Co-authoring desistance narratives: Analysing interactions in groupwork for addressing sexual offending

Abstract

Research and theory suggest that desistance narratives and pro-social identities are key to the process of desistance from crime. However, little research has examined how desistance narratives and related identities are produced in contexts other than research interviews or how core correctional skills intersect with the development of these narratives or identities. This study applies discourse analysis and conversation analysis to transcripts of 12 video-recordings of groupwork sessions for addressing sexual offending, examining how desistance narratives and identities are produced, and how practitioner skills and conversational styles intersect with their production. The analysis illustrates how criminal justice practitioners help to co-author desistance narratives through subtle and explicit aspects of interaction, although certain orientations to risk may limit this potential.

Keywords

conversation analysis desistance narrative probation

Introduction

To desist from offending, research suggests individuals need to construct a non-offending identity, one incompatible with offending behaviour and consistent with future prosocial aspirations, which coherently accounts for past offending (Giordano et al., 2002; Maruna, 2001; Rocque et al., 2016). Identity change, or ‘secondary desistance’, supports maintained abstinence from offending, rather than a temporary behavioural lapse (Giordano et al., 2002; Maruna and Farrall, 2004). Using life story interviews, several studies revealed characteristics of narrative identities of people desisting from offending (e.g. Gadd and Farrall, 2004; Giordano et al., 2002; King, 2013; Maruna, 2001).

But where do desistance narratives and non-offending identities come from? Given narratives and identities are social in nature, they are likely formed, at least partly, through social interaction. Desistance is influenced by how people’s identity change is recognised and reflected back to them by others, including people close to them (micro-level), their community (meso-level) and societal institutions (macro-level) (Nugent and Schinkel, 2016). However, it is not evident how interactions at a conversational level shape these narratives. Criminal justice interventions, that is, probation supervision and structured programmes, provide interactional spaces for people to ‘re-story’ their identities and life stories. As such, they may contribute to the formation, shaping or reinforcement of desistance narratives and non-offending identities. This article explores how interactions during sessions of a groupwork programme for addressing sexual offending contribute to constructing the identity characteristics evident in desistance narratives, discussing the implications for practice, theory and research.

Desistance narratives

Narrative identity refers to how identity is formed through people’s stories about their lives, where past experiences are coherently intertwined with current circumstances and future goals (McAdams and McLean, 2013; Maruna, 2015; Vaughan, 2007). These narratives are shaped, edited and refined through interactions with others. This active and interactive process is self-constituted and impacted by wider social and discursive influences (McAdams and McLean, 2013).

Narrative approaches have greatly influenced the treatment and conceptualisation of identity in criminological literature about desistance (e.g. Gadd and Farrall, 2004; Harris, 2014; Maruna, 2001; Vaughan, 2007). Using life story interviews, Maruna’s (2001) seminal work highlighted that people desisting from general offending developed a ‘redemption script’, where they learned from their past mistakes and were agentic in moving away from offending, often with a generative purpose, for example, giving back to the community. Those persistent in offending presented themselves as victims of circumstance, destined to live a troubled life. Central to the redemption script is the presentation of a true ‘core self’ characterised by an enduring positive identity, separate from offending behaviour. Maruna (2004) noted the explanatory style of those desisting contributed to this characterisation. Desisters attributed positive life events to stable, internal and global factors (e.g. ‘I got the job because I’m a hard worker’) and negative life events to unstable, external and specific factors (e.g. ‘I failed that exam because I was sick’), and vice versa for persisters. As such, desisters separated their true ‘core self’ from their previous offending behaviour, attributing offending behaviour to situational rather than dispositional factors.

Subsequent research notes narrative as central for building a non-offending identity. Gadd and Farrall (2004) found their male respondents narrated their offending and desistance in line with gendered norms by drawing on social discourses of masculinity (e.g. rebellious, troubled youth). This distanced their present self from their past, while maintaining coherence in their self-narrative. Presser (2004) showed men in her study used discourses of masculinity and morality to situate their behaviour, presenting themselves as moral. Vaughan (2007) argued narratives must morally reassess past behaviours to support desistance, not simply reconstruct past events. King (2013) proposed that a sense of moral agency in their early desistance narratives allowed her participants to identify future, non-offending identities. Changes in self-identity or how people view themselves is considered necessary for desistance (Rocque et al., 2016).

Most research concerns desistance from general offending; less is known about the narratives of those desisting from sexual offending. Farmer et al. (2016) analysed narratives of men convicted of child sexual abuse, demonstrating the separation of behaviours from the ‘core self’ also distinguished between those desisting or persisting in sexual offending. The ‘desisters’ presented situational rather than internal factors as primary causes for their offending and rejected the label ‘sex offender’ (Farmer et al., 2016; McAlinden et al., 2016). Farmer et al. (2016) highlighted that this explanation of offending enables individuals to manage their shame and develop a non-offending identity.

Farmer et al.’s (2016) participants were performing self-presentation, not merely rejecting the label ‘sexual offender’ but guiding others to the same assessment by constructing their behaviour as situational. People manage how they present themselves, to influence others’ impressions of them and pursue particular aims (Goffman, 1959). Situational explanations of behaviour perform moral self-presentation (e.g. justifying lateness because of heavy traffic, rather than personal tardiness) and fulfil a range of social functions (e.g. excusing, legitimising). Hulley (2016) found respondents’ ‘neutralizations’ (Sykes and Matza, 1957), for example, blaming offending on mental health issues, presented an acceptable moral identity separate from their sexual offending. Similarly, research on sexual offending highlights denial may function to maintain a person’s self-presentation so they are not ostracised (Blagden et al., 2014). Understood in this way, accounting for behaviours can be treated as building blocks of a broader narrative identity, a way of constructing a person’s character.

