Skip to main content
SearchLoginLogin or Signup

The meaning and use of weapons in English prisons

Weapon use is as risky in prison as it is in the community, but the type, use and meaning of weapons differs between these settings. Consequently, knowledge about community-based weapon violence may not generalise to prison contexts. Using a framework for understanding weapon...

Published onJan 23, 2024
The meaning and use of weapons in English prisons


Weapon use is as risky in prison as it is in the community, but the type, use and meaning of weapons differs between these settings. Consequently, knowledge about community-based weapon violence may not generalise to prison contexts. Using a framework for understanding weapon selection and use in the community, six prisoners with a history of weapon possession and use in prison discussed their selection and use of weapons. They viewed prison as a warlike environment where weapons were essential to reducing uncertainty and managing identity and where the advantages of weapon possession greatly outweighed the official consequences. The study also suggests that a framework developed for understanding weapon use in the community can be applied effectively in a prison context.


Weapons are used by human and primates as tools to physically control and harm others (Löffler et al., 2019). Worldwide, they contribute to considerable interpersonal harm from violence resulting in injury and death (Brennan et al., 2006). Annually, sharp objects and firearms are used in around 70% of the 450,000 homicides worldwide (United Nations Office for Drugs and Crime, 2019). In England and Wales, 21% of violent crimes against an adult in 2019/20 involved a weapon1, rising to around 70% for homicide (House of Commons Library, 2021). The strong and disproportionate association between weapon use and the harm inflicted in violent incidents indicates that a great deal of the burden of violent harm is attributable to weapon use and removing them from violent incidents would dramatically reduce this burden.

Weapon use is also a significant problem in custodial settings. Weapons used in prison can be modified everyday items, such as ‘shanks’ and ‘saps’, unmodified items, such as pool cues, dumb bells and hot water or offensive weapons per se, such as knives or batons (Lincoln et al., 2006). Across the adult and youth estate in England and Wales, between April 2018 and March 2022, the proportion of assaults that the Ministry of Justice recorded as involving a weapon was, on average, 24%2, accounting for almost 30,000 assaults or an average of around 6,000 assaults per year (Ministry of Justice, 2023).

Although information is collected about the prevalence of weapon use and the broad categories of weapons used (Ministry of Justice, 2023), little research has been undertaken to understand weapon-carrying and weapon selection behaviour in custody. Given the disproportionate levels of harm associated with weapon use in violent incidents, a better understanding of weapon-related decision-making and behaviour could be used to create safer environments for prisoners and prison staff.

Social-ecological explanations of weapon-carrying and use

Brennan (2019) noted how risk factors for weapon use extend across the social-ecological framework (Bronfenbrenner, 1979) incorporating individual, interpersonal and community factors. This framework purports that, although behaviour is enacted by individuals, that behaviour is affected by interpersonal, organisational, community and societal factors. Accordingly, addressing the causes of weapon-carrying requires a comprehensive appraisal of each level and their interaction (Bullock et al., 2023).

At a societal level, weapon use is correlated with the availability of weapons. In turn, weapon availability is negatively correlated with prohibitive legislation (Santanella-Tenorio et al., 2016) and positively correlated with the recency of conflict in that society (Muggah, 2007). Weapon-carrying may also be driven by higher crime rates but the causal order of a weapon carrying-crime rate relationship is difficult to determine (Siegel et al., 2013). At the granular level of community, weapon-carrying is almost impossible to measure, limiting the potential to understand the relationship between community characteristics and weapon use, but factors such as deprivation, availability of services, low social cohesion and crime rates are predictive of weapon-related violence. Again, the causal relationship between these factors is difficult to determine.

Within groups, gang membership and the proportion of one’s peers being in trouble with the police is predictive of weapon-carrying (Brennan, 2019; Emmet et al., 2018). Longitudinal research has found the introduction of a weapon into a social group by one individual has contagious effects on weapon-carrying across the group (Djikstra et al., 2010). At the individual level, the correlates of weapon-carrying are being male, being in late adolescence and early adulthood, substance misuse, low trust in the police and involvement in bullying or violence as a victim or a perpetrator (Brennan, 2019). In particular, the interaction between low trust in the police and exposure to violence has a particularly acute effect on later weapon-carrying (Brennan, 2021). Accounts of motivations for weapon-carrying describe how weapon-carrying offers a way to mitigate or overcome fear of victimization (Traynor, 2017), that weapons are a valuable instrument for making criminal activity, such as violence and robbery, more successful (Brennan, 2017) and they contribute to the construction of an aggressive social identity (Harding, 2020; Harcourt, 2006) that can generate respect, status and protection (Palasinski, 2012).

