Abstract
Keywords
Ethnography unlocks distinctive opportunities for researchers to re-think and re-evaluate long-established and existing research topics from different vantage points, permit new empirical knowledge and raise new research questions. In political science, this type of research has re-gained momentum and academic reflection in recent years (e.g. Wedeen, 2010), even if the methods associated with it – notably observation – have been pioneered a long time ago (e.g. Fenno, 1973, 1978). Scholarship on parliaments and legislatures has not been immune to this so-called ‘ethnographic turn’ (Brodkin, 2017) and wider methodological reflection (Bhattacharya et al., 2022a, 2022b; Crewe, 2021b). 1 For example, ethnographic studies have examined the Indian parliament and gender (Rai and Spary, 2019), everyday life in the European Parliament (EP) (Busby, 2013a, 2013b), hierarchies and relationships in the UK parliament (Crewe, 2005, 2015), African-American women’s experiences and political behaviour in US state legislatures (Brown, 2014), linguistic representation of constituency concerns in committees in the Parliament of the Republic of South Africa (Siebörger and Adendorff, 2015) and the role of hybrid professional identities in interpreting the Brazilian Congress (Brum Bernardes and Nunez, 2022).
Ethnographic studies on legislatures have made important and significant contributions to our understanding of political institutions. The above select examples also highlight a variety of analytical and empirical foci, theoretical positions and methods, and (inter-)disciplinary backgrounds of the authors. This raises the following interesting questions: What is the impact of making particular ethnographic research design decisions for studying political arenas such as parliaments? What are the trade-offs involved? We believe that these are important questions because, while ethnographic studies have become more common, there are still few detailed reflections about the benefits and pitfalls of ethnographic research when specifically applying it to high-level political institutions such as parliaments – and what pragmatic strategies are needed to maximise the leverage that such methods can bring. Parliaments are often distinctive, highly complex and multi-layered. The type and level of access to them will likely vary and, depending on what a researcher is able to negotiate, this will affect their research.
In this article, our purpose is to reflect on the diverse ways of practically doing ethnographic research, and the effects of different strategies, tools and research design decisions – to enhance discussions around the
To develop our argument, this article is based on our experiences of adopting ethnographic research on parliaments for over 10 years. We have different theoretical starting points, adopted different strategies and tools, and made different research design decisions. Our focus of analysis is around four key themes, to show that (1) parliaments offer different entry points for ethnographic research, (2) fieldwork needs to be adapted to organisational idiosyncracies, (3) there are particular strategies and tools to build relationships and to navigate obstacles – what we call our ‘ethnographic capital’ and (4) the choices do not end after we leave the research settings. These four themes form the main heart of the article. Before we examine them, we summarise what we mean by ethnography, why we do it and what it contributes, and a second section that gives greater context about our fieldwork experiences.
Summarising the case for interpretive ethnography
Ethnographic research has a long history, deep inter-disciplinary roots and diverse epistemological traditions (Aunger, 1995; Schatz, 2009a; Wedeen, 2010). While there is no consensus in defining or practising ethnography, most agree that it involves immersion in a social and/or political world of some kind to understand the lives of people under study. This can include: Living in the community being examined; learning a local language or dialect; participating in the daily life of the community through ordinary conversations and interaction; observing events (meetings, ceremonies, rituals, elections, protests); examining gossip, jokes, and other informal speech acts for their underlying assumptions; recording data in field notes that attempt to produce daily accounts of social and political life; in other ways letting trust and emotional engagement be of benefit to the research project (Bayard de Volo and Schatz, 2004: 267).
Some, especially ‘classical’ ethnographers, argue that these activities, and especially physical immersion and participant observation, are an essential and necessary part of ethnography (Malinowski, 2014[1922]). However, we follow a second, broader understanding of ethnography as a particular way of studying social and political worlds that focuses on and examines everyday lived realities of people under study to shed light on wider social and political practices, relationships and webs of meaning (Schatz, 2009b: 5–10). Physical immersion and observation (through participation or from a distance) are two particularly useful means to do this type of research. But, for us, ethnography is a ‘sensibility’ (Pader, 2014; Simmons and Rush Smith, 2017), where what matters is not necessarily the specific method used to gather data but how scholars analyse that material by paying attention to social and political meanings embedded within them.
Why focus on everyday practices, informal norms, beliefs and values in the first place? Following a long line of interpretive-inspired research across philosophy, the arts and humanities, and social sciences (e.g. Finlayson et al., 2004; Yanow and Schwartz-Shea, 2014), we argue that drawing attention to these factors holds significant explanatory value. First, we believe that social and political worlds are socially and politically constructed, requiring actors to interpret those worlds to create meaning (Berger and Luckmann, 1967). Second, meanings – for example, around gender equality or parliamentary activities – are socially embedded and inter-subjectively evaluated and assessed, requiring a relational focus between actors and their physical and social environments. And third, beliefs and practices are historically situated and entangled in power relationships. 2 Taken together, they point to a need to unpack politics as the contingent beliefs and values of people under study, embedded in wider webs of meaning and how these may inform and are shaped by institutional arrangements.
Ethnography has much to offer to interpretive social science. Conceptually, it allows us to look at social and political issues from new vantage points. We can ‘see’ the same research topic from different angles, spurring on new research questions, identifying previously unnoticed factors and so on (Rhodes et al., 2007a). Empirically, our new vantage points provide opportunities to give depth, nuance and texture to studies conducted in other research traditions, potentially even challenging existing scholarship or generalisations (Schatz, 2009b: 10). Methodologically, it invites critical reflection about the strategies and tools that we employ, and about the relationships between (and positionality of) the researcher and the researched. And normatively, ethnographic research is often motivated by foregrounding the voices of those that have otherwise been marginalised, to address power imbalances, or improve political life in some way.
Our arguments about the contributions of interpretive ethnography are not new. Nevertheless, and despite the wide-ranging methodological reflections on ethnography in general, and empirical applications to parliaments in particular, the implications of and practical questions around choosing ethnographic strategies and tools remain less well-understood. To highlight the main question in the title of our article, we now shift the focus of discussion from the inherent value of interpreting parliaments to asking how different strategies, tools and research design decisions affect ethnographic research.
