key: cord-0967487-ag8aq2hv authors: Jenkins, Katy; Romero Toledo, Hugo; Videla Oyarzo, Angélica title: Reflections on a failed participatory workshop in Northern Chile: Negotiating boycotts, benefits, and the UN declaration on the rights of indigenous people date: 2020-09-03 journal: Emot Space Soc DOI: 10.1016/j.emospa.2020.100721 sha: 7be43f6c50575cc56c886e509332ab6efb8022ec doc_id: 967487 cord_uid: ag8aq2hv In this paper, we critically analyse our experiences of initiating participatory research in the challenging context of the Atacama Desert, Northern Chile. We use our experience of organising participatory workshops with Aymara and Quechua women community leaders to reflect on the politics of participation/non-participation, and explore these experiences in light of our multiple and overlapping positionalities as Chilean/British, male/female, white/mestizo. In the light of one workshop being entirely unsuccessful, we discuss the ways in which our empirical and methodological thinking has nevertheless been enriched by this experience. We situate the challenges we faced in relation to negotiating the tensions presented by debates on decolonising research from our positions within the neoliberal academy, exploring the questions raised by indigenous women activists' research ‘refusal’, and critically reflect upon the emotional responses this situation elicited in each of us. We argue for the importance of embracing such apparent fieldwork ‘failures’ and, recognising the resulting emotional swirl of panic, anxiety and inadequacy that they produce, emphasise these experiences as illustrative of the inherent tensions around decolonising research, as well as an often inevitable element of conducting research with marginalised communities involved in socio-environmental conflicts. An empty seminar room, three anxious academics … reminding ourselves no one ever arrives on time in Chile … reviewing who had confirmed, checking for messages … remembering this is a normal part of fieldwork, nothing ever goes to plan … Finally, a participant, an Aymara woman community member, an activist and lawyer, arrives. We are briefly relieved, if one has come surely others will be on their way too? Then, somewhat apologetically, she explains that she is not surprised that no one is there, and that she has come to explain why no one is coming to the workshop, why we have been 'boycotted'. In this paper, we explore this scenario, the 'reasons' for the workshop's apparent failure, the role of emotion in negotiating the aftermath of this, and the implications for using participatory methodologies in the context of researching resistance to large scale resource extraction in Chile. Throughout, we emphasise the importance of 'writing in' emotion as a key way of making sense of fieldwork experiences in challenging contexts, reflecting on 'fieldwork failure' as "an event or experience that affects the research process in a manner perceived as negative by the researcher" (Harrowell et al., 2018: 231) . Positionality is central in situating these discussions; whilst our identities cut across global South and North origins, and we are of different ethnic backgrounds, none of us are indigenous, 1 and this has become increasingly significant in the course of this research, especially in the increasingly contentious political context of contemporary Chile. 2 Throughout the paper we include our personal perspectives (where indicated in bold) as well as excerpts from our field notes, aiming to make explicit our differing positionalities and emotional responses where appropriate, whilst recognising our commonalities as academic researchers (see Caretta, 2015 on the importance of making visible how multiple positionalities, of researchers, participants and research assistants, intersect in the context of collaborative and cross-cultural research). This paper reflects on a project led by Katy Jenkins and Hugo Romero Toledo, focused on exploring participatory methods as a means of developing new perspectives on gender and indigeneity in communities impacted by large-scale mining in the Atacama Desert, Northern Chile. The project was funded by a British Academy International Partnership and Mobility grant, combining international academic network-building with a small research element. We begin by outlining the context within which this research is situated. Drawing on literature around research with and in indigenous communities, we explore some of the key methodological and epistemological issues that emerge from our analysis of this key incident, recognising that this apparent 'failure' provides a useful lens for reflecting on the dynamics of research with indigenous communities and engaging with debates around decolonising knowledge production, particularly in often fraught contexts of socioenvironmental conflict. We conclude by re-visiting the importance of acknowledging the emotional impacts of failure and emphasise the need for caution in assuming that, as non-indigenous researchers, we can legitimately conduct participatory research with indigenous communities experiencing conflict. The neoliberal era inaugurated a pattern of conflicts with indigenous communities across Latin America, and in this case in the North of Chile, due to the deepening of inequalities inherited from the colonial and postcolonial processes of subjugation of first nation peoples (Bryan, 2012) . Through reforms to land, water and foreign direct investment, mining companies have strengthened their presence in indigenous territories (Prieto, 2016; , generating a massive process of dispossession of commons, and affecting notions of democracy and development (Bebbington et al., 2008) . At the same time, throughout Northern Chile, members of Andean communities were migrating to urban areas, to access education and health services, and the labour market (Gavilán, 2015; González et al., 2014) , driven by Pinochet-era policies that sought to depopulate rural areas and concentrate populations in urban centres where they could be more easily controlled (Gundermann, 2003) . Chile's neoliberal indigenous policy was mainly oriented to rural populations, aimed at promoting local development associated with agriculture and tourism, without a recognition of the Andean dynamics that connect the highlands with coastal areas, and long-established farming and ranching communities with more recent urban political organisations. In 1993, the democratic government passed the Indigenous Law, creating an agency to promote indigenous peoples' development, following a multicultural neoliberal approach that produced particular indigenous subjectivities, the so called indios permitidos (Hale, 2004) . Through this process, hundreds of indigenous communities were formed, and thousands of people started to register as indigenous, transforming themselves into subjects with special rights, particularly in relation to poverty alleviation. Simultaneously, mining, forestry and energy projects were expanding, putting increasing pressure on natural resources, especially water resources within indigenous territories (Budds, 2004) . A growing number of