The Craft of Doing Qualitative Research in Prisons

In this article we examine the characteristics, challenges and added value of qualitative prison research in a Belgian context. As the many dynamics and challenges of qualitative research are often underreported in academic publications, we pay particular attention to the research processes and the pains and gains of qualitative prison research. Firstly, drawing on experiences from several prison studies, we describe the different steps of gaining access to the field as a constant process of negotiation. Secondly, we discuss some of the dilemmas of prison research based on two ethnographic studies of prison staff. We end with discussion of the value added by a qualitative research approach to facilitate understanding of what is at stake in prisons and how this fits with a critical research position. two ethnographic studies of prison staff in that country. The process of gaining access to the field is described as a constant process of negotiation. We discuss the added value of qualitative research to understanding prison life and how this fits with a critical research position.


Introduction
Prisons are generally recognized as particular, complex and conflicted institutions. Fifteen years ago, Liebling (1999: 163) described prison research as 'an enterprise made complex by the human nature of the researchers and the researched. It is an intense, risk-laden, emotionally fraught environment. It makes demands on fieldworkers which are at times barely tolerable'. Until recently, literature focusing on the process of doing qualitative prison research and on the challenges this brings for the fieldworker was rather scarce and the craftsmanship 1 of doing prison research often faded away into long footnotes, appendices of books or research reports or in informal conversations between researchers (see for example Jacobs 1977;Mathiesen 1965;Sykes 1958). Interesting reflexive accounts were provided though, concerning, for example, the ethical and methodological implications of the 'choosing sides' dilemma (whether to side with staff or prisoners; or with some factions of prisoners rather than others : Jacobs 1977: 215-229); and the process of gradually engaging with and being trusted by long-term prisoners Taylor 1972, 1980). Today, prison studies more often pay explicit attention to the research process itself and to the position of the researcher in gathering and analysing the data (Bandyopadhyay 2010;Bosworth 1999;Crewe 2009;Genders and Player 1995;King and Liebling 2008;Liebling 1999;Liebling assisted by Arnold 2004;Liebling, Price and Elliott 1999;Nielsen 2010;Piacentini 2012 2010; Scheirs and Nuytiens 2013). Jewkes (2012Jewkes ( , 2014 points to the comparatively late awakening of academic attention in criminology to the emotional investment of the ethnographic researcher and pleads for greater emotional honesty in writing about prison research. The research team Penality and Society of the Vrije Universiteit Brussel (Belgium) prioritizes qualitative prison research, as this best suits our research questions which relate to understanding prison life, work, cultures and regimes, with a particular interest for staffprisoners interactions (Beyens and Boone 2013;Claes 2012;Snacken 2005;Snacken et al. 2000;Tournel 2014). Research is always guided by a set of beliefs about the world and about how it should be understood and studied (Bottoms 2008). These assumptions influence where and how we look for relevant facts and what we perceive (Bauwens, Kennes and Bauwens 2013;Dicristina 2006). They shape and rationalize decisions made during data-collection and dataanalysis and thus underpin our methodological approach. The net containing these ontological, epistemological and methodological premises may be termed 'paradigm', 'interpretive framework' (Denzin and Lincoln 1998: 26) or 'perspective'. Our research is grounded in a social constructionist perspective, which assumes a relativist ontology ('multiple realities' are constructed by the researcher(s) as well as by the participants in the research), a subjective epistemology (researcher and participants/respondents co-create understandings), and a naturalistic set of methodological procedures (Bauwens, Kennes and Bauwens 2013;De Koster et al. 2004;Lincoln 1998, 2005).
Given the congruence of these premises, our preference for qualitative research stressing 'the socially constructed nature of reality, the intimate relationship between the researcher and what is studied, and the situational constraints that shape inquiry' (Denzin and Lincoln 2005: 10) is obvious. The ethnographic researcher participates in people's daily lives for an extended period of time. Therefore observations and open or semi-structured interviews with staff as well as with prisoners are our preferred research methods. They allow us to study people in their natural surroundings, examining their everyday behaviour, interactions, routines and rituals, along with the artefacts and symbols that bring meaning to their lives (Noaks and Wincup 2004). This makes the researcher the principal research instrument, as (s)he establishes field relations and conducts and structures observations (Claes et al. 2013).
