Researching Imprisoned Persons: Views from Spain and Latin America

Abstract

Empirical research in criminology poses many challenges, as the present book attests. This is especially true when researching vulnerable populations such as the group addressed in this chapter: imprisoned persons. Carrying out studies with imprisoned persons has its own set of difficulties due to the context in which research is developed—the prison—and the vulnerabilities of the studied population—the prisoners—, who usually come from disadvantaged backgrounds and are subjected to the power of prison authorities. In this chapter, we use our experience conducting fieldwork inside prisons, mainly in Spain, but also in some countries of Latin America such as Brazil, El Salvador and Mexico, to reflect upon the specific challenges and ethical dilemmas of researching imprisoned persons. First, we address the difficulties of gaining access to prisons and the main obstacles we may find when dealing with the different gatekeepers and a distrustful study population. Secondly, methodological challenges are explored, with a specific focus on conducting surveys and interviews with inmates. The third section discusses the ethical and emotional aspects of prison research. The chapter concludes by reflecting upon the lessons learned from conducting research with imprisoned persons inside prisons.

1 Introduction: Researching Imprisoned Persons

Empirical research in criminology poses many challenges, as the present book attests. This is especially true when researching vulnerable populations such as the group addressed in this chapter: imprisoned persons, who usually come from disadvantaged backgrounds and are subjected to the power of prison authorities. Research on imprisoned persons is characterised by the variety of topics and methodological approaches adopted by prison researchers.

The first studies on imprisonment were sociological accounts of life in prison and the inmate subcultures that flourished inside: The Prison Community (Clemmer, 1940) or The Society of Captives (Sykes, 1958). Later, the lives of imprisoned persons—rather than the prison as an institution—became the focus of attention of psychologists and criminologists. These scholars have explored diverse topics such as inmates’ adaptation to imprisonment, prisoners’ relationships with prison staff, prisoners’ well-being, inmates’ misconduct, the use of solitary confinement, the effectiveness of prison policy and rehabilitation programmes and the effects of imprisonment on recidivism (see, e.g. Johnson et al., 2016; Jewkes et al., 2016; Wooldredge & Smith, 2018). Furthermore, all these issues have also been studied focusing on the specific prisoner populations such as female, long-term, older or mentally ill prisoners, or through the lens of race.

Regarding methodological approaches, qualitative and, particularly, ethnographic enquiry have provided significant insights on the meanings and implications of imprisonment (e.g. Cunha, 2014; Drake et al., 2015). On this point, it has been highlighted that qualitative research is the most appropriate approach ‘to make sense of the cultural, hierarchical, social and emotional dimensions of life … in prison’ (Beyens et al., 2015, p. 73).

On the other hand, quantitative approaches aimed at measuring, for example, prisoner psychosocial adaptation to imprisonment (Toch, 1977) or the Quality of Prison Life (Liebling, 2004) by means of surveys, questionnaires or structured interviews have also been relevant to our understanding of prisoners’ experiences. In this sense, quantitative, longitudinal studies are particularly important for unveiling the accumulative effects of imprisonment on prisoners’ lives and well-being, and on their future criminal behaviour (e.g. Dirkzwager et al., 2015).

Finally, it needs to be noted that although most research on prisoners has been developed by academics, convict criminologists have provided an invaluable contribution to the literature on imprisonment by combining personal experience with critical perspectives on penalty (Earle, 2016; see, for example, the Journal of Prisoners on Prisons).

Besides the usual challenges of criminological research, carrying out such investigations has its own set of difficulties due to the very nature of the penitentiary institution—i.e. high security, thick bureaucracy, strict routine, lack of transparency—and the characteristics of the prisoner population—coming mostly from disadvantaged backgrounds and being subjected to the power of prison authorities. In fact, there is plenty of literature reflecting on the challenges and experiences of prison research (e.g. Beyens et al., 2015; Liebling, 1999; Liebling et al., 2021). However, to our knowledge, none specifically captures the realities of Spain and Latin America.