indigenous and environmental conflicts emerged and, at the same time, the number of people self-identifying as indigenous increased, particularly in urban contexts, distant from the actual sites of extraction (Aravena and Á lvarez, 2012; Romero Toledo, 2018) . This process of indigenous identity formation has parallels across much of Andean Latin America and has been theorised as "re-ethification" (Van Kessel, 1980) , "identity articulation" (Clifford, 2013) , "indigenous emergence" (Bengoa, 2000) , "ethnisation" (Restrepo, 2013) , indigeneity (de la Cadena and Starn, 2009) and "indigenisation" (Ballón (2003) , among others. Whilst we do not have the scope here to capture the nuances of these debates, together these concepts capture a deep global process of political and cultural production of indigenous identity in the face of a constellation of social, political, economic and environmental inequalities associated with large-scale resource extraction, unequally distributed in terms of gender and place. The production of indigenous identity provides an opportunity for political visibility, creating a differentiated citizenship, becoming a territorial actor, negotiating development strategies, and claiming political autonomy (Postero, 2005; Garcia, 2005) . Against this backdrop, we aimed to explore the ways in which indigenous Aymara and Quechua women leaders were situated in relation to the mining industry, having observed differences between the discourse of male indigenous leaders confronting mining companies and extractivism, and that of indigenous women, who are becoming more prominent as community leaders. For example, we had noted that indigenous women leaders articulated a particular concern with conserving or reinvigorating Andean village communities as a means of ensuring the physical and cultural reproduction of their communities for future generations, in the face of the threat posed by ongoing extractive activities (Romero Toledo and Jenkins, 2017; Romero Toledo and Sambolin, 2019) . We aimed to explore how we might develop participatory methodologies in partnership with indigenous women community leaders to understand their concerns in relation to large scale resource extraction. We hoped to co-design small participatory pilot projects, to explore how processes of women's indigenous identity formation intersect with the expansion, intensification (and sometimes contraction) of large-scale mining projects. The workshops that we discuss here were the initial phase of negotiating this involvement, and were planned to involve mapping activities and focus group discussions to begin a dialogue with the women. In negotiating research with indigenous communities, the work of decolonial scholars is central in making sense of the epistemological and ethical challenges that this endeavour presents. Perspectives such as epistemicidio (de Sousa Santo, 2010) , extractivismo epistémico (Grosfoguel, 2016) , and epistemic injustice (Fricker, 2017) , enable us to understand how oppressive relationships are based on the naturalisation of the inferiorisation of voices, knowledges and actions of "others". These oppressive relationships intersect with the process of coloniality in the global South, understood as an integral part of the modernity process (see Castro-Gómez and Grosfoguel, 2007) . Therefore, "others" are conceived through the naturalisation of territorial, racial, cultural and epistemic hierarchies, that make possible relationships of dominance. Together, these hierarchies constitute the "dark side" of modernity (Restrepo and Rojas, 2010) , sustained by a pattern of colonial power (Quijano, 2001; Mignolo and Escobar, 2013) . Decolonial scholars such as Walsh (2005) , emphasise that the most appropriate space for knowledge construction is not necessarily the 2 In November 2018, the Chilean police killed Camilo Catrillanca, a young Mapuche man, who lived in a Mapuche community, and who was descended from a local Mapuche leader. This sparked widespread protests by indigenous people against the Chilean state. Following this, in October 2019, a massive social movement against neoliberalism began, sparked by student protests against metro fare increases. These protests brought the country to a standstill, and were highly active until COVID-19 arrived in the country, in early March 2020. academy, and that concepts such as rationality are not necessarily those which underpin expert knowledge. This is captured in argument about a "hybris de punto cero" from which the western Eurocentric constructs universalisms. As Radcliffe (2017, 329) emphasises, 'power relations in the colonial present permeate all forms of knowing about and understanding the world'. To confront this, Walsh (2007) , Mignolo (2007) and Grosfoguel (2007) , among others, have called for a geopolitical and body-political situated construction of knowledge, with the ability to develop other epistemic matrices through different forms and kinds of knowing. To decolonise means, on one hand, questioning from where someone speaks and constructs knowledge, and on the other, emphasises the importance of the situated knowledge of subalternised peoples, and others' genealogies of thought, that have historically been considered as inferior from western perspectives (Grosfoguel, 2012; Radcliffe, 2017) . However, as activist scholars such as Betasamosake Simpson and Silvia Rivera Cusicanqui recognise, this intellectual approach does not in itself ensure the avoidance of "extractivismo epistémico" (Grosfoguel, 2016) , and decolonial theoretical approaches also risk (re)producing language and conceptual infrastructures that themselves contribute to decontextualising and depoliticising social struggles, and that are not rooted in political solidarity with the people from whom they extract knowledge (Rivera Cusicanqui, 2010). Furthermore, even when research is entirely developed in other places, and in other languages, academia demands that ideas and findings should be situated in relation to dominant (overwhelmingly western) perspectives and scholars (Snow, 2018) . Whilst participatory approaches to research are not necessarily decolonial, they do hold out the promise of enabling marginalised communities to actively shape and co-produce research that affects their own communities. Further, we recognise that as non-indigenous scholars situated within mainstream academic institutions, we are not in a position to be able to truly decolonise our research. Nevertheless, a shift in power relations across multiple scales is inherent in adopting a participatory approach: Implicit in participatory approaches to research is the potential for transforming not just the process of knowledge production and the hierarchical relations that exist between university and community, between researchers and researched, but an expansion of the goals of traditional social research. (Zavala, 2013, 59) . In developing strategies that facilitate more inclusive research practices, techniques such as participatory mapping have enabled the construction of better understandings of the different territorial entities of indigenous peoples in