Our preference for qualitative research methods and particularly ethnography fits with this constructionist paradigm. We aim to better understand the cultural meanings of social actions to actors and normative bonds, which are often important in everyday social life (Bottoms 2008). This interpretative understanding requires attention to formal and informal interactions and a sensitive analysis of the particular in the local and social embeddedness of the routines and habits of the researched group. A long-term approach to slow and deep immersion in the penitentiary world is required to develop an understanding of and to interpret prison cultures; the experiences of staff and prisoners; their mutual interactions and their interactions with rules and regulations imposed from above; the impact of the implementation of new managerial structures; prisoners' rights; and new early release decision-making structures This has resulted in case studies which show, irrespective of the formal research aims and objectives set by the authorities, that each prison has its own dynamics, routines and cultures which are formed by the specific composition of the prison population and of staff characteristics and their interactions. Our choosing to prioritize qualitative research methods does not imply, however, that this approach should be used to the exclusion of the use of quantitative data found, for example, in judicial or penitentiary files or reports of disciplinary hearings (see for instance Snacken et al. 2000). two ethnographic studies of prison staff in that country. The process of gaining access to the field is described as a constant process of negotiation. We discuss the added value of qualitative research to understanding prison life and how this fits with a critical research position.

Gaining access
Getting free and easy access to the field is indispensable for qualitative researchers and particularly for ethnographers. The procedure can be time consuming and various gatekeepers must be convinced of the soundness of the research and the legitimacy of the researcher or her/his supervisor. In Belgium, receiving formal permission from the Prison Administration is a rather easy and smooth process compared with other jurisdictions. We have never been hampered in the research that we wanted to do, neither at a political level, nor at an administrative or operational level. This is probably also related to the fact that the Belgian government has no predetermined prison research agenda. 2 Receiving formal permission does not mean, however, that the field is prepared to receive and embrace the curious researcher. The researcher's way to the heart of the prison is paved with obstacles that need to be negotiated on a daily basis. A prison is dominated by security and safety issues and fieldwork must be negotiated on the floor, taking safety and security concerns into account. However, this does not mean that the researcher has a (legal) guarantee of protection, or that (s)he is protected when something goes wrong. To some degree fieldwork is always at some risk to the researcher. 3 However, the hardest challenge for the fieldworker is being accepted on the landings: 'Obtaining the right information and being trusted by the research subjects in the field is another, much more challenging job, which is terribly demanding, time consuming, and exhausting' (Beyens 2013: 17). In the following section we elaborate this complex and intense process of achieving and maintaining trust by prison staff, based on the experiences of two authors of this article and insights from literature.

The pains and gains of doing prison research
At the start of fieldwork, negotiations with local gatekeepers centre round 'the kind of person' you are. Several authors describe the prison structure as making it difficult for an outsider to assume anything other than a marginal position, preventing him/her from becoming 'one of them' (Crewe 2009;Jacobs 1977;Mathiesen 1965;Rowe 2014).
Tournel (2014) experienced the complexity of doing research in a 'low trust environment' (Liebling assisted by Arnold 2004) with prison officers who were highly suspicious of outsiders in general, and of researchers in particular. She found, as in earlier research, that prison officers felt unrewarded, both by their superiors and by the outside world, had a strong sense of social isolation (Crawley 2004) and were convinced that 'others' care more about prisoners than about them (Arnold, Liebling and Tait 2007;Jacobs and Retsky 1975). A long tradition of scholarship depicting prison officers through the eyes of prisoners or the prison administrators instead of focusing on their own issues has resulted in mistrust of academics (see for example Liebling, Price and Elliott 1999;O'Connor 1976;Toch 2011). This was exacerbated within the Belgian context by the recent implementation of the Prison Act, 2005 which granted some formal rights to prisoners and which was seen as the product of 'unworldly' academics.