In this chapter, we use our experience conducting fieldwork inside prisons, mainly in Spain, but also in some countries of Latin America, to reflect upon the specific challenges and ethical dilemmas of researching imprisoned persons in our contexts, with the aim of adding a new perspective to the above-mentioned literature. More specifically, we have conducted more than 400 surveys and interviews, inside closed prisons and also outside them, with people in semi-freedom conditions and on parole. In these interviews we have explored inmates’ perception of the quality of prison life (Güerri & Alarcón, 2021; Rodríguez et al., 2018), the pains of open prisons (Martí, 2018, 2019) or the role of prison officers (Güerri, 2020). One of the authors has also conducted research in prisons in Latin America; specifically, the countries of Brazil, El Salvador and Mexico.¹

The chapter is structured in the following parts. First, we address the difficulties of gaining access to prisons and the main obstacles we may find when dealing with the different gatekeepers and a distrustful study population. Secondly, methodological challenges are explored, with a specific focus on conducting surveys and interviews with inmates. The third section discusses the ethical and emotional aspects of prison research. Finally, we reflect upon the lessons learned from conducting research with imprisoned persons inside prisons.

2 Conducting Fieldwork in Prison: A Constant Process of Negotiation

One of the challenging aspects of conducting research with imprisoned persons is accessing the field and getting prisoners’ consent for their participation in the study, since there are many people who need to be convinced in successive stages. In fact, doing research in prison has been described as a ‘constant process of negotiation’ (Beyens et al., 2015, p. 68) in which we need to persuade several gatekeepers: first, we need to be authorised by the Prison Administration to access the prison; secondly, the specific details of in-site research will have to be negotiated informally with both prison management and prison officers and, finally, we need to gain the trust of prisoners.

2.1 Accessing the Field: Obtaining the Authorisation of the Prison Administration

Gaining access to the field is the first difficulty that we face when undertaking research on incarcerated people. Authorisation by the Prison Administration is required to enter a prison. Moreover, this authorisation also determines the degree of freedom that we will have to carry out our investigation, which becomes especially important in the case of qualitative approaches.

Obtaining permission to conduct research in prison is a rather challenging task. Depending on the jurisdiction or the political context, prison administrations may be more willing or more reluctant to grant access to researchers from the outside (González, 2012). For example, the opposition faced by Spanish researchers in the nineties (Ríos & Cabrera, 1998) contrasts with the ease of the process in Belgium (Beyens et al., 2015). Similarly, Wacquant (2002) denounced that the generous access granted to US prison sociologists until the seventies was withdrawn in the wake of the age of mass incarceration. In some Latin-American countries, it is difficult to know who to contact for authorisation and even then, there may not be any response or reason provided for rejection.

The main reasoning used by prison administrators for rejecting or imposing restrictions on researchers is security. The argument of security is frequently used in the prison context, from restricted access to activities and products—as King and McDermott’s (1990) humorous anecdote on ‘subversive geraniums’ demonstrates—² to turning down visits of journalists or inmates’ family members (Güerri et al., 2021). Thus, it is not surprising that we have faced difficulties such as being prevented from entering with voice recorders (which stands in the way of conducting in-depth interviews) or that it has been suggested that we exclude high-security wings from our research—all in the name of security.

Moreover, our authorisation to access and spend time inside prison may be limited due to a lack of sufficient human resources to guide us through the institution and ensure our safety without disturbing prison routines (see also Field et al., 2019). We have also been denied access to inmates’ files alleging data protection laws or, more recently, prevented from entering due to Covid-19.³

In addition, obtaining authorisation is usually a long process. First, the stipulated procedure to get the research project authorised requires sending a request detailing aspects such as the main objectives of the research, the methodology, the prison resources to be used, or the approval of the ethics committee of the research institution. Once the request is sent, the Prison Administration may take several months to respond, and then we may be asked to make adjustments that may require resending the request (see also Field et al., 2019).

A difficult, and even problematic, issue related to the need of an authorisation to enter the prison is finding the balance between the purpose of our research and the goals and limitations of the Prison Administration. For example, in Spain, where there is a considerable discursive emphasis in the constitutional goals of re-education and reintegration, we may find ourselves justifying our project to the administration in terms of rehabilitation, even when this is not the objective of our research. Similarly, in Canada, Hannah-Moffat explains that critical criminologists may have difficulties gaining entry to prisons as their research may be deemed ‘insufficiently practical’ or ‘too political’ (Hannah-Moffat, 2011, cit. in Kaufman, 2015).

