Abstract
Theoretical explanations of group offending have been hindered by a focus on rational actor models of social relationships. One consequence of this has been a neglect of the dynamics of social relations and their role in group offending and desistance. Drawing illustratively on two studies conducted in the West of Scotland, this article advances an integrated theoretical framework for the comparative study of group offending that moves beyond either individualizing or ‘gang’ frames dominating existing discourse, towards a thick understanding of situated social relations. By integrating Bourdieu’s concept of habitus with Donati’s relational realist framework, this article theoretically and empirically examines the dynamics of group offending relationships, what shapes them and the way they can, in turn, shape and affect offending and desistance trajectories.
Introduction
The notion that offending, particularly youth offending, mostly occurs in groups is an established ‘criminological fact’ (Schafer et al., 2014: 117). Yet, despite large criminological sub-fields devoted to the study of gangs, co-offending and desistance, there is a notable lack of theoretical analysis devoted to the dynamics of social relations that sustain these behaviours (see Goldsmith and Halsey, 2013; Weaver, 2015). Instead, academic and policy framings focus overwhelmingly on young people as individual rather than collective actors. Where group categories are employed, it is predominantly in terms of gang activity, which narrows recognition of the everyday human engagements and relational dynamics underpinning group offending (Fraser and Hobbs, 2017). In this article, we offer a theoretical vocabulary that reaches between and beyond existing approaches by centring the concept of social relations. Rather than approaching group offending through the lens of gangs, co-offending or desistance, we privilege the social relations that underpin and sometimes undercut these categories, examining the dynamics of group relationships, what shapes them and the way they, in turn, shape and affect offending and desistance trajectories.
Informed by our secondary theoretical analysis of data from two distinct but overlapping studies, we argue for the theoretical and empirical value of placing social relations at the heart of studies of group offending, and situating those relations in the socio-cultural contexts that shape them. In making this argument, we advance an integrated theoretical framework incorporating Bourdieu’s concept of habitus into Donati’s relational realist framework and develop an innovative conceptual schema that adapts Archer’s morphogenetic framework to illustrate the empirical application of our integrated theory. This approach, we argue, transcends existing analyses focusing solely on the network as a unit of analysis (as in gang or co-offending research), or on the individual (as in much criminal careers research), by treating individual and collective offending trajectories as fundamentally social activities, shaped by the historical and socio-cultural contexts in which they emerge.
Humanizing the subject
Over three decades ago, Reiss (1988: 117) argued that ‘understanding co-offending is central to understanding the etiology of crime and the effects of intervention strategies’. Substantial research has demonstrated that most offending occurs in partnerships (Carrington, 2009; McGloin et al., 2008), that both co- and group-offences often involve more serious consequences than individual offending (Carrington, 2002; Sweeten et al., 2013) and, especially for young people (Lantz and Hutchison, 2015), extend offending careers. Yet, the relational dynamics of group offending remain largely neglected by both research and policy (Van Mastrigt and Farrington, 2009). Instead, the study of group offending has principally proceeded from an individualized or network perspective (e.g. Haynie, 2001; Marti et al., 2021; Roman et al., 2021). The general over-reliance on positivistic methods in co-offending and gangs research with their attendant ‘aggregate descriptive analyses’ (McGloin and Nguyen, 2012: 464), and the privileging of rational actor approaches (Brotherton, 2015), has had the effect of decontextualizing the meaning and texture of social relations as they pertain to individual experience. This has resulted in an incomplete understanding of the significance of social relations in processes of desistance (Rosen and Cruz, 2019; Weaver, 2015). The primarily US-based literature on gang extrication (e.g. Decker et al., 2014; Pyrooz and Decker, 2011; Pyrooz et al., 2014; Sweeten et al., 2013), for example, tends to draw on life course and social control theories, and, to a lesser extent, social learning and strain. While describing ‘what changes’ (Sweeten et al., 2013: 475), such theories are limited in their capacity to explain how and why they change in the way that they do, by eliding the role of socio-economic, and cultural conditions altogether (Sanchez-Jankowski, 2003).