different countries, facilitated the recovery of territorial rights (Bjørn et al., 2013) , and strengthened communities' governance of their spaces. Such methods emphasise the importance of the construction of socially relevant and situated knowledge, with an emphasis on knowledge that is developed collectively, cooperatively, dialogically and horizontally (Kemmis and McTaggart, 2005) . As Linda Tuhiwai Smith (1999, 313) recognises, 'When Indigenous peoples become the researchers and not merely the researched, the activity of research is transformed. Questions are framed differently, priorities are ranked differently, problems are defined differently, and people participate on different terms.' We hoped to draw on some of these approacheswhich resonate with our own commitment to supporting and advancing the social, political, cultural and territorial rights of indigenous communities in Chileas we co-developed the focus of our research. Decolonising knowledge production requires what Hunt aptly describes as the 'subtlety of dancing between [the academic and indigenous] worlds' (Hunt, 2014: 28 cited in Basile et al., 2018) , and here we find decoloniality useful in reflecting on our emotional responses to the ethnographic refusal we faced, providing an important lens for thinking through the challenges of navigating this 'dance' in practice, and the challenges that we faced in opening up spaces to negotiate participatory research with indigenous women. The workshop that we discuss here, was due to take place in November 2017, in Iquique, Northern Chile. It was envisaged as a first step in bringing together indigenous women leaders to co-design and carry out some pilot projects exploring indigenous women's experiences of living with large-scale resource extraction. We hoped to develop the focus of the research in partnership with women leaders, and in accordance with their priorities. As the relatively small amount of funding we had secured was predominantly focused on developing academic networks, we had not negotiated the involvement of affected communities from the outset. However, we recognise that this initial disconnect has been fundamental in limiting our scope for working collaboratively as the project has developed. 3 The workshops in Iquique (one with Quechua women and one with Aymara women) were envisaged as the beginning of a collaborative process. 4 Initial ethical approval for the workshops had been sought and granted by Jenkins' University ethics committee, with the expectation of a further more detailed application were specific pilot projects developed. This two-step approach reflects the dilemma Snow (2018, 5) articulates -'a catch-22, where institutional ethics cannot be granted without local approval, but local ethics cannot be obtained without institutional approval'. We had previously, under the auspices of Romero's larger research project, conducted in-depth interviews with women (and men) indigenous community leaders, and Romero had researched extensively in the region between 2014 and 2017, developing reciprocal relationships with key local actors. Videla had also undertaken a successful preparatory visit to Iquique in October 2017, to nurture networks with women community leaders and to recruit workshop participants. Videla's positionality as a mestiza woman in an indigenous self-identification process, was also important in building trusting relationships in this context. Participants were originally contacted via known gatekeepers to take part in an interview (part of Romero's larger research project) and were subsequently invited by Videla to participate in our planned workshop. Printed information about the workshop, translated into Spanish and complete with institutional logos, was presented and explained to those who were invited. There was constant communication between invitees and the research team by email and messaging app to respond to queries in the days leading up to the workshop. We were expecting around 12-15 participants at each workshop and indeed, the evening before the Aymara women's workshop, Videla had received confirmation messages and enquiries as to whether additional participants could come along (which we had said they could) and if travel expenses would be reimbursed (yes). We were therefore expecting a reasonable turn-out, and this brings us back to the point at which we began our discussionan empty room and a woman leader explaining that our workshop had been boycotted. So, what lay behind this apparently collective decision not to participate in the workshop (and by extension the development of the research)? Below we begin by sketching out our interpretations of the day's events, drawing on reflections written by each of us at the time. We then explore some of the ways in which we have made sense of this situation, particularly drawing on literature on ethnographical refusal. We critically analyse issues around the strategic use of indigenous identity, and explore the apparent disconnect between academic and indigenous communities, examining the ways that the scenario we faced speaks to dilemmas around decolonising research and underlines the challenges of conducting participatory research with indigenous communities as non-indigenous researchers. Finally, we reflect on the emotional responses provoked in each of us, and argue for the importance of embracing such fieldwork failures and their emotional outcomes as learning opportunities to inform our approaches to future research. The one (non) participantwho was not one of our known interlocutorsexplained that there had been a 'discusión fuerte en el correo' ['an intense disagreement via email'] the previous evening, where apparently doubts had been raised as to whether the women should participate in our workshop. Whilst it was hard to glean the exact scenario, it appeared that someone with a degree of influence had decided that none of the women should participate. 5 As she explained this, the woman produced a document to justify or contextualise this nonparticipation, the UN Declaration on the Rights of Indigenous Peoples (UNDRIP), in particular article 31 on the protection of indigenous culture and knowledge, with the relevant section already highlighted (see Fig.1 ). Thus, she framed the 'boycott' very strongly in terms of a deliberate collective action driven by the need to protect their cultural heritage. She went on to explain that they were highly suspicious of academics, and that it was considered likely that we would either sell the knowledge that we gained to the mining companies, or use it to profit in some way ourselves, or our universities would. This reflects very legitimate and ongoing concerns around intellectual and cultural property rights in relation to research with indigenous peoples, rooted in a long history of inappropriate research conducted on indigenous communities across the global North and South (Castleden et al., 2012; Cochran et al., 2008) . The continual undermining of the UNDRIP in contexts of resource extraction, as discussed by Radcliffe (2019) , makes the use of this document particularly potent in this context. Considering