2.2 Arriving to, Entering and Navigating the Field: Dealing with Prison Staff

Once authorisation is granted, we need to get to the prison where our research will be conducted. The first thing one realises is that getting to a prison requires a lot of time and resources because closed prisons are situated in isolated and remote areas far from the city. Moreover, there are no convenient options of public transportation and, thus, you must have a car if you want to avoid wasting hours on a bus and being restricted by unsuitable timetables.⁴ The situation is different regarding open prisons, which are usually located in well-connected areas on the outskirts of the city because many inmates leave daily to go to work.

In our experience, the commencement of fieldwork in prison varies greatly depending on the managing style of the prison manager. On the first day, some managers would meet with us and show interest in the goals of the project, while others would limit themselves to processing the paperwork that was needed to give us our visitors’ ID. These first encounters with prison managers are also a new instance of negotiation where the spaces and times of the research project or the way in which inmates would be selected and approached are discussed.

This stage of negotiation is key for the methodological soundness of the research project, especially when it includes more than one site, as some prison managers may be more flexible than others, and even impose new conditions and restrictions to the research plan. For example, on one occasion (see Güerri, 2020), a prison manager did not allow the use of a voice recorder to interview prison officers on prison grounds—although it had been authorised by the prison administration—and it was only allowed for interviews at the external offices.

Moving around prison while conducting research is arduous due to the security measures that restrict free movement on the inside. In other contexts, it has been discussed whether researchers should or should not carry keys. On the one hand, keys are a symbol of power that places the researcher ‘too close to staff’ (King, 2000, p. 305, cited in Drake, 2012). On the other hand, keys allow free movement and release the researcher from staff supervision (Drake, 2012).

In our case, this was never a choice to be made. The characteristics of the Latin American prisons we visited—high levels of violence, disorganisation and large spaces—required that our safety be ensured by staff accompaniment. Whereas in Spain, most prisons have automated doors that need to be opened by the prison officer working the control cabin and, thus, having keys was never a possibility. Instead, the difference depended on whether prison managers thought we needed to be escorted around the prison or were open to us moving freely, as we explain below.

When researchers depend on prison officers to move around, entering the prison means having patience and waiting—sometimes for a long time—until someone can come to escort them. Being accompanied to the wings by prison officers brings about the opportunity to engage in fruitful, informal conversations with them, but these occasions were also used by some officers to let us know that we were not welcome or that our presence interfered with their work (for example, ‘I’ve had to leave a colleague alone in the wing to accompany you’).

Being allowed to move around autonomously does not mean, however, that our movements are free and unrestricted since every doorway—from the main gate to cell doors—is controlled by prison officers. This implied having to show our visitor ID and, sometimes, a letter of safe-conduct signed by the prison director every time we wanted to move from one space to another, and we were usually required to explain who we were and what we were doing in the prison. Moreover, this process of introducing and explaining ourselves had to be constantly repeated due to changing prison officer shifts.

Regardless of if we were escorted or not, once inside the wing we had relative freedom to move around the common areas at will. Thus, this difference had no substantial impact in the research process and its potential results. It needs to be noted, however, that the research experience felt very different indeed: not being made to wait more than strictly necessary and having autonomy was a huge contrast with the times in which we were entirely dependent on staff.

In addition to movement restrictions, time constraints were also important during our research in Spanish prisons. First, our authorisation was for a limited amount of time, and we had to hurry to get all the interviews needed. Secondly, keeping the routine and its daily procedures is as important to the maintenance of order as the control of space and, for this reason, we were only allowed inside the wings during ‘activity time’.⁵ Consequently, time limitations prevented us from observing moments such as the opening and closing of cells or mealtime and, even if we could interact freely with inmates who stayed in the wing during activity time, it complicated the task of gaining inmates’ trust progressively.

2.3 The Last Gatekeeper: Gaining the Trust of the Studied Population

Inmates generally distrust outsiders (Field et al., 2019). Moreover, the characteristics of the researcher (age, gender, race) may influence the willingness of inmates’ wanting to talk and opening to us (Beyens et al., 2015). That is why the process of gaining inmates’ trust has been described as complex, time-intensive, exhausting and dynamic (Beyens et al., 2015). For example, Drake (2012) approached inmates informally and asked if they would like to participate in a formal interview. Afterwards, she provided a document detailing the goals of the research and the consent form. Finally, the date for the interview was set and the formal conversation took place.