Some contemporary contributions have, nevertheless, hinted at the more expansive web of social relations underpinning group offending. Some gang scholarship has demonstrated the fluidity of gang identification and that gang affiliation is embedded within wider networks of social relationships (Pyrooz et al., 2014). There have been efforts to capture connections between and beyond gang members, to identify and explore the effects of these relations between gang/co-offending network members and non-gang/offending friendships, kinship and neighbourhood networks (Marti et al., 2021; Papachristos et al., 2015), and a growing recognition that co-offenders, street gangs and gang members cannot be understood independently from the networks and neighbourhoods they are embedded in (Lammers, 2018; Lopez-Aguado and Walker, 2021; Papachristos et al., 2015; Schaefer et al., 2014). Similarly, while recent scholarship has pointed to the formation of structured ‘inmate organizations’ in prisons internationally (Skarbek, 2014, 2016), the relationality of such groups can, in certain contexts, result in multiple trajectories including networks of solidarity and support (Biondi, 2016; Levenson-Estrada, 2013). While such relations are somewhat instrumental in the form depicted by US studies (Gundur, 2018; Skarbek, 2014), Darke (2018: 237) emphasizes the ‘continuing importance of interpersonal relations of reciprocity and trust’ in Brazilian prison gangs. 1
In this article, we seek to build on these accounts of group offending by developing a theoretical framework through which to interrogate the texture and temporality of social relations as they relate to both group offending and desistance trajectories. We aim to move beyond rational actor models that emphasize transactional or instrumentalized bases for social relationships—which Newburn (2014: 3) observed resulted in ‘somewhat dry, evacuated accounts of those worlds’—towards a holistic account that engages with the invisible cords that bind social action. We suggest that research into group offending might be enhanced by the call in critical gang studies towards ‘humanizing the subjects’ (Brotherton, 2015: 96). Such an approach conceives of group offending as constituted in and through social relations, shaped by distinct socio-historical and cultural contexts, and which are therefore subject to change. To re-humanize the subject, we therefore need to conceive of an alternative theoretical lens to open the black box of group offending and expose their dynamics to critical scrutiny.
Group offending and relational sociology
In what follows, we develop a new framing for the role of social relations in group offending through engagement with the relational sociology of Pierpaolo Donati. Donati argues that agents and structures do not exist in an isolated state, but as complex relational webs in which they are defined and conceptualized in reference to the social relation. From this perspective, any sociological investigation necessarily studies a phenomenon ‘that is derived from a relational context, is immersed in a relational context and brings about a relational context’ (Donati, 2011: 14). Donati conceptualizes social relations as those bonds maintained between subjects that constitute their reciprocal orientations towards each other. It is the ‘reality in between’, the invisible ties that bind, which ‘are both the product of concrete human beings and also that which helps to forge them’ (Donati, 2011: 61), ‘which depend on the[m]. . ., but at the same time goes beyond them and exceeds them’ (2011: 26).
Donati’s (2011) concept of social relations is concerned with relazione (reciprocal or mutual interaction), rather than rapporto (the statistical relations empirically established between independent variables, characteristic of much social network analysis). From this perspective, social relations are not straightforwardly reducible to influences of one person on another. Instead, understanding how social relations work requires an examination of their effects, specifically ‘the behaviour that none of the actors [individually] “brings” to the relation but which results from their interaction’ (Donati, 2011: 126). ‘Reciprocity’ is therefore central to social relations, suggesting as it does an ‘exchange of something’. Rather than a transaction, however (e.g. Skarbek, 2014; Weerman, 2003), this is conceived as ‘a reciprocal action in which something passes from ego to alter and vice versa, which generates a reciprocal link . . . between them’ (Donati, 2011: 73). The positive effects of such relationships are what Donati calls relational goods (e.g. loyalty, trust, solidarity, care) though they can also produce relational harms (e.g. control, fear and mistrust). Thus, distinct from rational actor approaches (Uhlaner, 1989) that construe relational goods as a function of instrumental action, relational goods in this context are understood as emergent effects of social relations, ‘produced by subjects who reflexively orient one towards the other to generate a relationship from which benefits derive for all those who participate in it’ (Donati, 2019: 238). To maintain relational goods, people adjust their behaviours and make changes in the way they relate to each other. In this way, relationships motivate individuals to behave in ways that they might not otherwise have done—pro-socially or otherwise. However, as Donati observes, social relations themselves are embedded in, and influenced by, a structural and cultural context—both of which require investigation.
Though Donati’s relational sociology contains within it an account of structure and culture, and its interplay with agency, here we suggest the value of bringing Bourdieu’s concept of ‘habitus’ to bear. This is because Donati’s relational sociology depends and builds upon the social ontology of critical realism, particularly that of Archer (e.g. Donati and Archer, 2015). Donati (2011) builds on Archer’s conceptualization of reflexivity as a mediatory mechanism between structure and individual agency (personal reflexivity) but advances a mode of collective group orientation (relational reflexivity). Notably, however, Archer’s representation of reflexivity has generated much critique (e.g. Akram, 2013; Elder-Vass, 2007; Farrugia and Woodman, 2015). As Farrugia and Woodman (2015: 626) argue, her ‘rejection of any pre-reflexive dimensions to subjectivity and social action leaves her unable to sociologically explain the genesis of “ultimate concerns”’, guiding reflexive deliberation and action (Archer, 2003). While Donati (2011) argues that relationships are the context within which our ultimate concerns arise, critiques of Archer—that some actions are determined with little or no conscious effort and that her account downplays the role of dispositions that social environment instils—are applicable to Donati’s relational sociology. Both theoretically and empirically, then, ‘a language of disposition is necessary to come to terms with deeply embodied social experiences, particularly those related to stigmatization and social exclusion’ (Farrugia and Woodman, 2015: 638).
