gestures, yelling: ‘Hey, hey, leave this site, leave this site, we don’t want you here,’ urging others to drive us away. As we got closer, a woman recognized my brother-in-law and reassured the miners we were friendly. My brother-in-law made introductions and miners quickly identified me as ‘wife of OB’ [husband’s household name]. My status as ‘wife of OB’ was significant for negotiating relationships and access. Miners opened-up about mining practices, including mining at night to evade detection. Due to familial relationships, I gained important insight and information, thereby leading to knowledge production on miners’ extractive practices.

Objects were also important in signaling positionality. While my marital relationship constructed me as an ‘insider,’ the pen, notebook, and papers I carried positioned me differently. Such materials – and ideas about them – were constant reminders of my otherness. In Ghana, people walking around with pens and paper are constructed as akrakyefoɔ [the educated], often assumed to be government workers or urban elites. Some miners questioned whether I was a state official. Others referred to me as nsemtwerɛnii [journalist], and jokingly asked me to ‘report’ not only land destruction, but also how miners struggle to dig through hard rock before accessing quartz veins. Miners were constantly curious about my notetaking. Like giraffes, some stretched their necks to read my notes. When I received phone calls, people joked I was calling the police. At other times, objects were connective – I used my phone to show pictures of my daughter, or people used my phone to play music. While I did not take my daughter to mining communities,¹⁰ showing pictures of her, usually in response to questions about my marital and fertility status, helped break down barriers, even with male miners, who would then share about their families.

‘I now consider you a daughter, I am no longer afraid of you’

State officials, including police, visit mining sites to discipline miners and secure bribes. One cloudy afternoon, a male miner shouted, ‘Scatter, they [state officials] are here!’ Suddenly, everybody ran helter-skelter. Adam, a mining site leader, told me to run. One woman quickly interjected, ‘But she is not one of us.’ Adam responded, ‘When they come, by the time you open your mouth to reveal your identity, they will have beaten you to death.’ Adam took my bag and I quickly followed. We climbed a mountain, running through farms. While climbing, I noticed many people were uncomfortable with my presence. One male miner murmured, ‘Why is she still following us? She will show them our hiding place.’ While these comments made me uncomfortable, I understood miners’ fear and concern.

On the mountain, we watched officials patrolling the site, seizing mining equipment and making phone calls. After three hours, several women decided to return home via a different route. I accompanied them and while descending the hill, they discussed everyday struggles, fears and experiences. They explained male miners persuaded them to work at dawn, despite other household responsibilities, so the mining day would end before officials arrived. Other women intentionally misrepresented identities by acting as farmers, carrying cutlasses and firewood.

Evading arrest together became critical in breaking down barriers, particularly with women, who became interested in the research and invited me to their homes. They began to confide in difficulties and desires. As one woman explained, ‘I now consider you a daughter, I am no longer afraid of you.’ This statement gave me comfort and reflected connection and trust. After our ordeal hiding from the police, women shared information more freely, thereby shaping the results of my dissertation, which links land-use, gender, and extractive practice. Connections and relations, including those forged through a mutual fear of state violence, were vital to enhancing my understanding of women’s experiences in mining.

Beyond land surface: Local ontologies in resource extraction

Recently, as scholars have sought to think through complex interrelations, Indigenous ontologies have been examined as an important competent of land and natural resource management (Hausermann, 2019; TallBear, 2015; Theriault, 2017). Sundberg (2014, p. 34) critiqued her previous scholarship for privileging western philosophies over Indigenous ontologies thereby reproducing, ‘colonial ways of knowing … further subordinating other ontologies.’ Delivering the keynote address at DOPE 2015, Kim Tall Bear stated, ‘As academic conversations seek new language to articulate relations between humans and nonhumans, they will benefit from indigenous standpoints that never forgot the relatedness of all things.’

