The Triple Bind of Narration: Fritz Schütze’s Biographical Interview in Prison Research and Beyond

Introduction

The strengths and weaknesses of various methods employed in prison research have been widely discussed in the pages of sociology, criminology and methodology journals over recent years. Notable contributions by Bosworth et al. (2005), Crewe (2009, 2014), Jewkes (2012, 2014), Liebling (1999, 2014) and Schlosser (2008) among many others commented on both the ethics and the methodology of prison research, shedding valuable light upon problems of access and identifying the physical, procedural and infrastructural constraints. Drawing on these earlier debates we offer another perspective on negotiating what Schlosser (2008) referred to as the ‘methodological landmines’ of prison research by analysing the method of biographical interview in this context. We focus specifically on the biographical interview method developed by the German sociologist Fritz Schütze in the 1980s, and by evaluating this method in detail we hope to advocate its wider application not only in prison sociology but also in social science research in general. Schütze’s (1983, 1984) biographical narrative interview ‘merges the objective features of subject’s life with the subjective meanings attached to life experiences’ (Denzin, 1989: 54–55), thus making concrete the ways in which individuals understand their own lives. The fundamental feature of the narrative interview is that the material obtained is not a sum of answers to questions, but a spontaneous and uninterrupted account in which the participant attempts to relate reflexively to their own life experience. Drawing on our experience of empirical study of repeat offenders in the Polish prison system, we portray the characteristics of this way of gathering data, and the conditions that require to be met for its proper use.

We use our analysis to support two key proposals. First, we argue that in social science research Schütze’s biographical interview method offers unique insights into the process of identity formation. Second, in relation to collecting data in the prison setting we contend that this method has a rehabilitation value in leading the interviewee/narrator to engage in reflexivity and re-evaluate their life through biography work. We also offer some practical advice on conducting biographical narrative interviews in the prison setting, given the difficult nature of the prison as locus of research.

Schütze’s Biographical Interview Method

The biographical method developed by Schütze counts as one of two biographically oriented streams in German sociology (Denzin, 1989) along with Oevermann’s objective hermeneutics. Schütze (1983, 1984) was at pains to offer a methodologically coherent conception of how to use the narrative interview, how to analyse the data collected, and even how to write up the final research report. His proposal appealed to the achievements of interpretative sociology – a tradition which includes ethnomethodology and conversational analysis, as well as phenomenology and Goffman’s (1961b) work on the sociology of interaction. In line with the tradition of interpretative sociology Schütze’s method has its roots in the tradition of the Chicago School, a movement marked by a commitment to bringing to bear the resources of ethnography and sociology upon the analysis of social problems experienced through the eyes of the individual (Faris, 1967).

The biographical interview implies the definition of the key participant as narrator rather than as interviewee. Each of the participants, researcher and narrator, plays a specific role – the narrator is at the centre of attention, and the researcher is primarily a listener, whose role in guiding the direction and nature of the issues picked up is a limited one. It is the person telling the story who constructs their own identity (Evans and Wallace, 2008; Farrant, 2014; Gelsthorpe, 2007; Presser, 2009), and it is story-telling above all that allows narrators to engage in biography work. Life stories constitute the subject matter of the biographical interview; most of the time these are cast in the context of specific problems or sets of relationships, like people’s job histories or war experiences (Schütze, 1992). The result is that the researcher gets a ‘thick’ narrative account, sometimes hours long, which they record and transcribe. Biographical interview creates specific conditions favouring the growth of a sense of security in the narrators; and in turn it makes it possible to gather information which the researcher would find it hard to elicit by direct questioning. A Schützean approach was successfully employed in research on migrant workers (Treichel and Schwelling, 2006), patients (Kolip et al., 2009) and religious converts (Jindra, 2011) among others.

