Uncomfortable Knowledge: Toward a Pedagogy of Reflexivity

Reflexivity is a hallmark of good ethnography and many consider it a defining characteristic of anthropology. It is thus surprising that anthropologists have not paid more attention to how we teach students to be reflexive. Many of us learn reflexivity by making mistakes in the field, yet discussions of anthropological faux pas and their potential contributions to reflexive learning are typically limited to informal settings and occluded or heavily curated within our research outputs. In this article we employ analytic tools from the theory of sociocultural viability, in particular the notions of clumsiness , elegance , and uncomfortable knowledge , to contribute to developing a more explicit pedagogy of reflexivity. Since reading ethnographies plays a major role in how we teach anthropology, we argue that anthropologists should do more in their publications to highlight how awkward moments can deepen reflexivity. To advance this agenda, we provide cases of uncomfortable knowledge drawn from our own field experiences, highlighting how the social, emotional and embodied awkwardness of each situation contributed to acquiring reflexive insights. This article is thus a call to initiate prospective researchers earlier into the messy backstage of anthropological research, including by clarifying how the embodied and affective aspects of our interactions offer potential for deepening reflexive knowledge. In the hopes of facilitating the development of our pedagogies of reflexivity, we conclude the text with four recommendations that we feel will encourage reflexive learning from awkward fieldwork encounters. I am thankful for the moments that we spent with the pastor, and for the wisdom he has shared with us. May we take this wisdom with us and do good with it. Let us remain safe and healthy after this meeting. Thank you and goodbye.