Learned through socialization, the traits of the primary habitus allow individuals to negotiate, or ‘improvise’ responses to everyday presenting situations (Wacquant, 2014); a matrix of perception through which the social world is approximated (Bourdieu and Wacquant, 1992). These dispositional traits are particular to the class-position of an individual, representing the internalization of influential objective social structures through their conversion into meaningful action. Habitus represents the cultural mechanism through which power and domination are ossified and sustained and is therefore both an artefact of entrenched economic inequality and embodied cultural practice; of ‘necessity internalized’ (Bourdieu, 1984: 166) as unconscious schemata (Wacquant, 2006). The concept therefore offers a means of approaching the dispositional remnants of deeply buried cultural structures that persist across time and space.
Notably, the concept can operate across multiple scales, enabling an understanding of social groups over time; it is a set of dispositional characteristics that is ‘both individual and collective’ (Bourdieu, 1998: 66). These forms of cultural reproduction occur through dynamic interaction between objective and subjective structures (Bourdieu, 1998), ‘in situations constrained by intergenerationally transmitted characteristics and objective circumstances’ (Robbins, 2015: xxiii). The ‘glacial force’ of habitus (Appadurai, 1996: 6) as a generative historical structure means that it is not amenable to short-run alteration; rather, the habitus ‘develops a history and generates its practices, for some time, even after the original material conditions which gave rise to it have disappeared’ (Nash, 1999: 184)—referred to as ‘hysteresis’ (Bourdieu, 1990), or ‘lag’ (Wacquant, 2016). As we illustrate, in the west of Scotland, the industrial habitus of stoicism, gameness and ‘hard’ masculinity has glided through history like a Clyde-built vessel; powered by the engine of enduring economic inequality and deindustrialization. 2
Contrary to scholars who present Bourdieu as a rigid determinist (Jenkins, 2002), habitus is neither unified nor consistent (Wacquant, 2016), but capable of reflexive reconstruction within the constraints of historical and biographical context. Nonetheless, critics have noted a lack of attentiveness to webs of group influence that shape individual decision making. As Atkinson (2016: 8) notes, ‘the intricate chains and webs of interaction, communication and association between people and things situated in concrete time-space’ are appreciably muted in Bourdieu’s work. It is here, we suggest, that dialogue between habitus and Donati’s relational sociology might productively sit, illuminating not only ‘how the possibilities for human flourishing are shaped by structural contradiction and social inequality’ (Farrugia and Woodman, 2015: 642), but as a way to understand social relations as both context (as the cultural and structural connections in a given context) and interaction (as the emergent effects in/of interactive dynamics) (Donati, 2011) in the study of group offending.
Methods
To illustrate the application of this framework, this article combines insights from two separate but overlapping studies conducted in the west of Scotland. The first data-set (2006–2010), an ethnographic study of a deindustrialized working-class community called Langview (pseudonym) in Glasgow, explored understandings, experiences and meanings of ‘gangs’ and group offending for a group of young men called the Langview Boys (Fraser, 2015)—Kev, Dylan, Gary, Mark and Willie—and comprised ethnographic observations in a youth project, street outreach setting and local high school. 3 This group had desisted from territorial gang activity through a refashioning of social relations. The second (2007–2013), a retrospective analysis of the individual life stories of six men—Seth, Evan, Jay, Andy, Jed and Harry—central to a street gang (‘the Del’) in a west of Scotland town called Coaston (pseudonym) in the 1970s–1980s, explored the role of their social relations in shaping and influencing offending and desistance (Weaver, 2015). Although situated in distinct temporal and geographical contexts, the data reveal remarkably similar sets of circumstances and relational dynamics. Both studies studied naturally forming groups of peers who engaged, to greater or lesser degrees, in group offending, thus facilitating an analysis of each group as a social relation. Unlike many studies of criminalized groups such as gangs elsewhere (Hagedorn, 2006; Vigil, 2002), both groups were ethnically white, reflecting the particular patterning of ethnicity and inequality in Scottish society (Fraser and Hagedorn, 2018).
The Langview Boys were a street-based group of young men aged 12–16, who spent much of their childhood together and among whom there was an intensely powerful bond, and with the area. Most engaged in territorial violence—several had stab-wounds and admitted to carrying a knife and, although none had received a serious charge, several had elder brothers in prison for violent offences. Rather than indicating a static identity, however, their gang violence and co-offending were dynamic and situational. Empirical data relating to their group dynamics, leisure lives and co-offending patterns were collected through participant-observation conducted over a four-year period, complemented by 18 recorded discussion groups. Data were coded and analysed inductively using the software NVivo. A theoretical model of cultural reproduction, drawing on the concept of ‘street habitus’ (Fraser, 2015), was developed.
For the second study, life-story interviews were conducted with six adult male members of the Del, then in their 40s, who were inseparable during early childhood, adolescence and early adulthood. The Del included friend and fraternal relations, whose shared history of persistent group offending, commencing in late childhood/early adolescence and continuing into adulthood comprised acquisitive offences including fraud, theft, reset and housebreaking; violent offences of assault, serious assault and possession of offensive weapons, armed robbery and attempted murder; and general public disorder offences. Life-story interviews involved participants in between two and four interviews, averaging five hours. Data were analysed using the Interpretive Phenomenological Analytic method using procedures outlined by Smith et al. (2009). While themes were generated inductively, during the analysis of the individual cases, the frequency with which each individual drew on their relationships with significant people in their lives prompted a theoretical analysis during cross-case analysis, revealing a dynamic relationship between the comparison of individual life stories and Donati’s (2011) relational sociology.