Custodial violence and weapons

Violence is one of the topics that features most prominently in prison research and the most extreme types of violence are often characterised by weapon use (Bottoms, 1999; Davies, 1982). Despite their importance in causing serious harm, the distinct utility and meaning of weapon carrying and use in prison has received insufficient attention.

At first glance, a comparison of statistics derived from custodial (Ministry of Justice, 2023) and community settings (House of Commons Library, 2021) indicates that the proportion of violence involving weapons in prison is slightly higher than that in the community (24% vs 21%). The non-reporting biases inherent in each statistic mean that these proportions are not comparable. Furthermore, closer examination of the list of weapons recorded by Ministry of Justice (Blunt instrument, Chemical incapacitant, Dangerous liquid, Excrement/urine, Firearm, Food, Knife/blade, Ligature, Other sharp instrument, Spitting, Thrown equipment, Thrown furniture, Other and Biting) and that recorded by Crime Survey for England and Wales (Bottle, Drinking glass, Knife, Screwdriver/stabbing implement, Stick/club/hitting implement, Pistol or handgun, Shotgun or rifle, Airgun/air rifle, Gun - can't say what sort, Syringe, Stones/bricks/concrete, Axes/swords/cleavers, Dogs, Car/Van other vehicle or Other) reveals that what constitutes a ‘weapon’ differs across the two settings. As around 40% of ‘weapon use’ in prisons does not involve an ex-corporeal object (e.g. biting, spitting, throwing excrement or urine), the distinction between what is and is not a weapon remains open to debate. Whatever the reason for this inconsistency, it suggests that community and custodial weapon use are two rather different contexts. By extension, their use and meaning may also vary.

No study has attempted to model predictors of weapon-carrying or weapon use in prison, but some associations use have been identified in quantitative and qualitative studies of prison violence. For example, weapon use correlates negatively with age (McCorkle, 1992). This is likely to be moderated through the hypermasculine identity formation that is frequently observed in younger or newly incarcerated prisoners. In North America, weapon-related violence is also correlated with gang involvement (Pyrooz et al., 2016) although Ricciardielli’s (2014) Canadian interview respondents suggested that weapon use was more likely in prisoners who were not gang-affiliated. Given the differing focus of the studies (perpetration and victimisation, respectively), it is possible that the data relates to differing forms of weapon-related behaviour or intention.

In terms of motivations for weapon-carrying, the common theme is that, in prison, weapons are kept and used for self-defence. A survey of prisoners in a Tennessee maximum security prison found that one-quarter of prisoners kept a weapon nearby (McCorkle, 1992). A factor analysis of defensive behaviours captured in that survey found that ‘getting tough to avoid victimisation’, lifting weights to build a physically intimidating identity and keeping a weapon nearby clustered as a set of ‘aggressive precautionary’ behaviours designed to avoid victimisation (ibid.). Similarly, an interviewee of Rochelau (2011) described prisoners keeping a ‘just-in-case shank’ (p.232) and Ricciardelli (2016) described the extensive punishment that a prisoner accepted for not giving up his knife.

Several other researchers have observed that, beyond having weapons accessible for use when in need, maintaining a reputation for weapon use also served defensive purposes (Fleisher, 1989). Crewe (2009) described how weapon-carrying and use could be both symbolic and instrumental: a ‘tool merchant’ could use weapon-carrying to construct a violent identity not for the purposes of achieving dominance or status but to ward off threat. Again, weapons were a feature of aggressive precaution – to avoid violence – rather than to be successful in assaulting another prisoner.

At least with the US literature, prison characteristics and prison officer activity were predictive of weapon-related behaviour to some degree. Ricciardelli (2016) described how because the population of maximum security prison was more transient and, consequently, unpredictable, prisoners were more likely to arm themselves in these settings in anticipation of violence. Weapon-related behaviour increased when the social order of the prison was upset. Crouch and Marquart (1989) detailed how the banning of the ‘building tender’ system3 created a power vacuum that was rapidly filled by gangs and the most violent prisoners. Weapon-carrying was also alluded to when prison governance was weak. This might be because prison officers or systems were not competent in preventing violence generally (Bottoms, 1999) or because surveillance was poor in some areas of the prison (Lowman,1986; O’Donnell and Edgar, 1999).

In a parallel with accounts of weapon-carrying in the community, it is noteworthy that there are few, if any, accounts from prisoners about keeping or acquiring a weapon for overtly offensive purposes. This may reflect a code of silence about violent activity, discomfort among researchers about asking challenging questions or response bias where the more aggressive prisoners do not participate in surveys or interviews. Although prisoners often discuss how they will avoid violence when possible, it is unlikely motivations for weapon use in prison are unambiguously defensive.

