Powerful yet Precarious: Rethinking Elite Interviewing Through Vulnerability,Absence and Disclosure

Abstract

Elite interviewing is often framed as a method for accessing those in power. This paper adopts an alternative view and repositions it as a forum where power and vulnerability are relationally co-produced, negotiated, and revealed by both interviewer and interviewees. Drawing on 110 interviews across three prison health research projects, this study offers a self-reflexive and empirically grounded exploration of the nuanced dynamics that shape elite interviews. This paper significantly departs from dominant assumptions that portray elites as authoritative, detached or evasive. Instead, it examines how interviewers can inspire elites to feel vulnerable to scrutiny or exposure. They do so using soft coercion in snowball sampling, the intimacy of online interviewing, the emotional labour of disclosure, and the reputational risks involved in sharing sensitive insights. The paper also highlights how the shrinking pool of senior actors, accelerated from burnout, institutional restructuring, and post-COVID-19 pandemic attrition, signals a loss of institutional memory, which is an under-recognised form of elite power. Methodologically, the study proposes the idea that elite interviews should now be viewed as emotionally textured, ethically complex, and relationally co-produced encounters. Theoretically, it redefines elite status as contingent, precarious, and contextually situated rather than fixed and unidirectional. This work’s significance extends beyond academia, offering insights into transparency, accountability, and policy learning that can inform improvements in a system that is often replete with secrecy and centralised control. By reframing elite interviewing as an interactional process shaped by both authority and exposure, this research contributes original conceptual insight, rigorous empirical analysis and practical relevance. It urges scholars to attend to the affective, ethical, and epistemological dimensions of elite research, and to treat elite interviewees’ silence, vulnerability, and absence from participation not as gaps in data, but as critical data in themselves.

Introduction

Interviewing elites has been an established qualitative research method for over six decades (Giddens, 1972; Kincaid & Bright, 1957; Osorio-Rauld et al., 2024). Elites, whether C-suite executives, major donors, or political leaders, are often regarded as uniquely positioned to provide insights inaccessible to the public or unavailable from lower-level organisational members (Ellinas, 2021; Fastenrath & Marx, 2024). Yet, despite its longstanding use, elite interviewing remains under-theorised, particularly in relation to the role of power.

As Ostrander (1995, p. 133) notes, social scientists still “too rarely ‘study up’”; that is, they often neglect to explore individuals who may possess high status. This omission reflects the challenges of accessing elites, largely due to their power (Desmond, 2004; England, 2002). Though many studies acknowledge elites’ power, they often treat it as fixed and unidirectional, thereby portraying elites as dominant figures who can control access, deflect questions, or even dictate how research findings are disseminated (Aguinis & Solarino, 2019; Sabot, 1999; Welch et al., 2002).

Social scientists have frequently conceptualised power as a fixed force predominantly concentrated in the hands of societal elites (Dexter, 1970; Welch et al., 2002). Researchers, though, are often positioned as subordinate to elites, and methodological prescriptions for studying elites often reinforce this perceived imbalance, such as through using flattery, not challenging elite responses, or “playing dumb” (McDowell, 1998; Rice, 2010; Velardo & Elliott, 2021). The tendency to glorify elite power while downplaying researcher agency is both empirically limiting and politically problematic, which can reinforce existing power hierarchies and limit critical inquiry.

Theoretical frameworks from the social sciences, however, suggest a more complex understanding of power. Lukes’ (1974) three dimensions of power (i.e., decision-making, non-decision-making, and ideological) are useful for analysing how elites exert control, not just through action, but also through their inaction and influence over prevailing thought or discourse. Foucault’s (1979) concept of power/knowledge further suggests how this picture is complicated, highlighting how elites are often powerful because of the specialised knowledge they possess (Goldman & Swayze, 2012; Petkov & Kaoullas, 2016). Giddens’s (1984) structuration theory similarly frames power as embedded in agency and conventional behaviours and practices.

Power in elite interviews is relationally dependent and situationally contingent. For instance, researchers, too, exercise power; they do so when shaping interviews, guiding narratives, and controlling dissemination of academic output (Morris, 2009). A more critical engagement with elite interviewing should acknowledge this reciprocity and explore how power is negotiated and co-produced in the research encounter.

Moreover, current literature pays insufficient attention to the vulnerabilities of elites. Though elites are often presumed to be invulnerable, emerging studies show that this is not always the case. Disclosure of sensitive information, reputational risk, or emotional strain during interviews can make elites uneasy or even distressed, thereby creating moments of vulnerability (Declercq & Ayala, 2017; Glas, 2021; Lancaster, 2016). Vulnerability is particularly salient when elites recount personal histories or reflect on decisions having moral or political weight (Neal & McLaughlin, 2009).

This paper examines the paradoxes, contradictions and nuances of power and vulnerability in elite interviewing. Although elite interviewing has traditionally emphasised the power of elite participants, this work argues for an enhanced understanding that recognises the shifting dynamics of both power and vulnerability on both sides of the research encounter. By investigating these dynamics, the current undertaking challenges conventional assumptions and advances a more reciprocal, context sensitive approach to studying elites.

The paper is grounded in the first author’s experience conducting semi-structured interviews with 110 policy elites across three prison health research projects conducted between 2016 and 2023 (Ismail & de Viggiani, 2018; Ismail, 2020; Ismail, 2022; Ismail, under review). These projects spanned both pre- and post-COVID-19 contexts and addressed the following areas: the use of direct regulation to implement the Healthy Prisons Agenda in England (2016), the impact of austerity on prison health governance and healthcare delivery (2017–2021), and an evaluation of COVID-19 mitigation measures in English prisons (2023). These projects are hereafter referred to as Project 1 (2016), Project 2 (2017–2021), and Project 3 (2023).