Schütze was interested both in what people had to tell, and also in how they told their story; and how narrators arrive at their own deep understanding of what they experience (Riemann, 2006). Schütze (1983) argued that when telling a story the narrator is constrained by the ‘triple bind of narration’ (dreifacher Zugzwang des Erzählens): the requirement to close (Zugzwang zur Gestaltschließung), the requirement to condense (Zugzwang zur Kondensierung) and the requirement to provide detail (Zugzwang zur Detaillierung). These three requirements mean that an extempore biographical account can be expected to consist of a coherent and condensed story line with well-defined turning points, and with all characters introduced into the story playing defined roles in it. Although the story has to be condensed, it still requires sufficient detail to be credible – hence the requirement to provide detail. Schütze argued that where narrators agree to give an account and embark upon it, their account of their reasons and actions is inherently trustworthy, and the triple bind of narration makes the story analytically valuable for the researcher (Schütze, 1982).

It needs to be emphasised that not every fragment of the interview is a narrative (Schütze, 1983). Narratives are only those passages which have a clearly marked beginning, maintained as such in the chronological ordering of the subject matter. The beginning is followed by a description of the development of events with a connecting thread, and the account comes to an end, which is clearly marked in terms of time. Schütze’s narrative interview comprises five stages: (1) starting the interview; (2) prompting the story through a generative question; (3) the telling of the story without any interruptions until a ‘coda’, an ending-formula, is used by the narrator; (4) asking questions; and (5) bringing the interview to an end. The first stage of the interview is important in creating appropriate conditions for the exercise, in spite of the fact that it does not usually get documented and generates no empirical material of its own. It is necessary to lay down clearly the ground rules – to define the roles of the participants and create a climate conducive to the narrator ‘opening up’. The researcher takes on the role of a person interested in what the narrator has to say, regardless of the account’s structure, or the issues discussed.

The methodological literature is clear on the point that many problems with getting good quality narrative are the result of a lack of active listening skills on the part of the interviewers (Farrant, 2014; Kaźmierska, 2004a). Comment is to be avoided for the actual duration of the storyteller’s account, so are questions of any kind. Only after the spontaneous biographical account draws to an end, in the fourth stage, does the researcher ask two types of questions – for clarification of narrative detail, and ‘theoretical’ questions which relate more closely to the aims of the study. No less important is the manner of bringing the interview to a conclusion. This phase is primarily a matter of addressing an ethical problem when the researcher normalises the interview situation. Winding up the meeting with the person who has obliged the researcher with their time and their trust, it is incumbent in turn to take the time to treat the narrator as an interactive partner, and not just a ‘story-telling machine’ (Kaźmierska, 2004b: 157).

Schütze’s method has been subject to further modifications by researchers coming from the psychoanalytic perspective. The most notable of these modifications is the free association narrative interview (FANI) advocated by Hollway and Jefferson (2008, 2013) and Wengraf (2001). Free association narrative interview draws on Schütze’s tradition by its emphasis on eliciting stories, the use of open-ended questions, avoiding ‘why’ questions and maintaining the discipline of staying as ‘invisible’ as possible, which does not imply a belief in an ‘objective’ interviewer (Hollway and Jefferson, 2013). In contrast to Schütze’s original method, however, FANI focuses on an unconscious, rather than a conscious logic of story. Discarding the narrative convention FANI allows the researcher to analyse incoherence and incongruity rather than the logical story line. Schütze’s method also has some links with psychosocial research (Clarke, 2002; Clarke and Hoggett, 2009) which rejects ‘top–down’ approaches to interviewing which fail to depict the complexity of human lives, and are unable to escape the inequality in power relations between the interviewer and the interviewee (the latter argument raised by scholars such as Crewe, 2014; Stanley, 1992). We have chosen however to use the Schützean method in which the analysis of extempore narrative gives access, not to the events of the past, nor to the unconscious motivations of the present, but to the socially available elements of identity which people naturally deploy when they say who they are, where they are from and why they came to be who they are.