Situational explanations of offending behaviour are consistent with normative excuse-making behaviour (Maruna and Copes, 2005), where people give situational accounts for their misdeeds, and dispositional accounts for those of others. Maruna and Mann (2006) question if criminology and criminal justice interventions unduly pathologise ‘excuses’ of people who offend, conflating these with offence supportive attitudes. Situational accounts for offending are often labelled justifications, excuses or cognitive distortions¹ and targeted in criminal justice interventions to reduce reoffending, despite insufficient evidence (Blagden et al., 2014; Maruna and Mann, 2006; Ware and Mann, 2012), particularly in relation to sexual offending (Beech and Mann, 2002; Hanson and Morton-Bourgon, 2005).

In sexual offending treatment programmes, a tension exists between the participants’ narratives, possibly situational accounts consistent with desistance narratives, and institutional requirements for people to take full responsibility for their offences, for example, avoiding minimisation and ‘excuses’ including situational factors (Kras and Blasko, 2016; Waldram, 2010; Ware and Mann, 2012). Programmes are called to be future-focussed, developing desistance narratives, and less focussed on clients accepting full responsibility for their past offending, which potentially hinders the desistance process (Blagden et al., 2014; Farmer et al., 2016; Maruna and Copes, 2005). However, little research has looked at how ‘excuses’ are dealt with during interactions between practitioners and participants of criminal justice interventions, instead relying on retrospective interviews (e.g. Blagden et al., 2014; Bullock and Condry, 2013) and assessments of practice/programmes (e.g. Beech et al., 1999; Marshall and Serran, 2004). A more detailed examination of how people account for their offending behaviour in the context of criminal justice interventions and how this is responded to by practitioners will improve understanding of the construction of desistance narratives (Kirkwood, 2016).

Narrative as interactionally contingent

Although Maruna (2001: 8) acknowledges that ‘self-narratives are developed through social interaction’, much narrative criminological research fails to address the interviewer’s role in narrative construction, as their contributions are often omitted or under-analysed (Kirkwood, 2016). As with any setting, the research interview is itself a site of identity construction because it ‘sets parameters and asks informants to respond within those parameters’ (Presser, 2004: 38). For instance, Presser (2004) noted her participants tailored their accounts to her gender and status as researcher. Carlsson (2012) emphasised imposing analysts’ categories, such as turning points, will bias participants’ responses, possibly distorting their importance. The stake and interest of conversational participants differs between contexts and influences the accounts that are produced (Potter and Hepburn, 2005), raising questions about the ecological validity, or generalisability beyond the interview context, of desistance research based on interviews. As identities are contextually performed, analysis of narrative identity should consider the interactional context (Korobov, 2014).

In criminal justice interventions, accounts are bound by the context and constitute social actions. For instance, Waldram (2010) notes practitioners use the concept of cognitive distortions to construct client identities to fit with treatment programme aims, contrasting with clients’ autobiographical narratives. People arrange their narratives to manage responsibility for their offending behaviour, consistent with desistance narratives (Auburn and Lea, 2003). Auburn (2005) and Auburn and Lea (2003) demonstrated clients of a groupwork programme for sexual offending skilfully weave their stories to avoid negative attributions, for example, that they are minimising their offence, while maintaining a separation between their present ‘true’ self and past behaviour. More recently, Kirkwood (2016) argued that analysing criminal justice interactions between social workers and clients allows access to the ‘black box’ of practice. Using discursive psychology, he demonstrated how identities are presented, negotiated, and rejected in the talk between facilitators and clients of a groupwork session addressing domestic abuse. The discursive context is therefore central to the interactive negotiation of narrative identity.

As narrative identities are contingent on context, it is important to explore how narratives are constructed within interaction. This is especially relevant in examining how primarily talk-based criminal justice interventions can influence desistance through encouraging certain narrative identities. As such, the following assumptions underpin this study: (1) desistance narratives are key to the process of desistance, (2) narrative identities are shaped through interaction, (3) accounts of events and behaviours constitute building blocks of narrative identity, (4) accounts produce and are sensitive to local interactional contexts, and (5) in criminal justice interventions accounts of offending behaviour are elicited, offered and responded to. Therefore, interaction in criminal justice interventions is fertile ground for examining how desistance narratives are cultivated. Discursive psychology and conversation analysis methodologies examine how interaction functions, making them suitable tools for exploring the construction of desistance narratives. How the talk-in-interaction during a groupwork programme addressing sexual offending may contribute to the construction of desistance narratives, posited by previous research, has not been examined. Our analysis explores how narratives of a ‘core self’ and situational accounts for offending, consistent with desistance narratives found in research interviews, are constructed in interactions during a groupwork programme addressing sexual offending.

Methodology

This study focuses on interactions within the Scottish groupwork programme for addressing sexual offending, ‘Moving Forward: Making Changes’. This rolling programme, run in the community by local authority criminal justice social work services, works with adult men (18+) convicted of sexual offences who are Court mandated to attend. It is influenced by the Good Lives Model (GLM), a strengths-based model positing there are universal goals all humans seek, for example, happiness, relatedness (see Ward and Maruna, 2007), and offending behaviour functions to achieve these goals, albeit harmfully. The GLM proposes clients should be supported to identify and achieve their goals prosocially, building on and developing their strengths and capacities. Clients are encouraged to consider their offending in the wider context of their lives, that is, what were they trying to achieve, identifying what life goals are important to them, for example, peace of mind, relatedness, and developing appropriate ways to achieve their goals. Ward and Marshall (2007) highlight that the construction of a prosocial narrative identity is central to this process and a core component of the GLM. As such, developing a new, prosocial identity is considered central to this programme (Scottish Government and Scottish Prison Service, 2013).

Research participants gave informed consent and identifying features were anonymised. The authors’ university and relevant local authority gave ethical approval. The first author watched 12 video-recorded groupwork sessions repeatedly, approximately 28 hours, transcribing these orthographically. The local authority routinely video records the groupwork sessions for internal quality assurance. Recordings were selected for this study on practical grounds, that is, best visual and audio quality. Each group had 2 facilitators and 4–6 men convicted of sexual offences, 18 men in total. Their offences included accessing indecent images of children, rape of adults, and child sexual abuse. The facilitator team involved two men and three women, resulting in mixed and same gender facilitator variation per session.