In the paper, “elite participants” refer to individuals strategically positioned within national and international organisations to influence prison health policymaking and practice. Our use of the word “elite” is not intended to imply a homogeneous power bloc. Rather, it denotes individuals whose institutional roles and accumulated expertise grant them unique access to information and a distinct capacity to shape outcomes within the specific context of prison health.

Participants were drawn from organisations central to prison health governance and strategy, including representatives from the World Health Organization (WHO), United Nations (UN), HM Prison and Probation Service, NHS England, the UK Health Security Agency, and the UK Scientific Advisory Group for Emergencies. Although the individuals interviewed across the three studies did not constitute a homogenous group, they shared critical characteristics relevant to their elite status within the policy landscape. As per Lilleker's (2003) definition, all participants were actively engaged in prison health policymaking. Furthermore, they held authoritative roles within their respective organisations and fields (Littig, 2009; Mikecz, 2012), possessing specialised professional expertise and access to critical resources (Khan, 2012). Their influence stemmed from significant social capital, strategic networks, and institutional positioning (Harvey, 2010), enabling them to operate in proximity to elected decision-makers. In comparison to the public or frontline staff, these individuals typically exercised greater influence over political and policy outcomes (Richards, 1996).

The foregoing characteristics of the elite participants reflect a central argument of this paper: that elite status in the context of prison health policy is inherently relational, contingent, and context dependent. This status is shaped less by absolute power and more by participants’ relative ability to mobilise influence, knowledge and resources within specific institutional and political structures relevant to the current study’s objectives.

This study represents a significant and original contribution to the field of qualitative social science and elite studies, offering new theoretical and methodological insights into the shifting dynamics of elite power and vulnerability. Drawing on a voluminous empirical dataset, which comprises 110 in-depth interviews, 6,693 minutes of recorded dialogue, and over 1.1 million words of transcribed narrative, this work is underpinned by methodological depth, analytical rigour, and reflexivity regarding ethical considerations.

This study departs from extant dominant approaches. Specifically, it emphasises the emotional, ethical, and institutional vulnerabilities of elite actors, particularly in high-stake policy contexts such as prison health. By critically examining how elites navigate pressures related to access, disclosure, burnout and institutional change, the research introduces new ways of understanding elite status not as fixed or monolithic, but as contextually situated and often precarious. The juxtaposition of power and vulnerability challenges prevailing assumptions in elite research, thereby making a conceptual contribution that is both timely and novel.

Methodologically, the research is innovative in its reflexive approach to elite interviews. It innovates through its capture of micro-level dynamics (e.g., impression management in online settings), soft coercion through use of snowball sampling, and the ethical dilemmas of reporting politically sensitive material. These insights are grounded in the lived researcher experience, thus offering a model for rigorous and ethically sound empiricism in politically sensitive contexts.

The significance of this work extends beyond academic discourse. It directly engages with ongoing policy concerns related to institutional memory, leadership turnover, and transparency in public health governance. The findings hold relevance for policymakers, think tanks, and advocacy organisations interested in prison reform, evidence-based policymaking, and ethical management of public health crises. Moreover, this study’s post-fieldwork engagement with elite participants aligns with principles of research impact, offering pathways for future knowledge mobilisation, collaborative engagement, and public benefit.

The remainder of this paper is organised around a series of interlinked thematic sections that explores how elite power and vulnerability are co-produced throughout the research process. Initially, the paper, examines the recruitment phase, highlighting how snowball sampling simultaneously facilitates access to elite participants and introduces subtle forms of pressure and exposure. It then explores the dynamics of the interview encounter itself, focusing on how online formats and the creation of safe, reflective spaces shape elites’ performance in the encounter and their disclosure. Subsequent sections examine ethical and emotional complexities of reporting findings, especially in contexts where participants’ institutional visibility complicates anonymity. A discussion is then offered about how elite interviews are a potential mechanism for political accountability, as well as how elites voicing dissent or criticising a comment poses risks for themselves. Finally, the paper reflects on a broader structural shift: the shrinking pool of elite participants due to burnout, resignation, and institutional attrition. Because of this phenomenon, the paper argues for rethinking of institutional memory as a critical yet increasingly fragile form of elite power. Collectively, the foregoing sections provide a reflexive, empirically grounded account of the shifting dynamics occurring in contemporary elite interviewing.

Rethinking Elite Power and Vulnerability Through Research Interaction

Although elites are often framed as powerful, this section explores how vulnerability, understood as exposure to risks, pressures, or uncertainties that can affect their reputation, authority, or decision-making autonomy, can surface in subtle and unexpected ways throughout the research process. Drawing on reflexive insights from three prison health studies, this paper examines how elite participants navigate the social pressures from snowball sampling, the disarming nature of online interviewing, and the safe space the interview itself can provide. Together, these reflections highlight the emotional texture of elite research encounters, where power and vulnerability coexist and where candid, human moments can emerge in even the most structured settings.

Vulnerability of Elites in Snowball Sampling

The use of snowball sampling in which existing participants assist in recruiting others from their professional networks (Denzin & Lincoln, 1994) illustrates the paradox of elites as being both powerful and vulnerable. In all three studies, this technique played a central role in accessing hard-to-reach policy elites. Referrals were provided to the first author both indirectly (through name suggestions) and directly (via email introductions). As Thuesen (2011, p. 620) notes, “networks, social capital, and trust are often paramount for gaining access to elites.” The effectiveness of snowball sampling was reflected in a recruitment success rate of 30% with nearly one in three individuals agreeing to be interviewed.