Prison as a Locus of Social Science Research

Before we discuss our findings we turn briefly to the subject of prison as a setting for social science research. Prisons generate the curiosity of academics, students and popular media, but research inside prison is attended with numerous practical difficulties which include negotiating access, seeking ethical approval and navigating the physical space of the prison environment (Coyle, 2002; Liebling, 2014; Piché et al., 2014; Reiter, 2014; Schlosser, 2008; Sloan and Drake, 2013). In many countries prisons are almost off-limits to social science, constituting an area shrouded in secrecy and mystery (Jewkes, 2013; Kamiński, 2004; Newbold et al., 2014). Prison is a total institution, independent of the political-social system within which it functions (Goffman, 1961a). The peculiar quality of prison and its absolute insulation from the rest of society is apparent in practice, in the way that a person from outside the ‘firm’ (such as the researcher) upsets the rhythm of prison life (Bosworth et al., 2005; Crewe, 2009; Ferrell and Hamm, 1998; Schlosser, 2008). Prison resists scrutiny because it is structurally and bureaucratically closed off from the rest of society (Reiter, 2014), but also because the researcher is often emotionally disconnected from the subject of their study (Crewe, 2014; Jewkes, 2014; Ugelvik, 2014). Hence, Reiter (2014: 417) refers to prison research as a ‘social black site’ – invisible both to the public and the researcher.

Another set of factors which make the conduct of research in prison difficult are those connected with the people the research is about – whether these are Prison Service staff or prisoners (Bosworth et al., 2005; Liebling, 1999). The common denominator for both groups is mistrust and reluctance to talk about their experience of prison life (Copes et al., 2012; Liebling, 1999). One major factor concerns the anxiety prisoners feel about the ‘real’ reason for gathering information and the purposes for which it may be used (Copes et al., 2012; Jewkes, 2012; Ugelvik, 2014). The last point relates to the ubiquity of power relationships in prison, extending naturally to the domain of research. Bosworth et al. (2005), Crawley (2011), Crewe (2014), Jewkes (2013, 2014) and Liebling (1999, 2014) criticised the privileging of methodological orientation at the expense of the emotional side of research, and the tendency to downplay the existence of issues of power, authority, vulnerability and despair which come to the surface in prison. Both Jewkes (2012, 2014) and Liebling (1999, 2014) called for the acknowledgement of the feelings of frustration, fear and outrage, which affect the research process – both on the part of the researcher, for example by being ‘burdened’ by the knowledge of unresolved crimes, and on that of the prisoner, by being ‘given a voice’ (Liebling, 1999; Piché et al., 2014; Stevens, 2012).

Methodological issues of prison research also feature strongly in researchers’ accounts: these include such matters as how concepts are to be operationalised, how research instruments can be designed to avoid implications of guilt (Schlosser, 2008) and how cognisance is to be taken of race and gender issues (Burman et al., 2001; Crewe, 2014; Earle, 2014; Jewkes, 2012). Nor can the use of narrative as a discursive construction of the world in prison be said to be uncontroversial: the main difficulty being the doubt to what extent individuals’ stories can be seen as separate from the culturally dominant story lines (Bowling, 2013; Farrant, 2014; Presser, 2009). Schlosser (2008: 1513) argued that asking an inmate to ‘tell your life story’ leaves researchers at an analytical disadvantage – a point which we had to bear in mind when designing our study. One important advantage of Schütze’s method, which can address some of the methodological problems associated with research inside prison, lies in the redefinition of the research situation, that is, the creation of an atmosphere of ‘ordinary conversation’. For example the implication of guilt discussed by Schlosser (2008) and the unequal power relationship explored by Jewkes (2012) and Ugelvik (2014) are reduced by the listener/researcher taking a step back, asking very few questions and allowing the narrator to be the ‘host’ of the research situation. Similarly the mistrust and reluctance to talk, which Copes et al. (2012) and Liebling (1999) commented on, can be overcome by making them feel safe, giving the narrator a sense of agency in telling their story and allowing them to ‘lead’ the listener. These are characteristics which give Schütze’s method advantage over other methods such as observation (which gives the researcher insights into the here and now of prison life but says little about their criminal past), or traditional interview which creates an atmosphere of interrogation. The next section of our article will draw on the findings from our study to demonstrate some of the advantages of Schütze’s method in practice.