The first author identified extracts relevant for analysing the development of desistance narratives by coding for interactions broadly featuring characteristics of such narratives, for example, situational accounts for offending, speaking about giving back, passive references to offending, presenting a ‘core self’. Coded extracts were then transcribed in greater phonological detail (Jefferson, 2004) and closely analysed, exploring how talk relating to desistance and identity arose. The second author checked for accuracy of transcription and interpretation against the video-recordings. The presented extracts, chosen for their clarity and brevity, are representative of a larger sample of similar identified patterns in that they are examples of how such interactions unfolded. The extracts are presented verbatim allowing the reader to judge the validity of the interpretation themselves, as is convention with these methods (Liddicoat, 2011).

To examine the talk-in-interaction, conversation analysis and discursive psychology (i.e. McKinlay and McVittie, 2008; Potter and Wetherell, 1987) were applied. These methods examine the micro-level utterance by utterance sequence of talk, looking at how people make sense of their conversations and what they are doing in their talk, for example, encouraging or censuring. They have been used to analyse interactions in general physician (GP) consultations (Heritage and Robinson, 2011), child protection helpline conversations (Hepburn and Potter, 2007), and mediation intake calls (Stokoe, 2013). Here, language is treated as actively constructing social reality and accomplishing social functions (Liddicoat, 2011; McKinlay and McVittie, 2008). For example, Kirkwood (2016) noted clients of a domestic abuse programme demonstrated ambivalence to pro-social identities, and practitioners responded to this resistance, encouraging acceptance of these identities. How narrative identity is conferred, contested and constructed is therefore visible within interaction. This article examines how accounts and narratives are offered up and responded to in the context of the interactions in this groupwork programme.

Analysis and findings

As outlined, attributing offending behaviour to situational, external and specific factors is consistent with desistance narratives (Maruna, 2001, 2004). Previous research implies such attributions originate in the speaker, rather than through interaction with others in specific contexts. However, for example, a young person accused of assault may give a different account speaking to their friend than to the police. The person being told the ‘story’ influences its shape, through their status, the context and their questions and responses (Drew, 2012). Specifically when interaction is bound in institutional talk, such as criminal justice social work settings, there are constraints on what contributions are allowable or relevant to achieve the institutional business (Heritage, 2005). Extract 1 highlights how subtle shifts in the facilitator’s language can encourage a situational account for offending, separating the self from the behaviour. In this extract, Brian is asked to reflect on the links between his broader life experiences and his offending. G# denotes group facilitators; all other names are clients.

Extract 1
1	G2:	We’ve got to know you a bit. It sounds like Evan’s kind of drilling down
2		a wee bit here though. What’s- I mean what’s relevant cause this is not just about
3		saying what was your childhood like it’s like it’s about working out what’s
4		relevant to kind of >how you how you< <why you’re here
5		[°somehow°>
6	Dale:	[wh what what made you basically offend
7	Evan:	Because-
8	Dale:	OR be in a situation that [you got done for-
9	Brian:	[eh::::
10	Dale:	you got charged as an offender or something [like that.
11	Evan:	[yeah because you are
12		an intelligent person and you’ve got a good head on your shoulders.
13	Brian:	Hmm

G2’s question (ll.1-5) highlights the institutional business at hand: requesting a narrative account to which Brian has access. Here, G2 avoids the interrogative ‘how’ instead repairing to ‘why’ (l.4). What does this repair achieve? ‘How’ questions are commonly answered by describing the means by which something occurs (Hayano, 2013), possibly implying an expectation that the client provides the formal account of his offence (Waldram, 2007). ‘Why’ questions can be more challenging, for instance, suggesting the situation does not accord with common sense, requiring respondents to justify their behaviour, with related social difficulties for the interaction (Bolden and Robinson, 2011); in this case, Brian is invited to provide an evaluative account that explains and perhaps justifies his ‘being here’ (i.e. why he offended). However, stating ‘why you’re here’ (l.4), rather than using more explicit statements (e.g. ‘why you committed a sexual offence’), moderates the challenging tone of the question by drawing on the group’s shared understanding of the ‘problem’. Demonstrating empathy, G2 treats the topic as sensitive by hedging her questions (‘kind of’ l.4), softening her statement through the utterance ‘somehow’ (l.5), and speaking more slowly and quietly. ‘Somehow’ (l.5) also implies a lack of agency, that Brian attending a programme for addressing sexual offending may be accidental. As such, the design of G2’s question, given her institutional status, enables an account focussing on situational factors (Heritage, 2005).

Conversation analysis is primarily concerned with what conversation participants do in interaction, rather than analyst categories, that is, how participants treat and make sense of each other’s talk (next turn proof procedure; Edwards, 2004a). Here Dale’s development of G2’s question evidences it is heard as permitting, even encouraging, a situational account. He makes the topic explicit (l.6), re-formulating the question in a manner that downplays Brian’s agency in offending and highlights the situational (ll.6, 8, 10). Line 6 could be requesting a situational cause for offending; something that made Brian offend. However, it is ambiguous; Brian could disclose he’s sexually attracted to children, a more stable, internal attribute. By re-formulating this to ‘be in a situation’ (l.8), the request for a situational account is made explicit. Finally, Dale’s self-correction from ‘you got done for-‘ (l.8) to ‘you got charged as’ (l.10) reduces agency and accountability by placing Brian as a passive actor in his arrest and subsequent conviction. Getting charged as an offender rather than being an offender or committing an offence allows deniability of the characteristics and predicates of the category ‘offender’. Brian is not labelled a ‘sexual offender’, a category which implies stable, internal traits of deviance and intractability (Levenson et al., 2007). Evan’s dispositional description of Brian as intelligent with a ‘good head’ (ll.11-12), presents Brian’s character or ‘core self’ as positive, positioning his offending behaviour as an aberration. The request for Brian’s account of ‘why [he’s] here’, constructed between the facilitator and the other group members, invites a situational account for his offending behaviour, separate from his ‘core self’. As such, normative excuse-making behaviour is enabled, alongside the maintenance or development of a prosocial identity, as Brian is positioned as someone ‘who should know better’.