Snowball sampling, however, can exert subtle forms of pressure on prospective participants. Invitations from trusted colleagues or superiors may be interpreted as endorsements, thereby creating a silent but palpable sense of obligation among these potential interviewees. This “soft coercion” was not always visible on the surface but often revealed itself in these individuals’ subsequent tone, body language, or offhand remarks. There was a moment in Project 2 when a senior civil servant, arriving flustered to the interview said candidly, “If I’m being totally honest, I’m here because the minister told me I had to be. And I tend to do what my [superior] tells me to do. So, shoot your next questions.” It was a disarming moment, equally humorous and deeply telling of the relational politics surrounding elite access.

Conscious of such dynamics, in accordance with ethical protocols, the voluntary nature of participation was emphasised in all communication. Prospective interviewees were given ample time to consider their involvement; once an introduction was made, gatekeepers (i.e., the individuals who made the initial referral) were excluded from further correspondence to protect participants’ autonomy and privacy. Throughout the invitation process, participation was consistently emphasised, both verbally and in writing, as being optional and confidential. A standard question in the initial exchange was also devised: “In what way does this study appeal to you?” The question was a gentle prompt to help potential participants reflect on their own motivations. Most individuals responded with genuine interest in the research topic, though such interest could coexist with other, less voluntary influences.

Another layer of elite vulnerability emerged from the risks of confidentiality inherent in snowball sampling, especially within the tightly knit networks of the UK’s health and justice policy landscape. Many participants knew one another professionally, and during the interviews, many referred to the “usual suspects.” Interviewees frequently tried, sometimes subtly, sometimes quite directly, to discern who else had already participated. Their questions ranged from “Has anyone from [X department] spoken to you yet?” to “Am I the only one saying this?” These queries were often laced with nervous laughter, pauses or hesitation.

In one interview, after gently declining to reveal any identifying information, the interviewer responded, “You know I can’t tell you that; it’s the researcher’s code.” The participant laughed and did not pursue the line of questioning further. But his laughter seemed to express a deeper unease: the fear of being exposed, of saying too much, or of being the outlier in a closely watched community. Navigating such moments required both the first author’s ethical clarity and emotional sensitivity, where he developed strategies to reassure participants without compromising their anonymity. For example, the following remark was offered to provide some validation to the interviewee without revealing specific details: “So far, others’ answers are quite similar to yours, though the interviews are still ongoing.” At times, a gentle steering of the conversation away from such inquiries was helpful to preserve both rapport and ethical boundaries.

The foregoing experiences revealed how elite participants, a group that other researchers have typically positioned as confident and in control, can experience very real vulnerability in research encounters. Their reputational stakes, institutional allegiances, and proximity to one another complicate the assumption of their absolute autonomy. What emerged through the interviews was a more complex picture of the elite interviewee: one who is powerful, certainly, but also cautious, reflexive, and at times, exposed.

Vulnerability During Data Collection: How Online Interviewing Limits Impression Management

Existing research on elite interviewing often draws from Goffman’s dramaturgical theory, particularly his concept of impression management: the way individuals consciously construct their self-image in social encounters (Goffman, 1959). Scholars have noted that in interviews, elites seek to project authority, competence, and moral legitimacy (Oydna & Bjorndal, 2022; Whiteside & Kelly, 2015). Others describe how elites often frame their participation as driven by intrinsic motivation, civic duty, or transparency (Gao & Zhang, 2021; Gilding, 2010).

Little attention has been paid to how elite engagement shifts when the interview moves online. Although platforms such as Zoom and Microsoft Teams (and previously Skype) undoubtedly ease scheduling and improve accessibility (Lo et al., 2016; Self, 2021; Vaagland, 2024), they can also diminish the impact of many of the visual and environmental cues elites traditionally used to assert their dominance. Goffman (1959, 1967) asserts that settings are part of an individual’s performance. In-person interviews and physical surroundings such as imposing offices, visible staff, and institutional architecture reinforce the symbolic power of elite actors (Ross, 2001). However, when elites are interviewed remotely, whether from an office or home, their “performance” is mediated by the digital frame. While a professional office background still conveys authority, the typical online interview scenario often reduces the broader environmental cues that contribute to a full “setting” performance, focusing attention more directly on the individual’s presence on screen rather than the encompassing physical space.

In contrast, in Project 3 that involved an evaluation of COVID-19 mitigation measures in English prisons, most elite participants joined the interview from their home. Children’s drawings, family photographs, kitchen counters, and living rooms became backdrops. The difference in “affective relationality” (i.e., the emotional texture of the researcher-participant interaction) was palpable. It was not just a change in setting but a subtle shift in dynamic. Many participants, clearly fatigued from back-to-back virtual meetings, arrived late, flustered, or apologetic. Some expressed difficulty transitioning mentally into the interview space. One began with, “I’m not sure I’ll be coherent—I’ve been on six calls already today.” Another participant glanced at the clock and said with a tired smile, “It’s 3 p.m., and I haven’t had my lunch yet. Do you mind if I eat while we talk?” These disclosures immediately softened the tone of our exchange. There were moments when children screamed for attention, pets leapt onto laps, or participants paused to answer a knock at the door. One senior official excused himself mid-interview because “the cat just knocked over the kettle.” These unscripted interruptions did more than disrupt. They revealed glimpses of elite interviewees’ everyday lives, making the interaction feel more equal, more real. In these moments, interviewees seemed to reduce their performative efforts in the interview and became more candid. The weariness of participants made them more human, and perhaps paradoxically, more open. Rather than espousing polished, prepared narratives, they often delivered fragmented, emotional or spontaneous reflections. The interviews thus became less like a performance but more like a conversation.