Consequently the researcher was identified with this strongly rejected and even despised prison IJCJ&SD 69 Online version via www.crimejusticejournal.com © 2015 4(1) reform which produced additional feelings of negativity towards her. She describes the process of gaining and maintaining trust as complex, time-intensive, exhausting and dynamic, with various stages presenting ethical, emotional and gender-related challenges. Three important aspects in gaining and maintaining trust are discussed here: introducing the researcher and the research; self-disclosure; and reciprocity. Drawing on the fieldwork experiences of Tournel and Kennes, we reflect about these three issues and look for similarities and differences between the two studies.
The researchers' role: From naïve student towards a competent professional When introducing themselves, Tournel and Kennes assumed the roles of rather naïve PhD students, new to the prison world and willing to learn from experienced prison actors in order to gain trust. The literature frequently suggests the adoption by the researcher of a rather passive, submissive role towards research subjects, taking care not to cultivate an image of incompetence to the 'point of being seen as a bumbling idiot who is not to be taken seriously' (Gurney 1985: 43). This is, however, difficult to balance and Gurney (1985) also points to the dilemmas faced by female (feminist) researchers trying to build rapport because they need to prove themselves to be nonthreatening and at the same time be seen as a competent professional. Tournel thus felt uneasy when confronted with sometimes wrong assumptions and prejudices in the minds of the prison officers which she could not challenge.
Once accepted, the fieldworker is advised to discard the naïve role and to take on a more competent, professional role. After six months of fieldwork, officers expected some feedback from Tournel, which she offered after a three week analysis of her preliminary data. Selfpresentation hence occurred within the interaction between researcher and research subjects (Gobo 2008;Jacobs 1977;Shaffir 1991).
Contrary to Tournel, Kennes did not experience any form of suspicion or hostility at the outset. He quickly established positive contacts with most prison officers who appreciated his committed approach and intentions for a long-term presence in the prison. Although his male gender might have helped acceptance, he too was confronted by the tension he experienced between playing his naïve student role and his identity as the academically informed researcher. He sometimes received questions about his data-gathering (how did he memorize all things happening in prison, did he write everything down at home, and so on), but these seemed to fade away as time went by. Nonetheless, this transition from being a 'naïve' researcher (in order to stimulate prison officers in explaining the way they do things) towards a competent professional was certainly not straightforward.
Another striking aspect was the research subjects trying to control the researcher. For example, Tournel experienced being observed by prison officers who told her that they knew which car she drove and which route she took to go home. They admitted that they had 'Googled' her and had discovered that she had given public lectures on prison issues and had published with her supervisor, whom they situated in the pro-prisoner camp due to her involvement in drafting the Prison Act, 2005. This last 'discovery' made it all the more difficult to maintain a naïve research position, but it also forced her to take a side. Kennes also had to tackle the 'stigma' of being a criminologist who, in the eyes of many prison officers, chose the side of prisoners. 4 Both researchers thus had to defend themselves in order to neutralize that stigma.
These examples show the unpredictable nature of fieldwork, which causes additional unforeseen stresses and forces the researcher to be creative on the spot in order not to undermine his or her research role in several possible ways.

Choosing and switching sides
Most of the prison studies that have been conducted over time in our research centre at the Vrije Universiteit Brussel have included the investigation of more than one actor in the same study (prisoners, prison officers, prison governors, and so on) in order to understand prison experiences from different angles (see for example Beyens and Boone 2013; Claes 2012; Kennes forthcoming; Snacken et al. 2000). Whether this approach is appropriate is, however, subject to discussion (Becker 1967;Liebling 2001). We agree with Liebling (2001) that studying more than one side is possible and even recommended from a constructionist perspective. The methodology may depend on the research questions but perhaps, even more so, on the characteristics or culture of the subject prison. A multiple sides approach may not be possible where strong tensions or real conflictual relationships exist between groups in prison (see for example Jacobs 1977). That was the reason for Tournel (2014) deciding to choose sides and devote her fieldwork exclusively to prison officers. She felt that including prisoner and/or governor views on the work performed by the officers could compromise the study. In other words, in the minds of prison officers, it would have confirmed the 'pro-prisoner' stigma that they assigned to academics and which enhanced the officers' initial fear that she might be spying for the governors or the prison administration and thus could use her research findings to damage them. Therefore her choice to focus uniquely on the officers seemed the most rewarding one. This can be illustrated by a single event. A brief encounter in the corridor with a prisoner governor, who simply asked how her research was going, was observed by officers. This resulted in repeated hostile remarks from the officers: 'Ah, there she is, our spy'; or 'maybe you can ask Hanne because she is good friends with the governor'. Recovering from this trust gap took time, necessitating constant proof of her independence, loyalty and engagement. Researchers obviously do not escape a prison officer culture which insists that there are hidden meanings in everything that happens in prison, however unimportant and inconsequential this might seem to an outsider.