Advancing ethnographic studies of parliaments: Our questions and data
We draw on our experiences, as summarised in Table 1, to identify comparative insights for conducting parliamentary ethnographies (relevant also to other political institutions). We are well suited to this given that we have directly engaged with different types of parliaments and we have made several different research design decisions and improvisations to gain access and work with actors. This allows us to compare and contrast strategies and experiences inter-subjectively (Folkes, 2022), especially the different trade-offs we made. That said, we acknowledge that our experiences are partial: the parliaments we have studied are established and institutionalised representative bodies in western Europe. Also, our experiences are embedded in political and higher education traditions that might be less applicable, or conceived differently, in other contexts (and we would welcome further contributions to explore such differences, or indeed similarities, in future). Nonetheless, we believe that our experiences have broader relevance given shared and recurring concerns throughout not only in our fieldwork but also in accounts of other parliamentary ethnographies that we have read and studied.
Ethnographic approaches and methods.
Geddes has worked on several parliaments. First, he conducted two Westminster-focused projects, specifically (1) on accountability in and by parliamentary committees, which included a 3-month placement in the UK House of Commons with a committee and over 45 formal semi-structured interviews (Geddes, 2020) and (2) on the role and use of evidence by committees, through a 12-month Parliamentary Academic Fellowship in which he spent 4 months working in Westminster as inquiry (co-)manager for a committee, and undertook 50 formal semi-structured interviews (Geddes, 2023). Second, though not a predefined ethnographic study, Geddes worked with the UK’s four legislatures for comparative research to examine knowledge exchange strategies by those legislatures and parliamentary actors’ perspectives (Beswick and Geddes, 2020). Finally, he has conducted ethnographic research on Germany’s federal parliament (the Bundestag), working as a visiting fellow at a research institute near the Regierungsviertel – governing quarter – of Berlin. Most of this last study took place at a distance without institutional access, focusing on interviewing actors, but eventually he was able to observe Members of Parliament (MPs)
Miller has worked on two legislatures, the UK House of Commons (Miller, 2021, 2022, 2023) and the EP (Miller, 2022, 2023). For the UK project, she formally spent 18 weeks there with access provided by a backbench, opposition MP who she shadowed while conducting 68 semi-structured interviews. This study explored how gender was reproduced ‘everyday’ in the UK House of Commons. For the EP study, she worked as a post-doctoral research fellow on a large collaborative project to analyse how gender was reproduced in teh EP’s political groups. She accessed the parliament via a 2-month academic placement with the EP’s Research Service. She accessed the groups via individual Member of the European Parliament (MEP) offices, making by-appointment requests to attend meetings, such as political group meetings or working group meetings. She also shadowed (Mikkonen and Miller, 2024) nine MEPs across five political groups for between half a day to 2 days. Over 140 interviews were conducted for the study in total. Her current research has followed gendered EU-UK parliamentary diplomacy through interviews, video analysis and field visits.
Based on these summaries alone, we can identify similarities and differences. We undertook our first fieldwork projects at a similar time and at the same institution: during the 2010 parliament of the UK House of Commons. However, our access differed significantly: for Geddes’ study of accountability by select committees, he negotiated an internship (via his PhD supervisor) to provide research assistance to a committee; for Miller’s study of gender in Parliament, she negotiated access (individually) within an MP’s office. Our subsequent research took us in different directions and places. In sum, our means of accessing parliaments included (1) working for a parliamentary committee, (2) as an assistant in a representative’s office, (3) through a placement via research services, (4) periods of shadowing representatives and (5) observing from a distance. These contrasting vantage points gave us opportunities to study everyday practices in parliaments. For all projects, we undertook formal, semi-structured interviews as well as countless informal conversations with MPs, permanent parliamentary officials and researchers, MPs’ staff, political group staff, political and parliamentary journalists, and service-sector staff (e.g. cleaners, bar staff, etc.).
On the basis of wider scholarship and our experiences, we identify and discuss four recurring themes based on our fieldnotes and themes raised in the existing literature: (1) negotiating access and consent, (2) organising fieldwork in political institutions, (3) reflecting on ethnographic capital and (4) ‘post-fieldwork’ considerations through analysing and publishing findings.
Negotiating access
Our starting point concerns how we enter parliaments and, more specifically, to highlight the difference between entry and access. In whatever way a researcher may enter an organisation or other research setting, this does not guarantee access or consent within that site. As Gusterson (1995: 191) memorably articulated in his study of laboratories and nuclear weapons, personnel were ‘polite’ and ‘friendly’ but also ‘profoundly unhelpful’ because – despite securing
Entry versus access
In our research, access and consent were constantly re-negotiated, regardless of ethical clearance by university processes or written agreements we had secured. We do not take organisational consent or entry for granted, but it is often not enough and very different when faced with the practical realities of ‘doing’ research. For instance, a formal agreement by administrative staff to study committees for Geddes required additional persuasion of individual committees (and their MPs and officials) to permit access to private meetings (not all of whom agreed). To give another example, one of us gained access to a meeting through a parliamentarian (member) of an All-Party Parliamentary Group (APPG). The APPG chair – unbeknown to us, had not been informed, so we accounted for our presence through fraught questioning, in front of APPG members, academic experts and witnesses. The setting for this encounter – a grand committee room in the Palace, with dark, wooden panels and oil paintings was remarkable. To give a final example, our attempt at observation through being embedded in a team was significantly curtailed within two weeks because – despite ethical clearance from both university and parliament – it became clear that members of the team were not fully briefed and were uncomfortable with being observed. In short: entry is not the same as access.
An alternative route to joining particular teams, units or groups is being hosted directly with parliamentarians. This also created challenges. The levels of autonomy that Miller was afforded were opaque and double-edged. On one hand, Miller was provided with a pass from a busy MEP and told ‘to make [her] own way’ around the parliament. But, conversely, this was not carte blanche access. Pass sponsors cannot guarantee informed consent all of the time – not least of their researchers who are in an employment (power) relationship, or adjacent parliamentarians’ offices with shared staff workspaces. In turn, we were initially cautious and leant on auto-ethnographic styles and formal interviews. Nevertheless, being situated with a parliamentarian afforded access that working with the parliamentary administration could not give. For example, it gave direct access to MPs and parliamentary party groups (PPGs) working in real time to negotiate positions or work out strategies for upcoming votes, with party discipline in mind. Access therefore makes a key difference, with trade-offs involved. In the case of the parliamentary administration, you are more likely to be able to observe the planning and organisation ahead of political battles; in the case of an MP or PPG, you will see those political battles happening. To overcome possible limitations and obstacles from situated locations, others have recommended ‘polymorphous engagement’ (Gusterson, 1997: 116) as a strategy. This requires collecting data from different (parliamentary) groups, teams, settings and participants, and in a variety of formats. This enables researchers to partially account for different views and institutional locations in the parliament.