Within these studies, weapons were typically presented as by-products of violence, inert objects used to inflict harm rather than something to understand on their own and there was no discussion of how weapon-carrying and use could be prevented in prisons. There appears to be no discussion of why certain weapons are chosen in different situations, nor is there a literature on how weapons per se become available in prisons. The discussion of accessibility of weapons is limited to noting that accessibility is a concern for prisoners with no insights into how prisoners maintain this access across the custodial environment. Despite the potential of these findings to be operationalised within a prison and to potentially reduce future violence, none offered any practical insight for reducing weapon use itself.

Theoretical framework: the four A’s of weapon selection

With so little examination of weapon-carrying in a prison setting on which to build, it has been necessary to import a model of weapon selection from the community. Fortunately, while weapons are certain to be used in different ways in custody and the community, there is considerable potential for generalising between the settings.

The preceding section has presented a comprehensive list of risk factors that sweeps across micro-, meso- and macro-level explanations for weapon-related behaviours. However, deterministic predictors such as these are often distally linked to the behaviour. While they are valuable for informing longer-term prevention efforts or for understanding causal relationships, they offer little insight for understanding issues such as why people choose certain weapons, how weapons are used and how weapon use can be prevented in the short-term.

Sidebottom et al. (2021) addressed this through the creation of a framework, coined the ‘Four A’s’, to explain how types of knives, specifically, are selected for use in violence in a community setting. They observed that the weapon types that are most prevalent in a community, kitchen knives, are disproportionately under-used in community violence than rarer, more prohibited knife types, such as machetes and combat knives. Adapting a model of access to health care (Penchansky & Thomas, 1981), the authors proposed that the decision to select a weapon type considers four factors: availability, attractiveness, affordability and accessibility. Availability relates to the prevalence of the weapon type within the community; attractiveness relates to the features of the weapon type that makes it attractive to users; affordability is the amount of tangible and intangible costs and risks to accessing that weapon type; and accessibility relates to the potential for accessing that weapon type for use. When, for example, availability is low, the affordability and accessibility are also low, but attractiveness may be heightened through its rarity or the extent to which it has been prohibited. A potential user might discount selecting this weapon type because the sourcing costs are too high or the punishment for being detected in possession of a highly prohibited weapon might be too great. If a weapon is highly attractive, perhaps because it signals dangerous intent, a user may seek to overcome the affordability costs or the accessibility effort. Attractiveness is resistant to manipulation, but in community settings, policy makers, police or other services may seek to manipulate any of the other three ‘levers’ in order to overcome the attractiveness of that weapon type. For example, banning the sale of a weapon may reduce its availability; the introduction of situational preventive measures could reduce its accessibility and an increase in punishment for possession of a particular weapon type would decrease its affordability.

The ‘Four A’s’ framework was developed for understanding knife selection in a community context but it has been applied effectively to inform the safe retail of knives and other weapons in shops and supermarkets. Although untested, its use may also be transferable to a custodial setting where the acquisition, creation and use of modified, unmodified and offensive per se weapons is a poorly understood issue.

Research Questions

This lack of weapons research in prisons, coupled with the limited research into weapon carrier motivations, demonstrates a literature gap with important implications for prisoner and prison staff safety. With no specific framework explaining custodial weapon-related decisions and behaviour, the 4As framework offers a structured opportunity to examine this phenomenon.

Using this framework, we derived the following questions to learn more about weapon selection in prisons:

  • What factors and process(es) influence(s) a prisoner to make, carry, or use a weapon? (attractiveness)

  • How do prisoners make decisions about their weapon selection and/or production? (availability and accessibility)

  • What are the costs to individuals, and others, associated with carrying and using/not carrying or using a weapon? (affordability)

An over-arching deductive question was:

  • To what extent is the 4As framework generalisable to and suitable for understanding weapon use in a prison setting?



The study employed a cross-sectional, qualitative, research design.