In some cases, this shift created moments of a deep connection between the interviewer and elite interviewee. A senior policymaker, reflecting on the unprecedented challenges of managing COVID-19 measures in prisons, sighed and shared a profound moment of vulnerability: “I’ve never told anyone this, but we were completely overwhelmed.” That admission of operational and emotional strain came from an individual in an authoritative, decision-making role was significant. This moment stood out, not only for its emotional weight but because it resembled something that had transcended the expected script. The digital medium, often dismissed as cold or distancing (Adams-Hutcheson & Longhurst, 2017; Seitz, 2015), had in this case fostered a kind of disarmed intimacy, offering a rare glimpse into the personal realities of those at the apex of policy implementation.

At the same time, navigating this new milieu brought with it fresh ethical responsibilities. Seeing elite participants in their home environments, dressed informally with signs of everyday life, blurred the boundaries of the professional and the personal. Interviewees’ emotional fatigue and disarming candour made them seem more human but also more exposed. The first author was acutely aware that interviewees’ openness did not equate with their informed consent for the revelation or dissemination of highly personal or sensitive information, and that trust, once extended, required care to be sustained. He often found himself softening the edges of his questions with phrases such as, “Would you be comfortable if I asked…?” This query was a small but deliberate gesture to respect their boundaries. Maintaining rapport without overstepping became a quiet act of calibration.

Online interviewing, then, is not just a logistical adaptation; it reshapes the entire research encounter. It exposes the constructed nature of impression management and creates new possibilities for vulnerability for both participants and researchers. The first author often found himself negotiating how much of himself to reveal, in return, to interviewees. Should he acknowledge the household noise in his own background, signalling shared disruptions of remote working life? Should he mirror the interviewee’s candour, joining them in a moment of mutual vulnerability, or simply preserve a more formal, professional distance? These choices were rarely trivial. Indeed, they shaped the tone of the encounter and reflected an ongoing ethical balancing act between empathy and detachment, and authenticity and rigour.

Ultimately, these digital encounters revealed a crucial insight: the power dynamics of elite interviews are not fixed but fluid. They are contingent based on location, technology, emotion, and the micro-interactions between the interviewer and interviewee. Rather than impairing the research, online interviews often softened the formality of the encounter. In such moments, the polished performance typically associated with elite interviewing gave way to something more candid and relational. Specifically, guardedness was diminished, and both the first author and participant engaged with enhanced openness and authenticity.

Vulnerability of Elites Stemming From the Safe Space of the Interview Encounter

Continuing with the theme of elites “letting their guard down,” the research encounter itself often functioned as a rare safe space for both interviewee reflection and emotional release. The notion that interviewees were “powerful yet vulnerable” became particularly evident in moments when participants described the interview as unexpectedly therapeutic. Interviewees often perceived the interview not merely as a point of data collection, but as a time for pause, debriefing, and introspection. Participants sighed audibly, lingered in silence, or smiled in ways that signalled the decrease of built-up tension. Such candidness is not unique to elite interviews; interviews more generally, such as those in long-form podcasts, can often prompt personal disclosure and reflection. What seemed distinctive in this study’s interviews, however, were the type of participant and the institutional context in which they occurred. Interviewees were senior policymakers, civil servants, and institutional leaders, often navigating high stake environments where strategic communication and impression management are second nature. Indeed, many began the interview with rehearsed, guarded responses, only to shift over the course of the conversation to more spontaneous, emotionally honest reflections. The formality of their public-face roles gave way to something more personal and unguarded. The contrast between elite participants’ expected performance and the emergent intimacy made these encounters empirically significant. The interviews created a rare backstage space away from press briefings, ministerial scrutiny, or organisational messaging where participants could speak more freely, and where the emotional texture of elite life came into sharper view.

Far from the often impersonal or combative dynamics sometimes associated with elite interviewing (Welch et al., 2002), the interactions were deeply humanised. The following excerpts from Project 2: Prisons’ Health Under Austerity (2017–2021) and Project 3: COVID-19 in Prisons (2023) highlight how participants experienced the interviews as opportunities for reflection, sometimes cathartic, sometimes emotionally taxing:

I would like to say thank you for choosing to ask me those questions. It’s been quite therapeutic unloading it all, so thank you very much. Because it’s a level of honesty isn’t it, in terms of kind of saying what you feel is wrong with things. Thank you very much for giving me that chance of talking (Regional Health and Justice Lead, Health Organisation).

[This interview] has been an interesting experience. These are things I’ve not thought about for a long time, and just reminiscing and reminding myself has been interesting, slightly [pause] nightmarish as well. It was a really difficult period of my professional life (Criminal Justice Lead, National Health Organisation).

These reflections challenged researchers’ traditional assumptions about elite control over interviews (Aguinis & Solarino, 2019; Dexter, 1970; Sabot, 1999; Welch et al., 2002). At the start of many interviews, particularly with high-level political or institutional interviewees, the tone was often tense, guarded, and at times almost journalistic, as if participants were anticipating a media-like interrogation. Responses were frequently initially polished, rehearsed, and clearly “on message.” But as the conversation deepened, especially when questions were framed within a reflective, respectful tone, individuals began expressing uncertainty, acknowledging that sometimes they had doubts about their own conduct, and at times showing visible signs of emotional vulnerability.

Paradoxically, some participants in the most authoritative positions, such as the heads of organisations, who might have been anticipated to be the most controlled and aloof often proved to be the most introspective. There was a tangible sense that, in the space of the interview, they were able to step away from the performance required in their professional lives. They implicitly invoked Goffman’s (1959) concepts of front-stage and backstage behaviour. Although elites often entered interviews with a prepared “official stance,” what unfolded was often far more spontaneous and candid, what Goffman might call the “backstage” self, that was temporarily revealed “behind closed doors.”