The fact that Kennes succeeded in including prison officers and governors in his study can be explained by their less oppositional relationship in the prison where he conducted his research. However, this choice was very demanding from the start and sometimes put him in precarious situations. He decided to start his observations with the prison officers, but mentioned from the beginning his intention to include more than one side in his research and to involve principal officers and governors later on in his study. By doing so he sought to avoid having research subjects who had come to trust him feeling betrayed afterwards. He still had to contend with, though, tensions between basic grade officers and wing officers on the one hand and principal officers, who supervise the former, on the other. Prison officers even spoke of two (opposing) 'clans' surrounding two principal officers. Researchers have to take these types of interactions into account while balancing relations in the field. Being invited for an informal event, being friendly to somebody from another clan, overhearing gossip, and so on are daily occurrences that can lead to explosive situations which have to be handled carefully. If we include different sides in one study, keeping this 'switching sides' aspect constantly in mind while being in the field is important. This requires consideration in advance on how to organise the fieldwork. Kennes decided to complete fieldwork with one group before commencing with another so as to observe different groups separately. So he first spent time with the officers, and then moved on to the governors. He tried to ease this transition by building a reputation of discretion and confidentiality throughout his contacts with prison staff of all ranks. As Hammersley and Atkinson (2006) indicate, considerable thought must be given to management of impressions. Impressions hampering access must be avoided or countered as much as possible, while those which facilitate should be encouraged, within the limits set by ethical considerations. Switching sides and having to cope with loyalty issues is challenging and can even be distressing for the researcher who investigates multiple sides within the one prison.
IJCJ&SD 71 Online version via www.crimejusticejournal.com © 2015 4(1) Self-disclosure: Not only a personal, but also a situational and geographical determined choice Self-disclosure or the unveiling of commonalities and information about the researcher to the research subjects is important for building rapport, for instilling a non-hierarchical relationship, and for gaining respect as a person (Bosworth et al. 2005;Dickson-Swift et al. 2007). In order to be accepted the researchers decided from the start to be as open as seemed appropriate about where they lived, about their weekend activities, and so on. In fact, avoiding some reciprocity about personal matters would have been impossible and unnatural, especially when following and observing the same prison officer or governor for more than a month. This personal investment also enhances the development of a constructive, professional rapport and can lead to friendships, a factor mentioned by Liebling (1999Liebling ( , 2001) and, a decade earlier, by Taylor (1991: 238): 'The objects of our studies are not objects at all. They are people who may become attached to us and to whom we may become attached'. In this age of social media separating personal and professional lives becomes even more complicated for the researcher. Researchers are searched on the internet by the researched and are invited to become friends on Facebook.
There are no guidelines for these new dilemmas and researchers react to them according to their personal values. Moreover, geographic proximity of the researcher's residence to the prison can be important. Tournel lived proximate to the prison: some officers knew relatives of hers, and she met others while shopping in the same stores or participating in cultural or other local activities. Hiding her personal life became therefore more untenable for her than for Kennes who did research further away from his hometown.