the ways in which indigeneity is, as Radcliffe (2017, 221) aptly describes, 'produced by particular people and institutions at particular times in power-drenched ways', we recognise that the act of visiting us and producing a physical copy of UNDRIP to underline the rights that they have been granted, as Aymara community members, by an international legal instrument, was a powerful way of demonstrating and enacting agency in the context of often circumscribed opportunities for them to do so. "This was not confrontational, she was pleasant, polite, and apologetic to some extent. But she was clear that the reason for no one coming was because they needed to protect their heritage, and that this was threatened by us as outsiders. […] She cited a history of researchers, consultants etc, coming and extracting knowledge from indigenous communities and using it for their own gain and the indigenous people seeing nothing from it. Ironic given that what we were trying to do was quite the opposite, but start a dialogue and a more participatory way of working." (Katy Jenkins' field notes, 24 th November 2017, Iquique) In exercising their right to decide not to take part (see also Velásquez, 2017 on indigenous women and refusals), the indigenous women demonstrate the ultimate power of participants in the research relationship, the power not to participate. The subaltern doesn't want to speak (Spivak, 1988 ) -at least not to us, or with us. As scholars committed to developing equitable research practices, this refusal is one that we must accept, and recognise as a fundamentally political act, despite the difficulties it may cause us. This brings us to the question at the heart of debates around decolonising knowledge production: is it appropriate that we as white or mestizo/a, non-indigenous, middle class scholars, even try to do research with indigenous Aymara and Quechua Chilean women, even though two of us are also Chilean, two of us are womenone of whom is herself in a process of recognising her own Mapuche heritage and identity -and when the research is premised on being as participatory and horizontal as possible? 6 : Angélica: I feel that, through this action, the Aymara women were underlining their stance of confronting organisations that are not part of their culture and ethnic identity. This makes me reflect on our positions as researchers: Do we deserve their confidence, if we are foreign to their culture and everyday reality? Could our research be considered as a misappropriation of their knowledge, or an interference with their social, cultural, political and economic networks? As captured in Videla's reflections, this experience has caused us to reflect on our personal and professional motivations and structural locations, as well as leading us to consider how "… even though the researcher aims at establishing honest, participative and open communication with participants by sharing her background and the research purpose, this goal is not always well understood." (Caretta, 2015, 501) . Were we misunderstood, or were we fully understood but refused and rejected? We can never entirely know, and perhaps it does not matter. Either way, the scenario raises some important theoretical, methodological and empirical issues, and we explore these in the remainder of this paper. The recent significant rise in indigenous self-identification is critical in today's Chile. According to Chilean Ministry of Social Development, Chile's indigenous population increased from 635,376 in 1996 (4.5% of the national population), to 1,694,870 in 2017 (9.5% of the national population). However, in the 2017 Chilean census, 2,185,792 people identify as indigenous (12.8% of the population). Understanding the neoliberal context of Chile is crucial in making sense of this scenario, and in the literature it is possible to identify three interconnected elements that contribute to this phenomenon: 1) the global struggle for indigenous recognition and rights, led by institutions such as the UN and ILO; 2) the increased agency of indigenous groups at the national and local levels, and their growing ability to articulate their demands in the context of the transition to democracy (Bengoa, 2009; Clifford, 2013) ; 3) the neoliberal multicultural policies promoted by the Chilean state and the production of indigenous subjects who are then able to apply for local development projects that create particular territories, discourses and practices related to the reproduction of neoliberalism and post-dictatorship democracy (Postero, 2007; Merlan et al., 2009) . Finally, in our own experience, it is also important to recognise the way in which large extractives companies create particular mechanisms to deal with indigenous communities, focused on indigenous recognition and social responsibility, whilst simultaneously engaging in an ongoing process of negotiation, not only for critical resources such as water and land, but also about how territories are represented, and the kinds of economic activities that can be developed (Hale, 2004; Romero-Toledo and Sambolín, 2019) . In Chile, the presence of extractive companies in very-local 5 It is not clear exactly who this discussion was between, and who ultimately decided no one should participate. Whether it was a decision made collectively by the women leaders themselves, or whether the discussion involved others, obviously changes the way in which this 'boycott' might be understood, but this information was not available to us. We cannot know whether perhaps male dirigentes felt threatened or uncomfortable with women dirigentes participating in our project, and chose to veto their involvement, or whether this was a decision made entirely by the women themselveswhich we have to assume it was in the absence of any other information. territories has brought several challenges to different actors. Romero Toledo (2018) uses the term "dynamics of negotiation/resistance" to explain how the state, mining companies, indigenous and non-indigenous communities, NGOs and academics, among others, co-produce territories and identities in highly creative and strategic ways. Mining companies have created special departments to deal with indigenous communities, with an emphasis on mitigating risk via economic compensationthrough projects focused on social issues (e.g. grants for digital connection, community centre renovations, support for folkloric bands) and the direct transfer of money (Romero Toledo and Sambolin, 2019), also known as "social investment". In this context, indigenous communities' opposition to mining, and the ensuing process of negotiation with mining companies, provides an opportunity to overcome poverty, to have access to services and to modernise old villages and roads, in a scenario characterised by a lack of state support and investment. However, this imposition of market logic and the establishment of corporate-community relationships based on the exchange of money and/or provision of direct benefits, has had a deep impact on the social interactions that occur in this territory, particularly in relation to notions of indigeneity and the