Using intention and intuition across the three research projects, the first author deliberately sought to enable such a shift. He adapted the nature of his professional role depending on the interviewee: he presented himself as a policy-oriented academic for political actors, a neutral health stakeholder for those in policy roles, or a thoughtful criminologist. These shifts were not manipulative but strategic; they reflected his effort to meet participants in their specific field. His prior professional experience as a prison health commissioner likely contributed to a level of credibility and empathy that helped him build trust. He could speak the language of policy and practice while maintaining interviewer distance, thereby leveraging the academic identity that promises rigour and impartiality without sacrificing warmth or reflexivity.

The interview space offered participants something that by their own account professional environments often lacked: time to pause, to reflect, and to express complex feelings without encountering institutional consequences. For some interviewees, the assurance of confidentiality, coupled with the tone of the interview, afforded revisiting difficult professional experiences. One participant, visibly moved, responded to a particularly sensitive question with a long pause, followed by “Hmmm… good grief, that’s a difficult question.” The first author remained silent, sensing that what followed would be meaningful.

Notwithstanding the foregoing interview characteristics, prevailing portrayals of elite participants as tightly controlling the research process through steering the conversation, deflecting the inquiry, and managing their own image (Burnham et al., 2004; Desmond, 2004; Mikecz, 2012; Ostrander, 1995) did indeed emerge in some interviews. This situation reflected that, across the three prison research projects, there was a much more nuanced interplay in which participants were not merely strategic actors, but individuals navigating institutional fatigue, political pressure, and personal emotion. As such, there were shifts within interviews as well as differences between them.

These experiences highlight the agency that researchers possess in shaping interviews: methodologically, ethically, and relationally. Embracing this agency with care helped the first author to create interview environments that were both safe and empirically rigorous. Vulnerability was not something to be extracted but something co-produced, emerging in the space between openness, trust, and the unpredictable flow of conversation.

The result was that the interviews served not only as sources of data but as meaningful encounters for both the first author and participants. In providing elites the space to reflect without encountering judgment, the research process became an act of care—one that blurred the boundary between the professional and the personal, as well as reflecting that qualitative research is, at its core, deeply human.

Exposure After the Interview: Confidentiality, Risk, and the Precariousness of Elite Disclosure

Although much scholarly attention in elite interviewing focuses on the research encounter itself (see the work of McDowell, 1998; Richards, 1996; Velardo & Elliott, 2021), vulnerability does not end once the interview concludes. For many elites, the most acute sense of risk emerging during the post-interview phase: when their words took a life beyond the private setting of the interview. Despite assurances of confidentiality, elites remained conscious of how easily their identities could be inferred from contextual clues, professional affiliations, or organisational roles. This section explores the reputation and political stakes of participation in elite research, revealing how power can shift towards the researcher during analysis and dissemination. This section reflects on the first author’s ethical dilemmas and emotional negotiations that arose when reporting politically sensitive insights and on the delicate balance between protecting participants and honouring the significance of what they chose to share.

Vulnerability of Elites due to the Exposure of Their Identities and Affiliations

Though elites are often perceived as shielded by institutional authority and personal status, such protection largely functions within the interview and not during data dissemination. Despite formal guarantees of confidentiality, many elite participants remained acutely aware of the risks associated with their visibility. Their unique roles, affiliations, and geographic or organisational contexts often made them easily identifiable through deductive disclosure, even without reference to their names. As Neal and McLaughlin (2009, p. 703) observe, this “unsettled nature of the power dynamic in the ‘upwards’ research encounter” points to how power can shift toward the researcher, especially when publication of findings holds reputational or political implications.

Vulnerability was frequently expressed through participants’ concern over attribution. During an interview with a consultant from an international health organisation in Project 3: COVID-19 in Prisons (2023), one interviewee hesitated before speaking openly:

Participant: The only thing I’m concerned about is kind of attribution and confidentiality. I wouldn’t want anything to be able to be attributed to me directly, because I do play quite a unique role in this field.

First author: Don’t worry, the quotes won’t be attributed to you or your organisation.

Participant: Okay. In which case I will be as candid as I care to be.

Similar concerns were encountered across all three projects, particularly from those occupying especially critical or senior positions. There was tension between participants’ desire to be candid and their fear of professional exposure. Assuring anonymity was not simply a procedural and ethical step; rather, it was a relational necessity. Clear explanation of how the data would be handled ensued, including how identifiable details would be redacted, paraphrased, or withheld entirely during analysis and reporting. The explanation often induced visible relief. As one criminal justice lead from a national institution stated, “I’m glad that this is anonymous; I can rip things apart.”

Such responses also created space for deeper, more authentic engagement. Participants’ “backstage” reflections (Goffman, 1959) emerged, revealing dissenting views, policy frustrations, and institutional tensions that would not surface in public-face conversations. At times, interviewees would lean in slightly and preface a remark with, “But don’t quote me on that,” or “This isn’t in the public domain yet.” Others offered disclosures prefaced with, “I’m only telling you this because I know you well enough,” thus signalling a level of trust that felt both rewarding and significant. Such moments did not occur in a vacuum. Instead, they were shaped by the first author’s presentation throughout the research process. The coupling of strategic and authentic positioning appeared to foster a sense of familiarity and legitimacy, allowing participants to move beyond formal scripts and engage in more open, candid exchanges. In this sense, the trust they extended to the first author was not only interpersonal but also grounded in the shared experience of working within, and sometimes around, complex institutional systems. Being perceived as “known” or “safe” did not erase the power asymmetries between the first author and participants. Instead, it helped create a space in which those asymmetries could be momentarily softened.