Blurring our status: Inside out and outside in
In the end, regardless of how we ‘got in’, an immediate new challenge is how we were then interpreted: as insider or outsider to the organisation. The impact of insider/outsider dynamics – and even the terminology – has been discussed in the literature (Brown, 2018; Schatz, 2009b: 6–9). What counts is not just your formal status, but your relations of power with other actors (reiterating the important distinction between entry and access), self-perceptions and the perceptions of others. 3 For his UK fieldwork, Geddes felt he was an insider with committees and integrated into respective teams (and was treated as such by staff), but an outsider for MPs and PPGs. Miller was an insider for her MP and, arguably, the political party she was situated within. But she was an outsider elsewhere. As Busby (2013b: 71) explains: ‘By positioning myself with(in) one group, I immediately closed off access to others’. In our case, while Miller was introduced to one MP’s network and their colleagues, the party label created difficulty in gaining a broader cross-section of interview participants (which can be ameliorated through effective – and lucky – snowballing). This does not necessarily need to be related to parties. In the UK House of Commons, being associated with the Speaker might be interpreted positively and open certain doors, but can also create difficulties in seeking to speak with the Speaker’s opponents. Such fortunes can change over time, too; for example, the Speaker’s authority might wane as their tenure extends until it might be counter-productive to rely on them for access to the organisation (see Childs, forthcoming).
There are, of course, strategies to gain access to other units or groups as part of research (e.g. noting to them that their voice is missing from a research project that all others have contributed towards), as well as trade-offs and obstacles. Miller engaged with similarly aged parliamentary researchers – the gatekeepers for ‘their’ elected representative – with the ability to prioritise invites. However, one assistant once declined Miller’s invitation to her MP to participate in an interview, noting further that Miller had used the Member’s title in the wrong order – important in the UK context. Ultimately, access was precarious – knowing that you can be asked to identify yourself or the parliamentary actor who had granted access – or at worst, be removed from a space, shows how power is felt in an anticipatory way.
Snowballing can be a useful technique to overcome challenges of access, but this technique could get complicated. For example, not all snowballing is ethical. In several cases we were given the private phone numbers of actors (journalists or MPs) to contact directly, even though they had not given their permission to share the contact information and might not want to speak with us. To give another example, a tense encounter occurred when a male MEP, despite reservations from Miller, unsuccessfully snowballed an interview with a female MEP who had already declined to participate in the research. The MEP declined, explaining that she was ambivalent about ‘this MeToo, MeBoo thing’, enjoyed receiving wolf-whistles and had felt personally embattled by feminist invitations to publicly debate her. Trade-offs are, therefore, inherent in snowballing processes: they coexist with elite control, defensiveness and partisanship. Furthermore, some parliamentary actors might not have access to networks, so a snowball approach might reproduce existing inequalities or create biases.
The ambiguity of being on the inside can have benefits but also brings challenges. For Geddes, being part of the parliamentary administration meant balancing a line of separation with colleagues. He was invited to social events and was assumed to be just another member of the team. Both of us had to negotiate studying these organisations while, in the process, building closer and personal connections with staff, sympathising with their problems and challenges and so on. This closeness, of course, built trust – but also created a challenge in how to navigate friendships while still studying the organisation and the people that worked within them. There is a trade-off, then – acute in ethnographic research – about how closely and personally entangled a researcher can become, while also gaining access. Anonymity, triangulation and a variety of other techniques became ever more important to manage tensions, as well as being explicit and clear with potential research participants about clear boundaries and continuous, informed consent when asking for their participation in interviews.
So far, we have summarised our experiences as they relate to the consequences of gaining entry. However, we might not get so far to enter at all. This was especially pronounced for Geddes’ early fieldwork attempts on the Bundestag. Based with the Institute for Parliamentary Research meant that he was only 30 minutes away on foot from the parliament – but this still brought significant physical and cognitive distance. There was never any doubt that Geddes was not on the ‘inside’. For much of the time, the parliament felt distant, opaque and hard to understand. It was only through follow-up visits – where he was able to shadow two MPs over two separate sitting weeks – that he got a foothold onto the estate. This took time, which is not uncommon in ethnographic research and often relies on key informants or contacts. Building contacts and trust, and then being able to conduct fieldwork, requires time. Sometimes it can take several months, if not years, to organise fieldwork. For political scientists, in particular, this is not always feasible given the disciplinary traditions of how many of us conduct research.
Implications: Access as a continuous conversation
There is no ‘right’ or ‘wrong’ way of entering the field and negotiating access, but being clear about the trade-offs and subsequent research outcomes is important. While this applies to all ethnographic research, there are specific implications for parliamentary research. For example, being part of the parliamentary administration gives a sense of being on the inside and being a part of the institution. It gave Geddes a particular, institutional-level and organisational-level view of activities. Meanwhile, Miller’s access via an MP’s office enabled her to interpret the everyday working lives and trade-offs between different demands on MPs more clearly. We do not know, of course, how our findings might be different if we had undertaken the same studies but with opposing access, but it is reasonable to assume that Geddes would not have a chapter on ‘hidden servants’ (Geddes, 2020: 77–97) but might instead have one on an MP’s office; Miller, meanwhile, might have emphasised the gendered organisational dynamics of the parliamentary administration in greater depth.
Ultimately, we both strategised and improvised with the access we had. We agree with Wedeen that ‘there is never nothing going on’ (2010: 256) in the spaces we
There were fraught moments, such as the aforementioned APPG moment, where being an ‘outsider’ was compounded by physical, topical and political dynamics. But it was also these ethnographic moments that helped us to interpret the confluence of powers in the field and to unpack events and relationships, such as the importance that actors attach to anonymity and confidential spaces. Being reflexive matters, vis-à-vis the aims of the study, as well as the type of parliament and how its functions are organised (Miller, 2022). We cannot cover the infinite possibilities of different access arrangements in political institutions but we do note that access arrangements are foundational features that form the basis: your access shapes relationships, the positionality and capital that you gain or lose, and how to present research findings.