The concept of a moral ‘core self’ was a common theme in the groupwork sessions. In positioning the offending behaviour as situational, and out of character, the ‘core self’ is protected from negative moral implications. Clients here commonly used the passive voice to distance their ‘core self’ from their offending behaviour, that is, ‘it just happened’, like Maruna’s (2001) observation. By using the passive voice, however, clients risk being seen to be minimising their offending behaviours. This could be problematic in a groupwork programme addressing sexual offending given the emphasis on people taking responsibility for their offences (Waldram, 2007; Ware and Mann, 2012). Here, through delicate discursive work, clients place offending behaviour as external to their self without being censured for minimising their offences, as demonstrated in Extract 2 below as Carl deliberates about disclosing his offending history to his girlfriend, a requirement of his licence conditions.

Extract 2
1	Carl:	And I and I think the fact which I’m not (.) agreeing and I’m not saying that it’s
2		any worse or any less than anything else
3	G4:	uh hm
4	Carl:	but the fact that like I was fourteen fifteen at the time
5	G4:	uh hm [I remember you saying
6	Carl:	[<I don’t know if that would be>
7	G4:	((clears throat))
8	Carl:	like eh more like accepting of it or if it would be worse the fact that
9	G4:	hmmm
10	Carl:	she’s got a seventeen year old son which (.) she could be thinking well when
11		he was like fourteen would he have made that same mistake how can
12		[he make that mistake=
13	G4:	[hmm
14	Carl:	=or sh she she could look at it people make mistakes I don’t know (.) but I’ve
15		(1)
16	G4:	Right eh certainly what I’m getting out eh a sense of here Carl is that your
17		relationship is very important for [you.
18	Carl:	[yeah
19	G4:	You you have a long longer term view of [this
20	Carl:	[yeah yeah

Through narrative reflexivity, providing a here and now commentary within the course of the narrative (Auburn, 2005), Carl manages the risk of being seen to minimise his offences: ‘which I’m not agreeing and I’m not saying that it’s any worse or less than anything else’ (ll.1-2). He places his offending as specific to when he was a teenager (l.4), providing temporal distance between his past and present. G4 aligns with Carl’s stance: ‘I remember you saying’ (l.5). Reference to the category ‘teenager’ also highlights associated stereotypical characteristics; that is, impulsive, irresponsible, risk taking and, importantly, a stage which one grows out of. The passive use of the word ‘it’ (l.8) (Maruna, 2001) and referring to his offending as a ‘mistake’ (l.11, l.12, l.14) further separates the behaviour from the self, implying it was an error, not intentional, and as such situational, external and specific. By referring to his girlfriend’s possible evaluation (l.14), Carl reports her possible reaction rather than his beliefs, further mitigating against being assessed as minimising. Carl’s lengthy and tentative construction of his offending behaviour is said with hedged explanations, possibly seeking indications the facilitators or other group members accept this narrative account, or in expectation of challenge. His account, however, is not challenged or problematized, instead G4 contextualises the importance of this relationship for Carl. G4’s minimal tokens (e.g. ‘hmmm’) align with Carl’s account, encouraging him to continue his story without endorsing it (Stivers, 2008). No direct challenge or rejection implies Carl’s account is accepted, institutionally reinforcing his story (Heritage, 2005). In doing such delicate discursive work clients can present as accountable while also attributing causes to external factors. Furthermore, these accounts are not merely given and received, but tentatively presented to the group for evaluation and adaptation.

At times, passive positioning is oriented to by the facilitators and other group members. In Extract 3 below, Frank outlines his goals for the programme under the Good Lives Model domains, specifically ‘Knowledge: Learning and Knowing’ (see Ward and Maruna, 2007).

Extract 3
1	Frank:	Well learn from my past (.) mistakes I’ve made °my offending°.
2		(5)
3		how it come to be .hh ((small shrug)) (.) hh
4	G5:	Because Frank you were saying just:: at the break just before coming into
5		this that actually (.) .hh that’s what keeps you going at the moment
6	Frank:	Yeah
7	G5:	is a real motivator for you is wanting a bet-
8	Frank:	>yeah that’s right<
9	G5:	better understanding of? °can you,°
10	Frank:	°why it all come to that yeah°
11		(3)
12	G5:	So a better understanding of why you came to offend.

Like Carl in Extract 2, Frank refers here to his offending behaviour as ‘mistakes’ (l.1), again implying his offending behaviours were errors, situational and specific. Here, however, he is presented as agentic (‘learn from my past’; ‘mistakes I’ve made’), actively referencing his offending through use of the possessive pronoun ‘my’, although it is whispered (l.1). Whispering can indicate upset, and is associated with crying (Hepburn, 2004). On line 3, Frank provides a more passive account (‘how it come to be’), one he echoes at line 10. Referring to offending behaviour as ‘it’ passively happening or arising, that is, ‘come to be’, again serves to separate the offending from the person. Frank’s passive account, which places his behaviour as external to himself, is not directly challenged; however, the facilitator modifies it through lexical substitution (Rae, 2008). Echoing Frank’s passive verb use, this modification aligns with Frank’s situational account while the pronoun ‘you’ places him as an active and accountable agent – ‘why you came to offend’ (l.12). Frank can distance himself from his offending behaviour, providing an opportunity to develop or maintain a positive ‘core self’, potentially facilitating the development of a desistance narrative, while still being held accountable for his behaviour.