The behind-the-scenes insights that the interviewees revealed often included politically sensitive information, such as disagreements with superiors, concerns about future policies, or even strained relationships with government ministers. The stakes were clear. A misstep in how this information was reported could not only harm the participant’s reputation, but potentially jeopardise their organisation or broader professional network. Many elite interviewees were seemingly operating as a “precarious elite,” an individual positioned at the intersection of influence and risk, where visibility could be as dangerous as power would be enabling.

Because of the foregoing situation, the first author was most circumspect when curating the report. Decisions regarding what to quote, what to anonymise, and what to omit in reporting data held significant ethical implications. Power, in this context, did not end in the interview but extended to interpretation and dissemination of the findings. As with many forms of qualitative research, particularly when participants are part of tightly networked communities, the ethical responsibility to protect confidentiality was marked. What distinguishes elite interviewing, however, is not the uniqueness of these obligations, but the specific nature of what is at stake: reputational damage, institutional fallout, or policy repercussions that could follow from even indirect attribution. Therefore, the first author’s role demanded not only methodological rigour, but also a heightened moral attentiveness. Specifically, he had to recognise that even anonymised data could carry traceable institutional fingerprints and that participants’ professional proximity to one another intensified the risk of deductive disclosure.

Ultimately, elite vulnerability underscored the ethical complexity of “studying up.” As such, the first author had to recognise that power can be fragile, is reputationally contingent, and is unevenly distributed, even among the most powerful participants. Managing these concerns with care was central to sustaining trust, ensuring ethical integrity, and producing knowledge that was credible and meaningful, yet solicitous.

Political Disclosure and the Dual Edge of Power and Vulnerability

Elite interviewing offers more than access to privileged knowledge; it also creates moments in which the boundaries between institutional loyalty and personal conviction blur. In such moments, elites can shift from being powerful gatekeepers to vulnerable disclosers, sharing insider knowledge, frustrations, and critiques that are rarely aired publicly. Despite the formal status and institutional authority that many of the participants held, they were often constrained by political structures, organisational pressures, and the secrecy that surrounds sensitive domains such as prison governance. Within this interview space, framed by anonymity, academic rigour, and perceived neutrality, many interviewees found room to speak freely.

Participants frequently described the interviews as opportunities to “say the unsaid,” sometimes likening the experience to whistleblowing. This was particularly evident in discussions around controversial topics such as austerity, ministerial influence, or the pandemic response in prisons. As one participant, for instance, noted, “I don’t think I can say it in this call because you’re recording it and using it (Senior Policy Lead, Health Organisation). This comment underscored the delicacy of balancing institutional loyalty with personal veracity. The interviewee’s concern echoed Goffman’s (1959) concept of the “veneer of consensus,” where front-stage performance is maintained to preserve organisational unity, even if backstage beliefs diverge significantly.

In prison related research, this tension is not merely present, but it is intensified by the highly centralised, security-laden, and often exclusionary nature of prison governance. Unlike more transparent policy domains, prison systems are characterised by closed decision-making processes, restricted public oversight, and political climates that discourage dissent (Liebling & Crewe, 2012; Simon, 2018). Though policy actors across many fields face institutional pressures to support prevailing agendas, the stakes in prison governance are often uniquely high. Participants in all three prison studies spoke of working within cultures of silence and risk aversion, where questioning official narratives, particularly around contentious issues such as austerity or COVID-19 measures (Ismail, 2022; Ismail, under review) was professionally sensitive, if not reputationally dangerous. Against this backdrop, the interview emerged not just as a methodological tool, but as a rare and protected space where policy elites could articulate discomfort, share insider perspectives, and quietly but meaningfully push back against the constraints of institutional orthodoxy. As one public health specialist remarked:

Your questions articulate some of the issues that we’ve all been skirting around… because it is a political topic, and we are civil servants. Whereas you’ve gone straight to the question (Prison Health Lead, Prison Health Under Austerity Project)

Another participant reflected on the interview as a rare opportunity to reflect openly on both failures and successes:

What appeals to me is being able to put that on record… hopefully be able to influence future responses to these issues (Health and Justice Lead, COVID-19 in Prisons Project).

These moments revealed how power and vulnerability coexist in elite roles. Participants wield influence, but they are also constrained owing to organisational narratives, political climates, and reputational risks. Some interviewees described synthesising their messages to be more “politically palatable” when speaking with government ministers. Others admitted that their government ministers had asked them to find evidence that supported predetermined political positions, regardless of their own judgment, hence echoing critiques of policy-based evidence-making (cf. Coakley & Todd, 2014; Hammersley, 2013).

Contrary to research suggesting that elites typically adhere to the party line or suppress personal opinion (Blix & Wettergren, 2015; Harvey, 2010), many participants spoke as individuals rather than institutional representatives. That is, they detached themselves from their formal roles and expressed views that they might not voice in official settings.

The first author had to be acutely conscious of his power as a researcher in these exchanges, especially when participants disclosed sensitive information. Interviewees’ candour often came with the expectation that he would amplify the substance of their concerns even as their identity was protected. For several participants, the interview felt like an act of advocacy. When mentioning an issue of major concern to him, one interviewee stated, “Let’s all band together and create a greater storm about this.” Another said, “It was my duty to participate in such research.” The resulting openness, however, carried risks. As such, the first author’s use of triangulation remained critical, not only to protect against faulty memory or self-serving accounts (Sabot, 1999; Todd, 2014), but also to ensure that disclosures were not misread or misused.

