Disturbing hierarchies. Sexual harassment and the politics of intimacy in fieldwork

This paper examines how sexual harassment is often mediated through the making of imagined complicities that are constructed to imply that an alliance/compliance underpins the relationship and ‘justifies’ the harassment. It is concerned with how the making and doing of intimacies engages with broader hierarchical structures of power, including structures of inequality. Fieldwork is viewed as a site in which the politics of intimacy exposes normative expectations and structures of inequality. Specifically, the discussion exposes how processes of Othering are mobilised by participants as a means to cultivate imagined complicities but expose discrimination. The paper calls for a reappraisal of the focus placed on building rapport and/or a sense of familiarity in qualitative research to take account of multiple forms of intimacies and risks they can entail. This is increasingly prescient in light of the renewed emphasis on participatory methods and co-production which entail closer working relationships.


Introduction
The experience of sexual harassment in research interactions is slowly gaining greater consideration, although it remains acutely lacking in formal methods teaching, and broader discussions within the academy. In particular, scholars have drawn on their own experiences in order to encourage reflection on aspects of safety and how we might manage these kinds of interactions (e.g. Green et al., 1993;Kloß 2017;Roguski and Tauri 2013;Zurbriggen 2002;Clark and Grant 2015), but also as a means to reflect on the role of power in research relationships and, especially, how feminist commitments can be met within hierarchical research relationships that are produced through gender and sexualisation (Gailey and Prohaska 2011). More recently, the #metoo movement has given momentum to open up discussions about violence in fieldwork settings in ways that draw connections to the machoistic and neoliberal culture of academic institutions more broadly whilst also calling for systemic and institutional change (e.g. Appert and Lawrence 2020;Berry et al., 2017;King et al., 2020;Schneider 2020). These more recent contributions have also begun to shift focus away from solely those types of gendered relations in which hierarchical features appear more pronounced. As will be discussed below, sexual harassment has been discussed most commonly in research with men in, for example, elite managerial positions, with sex market buyers, prisoners, or in research about gender and sex specifically. Whilst we can identify the beginnings of a shift, there is nevertheless a pressing need to expand work that considers sexual harassment in research in what might best, if a little clumsily, be described as 'ordinary' interactions.
In this paper, I reflect on a set of interviews and conversations that I conducted as part of a project exploring changes in policy, practice and activism broadly related to questions of race and ethnicity. The interviews and less formal conversations were, in the main, with civil servants, community activists, community voluntary workers and public and third sector professionals. The content of the discussions was not focused on any aspect of sex or gender and did not purposely include participants who we knew to have a history of sexual harassment or held significantly powerful roles. For clarity, I have adopted Kloß' definition of sexual harassment, which they characterise as, 'coercive behaviour, which may include gestures, actions, and other modes of verbal or nonverbal communication, with sexual connotations, which intimidate, humiliate, and exercise power over another person' (Kloß 2017: 4). In the instances described below, this includes a range of practices from light touching and verbal flirtations (including over complimenting), patronising talk, to offers of dates, marriage and sex.
The core aim of this paper is to examine the politics of intimacy and, specifically, how sexual harassment affects the making and doing of intimacies in fieldwork. Principally, I explore how unwanted intimacies and sexual harassment disturb hierarchies and destabilises research relationships. This is an exigent concern given the retained focus on rapport building (especially in methods teaching) and the shift towards more co-produced research which implies a closer (more intimate) shared practice. In doing so, I am concerned, not so much with how dynamics of researcher-participant relationships are salient to the personal encounter, but rather how they connect with broader hierarchical structures of power, including structures of inequality. Specifically, I examine how sexual harassment is often mediated through the making of imagined complicities that are used to imply that an alliance/compliance underpins the relationship and thus justifies the behaviour. I especially want to expose how processes of Othering (of other women) are mobilised by participants as a means to develop imagined/unwanted complicities and intimacies. Paying attention to these processes is important because it gives us insight into the contexts in which research takes place and has a bearing on what and how knowledge is produced.