In rural Ghana, I grew up around understandings of the world wherein spiritual, human and biophysical realms are deeply interconnected. In my native village, fishermen do not fish on Sundays for the lake deity to rest. The lake has been gendered as a male whose natal day is Sunday. Fishing on this day is ‘taboo.’ Fisherfolk who ignore the taboo may suffer serious consequences like death and sickness. The lake’s shore is also designated as asɔneyɛso (shrine), where people offer libation and sacrifice to the lake deity. Farming around the asɔneyɛso is prohibited because of its sacredness. These understandings deeply impact land and resource management. So, I was not surprised when miners began describing spiritual ontologies of gold.

One evening, for instance, Kwaku, a team leader, and I sat in front of his family house chatting.¹¹ Earlier that day, Kwaku’s boys (mining team) and another team had a fierce argument, an exchange I witnessed. The other team was reportedly using juju [spiritual intervention] to draw gold from neighboring pits. Acutely aware of such understandings, I asked why Kwaku’s team suspected the other gang was using spiritual means. Kwaku looked down at the concrete floor and responded, ‘This world is not there for nothing,’ giving credence to how physical and spiritual entities are deeply interconnected. In extraction, gold is viewed as having capacity to interact with spirits. Ritual sacrifices can thus hasten or transform gold production and ore spatiality.

Kwaku explained that while he suspected neighboring miners were using juju to pull gold to their site, his team also visited a spiritualist. They presented cola nut and gin as sacrificial items. The spiritualist, in turn, gave them a ‘spiritual liquid’ to sprinkle on the site. I was surprised by Kwaku’s admission and took it as a sign of trust; while people often accuse others of such practices, they rarely discuss their own involvement. This demonstrates the importance of relationships and affect in research processes (Neely & Nguse, 2015).

I learned other ways gold, spirits and taboos shape land-use practice. In Odumasi, mining does not occur on sacred days. Extraction can be delayed for weeks or months to make space for rituals. Understandings of sacred forest groves or waterways associated with spiritual entities can keep mining out of these sites (also Barre et al., 2009). One miner remarked, ‘our activities cause limited destruction to water bodies,’ differentiating ‘Ghanaian’ mining from more destructive Chinese operations that flank or dredge rivers.

Women also shared intimate stories about taboos. Menstruating women are forbidden from entering sites due to beliefs that blood will drive gold away. Whereas chicken blood via juju sacrifices can help attract gold to a site, women’s bodily processes are stigmatized, revealing the ways biology, gender and spiritualism combine in ways that can be contradictory and exclusionary. Women are also prohibited from mining underground. Women’s openness about menstruation and associated taboos was likely due to my positionality as a Ghanaian woman. These interactions helped me understand that traditional beliefs hugely shape gendered mining practices and livelihoods.

Discussion

While the transformative possibilities of feminist reflection on positionality and knowledge production are context specific, we offer several, broad implications for LCS scholars. First, positionality is deeply tied to method and knowledge production of land systems. In field-based contexts, participants’ willingness to engage in research yields rigorous sampling and data collection techniques, enhancing understanding of land-use, state-society tensions, dispossession, and so on. In the examples above, we directly linkour positionalities and research relations to specific methods and empirical insights.

Heidi (due to race, citizenship and other privileges) has never had problems recruiting participants in research abroad or finding governmental or academic collaborators. Janet, by contrast, encountered various challenges in the field despite being Ghanaian. In politicized mining environments, community members assumed Janet was an undercover journalist. State officials, moreover, were more dismissive of her research. Yet, when Janet was able to dismantle suspicions and gain trust, she formed close bonds with Kwaku and women miners and communicated directly in Twi, leading to new insights about how ontologies and spiritual practices shape extraction. Despite our differences, openness to self-reflection, in the moments and relations that constitute research, helps connect with others, thereby enhancing the stories we hear and, thus, the stories we tell.

Second and relatedly, reflection on positionality and methodology are co-constituted. Diverse research participants are often keen to show Heidi (white, educated, American) land, from coffee and cocoa production to subsistence farms, mining pits and dispossessed lands. Janet, on the other hand, traversed landscapes not as a method but while fleeing from police who would not differentiate her from miners during raids. ‘Knowing’ the landscape with rural people differs based on research relations, positionality and has a direct impact on method, including geospatial techniques.