Introduction
Most if not all anthropologists would agree that reflexivity is a keystone of anthropology (e.g., Bourke, 2021;Coleman, 2011;Davies, 2007, pp.254-272;Dellenborg, 2013;Engelke, 2017, pp.10-15;Spencer, 2011). A common definition of reflexivity is "the process of critical self-reflection on one's biases, theoretical predispositions, preferences, and so forth" (Schwandt, 2001, p.224); some anthropologists have however emphasized that reflexive praxis is emotional and embodied as well as cognitive (e.g., Cain & Scrivner, 2021;Spencer, 2011). Davies argues that reflexivity "must be developed and incorporated into research from the initial stage of selecting research topics through the interactions with others in the field to the final analytical and compositional processes," because it is "the only means" of "coming to know, however imperfectly, other aspects of social reality" (2007, p.254). Postmodernist, feminist, and indigenous anthropologists often claim a special attentiveness to reflexivity (e.g. Bourke, 2021;Spencer, 2011; see also Davies, 2007, pp.265-267), but reflexivity has also been described as inherent to cultural relativism and the ethnographic project (Engelke, 2017, pp.1-24;cf. Malinowski, 2005, pp.1-19). Reflexivity in scholarly texts helps the research community assess the conditions under which knowledge has been produced, which is particularly important given that an anthropological project is largely a unique methods assemblage (Law, 2004) generating a particular perspective and knowledge (Haraway, 1991). Ethically, reflexivity facilitates scholars' ability to consider the well-being or "co-presence" (Chua, 2015) of those they encounter during research. Finally, as a methodological tool, continual reflexive reassessment of research practice allows identification of blind spots in research programmes (Cummins and Brannon, 2021).
Given reflexivity's centrality to anthropology, it is surprising that anthropologists have not written more about how to teach it. Davies writes that we learn reflexivity from reading theory and others' texts, and from talking with research participants (2007, pp.254-272). Bourke, who emphasises the need for reflexivity in ethical anthropological practice and notes that it is often not taught at an institutional level, offers similar guidance: anthropologists should learn reflexivity from reading others' publications and thinking about their own positionality (2021, pp.30-33). Spencer, who writes that reflexivity is "somewhat black-boxed" in the practice and teaching of anthropology, argues by contrast that "the emotional has been continuously undermined by powerful scientific hegemonies" in anthropology and other social sciences (2011, p.70, 84), a position that resonates with Okely's claim that the cerebral has been privileged in representations of fieldwork practice (2007, p.65). Spencer advocates for explicit attention to the emotional aspects of reflexivity, including through "safe" discussions during research supervision, whereas Okely's contribution is a set of stories through which anthropologists reflect on their bodily experiences as sources of knowledge.
While few anthropologists explicitly connect emotional and embodied learning in the field to reflexive knowledge, many have argued for greater attention to embodied praxis (e.g., Dellenborg, 2013;Deltsou, 2020;Johansson & Montesi, 2021;McGranahan, 2014;Okely, 2007;Strauss, 2017). For example, McGranahan emphasizes engaging the body to help students develop an ethnographic sensibility, offering examples of how she uses discomfort "to bring some of the experiential, embodied, and empathetic aspects of ethnographic research into the classroom" (McGranahan, 2014, p.24; see also Deltsou, 2020). Yet as Spencer points out, anthropology students usually are "advised to read and discuss ethnographic monographs…mainly from the viewpoint of concepts, theories and the interpretation of data," to the detriment of "the messy sides of lived fieldwork experience, encounters and relationships" (2011, p.71; see also Johansson & Montesi, 2021;Okely 2007). Indeed, most ethnographic publications eschew detailed discussions of the anthropologist's clumsiness, awkwardness, and mistakes, not least because of concerns about professionalism and academic rigour (Spencer, 2011). A minority of anthropologists have been willing to describe in detail their discomfort, confusion, and awkwardness in the field in their academic publications, although such stories are a staple of anthropological memoirs or fiction (e.g. Barley, 1986aBarley, , 1986bBarley, , 1988Bohannon, 1954Bohannon, , 1966Kulick, 2019). These trends render a text-based pedagogy of reflexivity more difficult.
In this article, we collect uncomfortable learning incidents in the field into an explicit repertoire for pedagogical reasons. While this involves attention to the anthropologist's presence in the field, it does not imply a return to older modes of cross-cultural representation in which the anthropologist was the 'natural' protagonist with their ethnographic authoritative presence and gaze at the centre of narratives and analyses. This previously dominant way of writing ethnography has been questioned for long, not least through the debates instigated by James Clifford's and George Marcus influential Writing Culture: The Poetics and Politics of Ethnography (1986). Nor is the reflexivity we advocate for in this article of an autoethnographic kind, and it does not involve the kind of anthropological narcissistic and self-absorbing navel-gazing that was characteristic of some the debates following the publication of Writing Culture. In such accounts anthropological subjectivity and corporality turned into the primary objects of study and anthropological texts became increasingly difficult to differentiate from psychological or philosophical texts, or, for that matter, poetry.
What we argue for, then, is not the insertion of 'more anthropologist' in anthropological texts, but the reflexive and analytical harnessing of moments when things go wrong in the field, moments when people react to our clumsiness and we find ourselves overwhelmed with awkward and uncomfortable sensations, which subsequently can be approached analytically in order to generate a form of out of body experience in which the inappropriateness or awkwardness of our deeds, gestures, or words are understood in specific concrete contexts. Focusing on moments of interaction and relationality, we can thus come close to see ourselves as others see us, which in turn engenders learning, not primarily of ourselves, but of specific relational social settings.
In this article, we utilise theoretical language from the Mary Douglas derived theory of sociocultural viability 1 , which is a tool for analysing the interactions of different worldviews at any social scale (cf. 6 & Richards 2017), to conceptualise: 1) why anthropologists avoid discussing awkward field encounters and 2) how such interactions can lead to reflexive insights. In this language, we call on anthropologists to recognise the role of uncomfortable knowledge in our reflexive learning and, thus, pedagogies of reflexivity. For us, uncomfortable knowledge is information that directly challenges our implicit and explicit understandings of a given situation and our roles in it. The collection of examples we provide below demonstrate the elegance of our initial assumptions and how interactions with interlocutors revealed these to be oversimplifying and partial. These encounters were uncomfortable at the time, but each (eventually) gave rise to new understandings of the field and who we were in our fieldwork social relationships, moving reflexivity beyond a cognitive and intellectual process to engage the body and emotions. In other words, like other anthropologists we see the fieldworker's discomfiting interactions as central to knowledge production (e.g. Dellenborg, 2013;McGranahan, 2014;Okely, 2007;Spencer, 2011;Watson, 2012), and in particular reflexive knowledge. We advocate for writing openly about the role that uncomfortable knowledge plays in reflexive learning as a necessary step toward teaching our students to become more reflexive.
In the next section, we draw on the theory of sociocultural viability to illuminate the preference for elegance in ethnographic publications and the clumsiness that interactions in the field sometimes yield. We then present examples of uncomfortable knowledge drawn from our own field mishaps and discuss the reflexive learning they enabled. We conclude by considering how acknowledging and embracing discomfort in participant-observation data collection might contribute to developing pedagogies of reflexivity.