By linking both data-sets, we seek to explore continuity and change in the social relations that underpin group offending. This comparative methodology has two components. The first of these, referred to as the ‘heuristic revisit’ (Burawoy, 2003) or ‘comparative ethnography’ (Sallaz, 2009), bring together comparable data-sets from different fieldsites to enable analysis of comparable phenomena. Following Van Hellemont and Densley (2019), original fieldnotes and transcripts were shared between the authors and recoded using the theoretical framework presented in Figure 1, focused on the relationship between conditioning structures, network interaction and effects of interaction. As Van Hellemont and Densley (2019: 183) note, though such approaches stop short of a full exchange, the method has the benefit of transcending individual studies to ‘theorize gang processes that would be difficult to capture quantitatively, even in a multi-site study’. The long-term, qualitative and relational nature of the data illuminates the complexity and contingency of social relations, in context, which quantitative data-sets seldom allow.
Overview of investigative framework.
The second approach, which seeks comparison over time, is associated with the so-called ‘punctuated revisit’, in which ‘the same ethnographer conducts separated stints of field work in the same site over a number of years’ (Burawoy, 2003: 670); however, the method has also been utilized by a researcher revisiting the fieldsite of a prior study (Sallaz, 2009). This methodological strand seeks to build a longitudinal understanding of a particular field setting, to inform theoretical understandings of continuity and change. Though our fieldsites were geographically separated by 40 kilometres, as Ward et al. (2017) argue the social, economic and cultural history of the west of Scotland is so closely intertwined that it can reasonably be discussed as a holistic entity. Following Burawoy, our analysis utilized ‘inductive generalization . . . to seek out common patterns among diverse cases’ (Burawoy, 2009: 49), uncovering continuities in the structural, cultural and interactional milieu framing participants lives. In excavating these lifeworlds our ambition is not only empirical but theoretical, to reveal ‘the most profoundly buried structures of the various social worlds which constitute the social universe as well as the “mechanisms” which tend to ensure their reproduction or their transformation’ (Bourdieu and Wacquant, 1992: 7). The investigative framework, developed iteratively through the process of coding and analysis, is illustrated in Figure 1.
The framework, adapted from Archer’s (2010) morphogenetic sequence, seeks to investigate the dynamic interplay between structural, cultural and interactional drivers of social behaviour and their consequences for reproduction or elaboration over time. 4 Essentially, this is a cyclical process comprising three temporally sequential phases (Archer, 2010): structural conditioning (T1), involving interaction between structural conditions and cultural context; interactions in networks (T2>3); and outcomes (T4), in the form of observable effects of interactions, which may result in the cycle of social reproduction restarting, or in structural elaboration (T4>T1).
The conditioning influence of the structural/cultural context (T1) shapes the social environment and, in turn, individual circumstances such that some courses of action are enabled and others constrained. The properties of the structural/cultural context (conditioning structures), in this model, include the results of past actions, the accessibility of roles and resources, and the prevalence of internalized beliefs and dispositions that guide action and shape interactional norms (Bourdieu, 2005; Donati, 2011). Conditioning structures are mediated through the conduit of personal reflexivity, which guides action in the space between these enablements and constraints (T2>3). How people respond, and the effects of those actions, influence whether or not the conditioning structures stay the same (social reproduction) or change (structural elaboration) (Archer, 2010) (T4>1). In Bourdieusian terms, when the ‘dispositional architecture’ of the habitus is altered or adjusted, ‘according to both “new experiences” (Bourdieu, 2000: 161) and . . . via conscious, intentional self-fashioning’ (Friedman, 2015: 131), this is referred to as a habitus clive. As we elaborate below, this model thus accounts for the continuation of deep-seated structural conditions and individual trajectories that challenge or subvert related dispositional traits.
Beyond the individual, conditioning structures also influence the nature and form a given social relation takes in terms of prescribing the sets of relational rules advising how one should behave towards others, according to the norms and values that the context requires; norms which individuals may have internalized, and reproduce, or which they negotiate reflexively in a relational context, to bring about change. Conditioning structures thus shape the interactions and manner of relating in networks (T2>3), the nature of which then influences the outcomes to effect either structural elaboration or reproduction; in this case, desistance from or persistence in group-offending (so T4 returns to T1). Hereafter, we apply this model through examination of conditioning structures/habitus (T1), interactions in networks (T2>T3) and outcomes (T4) as they pertain to the Langview Boys and the Del.