Our experiences are quite different to those described in the above-mentioned literature. In Spain, our time constraints made it difficult to gain inmates’ trust in a gradual manner. On the contrary, we would directly call the person and sit with them in the designated space—i.e. the dining area or an office, depending on the prison. Being called by prison officers raised suspicion on some inmates (‘why have I been called?’), so we always explained that we were academics carrying out an independent research project and they had been randomly selected out of the complete list of inmates. Subsequently, it was made clear that their participation was completely voluntary and that their acceptance or refusal to participate would not have any consequences, either positive or negative, and, therefore, they could leave if they wanted to (only a few of them did). This was an important step as we did not want inmates to feel coerced into participating because they had been summoned by prison staff.⁶

Prisoner dynamics in El Salvador, characterised by the presence of gangs and strong power relations among prisoners, made it necessary to speak first with the inmates who had informal control of the wing (delegates of the wing and/or gang leaders).⁷ We presented the objectives and characteristics of the study to these delegates and leaders to gain their trust, emphasising that ours was an external research and that the authorities and judges would not have access to the data, that is, that the research was anonymous and confidential.⁸ After obtaining their approval, we asked the delegates/leaders to inform the rest of the inmates that we would be conducting these surveys in the following days and that we could be trusted. This served, first of all, so that the inmates would already be aware that an external survey was being conducted and so would not be caught off guard, which likely would have generated more distrust. On the other hand, it also served to ensure that the inmates already had the approval of their module delegates and/or gang leaders to participate in the study. Had we not talked to the inmate in control of the wing, it is unlikely that the prisoners would have participated in our research.

Even in the cases in which we have gained the trust of the studied population, there are certain topics that are difficult to explore. Previous research has reported prisoner reluctance to self-report in topics such as homosexual behaviour, drug use, or victimisation due to the existence of taboos or fear of reprisals (Field et al., 2019). In our research in both Spanish and Latin American prisons, we found that some of our interviewees were clearly uncomfortable responding to questions related to self-injuries and suicide attempts.

3 Employed Techniques

As mentioned above, in some contexts the use of ethnographies (and also large surveys) has predominated. In our case, the research in which we have participated has relied mainly on surveys and interviews, partly due to the limitations discussed in the previous section. For this reason, this section dedicated to the techniques employed in prison research focuses mainly on interviews and surveys, but without underestimating the importance of observation.

A key aspect of any research design is the selection of the sample. When researching imprisoned persons, we may face a number of difficulties which are related to the characteristics of the prisons and the prison population. These difficulties will depend on the type of sample we need, especially on whether we want it to be representative or not.⁹ We discuss four of the most common difficulties below.

First, as researchers it is not always easy to know in advance how the prison population is distributed in the different prisons of a region, as this information is not necessarily published by the prison administrations. This is important because we may be interested in interviewing one type of population (for example, people in pre-trial detention, female prisoners or people convicted of a violent crime) rather than another, and so we need to know how to find these people. In this regard, we researchers are confronted with the consequences of the lack of transparency that characterises prisons, a widespread concern due to the effects it has mainly on incarcerated persons.¹⁰

Thus, if we cannot have this information in advance, we may need to have an initial meeting with the representatives of the prison administration in order to find out how the prison system is organised and to be able to select the prisons we want to study appropriately. Furthermore, we may also need to meet with the managers of the selected prisons in order to acquire specific information about their inmate population to determine which wings will be the focus of our investigation.

Secondly, there are difficulties in capturing certain profiles of the prisoner population. For example, in many countries, including people in solitary confinement in the sample can be very complicated because security controls are more restrictive (see, for example, Drake, 2012) and prison authorities may be more reluctant to allow them to be interviewed. Along the same lines, certain inmates, such as mental health patients, violent offenders and women, are normally placed only in specific prisons and it is generally necessary ‘to go and look for them’. Sometimes there are restrictions to meet with ‘vulnerable’ prisoners, which in Kauffman’s case meant no interviewing those charged with sex offences (2015, p. 57). All in all, this can make it difficult for researchers to gain access to certain people and it may cause certain groups of prisoners to be excluded or under-researched. For example, Daniels et al. (2015, cited by Abbott et al. 2018) found low reporting of data collection processes in research with violent offenders.