Seeking Clumsiness
In anthropology, and other fieldwork-based disciplines, stories of the researcher's humiliation, despair and failure are more often heard backstage in the places where anthropologists congregate -at conference dinners, work seminars, coffee breaks, the conference hotel bar -than found in peer-reviewed publications (Hilgartner, 2000;Johansson & Montesi, 2021;Spencer, 2011; see also Okely, 2007). This derives in part from our understandings of what scholarship and professionalism entail (Okely, 2007;Johansson & Montesi, 2021;Spencer, 2011). When it comes to publication, the peer review process, in which reviewers are asked to evaluate a manuscript's originality, theoretical contribution, use of existing scholarship, research methods, and coherence, leads us to prioritise a work's analysis, argumentation, and production and use of data. Stories about the anthropologist's mistaken assumptions, miscommunications, and difficult emotions do not contribute to this "scientific image" (Spencer 2011, p.77; see also Johansson & Montesi, 2021;Okely, 2007), leading to calls for anthropologists to have the space to reflect on their embodied experiences without damage to their professional reputations (Cain & Scrivner, 2021).
In the language of the theory of sociocultural viability, the peer-review process and conventions of scholarly knowledge production privilege ethnographies that are elegant. An elegant ethnography clearly expresses a specific analytic understanding of the group or phenomenon under study (see Verweij et al. 2011, p.6). The author intentionally eschews and excludes a multitude of alternative interpretations and arguments that could be applied, working to convince readers that the particular theoretical apparatus and narrative is productive and even the "right" interpretation.
The theory of sociocultural viability offers a typology of social forms and their concomitant world views that draws inspiration from the classificatory schemes of theorists such as Durkheim and Weber (Verweij et al. 2011, p.2). It is a strategy for mediating between universalising theories predicated on the notion that all humans are the same (such as rational choice theory) and theories that posit the fundamental incommensurability and radical uniqueness of social groups (such as poststructuralism). The theory of sociocultural viability proposes four ideal types of social organisation -none of which are thought to exist in the world in a pristine state -that support four distinctive epistemologies (6 and Richards, 2017;. The particular features of these ideal types are unimportant for our present purposes, and indeed, adherents of the theory of sociocultural viability do not always employ them (e.g. Thompson, 2008a). Here, the key implication is that people belonging to different forms of solidarity depart from different premises about how things work, which are anchored in their specific modes of social organisation. Further, as Verweij et al. write, "Each way of organizing and perceiving distils certain elements of experience and wisdom that are missed by the others. Each way of organizing and perceiving provides a clear expression of the way in which a significant proportion of the populace feels we should live with one another and with nature" (2011, p.6).
In the theory of sociocultural viability, when members of different solidarities come into contact, conflict may arise. Epistemologies and ontologies clash and their representatives may seek to impose their own perspective over others. When one perspective dominates, the result is elegant but partial (cf. Strathern, 1991). However, interactions between social solidarities also have the potential to yield greater detail in understanding, enhanced representation of diversity, and more responsive action. This is because each solidarity's narrative contains unique knowledge missed by the others. Put another way, because any model of (messy) reality is partial, there is always residual information. Different perspectives will have different residuals of knowledge, as the whole can never be fully grasped within an elegant framing. Thus, within the theory of sociocultural viability, when it comes to grasping and resolving complex conflicts, it is desirable that knowledge drawn from each of the social solidarities is present and respected by members of different social solidarities. By creating institutions receptive and responsive to the knowledge of different social solidarities, elegant situations become clumsier and can produce better understandings and more appropriate actions (Thompson, 2008a;Thompson, 2008b;. In any case, an overly elegant understanding of a situation is liable to be destabilised when information beyond its paradigmatic limits breaks through, producing uncomfortable knowledge which, given responsive institutions, can lead to "clumsy solutions" (Thompson, 2008a). Other attempts to grasp and bridge alterity such as the so-called ontological turn resemble this point of view (cf. Singleton, 2021;Chua, 2015).
In the following we draw on the idea of diverging models of reality and their concomitant epistemological and ontological differences to explore a set of destabilising moments in our fieldwork. The interactions varied in their levels of publicness, duration, and discomfort for participants (anthropologist and interlocutors), but all manifest uncomfortable knowledge that facilitated deeper and more embodied forms of reflexivity. Awkwardness, embarrassment and emotions such as anger marked moments when the anthropologist's elegant and partial understanding was challenged by views and practices deriving from a solidarity grounded in different principles. By describing these experiences in detail, we show how uncomfortable knowledge has the potential to enable reflexive learning, in which our analytical and theoretical models were disrupted by information previously invisible to us.

Cognitive Clashes
McGranahan writes that ethnography is "an open-minded, open-ended collection and celebration of the excess and messiness of human life" (2014, p.24; cf. Van Maanen 2011). It requires, in Deltsou's words, reflecting on new topics and "moving between subjectivities" (2020, pp.29-30). The intersubjective component of anthropological praxis has been emphasised since Malinowski first advocated for participant-observation fieldwork. In his vision of what became the discipline's central method and mode of knowing, the anthropologist enters the field equipped with training in theory rather than "preconceived ideas," and is capable "of changing his views constantly and casting them off ungrudgingly under the pressure of evidence" (2005, p.9). Yet as others have argued, training in theory does not always equate to the ability to cast off preconceived ideas or achieve deep understandings of another's lifeworld (e.g., Dellenborg, 2013;Spencer, 2011). In some cases, clumsiness, awkwardness, hurt, and offence are required for reflexive learning.