Methodology

We draw on an empirical study designed and conducted by the first author in four medium-security prisons in Poland over a period of 12 months in 2011–2012. The aim of the study was to understand the processual nature of repeat offending, and gain insights into the formation of habitual offender identity. We aimed to reconstruct biographical experiences, and compare the ‘strategies for action’ used by offenders throughout their lives. The themes investigated included offenders’ subculture, labelling and the mechanisms of adaptation to prison life as they constitute an integral part of repeat offenders’ identity (Kamiński, 2004) as well as attempts to break away from the cycle of re-offending. The participants, all men, were serving prison terms for a range of offences from burglary and motor vehicle theft to armed robbery and grievous bodily harm. Almost all had problems of drug and alcohol dependence. The shortest total time incarcerated for any participant was five years, the longest 31. All interviewees were habitual offenders who had served more than three custodial sentences. Access to the prisons was secured through senior management of the Regional Prison Service Inspectorate in Poland. Special exemption had to be secured to allow the use of electronic audio recording equipment inside prison, which is normally prohibited. All ethics procedures set by the principal investigator’s home institution were followed, and the study had the approval of the Prison Service authorities. The prisoners were informed by the education tutor of the possibility of taking part in conversation-interviews conducted by an academic researcher. No incentives were offered for taking part in this research. A total of 38 prisoners took part in individual interviews, which lasted from two to four hours each. Only 23 transcriptions generated material in the form of a biographical narrative, amounting in total to more than 460 pages of text. The material obtained from the other 15 interviews, which did not contain biographical narratives, could only be analysed following the guidelines of grounded theory.

In the course of research we made use of practical guidelines offered by Schütze (1984) on the processual structures and homology between the order of experience and the order of narration. Analysis of empirical material was undertaken on the basis of the standard procedures of the methodology of grounded theory (Glaser, 1978; Glaser and Strauss, 1967). This included the open and selective coding of the collected information, the preparation and sorting of theoretical notes and the generation and integration of theoretical explanations with close reference both to the empirical data and to the theoretical insight gained (Glaser’s (1978) method of theoretical sensitivity). Examples of codes include: breaking away from the habitual offender stereotype (establishing a relationship, dropping old associates); confronting the stereotype (helping other offenders straighten out, leading a self-help group for substance abusers); subterfuge (more effective ways of evading justice, more efficient ways of being a criminal); the role of women in preventing the return to crime; the released prisoner’s search for a non-criminal environment; the search for an alternative (to the life of crime).

Making a connection between these two methodologies was possible because of the coherence at the level of the assumption that social reality has a processual character. The analytic procedures of grounded theory allowed us to go beyond the individual life stories of repeat offenders by uncovering common patterns and mechanisms which shape their social relations as well as their logically guided actions. By means of conceptualisation, our method made possible the reconstruction of the dependency between the shape of a biography and the social conditions that determine the wider context of the person’s activities and experiences.

Findings

To illustrate how the ‘triple bind of narration’ allowed us to reconstruct the dynamics of individual events in the broad context of the life experiences we focus on one theme – emotional relationship events such as marriage, the birth of a child and the growth of strong emotional ties with a girlfriend’s children. In investigating these relationships we were interested in the events leading to cessation of criminal activity. Almost all of our interviewees had experience of strong relationships or marriage during periods in between spells in prison, and assigned a ‘sanctuary’ characteristic to these events; they saw them as a protection against returning to prison, and also against returning to crime. We argue that it was possible to capture the dynamics of these experiences by reliance upon three key features of the narrative interview. The first of these is the requirement to condense, which means that since narrators cannot tell the whole story of their life, they are driven to talk about the most important events that relate to the ‘turning points’ of their life story such as relationships with women. In the times when our narrators were setting up a home or were in a strong relationship, they were ‘surviving’ in freedom for relatively long periods of time. When incarcerated, the narrators attached particular value to ‘better days’ and periods of freedom, and it was precisely the stability of home and family, and support lent them by women, which offered them a real chance for their next attempt to avoid repeat offending and stay out of prison. One narrator reflected on his marriage:

Later I got a bit of peace, when I got married. I got married and settled down. I absolutely cut myself off from bad company. Family, wife, I had a baby … I told myself, you need to sit on your bum, pardon the expression. So I had a proper job, public sector, but you know you’re never going to get rich on that sort of money. […] How stupid I was to think like that! Why did I go back to nicking stuff? […] I used to have such a good life. On Sundays we all go to the allotment – me and the wife and the kid.