Organising fieldwork
The multifaceted nature of political organisations, especially parliaments, can affect the type, level and (dis)continuity of access for ethnographic research. Here, we discuss how these very particularities affect the way that ethnographers relate to organisations and its actors, especially parliaments’ (1) simultaneously open and opaque nature and (2) distinctive organisational rhythms.
Semi-public? Semi-closed?
Parliaments are, in principle, semi-public organisations, with estates open to a significant array of actors beyond MPs, their (and their party’s) staff, and officials from the parliamentary administration. Other actors include political journalists; representatives from interest groups and lobby organisations; construction workers or IT staff to fix physical or technological issues; school groups and members of the public; interns to support the work of an MP; service-sector staff, such as cleaners; or perhaps even ethnographers conducting fieldwork. It is not usually easy to distinguish between actors, and so there is often an assumption that people regulate their behaviour because they are ‘on show’, on the ‘front stage’, with every act and gesture potentially being scanned for meaning (Geddes, 2020; Goffman, 1990[1956]). This means that parliamentary actors are likely de-sensitised to observation; they are used to it. It also enables anonymous observation with relatively low ethical risks, provided that anonymity is respected. But, MPs and officials are consequently also likely to enforce distinctions more stringently between insiders and outsiders, and to especially welcome private and confidential spaces – where observation is difficult. Furthermore, what is considered secretive or sensitive may vary by the type of parliaments, or even the political group within the parliament.
The semi-openness of parliaments has several impacts on fieldwork and can be paradoxical (Leston-Bandeira, forthcoming; cf. Teixeira et al., 2020). In one sense, semi-open spaces can be fertile ground for observation. In the UK’s parliament, the atrium of Portcullis House is a perfect example. A large and open, glass-roofed space with lots of venues at its edges (cafes, restaurants, a post office and other services) and lots of tables for meetings (over coffee, lunch, etc.) means that it is an actively used space. We could observe who is meeting who, who wants to be seen with who, and who avoids this place. Other spaces might be used in other ways; informal settings can be used to exchange information or mobilise political support (Norton, 2019; Psarra et al., 2023). Geddes once accidentally interrupted a meeting in a corridor between government-side MPs that were plotting a rebellion (as confirmed in the next day’s papers). Meanwhile, Miller observed MPs engaged in intimate conversations with journalists between pillars of a colonnade in New Palace Yard. The built environment contributed to our interpretations of the atmosphere and intrigue of these exchanges.
Spaces can also be exclusive. In the context of the Indian Parliament, Rai and Spary (2019: 26) noted ‘how the parliamentary space is being securitized as we tried to take photographs’. Simply being around parliamentary infrastructures and inside their spaces can therefore be insightful to interpret how spaces are regulated, which bodies are permitted where and why (see also Puwar, 2004). Indeed, space can also be used strategically, or it might be simply less functional for observing meetings. Access could be denied because of the location or type of space, explaining that, especially in historic, less adaptable, buildings, there is ‘no room’ for an observer even if in principle they weren’t against it. ‘Observing at a distance’ occurred in Miller’s recent research, watching proceedings from a ‘listening room’. In cases where researchers have no physical access, parliaments can feel especially closed. For Geddes’ fieldwork of the German Bundestag, the lack of a pass or continuous access created a sense of ‘glass walls’: glassy infrastructure that ostensibly invited people into the parliament, but one where you can only observe (often with prior approval) without necessarily engaging or experiencing in the life of the parliament. Indeed, even in ‘open’ buildings, such as the Scottish Parliament, light, metaphorically, coincides with shadows. 4
Parliamentary spaces can be public – as can their work. Unlike other elite organisations (Monahan and Fisher, 2015: 715), this can present opportunities because much data can be gleaned online to help with recruitment and interpretation, such as public biographies of MPs, use of Twitter, voting records and more (some parliamentary websites are more usable than others). We used such information to call and email offices for interview requests, but also used the official cataloguing of their behaviour to observe them at a distance – something that is not necessarily possible for other elite ethnographies (and not necessarily without its problems; see Worthy and Langehennig, 2022). It also enables remote ethnographic research, which Geddes used for his German fieldwork, especially watching parliamentary proceedings and committee hearings (though we also acknowledge that not all work is public; for example, German committees deliberate in private; on opening the ‘black box’ of committees, see Siefken and Rommetvedt, 2021). While observing committee proceedings online has the benefits of replaying exchanges, watching online in real time allows researchers to additionally share field members’ anticipation (though time zone differences may complicate this endeavour). 5
Organisational rhythms
Each legislature has unique organisational features, which can have a critical impact on how everyday parliamentary life – and hence fieldwork – is organised. One basic example is the parliamentary timetable. In the UK House of Commons, time is often structured through a particular weekly rhythm in which most business takes place between Monday afternoons until Thursday afternoons; Monday mornings and Fridays tend to be quieter with MPs in constituencies and – since COVID-19 – many staff working more flexibly from home. Meanwhile, the Bundestag is organised into stricter sitting and non-sitting weeks that usually alternate across the year; in the former, MPs and staff will be on the parliamentary estate throughout the week with set days for particular meetings (working groups meet on Tuesdays, committees usually on Wednesday, etc.). In non-sitting weeks, MPs are in their constituencies. And to give a final example, Brown (2014: 193–194) explains how the Maryland legislature comprises ‘part-time representatives who dedicate an annual ninety-day period to law making . . . [and that] Maryland’s short legislative session requires a structure that facilitates law making at a relatively quick pace’. Not only do elected members have particular routines or rhythms, so do parliamentary staff, who will be affected by those timetables but also have annual strategies to work towards and annual events to plan (e.g. a State Opening of Parliament (Johnson et al., 2014)).