Indigeneity and the Blushing Cheeks of a Young Anthropologist (Burman)
As a young anthropology student, I was a member of the local Gothenburg section of the International Work Group for Indigenous Affairs (IWGIA) and very committed to supporting indigenous peoples in their struggles for social, political and territorial rights. At the age of 23, in 2000, I was offered a two-year volunteer position within an indigenous Aymara organization in Bolivia. My job was teaching at a rural institute for higher education and participating in the organization's workshops and other activities in rural areas. My supposed expertise was within the area of "indigenous peoples' rights". I was thrilled. I would have the opportunity to return to the Bolivian Andes to work with indigenous people in their struggles.
After a few months in Bolivia, I was invited to a meeting at the National Council of Ayllus and Markas of Qullasuyu (CONAMAQ) headquarters in La Paz. CONAMAQ was then a recently founded indigenous organization that had crystallized in the late 1990s from the mobilizations around the reconstitution and rearticulation of the ayllus, the traditional Andean sociopolitical territorial nuclei. Again, I was thrilled. I would meet the highest-ranking authorities of the Andean indigenous peoples. As I entered the small meeting room, I was assigned a seat on the right-hand side of the room, where the male authorities (mallkus) were seated on a long bench with their backs against the wall; on the left-hand side sat the female authorities (mama t'allas). Over the course of two hours, the authorities would stand up and speak one by one. At the end of their speeches, they all called out things like "Jallalla CONAMAQ!" ("Long live CONAMAQ!") and "Jallalla the Aymara nation!" whereupon everyone responded: "Jallalla!" There was an enthusiastic atmosphere in the room, a sense of joy and shared commitment. Feeling privileged and proud to be at this significant event, I absorbed it all. Finally, I was invited to say a few words. I wanted to be as good as anyone else, so I ended my speech by exclaiming: "Jallalla the struggle of the indigenous peoples!" I had expected a collective "Jallalla!" from those present. Instead, there was dead silence.
The authorities looked confused. "The indigenous peoples?" asked the apu mallku, the principal leader. I nodded, "Yeah?" After what seemed like an eternity, he too nodded his head and said: "Yes, you're right. The struggle of the indigenous peoples is similar to our struggle! So sure -jallalla brother!" With cheeks glowing, I sat down and the meeting continued. I understood that I had said something awkward, but did not understand what the apu mallku meant by saying that the struggle of the indigenous peoples was similar to their struggle. To me, the Aymara and Quechua authorities who were there in the small room were all indigenous authorities; consequently, the struggle of the indigenous peoples was their struggle. How could it be otherwise? I bore my embarrassed confusion for some time. Gradually, however, I started to understand the historical and political reasons why Aymara and Quechua people did not identify as being "indigenous." More than a decade later, I published an article about the dangers of supposing that external analytical or legal categories mirror people's own sense of identity and ways of talking about themselves (Burman, 2014). This uncomfortable interaction was my first step toward understanding that the concept I had taken for granted was problematic to the very people I had applied it to.
I have used this anecdote in teaching in Sweden and Bolivia as an example of epistemological and conceptual discomfort. My Swedish students seem to receive it as a cautionary tale or a comforting anecdote to take with them into their own imminent and potentially clumsy field encounters. Perhaps it reveals discomfort as an underappreciated condition for reflexive learning, for understanding that one's external analytical categories may diverge sharply from others' sense of who they are. While my Bolivian students also reflect critically on the anecdote, they tend to see the situation more straightforwardly: an anything-but-elegant gringo anthropologist saying stupid things. Sometimes, their hilarious laughter makes me recall the feeling of being a young anthropologist with blushing cheeks.

The One Where I Lost My Rag (Singleton)
Like many social scientists, much of my work is based on an epistemological perspective of the basic equality of all humanity and a commitment to the dissolution of inequities (of course, whether I do this is another matter). A silly presumption with this on my part is that most marginalised peoples share a similar commitment to social justice. A particular case shook me as I lost my composure with a key interlocutor during my work on migration and integration in Sweden. I will not give precise details as I wish my interlocutor to remain anonymous. To summarise, I was chatting over lunch with the interlocutor, someone who had fled conflict and now works in integration projects, and another immigrant. Conversion turned to the coming elections and the interlocutor explained that he intended to vote for the far-right populist party, the Sweden Democrats (SD). Immediately I felt the blood rushing to my cheeks and a heated argument ensured. My interlocutor said that the influx of Muslim immigrants was Sweden's biggest contemporary societal issue and only SD were prepared to take action. This surprised and angered me. I had assumed the interlocutor's job and, more lazily, his own traumatic background would make him more sympathetic to the plight of many Muslim refugees. Furthermore, the interlocutor seemed completely unconcerned with many other issues that I articulated about SD, whom I consider to be dishonestly, callously and dangerously undermining many of the foundations upon which Swedish society and democracy are based. Put another way -keeping Muslims out was more important to the interlocutor than maintaining democratic and social norms. In his eyes, I was naïve.
Once I had calmed down, I was able to analyse the situation. The interaction's emotional tone clearly indicated the appearance of uncomfortable knowledge. The event revealed the overly elegant framing I had been making of this situation, concealing what were arguably racist and definitely ignorant assumptions about my interlocutors. Many people around the world live in societies where social difference and inequality are givens, closely woven into the social fabric (cf Thompson, 1979Thompson, [2017, pp.158-60). Likewise, many people live lives as best they can in iniquitous, undemocratic contexts and necessarily some people benefit more or less than others in these situations. I now realise it is absurd to imagine that simply crossing a national border will instantly transform someone into a passionate pro-democracy advocate, with a commitment to the emancipation of society. Furthermore, when coming from conflict situations, persecuted people may be willing to burn the bridges that facilitated their arrival rather than allowing perceived enemies to arrive, no matter the cost. This was deeply uncomfortable to me, involved as I was in an academic project rooted in the fundamental equality of humanity. More broadly, this incident feeds into concerns I have regarding the methodological limitations of equality-based perspectives in a world where many people are perfectly happy with caste, gender and class differences. The elegant assumption that everyone actually regards egalitarianism as a utopia or that people mean the same thing with egalitarianism cannot withstand the encounter between different worldviews.