The research was conducted in a single category B adult male prison in a large English city with a capacity of over 1,000 prisoners. Eligibility criteria were having been adjudicated4 once or more for weapon carrying or being the subject of a prison security intelligence report regarding weapon behaviour. An initial eligible population of 17 prisoners were identified (1.5% of the prison population), reduced to 11 due to imminent release, risk to females (interviews were undertaken by a female researcher who worked as a forensic psychologist in training in the prison) or having a current working relationship with the researcher. Nine prisoners were approached initially: of these, three declined but six (35% of the eligible population) agreed. This number was enough to achieve data saturation and the remaining three prisoners were not approached. Respondents ranged in age from 18 to 44 years (median = 25 years, interquartile range: 4.25 years). Five of the six were remand prisoners and one was on recall. All were of different ethnicities: one was white, two were black, two were Asian and one was another ethnicity. All had been charged with, although not necessary convicted of, some form of violent offence. Their average time in the prison was 6 months, which is typical for a remand prison.


A semi-structured, interview schedule was developed based on the 4A’s framework (Sidebottom et al., 2021).


Ethical approval was obtained from the prison’s research ethics committee. Potential respondents were approached and taken to a private space away from their wing location where the research was discussed and they were offered to option of taking part. This ensured that others were not aware of the subject of the conversation. They were given information about the research and offered the option of taking away a sanitised information sheet (containing no reference to weapons) to read and consider participating. They were given the option of 48 hours to consider participation after which an interview time was agreed with those who wished to take part. Participants completed semi-structured interviews, at pre-agreed times within the prison’s operational regime. Rooms were selected so the participants could speak candidly without fear of being overheard. Interviews lasted between 33 and 64 minutes (M = 48.5, SD = 21.9) and were conducted over a three-week period between October and November 2022. The interviews were recorded and transcribed by the first author. The interview questions approximately followed the schedule mentioned above and in their responses, prisoners were not limited to their experience in the current prison.

Analytic strategy

To investigate emerging themes, the researcher coded the data using Thematic Analysis (Braun & Clarke, 2021). Preliminary patterns between responses were identified and coded before classification and development into emergent themes. The relationship between each theme was examined through re-coding. Finally, they were related back to the research questions. Due to the deductive design, the data and developed themes were considered within the context of the 4As framework.

Analysis and discussion

Three superordinate themes emerged from the data: ‘Prison as war’, ‘Weapon use as identity management’ and ‘Imbalanced costs and benefits’. They are described and interpreted below with reference to the literature.

Prison as war

In all interviews prison was likened to the experience of being at war, i.e. in a state of near-constant threat and opportunity for violence. Attractiveness, availability and accessibility were linked within a war-like construction of the prison environment. Because availability of weapons per se was highly restricted, individuals demonstrated skill and ingenuity in creating novel and highly lethal weapons from everyday prison items. The attractiveness of particular weapon types was largely based on their potential to do harm although attractiveness depended on context. Participants noted how the constant threat of violence impacted on the need for weapons to have high accessibility, which required careful strategising and linked to skills in the manufacture of makeshift weapons.

Offensive and defensive strategies

For all participants, weapons served dual functions: causing harm and protection. P3 described this as, “they’re at risk or...wanna cause risk”. Weapons were used “for [their] own protection” (P2), while also being “something that can cause damage to a person” (P6). Each participant referred to how these functions were not mutually exclusive and could serve both purposes, sometimes in the same incident.

Research on weapon-carrying in prison has tended to focus on weapon carrying and use as precautionary (McCorkle, 1992) but this sample in custody reported a more balanced set of motives – both offensive and defensive – and were more forthcoming with descriptions of violent planning and intentions. This observation echoes the multi-functionality of weapons in the community to which Harcourt (2006) and Brennan (2017; 2021) have referred.

Harm and control

The common denominator of what made a weapon type attractive was potential to cause serious injury. At the top of the hierarchy of attractiveness appeared to be offensive weapons per se (‘road knives’; P1) but their availability was close to zero in the prison. Consequently, weapons were created or modified from everyday objects that were able to inflict varying levels of harm. While sharp objects were the most frequently discussed, ‘kettling’5 was described as one of the worst forms of assault. Although the respondent did not elucidate, presumably this was because of the potential for disfiguring injury as well as affecting a wide area of the body.

There was a limit to which some weapon types were attractive when they fell below a threshold of potential harm. Potting6 was an example of this: “it’s not really...harmful” (P5), and “like it can fuck you up so I would understand if people say it’s a weapon but I wouldn’t class it as a weapon” (P6). However, the limited potential for serious harm, coupled with the associated humiliation, made potting an attractive weapon type when assaulting a staff member: “yeah people will just spit on a gov or throw piss on a gov, I don’t really see some people going to use weapons on the govs” (P6), although P1’s account of an officer being ‘kettled’ contradicts this. This indicates that weapon selection in prison was dependent on context and victim/adversary type with the potential to control the amount of possible harm being a consideration alongside harm potential.