Clients were also explicitly encouraged by facilitators and other group members to accept a narrative of having a good ‘core self’ separate from their offending behaviours. However, as Kirkwood (2016) noted, clients may demonstrate ambivalence to accepting prosocial identities, an ambivalence facilitators orient to, highlighting positive aspects of self to enable narratives of change inherent in ‘secondary desistance’ (Maruna and Farrall, 2004). This encouragement is evident in Extract 4. Fred outlines his goals for the treatment programme, under the goal ‘Happiness’. Twice G1 points towards previous comments recorded on a flipchart visible to the group to highlight his point.

Extract 4
1	Fred:	To look at myself in the mirror and say I am a good person
2		(1)
3	G1:	Hmm
4	Fred:	(be happy then) where now I’m looking in the mirror thinking °nah° not
5		doing it for me
6		(3)
7		(that’s what I get/ just full of guilt)
8	G1:	So so your goal in terms of happiness what I’m picking up from that is
9		that (.) eh eh eh something about (.) this ((points to flipchart)) I wonder
10		if it’s connecting to this again. You know you want to tell yourself that
11		(.) you’re (.) and this (.) ((points to flipchart)) that you’re you’re not
12		someone who’s defined by your offences that Brian said you know that
13		you’re someone else (.) you’re a good person
14	Fred:	yeah °aye°

The comparison Fred makes between his future hope (l.1) and current experience (ll.4&5) implies he considers himself a bad person because of his offending behaviour, thus offering a global, general and negative assessment of his character. Orienting to this, G1 reformulates Fred’s statement, portraying the offending behaviour as part of Fred’s history not the totality of it (ll.11-13). G1 proposes separating the behaviour from self, rejecting categorisation as a ‘sex offender’ and offers a moral ‘core self’ – ‘you’re a good person’ (l.13). Fred agrees with this characterisation (l.14). Referencing other group members’ contributions, by referring to the flipchart and Brian’s statement, strengthens G1’s formulation. Furthermore, the group members’ responses socially ratify this way of constructing the narrative, to separate the behaviour from a positive ‘core self’ as Maruna (2001) proposed desisters do.

Promoting narratives that separate offending behaviour from the person, highlight a positive ‘core self’ and place offending behaviour as external, specific and situational is constrained by priorities of risk management and public protection. That is, although these narratives are constructed in the talk-in-interaction during sessions of this groupwork programme, they are tempered and constricted by risk discourse. With extract 5, we present a ‘deviant case analysis’ (Edwards, 2004b), illustrating that practitioner responses do not always simply accept external, specific, situational accounts of offending. Here, Evan is introduced to the module called ‘Relationship Skills’, which explores individuals’ risks, needs and strengths for developing and maintaining successful relationships. Evan is asked to consider what has been problematic for him in relationships, in relation to his offending.

Extract 5
1	G2:	What links do you make Evan between relationships in your life and how you
2		came to offend an- and the situation that you did that?
3	Evan:	Just trying to be close with someone (that’s all really)
4	G2:	Okay okay
5	Evan:	I shouldn’t- someone showing me a little bit of eh (.) maybe not kindness but
6		interest.
7	G4	[hmm
8	G4:	[mh hmm
9	Evan:	I didn’t know the person at the time you know I will go into more detail at
10		the time who it was.
11	G4:	mh hmm
12	Evan:	Em (.) I shouldn’t have got close but they got close with me first and I took
13		that one step closer and I shouldn’t have done that one step [closer.
14	G4:	[hmm
15	Evan:	I never started anything I know that for a [start
16	G4:	[hmm
17	G2:	But that area though of (.) being close feeling close
18	Evan:	yeah
19	G2:	is an area which you can link to your offending >which is linked to
20		relationship
21	Evan:	Yeah [yeah
22	G2:	[styles< and an-
23	Evan:	It’s like jumping into one relationship from the shop
24	G2	ah hah

Like Extract 1, saying ‘how you came to offend’ and ‘the situation you did that’ (l.2) implies circumstantial reasons for Evan’s offending behaviour while also placing him as accountable through repeated use of the pronoun ‘you’. However, questioning the association between Evan’s relationships and his offending behaviour also allows for wider, global attributions, including possible dispositional ones (e.g. his attachment style or view of women). G2’s formulation of Evan’s account (ll.17-22) strengthens the implication that Evan’s offending is linked to more enduring traits. Formulation is when a version of events is proposed that follows from another’s own account but introduces a transformation (Antaki, 2008). G2 interrupts Evan’s justifying account by extending his sentence, ‘but that’ (l.17). This technique can adjust the focus of a previous speaker’s statements in a non-challenging way, appearing as a mere continuation of their talk (Vehviläinen, 2003). Avoiding direct challenge, G2’s formulation sidesteps Evan’s justifications and minimisations (‘I didn’t know the person’ l.9; ‘they got close with me first’ l.12; ‘I never started anything’ l.15) and makes ‘being close feeling close’ (l.17) relevant to his offending. This refocuses to the topic of the exercise and identifies an area of need for Evan, having intimacy in relationships, which he may have pursued in a harmful manner through his offending behaviour, as theorised by the GLM.

G2’s orientation to relationship styles (ll.20-22) dismisses a purely situational account for Evan’s offending, implying it was also a function of his learned and enduring relational behaviours which may indicate areas of risk, for example, hostility towards women, sex as emotional coping (Hanson et al., 2007). This balance of accounting for offending behaviour and being accountable for offending behaviour is central throughout the data as the personal and institutional narratives meet, echoing Waldram’s (2007, 2010) observations. However, contrary to Waldram, the personal and institutional narratives here do not appear to clash, even if they may be in tension. Instead, as these are brought into interaction, they appear to shape stories that incorporate both change and risk, as different elements are oriented to by the interlocutors and thus made relevant to the overall narrative.

Discussion

Through detailed analysis of interaction in a criminal justice setting, this study illustrates how criminal justice practitioners shape desistance narratives through subtle and explicit aspects of their talk. Facilitators and other group members ask questions in ways that invite situational accounts (extract 1). This is evident in how speakers repair and reword their contributions to tone down the agency implied by questions about offending behaviour. They can also affirm a moral ‘core self’, which further reinforces a situational account, as it suggests the individual is characteristically disposed to acting morally and their offending behaviour is an aberration.