Preserving Old Houses (Gillette)
During the mid-1990s, I did field research in the Xi'an Muslim district in Northwest China. Xi'an is a very old city and Muslims have been living there for a thousand years. During my fieldwork the Muslim district had several old houses and mosques constructed in what can be called traditional styles (Gillette 2000, chapters two and three). The area stood out for its quantities of older architecture; historic buildings in much of the rest of the city had been replaced with Soviet modernist construction. I met families who lived in two-hundred year old wooden houses. One of the first times I was invited into one, I waxed enthusiastic. It was lovely. I said something about how important it was to preserve such valuable historic buildings.
The woman showing me the house looked at me in silence. Then she grabbed my hand and took me to the toilet. It was an outhouse, a simple wooden structure with two holes as commodes. It stank. There was no electricity and no running water. In fact, the whole house lacked electricity and running water, and the rooms were heated with small portable coal stoves. She looked me in the eye and said, "How would you like to live here? We don't want to live like this. We want running water, proper toilets, electricity, and central heating." Her son added, "We're not going to preserve this house. As soon as we have enough money we are going to tear it down and build a modern house." This interaction took me by surprise. I had never thought about the burdens that historic preservation might put on a family. I hadn't questioned how "good" preserving old houses was, or for whom or what it might be "good." Recognising what a luxury it was to be able to afford historic preservation without requiring someone to live in a house with a malodorous commode and its concomitant labour, was uncomfortable knowledge. I was 26 years old and shocked to see my own unconscious colonialism. The encounter was clumsy and unsettling -and moved me toward a more encompassing and demanding form of reflexivity.

The Atheist Prayer (Blanes) 2
Back in 2008, my colleague Ramon and I made a short visit to Mbanza Kongo, former capital of the ancient Kingdom of Kongo, located today in the province of Zaire in northern Angola. Despite its glorious history, it is today a typical mid-sized Angolan town, suffering from decades of exodus to the capital Luanda and lack of infrastructural investment. On the other hand, it has remained a symbolically and politically relevant site, not only due to its royal history, but also to the recent listing of its urban centre as a World Heritage Site.
Ramon and I were coresearching Christian and other prophetic churches in Angola and how they conceived this ancient capital within their current spiritual and political imaginaries. We spent our days visiting the Kimbanguist, Mpadist and Tokoist churches, attending their services and interviewing their members. One visit was to the pastor of the Tokoist Church. This movement, one of the main Christian churches at a national level (cf. Blanes, 2014), was not the most popular in town, so when we knocked at his door, the pastor was happy to welcome us. He showed us around the compound, where we saw children being schooled in the backyard, and pointed to the door of the local tabernacle, a hallowed ground for the church members. Then the three of us entered a room and sat down for a long conversation. One fascinating topic was the pastor's description of heaven, which he had seen in dreams.
My most striking memory of the encounter arrived at the end. When the conversation was over, the pastor decided that we should perform a collective prayer. Although we had been clear about our academic background, this did not prevent him from not only suggesting this prayer, but also asking me to conduct it while we held hands. Up to that moment, I had attended dozens of religious services, in churches with many theological, spiritual and ritual traditions, but never been asked outright to "say a prayer." In the split second from request to response, the following crossed my mind: I am familiar with the Tokoist liturgy, and could just recite the Lord's Prayer as is common at the closure of their services; I am an atheist. I do not feel very comfortable uttering an insincere prayer. It is too late to explain any of this to the pastor. Do something. Improvise. I rapidly decided not to make an actual Christian prayer, but to utter some sentences that reflected a sense of thankfulness for having been welcomed by the pastor and spending time learning from his wisdom. Using the same solemn tone of a Christian prayer, I said something along the lines of: I am thankful for the moments that we spent with the pastor, and for the wisdom he has shared with us. May we take this wisdom with us and do good with it. Let us remain safe and healthy after this meeting. Thank you and goodbye.
After I finished, the pastor's puzzlement spread across his face. However, he did not comment, and eventually seemed accepting. I felt relieved and embarrassed. Afterwards, returning home, Ramon laughed and said it was the first time he had ever heard an "atheist prayer." Every now and then Ramon and I recall this episode. While it makes us laugh, it is also a constant reminder of how ethnographic awkwardness can be productive, pushing us into a reflective, critical space. In this case, it made me realize how divergent the motivations and expectations of the participants to an ethnographic encounter can be, and how such interactions can have plural outcomes and interpretations according to whom is present. In anthropological research in religious contexts, such possible divergences can lead to uncomfortable questions about personal conviction and in/sincerity that are the basis of reflexivity (cf. Blanes, 2006).