This brings us to the issue of how to avoid 'going native' (that is, assuming the cultural traits of those being researched) and instead balance on the thin line between being sufficiently involved and remaining appropriately detached from our research subjects. This difficult balancing act is discussed in several ethnographic studies. 'Research in any human environment without subjective feeling is almost impossible (particularly, I would argue, in a prison)' (Liebling 1999: 149). While we agree with Snoek (2008) who states that a good researcher needs to be involved, not getting too close or not being occasionally overwhelmed by the feelings or experiences of research subjects can be difficult for researchers. Too close an association only becomes problematic when the research subjects' views, rhetoric or behaviour uncritically take over. For example, after some months of fieldwork, supervisors warned one researcher about a propensity to copy the prisoner officers' depersonalizing language with regard to prisoners. Support for the researcher was received through sharing field experiences with colleagues, which also initiated a process of (self) reflection (Phillips and Earle 2010;Scheirs and Nuytiens 2013). Withdrawing from the field for three weeks also helped to preserve emotional well-being after some distressing incidents with prisoners had been experienced. Several authors recommend inserting one or more 'time out' phases when doing intensive prolonged fieldwork (Dickson-Swift et al. 2007;Gilbert 2001;King 2000). Aware of the fact that researchers and participants co-create understandings of the studied social world (cf. subjective epistemology), we ask whether it is always possible to recognize when we are getting too involved. The advice of experienced colleagues can be beneficial.

Quid pro quo: About reciprocity and engagement
The importance of reciprocity and engagement has been raised in (feminist) methodological literature (Dickson-Swift et al. 2007;Gurney 1985;Harrison, MacGibbon and Morton 2001).
Since the prison is a controlled environment where little input of outsiders is allowed, only limited forms of reciprocity are possible. We identified six that were practiced. The first was to shake hands -an important cultural habit -with all officers, irrespective of rank. Secondly, Tournel's presence and 'listening ear' provided a space and forum where officers could tell personal stories and express feelings (Arendell 1997;Lee 1997;Liebling 1999 feelings of anxiety, grief or guilt within a culture that is intolerant towards human vulnerabilities. A prison officer thus told her how he still suffered from an incident of severe physical aggression by a prisoner. The incident still puzzled him because there was no particular reason for this aggression. Whether this form of disclosure is typical of the female researcher in a male prison is not clear. 5 Thirdly, showing commitment to the field by participating in all shifts, including night services and weekends, created respect in the eyes of research subjects. Fourthly, both researchers looked for jobs to assist the officers in their daily work to gain trust and respect (see also Declercq 2005). They elected not to carry keys in any situation (see also the discussion by Crewe 2009;Earle 2014;Jewkes 2014;King and Liebling 2008) but assisted in the distribution of meals to the cells. This increased their legitimacy as committed researchers who wanted to understand the work of the prison officer. Finally, the manner of leaving the field has to be considered and discussed in advance. To respond to the many questions of prison officers during the period of the fieldwork, Tournel organised a collective feedback session about her findings for all prison officers involved in her study. The decision to ask for feedback was primarily motivated by ethical concerns of not leaving the field abruptly (Taylor 1991), but could also be a test of the adequacy of the findings. The prison officers essentially concurred with her conclusion that prison officers are subjected to detention harm. We agree however with Rock (2011) who stated that the comments of insiders can often be helpful, but that they sometimes can be swayed or converted or too polite to contest a researcher's conclusions. From this position the opinion of the research subjects about the conclusions may only be regarded as one voice and not as the final arbiter of an account.

Fieldwork as an emotional minefield