The specific way in which people refer to their offences, for example, ‘mistakes’ (extracts 2 and 3), can emphasise the aberrant nature of their offending behaviour. However, in the context of criminal justice interventions, people orient to the problematic potential for such accounts to be seen as ‘minimising’ or ‘excusing’ their behaviour. In this instance, people use ‘narrative reflexivity’ (Auburn, 2005) in explicitly commenting on their accounts and managing these problematic interpretations. This demonstrates their moral awareness and heads off trouble. The clients also manage potential problems in their accounts by emphasising the potential learning from their past offending behaviour. Past harmful acts are reconstrued positively as an opportunity to learn and improve, while clients are held accountable. This is particularly evident where groupworkers distinguish between harmful past acts and a morally good ‘core self’, affirming the individual is not defined by his offences, reinforcing hope and the potential and commitment to change (extract 4). Emphasising learning from past behaviour as a vehicle for change echoes Maruna’s (2001) findings, where desisters’ highlighted their experiences of offending as central to making them who they are today.

However, given the dominant concern with risk in criminal justice interventions, accounts emphasising situational explanations for past offences and a positive ‘core self’ are not permitted to be produced unregulated. At times, facilitators orient to aspects of accounts suggesting continuing patterns of behaviour that may sustain future offending (extract 5). In these cases, practitioners withhold affirmation of the accounts (e.g. through minimal responses such as ‘hmm’), instead emphasising the problematic nature of such patterns of behaviour possibly linked to ongoing risk of offending. Narratives of desistance and narratives of risk can be in tension in this context, where there is an institutional responsibility to assess and manage risk.

This analysis illustrates how criminal justice practitioners are actively involved in the process of re-storying people’s narrative identities. In Cavarero’s (2014) terms, a person is the protagonist of their life story, but not necessarily its narrator; rather, multiple voices tell the narrative. Using a different metaphor, we suggest that the co-construction of desistance narratives can be understood as a form of sculpting, as the individual, practitioners and peers are working with the same materials (the individual’s past behaviour, present intentions and future aspirations), collaboratively crafting an institutionally, socially and personally acceptable narrative. Rather than considering secondary/identity desistance and tertiary/relational desistance (Maruna and Farrall, 2004; Nugent and Schinkel, 2016) as separate, although intertwined, processes, here it is evident that desistance identities are shaped in relational contexts, through dialogue and interaction. The analysis of identity in interaction provides an understanding of how interactions shape or reinforce identities, which offer ways of making meaning and directing behaviour. Practitioners translate macro-level discourse regarding risk and offending behaviour into the micro-level interaction. Whereas others have suggested individuals’ autobiographical narratives come into conflict with the institutional narratives (Waldram, 2007), this study demonstrates collaborative interactional work shaping a narrative that is acceptable to the individual and the institution, providing a recognisable self that reflects societal norms and mitigates risk.

Our research supports previous work critiquing the notion of cognitive distortions (Auburn, 2010; Auburn and Lea, 2003; Maruna and Mann, 2006), further illustrating situational accounts are not necessarily pathological, but rather normative, and potentially support the creation of liveable self-identities as the foundation for desistance narratives. However, criminal justice contexts where the risk paradigm predominates may be unsupportive of desistance narratives, instead reinforcing self-identities predicated on continuous risk of offending (McNeill, 2016). The need for risk assessment and management is not negated, but corresponding balanced engagement with narratives of change and desistance is necessary. Moreover, desistance narratives are probably unsustainable when society and other institutions treat risk as ever-present and people convicted of sexual offences as intractable, as is common-place (Harris, 2017; Levenson, 2018). As we’ve argued elsewhere (Kirkwood, 2016), narrative reconstruction alone is not sufficient for desistance; behavioural change, skills development and pro-social opportunities are also necessary (McNeill, 2006). While recent research suggests identity change plays a causal role in desistance (Rocque et al., 2016), further longitudinal research is required to examine the impact of interactional aspects of criminal justice practice on desistance.

Besides demonstrating the GLM’s strengths-based focus and intention to foster pro-social identities (Ward and Marshall, 2007), the interactional style evident in this setting reflects Braithwaite’s (1989) model of reintegrative shaming, which condemns the harmful act, emphasising the moral goodness of the individual and supporting them with positive change. It also reflects core social work skills, such as active listening, demonstrating empathy, and encouraging self-efficacy (Raynor and Vanstone, 2015). Interestingly, other group members also displayed these skills. They were possibly socialised into pro-social ways of responding or demonstrating modes of interacting common in everyday interaction where people show alignment, warmth and support. Further research could explore interactional styles in different criminal justice settings with different client groups and connections with desistance narratives. Effective social work skills are likely to support desistance narratives, in being non-stigmatising, empowering and working with people’s own understanding of themselves. Moreover, everyday interaction can support desistance narratives where it affirms people’s essential goodness while censuring their harmful behaviour, supporting their potential to change.

Footnotes

Appendix 1

Transcription notation adapted from Jefferson (2004).

Acknowledgements

We would like to thank those who supported and participated in the research. Thank you also to Eric Laurier, Bill Whyte, Fergus McNeill, Briege Nugent, John Devaney, Bethan Benwell and Shawn Bodden, to members of the Scottish Ethnomethodology, Discourse and Talk (SEDIT) and Ethnomethodology (EMCA) Doctoral Network research groups, and the anonymous reviewers, for their helpful advice.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Economic and Social Research Council as a collaborative studentship with the Risk Management Authority and a local authority, in Scotland [grant number 1611091].

ORCID iD

Eve Mullins

Notes

Author biographies

Eve Mullins is completing her PhD at the University of Edinburgh, exploring social work practice skills in a groupwork programme addressing sexual offending. She is also a qualified social worker, with experience working in Criminal Justice services, and is a lecturer in Social Work at Edinburgh Napier University.

