‘The Eyes Have It’: Using Parliamentary Ethnography to Examine Gender in the UK House of Commons’ ‘Working Worlds’

Abstract

Methodological possibilities into exploring the reproduction of gender in parliaments, executives, and political parties are plural and exciting. Given that a performative approach to gender involves observing how gender is repeated over time; situates agency in institutional context; and constantly revises new questions and areas of analysis, parliamentary ethnography is a pertinent methodology to employ since it involves immersive methods. This chapter provides a standalone methodological chapter detailing what a parliamentary ethnography might involve and maps the ‘state of the art’ of parliamentary ethnography. The chapter discusses the logistics of the ethnography in terms of entry and access, participation, recording, elite interviews and documentary analysis. Finally, it maps the ‘field’ by introducing three ‘working worlds’ in the UK House of Commons—that is, the configuration of actors in the parliamentary workplace: MPs, the parliamentary administration and parliamentary researchers. The methodological chapter is combined with a methodological appendix which grounds this discussion in meta-theory, power relations and reflexivity.

Appendices

Appendix: Methodological Reflection, Ethnography

See Tables 3.7, 3.8, 3.9, 3.10, 3.11, 3.12, 3.13, 3.14, 3.15, 3.16, 3.17, 3.18, 3.19.

Table 3.7 Members’ staff interviewed parliamentary researchers

Table 3.8 Members of Parliament Interviewed Members of Parliament

Table 3.9 Members of the House Service Interviewed Members of the House Service

Table 3.10 Extra-commons participants interviewed

Table 3.11 General observation diary

Table 3.12 List of parliamentary researchers interviewed

Table 3.13 List of MPs interviewed

Table 3.14 House Service participants interviewed

Table 3.15 Extra-Westminster participants interviewed

Table 3.16 Illustrative interview questions for parliamentary researchers

Table 3.17 Illustrative interview questions for MPs

Table 3.18 Illustrative interview questions for participants from the House Service

Table 3.19 Indicative list of documents consulted

Having outlined the methodology in this chapter 3, this reflective note now outlines: (1) the meta-theoretical position taken, situating Butler and parliamentary ethnography; (2) the nuts and bolts of the specific methods outlined in this chapter 3; (3) the processes of analysis; and (4) criteria by which I assessed the quality of the interpretive research.

Butler and Parliaments—Interpretative Parliamentary Ethnography

Parliamentary ethnography has much to share with Butler; however, the decision to couple Butler with ethnography is not uncontentious and this dissonance is addressed here. There are a continua of research practices in ethnography. Some ethnographies are situated at the interpretivist end and some towards the more positivist end of social science epistemology. Butler aims at subjectless critique, echoing a similar of Foucault’s critique of humanist anthropology. However, I would also argue that she encourages us to acknowledge those actors who are ‘interminably spectral’ (2004, p. 34)—in this case unelected staff and researchers, in addition to MPs as the ‘presumptive centre’ of power. Ethnography allowed me to observe and partially represent the interactions between different working worlds in the UK House of Commons and the gendering therein and so ethnography was a powerful pairing with Butler.

In addition to the potentials for attending to spectral actors, the anti-foundational ontological position of gender performed every day in acts marries with my epistemological position—which is that a researcher’s vantage point needs to be as closest to ‘knowledge’ as possible, prioritising embeddedness, rather than detachment in order to produce fine-grained analyses. Therefore, ethnography is an important methodology for gender scholars if we accept that gender is iterated over time and that change and agency are identity contingent.

Writing up a parliamentary ethnography using Butler presents a clear issue about whether Butler’s verboseness can optimally speak to everyday issues since she ‘projects an aura of esotericism’ (Fraser 1995, p. 67). Indeed, in many respects, ‘subject position’ does not feel very human. I have argued in this book that far from being epiphenomenal or esoteric, identity-building practices, reputations and gender performance are actually very important and that naming these and providing a language to describe insidious power relations where there may otherwise not be a language (McNay 2014) frees energy for other productive pursuits. In particular, the logic of iterability and writing the body into parliaments as settings is important.