potential economic benefits attached to this. (Radcliffe, 2020; Vel, 2014) . The tensions that arose in developing our research, illustrate that these changes have also fundamentally shaped the possibilities and challenges for conducting research with and in indigenous communities in the areas of influence of the mining companies, particularly in terms of the expectations of indigenous communities and what we might consider as their strategic use of indigenous identity in ruling in or out particular interactions: "She claimed that the location of the meeting was not the right place for the Aymara people, and even the food -biscuits, coffee, tea and orange juice -was not right. She said that we should provide traditional Aymara meals, such as "picante de conejo" or "calapurca". She rejected our workshop, and, to some degree, she questioned our experiences of fieldwork with Aymara people (Hugo Romero Toledo's field notes, 24 th November 2017, Iquique). Despite their experience as interlocutors working across indigenous and non-indigenous spaces, the woman lawyer drew on quite essentialist constructions of indigeneity, to exert agency over the types of interaction that are permissible, and to symbolically frame us as 'outsiders', undoing the access we thought we had secured and undermining our legitimacy as researchers. This highlights the ways in which indigenous identities may be deployed strategically to grant or deny access to particular spaces, emphasising the agency of indigenous actors, organisations and interlocutors, and disrupting established academic and NGO practices and assumptions around where and how we negotiate and structure research relationships (see also Snow, 2018) . When the Aymara woman came to the workshop, she was at the same time: Aymara, woman, lawyer, leader, member of a community, member of an Aymara women's group, rural, urban and semi-urban, ancestral, traditional and modern. As non-indigenous academics, are we really adequately equipped to understand the multiple and overlapping subjectivities that she represents? We certainly cannot take for granted the indigenous interests she embodies as she moves across contexts and spaces. This scenario causes us to reflect upon broader disconnects between academia and indigenous actors, in the current neoliberal scenario, particularly in relation to agendas, timeframes and ways of working, that structure the ways in which research opportunities may be understood. For example, we should recognise that it may be that the research we are promoting is not an attractive proposition, perhaps because we are not offering something concrete to overcome poverty, or to understand the "hard" impacts of extractivism that could directly help in the communities' struggles against the mining companies. Communities want knowledge to use in their fight for the protection of their environment and benefits for their members -studies on water quality, air pollution, and archaeological sites. Neoliberalism has generated a deep asymmetry of power at the local scale: while the companies can compensate and invest in order to have durable and long term relationships with communities (to facilitate their extraction activities), academia in general, and social science in particular, wants to understand processes with limited resources and time, and is not well equipped to offer immediate solutions. While mining companies are permanent residents of the territory, a highly visible feature of the landscape, academics are only visitors, able to dedicate short periods of time to establishing relationships, and to offer much less tangible and attractive support for communities. In this context, knowledge and Fig. 1 . UN Declaration on the Rights of Indigenous Peoples, Article 31.1 "Indigenous peoples have the right to maintain, control, protect and develop their cultural heritage, traditional knowledge and traditional cultural expressions, as well as the manifestations of their sciences, technologies and cultures, including human and genetic resources, seeds, medicines, knowledge of the properties of fauna and flora, oral traditions, literatures, designs, sports and traditional games and visual and performing arts. They also have the right to maintain, control, protect and develop their intellectual property over such cultural heritage, traditional knowledge, and traditional cultural expressions." Photo: Videla, 2017. knowledge production are commoditised, and with limited time and energies, we should not be surprised that indigenous actors make strategic choices about the most useful and profitable activities to pursue. Hugo: How many actors are doing similar activities with indigenous communities at the same time and in the same territory? What kind of benefits do they feel that they receive each time that they participate in these activities? Today's indigenous Chilean communities are clearly professionalised in the heat of battle with the mining companies. They decide with who, when, where and how they want to speak, and what they want to say. We had hoped that our workshop would be an opportunity to begin a dialogue around future possibilities for co-productive ways of working, to develop opportunities to support the capacity-building of indigenous women. Nonetheless, the questions asked by potential participants underline the extent to which 'participation' in this context is conditioned by a particular set of neoliberal subjectivities. When coordinating with leaders to secure their participation, questions included: Is it a training session? Will there be a certificate of participation? Are travel and food costs covered? All these queries emanate from a particular way of doing things in Chile, which conditions the relationships that indigenous peoples establish with institutions and projects, mediated by the interests and benefits that emanate from assistentialist governments in a neoliberal context. We also recognise that the construction of knowledge by/with academia may be perceived as less important by communities, because they now have relationships with other actors that also construct knowledge, such as NGOs and consultants. Academia does not have a monopoly on knowledge production in this fraught and potentially lucrative context, challenges Kirsch (2002) also recognises in relation to navigating knowledge production and community-mining company relations in Papua New Guinea. A further element to this dynamic is the often highly individualised way in which socio-environmental conflicts are negotiated in the context of neoliberal Chile. Interviews with community leaders in Northern Chile (Romero Toledo and Jenkins, 2017), highlighted the extent to which communities are not necessarily united or organised in confronting mining companies, with relatively little accountability of community leaders. Community 'representatives' negotiating with the mine may be self-nominated and do not necessarily represent the interests of the whole community (particularly given the divisive nature of such conflicts); a gendered dynamic is also evident here. The women we had invited to participate in our workshops were leaders and