Sexual harassment and research methods literature
Despite recent interventions, there remains an overwhelming lack of attention paid to sexual harassment in mainstream discussions in the academy and it is a topic that is almost always absent in analyses across disciplines, including Sociology where this author is situated. Until recently (post #metoo) sexual harassment typically only emerged in discussions about gendered relations in research in which hierarchical features appear more pronounced or form the main purpose of the research itself. For example, there is a body of literature that discusses sexual harassment in relation to research with men who have a history of violence against women (Presser 2005;Gailey and Prohaska 2011), or hold elite positions (Lee 1997), are involved in commercial sex (Grenz 2005;Walby 2010), or in research that examines specific gendered dynamics in, for example, the labour market or personal/family life (Arendell 1997;Gurney 1985;Pini 2005), or takes place in male-dominated settings (Soyer 2014;Baird 2017). This continues to be an important gap to address for several reasons. First, when debates emerge out of contexts in which hierarchies are signalled to be more pronounced, there is a risk that this will embolden assumptions about what kind of men are most likely to engage in these behaviours. Green et al. (1993), for example, describe how they were warned about interviewing drug users and prisoners but that no one had thought to warn them about interviewing policemen or prison officers. Such an omission lends itself to the deepening of clumsy and essentialising constructions of gender, which become more problematic when we think this through intersectionally to consider how class, race and other social identities may lead to false conclusions. The hypersexualisation of black and working class men and women, for example, indicates the potential for some bodies to be overly policed and falsely ascribed with higher potential for risk, or, to be held more accountable for the sexual harassment they experience (see, for example, Reynolds 2016; Slatton and Spates 2016 for a discussion of the impact of hypersexualisation). Second, if we only consider the implications of sexual harassment within particular types of fieldwork then we risk overlooking the impact this has on knowledge production more broadly. We thus risk ignoring the politics of intimacy and what 'mundane' occurrences of sexual harassment can tell us about the context in which research is situated, and our participants' ways of seeing and interpreting the world that would prove useful for our analyses.
Only recently have we begun to see research that captures how sexual harassment (and other forms of violence) takes place in what we might clumsily call 'ordinary' research interactions, and with men that are not connoted with violence by the researcher at the outset. Paying attention to the presence or persistence of sexual harassment in all types of research settings can help move us to a point from which we can consider when and how it emerges, but also what this can tell us about the making and doing of intimacies in such interactions. A precedent has, if in a somewhat tangential way, been set by the ways in which we recognise the importance of paying attention to when laughter, hesitations, mumbling or pauses occur in our conversations with research participants. Attending to these nonverbal cues, or divergences from the core narrative, have been included as important elements that can bring greater depth to our analysis. Research, on how people talk about race, for example, has usefully highlighted how these instances tell us something far greater than what is un/said (see Harries, 2017 for examples of this). They are often brought to the fore to highlight the role of dominant or marginalised experiences through the analysis of struggle, discomfort, bravado or shame, thus enabling a more rounded understanding of what is being said (or not). Hence, this work can point us not only to the salience of where there is a hierarchy present in the personal interaction itself, but also tells us about the wider political context in which these interactions take place and the need for these moments to form part of our analyses. Why then has the same attention not been given to moments of sexual innuendo, flirtation and propositions? The making and doing of intimacies cannot, after all, be fully understood when confined to the personal. They are entangled with complex questions of power.
There is considerable credence to the argument that sexual harassment is rarely accounted for or even recognised within the academy because individual researchers experience it as an extension of routine everyday life. Indeed, it does not stand out just because it emerges during our interviews or ethnographic encounters (Harries 2016;Gurney 1985;Green et al., 1993;Schwalbe and Wolkomir 2001). These instances are typically entirely normalised. However, to understand their absence more broadly we must also reflect on the failure of core research methods training to address issues of gender in general, let alone issues related to sexuality and sexualisation (Orrico 2015), and intersectionality (Berry et al., 2017). Hence, sexual harassment goes undebated not solely because individuals fail to recognise it but because, as Hanson and Richards (2017) note, these absences are situated and reproduced within the structures of the academy, which is itself driven by a neoliberal machoistic culture. This is most acutely illustrated by the normalisation of patriarchal power relations within higher education and the 'institutional airbrushing' which glosses over and conceals sexual harassment and assault (Phipps 2020). Indeed, despite important and bold interventions by feminist methodologists, a white androcentric perspective nevertheless continues to shape the epistemological foundations of research (Berry et al., 2017;Hanson and Richards 2017) and this in turn leads to the construction of ideological norms about what research, and especially good quality social research, looks like. Institutional structures do not facilitate deviation from what is conceived as fitting in to the normative idealised trope of a researcher (Berry et al., 2017). These risks are more pronounced for black women, whose knowledge is delegitimised in the academy broadly speaking, but are also not validated within white-led feminist debates that dominate and around which our institutions are structured (Alexander and Mohanty 1997;Buggs et al., 2020;Hill Collins 2000;Lewis 2013). To understand this more closely, we only need to reflect on what a 'good researcher' looks like, and then specifically what implications this has for how we account for (or not) sexualisation and sexual harassment in research.