An additional consequence of being ‘at war’ was the need to always have a weapon within easy access: “I grab my weapon before I brush my teeth” (P1), “just always having a guard up” (P5), “you don’t wanna be the one where they’ve got the weapon and you don’t” (P3). P1 noted not having a weapon, “…feels like you’re naked like you’re not shielded” while, with it present, “you feel’ve got a shield now”. Brennan (2021) hypothesised that, rather than defining the motive for weapon-carrying as either offensive or defensive, ‘managing uncertainty’ is a more efficient and realistic way to capture the dynamic value of a weapon in a hyperviolent environment.

Maintaining access to a weapon required strategic planning. When the interviewer asked if weapons were stashed in communal areas of the prison wing, P6 said this was unlikely because they would be difficult to hide and could be used against them: “some people are not gunna want it around the wing…not gunna want other people to know” (P6) but P3 and P5 suggested that weapons being stashed in communal areas happened occasionally and many prisoners knew it was in their best interests to ignore its presence for fear of repercussions: “I’ve seen it happen like someone’s just seen n just ah yeah you know there’s a thing there just like, know what I mean like it’s not mine, I’m not involved” (P5). This observation echoes an account from Ricciardelli’s respondent who described holding weapons for other prisoners and accepting that any punishment they incurred was just a cost of staying safe in prison.

The relationship between prisoners, their environment and weapons

In this sample, weapon-related decision-making and behaviour was strongly affected by their environment. Firstly, danger was pervasive and they were highly sensitive to the threat of violence: “it’s a dog-eat-dog world” (P1), “certain have to roll with it 24s” (P3), “the prisons aren’t safe enough…. a very, very dangerous place...nowadays its bad” (P2). Accordingly, they rationalised weapon-carrying as a necessary act of self-preservation in a hyperviolent place.

These observations echoed those of Ricciardelli’s respondents who had been in the Canadian prison system where maximum security prisons that held prisoners before transfer to the ‘home’ prison were more risky and required a higher level of vigilance and, by extension, weapon-carrying. In addition to the place-based distribution of risk, veteran prisoners in our sample described how risk in prisons has changed over time. Unfortunately, changes in the measurement of weapon use in prison violence from 2018 onwards prevents this from being verified, but the pre-change years did show increases in the prevalence of weapon use from 2012 to 2017. The proportion of assault incidents involving a weapon have remained stable – between 22% and 26% - since the new measures were introduced, suggesting that weapon use prevalence has not changed disproportionately over this short period. However, these figures are too aggregated and crude to dispute a claim that prisons have become more dangerous.

Some participants reported a lack of trust in staff ability to prevent serious violence, justifying the carrying and using weapons. For these respondents, staff were not viewed as protectors but as uninterested or incapable guardians: “I’ve seen officers see someone get stabbed up in cell and just walk past…They don’t want the aggro”7 (P2), “remember these govs are civilians...they just come to work and go home” (P1). This also mirrors observations from the US prison literature where respondents relied on prison officers to be competent so that the social order could be maintained and violent conflict avoided. When the social order became unbalanced, power struggles involving serious violence occurred before equilibrium was restored. This observation echoes reasoning by weapon-carriers in high-violence communities (Harcourt, 2006; Palianksi, 2012) where police are demonstrably incompetent or unwilling to prevent violence. For those respondent, individual safety becoming a personal rather than a state responsibility justified weapon-carrying and use.

Mirroring generals planning battles based on terrain, weapon selection depended on where it would be used. According to P1, “it depends on the situation where its gunna be a cell, it’s a confined space so there’s less ability agility, agility to activate your weapon d’ya get what I mean, off wing, its perfect because there’s so much space”. For P2, the availability of weapon types differed depending on the prison, “in jails where…you can cook…I’ve seen people pour fat in pots, they put plastic bags in…melt them down…chuck them in people faces”.

In a departure from or an advance on the 4As framework, the respondents in custody indicated that the attractiveness of a weapon type shifted depending on where they planned to use it. The available space affected how suitable a weapon type was and the accessibility of different weapon types changed by location in a way that has not been observed and is unlikely to be mirrored in a community setting. While accessibility and attractiveness are somewhat crystallised in the 4A’s framework, the restrictions on availability and accessibility of weapon types in different areas of a prison suggest there is a more fluid relationship between these domains than in community contexts.

Skilled weapon use

With low availability of weapons per se, weapon manufacture and use were viewed as skilled pursuits. This theme contained two features. Firstly, the respondents noted how any motivated individual could overcome challenges in availability by adapting items to become weapons. Nonetheless, there was skill in making more novel and lethal weapons and these features contributed to their attractiveness. Second, once in possession of a weapon, individuals had to be creative about their how they stored and used it.