Post-fieldwork engagement became a continuation of the research relationship. All participants were kept informed of the project’s outputs and how their data would be used. Keeping participants informed of how findings were reported served the purpose not only of keeping an implicit promise, but also of safeguarding trust and maintaining access for possible future research. Accordingly, the first author had to walk a fine line: remaining analytically rigorous and accountable, while upholding the integrity of participants’ disclosures. The reciprocal relationship with the elite interviewees challenged assumptions in the literature that portray elite interviews as extractive or one-sided (Herzog, 1995; Mikecz, 2012). Such engagement is not just a matter of good ethics, it is also an institutional expectation. In fact, funders and frameworks such as the UK’s Research Excellence Framework (REF) increasingly demand demonstrable public benefit. Consistent with such demands, when conducted reflexively and ethically, elite interviewing can enhance both scholarly insight and real-world accountability.

Ultimately, what emerged from the interviews was not simply information, but a shifting relational terrain in which power was negotiated, vulnerability revealed, and knowledge co-produced. These encounters infer that elite status does not preclude risk and that the role of the researcher is not only to record what is said, but to honour the complexity of what is shared.

Emerging Vulnerability: The Shrinking Pool of Elites and the Loss of Institutional Memory

Elites are often assumed to be enduring sources of expert insight—fixtures within institutions that outlast political cycles. Yet, during the fieldwork for Project 3: COVID-19 in Prisons (2023), the first author encountered a very different reality; it was marked by instability, absence, and emotional fatigue. A confluence of macro-level forces (i.e., economic austerity, organisational downsizing, and shifting work cultures in the post-pandemic era) had not only reduced the number of accessible elite actors but also eroded a less visible but equally crucial resource: institutional memory.

In the UK, ongoing austerity measures and government restructuring have significantly hollowed out parts of the public sector. Many civil servants, policymakers, and health leads with years of experience are leaving their posts, often without replacement. This has been reinforced by the current government decision to abolish NHS England (Department of Health and Social Care, 2025) which commissions prison healthcare services in 118 prison establishments in England. These departures are not isolated cases. Of the 117 individuals invited for interviews in Project 3, 29 (25%) had either resigned, retired, or could not be reached. Follow-up correspondence with several institutions confirmed that these individuals had not been replaced, often due to cost-saving measures.

Initially, the first author regarded these issues as barriers to recruitment. However, with each out-of-office reply and follow-up email that bounced back, such potential participants represented a loss of expertise, relational networks, and historical understanding. What became clear was that institutional memory is not simply background knowledge. Rather, it encompasses the ability to recall prior decisions, contextualise policy shifts, and understand the long arcs of institutional logic. When this memory is lost, so too is the continuity of critical insight that shapes effective governance.

Loss of institutional memory was particularly concerning in the prison health field, where decision-making structures are already centralised and often opaque. Participants often mentioned the lack of institutional learning following the COVID-19 pandemic, noting that “things haven’t moved on nearly as quickly as we would have hoped.” The fragility of policy progress seemed directly tied to the turnover of key personnel and the absence of structured knowledge being passed on to an incumbent’s replacement. The consequences of this problem were not abstract. As one participant bluntly asserted, “It’s developed a kind of new normal for prisons, which we’re very worried about.” The implication was that incarcerated people were in danger as a result.

Compounding the foregoing were broader shifts in workplace culture and institutional capacity. Some global discourse points to “quiet quitting” as a new ethos, which is marked by employees refusing to go above and beyond their core duties (Mahand & Caldwell, 2023). However, what was observed in the field during Project 3 was not so much disengagement but emotional and operational overload. Thirty-seven (32%) of those contacted declined interviews, often citing overwhelming workloads and burnout. One civil servant remarked, “My plate is rather full these days with urgent requests from the top, so a research interview like this is at the bottom of my priority list.” Another noted, “It’s not possible to take part unless I sacrifice my sanity.” These were not signs of indifference, but of saturation, symptoms seemingly reflecting leaner institutions stretched by staff shortages and post-pandemic demands. Rather than reflective of withdrawal, these responses highlighted a different kind of elite vulnerability: being too overextended to participate in research at all.

In several cases, participants who initially agreed to participate later disengaged from the research process: they either ceased communication with the first author or quietly withdrew without offering an explanation. Although such withdrawal is not uncommon in qualitative research, its timing and recurrence prompted the first author to engage in further reflection. He discovered that these instances clustered around periods of heightened institutional pressure and policy uncertainty, often involving individuals who had initially expressed enthusiasm for the study. One e-mail message simply stated, “Sorry, got too much on. Hope you managed to secure participation from others.” Although brief and courteous, such responses raised questions about what such silence might signify.

Exercising caution in interpreting non-response in the three prison studies is important. Withdrawal may reflect neither vulnerability nor hostility, but rather a desire to avoid the emotional or cognitive effort of formally declining. Yet, in the context of elite interviewing, where participants often hold roles of public responsibility and professional visibility, disengagement can carry additional resonance. The silence may reflect reputational concerns, organisational sensitivities, or broader fatigue with the demands of reflective engagement, especially on politically or ethically complex topics.

Although ascribing a single meaning to these withdrawals may be premature, they do point to the limits of access and disclosure in elite research. Rather than dismissing them as routine attrition, they should be viewed as part of a broader relational terrain of elite interviewing. Silence, too, becomes data, and non-participation may reflect the same pressures that shape what is articulated in interviews when elites do agree to speak.

The marked number of withdrawals prompted the first author to reconsider how to conceptualise the word “elite” in elite interviewing. If power is assumed to reside in access, status, or expertise, what happens when these individuals are no longer present or emotionally capable of participating? This does not mean that they cease to be elites in the institutional sense. Instead, it does expose the precariousness beneath the surface of presumed influence.