Sexual harassment in the field
This discussion draws on a set of interviews and ethnographic encounters that I, a white heterosexual woman, conducted with men working in the public and community sectors. These research interactions were part of a group study exploring changes in policy, practice and activism broadly related to questions of race and ethnicity. They took place largely in neighbourhoods in four cities of the UK: Cardiff, Glasgow, Manchester and London. Neighbourhoods were chosen in the four cities because they each had long histories of migration and had been at the receiving end of numerous public and policy discourses and interventions. The discussion draws from interviews and fieldnotes that detail formal and informal conversations with male participants. The interviews and less formal conversations are, in the main, with civil servants, community activists, community voluntary workers and public and third sector professionals. In total, I interviewed 48 people, 22 of which were men and from various ages and ethnicities. I also have 52 fieldnote entries that detail individual or group conversations of varying lengths that involved one or more men. Whilst this was a group study, all these interactions were conducted on a solo basis and, for the most part, the author was also the only member of the team present in the research site at the time they took place.
The research on which this discussion is based did not seek to explore sex, sexualisation or sexual harassment. Gender was a matter of concern in the sense that we were interested in how gender intersects with race and other social experiences to shape policy and practice (see, for example, Harries et al., 2020). It is also important to note that the thinking that went into this discussion did not start until after most of the data collection had been completed and interviews were transcribed. Moreover, it was only after reading through transcripts that had been typed up by an external transcriber that this topic became of interest worth further analysis. Hence, this experience consolidates the idea that sexual harassment goes largely unnoticed and unremarked on, which, as I will go on to discuss, means that we also avoid addressing the politics of making and doing intimacies and how this is entangled with power and hierarchy beyond the personal interaction. Specifically, it was the following extract that encouraged this analysis: This extract is taken from an interview with a housing officer. The interview was largely about local housing provision and policy, to include how this intersects with integration strategies. Much of the focus had been on the changing migrant population and representation of, and organisational interaction with, people categorised as ethnic minorities. It also touched on local discourses of migrants living locally and how this has affected housing allocations. The content of the interview centered on these issues and so, reading the full interview in print, I was struck by how unexpectedly the question appears. Furthermore, listening back to the recording there is nothing from the tone to suggest this proposition is coming and, since I switch the tape off quite abruptly in response, indicating my discomfort, I do not know what I said in reply. Following the interview, I did also receive two emails giving me advice of places to go out to listen to music and an invitation to join me at some point (implying a shared taste), which I ignored.
As with most of the other interviews in this project, we had not exchanged much information about each other beyond what was necessary to organise the practicalities of when and where to meet. The interview took place in an office kitchen which was empty on arrival but occasionally had people coming and going. The semi-private space used was also a quite typical type of interview location in this project as few of the level of workers we were interviewing had their own personal offices. In many ways, this is counter to findings in other projects. Pini (2005), for example, found that sexualisation practices tended to occur only when her male participants were speaking informally to other men in front of her, rather than during the one-to-one interview itself. My participants, however, were very openly flirtatious and suggestive during recorded interviews as indicated in the example above.
In total I was directly 'asked out' by three different men in the course of interviewing them. I received one marriage proposal and two suggestions of sex. More common were insidious forms of 'flirting', which often included follow-up emails or texts after interviews. The development of over-familiar and often unwanted intimacies occurs in various ways. In ways that can be quite subtle, or those that can be crudely stated and can rapidly escalate. All were examples that are familiar in everyday life. They included attempts to be chivalrous (see also Presser, 2005), and/or condescending (see also Pini 2005) and often incorporated gentle touching on the arm or lower back. As will become clear in the following discussion, many of these practices also relied on producing me as a particular kind of womanreliant on how assumed class, ethnic, sexual and national identities were used to construct me simultaneously as heterosexual, available, naïve and caring. For example, by interpellating ideas about white Welshness with sexuality and, simultaneously, promiscuity and motherliness (see Aaron 1994;Harries 2016). What is significant is that these practices all occurred in ordinary research interactions and what is common to all these examples, is the sense of over-familiarity that was assumed and adopted by interviewees and the absence, for the most part, of my recognition of them at the time. Indeed, over-familiarity is the key characteristic of the intimacies being done and made here. The idea that these men already know me, know my tastes and what I need, know me well enough to touch me is a common dynamic but I want to focus on how this also produces a set of assumed complicities.