active community members but did not necessarily come as representatives of particular organisations. It was clear from our workshop with Quechua women the previous day that each had their own personal agenda and did not particularly coordinate with other communities also opposing and/or negotiating with mining companies; it is obviously to the benefit of mining companies to encourage this competitive approach. We recognise that such an individualised context is not conducive to the development of participatory ways of working -that entail an input of time, ongoing collaboration and shared goals, and that come with uncertain and often unforeseen outcomes. In this context, where access to knowledge translates into competitive advantage and/or directly into financial compensation, the incentive and motivation to collaborate and to share knowledge is severely diminished. Participatory research is often held up as the pinnacle of 'good' development research practice -emphasising co-production of research, the collection of data that is more representative of the perspectives of the community being researched, and the production of research that is useful to those participants as well as to the academics who produce it (Wynne-Jones et al., 2015; Kindon et al., 2007) . However, embedded within this approach is an assumption that such research will be inherently more acceptable to research participants, and that they will want, and be able, to actively participate. Our experience resonates with those who suggest that a more cautious approach is necessary, particularly when negotiating collaborations with indigenous communities, and particularly indigenous women, whose historical exclusion from much scholarly research makes this an especially fraught issue (Basile et al., 2018) . Whilst we cannot be 100% certain (given their non-participation), we hypothesise that had we planned to carry out individual interviews with women leaders, we would have met little or no opposition and been able to successfully undertake our data collection. Indeed, on several previous occasions we had all, either jointly or separately, done just that, and encountered no significant obstacles. Thus, we identify the collective nature of participatory methods as a possible barrier to securing participation. The demands that this approach places on participants to work together, share ideas and develop collective responses and projects may also be threatening (or at least unappealing) to potential participants. In a context where resources (financial, material, social, physical) are scarce and contested, and communities are often pitted against each other to compete for limited funds and compensation, community leaders cannot necessarily be expected or relied on to act in solidarity with one another (or even with their own communities). Allegiances and power relations are tenuous and shifting, and thus conducting meaningful participatory research, which relies on a sense of collective identity and shared objectives, may be ambitious or even (as in this case) unachievable, compared to the logistical ease of a one-off interview. We suggest that in setting out to develop a collective and co-produced set of ideas, we left ourselves more vulnerable to the rejection of our proposal, which required sustained 'buy in' from multiple actors across a shared agenda. We therefore surmise that it was proposing to conduct specifically participatory research that led us to encounter difficulties with nonparticipation, particularly given the proposed engagement of participants over a longer period, requiring an ongoing and collective commitment. The inherently more in-depth nature of participatory research, supposedly its strength, may be perceived by potential participants as more 'risky' and as carrying a greater danger of participants 'exposing' themselves, their beliefs and their lives (see also Zahara, 2016) . The reaction from our Aymara contact and her use of UNDRIP, suggests that cultural heritage was perceived to be 'at risk' of appropriation should women leaders have taken part. 7 In contrast, individual interviews are a more known quantity, where the interviewee can feel 'in control' of what information is shared and what is withheld. Indeed, we have observed in various contexts that activists and community leaders are often familiar with giving interviews and have a well-practised narrative to deploy in this situation, enabling the interviewee to limit the information in the public domain. Participatory research disrupts these comfortable 'known' research patterns, and may be perceived as less desirable in a context, such as in relation to resource conflicts, where control (of information, of land, of knowledge) is crucial. As Tuck and Yang (2014, 225) reflect, 'Refusal, and stances of refusal in research, are attempts to place limits on conquest and the colonization of knowledge by marking what is off limits, what is not up for grabs or discussion, what is sacred, and what can't be known'. Whilst we might interpret this situation as our collective 'failure', and with hindsight we recognise that we should have been more cautious in our ambitions and initial engagements, we also emphasise the risks inherent in these ways of working may make them less feasible to use in situations of intense socio-environmental conflict. Undertaking qualitative fieldwork, particularly in contexts of 7 Or, at the very least, the UNDRIP provided a powerful and effective way of declining to participate. conflict and insecurity, is always a challenging process that requires changes of plan, flexibility and reflexivity as researchers adapt to unexpected situations. It is important to recognise that such processes are 'normal'; despite extensive preparation for fieldwork, things go wrong, do not work as intended or require modification. Negotiating research with indigenous participants adds an additional layer of complexity, causing us to re-examine our academic praxis and question our legitimacy as researchers. It is vital to recognise the emotions embedded in these experiences as a valid and productive part of the research process (see special issue by Baillie Smith and Jenkins, 2012) . In this case, our 'failure' provoked a range of emotional responses stemming from our differently positioned subjectivities: Katy: Even in the process of actively trying to establish equitable relationships to enable the co-production of knowledge that will be useful to participants, we were found to be severely lacking. And not even us personally, but we stand in for decades of colonization and academic extraction of knowledge … It is painful to be perceived in this way, and forces me to consider how I could have made more explicit my commitment to working in solidarity with women activists, and to the co-production of knowledge. Hugo: I felt like an extractive academic. I always seek the support of communities that are affected by extractivism to highlight their everyday acts of resistance, and how they are reconstructing their ancestral territories. Until this experience, I never felt that I was part of the problem and I was devastated. I