I assembled the fieldwork materials into a NVivo project. NVivo is a data management package; it allowed me to collate the data in one place, create analytic memos, and make links across documents. I could also run computer-aided searches to navigate terms. This was not a substitute for fine-grained analysis. Researchers often feel obliged to engage in a ‘quasi-industrial process’ of carefully ‘cleaning up’ unwieldy data to meet disciplinary standards when it is often unruly or:

sprawling, sticky, lumpy, impure and distinctly rough around the edges – not a substance that…[is]…easily handled, let alone neatly decanted into the stringently delimited and sterile confines of a theoretical framework. (Wilkinson 2013b, p. 389)

Furthermore, ethnographies of parliaments may aim at some type of political—in this case, feminist change, to bring about social justice. However, Butler’s Foucauldian conception of power has been questioned for its theoreticism and inability to create an actionable programme for change (Lowndes and Roberts 2013, p. 88; Nussbaum 1999; Hennessy 1993; Ebert 1992; Hartsock 1985, p. 38; Lather 1991, p. 8) and that ‘she prefers the sexy acts of parodic subversion to any lasting material or institutional change’ (Nussbaum 1999, p. 211, emphasis added). Butler stressed that programmatic assessments belie contingency and context (Butler and Bell 1999, pp. 166–167). However, it would be incorrect to suggest that Butlerian scholars have sought discursive change in and of itself.

Using Butler has consequences for the types of claims that are possible to make from the parliamentary ethnography. Many post-structuralists are critical of a representational politics with faithfulness to an ‘original’. Performativity is seen to be a non-representational theory; rather than the ‘what’, theorists of performativity are interested in the ‘how’. Butler is interested in what is at the limits of representation. Therefore, the book does not seek to provide ‘men’ and ‘women’ or distinct ‘styles’ but to open up some of the ways that gender was spoken about, performed, in relation to gendered/gendering rules. The empirical discussion and framework can be taken up, critiqued and reworked by others.

Linked to this politics of representation, uncovering the informal ‘can present difficulties of access and confidentiality’ (Chappell and Waylen 2013, p. 2) and so I followed observations through interviews to see if people would be willing to discuss them in an interview context, omitted them or found an open source such as the FOI request on the reduced to two posts rule-in-use. As mentioned, I guaranteed anonymity to participants. This is because I was vigilant that academic work can become ‘performative’ in terms of its re-citation.⁶ Furthermore, when including, I followed Denscombe’s two criteria towards the respectful inclusion of field notes that: (1) participants should not suffer as a consequence of inclusion, and (2) they should not be identifiable (2014, p. 218).

Reflexivity was practised throughout the research. Warren stresses: ‘It is not “any researcher” who produces a particular ethnography, it is you’ (1988, p. 65). Given the exposure to ‘flesh and blood’ elite actors, it would be remiss not to comment on this, given my ‘fleshy’ analytical framework. Whilst over-reflexive accounts may lead to ‘full and uncompromising self-reference’ (Davies 1999, p. 7), Wilkinson suggests ‘the researcher herself becomes a source of data that can contribute additional layers to the thick description that is gradually being developed’ (2013a, p. 132). It is ironic when interpretivist scholars fail to discuss their ‘own beliefs, subjectivities, and subject positions, and how these may impact on the research process’ (Chappell and Waylen 2013, p. 11).

A reflexive attitude was taken to inter-subjective power. Butler suggests that ‘we’re undone by each other. And if we’re not, we’re missing something’ (2004, p. 23). A researcher can undo a participant by asking an insensitive question. My academic identity was recited along a continuum of unthreatening to dangerous identities. I was aware that women can struggle to build authority as researchers:

A female researcher may not discuss the issue of gender in presenting her fieldwork experience for a variety of reasons…[She] may overlook or even deny difficulties she experienced in the field to avoid having her work appear unsound. (Gurney 1985, p. 44)

In terms of less threatening identities, being young(er), blonde-haired, unmarried, having a small build and pursuing gender research does not appear to be threatening. MPs discussed their daughters’ applications to universities; asked about the courses that I taught on; and asked about my experiences of being a woman in higher education. T’hart’s identity was an ‘egg head’ in the field (2007, p. 55). I was ‘dear’ed⁷ at in the field, even by ‘enlightened’ research participants. Police officers laughed amongst themselves after I missed a diversion: ‘we were going to see how far she got’. I was met with nurturing receptions when conducting interviews in constituency offices: ‘Watch those traffic lights, lov’⁸ One male researcher said ‘Cherry you’re just so nice!’ and another said ‘Little Cherry’ who wouldn’t let me stand my round in the researchers’ bar. This had connotations of being ‘naïve, blonde and bubbly’, a ‘poor student’. When asking one respondent about the presentational styles of MPs, she offered me some ‘gentle’ advice:

You want to present yourself as calm and capable but you’re doing this [slumps] and you’re doing this [bobs head] and I don’t know how old you are but it makes you look girlish⁹

This demonstrates that fieldwork at time, can be ‘hauntingly personal’ (Van Maanen et al. 1988a, p. 5), and a researcher can also be ‘undone’ positively and negatively by her participants (Butler 2004). I scratched my neck and an older man at a workshop with a press pass suggested ‘[i]n the old days, you would have had a noose around your neck for doing that’.¹⁰ I was also mistaken for a journalist rather than a PhD researcher: a subject position of being dangerous. This might be because by wearing old suits from a clearance store, my own descriptive marks and stylisation of the body were perhaps outside of the cycle of younger high street fashion. Generally, women tended to give more eye contact in interviews and they were more dialogical. One male researcher kept looking out of the window and gave the body signals of looking disinterested. Yet a male MP positioned me as expert as academic kindred-spirit: ‘people here would look very blankly at you if they tried to understand it. I can see you do. You’re the exception¹¹’. O’Reilly suggests ‘there is no escaping your own body and this will affect how you are received and how participants interact with you in the field’ (2012, p. 100). This was also notable in Skype interviews where there is no escaping your own body and indeed may be a more frequently used practice in political science interview research.¹² Brewer suggests that gender is the ‘primary identity’ in fieldwork and that female researchers can be treated as ‘gofers, mascots or surrogate granddaughters’ in the field (2000, p. 99). Whilst this is a valid point that women may be treated differently in the field, treating gender as a ‘primary’ identity ignores other aspects of identity such as class, race and sexuality (Butler 2011, p. 122). These positionalities may be affected by the parliament, for example, as a UK researcher in the UK parliament, accent is something that may be a qualitatively different marker than if I was conducting research in another country's parliament.

The tactic of flattering politicians (Crewe 2015b, p. 3) is possibly fruitful but also has gendered and power implications. This proved calamitous for myself when discussing mentoring schemes with a male MP. The MP questioned his mentoring skills and I replied with: ‘you can mentor me, then¹³’ and I panicked that this sounded sexualised. In interviews, I did, however, adopt active listening techniques to put myself in the respondent’s shoes, to be respectful of their experiences and to be a fellow traveller alongside their recollection of their experiences.

I was reassured to read that the phrase: ‘that’s interesting in itself’ (Rapport 2014, p. 60) has been used frequently by other ethnographers because it was a phrase that I often used, if a respondent presented a claim counterintuitive to (what they thought) I expected them to respond with and they thought I would be disappointed. Oakley argued for sisterly interviews (1981). However, I found that there were risks of ‘closure to this mutuality’ (Rapport 2014, p. 53). An example of where I could have managed the interview better is with a working-class participant where she discusses her presence on a committee and men speaking behind their hands asking: ‘where is she from?’. I replied:

CM::

Where are they from? [both laugh]¹⁴

I could have optimised the conversation to ask: ‘why do you think that they asked you that?’ However, on the other hand, empathy, and understanding, revealed that often there is a time-lag to acknowledge sexism.

CM::

Have you experienced sexism?

P::

Yes…I walk out of the meetings and I retrospectively think: “I should have been more assertive” and it’s difficult to be assured that it is sexism.

CM::

This idea of identifying sexism in retrospect is interesting, because…it has a longer timeframe than just the moment. Something similar happened to me yesterday!

P::

Exactly because you end up channelling the anger in the wrong direction.¹⁵

Despite epistemic risks of over-familiarity when studying parliaments as gendered workplaces, arguably, the researcher's second response resonates with Wulff’s comments surrounding collaboration in interviews:

the interviewer and the interviewee trigger each other into an exchange of escalating states of creativity beneficial for the interviewer’s research process as well as for the interviewee in the form of potential new personal or professional insights. When it works, this can be seen as a synergy situation, as the two people involved would not have reached these particular insights independently. (Wulff 2014, p. 163)

The weight that we should put on subjugated knowledge is a key concern for feminists. In choosing evidence to discuss within the book, I selected examples that were multifaceted enough to provide both content, but were also generative of new insights using analytical leverage.