Geospatial tools such as GIS and remote sensing wield significant power (Bryan, 2011; Fernandez-Diaz et al., 2018; Fraser, 2007). Heidi worked on a project (not as PI or Co-PI) where researchers were tasked with mapping ‘everyday activities’ for case/control matches around a tropical disease. It quickly became clear how stigmatizing and problematic this practice was, creating complicated project politics for those who refused to carry out the mapping method. As Nichter (2008, p. 13) explains of health-based research, those ‘who seek to … investigate … certain diseases can easily become complicit in “othering projects.”’ Beyond researcher positionality and power in mobile methodologies (Fraser, 2007), we must be attentive to the complexity of newly created spaces and subjectivities (re)produced through research practice (Ferring, 2013).

While there has been important reflection on the ethics of geospatial and mapping technologies (Carolan, 2004; Fernandez-Diaz & Cohen, 2020; Fisher et al., 2021; Fox et al., 2005; Myers, 2010; Propen, 2006; Slonecker et al., 1998; Wasowski, 1991),¹² there is still space for critical reflection on the political nature of geospatial methods and tools. As Brannstrom and Vadjunec (2013, p. 10) put it, ‘Land change science products develop political lives because they enter worlds of resource access, where people, across the spectrum from elite to marginalized, use them to control, define, and exclude access to resources and territory.’

Mixed methods research combined with reflection on positionality can shape careful use of geospatial techniques. We do not publish sensitive data that could put already vulnerable people at risk of land dispossession or other deleterious implications; this knowledge – of how and what to publish – is produced through relational and self-reflective qualitative research. For instance, Heidi delayed publishing maps of coffee production in Veracruz until government subsidies, based on acreage in coffee production, changed. In Ghana, sacred forests were not indicated on maps due to community concerns that they would be destroyed for gold mining. And when district-level bureaucrats requested maps of abandoned, water-logged pits to lobby for additional malaria resources from Accra, we made them. In other words, research relations and qualitative data help inform how maps are created, including spatial representations that protect, and in some cases assist, rural research participants. Considering the many ways specific mapping projects have gone very wrong (Dunbar-Ortiz, 2015; Wainwright, 2012), reflections on positionality, power and geospatial techniques are critical to LCS scholars and geographers. Attending to existing (and emergent) dynamics of power in mapping and other geospatial techniques, and the importance of ‘nonexploitative collaboration’ (Elwood, 2006, p. 197) is a critical dimension of research practice.

Thirdly, reflections on positionally can guide inclusive research practices. Speaking openly with research assistants and participants about identities and power imbalances promotes transparency, dialogue, and space for people to voice desires, frustrations and needs. When taking on research assistants or students, Heidi always discusses short and long-term goals as well as limitations or challenges, including constrained financial resources, mental health, childcare, and so on. Strategies are developed to meet goals that include proactively dealing with existing and anticipated challenges. A similar process happens in research communities, with individuals, teachers and/or local leaders. These conversations require reflection on privilege, struggle and limitation while helping create space for honest dialogue and goal setting that acknowledges mutual support.

Positionalities, moreover, shift over space and time. Whereas Heidi’s international research is enabled, in part, by intersecting privileges, in other spaces (conferences, job talks) she has been dismissed as a graduate student, junior scholar, ethnographer and/or woman. These interactions have policed the boundaries of what ‘counts’ as valuable and viable knowledge on land-use change. Pulido (2002, p. 46) writes, an ‘often-overlooked dimension of a white discipline is the limited set of experiences that inform the discourse.’ All knowledge, including method, is partial and situated. The critiques Heidi experienced in academic spaces fail to understand the exciting possibilities of collaborative, interdisciplinary and mixed methods research on land, offering important lessons for LCS scholars. When we reflect on practices and positionalities, in the field and academic spaces more broadly, we help create more inclusive spaces, in terms of social differences, epistemology and methodology. As Janet demonstrates, moreover, making space for Indigenous people and/or scholars from the Global South helps elucidate diverse understandings and uses of land, including spiritual ontologies.