Discomfited Bodies
Conducting anthropological research involves placing oneself in a web of power relations, a topic that has generated numerous discussions and reflections (e.g., Bourke, 2021;Burman, 2017;Kulick 2019). Yet practitioners of what has been called "the Church of Anthropology" tend to assume a certain moral superiority, claiming "that anthropology itself is a source of virtue, doing and intending what is right" (Jöhncke, 2021, p.202). This feature of the anthropological habitus (Coleman, 2011) may encourage blindness or immunity to some of the consequences of anthropologists' ways of being in the field.

The One Where I Learn How Easy it is to Hurt Someone (Singleton)
This event took place in Ocho Rios, a small town on Jamaica's north coast. I attended a focus group discussion with a support group for people living with HIV/AIDS (PLWHA) and was killing time with my contact, an attractive, vivacious, witty and friendly young woman about my age. We walked along the seafront, admiring a few colonial landmarks, and as we strolled side by side, my hand brushed against hers. Nervous and hyperconscious, I worried this inadvertent gesture would be misconstrued as some sort of clumsy sexual approach or something, so pulled my hand away quickly. She looked at me and asked if I was scared of catching HIV from her. Feeling like an idiot and even rude, I apologised and stammered out an explanation "I just don't like touching people." The moment passed, we continued to have a nice time, the focus group went well and we had positive communication afterwards.
Even so, this event still bothers me. Conducting research on HIV-related issues, I know stigma is very real issue for many people living with HIV/AIDS (PLWHA), in Jamaica and elsewhere, and can have violent consequences. Much HIV/AIDS advocacy tries to encourage PLWHA to live as normal a life as possible (Singleton, 2013). I had taken ethics courses and prepared for this research, but this uncomfortable moment yielded embodied knowledge about how a small, inadvertent action and reaction could reinforce patterns of stigma and discrimination in a interlocutor's life. Despite the fact that I thought I was reflexive, the surprise from this brief bodily contact generated learning that my physical presence could become part of stigmatisation in ways which I had not truly known. My reflexivity was too elegant: I was intellectually aware of injustice but had not fully engaged with its embodied reality (cf. Ese-osa Idahosa and Bradbury, 2020, p.34). No research I do justifies hurt. In retrospect, this contact became an opportunity to explore the deeper meanings of stigma for this young woman and others in similar positions.

Buying Books (Gillette)
One day during fieldwork in Xi'an I stopped by a small privately-owned bookstore next to a mosque. I had been visiting a family in the neighbourhood and an interlocutor recommended that I read a book that was for sale in this shop, which specialised in religious literature and literature about Chinese Muslims. After browsing for a while, I found several books and pamphlets to buy. I took them to the man minding the store. He examined the titles, and then removed several. "You can't buy these," he said. "These others you can buy." Then he told me the price.
I was surprised. "What do you mean?" I asked. "I'd like to buy all of them." He looked at me closely. "Are you a Muslim?" he asked. "No," I said. "Then you can't buy them," he replied.
I explained that I was a researcher and trying to learn about Chinese Muslims, the neighbourhood, Islamic practice, local history. He shook his head. "You can't buy these," he repeated. I explained that I grew up in a Catholic family, and so was part of the same monotheistic tradition. "No," he said. "You can buy the other books, but not these." I bought the remaining materials and left. I was flamingly angry riding my bike home and for a long time afterwards. Twenty-five years later, the experience remains powerful. His "no" was absolute. It was simultaneously personal and impersonal: it had everything to do with me and nothing to do with me. There was no workaround, no possibility for me to have the dignity or humanity or sacredness that would allow me to buy those books. I was unworthy and could do nothing to change it.
This (for me) uncomfortable interaction gave me a different view of who I was in the field to the people I sought to learn from. I had to confront the multiple facets of who I was to my interlocutors. I was used to being treated as a privileged person in the field. At that time there were relatively few foreigners in China and even fewer who spoke Chinese. I knew that locals talked to me, invited me into their homes, and allowed me to join in their activities because I was white, American, spoke Chinese, and had good entertainment value, as Barley discusses (1986a, pp.57-58). The patina of my graduate institution (Harvard) further eased my reception. I tried to be conscious of the privileges I was accorded and strove not to abuse them, but I benefitted enormously from how my interlocuters saw me -most of the time. This interaction made me recognise that while I could be a status symbol, I was also profoundly "dirty," in the same way that this particular minority group saw the surrounding Chinese majority population as dirty (see Author B, 2000).
In time, this uncomfortable knowledge cast new light on the anthropological project for me. The anthropologist does not "earn" access to the field, even though the many physical discomforts of fieldwork might lead us to think we do (cf. Johansson & Montesi, 2021). As Kulick writes (2019, 260-270), while we live in a world that binds us to one another, no particular group or individual is bound to engage, edify, or even sell books to any other.