Rhythms are not random, nor are they inconsequential. They will permit access in different ways and at different times, which affect how fieldwork takes place. Importantly, ethnographic research allows researchers to experience the rhythm as the actors do; it is their lived experience in which we can participate, understanding the pressures of the mid-week and the calmer and quieter days. Both of us adapted to the working weeks, allowing us to better understand how MPs and officials prioritised their time and to observe coping strategies, especially enhanced through sustained shadowing of MPs over separate sitting weeks in Geddes’ case. For Miller’s research on the EP, ‘group week’ was key to study PPGs. These groups had two types of leaderships: political groups leaderships and administrative leaderships. In one group, Miller held long meetings with, and produced documentation for, a powerful secretary general and their assistant who promised to put the question of access to group leaders. This never materialised. In another case, a different secretary general helped considerably with access. Geddes faced similar situations in his fieldwork. Miller organised fieldwork around MEPs who liked to attend fitness classes or an MEP’s weekly walk to the Brussels Midi Eurostar departures. These presented important, ethnographic moments, allowing Miller to unpack how actors negotiated their priorities around institutional norms and rules; meanwhile reminding us that we are at the mercy of the rhythms of individual actors and how they navigate parliaments.
Rhythms are also affected by physical infrastructure and technological advances. The EP is distinctive in this regard in being a multi-sited parliament. Miller had to arrange a trip to Strasbourg as well as the Brussels location. Indeed, even within a parliament the ‘office politics’ can matter. For Miller, the Brexit Party was located into offices outside the main Alterio Spinelli building (the main building of parliamentary offices). This entailed often windy walks in the rain to interview Brexit Party MEPs – but also created its own subculture and a sense of community for these ‘outsider’ MEPs. In general, Miller found Strasbourg to be a rabbit warren, in which political groups would disappear into their private meeting rooms, creating – as mentioned previously – glass walls between herself and her parliament. They sought alternative opportunities for immersion, such as attending the group leaders’ press conferences and informal exchanges with interpreters. However, a stronger strategy to have mitigated Strasbourg’s ‘glass walls’ may have been employing shadowing techniques in this parliamentary location (Busby, 2013b: 61).
The way that spaces are used is also affected by technological changes, especially the growth in use of messaging apps and meetings conducted online. The COVID-19 pandemic accelerated such trends significantly and affected both of our fieldwork in terms of when and where research could take place. The long-term effects of such flexible working are not yet known but, based on our embedded fieldwork, we can say that it has led to fewer spontaneous connections and meetings, and far less traffic on the parliamentary estate. 6
Implications: Adapting to parliaments as organisations
Ethnographers need to build time for entry, get used to the field and adapt to organisational routines. This makes sense across ethnographic research. But there are specific practical considerations to undertake a parliamentary-focused ethnography. Specifically, our point is that you can read all sorts of summaries of how parliaments work in textbooks, whether it is committee structures or the legislative process, but nothing quite prepares you for experiencing the rhythms of parliamentary life until you are there. Understanding and engaging with the physical infrastructure, in particular, can raise issues about where one can go and what is restricted. This then might lead to questions about what is open, what can be studied, where and under what circumstances. Crucially, of course, these adaptations need to be made by researchers themselves and depend on the topic under investigation.
We are aware that we have been fortunate with regard to this aspect of research – often thanks to our ‘capital’, which we develop in the following section. Not only did we have institutional positions and responsibilities that gave us time for fieldwork, but we were also able to invest our personal time and resources into fieldwork that others may not be able to do. Without caring responsibilities, we could behave flexibly and shape our daily lives to the needs of parliaments. We were mobile and travelled internationally to do our research. Others are not so lucky, which – given that social care predominantly still falls on women – is a gendered issue. Some parliaments (attempt to) align their recesses with school holidays, such as the Scottish Parliament; others do not. Researchers with disabilities may also encounter challenges at each stage of the research, such as planning the support needed during funding, to research interactions in the field (Brown and Boardman, 2011; Schwartz-Shea and Yanow, 2012: 118). Practical details about the fieldwork may be asked well in advance, but the precise dates of entry to the parliament and activities involved may not always be predictable.
Reflecting on our ethnographic capital in parliaments
A significant, and widely acknowledged, theme in interpretive ethnographic research is ‘positionality’, that is, being reflexive about the impact of the researcher’s identity and their position on fieldwork, research participants and the field itself. These are often part and parcel of extended ethnographic writing (in our research, see: Miller, 2021: 84–6; Geddes, 2020: 149–55) and in ethnographic methodological literature (e.g. Davies, 2008). However, we follow Reyes (2020) in arguing that we need to acknowledge how we actively draw on both our visible and invisible traits, as well as how they affect research. Reyes has adopted the concept of an ‘ethnographic toolkit’ from which researchers draw; elsewhere, Cunliffe and Alcadipani (2016: 241) identify ‘reputational capital’ to argue that different types of capital may facilitate particular kinds of access. We build on these interventions to argue that ethnographic research is strongly shaped by ‘ethnographic capital’ (we prefer ‘capital’ over ‘toolkit’ because it better emphasises that our identities are embedded in social worlds rather than ‘tools’ we can pick up). 7 This has particular implications for a parliamentary setting.
Institutional resources and roles in research
Institutional affiliations and their impacts can be significant. At the start of our careers, we were both PhD students from Russell Group universities in the UK (a group of research-intensive universities often perceived as prestigious). We moved on to conduct research through other opportunities: as post-doctoral researcher for highly prized European Research Council grants, or as Parliamentary Academic Fellow as part of the UK Parliament’s prestigious fellowship scheme. These arguably gave us significant capital that we used to our advantage. Aside from direct financial resources, we were able to mention grants and institutions in our covering letters and consent forms. Interviewees noted that they accepted interview requests because of the geographical location of the institution (e.g. a university based in their constituency), being a university alumni or interest in the grant funder’s work. Institutions also have other reputational advantages. For example, being affiliated with a Centre for European Studies undoubtedly shaped perspectives of Miller’s recent research on EU27-UK parliamentary diplomacy.
Returning to our first empirical section about access, we were also able to use institutional resources to blur the distinction between the ‘inside’ and ‘outside’ by being associated directly with the parliaments we were studying – as MPs’ staff, as parliamentary fellows, as committee assistants and so on. This was important ethnographic capital for several reasons, and which highlights a contrast between different types of access. First, through an institutionalised fellowship with the House of Commons (organised by the Parliamentary Office for Science and Technology), Geddes could rely on a formal fellowship agreement by the parliamentary administration to reassure participants and persuade actors to contribute to interviews. The research would, Geddes promised, feed into recommendations to increase the UK Parliament’s effectiveness. Miller was fortunate that the MP who provided access snowballed interviews to her immediate circle. However, achieving cooperation across the political spectrum required more creative ways of working (reminding us that political associations can increase but also diminish capital), such as working with alumni from her university (see also Brown, 2018). Relatedly – and second – an important symbolic change was to use parliamentary email accounts as opposed to university or personal ones. We were potentially seen as part of the parliamentary infrastructure and meant that our emails were not filtered as for ‘outsiders’. Without any direct institutional capital, access can be more difficult – as Geddes found with respect to studying the Bundestag where positive interview responses were more difficult to come by.