Two more examples are detailed below in order to help illustrate how a sense of overfamiliarity is cultivated to also imply a collaborative relationship around a broad set of social norms and how this produces an imagined complicitynot solely in the sexual contractbut a complicity in the production of a set of, often problematic, social norms. The following, for example, demonstrates the intrusive nature of sexual harassment that mimics a space, imagined by the harasser, as one in which we share a level of intimate familiarity that relies on 'shared' normative codes.
I This fragment of conversation and the relentless pressure it emits will be familiar to many when sexuality and availability is determinedly sought for in a way that suits. In the continuing interaction and in the numerous texts and emails he sends later, he persistently questions my role as a working woman and portrays himself as my potential saviour with repeated marriage proposals that will, he explains, facilitate my chance to give up work and look after him instead. As part of this, he insists he wants to marry 'a Welsh woman' and draws on a set of tired tropes of white Welsh matriarchal 'mam' figures to do thisthe significance of which I will return to in the final part of the paper. These interactions with X also highlight the significance of the tension between wanting to seek some level of rapport with those we encounter in our research and the difficulty in encountering sexual harassment in response. It also highlights the complications and difficulties in relation to managing, or negotiating, the closing down of these spacesin this example in terms of the local community hub space which I feel I can no longer return to.
The second illustration, includes a section of conversation with Y. In our conversation, he describes how, as a black man, his experience of racism had evolved over the years in relation to shifting local and global contexts. Part way through the interview he then proceeds to offer the following: He repeats this sentiment a moment later with an ambiguous use of 'you': Yeah. You know, you're okay. You're a white female, I fancy you and it's just the males (both laugh)-And then again six transcribed pages later: … You know, it's like I was saying, that hypocritical approach. You know, it was like when I was going to go and hate every white man out there, but it's okay because I'm into white women (laughs), you know.
At the end of the interview, as we leave the café he reiterates again (now off tape) that he has a preference for white women. This time Welsh is used directly as a prefix, whilst only implied during the interview. As we walk along the street, now dark because it is around 5p.m. and in winter, he invites me to go for a drink with him. I decline giving an excuse that I have to do more work. He asks me which hotel I am staying in and I respond suitably vaguely with something like 'in the city centre' and quickly make an exit into a clothing shop. This extract diverts attention from but does not invalidate his experiences of racism. It is used here to illustrate how preference is constructed. This exchange is not so much about me, but what he thinks will appeal to me and speaks to how he relies on particular constructions of racialized difference to do this. The reflections on this and other transcripts thus spark questions that are not restricted to, how does this make me feel and how does this affect the research relationship? It raises questions about the very articulation of sex and race and how the construction of preference produces black women in a pejorative way, precisely because they are rendered absent. These themes will be explored more fully in the remainder of the paper.