Thirdly, in the case we need a representative sample, another challenge is the need to have the complete list of inmates of each prison.¹¹ For privacy reasons, it may not be possible to have an informative list (i.e. with names and surnames) outside the prison and, therefore, carry out the selection of the sample from our computers. There are two possible solutions for this: either we ask for an anonymised list using only their prison ID number, or we ask for permission to carry out the selection on a computer in the prison itself. The first option makes it more difficult to find the people selected when we start to carry out the surveys, since we won’t have their names, while the second involves asking the prison to lend us a computer and using part of the time we spend at the prison to carry out the selection in situ. Both options are useful and, in any case, depend on the prison, although in our opinion it is preferable to try to obtain the list with names because this facilitates the process of identifying the persons selected.

Moreover, we may not have the option of having the list of inmates, but the authorities may allow researchers to select the participants. In this case, one possibility to try to make our sample representative is to select in each wing, for example, one person out of five. However, sometimes it may not be possible for the researchers to make the selection at all because the prison authorities do not allow it and it is the prison authorities who choose the participants. In this case, there is no choice but to point out this difficulty in our research and to make it clear that this is a sample that is most likely not representative.

Finally, it is also common that at the time of conducting surveys in prisons, some of the people selected as part of the sample are not available because they have been transferred to another prison or module, are doing an activity outside the module, are on leave, have finished their sentence or are serving a temporary isolation sanction, among other reasons.¹² Furthermore, depending on the context, it is likely that some inmates refuse to be surveyed. For example, in certain prisons, such as the ones we visited in Spain, a few inmates were afraid of possible reprisals, either from the prison administration or from other prisoners, for ‘talking too much’. For all these reasons, it is necessary to have a list of randomly pre-selected substitutes ready.

Additionally, we need to be careful when choosing the method for our data collection. It is usual that prison researchers use self-administered surveys to collect inmates’ views. However, this approach ignores the fact that people who are illiterate cannot participate in self-administered surveys. For example, in Spain, 11.8% of those incarcerated are illiterate and 31.1% have not completed elementary education (Gutiérrez et al., 2010), and we found that having their perspective was especially important because almost everything that one might want to do or request in Spanish prisons requires filling out a form (Güerri & Larrauri, 2022). Thus, during our research on quality of life in Spanish prisons, we decided to implement the questionnaire as a structured interview so we would not miss this important segment of the prison population (Rodríguez et al., 2018). The implementation of this strategy was useful because, in addition to being able to get the opinions of all inmates regardless of their educational level, it enriched the responses with qualitative information that turned out highly relevant to interpret the quantitative results.

A further problem is how to capture those who are not fluent in the official language of the country where they are imprisoned. This may be solved by translating our questionnaire to different languages or including researchers who can speak the main languages spoken by the prison population in our team (e.g. Beyens & Boone, 2015; Brouwer, 2020, p. 710). However, these resources may not be available for every researcher, and it is impossible for our team to speak every language that may be needed. Thus, in many cases, it is likely that the language barrier cannot be overcome. This is a problem especially when it implies excluding an important part of the prison population that we are studying, as it may happen with indigenous or migrant communities. On this matter, Field et al. (2019, p. 145), regarding aboriginal people in Australian prisons, argue that ‘researchers have a responsibility to familiarise themselves with the demographics of the prison population they are sampling and ensure they approach research in a manner that is culturally competent and safe for vulnerable ethnic groups’. Along the same lines, in Spain, a considerable proportion of the prison population are foreign nationals from non-Spanish speaking countries (approx. 29.8% between 2013 and 2016),¹³ which reinforced our decision to implement our questionnaire as a structured interview (Rodríguez et al., 2018): interviewing inmates allowed us to make as many clarifications of concepts as necessary, and, in this way, we were able to interview inmates who could speak Spanish but were not completely fluent on the language.

Another issue related to language is the use of slang by the interviewees. Prison slang is usually unknown for the researcher, at least during the first visits. In our view, it is advisable to ask prisoners about the concepts we do not understand and to incorporate these words into our vocabulary. This will be useful to make sure that we understand what prisoners are telling us and demonstrate closeness to their reality.