Third, the nature of the relationship with parliamentary organisations can afford protections or resources. In his UK-based fieldwork, Geddes felt that he was treated as being in the same category as other staff in the eyes of parliamentarians, and protected by the House of Commons Service through its contractual arrangements. By contrast, Miller’s parliamentary pass was contingent on an MP’s sponsorship. Furthermore, at the time of her research, parliamentary researchers were in a vulnerable position with significant power imbalances that had been subject to media controversies and investigations regarding sexual harassment and bullying (Cox, 2018; White, 2019). Such protections were introduced post-fieldwork and were strengthened for both categories of staff (the Independent Complaints and Grievance Scheme now underpinned by a behavioural code that also notably covers visitors), but demonstrate that relationships of power are significant considerations for political research. This echoes our earlier point about trade-offs and different ways of entering and accessing institutions – in our case, it placed us in different relationships vis-à-vis the institutions under study.
Our ability to take advantage of institutional affiliations was further mediated by institutional positions and roles. As PhD students, we did not have the prestige – in terms of reputation or resources – that a professor might. This had a direct impact. For example, Geddes was asked by the chair of a committee to ‘fetch the tea’ for his meeting, noting that this was ‘an intern’s job’ (after the meeting, parliamentary staff spoke of their fury); as parliamentary fellow, MPs showed far more deference. Nevertheless, we strategically used our early-career status to our advantage: we were able to get away with ‘basic’ questions to MPs and officials, obscuring naïve points. This was no doubt related to our visible traits, notably our age.
Identities and traits
Ethnographic capital is affected by physical/audible traits such as gender, class, ethnicity, citizenship, accents and age (Sarikakis, 2003). Age has been a continuous factor for us. Even long after we completed our doctoral studies, we continued to be mistaken for PhD students, which we explain through our younger ages in completing our research projects and being told by others that we look younger than our ages. Although being placed lower on the academic career ladder may have disadvantages, we sought to flip this trade-off to our advantage: we employed interview techniques that placed us in apparently learning roles and our probing questions were not interpreted as threatening or argumentative. Our ‘lower’ positions allowed us to appeal to the magnanimity of interviewees to give us access to closed spaces (e.g. showing us PPG spaces in the German Bundestag, explain how things work in the House of Commons tearoom, or further information about the EP Conference of Presidents).
Our identities indirectly – or unintentionally – affected our research. Geddes, for instance, was working in the UK Parliament as a white man with a neutral/middle class accent and an upmarket suit; he did not look or sound out of place. It is perhaps no surprise, then, that he did not pick up the same gendered dynamics as Miller in her research on gender in the House of Commons (compare Geddes, 2020, with Miller, 2021). Other scholars have pointed out such dynamics: for example, Childs and Palmieri (2023: 182) note that audiences of women MPs ‘required some kind of acknowledgement of cultural affinity [and] Palmieri felt compelled to declare her African ancestry’. Miller, too, felt compelled to share familial information when spending time with political families. Such dynamics can become tools for interpretive research, as in the case of Brown (2018: 29), who ‘did not challenge their [elites] (mis)conceptions’ and ‘willingly allowed them to map their stereotypes onto me in order to gain access to their world’. Smrek (2022: appendix) notes that his gender could have stymied access, even if he was not extraordinary in being a man in the Czech parliament. Meanwhile, for Miller, being part of a team allowed her to discuss these identities and traits with colleagues. Such issues are significant given the gendered, classed and raced nature of these elite political organisations (Puwar, 2004).
Implications: Performing the role of researcher
Our affiliations and traits (visible and invisible) combined in various ways that we sought to ensure facilitated access to parliaments. Put differently, we sought to ‘perform’ a particular role that would enable us to appear non-threatening, even part of the in-group. We wanted to seem like any other researcher on a parliamentary estate. To that end, we dressed and spoke in ways that allowed us to fit into the infrastructure – muting our accents, not voicing opinions, locking social media accounts, dressing smartly and so on. We made the choice that we wanted to be interpreted as insiders, even if we were not. These things are not always cost-free – such as buying a professional wardrobe as a PhD student or setting up professional websites with personalised domain names. Others may also criticise such moves as compromising our identities to ‘fit in’ and acquiesce to dominant power relations.
Importantly, the way that we might use our capital changed over time: we had different capitals depending on our roles and positions. Our capitals may also be impacted by the fortunes of wider political trends – being associated with one group or institution at one point might be interpreted as a boon; at others, a cost (see above and Childs, forthcoming). Ethnographic capital is not fixed but relational and is needed at different times: for Miller, reputational capital was less necessary in the House of Commons given her existing access via her sponsoring MP; but in the EP it was far more important to gain access. The same was true for Geddes, who has become trusted among some managerial staff in the UK Parliament, but remains unknown in the German Bundestag (though two interviewees accepted interview requests on the basis of their interest in Scotland and Scottish politics, where Geddes’ institution is based).
The implications of this discussion are fairly simple: that we need to be reflexive about our positionality for research and teaching; something that is widely shared across most ethnographic research. However, we argue, furthermore, that this affects our research strategies directly. This is especially important given the specific and political power dynamics and relationships at play in parliamentary settings, including the institutional affiliations one may or may not have, as well as how specific identities are marginalised in these high-level political institutions. That said, this applies to ethnographic research more generally, in which identities and resources can be deployed strategically; similarly, other types of qualitative research on parliaments will entail making choices about how to ‘perform’ the role of, for instance, interviewer.
Analysing and publishing findings about parliamentary elites
Our final theme – analysing, presenting, publishing and exchanging ethnographic data on and with legislatures – raises professional and reputational risks for both researcher and researched. In short, we point to the ongoing ethical and methodological considerations and decisions that we need to make, which do not disappear after we leave our research settings (for a discussion, see Niemi, 2010).