The One Where I Felt Alienated from My Interlocutor (Singleton)
An acutely awkward moment occurred during my PhD research while I was conducting fieldwork in the Faroe Islands on whaling practices (grindadráp). Circumstances had dragged me somewhat unwillingly into looking at an ongoing antiwhaling campaign. Launched by the controversial environmentalist organisation Sea Shepherd Conservation Society (SSCS), Grindstop 2014 was the talk of the Faroes that summer. Groups of uniformed, foreign volunteers constantly monitored Faroese whaling bays, while boat-borne teams stood ready to intervene if a whale drive occurred (see Singleton, 2016). As part of my research, I sought interviews with SSCS and eventually secured an interview with a senior member. Taking place in the social science kitchen at the University of the Faroe Islands in Torshavn, it was a strange and uncomfortable experience. I was nervous as the tone of the campaign was often heated and aggressive and was immediately caught off guard when two people arrived rather than the expected one, the interviewee plus another SSCS member, a tough-looking woman who said almost nothing throughout but sat to the side and looked on. Slightly rattled, I began asking questions. The conversation was largely good humoured and it was interesting to hear the pro-SSCS viewpoint (many Faroese, including those that oppose whaling, were critical of them). However, the interlocutors' rhetorical tactics were surprising. In any discussion of, for example, a pro-whaling talking point, the interlocutor would situate us -the interview participants -on one side or another. She repeatedly used the formulation "YOU'RE saying" in her responses. This, and a general aggressive and strident tone, discomforted me, a feeling which was exacerbated by the unnerving third person's presence. I never really felt that we were able to understand one another. Communication stalled at the surface level. This moment made me aware of a potential methodological weakness: what if our worlds cannot meet? Put differently, what if some uncomfortable knowledge can never be fully grasped? Analytically, I interpret this interlocutor's discourse as an example of "enclaving" (cf. Richards, 1999). Her rhetoric latently functioned to maintain the boundaries of group solidarity and forced me to make a choice. If I wanted to understand SSCS's social world from an emic perspective, then I would need to internalise the view that the act of killing whales is murder. Yet the SSCS depiction of the Faroes as a sinister, violent place where holders of antiwhaling views are persecuted jarred with my personal experiences of pro-whaling interlocutors. Members of this latter group at times expressed empathy for SSCS activists. This discomfiting interview revealed that what I considered "objective" was actually inherently elegant. A desire to see both sides of this particular quarrel meant I could never truly achieve an emic SSCS perspective. Unlike my elegant understanding of the researcher's task, actual research practice placed me on one side in a conflict. The surprise was realising that I had to choose between two elegant lenses.

Grieving (Gillette)
Long-term fieldwork gives anthropologists the opportunity to form lasting relationships with the people they meet. One person I became close to during fieldwork, and who in fact played a key role in getting me through the bureaucracy required for doing research in China, was Big Brother. Big Brother was vigorous, intellectually curious, and chatty. He was also a chain smoker. He died of lung cancer four years after we first met. The time from diagnosis to death was rapid -I saw him once post-diagnosis and when I returned a few months later he was dead.
Arriving in Xi'an after Big Brother died, I first visited his grieving wife and children. Later I went to the house of another family with whom I am close. They had met Big Brother's wife. I sat down with the mother, whom I called Little Aunty, and we got to talking. I told her about Big Brother's death and began crying. Tears rolled down my cheeks and I put my head in my hands and sobbed.
Somewhat later, when other members of the family came in and then some neighbours, Little Aunty told them the story of me crying. She gave them a demonstration, putting her face in her hands and lifting and lowering her shoulders and making noise. She probably told this story three or four times while I was there. Perhaps it was recounted later to other audiences.
Little Aunty's performances led me to an embodied understanding of our differences that I had lacked before. I had attended funerals in this neighbourhood and seen very different grieving practices. I had heard the laments that women made and watched the community practices through which women (and men) mourned during funeral and burial rituals. Reflecting about this experience now, I suppose I assumed that everyone cried the way I did when they were behind closed doors -if I thought about it at all. This experience forced me to abandon my elegant view of grieving. It enabled deeper -if uncomfortable -knowledge about the limits to an ethnographer's ability to develop intuitive embodied understandings of the people with whom they do research.