The possibility of tracing the dynamics of offenders’ relationships was reinforced by the drive to complete the account – the requirement to close. If the narrator brought a person into the story, they felt the need to show the role this person had played in their life and to show the person’s significance in the context of changes in their identity. An example would be the situation where the narrator remembered a woman who was a witness at his trial and who turned up next in the context of a story about fondness for animals. Finally the narrator felt himself ‘obliged’ to clarify who this woman was in his life and accord her a role of special significance in his autobiography:

I love dogs, when I am in prison I miss dogs. I had a pit bull recently. Beata – the woman I was telling you about earlier – is looking after him [a sigh]. I lived with this girl … I can’t say a word against her, because she kept the house clean, cooked well – even though she came from a pathological family.

An extended narrative devoted to a woman, the immediate trigger for which was a story about something else entirely (fondness for animals), finally included a reconstruction of three years of this offender’s life. He spelt out how, bolstered by a close relationship with her, he had stayed away from criminality; how he had used avoidance strategies to stay clear of the haunts of his criminal cronies, how he had dealt with the temptation to steal and finally what sort of conditions had seen the ties that bound him to her weaken, and his return to a life of crime. In contrast to the traditional interview, Schütze’s biographical interview gave the narrator an opportunity to tell a story and reflect on the events from the past.

The third requirement, the requirement to provide detail, also operated to help reconstruct the relationship dynamics that had existed between habitual offenders and the women in their lives. Narrators illustrated sequences of events that had accompanied the development, course and breakdown of their marriages with detailed descriptions of their lived experience. At times past events were achieving a present-time dimension as if they were going on here and now. Portrayals of dialogue rang true as recalled against the background of painful events, bringing to life in detail the emotions that accompanied them. The following quote illustrates this point:

You’ve been off with your pals again! – she says. I say to her, don’t hold it against me – I’m not the sort of guy who can sit at home all day. But she says to me, get a grip of yourself – you’ve got a family, you have a child who’s looking at you. Is that right? I say, the kid’s looking at you giving me dog’s abuse. I see her picking up the phone and I remember then thinking to myself – this marriage is over.

This spontaneous narrative, made possible by the researcher’s passive role, is a boiling pot of emotions. The details that accompanied his recollection, and the strong emotions it evoked, clearly meant that these were unusually important experiences in his life and that he recognised them as having crucial significance in his life story.

The dynamics of relationships (development, course and breakdown) were recoverable thanks to narrators’ disclosures of the motives behind their actions, and their perceived dilemmas and doubts. The next passage illustrates these points; the narrator had become a father at the age of 16:

She got pregnant by me, and our daughter was born. I remember I felt really good about that and stuff … how nicking stuff was very much in the background, at the beginning. And then I was trying already for my girlfriend’s sake. Taking into account that I’m not seen as the kind of guy that’s going to let himself be shut up in the house, at that age, OK? […] Well, my daughter was 18 months, the general falling-out began, various battles and … emotional. We were just 17 years old. I found, what with the drugs, with the larceny, with the whole thing, that for me I was at home just four hours … OK? But I was only there to sleep, that’s the truth, to get a feed, yes to get a good feed, not to give her any money and in general, to be honest about it. I was simply taking advantage … because it was … whether even later on I … by the time I was in jail, I joined the dots and thought to myself what these things meant … what all I had done, OK? I didn’t appreciate Anka at all, did I?