Lastly, it is worth remembering the importance of observation. Regardless of whether we are conducting surveys or interviews, observing what each prison is like and what goes on inside gives us invaluable insights. As researchers, we are interested in seeing the dynamics between staff and prisoners as well as between prisoners themselves, and witnessing certain situations can help us understand better what prison life is like. For example, on one occasion, one of the authors was conducting research on open prisons, where prisoners are usually allowed to leave daily to work in the community. On a visit she made to an open prison in Spain, she met a prison officer who was attending a prisoner who had left the hospital with an intravenous line (IV) still in his arm for fear of not arriving back to the prison on time. According to the literature (Martí, 2018; Shammas, 2014), prisoners in open prisons often experience stress and anxiety because of the challenge of reconciling prison and work schedules. This scene illustrated this pain of semi-liberty very well and demonstrates how observation can complement techniques such as surveys and interviews.

For the purpose of analysing data, it is often useful for researchers to make a schematic map of the different areas of the prison and what each module is used for, since this information is not usually public. Since we will not always have the option of returning, it is especially advisable to note down as much information as possible on the visits we make. In addition, it is useful to keep a fieldwork diary, which can also be enriched by the experience of all the researchers in the case of a team. For example, in some of our investigations, we would meet after each day to exchange impressions on the interviews we had conducted and to share our reflections.

4 Ethical and Emotional Aspects

Throughout our prison research, a number of emotional and ethical dilemmas frequently arise that must be addressed. As Field et al. (2019, p. 139) state, the correctional environment represents an ‘ethical minefield’ for researchers. These challenges may appear at different points in the research, from the design of the project, during implementation, or in the data analysis phase.

Thus, when we design the research, one of the first ethical questions that emerges is whether we should compensate participants financially for the time dedicated to the study. This is an extensive debate in the social sciences (e.g. Field et al., 2019) and depending on the discipline and geographic context, there are often different positions in this regard. In the review carried out by Abbott et al. (2018), they found a minority of studies that reported monetary or other participant incentives and a comparable number of studies stating they were not given, while most studies did not mention this question at all.

On one hand, there is concern about how offering a monetary incentive can influence inmates’ consent, as those in need of money may participate even if they originally felt disinclined to. This creates a twofold ethical problem. Regarding our research results, it is considered that remunerating participation can have a selection effect (more participation of those with lower resources) and provide worse answers (since participants do not have an intrinsic motivation to collaborate, they may not answer the survey carefully). More importantly, in the specific case of prisons, remuneration has also been questioned due to its potential coercive influence on inmates given the importance of money inside prisons. Not having money in prison may imply asking for favours—that have to be returned. Thus, offering an economic incentive may have a coercive effect on the most vulnerable prisoners (Grant & Sugarman, 2004).

On the other hand, it has been argued that the fact that there is an economic motivation does not necessarily mean that there is no intrinsic motivation or that inmates feel coerced to collaborate with the study. Furthermore, if we don’t reward participation, it may seem that we are taking advantage of them. However, monetary rewards are not the only option. In this sense, some feminist researchers have emphasised ‘the importance of reciprocity, egalitarianism and sharing in research’, although it is not always clear how to put such ideals into practice (Bosworth et al., 2005, p. 255).

In sum, it is always important for researchers to consider the implications of each option and assess the circumstances, since other factors, such as budget, may interfere apart from ethical issues. In our case, we have usually tried to remunerate inmates for their participation because we believe in the importance of appreciating the time they devote to us. However, providing this remuneration has not always been possible due to the lack of budget or to bureaucratic hurdles. It is important to note that, in these occasions, we have not experienced the lack of remuneration as a problem for the research. Moreover, regardless of whether they are financially rewarded or not, it is important to make it clear to prisoners that their choice regarding participation will not benefit (or worsen) their current situation.

Other ethical and emotional conundrums arise during implementation due to the imbalances of power that characterise prison life. As we have already mentioned, research in prisons is strongly conditioned by the issue of security and by the power relations that exist, both among prisoners themselves and between inmates and the prison authorities. In this context, it is important that we conduct our research in a way that protects the privacy and integrity of the participants and creates a suitable environment for the interviews or surveys.