The trouble with the ideal 'feminine' researcher and the need for intersectionality
One of the dominant frames through which hierarchies in research has been addressed in the academy is through the emphasis on the need for open and 'equal' research interactions. As noted above, this has often, although not exclusively, been generated within feminist methodological debates. At the heart of this is the notion that good quality, or successful qualitative research, must entail an adequate level of rapport with our participants, which can be poorly interpreted as a very crude and simplified understanding of ethics as 'do no harm' (Mun Wong 1998). This translates into methods teaching that encourages an intimacy (usually ill-defined) with research participants as a marker of serious and rigorous research (see, for example, (Pitts and Miller-Day, 2007)). It is pertinent that this way of apportioning value to certain kinds of research relationships over others has been largely derived from thinking through the relationship between white heterosexual interviewers and interviewees (see, in particular, the early work of Oakley 1981), and this continues to steer key discussions on the gendered dynamics of qualitative research relationships and hierarchies (Berry et al., 2017;Birch et al., 2012). In focussing on white heterosexual relationships, this work has emboldened what was perhaps an already implicit assumption about the notion of commonality and common experience between women (Stanley and Wise 1990), through the over emphasis on the centrality of white 'feminine' heterosexual traits from which to build research relationships (see Harries 2016 andGreen et al., 1993 for critique of this). This is reaffirmed when the role of the researcher is constructed as one that mimics traditional notions of femininity. White feminists, for example, reassert this link when they argue that interviewing is an 'intrinsically female' practice (Smart 1984). These arguments not only resemble crude ways of understanding femininity and research but encourage the status of researcher to be aligned with the extent to which they can perform being 'acquiescent, attentive, and assenting' (Green et al., 1993: 630). Indeed, the very practice of asking questions as a woman risks appearing to consolidate this link when women, and especially marginalised women, can be perceived as naïve and less expert than white heterosexual men (Berry et al., 2017;Hill Collins 2000;Iknonen et al.).
The problems with dominant perceptions of good qualitative research are manifold because, I would assert that somewhat contradictorily, in addition to being 'feminine' and dutifully acquiescent, there are signs of a simultaneous expectation that researchers should be courageous and take risks to gain access to participants and get data. Qualitative researchers are, for instance, scrutinised and judged for the depth of the ethnographic encounter (Irwin 2006), which extends to that which has a 'voyeuristic and exoticizing quality to it' (Hanson and Richards 2017: 596). When placed next to each other these joint expectations present a set of obvious potential problems. I would suggest that by encouraging researchers to be both submissive and risk-taking reflects how the practice of women doing qualitative research continues to implicitly allude to the intimate, and even sexual or provocative nature of research which was made more explicit in earlier methods literature.
The growing body of literature detailing women's experiences of sexual harassment in research brings to light the conflict entailed in dealing with expectations of being a good researcher when intimacies are shaped by unwanted sexual advances. This becomes obvious when we read about those experiences. Most troubling within these accounts is the implied sense that sexualised interactions are a necessary part of getting 'good data' because we are trained to develop rapport and/or a sense of familiarity to produce more equal research relationships. Gurney (1985) describes, for example, how research methods training often problematically leads women to feel they must make 'necessary sacrifices' by putting up with sexual harassment in order to get sufficiently close to participants and subsequently be perceived as doing a good job. Hence, mainstream research methods teaching and practice, typically ignores both the implications of risk when building intimacy and the extent to which a researcher's embodiment in the field matters (Hanson and Richards 2017;King et al., 2020), and how this varies for different bodies. Berry et al. (2017: 539) however note that even non-mainstream activist anthropology can 'replicate that which they critique, by silencing the racialized, gendered researcher's embodied experience or by inscribing it in new colonial narratives.' There is a further contradiction implicated here because talking about experiences of sexual harassment risks losing reputation because women are so often interpellated as troublemakers or 'killjoys' (Ahmed, 2014;Whitley and Page, 2015), yet the ways in which qualitative research is taught suggests that we need to use our bodies and develop intimacies or 'rapport' in order to be adequate researchers (Mun Wong 1998). Similarly, Hanson and Richards (2017) note how researchers are taught to be reflexive but also feel pressure to do work in ways that will be valued by the academy and therefore do not, or are not encouraged to, apply reflexivity to their own experiences of sexual harassment. There is also a 'certain obligation to be "nice" when doing interviews' (Green et al., 1993;Gurney 1985) but in fact we do not have to like our participants (Davids 2014) and indeed it is worth bearing in mind that our participants can, and do, make comments that are racist and sexist (see Bott 2010;Mäkinen 2016). Hence, we need to consider the politics of intimacy in research contexts beyond the traditional comprehensions of power hierarchies. Developing rapport or close relationships with our participants is very often a performance that 'always has the potential to be exploitative' (Thwaites 2017: 1). Any sense of equality emerging from being 'nice' is therefore also largely performative (Ikonen and Ojala, 2007;Ramazanoglu and Holland, 2002). These aspects are vital to bear in mind as we witness the development of a range of innovative methods that attempt, if not to develop rapport, certainly attempt to create a research environment that appears more 'natural' and familiar to participants, such as walking interviews, discussions of familiar personal objects etc., and one in which co-production and participatory methods are also increasingly common (see, for example, O'Neill and Roberts, 2019). That is not to say that these methods are not valuable, only that closer consideration of how relationships develop is needed since their very nature entails the making and doing of some form of intimacy through the shared practice of doing research.