There prevails almost universal agreement in criminology literature that relationships with women are among the most important predictors of reduction in males’ criminal activity (Beijers et al., 2012; Christian and Kennedy, 2011; Cid and Marti, 2012; McGloin et al., 2011; Skardhamar et al., 2014). Our research went beyond this to reconstruct the dynamics of this phenomenon: to capture the relationship in its beginnings (meeting the woman, deciding to form a relationship) through its course (daily life, sources of joy and satisfaction, building a life together) to its breakdown (predisposing causes of increasing turbulence in the relationship and attempts to overcome the problems involved) right through to the consequences of marriage break-up (divorce, return to a life of crime). Our analysis allowed us to specify what part involvement in crime played as part of the offenders’ life cycle, and what intensity it showed, depending on whether they formed close relationships and established a family or not. Data which we collected from narratives, unhampered by questions, devoted to the narrator’s whole life, allowed us to reconstruct the circumstances in which successful relationships were formed, the difficulties associated with breaking with the criminal environment, the strategies developed to maintain family stability and finally the factors that led the relationships to break down. Such data would be difficult to obtain using other methods, and they enabled us to recreate the process of withdrawal from criminal activity and the ‘remissions’ afforded by close relationships with women and family, and then to show the course of events, often violent in its strength and rapidity, by which criminal habits broke out of their sleep and back to life. This was often concurrent with the loosening of emotional and social bonds with the family. Our research poignantly illustrated our narrators’ awareness of this in the language they used as they recalled activities shared with women and aspirations to build up close relations with them: ‘she was my last chance’; ‘the women in my life are my one and only chance’; ‘thanks to her I had some sort of chance in life’. Our research has thrown some interesting light upon these relationships, making it possible, for example, to identify the complex determinants of success in their development as well as the reasons why they fail. The narrators demonstrated a high level of reflective consciousness developed by the isolation in prison. One participant noted: ‘in prison you mainly think about how things used to be, what happened, what harm you did to the people nearest to you and how you might have lived your life differently’. The analysis allowed us to gain insights into offenders’ relationships with women, and how these relationships affected their criminal activity; however, the analysis was only possible because of the nature of Schutze’s method of data collection. Using Schütze’s biographical interview in our study we gained new insights into the formation of repeat offender identity, as it allowed the narrator to locate particular experiences in the context of the entirety of their life experience.

Schütze’s biographical interview had other advantages as well. One advantage of this ‘ordinary conversation’ was that in the construction of their own story, the narrators felt safe – something which could be read off their changed posture and tone of voice. The interviewees in telling their stories provided very detailed descriptions of events and situations. As the interview progressed the wish to supply narrative detail often led them to reveal information about the prison subculture, or other ‘forbidden topics’ related to the life of crime. Some hesitation to speak openly was evident when in the course of the narrative interview it was palpably clear that the narrative stopped; a long silence, a sudden change of subject or even an explicit refusal to continue or to enlarge upon the events discussed, all for reasons unspecified. In most cases, however, narrators opened up, and their desire to achieve closure in their story overcame their discretion, delicacy and anxiety. At times we felt ethically burdened by information we received, but disclosures were sometimes required for the coherence of the narratives and were more often than not forthcoming, affording insight into the cause-and-effect relationships which caused criminals to re-offend. The need to forge a coherent story, and the need to ‘explain oneself’ (Presser, 2009) in such cases trumped the need for secrecy.

The above observations centred upon the method’s potential to elicit unique and valuable data, but the use of this method also had implications for the rehabilitation of the prisoners. Making prisoners engage in a free-flow narrative about their lives in crime allowed them to reduce feelings of discursive and ideological confinement (McKendy, 2006), by beginning to re-tell their lives. Our narrators were happy to step into the new role of agents who constructed their own story, which in many cases was different from the one told in court, and silenced in prison (O’Connor, 2003). When telling their own stories they could temporarily break away from the label of repeat offender. The following quotation suggests that the status of a deviant has a symbolic value – in the perception of many people it signifies a range of negative characteristics and a lack of any positive ones:

People only see prisoners in a bad light … in the worst light possible. They’ve never done time, they don’t know what it’s like inside and what crooks are like. People just watch American movies and think to themselves – violence … cruelty … and they think everybody in prison is like that, right? But there are people here with feelings, with families, with the problems of ordinary life.

The interviews induced the narrators to reflect more deeply and to put in question the meanings both of particular events and of pervasive life situations. Here is a narrator reflecting on what he sees as his own victimisation:

The way I see it now, you can see for yourself, I’m like a sort of waste bin. I never stood up for myself and I let folks come after me and take advantage of the fact that I’m not making trouble. And everybody thought they could throw whatever they like into this bin.