Steve Kirkwood is a senior lecturer in Social Work at the University of Edinburgh. His research addresses criminal justice, restorative justice, desistance from offending, and the integration of refugees, often using discourse analysis or evaluation research methods. He has published widely in criminology, psychology and social work.

References

Antaki

(2008) Formulations in psychotherapy. In: Peräkylä

Antaki

Vehviläinen

, et al. (eds) Conversation Analysis and Psychotherapy. Cambridge: Cambridge University Press, pp. 1591–1599.

Auburn

(2005) Narrative reflexivity as a repair device for discounting ‘cognitive distortions’ in sex offender treatment. Discourse and Society 16(5): 697–718.

Auburn

(2010) Cognitive distortions as social practices: An examination of cognitive distortions in sex offender treatment from a discursive psychology perspective. Psychology, Crime & Law 16(1–2): 103–123.

Auburn

Lea

(2003) Doing cognitive distortions: A discursive psychology analysis of sex offender treatment talk. British Journal of Social Psychology 42(2): 281–298.

Beech

Fisher

Beckett

(1999) STEP 3: An Evaluation of the Prison Sexual Offender Treatment Program. Available at: https://webarchive.nationalarchives.gov.uk/20110218143308/http://rds.homeoffice.gov.uk/rds/pdfs/occ-step3.pdf

Beech

Mann

(2002) Recent developments in the assessment and treatment of sexual offenders. In: McGuire

(ed.) Offender Rehabilitation and Treatment: Effective Programmes and Policies to Reduce Re-offending. Chichester: John Wiley & Sons, pp.259–288.

Blagden

Winder

Gregson

, et al. (2014) Making sense of denial in sexual offenders: A qualitative phenomenological and repertory grid analysis. Journal of Interpersonal Violence 29(9): 1698–1731.

Bolden

Robinson

(2011) Soliciting accounts with why-interrogatives in conversation. Journal of Communication 61(1): 94–119.

Braithwaite

(1989) Crime, Shame, and Reintegration. Cambridge: Cambridge University Press.

10.

Bullock

Condry

(2013) Responding to denial, minimization and blame in correctional settings: The ‘real world’ implications of offender neutralizations. European Journal of Criminology 10(5): 572–590.

11.

Carlsson

(2012) Using ‘turning points’ to understand processes of change in offending: Notes from a Swedish study on life courses and crime. British Journal of Criminology 52: 1–16.

12.

Cavarero

(2014) Relating Narratives: Storytelling and Selfhood. Oxford: Routledge.

13.

Drew

(2012) Turn design. In: Sidnell

Stivers

(eds) The Handbook of Conversation Analysis. Chichester: John Wiley & Sons, pp.131–149.

14.

Edwards

(2004a) Proof procedure. In: Lewis-Beck

Bryman

Futing Liao

(eds) The SAGE Encyclopedia of Social Science Research Methods. Thousand Oaks, CA: SAGE, pp. 874–875.

15.

Edwards

(2004b) Deviant case analysis. In: Lewis-Beck

Bryman

Futing Liao

(eds) The SAGE Encyclopedia of Social Science Research Methods. Thousand Oaks, CA: SAGE, pp. 259–260.

16.

Farmer

Mcalinden

A-M

Maruna

(2016) Sex offending and situational motivation: Findings from a qualitative analysis of desistance from sexual offending. International Journal of Offender Therapy and Comparative Criminology 60(15): 1756–1775.

17.

Gadd

Farrall

(2004) Criminal careers, desistance and subjectivity: Interpreting men’s narratives of change. Theoretical Criminology 8(2): 123–156.

18.

Giordano

Cernkovich

Rudolph

(2002) Gender, crime, and desistance: Toward a theory of cognitive transformation. American Journal of Sociology 107(4): 990–1064.

19.

Goffman

(1959) The Presentation of Self in Everyday Life. New York: Anchor Books.

20.

Hanson

Morton- Bourgon

(2005) The characteristics of persistent sexual offenders: A meta-analysis of recidivism studies. Journal of Consulting and Clinical Psychology 73(6): 1154.

21.

Hanson

Harris

AJR

Scott

T-L

, et al. (2007) Assessing the risk of sexual offenders on community supervision: the dynamic supervision project 2007-05. Public Safety Canada. Available at: https://www.academia.edu/2147102/Assessing_the_risk_of_sexual_offenders_on_community_supervision_The_Dynamic_Supervision_Project

22.

Harris

(2014) Tales from the trenches: Zooming in on the process of desistance from sexual offending. In: McCartan

(ed.) Responding to Sexual Offending. London: Palgrave Macmillan, pp.170–188.

23.

Harris

(2017) Desistance from Sexual Offending: Narratives of Retirement, Regulation and Recovery. Cham: Palgrave Macmillan.

24.

Hayano

(2013) Question design in conversation. In: Sidnell

Stivers

(eds) The Handbook of Conversation Analysis. Chichester: John Wiley & Sons.

25.

Hepburn

(2004) Crying: Notes on description, transcription, and interaction. Research on Language and Social Interaction 37(3): 251–290.

26.

Hepburn

Potter

(2007) Crying receipts: Time, empathy, and institutional practice. Research on Language and Social Interaction 40(1): 89–116.

27.

Heritage

(2005) Conversation analysis and institutional talk. In: Fitch

Sanders

(eds) Handbook of Language and Social Interaction. Mahwah, NJ: Lawrence Erlbaum Associates, pp. 103–147.

28.

Heritage

Robinson

(2011) ‘Some’ versus ‘any’ medical issues: Encouraging patients to reveal their unmet concerns. In: Antaki

(ed.) Applied Conversation Analysis: Intervention and Change in Institutional Talk. London: Palgrave Macmillan, pp. 15–31.

29.

Hulley

(2016) ‘While this does not in any way excuse my conduct . . .’: The role of treatment and neutralizations in desistance from sexual offending. International Journal of Offender Therapy and Comparative Criminology 60(15): 1776–1790.

30.