Conditioning structures (T1)
This section explicates the nature of the conditioning structures, shaping interactions within this socio-cultural context. Langview and Coaston are settlements associated with industrial masculinities and territorial ‘gangs’. Built in the late 19th century to house workers and their families from manufacturing plants, Langview’s proximity to jobs, quality housing stock and leisure facilities meant that, in the 1960s, it was viewed as a respectable working-class community. Coaston’s population totals approximately 12,000 and, in the 1960s and early 1970s, had a number of industries providing mass employment including shipyards, factories and a power station. Since, the process of deindustrialization has devastated both areas (Torrance, 2009). Like other post-industrial regions (McDowell, 2012; Rutherford, 1988), both areas have been long associated with particular forms of ‘hyper-masculinity’ (Ward et al., 2017). These structural factors influence the accessibility of roles and related resources, the nature of inter-generationally internalized beliefs and relational rules of interaction. Together, they influence the situations of action, as well as the nature and dynamics of social relations.
For both groups, in the context of limited resources, constrained social and geographical mobility, territorial spaces became imbued with meaning through individual and collective memory. This progressively informed both personal and collective identities, embedded in, and reinforced by, the relationships which developed there (Weaver, 2015). These relations and the behaviours they engender were influenced by the relational dynamics and behavioural ‘patterns’ they were exposed to. Such interactional norms, referred to in the model as T2>3, are reproduced in a context in which capacities for violence are an accepted cultural practice emerging in the context of enduring economic inequalities. Indeed, there remains ‘cachet in Glasgow in having the reputation of a “hardman”’, a ‘symbolic compensation’ for the violence of poverty (Damer, 1992: 52). In this context, the ‘street’ is a place in which identities are negotiated, reputations earned and status conferred: Everybody hung about our street, an area that was rough. There was a lot of poor families in it, so there was a lot of kids hanging about the streets and getting up to things—a kind of gang culture, from an early age . . . life was on the street then . . . that was who you were, that was what we all did. (Seth)
Shared experiences and territorial locales can be an equalizing experience wherein acquired status depends less on accomplishments and material trappings and more on one’s personal and social resources. Friendship bonds are facilitated by not only restricted geographical mobility but perceptions of similarity and processes of identification; people usually select and identify more strongly with friends who are similar to themselves, including similar socio-economic backgrounds (Allan, 1998). This is because by ‘building affinities with others who occupy a similar social and economic location, individuals affirm their own position, cement their status and give substance to their identities’ (Allan, 1998: 693–694). In the following excerpts, both groups reflect on the role of friendship and kinship in their socialization into group offending:
Why do you think you got involved in [fighting]?
Ah think it wis jist [was just], aw ma pals were daein [doing] it. Ah’d known aw [all] ma pals fae [from] when we were all in cots . . . ye’d jist [you’d just] like grow up wi them an then that wis jist what happened.
I’d three older brothers that were well known locally for fighting, for offending, for violence, all that kind of thing so in a lot of ways I don’t think I had a choice, it was all just there. I just more or less grew into that whole life, same as my brothers and friends.
While, on the face of it, these accounts resonate with social learning theory, here we suggest a deeper historical resonance. Both excerpts reveal participants’ perceptions of the inevitability of their involvement in group offending, be it violent or acquisitive, indicative not only of the influence of social relations in prompting or guiding action, but how such norms and expectations constrained freedom. These constraints around what one can be and do delimit possibilities of alternative courses of action, and, in turn, reproduce classed and cultural trajectories.
As Bourdieu suggests, this collective habitus emerges from shared experiences of oppression. Their common perception of the world emerged from the structural inequalities embedded in the west of Scotland, and a class culture exhibiting hostility towards social mobility (Finlay, 2003) and those seen to be ‘rising above their status’ (Willis, 1977). In ‘these practices of daily life, “meritocracy” stands for a threat to solidarity [and] . . . social mobility carries social costs’ (Sennett, 2003: 98). Structural and cultural conditions thus shape individual perceptions about what is possible and meaningful to them, which may be more or less conscious, but which nonetheless shape not only their situations of action but, correspondingly, exert a conditioning influence on their aspirations and identities: There was a total feeling of hopelessness that you could never get a job . . . that was instilled in me with my Dad not working . . . you could never go to college . . . It wasn’t a kind of thing that was done, going back for education . . . Not to say that nobody ever did but, hey, in our thinking you didnae [didn’t]. (Seth)
However, belonging to the group and participating in their collective actions represented, in the absence of alternative means, a source of respect and social recognition and operated as a point of resistance to the stigma attributed to them by others in public spheres such as school and the community, ameliorating experiences of marginality and powerlessness. As Evan recalled, ‘you got a bit—when you done something notorious—know you were . . . (gestures) and “that was good [Evan]” know? . . . “you’re the man”, know?’ Similarly, two of the Langview Boys discuss being the ‘talk of the town’ after a fight:
Did that make you feel good though?
Aye. (laughs)
He’s still feeling good the noo [now].
Did everyone talk aboot [about] it afterwards?
Aye. Saying ma name.
He wis the big man.
Both groups thus actively and self-consciously appropriated elements of an idealized, if not exaggerated, configuration of hegemonic ‘traditional’ working-class masculinity (Connell, 2002) in their pursuit of status, respect and recognition. These elements informed not only their emergent identities, interactions and actions but also the relational dynamics that structured social relations within their group.