In engaging in biography work narrators gained an understanding of their own actions and often for the first time discussed difficult moments in their lives: they attempted to justify their crimes, and commented on their decisions to act in certain ways. In this way the process of constructing their biography can be argued to have had potentially more than transient therapeutic value. Any such value could be put down to the operation of the drive for closure which puts a premium on achieved reasoning and causal accounts. What we also noticed was that reflection on the repetitive nature of their own offending appeared to lead the interviewees to break away from the processes of rationalisation and neutralisation of their crimes (Sykes and Matza, 1957). Being the focus of the researcher’s attention appeared to make them feel needed and valued as bearers of knowledge which the world outside was trying to gain. Many of the interviewees commented on the interview being their first opportunity to talk about their lives to someone else in prison. Maybe this attention and acceptance is itself a motivation to look at oneself through another’s eyes, like the man who expressed the following insight:

My wife was on at me the whole time about the money I gave her. ‘It’s dirty, same as you are!’ she would say. I used to tell her, I bring money home, and I give it to her for housekeeping, and she’s mean not to appreciate that … Now I see that’s not the way it should be. Now I see I was just kidding myself on.

An element which facilitated this openness was that we deliberately avoided consulting our interviewees’ files in advance of the interviews. We did not have preconceived ideas about their criminal history or information about the crimes committed, hence we avoided any implication of guilt – one of the difficulties in prison research discussed by Schlosser (2008). Admittedly, when faced with a ‘stranger’ some prisoners took the role of the ‘expert’ who cannot share their knowledge, because it is ‘secret’, but the great majority of them entered the role of a reflective interviewee interpreting what needed to be explained, then resuming the thread of their biography. This readiness to interpret resonated with Schütze’s requirement to condense and requirement to provide detail.

Although Schütze’s biographical interview is not the only method which promotes the participant’s reflexivity, the unique reciprocity and reflexivity between the narrator and the researcher is one of the method’s greatest strengths. Schütze’s perspective on biography as a sequential development takes account of the emotional side of this relationship (Gray, 2008). Giddens (1991) remarked that reflexive processes inherently involved change since they put the world in question as well as ourselves. As far as the narrators in our study are concerned, it was easy to see that the act of storytelling with its imperatives of closure, relevance and detail were able to play an important role in making offenders understand the rationale of their own decisions. It is just as true, though maybe less immediately obvious, that the role of the investigator in the Schützean scheme deeply meshed in a certain reflexivity which is imposed by the need at all times to avoid as far as possible imposing interpretations suggested by one’s own social, cultural and emotional particularity. This situated reflexivity called for ‘conscious and consistent efforts to view the subject matter from different angles’ (Alvesson, 2011: 106), working with multiple interpretations (Adams, 2006; Crewe, 2014; Riach, 2009), being suspicious of their own presuppositions (Clarke and Hoggett, 2009) or ‘altering one’s life as a response to knowledge about one’s circumstances’ (Holmes, 2010: 139). Choosing Schütze’s method has implications for the conditions under which research is undertaken and requires at every turn modifications of the research approach: this goes to presumptions, to target outcomes, to theoretical context, to methodology but also to the emotional sides of research (Hollway and Jefferson, 2013). Prison conditions add their own requirements to the mix, more than just appropriate preparation of the researcher’s theoretical and methodological organising base (Sloan and Drake, 2013). The researcher has to take a position which is on the one hand friendly and flexible, and on the other hand maintains distance: the role has definite boundaries. Meanwhile offenders are apt to deploy their own techniques to influence the situation and the researcher, not the least of which is the telling of ‘larger than life’ stories. Showing respect which is not perceived as weakness, and interest which is not perceived as credulity is one of the important skills of the interviewer’s role.