Ethnography has been described as ‘arrogant enterprise, taking hit and run’ (Reinharz 1983, p. 80; see also Agar 1980, p. 41; Stacey 1988). Sustaining synergies in the field has to be balanced with an appreciation of field members’ time pressures as a functioning workplace. I set up synergies and provided information on points of contact and pieces of research that had been conducted in the areas of interest to respondents.

So far I have provided an optimistic account of parliamentary ethnography, but I would like to qualify what I have said thus far. All ethnographies are partial More eyes need to have it, so feminists do not turn a blind eye to intersections, power and those iterabilities that a white graduate female researcher might not capture or access. This is because subjects are multiply positioned. The work for this book was not intersectional, and therefore, a more systematic and in-depth intersectional critique must be enacted of everyday practice. In addition, ethnography needs institutional support such as funding and research leave; an awareness of gendered contingencies in practice; and in the history of ethnography. Conducting an ethnography clearly presents opportunities and challenges and was traditionally the preserve of ‘lone ranger’ male anthropologists (Kuklik in Wedeen 2009, p. 75). One must ‘cut your life down to the bone as much as you can afford to cut it down’ (Goffman 1989, p. 127). It is not surprising then that Park’s oft-cited instruction was to men: ‘[i]n short, gentlemen [sic], go get the seat of your pants dirty in real research’ (in Silverman 2011, p. 19) or that Ottenberg described his field notes as ‘an extension of me – like an extra penis’ (1990, p. 141). Goffman suggested that as well as ‘getting into place’, ethnographers have the imperative of ‘exploiting place’ (1989, p. 129). This has all sorts of ramifications for power relations. Furthermore, conflict methodology is ‘a willingness to utilize any and all situationally available techniques to gather data’ (Lundman and McFarlane 1976, p. 507). This instrumentalism negates personal safety and research ethics. Strategies have been developed for addressing sexual and gender harassment in the field (Warren 1988). Furthermore, elite ethnography has been described as: ‘a young person’s game. It requires a degree of physical stamina and psychological adaptability that, taken together, are optimized in people of their twenties and thirties more than in their forties and fifties (even though people in their forties do have the advantage of being closest to the average age of House Members’ (Fenno 1978, p. 253). Therefore, ethnography can require a wraparound commitment and research institutions could perhaps support researchers more in both practical and immaterial ways when undertaking fieldwork.

Now that I have reflected on Butler and parliamentary ethnography, I will discuss how generalisable my analysis is. In terms of applicability, the quality of qualitative research is not based upon generalisability (Punch 1998, p. 154). Bevir and Rhodes favour ‘philosophical rigour’ (2006b, p. 81). I have developed an empirical framework of the career cycle, citizenship and public service that can be used in different settings, but parliaments come in different types and stress different functions, so it needs to be combined with qualitative research. I also conducted a degree of purposive sampling across men and women and political parties as well as power and identity.

• (2) Participant characteristics

I will now discuss the meta-approach to recruitment to participate in the interviews from the three ‘worlds’ outlined above. Sampling approaches were relevant at different periods of the study. My sampling was theoretically consistent with my epistemological approach: that if we are to attend to power at the capillaries of the House of Commons, then we must attend to differently situated actors. Like Harvey (2011, p. 434), many of my interviewees referred me to contacts and so this precludes a more purposive sample. Requests were made on the basis of five broad areas. These were: (1) gender, (2) party, (3) power and positionality, (4) theory and (5) identity. Firstly, gender is not a synonym for women (Lovenduski 1996, p. 4). Ramazonoglu and Holland (2002, p. 5) study ‘gendered social lives’ rather than ‘women’. I moved back and forth, engaging with both men and women as subjects and beneficiaries of knowledge production by asking about their experiences. Secondly, there was a degree of quota sampling across parties where I spoke with actors in five political parties. Thirdly, I sampled across power and positionality. Moving away from methodological elitism and investigating power at the capillaries would have included staff on lower incomes, such as cleaners, but as an early career researcher, I did not have the profile to have carried this out with trust. I used the evidence submitted to the Governance of the House report, but acknowledge that many contractors and lower-paid staff may not have felt a sense of ownership or submitted evidence to the inquiry. I spoke to a union member who may have had insights about this group but this can also be problematic for making representative claims. To maximise positionality, if somebody had served on two or three select committees, then this provided positionality to examine the different dynamics. Fourthly, theoretical sampling is most congruent with the type of ethnography that I was conducting, to gain informed comment on aspects of gender identity. In terms of meeting ‘gendered actors working with the rules’ (Gains and Lowndes 2014), I interviewed those who had taken equality initiatives through the Commons—such as being involved in an APPG, taken a Bill through and been a member of a Workplace Equality Network as well as everyday actors. Fifthly, I sampled across identity and included ‘outliers’ by listening for atypical views from participants who I met in situ:

Be sure to include dissidents, cranks, deviants, marginals isolates –people with different points of view from the mainstream, people less committed to tranquillity and equilibrium in the setting. (Miles et al. 2014, p. 298)

Tables Nine to Twelve show the respondents spoken to. In order to protect the anonymity of interviewees, I have not included specific job titles.

The limitations of the sample arose from the serendipitous nature of interactions with participants and the snowball nature of recruitment. Most of my respondents were white and from the UK. There were two respondents from Black Asian Minority Ethnic (BAME)¹⁶ backgrounds, though requests were sent to try to have a more representative group and to analyse superordinate and intra-category intersections (Brown 2018). For MPs, I had a greater number of Opposition members, though this matched Emma Crewe’s sample and equal numbers of requests were sent. For the House Service, I had a larger representation from the DCCS department, which is because my initial explorations were on the select committee as a gendered scene—that had been less explored in academic literature. For parliamentary researchers, I had a large representation of gay, male government researchers, due to the snowball nature of the recruitment. Illustrative questions that were asked to participants are set out in Tables 3.16–3.18 in the appendix to this chapter.

2.1 Observations

This section discusses more specifically the observations. Participant observers start out fairly non-selective in terms of what they observe (Denscombe 2014, p. 207) and work towards a progressive focussing. Table 3.11 refers to the scope and planning of my research engagements. Macro-observations were large and in-discriminatory to get an ‘overall feel’ for the situation, as a ‘scene-setting’ device. Denscombe calls this ‘holistic observation’ (2014, p. 207). Macro-observation included chamber performances, spending time in the various outlets on the estate, accessing open sources of information, ‘unstructured soaking’ and general unobtrusive methods to map the organisational structure and hierarchy. The meso-level of the diary included pre-planned, more focused observations of parliamentary business announced on the UK Parliament website, Mark D’Arcy’s week ahead and Benedict Brogan’s former Telegraph blog. In terms of micro-level activity, I attended serendipitous events that I was invited to by the member, coffees, events advertised on parliamentary intranet, staff networks, and think tank events that were focussed towards the themes that were emerging in the research. Table 3.11 outlines a general observation diary.

2.2 (Elite) Interviews

The strategy to sampling was discussed above. Tables 3.12, 3.13, 3.14 and 3.15 provide details on the date and format of the interviews.

Tables 3.16, 3.17 and 3.18 should only be treated as examples of common questions that were asked in the interviews. I took interview guides, having researched the background for each participant, but the interview guide was used as a prompt to initiate dialogue rather than as unvarying. This means that not all interviews are directly comparable or quantifiable. This is because in practice, the ethnographic research was exploratory and followed issues that participants volunteered. I probed participants’ views, even regarding my own observations from the field. The questions were capillary to elicit ‘everyday’ information relevant to the context and background of the interviewee, and what was happening on the estate during the fieldwork.

2.3 Documentary Analysis

Table 3.19 outlines the different methodologies in these documents also need to be attended to. The soft methodologies involved in House of Commons interview projects were criticised as a ‘snapshot’ by one respondent,¹⁷ but this response was revealing. Overall, the combined observations, interviews and documents provided a rich set of information. This section has discussed three forms of generation that took place within the field. The next section discusses the quality of the generation of insights and issues of power and (inter) subjectivity.

To conclude, to conduct ethnography in the UK House of Commons was an immense privilege. Butler is interested in what is at the limits of representation. This is only a very partial study and from one set of eyes behind the spectacles. I firmly feel that ethnography is productive for knowledge and ideas to change institutions and would thoroughly recommend this to other scholars and hope that higher education institutions provide the resources to support the time-intensive placement.