Jefferson

(2004) Glossary of transcript symbols with an introduction. In: Lerner

(ed.) Conversation Analysis: Studies from the First Generation. Amsterdam: John Benjamins.

31.

King

(2013) Early desistance narratives: A qualitative analysis of probationers’ transitions towards desistance. Punishment & Society 15(2): 147–165.

32.

Kirkwood

(2016) Desistance in action: An interactional approach to criminal justice practice and desistance from offending. Theoretical Criminology 20(2): 220–237.

33.

Korobov

(2014) Identities as an interactional process. In: McLean

Syed

(eds) The Oxford Handbook of Identity Development. Oxford: Oxford University Press. Available at: https://doi.org/10.1093/oxfordhb/9780199936564.013.030

34.

Kras

Blasko

(2016) Pathways to desistance among men convicted of sexual offenses: Linking post hoc accounts of offending behavior and outcomes. International Journal of Offender Therapy and Comparative Criminology 60(15): 1738–1755.

35.

Levenson

(2018) Sex offender management policies and evidence-based recommendations for registry reform. Current Psychiatry Reports 20(3): 2021.

36.

Levenson

Brannon

Fortney

, et al. (2007) Public perceptions about sex offenders and community protection policies. Analyses of Social Issues and Public Policy 7(1): 137–161.

37.

Liddicoat

(2011) An Introduction to Conversation Analysis, 2nd edn. London: Continuum International Publishing.

38.

McAdams

McLean

(2013) Narrative identity. Current Directions in Psychological Science 22(3): 233–238.

39.

McAlinden

Farmer

Maruna

(2016) Desistance from sexual offending: Do the mainstream theories apply? Criminology and Criminal Justice 17: 266–283.

40.

McKinlay

McVittie

(2008) Social Psychology and Discourse. Oxford: Wiley-Blackwell.

41.

McNeill

(2006) A desistance paradigm for offender management. Criminology and Criminal Justice 6(1): 39–62.

42.

McNeill

(2016) The collateral consequences of risk. In: Trotter

McIvor

McNeill

(eds) Beyond the Risk Paradigm in Criminal Justice. London; New York: Palgrave Macmillian, pp. 143–157.

43.

Marshall

Serran

(2004) The role of the therapist in offender treatment. Psychology, Crime & Law 10(3): 309–320.

44.

Maruna

(2001) Making Good: How Ex-convicts Reform and Rebuild Their Lives. Washington, DC: American Psychological Association.

45.

Maruna

(2004) Desistance from crime and explanatory style a new direction in the psychology of reform. Journal of Contemporary Criminal Justice 20(2): 184–200.

46.

Maruna

(2015) Foreword: Narrative criminology as the new mainstream. In: Presser

Sandberg

(eds) Narrative Criminology: Understanding Stories of Crime. New York: New York University Press, pp. vii–x.

47.

Maruna

Copes

(2005) What have we learned from five decades of neutralization research? Crime and Justice 32(2005): 221–320.

48.

Maruna

Farrall

(2004) Desistance from crime: A theoretical reformulation. Kölner Zeitschrift Für Soziologie Und Sozialpsychologie 43: 171–194.

49.

Maruna

Mann

(2006) A fundamental attribution error? Rethinking cognitive distortions†. Legal and Criminological Psychology 11(2): 155–177.

50.

Nugent

Schinkel

(2016) The pains of desistance. Criminology and Criminal Justice 16(5): 568–584.

51.

Potter

Hepburn

(2005) Qualitative interviews in psychology: Problems and possibilities. Qualitative Research in Psychology 2(4): 281–307.

52.

Potter

Wetherell

(1987) Discourse and Social Psychology: Beyond Attitudes and Behaviour. London: SAGE.

53.

Presser

(2004) Violent offenders, moral selves: Constructing identities and accounts in the research interview. Social Problems 51(1): 82–101.

54.

Rae

(2008) Lexical substitution as a therapeutic resource. In: Peräkylä

Antaki

Vehviläinen

, et al. (eds) Conversation Analysis and Psychotherapy. Cambridge: Cambridge University Press, pp. 62–79.

55.

Raynor

Vanstone

(2015) Moving away from social work and half way back again: New research on skills in probation. British Journal of Social Work 46(4): 1–17.

56.

Rocque

Posick

Paternoster

(2016) Identities through time: An exploration of identity change as a cause of desistance. Justice Quarterly 33(1): 45–72.

57.

Scottish Government and Scottish Prison Service (2013) The Moving Forward: Making Changes Programme. Theory Manual. Edinburgh.

58.

Stivers

(2008) Stance, alignment, and affiliation during storytelling: When nodding is a token of affiliation. Research on Language and Social Interaction 41(1): 31–57.

59.

Stokoe

(2013) Overcoming barriers to mediation in intake calls to services: Research-based strategies for mediators. Negotiation Journal 29(3): 289–314.

60.

Sykes

Matza

(1957) Techniques of neutralization: A theory of delinquency. American Sociological Review 22(6): 664–670.

61.

Vaughan

(2007) The internal narrative of desistance. British Journal of Criminology 47(3): 390–404.

62.

Vehviläinen

(2003) Preparing and delivering interpretations in psychoanalytic interaction. Text & Talk 23(4): 573–606.

63.

Waldram

(2007) Narrative and the construction of ‘truth’ in a prison-based treatment program for sexual offenders. Ethnography 8(2): 145–169.

64.

Waldram

(2010) Moral agency, cognitive distortion, and narrative strategy in the rehabilitation of sexual offenders. Ethos 38(3): 251–274.

65.

Ward

Marshall

(2007) Narrative identity and offender rehabilitation. International Journal of Offender Therapy and Comparative Criminology 51(3): 279–297.

66.

Ward

Maruna

(2007) Rehabilitation. London: Routledge.

67.

Ware

Mann

(2012) How should ‘acceptance of responsibility’ be addressed in sexual offending treatment programs? Aggression and Violent Behavior 17(4): 279–288.