We have to acknowledge certain limitations of our study, and of the method in the prison context. One problem related to the fact that some prisoners had frequent and ready recourse to set phrases and formulae, linguistic camouflage which they saw as safe. This applied particularly to those prisoners with a long history of interaction with correctional institutions. Here is one example (emphases on ‘institutional’ terminology added):

I come from a family *** well … a pathological family, because this family of mine … err … it matters a bit later, but as I remember it, well err … they liked a good bucket. To be precise, mum liked a good bucket. Dad didn’t drink, but the problems there, it was like there was absolutely no emotional connection, right? Mum had *** some other guy. They broke up, moved out, and … don’t know how to say this, well …. That’s what you call loading stuff on the family, yes?

In this quote the language used by the narrator reflects the language used by prison staff, that is, discusses the events from his ‘deviant’ life drawing on the rhetoric of correctional institutions.

The last phase of the biographical interview had its own difficulties. Just as the process had begun by opening a channel of communication through the hostile conditions of the total institution, the end of the process also signalised the closing of this channel and the re-erection of physical and psychological barriers. What had been a momentary freedom of expression was now over, for both the narrator and the researcher.

Conclusions

Two key points emerge from our discussion – one related to the use of Schütze’s method in social science research in general, and one in relation to prison research. First, we see Schütze’s method as having an important place in social science research as it affords access to the ways in which people understand their life experience, and this in turn gives us insight into identity formation. The strength of the method lies in its reflexivity: it is reflexive in that it rests upon a relationship between narrator and researcher, a relationship between the narrator and the generalised other (Schütze, 1983), and in the sense that the narrator is inherently led to reflect upon their own actions and experiences. It is not a means of access to a further reality, whether this is a reality of past events or a reality of unconscious psychic defences. Instead it relies on the interplay between the narrative pressure and reflexivity. The ‘triple bind of narration’ inherent in narrative – the requirement to close, to condense and to provide detail – makes the account of the narrator valuable for analytical purposes. The point of reconstructing a course of events is to reproduce remembered event sequences that reveal the special continuity of a life as experienced by the narrator. The advantage of Schütze’s interview is thus not that it enables us to gain information about something that has happened, but rather that thanks to the powerful entanglements of narrative, the narrator makes evident the cognitive process that structured their experience at that point (Küsters, 2009).

Second, although special constraints apply to prison research, we found that the methodology of biographical interview proposed by Schütze goes some way towards mitigating their effects and has advantages over other methods employed in prison research. Used to questioning on the subject of their experiences in crime, during traditional research interviews, prisoners often present their life history in formulaic and stereotypical terms geared to the preoccupations of the criminal justice system. Embellished with snatches from the Penal Code, this ‘official’ biography is something rehearsed for the benefit of Parole Boards and teems with descriptions of situations in terms which neutralise their criminal activity (Sykes and Matza, 1957), together with numerous notes of regret and sorrow. Our experience with the Schützean method was different: interview participants constructed their stories using a wider range of categories and identified causes and effects of their actions which crystallised a version of their own perceptions of their careers, a version, which we maintain, allowed us to glean insight into the patterns and regularities of repeat offending, and suggest ways in which the vicious circle of re-offending can be stopped.

It is not as if we can afford to neglect any method which makes a credible promise of understanding prisoners and their world, a world made harder to understand by the ambivalence with which we view it (Zara and Farrington, 2014). This is a world which at the same time repels and attracts, horrifies and fascinates (Hare, 2011). We acknowledge that in person-to-person contact with the prisoner we experienced very ambivalent emotions related to thinking that our interlocutor is an ‘ordinary’ human being, with no special features that set them apart. Throughout the process of conducting research, however, we tried to avoid falling into the trap of justification of our participants, and instead focus our attention on our understanding of the patterns of re-offending.

It would be naive to claim that the stories told by prisoners are entirely unedited and authentic narratives, and especially the ones elicited in prison are located in a broader socially situated discourse (Presser, 2009), but undoubtedly they offer new opportunities to understand the human being behind the offender label and give us insights into what makes people re-offend. We found that Schütze’s biographical interview may have a rehabilitation value for the prisoners themselves, and although the process of rehabilitation through narrative merits a different article, it is worth remembering that as researchers we are able to make an impact not only through the publication of our findings but also through the process of conducting research.