started to question how we go to fieldwork, how we "recruit" people for our research, how we collect data, and then, because we are faced a whole range of difficulties, how we disappear. Snow's (2018) work on being a Settler Ally is especially useful here in enabling us to reflect on our positionalities, and how we might enact these differently in negotiating future researchrecognising that, when working with indigenous participants, ongoing critical reflection is 'vital to renegotiating and repairing our inevitable shortcomings and errors' (Snow, 2018, 9) . In the moment of experiencing our potential participants' refusal to collaborate, we felt judged to be on a par with the extractive companies whose behaviour we sought to critique, an unsettling experience that has stayed with us beyond the research 'moment' -an 'affective residue' (Harrowell et al., 2018 ) -as our reflections above attest to. Such anxieties and uncertainties are seldom reflected in final publications, but we argue that it is important to recognise this swirl of emotions that surrounds fieldwork failure and its aftermath. In the field, often isolated from institutional and collegial sources of support, such scenarios can be stressful and anxiety-inducing (Klocker, 2015) . In our case, as three colleagues together, we were able to collectively reflect on this failure, and to support each other in understanding the reasons behind it and using it as a learning process. This enabled us, to some extent, not to succumb to the panic of having 'failed'. We have learnt a great deal from this so-called failure, both in terms of gaining a better understanding of the empirical context, and in terms of methodological insight, causing us to re-examine how we are positioned in relation to debates on decolonising knowledge and how/whether as non-indigenous scholars we can/should pursue research with indigenous communities at allas Zavala (2013, 60) cogently observes, 'How do these modes of participation ensure that the interests of historically marginalised peoples are represented, especially when they are carried out within colonizing spaces, such as universities and public school bureaucracies?' Angélica: This is not necessarily an act of personal rejection, but about what each of us represent. It could be because someone is from a colonial country, or a large university, the rejection is of those that have subalternised them -western society. They didn't reject me, based on my physical indigenous characteristics as a mixed-race woman, but nevertheless caused me to question when is the right time to conduct research with indigenous people? What kind of skills do researchers need to conduct research in this context? Should we wait for the communities call us? As Videla emphasises, timing, and timeliness play a role here too; our attempted collaboration came at a time of significant social and political unrest in relation to indigenous communities in Chile, with nonindigenous researchers being regarded with particular suspicionframed either as private consultants exploiting indigenous communities for commercial profit, or as undertaking 'epistemic extractivism' (Grosfoguel, 2016) , as our indigenous interlocutor's explanation attests to. Throughout this paper we analyse failure as presenting an important opportunity to improve our research, in terms of critically reflecting on social context, theory, and methodology. In thinking through failure and our emotional responses to it, we aim to counter a tendency for less successful parts of the research process to be obscured or hidden. In the pressured and highly individualised environment of the neoliberal academy, failure is a risky business, with potential implications for promotions, publications and future bidding (Berg et al., 2016) . However, following Harrowell et al. (2018) , we argue that fails should be expected as a normal and often productive part of undertaking fieldwork and of academia more broadly. We suggest that such failures are characteristic of ongoing attempts to decolonise research, a messy, partial and inevitably fraught process in which we need to be prepared to take risks and to deal with failure, rejection and refusal (Radcliffe, 2017) , especially in relation to historically vulnerable and marginalised groups who may be cautious of establishing relationships with actors who are foreign to the territory they are defending. In this sense, our research cannot escape the way in which coloniality produces knowledge about 'others', assuming that academic procedures designed far from these territories are appropriate for working with different people and cultures -not only in physical terms, but also symbolically, such as official documents elaborated in foreign languages and based on Western ethical protocols, that frame why, who and how 'others' can participate. In this colonial exercise, we assume that some research techniques are more positive than others for working with marginalised people -such as participatory techniques to co-produce knowledge -and also that some frameworks are more adequate to work with 'others', such as mainstream decolonial perspectives. However, despite our personal positionalities, sympathetic theories, and participatory methods, there remains a great distance between academia and the communities. Our research also cannot escape the dynamics of negotiation/conflict that pervade the relationship between the State, large investment companies, and indigenous and non-indigenous communities in territories where extractivism takes place. We will always be outsiders, and probably the least important actor in scenarios of environmental and territorial conflicts. This leads us to concur with Coombes et al.'s (2014) recognition that, 'collaborations and ethics in Indigenous Geographies unsettle our journey as researchers' (p.851), often in profound, unanticipated, emotionally intense, and messy (Klocker, 2015) ways, pushing us to re-visit how we can make our work relevant to indigenous women and how to facilitate collaborative and equitable ways of doing so, beyond participatory approaches. Indeed, this requires an ongoing consideration of whether it is even appropriate for us to be the ones developing such research (Radcliffe, 1994) , especially in light of the academic, professional development and agency that indigenous groups in Chile now have (Yopo, 2012) . As we reflect on the learning that emerges from this somewhat bruising experience, we emphasise the need for caution around the assumed value of participatory methods, especially in scenarios with significant socio-environmental conflict. We argue that it is essential to recognise that this sort of research opens up both researchers and research participants to greater riskrisk of exposure and risk of appropriation of knowledge (participants); risk of mis-understanding, failure and rejection (researchers). As researchers, whether from the global South or North, we need to be cogniscent of the multiple dynamics at work when we negotiate access and participant engagement, and remember that this is an ongoing process whereby consent may be withdrawn at