Interactions in networks (T2>T3)
The previous section showed how socio-economic conditions and cultural norms in the west of Scotland informed the gendered identities of the Langview Boys and the Del, despite the generational divide. Here, we demonstrate their influence on relational dynamics and social experiences, often absent from research into group offending. Accordingly, we explain the character of their social relations, and how they are shaped by those conditioning structures elaborated above. We also illuminate the dynamics and effects of their interactions, highlighting the relational rules that structure those relations and interactions. While illustrated in reference to the Langview Boys and the Del, this model as a socio-theoretical, analytic and explanatory framework is transferable to other groups, in other contexts, to comparatively explore not only commonalities but differences. Our premise is that this model also enables us to move beyond thin descriptions of human activity, and the factors that underpin them, to a thick understanding of social relationality, as both context, (conditioning structures) (T1) and interaction (T2>3), and how the former shapes those activities, which in turn, influence the outcomes (T4>1).
For the Del, experience of childhood trauma, including exposure to domestic, physical and/or sexual abuse, were identified as significant influences on individuals’ involvement in group offending. Resultantly, the streets provided a refuge, and association with a gang afforded a sense of protection, belonging and ultimately power, which ameliorated the sense of powerlessness they experienced and trauma endured. The insecurity, fear and threat characterizing their early lives and the frequency and intensity of their associations with each other, transformed their relationships into a stronger, more fraternal relationship: When I started getting about in gangs who seemed to have some sort of loyalty to one another . . . that helped in a way . . . you felt you belonged somewhere . . . you were a part of a crowd . . . that was a little bit protective, you were sort of indestructible. (Evan)
The nature of social relationality in relation to notions of solidarity and protection here, is markedly less instrumental, rational and transactional than depicted in many contemporary US studies, and more fraternal, emotional and reciprocal. Somewhat reminiscent of Whyte’s (1993 [1943]) ‘corner boys’, the Del’s and the Langview Boys’ lifestyles did not entirely cohere around criminality and violence, particularly during their earlier years. Rather, as Matza (1964) suggested, both groups drifted between conventional and criminal activities; their main concerns were on the collective context of their shared pursuits, on being together, as friends, of which group offending was a part and for some, over time, increasingly so. The Langview Boys spent much of their time pinballing between the streets, youth centre and park, looking for diversion through dares and risks: ‘There was carefree times . . . we used to do normal things that kids would do . . . it wasn’t all just hanging about and causing trouble. It was kind of normal things . . . doing just normal things’ (Seth). Among the Langview Boys, certain interactional dynamics, shaped by relational rules, mattered, and these influenced ‘pecking orders’ (Phillips, 2003). There was an intense desire to be involved in, and know of, everything concerning the group. Their investment in, and connection to, the group motivated participation in every activity they engaged in, including risk-taking and violent behaviours, to avoid missing anything that might have a bearing on their status in the group.
Friendships create obligations and are causally influential. Friendship can mitigate negative emotionality and stigma, and give confidence and impetus to act in a way an individual might not alone, to maintain the emergent relational goods (Donati, 2011), the pursuit of which may be more or less conscious, but which encourages or discourages certain actions of individuals-in-relation. Such actions, across both groups, were motivated by a mutual orientation towards relational goods of social trust, loyalty, solidarity and connectedness all of which rest ‘implicitly on some background of shared expectations of reciprocity’ (Putnam, 2000: 136) and from which other ends, information or resources, were derived as secondary emergent effects (Donati, 2006). Maintaining these relational goods meant fulfilling specific expectations and obligations, or, as Donati puts it, the relational rules guiding both action and interaction. For example, it was expected that you would support your friends if they were caught up in a violent incident: my pals, they meant everything. Not just to me, it was the same fae them as well. We used to call ourselves ‘the boys’ and if any of us were in trouble—the rest of us would all go. It was a tight group . . . we all had that commitment to each other. (Jay)
See if someb’dae [somebody] tried tae do [attack] us, Ah wouldnae be scared cos Ah know aw ma pals would back me up.
Have you ever had to put that to the test though?
One time up Hillside. Ah didnae want tae fight, but James got caught . . . they aw started running doon [down] . . . and when we looked around we couldnae find James. An that’s when Ah thought ‘they’ve caught James’. Then Ah done that ‘let’s fucking go!’ Me, Willie, Daz and Dylan . . .
When we thought oor [our] pal got caught, that’s when we all came round . . . see when that adrenaline’s kicking through ye . . .
We were all bottling it at the time, but see when we thought oor pal got caught, that’s when we thought ‘naw, we’re no taking this, we want tae back up our pal here’.
Far from the fixed, static, violent ‘other’ pervading the gang literature, across both studies such displays of solidarity established a sense of identity, represented a source of community and supported a way of expressing and experiencing belonging and loyalty to a group in difficult circumstances. This exaggerated representation of masculinity thus informed the social relations in which they participated as both a context (the cultural and social conditions, discussed above) (T1)) and as interaction (the effects in, and of, interactive dynamics (T2–3)). Their emergent gender identities and associated practices were infused into relational rules, influencing the kinds of bonds generated between them, guiding the form and nature of their relationships, interactions and the actions they gave rise to. While their behaviour ‘wasnae all bad’ (Jay), over time, the Del’s activities were increasingly characterized by anti-social behaviours including mobbing and rioting, fire-raising and vandalism. As they matured through adolescence, the places and spaces they occupied similarly changed, from streets to pubs, and so did the nature and intensity of their offending, with violent and acquisitive offences eclipsing earlier patterns of delinquent or anti-social behaviour, and which ultimately exerted a constraining conditional influence on their situations of actions, further down the line.