In accordance with Nielsen (2010) we emphasize the dynamic character of the process of building trust, which is not only related to incidents that may challenge confidence but which is also time-bound. Emotional challenges are not limited to this trust building process but are inherent to (doing research in) a prison environment (Jewkes 2014;Liebling 1999;Phillips and Earle 2010;Scheirs and Nuytiens 2013). Prisons are places of suffering, machismo, verbal and physical aggression, crime and abuse. Personal (emotional) demands were omnipresent in both prison studies: these were associated with tiredness due to shift work; testing behaviour from different stakeholders; and so on. During fieldwork, incidents such as suicide attempts or acts of verbal and physical aggression may provoke intense emotions in the field, including for the researcher, and it would be naïve to ignore this or expect that the researcher can escape these tensions. Emotional challenges were particularly related to gender (see also Claes et al. 2013). Tournel was subjected on a daily basis to numerous sexualized gestures and remarks, shouting, name calling, and so on from prisoners. This caused additional stress and, she noted, made her irritated and angry, and also instigated a more punitive attitude towards prisoners than she had anticipated, even though this ran counter her personal beliefs (Cowburn 1998). On the one hand, name calling, whistling, shouting sexist or derogatory remarks and making sexualized gestures seem so commonplace in a prison context that their use goes unquestioned. On the other hand, while male prison officers never uttered direct sexual remarks or propositions, they often indirectly signalled her subordinated position as a female (Gurney 1985). Realizing that she had overheard sex-related jokes between officers about her, they told her: 'You have to be able to deal with that kind of behaviour in a prison huh [laughter]'. Even formulated as a joke, such comments also hold a hint of hostility (Nielsen 2011). Reactions to this kind of behaviour can vary from fatalistic to confrontational, but the negotiation of which position to take is difficult and has consequences at a personal level, as well as influencing the 'breaking or making' of the research. The researcher seems to lose irrespective of the stance taken: either she compromises her beliefs by being non-confrontational; or she risks compromising the research project by not conforming to the misogynist macho culture. Contrary to Tournel, Kennes did not experience emotional challenges in relation to his gender, and we agree with Liebling (1999: 160) that there can be distinct differences between man-to-man and man-IJCJ&SD 73 Online version via www.crimejusticejournal.com © 2015 4(1) woman conversations. As Jewkes (2012) indicates, accounts of emotional difficulties experienced by male researchers remain rather invisible in literature. A possible explanation might be that these experiences are more commonly reported from fieldwork situations with female researchers in a male prison and that gender issues are more striking in interactions between female researchers and male prisoners. Jewkes (2012) rightly warns us not to focus too narrowly on negative emotions, but to be attentive to more positive experiences like the creation of friendships, acts of solidarity, chivalry of officers, and so on.
The account of the emotional challenges and their impact on research is paramount from an epistemological and methodological point of view, and like Kleinman (1991), these experiences gave direction to our fieldwork and analysis. Liebling (2001) has argued that the mutual experience of such incidents and emotions is crucial in gaining trust and respect from the research subjects (cf. 'affectivity'); other authors use the label 'emotional engagement' (Coffey 1999;Kerosuo 2007;Lofland and Lofland 1995). Being auto reflexive about these experiences enhances the understanding of the difficult position of the researcher in a prison setting. But we also agree with Crewe (2014: 401) that 'autoethnographic reflexivity and emotional disclosure come more easily to some researchers than others' without necessarily leading to better research.

Reflections
Without wanting to contribute to unfruitful methodological 'pigeonholing' (Bottoms 2008: 81), and despite the challenging nature of qualitative research, this type of approach is important for making 'intelligible' (Liebling 2014: 481) what is at stake in prison. Qualitative research not only fits better with our constructionist approach, but also with our personal preferences for a 'humanistic style of research', which tries to depict the painful and emotional aspect of the social practices we study (Liebling 2014: 481). Quantitative data can describe some aspects of prison life and are useful if we want to make statements about certain facts, incidents, categories or decisions, but are less appropriate to make sense of the cultural, hierarchical, social and emotional dimensions of life and work in prison. From our experiences, prisoners and staff prefer qualitative data collection from the standard measurements (see also Liebling 2014: 481), because they experience it as more rewarding.