Institutional memory is a collective and relational form of elite capital that is acquired over time, embedded in networks, and crucial to policy learning. Its loss poses not only challenges for research access but also for governance and accountability. Participants were not just interviewees; rather, they were, in many cases, the last stewards of knowledge in their respective institutions about what had worked, failed, or been quietly buried.

From a methodological perspective, the shrinking pool of elites raises important questions about the future of elite interviewing, not necessarily because it impedes such research, but because it shifts its terrain. As institutional memory erodes, the researcher’s task becomes to trace what was known as well as to document what is being forgotten and how that absence reshapes governance. Refusals, retirements, and disengagement are not just hurdles but seemingly signs of deeper systemic exhaustion. Thus, the research challenge is not access alone, but how to make sense of policymaking environments that are increasingly defined by transience, fragmentation, and loss of long-term expertise. As such, elite interviewing becomes less about retrieving facts from enduring informants but more about capturing a time in flux, one marked as much by silence, absence, and rupture, as by disclosure.

The foregoing are not just logistical or access issues; they are epistemological ones. They require scholars to think critically about where knowledge resides, how it circulates, and what knowledge is lost when systems are hollowed out. Elite interviewing, in this context, is no longer simply about “studying up” to power. It is also about tracking its dispersal, understanding its absences, and witnessing the quiet ways in which power retreats.

In reflecting on these shifts, we suggest that although the foundational task of studying elites remains truly critical, the context in which this work is conducted is evolving. This is particularly so in public sector domains where institutional churn, emotional fatigue, and policy volatility are increasingly visible. As elite populations in government shrink or become more transient, the challenge lies in gaining access as well as attending to the emotional, structural, and institutional conditions under which elites are now asked to speak. Therefore, scholars need to recognise that, at times, silence or absence may carry as much empirical weight as participation.

Conclusion: Reframing Elite Power Through Vulnerability, Absence and Interaction

This paper has argued that elite interviewing is not simply a matter of accessing power, but it entails navigating the complex and shifting terrain where power and vulnerability intersect. Much earlier work on elite interviewing often conceptualised power as a fixed, unidirectional force, portraying elites as dominant figures who could rigidly control researchers’ access, deflect interview questions, or even dictate how findings are framed and disseminated (Aguinis & Solarino, 2019; Dexter, 1970; Sabot, 1999; Welch et al., 2002). Researchers were frequently positioned as subordinate, with methodological prescriptions often reinforcing this perceived imbalance (e.g., flattering the elite interviewee, not disagreeing with an elite’s comment) (McDowell, 1998; Rice, 2010; Velardo & Elliott, 2021). Drawing on datasets comprising 110 interviews across three studies on prison health, our findings present a more nuanced picture, demonstrating that elite participants are not fixed repositories of influence but are actors negotiating institutional loyalties, emotional strain, reputational risk, and at times, the fading traces of institutional memory. Our work enriches the conceptualisation of power and vulnerability in the study of elite interviewing, highlighting how decision-making, non-decision-making, and ideological influence (Lukes, 1974) manifest in both overt and subtle ways during research encounters.

Across each stage of the research encounter, from recruitment via snowball sampling to unguarded moments in digital interviews and backstage reflections in trusted spaces, elites revealed themselves as both powerful and exposed. This outcome stands in stark contrast to portrayals of uniformly controlled and impenetrable subjects. Their comments, often carried the weight of organisational critique, moral conflict, or quiet advocacy. Although their authority granted them access to privileged knowledge, their positions also rendered them uniquely precarious, subject to internal scrutiny, political constraint, and burnout in an era of institutional contraction and change. Accordingly, Foucault’s (1979) notion of power/knowledge becomes particularly salient, as elites’ access to specialised knowledge simultaneously empowers and constrains them.

What emerged was a nuanced and humanised account of elite participation, where impression management could falter, where digital disarmament enabled intimacy, and where the act of speaking felt at times like resistance, therapy or even whistleblowing for participants. This general outcome challenges previous methodological advice that often focused on overcoming elites’ perceived control (McDowell, 1998; Rice, 2010). Specifically, we revealed how candidness can spontaneously emerge. Far from being detached, the interviews were affective, co-produced, and charged with ethical responsibility. Research, therefore, is not only about managing access and data, but it is also about negotiating emotion, ambiguity, and trust, before, during, and long after the interview itself. In line with Giddens’s (1984) structuration theory, these moments demonstrate how power is embedded in everyday practices and how elites simultaneously reproduce and challenge institutional norms through their interactions with researchers.

This study invites a rethinking of institutional memory as a form of elite power, one that is accumulated relationally and often lost silently. The shrinking pool of experienced elites, driven by systemic burnout, organisational downsizing, and post-pandemic disillusionment, represents more than a methodological challenge. It signals a deeper transformation in the conditions under which policy knowledge is created, stored, and made available.

Collectively, this study’s insights suggest not that elite interviewing is undergoing a radical transformation, but that its emotional, ethical, and institutional complexities have been historically underexamined. By highlighting the fragility and dynamism of elite power in the research encounter, our work directly challenges the often one-dimensional portrayals found in earlier elite interviewing literature. The current undertaking contributes an enhanced, more reflexive account of elite interaction, where online formats, institutional strain, and emotional candour converge in ways not previously documented. We call for augmented methodological attentiveness to the emotional labour, subtle silences, and contextual conditions that shape elite participation, recognising that absence, uncertainty, and vulnerability are as empirically significant as access and authority.

Ultimately, to study elites is not only to study power. It is also to study its power’s limits: the cracks in its performance, the exhaustion in its delivery, the stories withheld, and the moments where influence gives way to fragility. In those spaces, rich insight lies not in extracting answers, but in witnessing how elites live, navigate, and sometimes quietly resist the roles that they are expected to perform.