Effects of interactions (T4–T1)
Above, we clarified the nature of the social relations between the Langview Boys and between the Del, focusing on their interactive dynamics, themselves shaped by the enablements and constraints of their conditioning structures. Together these effect the outcomes at T4, which form conditional influences at the next stage in the cycle (e.g. the return from T4 to T1). As explained previously, individual responses to these enablements and constraints shape interactions and collective actions, which together are decisive for whether or not the conditional influence exerted is similar to the initial T1 (reproduction), or is distinctively different (elaboration) (Archer, 2010). Here then, we illustrate this third phase of the temporal process.
Any grouping of individuals-in-relation will necessarily mean variations in the relationships, status, involvement and meaning of belongingness among group members, which accounts for the heterogeneity of experience and effects of participation. While, as Sweeten et al. (2013) observe, for some, gang identities may become a more fixed aspect of identity and lifestyle, many will ‘mature out through a process of gradual disaffiliation and breaking away from the gang’ (Vigil, 1988: 106–107) as Jay suggests in reference to those less highly invested in the Del: You got some people that dropped off . . . They probably thought enough was enough . . . They maybe just stopped hanging out with us or they’ve may be been working away, so maybe just an opportunity arose for them and they took it—but I wouldnae say that was any of the real inner circle of pals—that was the ones that were just there or thereabouts.
Among the Langview Boys, Michael had been involved in gang violence for a prolonged period in his early teens, resulting in serious injuries, a local reputation and a police record. Nonetheless, when interviewed, Michael had withdrawn as an outcome of his reflections on the circularity and pointlessness of gang involvement. Yet, due to his previous friendships, and his continued location in Langview, he still found himself drawn into violent encounters: ‘it never goes away, cos you’re still in the area’, resonating with wider studies into gang extrication and desistance (e.g. Sweeten et al., 2013).
For the Del, a violent intra-gang feud divided the group and the violence between them escalated in frequency and intensity. How individuals and groups react and reflexively respond to changes in conditioning structures and social relations, mediated through the lens of their changing personal priorities or goals can explain differences in behaviours or outcomes. The Del splintered and some individuals responded by proceeding to offend alone, others continued to co-offend with others outwith the original group, yet others joined new groups, while another faction formed a ‘revised group’ compromising members of the original group. 5
In the context of enduring economic and structural constraints in the west of Scotland, and to escape the escalating violence, a number of the Del (the ‘revised group’) relocated to London to seize opportunities presented by the 1980s construction boom, representing a significant change in their conditioning structures, and the opportunities available to them. Adam was the first to move to access employment in steel-fixing, motivated by a reflexive intention to extricate from offending, to desist, and distance himself from the ‘relational bads’ (Donati, 2011) the feud generated. His resolve was further underpinned by his emotional connection to his partner and a desire to maintain their emergent ‘relational goods’, which continued offending and its outcomes threatened. Nonetheless, concerned to support his friends, Adam encouraged them to relocate and trained them in steel-fixing. Adam’s concern for his friends can be construed as evidence of his application of reflexivity not simply to himself or his individual social mobility but to his relationships. Having been a leader in the group, he now exercised leadership differently, consistent with Donati’s (2011) concept of relational reflexivity. 6
While economic and social changes to their conditioning structures (facilitated by employment) were enabled by the construction boom, the development of the necessary skills, and their capacity to access these opportunities, emerged from the reciprocal exchange of support and resources among the revised group. The changes in their conditioning structures were outcomes of Adam’s relational reflexivity and of the collaborative efforts (interactive dynamics) and reflexivity of the revised group. Re-establishing a collaborative relational network in a new location facilitated the re-emergence of the relational goods of social trust, solidarity and social connectedness threatened by the feud, from which other ends, including new knowledge and skills, employment and economic resources, were derived as secondary emergent effects (Donati, 2006).
Within this process there is evidence of continuity and change; change in context, opportunities and ‘field’, but continuity in habitus permeating the new field they encountered. Within the male-dominated environment and hard-working culture of the steel-fixing industry, work represented an alternative means of accomplishing masculinity, and acquiring self-respect and social recognition consistent with their idealized configuration of what it meant to be a man, and among men with whom they shared a culturally and classed recognizable habitus. The opportunities presented by the move altered their conditions of existence (Bourdieu, 1980), or conditioning structures (Donati, 2011), in the form of new weekly routines, new social relationships and employment-based networks, economic stability and concrete opportunities for new experiences, triggering a reflexive re-evaluation of the value of participation in offending, and an openness to new possibilities. These opportunity changes influenced shifts in individuals’ priorities and behaviour, which correspondingly exerted a constraint on the behaviour of others, who found they had less support from their desisting peers for offending. This reflected and generated changes in interactive dynamics and relational rules in this revised relational context, to which they responded by modifying their behaviour in the form of desistance, motivated by a desire to support each other and to maintain their relationships.