“Guys are creative in jail innit”, according to P4. For P1, “anyone can make a weapon, it’s just how you make it”. P2 believed that, “you gotta have someone who can make…a better weapon.” ‘Better’ in this context related to accessibility: “making it small enough so you can hide it” (P2) and attractiveness in the form of originality or creativity: “I might hear or see raaaa that’s crazy that’s a mad weapon...that one’s for me” (P3). While respondents noted that some individuals in prison were skilled at creating makeshift weapons, most prisoners could make something effective and this was made easier by long times spent locked in cells: “it’s pretty much easy in a way to make one, especially when you’re on the, in a cell for pretty much most of the day so like you’re gunna think, you’re gunna start thinking about how can I do that, how can I do this” (P5).

Beyond skill in making and concealing weapons, there was skills in their use. Decision-making the more skilled weapon users apart from others: “For a normal person it’s very quick...but for someone that rolls around with [weapons] it’s slow” (P3). This informed planning, “this one suits me, I like that, I know how to use it right” (P3) and tactics, “you can’t get close to me” (P1). The participants described how they would adapt their chosen weapon based on their own knowledge and their tactical approach within a certain context. For example, a plug on its wire would allow for speed and prevent someone getting close but this would only be selected if a weapon carrier knew how to use it: “how good they are with that weapon, literally, if you feel like you’re better with a plug, swinging a plug from a distance you’re better, if you feel like you’re better with a blade on a toothbrush up close you’re better, or it depends on the situation where its gunna be innit…in a cell or in the open” (P3).

Although Harcourt (2006) discussed the function of guns and their attractiveness, he did not mention the notion of skill and ability beyond having respect for firearms. This could be because participants in Harcourt’s study did not see using a gun as a skilled pursuit in the same way that the participants in the present study viewed using an improvised weapon at close quarters. Nevertheless, the present findings suggest individuals chose weapons based on what was present in the environment at the time of need (availability) and what they could retrieve (accessibility) converging with what best serves their purposes (attractiveness). This supports the assertion (Sidebottom et al., 2021) that the 4As intersect and inform each other rather than being four distinct domains.

Linking to previous knowledge about custodial violence, theories have explored the prison environment and its impact on those it imprisons (The Deprivation Model; Sykes, 1958; Goffman, 1961) as well as the characteristics of the imprisoned (The Importation Model; Irwin & Cressey, 1962). The superordinate theme of ‘war’ which was identified in this study incorporates both these aspects. The participants explored the idea that their environment was a dangerous place that required them to protect themselves. This adds evidence to the idea that prison culture contributes to violence (Colvin, 1992). Yet, as Dilulio (1987) stressed, there seemed to be more factors involved than just the prison environment. This could be an interaction between a perception of the environment as dangerous and the feelings of the weapon carriers which they ‘imported’ into the prison. There was a sense of feeling that something was missing “…feels like you’re naked like you’re not shielded” (P1) when they were not carrying a weapon. Payne (2015) refers to the emotionality of war and perhaps the feelings of the weapon carriers interacted with their perceived skills and the environment and terrain to inform decisions about the need to carry a weapon in custody.

Weapon use as identity management

The attractiveness domain dominated this superordinate theme. Again, the idea that weapons are multi-functional was present. Weapons offered the participants a symbol through which to define themselves both as individuals and in the way they navigated a social world.

Constructing identity through weapon selection

This theme was concerned with what weapon type said about an individual: “You can tell a person by the weapon they use” (P3). P1 saw weapons as a symbol which would help them put out a certain persona and help them to navigate the prison environment: “You come to jail and present yourself like...‘I’m a hard boy’” and “You can tell a person by how they use it [weapon] all-in their heart is” (P3). P1 reported, “I got it so if someone wants to try play with me…I’m not gunna have it”, and P3 explained that “everyone has their own weapon” exploring the idea that a weapon held individual significance for its carrier and that it represented them. There appeared to be a relationship between the type of weapon and its potential as a precautionary device. P1 noted how a knife – in this case, a weapon per se – has more protective value than other types of weapon, although the knife alone would not protect a person: “if you stab someone with a knife obviously then people are gonna know aight cool hes got a knife so we’re not going to play with him, we’re not going to test him but there’s other guys that might have a knife and think raa we’re still going to test you”.