But do the efforts of the qualitative researcher lead to 'better science'? It certainly leads to a different form of knowledge, and a quantitative-qualitative divide in some academic jurisdictions means that the battle for scientific recognition endures. This is illustrated by the more quantitative American versus more qualitative British approach to prison research. In this regard Jacques (2014) observes that qualitative research is published less frequently in American and international criminology and criminal justice journals than quantitative research. Qualitative research requires a longer time perspective, but Jacques (2014) also sees language differences between quantitative and qualitative researchers as a possible explanation for the latter's lack of academic esteem. As people mostly use a qualitative style of language to communicate on a day-to-day basis, qualitative research findings may be perceived as less sophisticated scientifically than the positivist approaches. As Liebling (2014) and the other contributors to Jewkes' (2014) special issue of Qualitative Inquiry on 'Doing Prison Research Differently' illustrate, there is great potential for storytelling or narrative accounts to humanize both the research process and the penal world. Telling stories and explaining narratives take time and need space. However, to meet the standards of academic journals, these stories have to be adapted to the formats of classic articles and summarized to 5000 to 8000 words. Further, the conditions for presenting a research project in order to get funding are strict and the standard selection and success criteria are not well adapted to the requirements and pace of qualitative research. These impediments have to be linked with the current state of academia, with its pressure to produce quick and quality output via publications in highly ranked journals, IJCJ&SD 74 Online version via www.crimejusticejournal.com © 2015 4(1) which are not always sympathetic to or familiar with the purposes and approach of qualitative research. In reaction to this development the Belgian philosopher Isabelle Stengers (2013) has voiced 'a plea for a slow science'. She interprets the 'slowing down' as being able to 'lose one's time' with questions that do not directly contribute to the immediate and evaluable progress of the field. We have to accept that 'messiness' is not the same as 'defect' but 'as what we have to learn to live and think in and with'. This is contrary to the straightforwardness that is often demanded from researchers nowadays. To be funded, an important evaluation criterion, for example, is the feasibility of the research, which leaves less freedom for exploring new questions or 'slow' qualitative research methods. To convince the reviewers in the context of accountability, every research step has to be calculated in hours, days, weeks, and months, and access to the field has to be prepared and ensured in advance (see also Liebling 1999Liebling , 2014. While good preparation for new research projects is important -and accountability is essential -the space for searching and failing is still a crucial part of doing research (Beyens 2013).
Although we choose the qualitative approach with conviction, we also underline the limits of immersion in the carceral world. We agree with Earle (2014) who points out that there is a difference between 'serving time' and 'spending time' in prison. Whatever researchers might do, their experiences cannot be compared with that which prisoners and also prison staff undergo and feel. There always will be social distance between the research subject and researchers, many of whom have socio-demographic characteristics that are far removed from that of most prisoners. This realisation can be humbling for researchers. Being 'in' and even being accepted does not mean that the social and cultural distance between researchers and research subjects can be ignored or eliminated. This distance also extends, but to a lesser extent, to research with prison officers as subjects.
Prison researchers are also caught in crossfire, having to choose sides (Becker 1967). We have discussed how the empirical study of multiple sides is possible under the right circumstances. And contrary to Scott (2014) we do not see an ethical dilemma in studying the position of the prison staff and prison managers, who are indeed in a hierarchical superordinate position compared to the prisoners. Scott narrows the question of choosing sides for the prisoners to the question of 'less or more suffering'. The reason why he denounces research with prison staff is because he believes they 'fail to acknowledge the greater suffering of those below them' (Scott 2014: 31). He states that the voices of those in power are deemed to be more legitimate: … research should not be a process of reconciliation or aim to justify the practices of the penal apparatus of the capitalist state, improve human resources management or some further utilitarian goal. Nor must it be to uncritically reproduce or condone exploitative power relations or naturalise their position. (Scott 2014: 31) It is the researcher's duty to uncover exploitation but if critical criminologists only study those who 'suffer most', this suggests that the suffering of prison staff is of lesser or no importance. It is the responsibility of the researcher to not only try to understand and explain how and why exploitation occurs but also how negotiation takes place in a modern total institution. Giving voice to the experiences, the constraints and worries of prison staff or of the 'powerful' while exercising their power has to be clearly distinguished from speaking the language of the 'officials'.
Studying the microphysics of prison life does not discharge the critical penologist from putting on the political agenda the question of legitimacy of the use of imprisonment. As criminologists we should never lose sight of the broader political and macro sociological picture of the political economy with its rising prison populations. Moreover, going into the prison and speaking with all the stakeholders gives the researcher a more legitimate position to take a stand in the IJCJ&SD 75 Online version via www.crimejusticejournal.com © 2015 4(1) political debate on reducing or abolishing the use of imprisonment. And although we have to be circumspect about our abilities to influence penal policies, our empirical research experiences give us an authoritative voice when participating in the public debate concerning imprisonment and for putting systemic questions with regard to expansionist policies on the agenda. Our criticism of the prison system and penal policies has never prevented us from doing the research of our choice. Being a critical criminologist goes beyond writing academic articles or reports about the results of a particular research project. Rather, there is the constant endeavour of writing opinion pieces in newspapers, debating with politicians and prison administrators in the media and at conferences, and of participating in discussion about radical alternative projects by and with practitioners (Claus et al. 2013