For example, we have already mentioned the need to speak with the inmates who controlled the wings in a prison of Latin America in order to be able to conduct our research: without such ‘permission’, prisoners who spoke with us could have been endangered. However, such threats also came from prison staff themselves. In the course of the mentioned research, our team learned that the inmates had been threatened by the prison management: they were not allowed to refuse to participate in the surveys but they were also forbidden to speak negatively of the staff or the prison. We had formed a team of about ten interviewers and spread out in a large room, each of us sitting on one side of a table, with the interviewee sitting on the opposite side. At first, we gave no indication as to which side the interviewer and the interviewee should sit on, but one day, we realised that there were several custodians standing at the back of the room. They were standing at a distance (as we had asked them to do so that they could not overhear conversations) but they were looking at us in a way that we found threatening, creating an atmosphere that was not appropriate for an interview. Knowing about the threats, we decided that the interviewers would sit facing the guards, so that the prisoners would not have to see them while being interviewed. In this way, we tried to ensure that this situation affected the people being interviewed as little as possible.¹⁴

The above example illustrates one of the ways in which prison staff can interfere with fieldwork and affect data collection process, privacy and confidentiality (see other examples in Abbott et al., 2018, p. 5). At the same time, this anecdote demonstrates that there are some measures that we can take in this type of situation within the limited manoeuvrability that researchers have in prisons: being careful with our notes, asking another person to move away from the table if we suspect that they are eavesdropping, or modifying the layout of the space, among others.

Another ethical dilemma that frequently arises has to do with the life stories and conditions in which many prisoners live. It is not uncommon for some prisoners to explain extremely difficult situations to us, and even ask for our help. For example, one of the authors was told by a foreign prisoner that he had hepatitis (which was also evident in the state of his eyes) and begged her to help him saying that he was going to be deported and that in his country of origin he would not be able to afford his medication and would die. On another occasion, after asking a prisoner if he had ever engaged in self-injury, he showed us his arm, which was full of scars caused by self-inflicted cuts. The prisoner told us that, during a previous sentence, he had received news that almost his whole family had been killed during a bombing in Syria. Additionally, in some countries the conditions are unquestionably appalling, and we have interviewed people with serious health problems because they have been locked up for years without being able to move or even see the sun. What do we as researchers do in such situations? What should we do? What can we do?

Our position is that when we do research we should stick to our role as researchers. We cannot and should not provide individual assistance, which we make clear before conducting the interviews, and reiterate if necessary during them. However, this does not mean that we cannot listen and show empathy, or refer them to the people who could help them (for example, in Spain there are several associations committed to assisting inmates). We can also become involved as academics by making an effort to ensure that our research has a positive impact on the conditions of prisoners and the functioning of the prison. As Beyens states, ‘being a critical criminologist goes beyond writing academic articles or reports (...) 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’ (2015, p. 74, based on Claus et al., 2013). In other words, those who wish to participate beyond strictly academic activities have several means to do so.

We also need to consider the emotional impact that testimonies such as the ones described above may have on the researcher. Witnessing the suffering of those incarcerated or feeling that we are in a dangerous environment can affect our judgement when conducting fieldwork and analysing the data and, therefore, reflexivity is needed. There are, indeed, some interesting essays on the emotional impact that such situations may have on researchers, not only negatively but also in a positive way (see, for example, Jewkes, 2012). In this sense, we find that acknowledging the relevance of emotions in prison research is of utter importance.

Similarly, our position of privilege as researchers and the characteristics that make up our identity—i.e. gender, race, nationality, age, class…—also should be taken into account while conducting fieldwork and trying to understand our data. For example, our first experience conducting fieldwork was soon after graduation. Being young women, prison staff often showed patronising behaviour. However, we were able to take advantage of the situation and get answers to many questions since we were not perceived as a threat and many prison officers were happy to ‘lecture’ us on how prisons really work. In addition, we believe that doing research in Latin American prisons while being European—which is evident due to our Spaniard accent—has made the people we interviewed trust us more quickly. Probably this is because being clearly foreign researchers (and also European foreigners, who are in a more privileged position), it is more plausible that the study is independent and not some kind of deception on the part of the Administration.