Learning Reflexivity
Anthropological research is predicated on the idea that it is possible for diverse groups of people, who belong to different social solidarities and possess different world views, to understand one another. From the perspective of theory of sociocultural viability, anthropological research is usually clumsy rather than elegant in aspiration, seeking to consider "social and political processes from a variety of normative viewpoints" (Verweij, Thompson & Engel, 2011, p.246). An affinity for awkwardness is arguably the hallmark of the anthropologist (cf. Chua and Mathur, 2018, p.1;de Pina-Cabral, 2018;Johansson & Montesi, 2021;Van Maanen 2011). Yet academic publications in anthropology, as in other scientific disciplines, tend toward elegance, which hampers teaching and learning the cognitive, affective, and embodied critical self-consciousness that we call reflexivity. We contend that anthropologists who seek to develop pedagogies of reflexivity must become more open about their uncomfortable knowledge, describing their awkward encounters not only during conversations over coffee or supervisions, but in their public presentations and ethnographies. Reflexivity, in practice and pedagogy, is foreclosed by overly elegant understandings and/or representations of the research situation. Instead, reflexivity emerges as uncomfortable embodied knowledge at the nexus of what McGranahan calls "destabilizing moments" "where two things are ostensibly not coming together" (2014, p.29).
Encounters such as those we describe here can be painful, frustrating, or amusing, but they offer enormous potential for learning. Difficult emotions highlight the socially embedded nature of ethnographic research (Chua & Mathur, 2018), and can be signals marking the opportunity to generate knowledge that is uncomfortable yet transformative (cf. Spencer, 2011). A maxim of ritual studies is to pay attention to changing emotional situations, to crescendos and climaxes as key to ritual practice (Grimes, 2014). The researcher seeking to learn reflexivity can take a similar approach, tracking the emotional and conversational energy flowing in an interaction (cf. Collins, 2004). When confronted with discomfort, the anthropologist can ask what it was that made the situation unsettling from her own and other participants' standpoints. Such a question draws on the scholarly tradition that moral systems are revealed through how a given group deals with that which is considered polluting (6 and Richards, 2017, Douglas, 1966[2002). When the situation permits, shared discussion about these encounters among all participants might help those involved to tease out the contours of the "patchwork" (cf. Chua, 2021) of clashing solidarities and worldviews, perhaps stimulating new uncomfortable knowledges for everyone.
In recent years, anthropologists have sought to identify how to practice our discipline with reflexivity, respect, and ethical relations (e.g. Bourke, 2021;Cummins and Brannon, 2021;Ese-osa Idahosa & Bradbury, 2020). Some have questioned an apparent disciplinary moral superiority, the assumption that doing anthropology is doing good (Jöhncke, 2021). To these calls for renewal, we add our own plea for recognizing and discussing uncomfortable knowledge in our professional presentations, publications, and classrooms. We echo others' calls to allow prospective students "backstage" earlier in their training, exposing them to the awkwardness and emotions of reflexive science (cf. Cummins & Brannon, 2021;Ese-osa Idahosa & Bradbury, 2020, p.42) as well as the other discomforts of field work (Johansson & Montesi, 2021). Teaching is a central site where anthropologists compare, reflect, and talk about what anthropology is, playing an important role in self-making for ourselves and our students (Coleman 2011, p.4, 9). Acknowledging the limits of elegance in the production of knowledge and embracing discomfort as central to reflexive insights and ethnography's transformative capacity, should become mainstream.

Toward a Pedagogy of Reflexivity
As Myerhoff and Ruby write, reflexivity "turns us to contemplate ourselves just as we may be beginning to realize that we have no clear idea of what we are doing" (1982, p.1). As we argue here, anthropologists could contribute to developing pedagogies of reflexivity by speaking openly about their discomfiting experiences with students and offering concrete examples in their publications about how uncomfortable knowledge led to more profound reflexivity. To further advance our shared capacity for teaching and learning reflexivity, we conclude with four recommendations for anthropologists and their students.
1. Include uncomfortable knowledge in the data collection. Writing and other forms of recording have a role to play in making reflexive learning possible. Many of the interactions we discuss here were fleeting and of limited significance to the questions that brought us to the field. Yet, because we wrote about these encounters in our field notes, the uncomfortable knowledge that they facilitated remained available. Anthropologists everywhere should be encouraged to observe their moments of discomfort and interrogate their causes. We suggest seeing strong emotions such as anger or embarrassment as signals marking the possibility for enhancing reflexive knowledge, moving reflexivity beyond the intellectual to become embodied. Ask students: how can we 'make use' of our ethnography-derived emotions in our journal writing?
2. Incorporate the concept of uncomfortable knowledge into teaching. Others have observed that anthropologists often underplay discomfort, including in physical terms (e.g., Dellenborg, 2013;Johansson & Montesi., 2021;Okely, 2007;cf. Kulick, 2019). To this we add the need to attend more to discomfort as a site yielding possibilities for reflexive learning. Our stories show that uncomfortable knowledge occurs in social interactions and relationships. The anthropologist does not always experience these interactions and relationships as positive or friendly. She may not feel good about who she was in those encounters. Mistakes and awkwardness are intrinsic to the methodology and they can be a window onto deeper, more embodied knowledge, including our ability to see who we are in our field site through someone else's eyes. In the classroom, discuss episodes of discomfort as 'events', with an ad hoc reflection on the motivations and expectations behind each reaction in the event.
3. Take your time: reflexive learning is a process of coming to know. Discomfort can be an indication you need to see yourself differently -and this reflexive learning can take a while. To generate reflexive learning from discomfiting encounters, the anthropologist must scrutinise her presence and her reactions, perhaps repeatedly returning to an incident over a long period. Use emotional intensity as your guide to opportunities for learning, even if such knowledge is hard-won and slow to develop. In the classroom, use the embarrassing events for a discussion on ethnographic positionality and its shifting condition.

Recognise limits.
A reality of anthropological research is that one may confront limits to the extent one likes one's interlocutors or is willing to accept their values. We may even find, using the language of the theory of sociocultural viability, that fieldwork as a practice and anthropology as a discipline are elegant and there are limits to how clumsy we are willing to be. This realisation can be painful. Yet, this discomfort leads to an important question for reflexive learning: how far will you go to understand another person and what compromises to your own views and values are you willing to make for this understanding? What are the epistemological consequences of those limitations for the overall research project?