One of the key means through which we might challenge these problematic overgeneralisations of the ideal researcher type is by drawing on work that attempts to mediate hierarchies by paying attention to the differentiated positions of researchers. This work usefully alerts us to the way in which our research relationships need to be thought through intersectionally to consider how different bodies produce different hierarchies. Black feminist methodologies point to the importance of this, for example, when they highlight how the production of knowledge is always associated with a white position (Berry et al., 2017;Hill Collins 2000;Hill Collins and Bilge 2016). This tells us that it is necessary to recognise how one's position affects the dynamic of an individual interaction, but also that is vital to consider how power and knowledge continues to be regarded as always entwined with the body (Letherby, 2003(Letherby, , 2020 but that this varies for different bodies. The following discussion draws this idea out more fully.

Politics of intimacy and the process of othering
Creating intimacy or rapport is seen as a means to redress power imbalances in research but in mainstream research methods teaching there is rarely clarification of what this means. Nor is there recognition of the multiple forms of intimacy that can emerge. Whilst the practice of getting to know someone can be mutually positive, intimacies are never neutral. The doing and making of intimacies also tests boundaries and exposes hierarchies. Of interest here is the way in which the fieldwork encounter becomes a site in which the politics of intimacy exposes normative expectations and structures of inequality. The interactions detailed above, for example, illustrate how intimacies are established in a way that tell us about how gender, sexuality and race are reproduced and operate as well as how they shape the personal interaction. In other words, the salience of the sexual harassment being explored here goes beyond the personal encounter and intersects and engages with broader hierarchical structures to shape and produce particular kinds of knowledge.
Specifically, the following explores how processes of Othering are mobilised by participants as a means to cultivate imagined alliances/complicities. When participants express preference when making their propositions, they also divulge conceptions of difference and processes of Othering that reveal wider inequalities and discriminatory practices. Without paying attention to these instances of sexual harassment all this risk going unnoticed. Indeed, feminist ethnographers illuminate the potential usefulness and indeed the necessity of such a practice when they reflect on how their (albeit nonsexualised) relationships to their informants shape their own knowledge production and interpretation of the data (see, for example, (Buitelaar, 2014); Davids 2014; (Nencel, 2014;Schrijvers, 1993;van Stepele, 2014;Willemse, 2014)).
To illustrate how this works it is useful to reflect on how, in several instances, I was held up in comparison to other women. For example, we can clearly see from the propositions detailed above that heteronormativity is emphasised through their very articulation and, no doubt, there are numerous important things emerging here that we could draw out further. Here I focus on how research participants often attempted to develop a common alliance/complicity through views on ethnic minorities. The following thus highlights the value of drawing out the connection between processes of sexualisation, complicity and the process of Othering and the diminution of 'Others' that do not 'fit'.
Numerous examples illustrate how assumptions made about gender and sexuality are articulated through race and nationhood and become interwoven with assumptions about women who are constructed as 'different', or as 'Other' (see, for example, Hill Collins and Bilge 2016;Yuval-Davis 1993). In the interactions in Cardiff, this was reproduced at a local level and exacerbated through the understanding that white Welsh women are both fecund and subsequently are the 'root' of the community, a notion compounded by the ways in which Welshness is often articulated with working classness. 2 The men tell me that 'we' (white Welsh women) are understood to be the 'Celtic matriarch' of 'the community' and as such white (typically coded as 'local') Welsh women are proclaimed as the (re)producers of the local working-class community: These images of a 'matriarchal' community are somewhat perversely intended to be complimentary. They carry the broader message that women 'like me' are 'desirable', especially when coupled with their flirtations. But, crucially they also invoke a nostalgia for a romantic depiction of 'local' women that belongs to the past which are contrasted to the rupture of diversity and 'Othered' bodies that cannot fit this trope. The resonances of these romantic depictions linger. X above, after all, justified his desire to marry or have sex with me because I am 'local' 'Welsh' (white) and Y was more explicit when he explained his preference for 'local' women by emphasising that he had his children with white Welsh women. Local black and Asian women are in contrast described as not understanding these 'ways' of the 'old community. These comparisons, which also entail processes of Othering, are used as if they are complimentary and as if this can deepen an alliance between us based on a shared understanding of racialised hierarchies. If we do not allow ourselves to be distracted entirely by the problematic interaction itself then we can recognise how they also tell us something about the women who are not spoken of in the same way, but only appear as contrast.