Participants also spoke of rejecting some weapon types due to what they represent. Potting – a uniquely prison-based form of armed violence – was an example of this, that carried a stigma and was seen by P5 as unacceptable: “you’re just being a scumbag” (P5). Additionally, an older respondent rejected the idea that a weapon was valuable in constructing identity or, at least with hindsight, that the connection was a mistake: “It don’t really make you feel better about yourself” (P6), and “You ain’t got nothing to prove, you just have to be yourself” (P1).

Some respondents also noted how not using a weapon in violence when the other combatant is armed can demonstrate ‘heart’ and build a hypermasculine identity: “obviously there’s guys that’s had weapons and got knocked out by a guy that’s just got his fists but that will just show, right he’s not one to play with, his fists are his weapons” (P1). This finding echoes one alluded to by a minority of respondents in Harcourt’s interviews (2006) who also suggested that not using weapons or being armed in highly violent contexts was a display of courage and strength.

Brennan (2018) posited that weapons can act as a means by which a person expresses their identity. He references Harcourt (2006) whose participants linked the dangerousness of their guns to how others perceived their power and masculinity. The present study’s findings reinforce the position that weapons provide a means whereby the bearer can project and shape an identity. In a verification that weapons do shape public identity, respondents described how the weapon choices of others gave them information about the dangerousness, vulnerability and ‘code’ of that person.

Weapons and group membership

Weapon use and weapon selection served in some way to strengthen group membership. This subtheme was heavily dominated by the attractiveness domain. According to P4, “Everyone just wants to get involved [in weapon violence], just to say that they did”. P1 echoed this, “Everybody wants to be a part of something”. This can offer a sense of protection, “You’re vulnerable...there’s a group of’ll try join them” (P5), and offer topics to talk about after the fact, “like fighting and stuff, talking about it...just to have something to say…” (P4). Being part of the group, however, meant following group norms, “some people just want [a weapon] to fit in” (P6). They get respect, “respect goes a long way” (P1) and establish control over others, “that gang, they’ll make sure that people know…they run that wing…that prison” (P2). There was a sense of pressure to adhere to the group norms, which involved weapon use: “in jail you have to fight” (P6). These observations are consistent with US prison research on how members of affiliated groups must fight to protect or avenge other group members or suffer consequences.

As illustrated in the themes presented above, weapon selection, carrying and use was used by this sample to create individual identity and to forge links with others. These observations are not novel in community contexts. Harding (2020) explored the way using knives acted as a source of “street capital” (p. 46) which could also discharge tensions and elevate the carrier’s social standing. Competition in this social space required a level of authenticity to prove themselves and one way this achieved is through knife carrying (Harding, 2020). Harcourt (2006) explored the notion that weapons specifically offer both a sense of belonging but also requires adherence to group norms. His participants viewed weapons as something important shared amongst the group. Adhering to that group norm indicated trustworthiness and facilitated in-group access (Harcourt, 2006). In a prison setting, the same processes are evident, but may be accentuated by the near-constant threat of violence and the need to form alliances rapidly and to maintain personal and group identities that are seen to offer protection. For these reasons, the domains of availability, accessibility, affordability and attractiveness are all present in a prison setting, but the way in which they interact, such as weapon accessibility being a priority over attractiveness, may be different from those ways in the community.

Imbalanced costs and benefits

The final superordinate theme was dominated by affordability. Respondents alluded to a striking a balance between the costs and benefits weapons offer their carrier in a way that was often rational. The formal costs – adjudications and punishments for weapon possession – were viewed as insignificant compared to the benefits (or avoided costs) of being armed when the need arose.

The balancing act

Participants seemed to analyse the costs and benefits of weapon carrying and use and in one case, quite literally: “it’s like a scales” (P1). Benefits of having access to weapons included its equalising effect in confrontations: “it’s fair like that” (P3). There were also seductive and hedonic effects: “I won’t say it’s more exciting to talk about, but it kinda is” (P4), “I’d say more thrill, cause you’re gunna get caught” (P5). For others, protection was the benefit offered: “the only reward is that you are safe” (P3). The costs of accessibility was appraised. These could be financial: “You can buy weapons8 on the canteen sheet” (P1) and you can pay for one: “you can buy ready-made ones innit…so to get a knife made like obviously a tuna can or a mackerel can they make it in to, to get that made they’d charge you like £50” (P1). One respondent described a market in weapons but that these commodities were very cheap. Referring to a previous prison, P4 stated, “I see guys selling knives for canteen n that, minor things innit just couple packs of noodles or something” (P4).