Concluding discussion
In keeping with critical gang studies (e.g. Brotherton, 2015; Hagedorn, 2008), our aim has been to advance a theoretical framework that both contextualizes and explains the social dynamics of group offending, and which is broad enough to accommodate cultural divergence and difference, as well as comparative analysis. We suggest that the situated study of the everyday social relations that pattern people’s lives and underpin group offending and desistance has significant potential to challenge the narrow framings characteristic of much contemporary research, policy and practice and to create a more nuanced, situated and social understanding of these dynamics, their effects and responses to group offending. Our groups were not corporate actors, engaged in transactional relationships, motivated by self-interest and oriented to maximizing personal benefit (e.g. Skarbek, 2014; Weerman, 2003); they were multiply marginalized, disenfranchised and disadvantaged people and it was in and through their friendships that they found the sense of belonging, status and respect denied to them in other spheres. The very human emotional need for social relatedness, belongingness, is a powerful driver of behaviour. What is missing in the existing research literatures, and what we have sought to advance, is such an appreciation and elaboration of the situated, social dynamics of group offending, and how they are suffused with normative concerns linked to the character and obligations of reciprocity in social groups. What this analysis therefore contributes to understandings of group offending more broadly is the impact that a given social relation has on individuals’ behaviour. That impact is attributable to the bonds maintained between people that constitute their reciprocal orientations towards each other and to the emergent effects of their interactive dynamics, which are shaped by and shape their conditions of existence. We propose that taking the social relation, the group, rather than the individual or network, as the primary unit of analysis can reveal a more nuanced understanding of the dynamics of collectives or social groups in general, and those involved in group offending in particular. The model proposed, therefore, facilitates an analysis of the social dynamics of group offending, and the socio-cultural influences that shape them, which our study reveals are, in this context, enduring across generations.
Relatedly, while we agree with Schaefer et al. (2014: 119) that ‘the spatial context in which youth are embedded affects their social interactions in the form of co-offending’, we have elaborated both how and why we need to move beyond identifying the neighbourhood characteristics and conducive contexts facilitating group offending. Rather, we need to conceptualize social environments as places and spaces in which identities are formed (Lopez-Aguado and Walker, 2021) and cultures and relations are re-/produced, shaping particular norms of interaction and action. It seems to us that focusing solely on the activity of group offending to the elision of other related aspects of people’s identities, activities and relationships, precludes a fuller analysis of human (inter-)action and the processes that shape them. Analysis of these localized social relations, informal networks and situated trajectories also has the potential to challenge not only one-dimensional views of gangs, but also other criminal collaborations such as organized criminal groups. As Smith (2019: 7) recently demonstrated in a historical, gendered network analysis of Prohibition-era Chicago, organized crime then was not, as previously thought, singularly composed of strict hierarchies and distinct groups but a ‘small, loosely clustered, non-hierarchical, decentralized group of politicians, police officers, collectors, and illicit business owners’.
Rather than advancing a general theory of group offending, we have advanced an integrated socio-theoretical framework through which to empirically, and comparatively, analyse group relations, interactions and associated behaviours in context. Our conceptual schema, uniquely, brings Bourdieu’s concept of habitus into Donati’s relational realist framework, utilizing Archer’s morphogenetic sequence or three-part cycle of change, illustrating the interplay between conditioning structures, interactions in networks and their combined effects. Consequently, it offers a means of bottom–up theory-building that is rooted in particular geographies, as opposed to a top–down model that evacuates cultural context. This article thus responds to the need for a more inclusive, and expansive theoretical framework to facilitate global comparative analysis of similarity and difference. We contribute, then, a framework for theoretical and empirical analysis that can elucidate localized meanings, historical antecedents and cultural influences—all of which are largely absent from research into group offending. This framework also enables comparative analyses, in similar and distinct jurisdictions, potentially encouraging global comparative research less predicated on the discovery of similarities (Fraser and Hagedorn, 2018), but on surfacing difference, and embedding explanations for group offending in ‘space, time and social setting’ (Brotherton, 2015: 25).
Beyond the empirical and theoretical implications of this work, unless policymakers and practitioners engage constructively with these relational and cultural contexts and conditions, their efforts to influence individual behaviours may be undermined. If responses to co-offending are merely punishing or incapacitating, they are likely to have the effect of compounding conditions of marginality that underpin participation in group offending. Desistance from group offending is likely to be encouraged through processes and responses that are inclusive and allow people to fulfil their reciprocal obligations. The importance of these social relations to individuals who co-offend underlines the need to ensure that a focus on social relationality, reciprocity and solidarity, and social integration inform the means through which policymakers and practitioners respond to group offending and support change.
Footnotes
Funding
The authors received no financial support for the research, authorship and/or publication of this article.