Conclusions

The LCS community has diversified tremendously in recent years, epistemologically and in terms of social differences among members. The Global Land Programme intentionally broadened the field’s breadth of inquiry and includes women, people of color and scholars/students from the Global South in conferences, committees, and networking (Brannstrom & Vadjunec, 2013). This is important work that will help create more inclusive spaces to think through complicated land dynamics that are clearly the outcome of complex biophysical, cultural, and political–economic interactions.

Reflections on positionality, in ‘the field’ and academia more broadly, can help bolster efforts to make LCS more inclusive and interdisciplinary. This article has addressed fruitful cross-pollinations between positionality, method and knowledge production on land. We have also addressed how positionalities shift across space and time, and the ways we have benefitted from privileges in particular sites and experienced anxiety and vulnerability in others, including vis-à-vis others’ efforts to police the boundaries of ‘viable’ LCS research. We acknowledge critical self-reflection on positionality is difficult work. An openness to feminist perspectives on knowledge production and positionality, as well as a commitment to talking openly across difference – with students, collaborators, colleagues and community members – is crucial. Like most things, these conversations feel less awkward the more they are practiced.

Validation and disruption are also important. Validating others’ efforts to reflect on positionality, through careful listening and support, is vital to normalization of these practices in classrooms, conferences, field sites and so on. Likewise, disrupting performances of white, heteropatriarchy is also important. This may take the form of defending students at conferences, not leaving job candidates or students alone with known aggressors when institutions fail to act, and openly discussing positionality. These practices, moreover, should be taken up by scholars with the most academic power, i.e. tenured faculty members, and not be the burden of women and people of color alone.

Acknowledgments

We would like to thank Steve Leisz, John Lindenbaum, Carrie Mott, Darla Munroe and the two anonymous reviewers for helping strengthen this piece.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. Names of research participants are pseudonyms.

2. In Mexico, coffee is largely grown beneath shade trees. ‘Coffee-forests’ possess diverse environmental benefits, from high biodiversity and carbon sequestration to water catchment and wildlife corridors.

3. As I developed relationships with state officials, I learned to trust several outside the office. I accompanied these individuals (along with interns and students) to experimental fields and other state research sites.

4. A wide search and review of literature in June 2021 did not find any papers identifying as ‘land-change science’ that deal – even cursorily – with positionality, although it is possible we missed important articles.

5. Holmes (2014) describes the racial hierarchy of berry-picking in Washington State, including privileges afforded white teens compared to Indigenous Mexicans.

6. Research assistants have been of various ethnicities, gender identities, education levels, disciplinary training, nationalities, etc. Discussing how their positionalities impact research is beyond this article’s scope.

7. I design research around community concerns, create reports and graphics for bureaucrats who lobby for resources to do their jobs effectively, fund locally determined projects like libraries and shaded lory stops, help Ghanaian and Mexican students complete degrees and apply to graduate programs, support community members with health, lawyer and education fees. These practices are always carried out in conversation and collaboration with Ghanaian and Mexican colleagues. While such practices do not undo uneven power dynamics that constitute research, they can be productive and promote sustained, open conversation with students, collaborators and community members.

8. When I go into gas stations, my oldest child will say, ‘Mom, don’t talk to too many people in there. I’m coming to get you after 7 minutes.’ Ability to talk across difference in various sites is something I value deeply and practice often, even unwittingly and to the annoyance of patient people in the car.

9. Kumasi is capital of Ashanti Region.

10. I was uncomfortable taking my daughter to mining communities for fear of vehicular accidents, a leading cause of death in Ghana. As a mother-researcher, I rode a motorbike on rough roads in isolated areas, but was uncomfortable with my child doing so. I made these decisions with empathy toward community members who could not make similar choices.

11. Initially, Kwaku was also deeply skeptical of my research. Over time, we became close and bonded over having children of similar ages. He taught me a great deal about the role of spiritual ontologies in gold mining.

12. Much of this literature is from archeology.