In sum, reflexivity is essential in prison research. However, we believe that we have to be wary that the acknowledgement of our emotions and positionality does not become the centre of the debate, since this would displace the focus from what really matters—prison staff and, especially, prisoners, that is, those who can really be affected by the research (aggressions, reprisals, threats...). In other words, putting the researcher at the centre of the debate when we talk about emotional impact in the context of prisons seems inappropriate to us because it disregards our position of privilege. This does not mean that we deny that this type of research can have an emotional impact on researchers. This impact is real, and moreover, it can vary among researchers whose positions of privilege may also differ according to personal experiences or socio-economic background. What we argue is that academically it should not be the main point, unless it is to reflect about how this impact may affect our objectivity when conducting fieldwork and analysing data.

5 Research in Prison and Research with Prisoners: Lessons Learned Through Challenging Interactions Within a Challenging Context

As this book attests, empirical research in criminology poses many challenges. Notwithstanding, besides the usual hurdles of criminological research, carrying out studies with imprisoned persons has its own set of difficulties due to the context in which research is developed—the prison—and the vulnerabilities of the studied population—the prisoners.

In the first place, studying the realities of imprisoned persons requires, on most occasions,¹⁵ carrying out research inside prisons, that is, inside institutions which are heavily bureaucratised, highly securitised, rigidly routinised and, generally, lacking in transparency. The implications of these prison characteristics on imprisoned people have been widely studied but, as we have shown in this article, they also condition researchers.

For example, due to bureaucratisation, we need to follow long, intricate and tiresome administrative procedures to get permission for our fieldwork. But the inflexibility of bureaucratic regulations can have other negative consequences, such as making it almost impossible to give financial retribution to inmates who had helped us with our research.¹⁶ Granting security is a constant concern of prison administrators and, thus, it will determine what we are allowed to do (e.g. to record interviews), or which areas of prison we are authorised to access (e.g. high-security wings). In this sense, a structured routine is essential to order, but researchers may disturb this routine or occupy human resources that are needed to implement it. Thus, our presence may be limited to certain periods in which the disturbance we cause will be minimal, limiting our chances of non-participant observation. Finally, the lack of transparency that characterises many prison systems hinders sample selection and makes it difficult to know how prisons work, which is especially troublesome if it is our first time doing fieldwork inside prisons.

The second source of difficulties related to researching imprisoned persons are the characteristics of the prisoners themselves, who usually come from disadvantaged backgrounds and, once in prison, are subjected to the power of prison authorities.

As we have explained, illiterate persons cannot fill self-administered surveys and there are language barriers due to the high proportion of aboriginal persons or foreign nationals in certain countries. Thus, researchers need to adapt their research techniques and strategies or, otherwise, some imprisoned persons will remain invisible. A more general question related to language issues is that researchers may also find that they need to learn prison jargon in order to communicate better with prisoners. Another sensitive aspect is that some vulnerable prisoners share life histories with a high emotional component. These situations raise ethical questions about how researchers should react and need to be acknowledged and reflected on since they may jeopardise the objectivity of the researcher when analysing the collected data. Finally, the imbalance of power between inmates and researchers and, most notably, between the prison administration and inmates, also raises ethical questions about the validity of the consent granted by inmates and the possible repercussions their participation may have for them.

Some of these problems and dilemmas have been widely discussed in previous literature. Notwithstanding, we would like to highlight that, in our view, essays reflecting on the challenges of prison research have not addressed certain methodological limitations that researching inside prisons entail. For example, in Spain, selecting a purely random and representative sample seems virtually impossible due to the difficulty of accessing the lists of inmates, the limited availability of those persons who are engaged in more activities, or the high mobility between modules and prisons. We were also surprised by the fact that self-administered surveys are widely used but the issue of illiterate inmates is never discussed. Similarly, only those investigations focused specifically on aboriginal or foreign national populations seem to acknowledge the problem presented by language barriers.

In this sense, we coincide with the concerns voiced by Abbott et al. (2018) about many prison studies not offering all the details about their methodology and the troubles they had to face. Entering the prison and conducting research with imprisoned persons is a challenging task regardless of the country in which the researcher is based. Thus, we believe that being open and honest about the difficulties we encountered, and sharing the lessons we learned, is indispensable to fully understand not only the endeavour that prison research entails but prison life itself.