The experiences of these 'Othered' women are produced within these attempts at gratification and the seeking of common ground. In Manchester, Z, a white community centre manager takes on a paternalistic/sexualising tone in our conversations. He imagines a complicity between us through our white-Britishness and this is rendered most evident when he begins to stare and evaluate a group of women that enter the room where we are speaking and is exacerbated when he invites me to join him in this exercise: Z: This is [Author] who is coming to do some research.
[Name] is from the Czech Republic and has come towell, you came looking for work; you came with your family. You are a fantastic volunteer who has improved your English a lot and so on and so forth [this is spoken in an affected accent to over enunciate words, miss out pronouns etc so she might 'understand'].
It is 'fun to speculate' he says. This invitation to shared voyeurism is made possible because of the relationship that he thinks we have established. We can see how in each of the cases detailed above, comparisons are made to women who are more marginalised than myself and who are constructed through dominant coded norms as 'Other'. We are both sexualised by Z but in different ways. We are all treated to the same paternalistic tone and unwanted touching but attempts to enable intimacy through shared voyeurism of the 'other' places me higher in his imagined hierarchy. The sexual proposition itself stands in for the construction of a set of positions that are established as a means to develop intimacies. Sexual propositions can thus tell us about our participants' ways of seeing, beyond the interview that can in turn shape knowledge production if we pay attention to it. Crucially then, the pressure to ignore these sexualised types of interactions means that we not only miss the opportunity to confront sexual harassment (if not in situ but in our methods teaching and practice) but also the opportunity to confront these processes of Othering on which they are founded.

Conclusion: Attending to the silences
The central aim of this paper was to highlight how the politics of intimacies produced in research interactions have implications for what and how knowledge is (re)produced and how the making and doing of intimacies engage with broader hierarchical structures of power, including structures of inequality. The paper examined how sexual harassment is often mediated through the making of imagined complicities that are used to imply that an alliance/compliance underpins the relationship and justifies the behaviour. The research interaction is viewed as a site in which the politics of intimacy exposes normative expectations and structures of inequality. Specifically, it exposed how processes of Othering are mobilised by participants as a means to cultivate imagined complicities and can form vital clues into wider social inequalities and discriminatory practices in which the research is situated.
The paper calls for a reappraisal of the focus placed on building rapport and/or a sense of familiarity in qualitative research, which are increasingly prescient in light of the renewed emphasis on co-production. The discussion raised a clear need to reassess the methodological literature that argues for the importance of building rapport and close relationships with our interviewees without defining what this means and how this is different for different bodies. This will be increasingly salient in the move towards coproductiona practice that entails closeness and intimacy through the shared practice of doing research. This need is driven by the problematic contradictions inherent in this way of advocating what is good research by virtue of the extent to which some level of intimacy is developed between researcher and participant. Indeed, this paper highlights how there is a clear tension present between the position that qualitative approaches should be based on openness, rapport and sharedness with participants and the risks and problems that this fails to take account of. Despite the slowly expanding body of work that examines sexual harassment in fieldwork it remains a largely overlooked concern. This discussion has highlighted how the omission partly stems from the various pressures placed on women to perform particular idealised roles in order to do good research. These idealised roles, often encouraged in research methods training and through the emphasis on the need for closeness and rapport with our participants, fail to account for how some bodies are more able to perform such roles but also do nothing to confront either their essentialised nature, or the risks of advocating for them. The discussion thus alerts us to the way these omissions are intertwined with how research is perceived and valued within the academy more broadly.

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/ or publication of this article.

Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Economic and Social Research Council (ES/K002198/1).

ORCID iD
Bethan Harries  https://orcid.org/0000-0003-1668-8391 Notes 1. Pseudonyms are ubiquitous and are usually used in order to give life to someone's story (Lahman et al., 2015). Pseudonyms are not used here to emphasise the universality of these experiences. 2. The significance to the understanding of the operation of gender and sexuality locally is discussed in full elsewhere (see Author). See also Mannay (2015) for a discussion on the articulation of class and Welshness.