Voice and plurality, transparency and anonymity in a parliament
Data collection and analysis need to be carefully and accurately explained in ethnographic research. In part, because a lot ‘hangs’ on the ethnographic research in the sense that the research is intimately connected to participants: it is their real and lived worlds about which we, as ethnographers, are writing. Transparency supports the rigour in conclusions, in which scholars are able to outline the processes and analytical frames used. It additionally has pedagogical value for future ethnographic work and to ensure that the lessons learned can be adopted by future researchers. And yet, it requires a necessary balance and deft handling with anonymity to protect the privacy and consent of participants.
Transparency can be approached in different ways, with different trade-offs (Reyes, 2018). Smrek (2022: 5) made his interviews open upon request but did not share fieldnotes since ‘their value for replication purposes is limited’. We shared meta-data about our research and interviews – such as timeframes, number and types of interviews and so on, but kept fieldwork diaries and interview transcripts private and confidential. Indeed, the status of fieldwork diaries is an overlooked issue in ethnographic research. We used them for immediate ‘scratch notes’ that were developed, usually later in the day, to fully and more expansive summaries. Geddes number-coded paragraphs and cited these in his research (even though those codes are not usable to readers they fostered personal accountability). For collaborative ethnographic research, this is necessarily different: Miller used a generalised observation protocol to account for different meetings and actors she observed, and shared with other project members via software (ATLAS.ti). Such data are sensitive – particularly in a political and parliamentary context – so how it is gathered and kept will impact the research. 8 Specifically, it will affect how much information can be traced back if published, requiring choices about transparency and the levels of details.
A key and distinctive consideration for elite-focused ethnographic research, such as on parliaments, is the level of anonymity for people (and, in some cases, documents). It matters for elites and parliaments because of their political and public nature. Niemi (2010) considered not naming the Finnish parliament altogether. In our cases, we did, but with different levels of specificity. For Geddes’ first ethnographic project, he noted the parliamentary timeframe (2010–2015) and that he was based with a committee, but not which one. This afforded far more freedom to draw on fieldwork notes and interview data, particularly as the PhD was subsequently published under embargo and the book appeared five years after the end of the parliamentary timeframe in 2020. By contrast, Geddes was open about the committee of his second period of fieldwork. This significantly curtailed his ability to make use of any observations for research purposes given the lack of protection for actors that he had observed. The trade-off here led to greater transparency but less publishable research.
Miller’s choices were slightly different. For her PhD, she worked through the strategy of specifying broad, aggregate groups. She included dates and government, opposition and minority oppositions and the department for administrative staff. She embargoed her PhD, particularly given the bullying and harassment scandal at the time of her completion, which sensitively connected to her research on the gendered nature of the House of Commons. 9 For Miller’s post-doctoral contribution to the team project, the political groups were named. This highlights the politically sensitive choices that parliamentary ethnographers make when publishing findings. 10 The greater the levels of transparency, as often sought by political science departments, the less likely it is for data to be anonymous – which matters for some participants (often impartial officials in parliamentary administrations) but not for all.
Issues round transparency and anonymity are affected by wider traditions and norms around ethics regimes in higher education and in political cultures. In the UK higher education sector, ethical review for research is considerable and process-based. The use of written consent forms for elites is common, to which both authors can attest. The same is not necessarily true elsewhere. In his German fieldwork, Geddes relied on oral rather than written consent when conducting fieldwork, where this practice is far more common.
Research interwoven with democratic values
We are not only passionate about ethnographic research, but also its link directly to parliamentary democracy; our research overlaps with some interests of the organisations we are studying that emphasises both our motivations and the research value. This is not necessarily unique to us; ethnographic research generally has often been interwoven with normative values, especially to foreground the interpretations of marginalised groups and create change (Childs, 2024). This applies in our research in several ways, in which we believe that political science should engage with and contribute to wider societal debates and challenges.
As Crewe (2021a: 14) has pointed out, ‘who to talk to and in what depth’ are important in influencing accounts. These choices are affected by the normative values we hold about democracy and politics. Miller, in the House of Commons, found that due to her age and research topics (gender and power), parliamentary researchers and officials – those in different institutional positions and entangled in different power relationships became key to understanding how far the parliament was a gendered workplace. In Geddes’ case, his belief in the importance of relational accountability in politics through parliaments meant that he emphasised practices of scrutiny rather than merely outcomes or uptake of recommendations. Brown (2014, 2022: 565) had a normative starting point: she wanted to centre black women’s narratives in political science research. Elsewhere, Crewe (2021a) discusses how her wish to defend politics and the UK Parliament persuaded actors to speak with her and explain their perspectives (more recently, she has noted how global and national breakdowns may foster more critical coalitions and networks; see Crewe et al., 2022: 754).
Some parliaments encourage knowledge exchange with researchers to improve their functioning. For example, Geddes’ parliamentary fellowship was based on the Select Committee Team’s objective to improve the effectiveness of its scrutiny. This was the outcome of longstanding relationships with parliamentary officials in the UK Parliament, having co-authored with staff (Geddes and Mulley, 2018) and presented research to the practitioner-focused Study of Parliament Group. Knowledge exchange and impact (KEI) is shaped by the wider university context, which is particularly strong in the UK, but nevertheless Miller’s EUGendDem project contributions included seminar presentations to the EP’s party groups in open and closed settings, to share knowledge and to participate in roundtables.
Implications: In search of a better democracy
Analysing and writing up findings, and seeking to publish them, are ways through which we continue to engage with the field and the people under study – especially how we choose to cite people, whose voices we privilege, and the venues and formats of research outputs. The relationships that have been built during fieldwork periods do not end when the ethnographer shifts the dial from immersion to analysis and publication. Ethnographers will work with the researched to write up papers or draft chapters, giving further opportunities to negotiate access and consent, test out ideas and gain feedback.
With respect to parliaments, there are specific and distinctive issues relating to publishing ethnographic material, not least the values placed on confidentiality, trust and anonymity given the politically sensitive nature of the research (see also Busby, 2011). At the same time, it is arguably in the nature of parliamentary-focused ethnographies that bring opportunities, too, especially for knowledge exchange and impact. It can take different forms: Geddes (2017) provided a programme note for a theatre play while Miller (2021: 141) achieved permission from an artist to reproduce a piece of artwork, ‘the Vote Office’, in her monograph to generate interpretive responses from readers. More traditionally, we reflected on improving parliamentary practices through evidence submissions to parliamentary inquiries or working with parliamentary actors to identify better practices for scrutiny (e.g. Geddes, 2023).