any stage. This is even more pertinent when potential participants are from indigenous communities, who have historically had little agency or decision-making power, and for whom processes of so-called 'consultation' have often been hollow (Radcliffe, 2020) . Finally, we emphasise the need to problematise and continually revisit the 'benefits' that our research may provide to indigenous peoples and to reflect on the emotional impact that 'refusal' may have on both researchers and the researched. Our own emotional responses will certainly shape our future research interactions with (potential) indigenous participants. Although painful to us as researchers, we recognise that such encounters are legitimate and important 'refusals' that underline the capacity of indigenous communities to actively decide and to shape the scope of research with non-indigenous scholars. We cannot take for granted that co-production of knowledge with academics is either useful or desirable to indigenous communities and activists, who may choose to prioritise more obviously, or immediately, productive encounters with 'others' (Le Bonniec, 2015) . This work was supported by a British Academy International Partnership and Mobility grant [grant number PM150114]. None. Juventud, migración y discriminación en el Chile contemporáneo Indigenización versus faccionalismo. Programa Andino de derechos humanos -universidad andina simón bolívar Co-construction of a data collection tool: a case study with atikamekw women Mining and social movements: struggles over livelihood and rural territorial development in the andes La emergencia indígena en América Latina ¿Una segunda etapa de la Emergencia Indígena en América Latina? Producing anxiety in the neoliberal university Territorialidad, mapeo participativo y política sobre los recursos naturales: la experiencia de América Latina Rethinking territory: social justice and neoliberalism in Latin America's territorial turn Power, nature and neoliberalism: the political ecology of water in Chile Situated knowledge in cross-cultural, cross-language research: a collaborative reflexive analysis of researcher, assistant and participant subjectivities i spent the first year drinking tea": exploring Canadian university researchers' perspectives on community-based participatory research involving Indigenous peoples El giro decolonial: Reflexiones para una diversidad epistémica más allá del capitalismo global El giro decolonial: Reflexiones para una diversidad epistémica más allá del capitalismo global Returns: Becoming Indigenous in the Twenty-First Century Indigenous ways of knowing: implications for participatory research and community Indigenous geographies III: methodological innovation and the unsettling of participatory research Indigeneity: problematics, experiences and agendas in the new millenium Epistemologías del sur. Siglo XXI Epistemic Injustice: Power and the Ethics of Knowing Making Indigenous Citizens: Identities, Education, and Multicultural Development in Peru Ser indio y ciudadano en el norte chileno: etnicidad y estatus de pertenencia a la comunidad nacional y política Comunidad indígena y construcción histórica del espacio entre los Aymara del Norte de Chile The epistemic decolonial turn: beyond political-economy paradigms Decolonizing western uni-versalisms: decolonial pluri-versalism from Aimé Césaire to the Zapatistas Del «extravismo económico» al «extravismo epistémico» y al «extravismo ontológico»: una forma destructiva de conocer, ser y estar en el mundo Sociedades indígenas, municipio y etnicidad: La transformación de los espacios políticos locales andinos en Chile Rethinking indigenous politics in the era of the "indio permitido. NACLA Rep Making space for failure in geographic research Anthropology and advocacy: a case study of the campaign against the ok tedi mine 2007. Participatory Action Research Approaches and Methods: Connecting People, Participation and Place. Routledge Participatory action research: the distress of (not) making a difference Un antropólogo inocente en búsqueda de exotismo en las cortes de justicia: reflexiones acerca de las posibilidades de llevar una etnografía del campo jurídico en contexto intercultural en el sur de Chile Indigeneity: global and local El giro decolonial: Reflexiones para una diversidad epistémica más allá del capitalismo global Globalization and the Decolonial Option Now We Are Citizens: Indigenous Politics in Postmulticultural Bolivia Transando el agua, produciendo territorios e identidades indígenas: el modelo de aguas chileno y los atacameños de Calama Colonialidad del poder, globalización y democracia. Utopías, nuestra bandera: revista de debate político 188 Representing) post-colonial women: authority, difference and feminisms Geography and indigeneity I: indigenity, coloniality and knowledge Geography and indigeneity III: Co-articulation of colonialism and capitalism in indigeneity's economies Etnización de la negridad: la invención de las 'comunidades negras' como grupo étnico en Colombia Inflexión Decolonial: Fuentes, Conceptos Y Cuestionamientos Ch'ixinakax utxiwa: una reflexión sobre prácticas y discursos descolonizadores Etnícídades, etno-territorios y conflictos mineros: aportes para una geografía humana de los aymaras en Indigenous communities' relationships with the mining industry in the Atacama Desert: exploring processes of (re)ethnification and strategic indigeneity Indigeneidad y territorio: los aymaras y quechuas en el Norte de Chile. Scripta Nova. Revista Electrónica de Geografía y Ciencias Sociales 23 Explorando conflictos entre comunidades indígenas y la industria minera en Chile: las transformaciones socioambientales de la región de Tarapacá y el caso de Lagunillas What does being a settler ally in research mean? A graduate student's experience learning from and working within indigenous research paradigms Marxism and the Interpretation of Culture R-words: refusing research Holocausto al progreso: Los aymaras de Tarapacá. Centro de Estudios y Documentación Latinoamericana Chile's violations of the Diaguita Indigenous Peoples' environmental and human rights and the need to enforce obligations to obtain free, prior, and informed consent Enacting refusals: mestiza women's anti-mining activism in Andean Ecuador Interculturalidad, conocimientos y decolonialidad El giro decolonial: Reflexiones para una diversidad epistémica más allá del capitalismo global Practising participatory geographies: potentials, problems and politics Políticas sociales y pueblos indígenas en Chile: aproximación crítica desde la noción de agencia Ethnographic refusal: a how to guide. Discard Stud What do we mean by decolonizing research strategies? Lessons from decolonizing, indigenous research projects in New Zealand and Latin America This paper was originally presented at the RGS-IBG International Conference 2018, as part of a panel on 'Reclaiming failure in geography: academic honesty in a neoliberal world'. Thanks are due to the panel convenors and participants in this session, and to two anonymouos reviewers, for their constructive feedback that helped shape the final version of this paper.