Ethnography strongly encourages direct and long-term engagement with actors, which means that research findings are likely to make their way into the parliamentary settings and potentially shape the thinking of actors inside these organisations. This is reinforced by the very foundational nature of ethnographic research that often aims to make a real contribution to understanding the world and hope to make it better. For political scientists, this avenue can be highly appealing in the United Kingdom and elsewhere with high demands for demonstrating wider societal benefits of research.
Bringing things together: Studying parliaments ethnographically
In this article, we have offered critical and comparative reflections of our own ethnographic research on parliaments. There are, of course, more and cross-cutting themes that go beyond the four main sections that we have sought to analyse above. This includes, for example, the ongoing and important power relations that pervade social science research. There are power imbalances within research teams as well as between different levels of researchers (e.g. PhD researchers and professors) and elites. Different identities and groups might be structurally disadvantaged in elite settings such as parliaments – for example, women, ethnic minorities – that ethnographers seek to point out by giving them a more central voice in accounts of parliamentary politics (let alone ethnographers’ own identities that mediate their research). Power relations affect the way that we, as researchers, access institutions, the capacities and resources we have to organise fieldwork, and more. This was most clear in our section on ‘ethnographic capital’. And while power relationships between researcher, participants, research setting, subject matter and so on, have been widely studied in ethnographic research (Archer and Souleles, 2021), we note that there are distinctive factors with respect to parliaments that – in liberal democracies – strive to achieve a balance between transparency and accountability.
A second cross-cutting theme is that the wider university and higher education setting is important. In the UK context, a political science PhD is expected to take no more than four (and ideally three) years of full-time study – which makes ethnographically-immersive methods more difficult, owing to their resource and time-intensive nature (but not impossible). Furthermore, UK higher education has been affected by a strong set of incentives for academic researchers to engage with practitioners to ensure their research is policy-relevant (Smith et al., 2020). Other higher education landscapes are not bound by this, and have different cultures, but might perhaps have – as in the United Kingdom – different levels of prestige or ‘capital’ attached to institutional affiliations. In short: different positions, affiliations and research cultures will affect relationships that can be fostered with parliaments. This is mediated yet further by funding capacity for research of this kind. Historically, research councils have been reticent to grant the considerable resources investments required.
And while the research landscape is important, so is that of parliaments and their institutional attitudes to academic engagement. In recent years, some parliaments have opened up further opportunities for collaboration, though we note that our experiences stem largely from western European settings. For example, the UK’s four legislatures have developed knowledge exchange strategies to enhance academic engagement and harness the so-called impact agenda (Beswick and Geddes, 2020; Smith et al., 2020); the Austrian parliament has introduced a ‘Day of Research’ and a related ‘Year of Research’ to promote parliamentary fellowships similar to that offered in Westminster and the EP (Parlament Österreich, 2023). Such schemes can facilitate ethnographic research – as done by Geddes (2023), Leston-Bandeira (2019) or Bennister (2021). Lobbying scandals or COVID-19 may mute some of these exciting developments, but nevertheless they suggest a mixed and changing landscape of opportunities.
Concluding remarks
Parliamentary ethnography has become increasingly widespread and there are further opportunities to expand its use in legislative studies. Its significant potential requires more reflection on the consequences of adopting particular ethnographic strategies. This article is one clear attempt to do so by linking discussions of undertaking qualitative parliamentary research with methodological practices and choices in ethnographic research design. Specifically – to highlight the core argument – we show that different choices and levels of access affect relationships between researcher and parliament, and the eventual outcomes of the research. By focusing on four themes, we have highlighted trade-offs across ethnographic research on parliaments.
In our first theme on entry and access, we have shown there are different choices and trade-offs in being embedded in the administration, working for MPs, shadowing representatives or observing as a spectator (among different choices). This has highlighted a perennial concern in ethnographic research about accessing research settings, but with a particular discussion on relevance for parliamentary-focused research. It also speaks of wider qualitative attempts to research parliaments, such as undertaking interviews (for a detailed discussion, see Cowley, 2022). While our discussion focuses on parliaments, there are broader relevant considerations for undertaking different types of qualitative research on political institutions.
In our second theme on organising fieldwork, we have demonstrated that researchers need to make adaptations to the rhythms and cycles of the research settings. Once again, there are broader ethnographic debates on this issue, but also distinctive parliamentary concerns, not least the structure of the organisation of parliamentary business and the cycles for how, when and where the specific parliament or legislature may meet. Researchers need to make choices about how far they can and want to align their patterns with those parliamentary rhythms. This speaks of wider choices that social scientists need to make with respect to ethnographic research on political institutions.
In our third theme on ethnographic capital, we have identified broader lessons from methodological debates on ethnography to explain distinctive issues facing political institutions and especially parliaments. We have identified our personal journeys and research design decisions through which we have sought to deploy our strengths and mitigate challenges. While there are specific parliamentary issues here about institutional resources and capital that may or may not be available (e.g. fellowship schemes with contractual protections for different actors), there are broader lessons for breaking down obstacles in ethnographic research.
And finally, in our fourth theme on analysing the findings, we have indicated the continual interwoven nature of the researcher with the research setting. Once again, we have applied general lessons and debates in ethnographic research to our specific elite, parliamentary setting. Here, actors are especially protective over their anonymity and concerned about political implications of the research (see also Busby, 2011). Similar lessons would apply in other politically focused ethnographic research.
In sum, what we demonstrate in our four themes are broad trade-offs and choices about practically undertaking parliamentary ethnographies – focusing on the ‘how’ of interpreting parliaments. We do not believe that there is a ‘right’ or ‘wrong’ way to do parliamentary ethnographies, nor that the outcomes are therefore ‘subjective’ or less relevant; rather, the aim of this article is to highlight that there are manifold routes but that each of them requires practical decisions about research design. Such choices inevitably lead to different types of research and potentially affect the outcomes of said research. But all of them make a contribution. Indeed, being there, in the thick of it, part of the action, and observing things happen in real time – metaphorically or literally – all of these things will give us new insights into political institutions. In the case for both of us, for our successful and not so successful forays into ethnographic research, it is observing how abstract ideas in books and articles play out in reality that is the most rewarding part of the research process. We acknowledge not everything will go to plan – indeed, that is one of the special things about doing ethnographic research in the first place.
Footnotes
Funding
ORCID iD