The benefits the participants referred to corresponded to those observed in the literature. Brennan et al. (2010) referred to the higher levels of sensation seeking behaviour in weapon carriers and Harcourt (2006) noted that young gun carriers gained positive feelings, fun and power all of which are seductive. While these were alluded to in the interviews, they were not the dominant features of the informal cost-benefit analysis undertaken by the respondents. Moreso, the benefit of weapons was in how they offered a form of protection and equalised fights (Marfleet, 2008).

Most respondents discussed how there was little meaningful consequence of being founded in possession of a weapon: “lose your TV innit, go on basic, that’s all…” (P4). Even when caught using a weapon, the benefits outweighed the costs: “an outside’re just going to do a couple more years in jail but you’re still alive” (P1); “reputation…weighs up more than added days, nickings [prison adjudication]” (P2).

The reflections from participants, which disregarded the formal consequences offered by prison authorities, mirror Palasinski’s (2013) findings of weapon-carriers in the community: the young men who took part in his study accepted that there would be legal ramifications but chose to engage in weapon carrying regardless. They also mirror observations from the prison literature where the value of consequences, such as time in isolation or additional charges, were far outweighed by the benefits to safety and respect and prisoners were willing to accept these consequences in order to maintain a code of silence, to retain access to a weapon, to avoid disrespect or serious injury (Ricciardelli, 2016).

Unavoidable weapon use

The other side of the cost and benefits theme was that the decision to carry a weapon was not one taken freely: “People don’t wanna be the way they turn out…. It’s sad that people have to carry weapons to feel safe in prisons” (P2). Additionally, those who want to do things differently, cannot, “They come to jail...wanna do this humble ting, just to ride the bird but it doesn’t last” (P1).

In an echo of the mutual assured destruction of arms races, one respondent suggested that most people would prefer to fight without weapons because of the harm they cause and because the potential legal implications for the user were substantial. However, the fear that others might have a weapon made weapon-carrying a necessary cost: “if I thought in my head d’you know what nahh I’m gunna fuck you up like, fuck that, I don’t need it, I don’t need a shank, why would I catch another stabbing charge…but in the back of your head, you’re thinking…what bout if he’s got a shank, I’d rather just shank him rather, its what, its like you’d rather kill than get killed” (P5).

This trapped feeling seemed to dissipate with age. Weapon carrying and use was seen as a young person’s activity: “a youth offender…They’re more gassed, more hot headed but…an older person yeah 100%...they think about it” (P5). Youth clouded judgement, “guys are just young, they’re not thinking straight” (P4), whereas, “older people will calm it down” (P4). Furthermore, for P2, “when I look back, there’s no just causes misery” and consequences are clear, “I done [more than ten] years in prison, seeing the aftermath that it causes for other families...I wouldn’t want anyone to go through that” (P2). Over time, the costs became more pivotal and weapons less attractive.

That offending lifestyles can trap people is not a novel concept (Holligan et al., 2017; Reid, 2023). Harding (2020) notes a pressure experienced by gang members causing them to feel “trapped in the system” (p. 41). Multiple prison researchers have described how prisoners would avoid violence if they could but were trapped by the violent social order and where the consequence of seeking to avoid violence was often more violence (Bottoms, 1999; Crewe, 2009).


The use of a weapon in violence is a strong predictor of serious harm both in community and prison settings. As a feature of community violence, weapon use has received a considerable amount of attention, but by comparison, the scale of research on weapon use in prison is small and underdeveloped. Through the accounts of a sample of experienced weapon carriers and users recruited from a single prison in England, but who had spent time in a range of prisons in England, we described how, for this sample, weapons were used to construct identity and to avoid or engage in serious violence. The accounts also detailed how factors like motivations, group membership, and prison environment shaped weapon-related decisions. Using a framework for understanding weapon use in community settings, we also described how factors such as attractiveness, availability, accessibility, and affordability interacted in the selection, use and understanding of different weapon types in prison.

In prisons, as in community settings, weapons were not wholly offensive or defensive. Their purpose was often dynamic, shifting across context and a weapon could occupy offensive and defensive positions at the same time. Rather than classifying motivations for weapon use as offensive or defensive, a more accommodating explanation for weapons in prison is that they help to manage the uncertainty and precarity of a hyperviolent milieu (Brennan, 2021). The sample were recruited based on their experience of a relatively extreme form of violence and presented prison environments that were more violent and risk-laden than those described in other prison research (Bottoms, 1999; Crewe, 2012). This is epitomised by the superordinate theme of ‘prison as war’ wherein prisoners made judgements about weapon selection rooted in the harm the weapon could do and the need to always be prepared for violence. Accordingly, with low availability of weapon types limiting choice, the attractiveness of a particular weapon type was often less salient than its accessibility. The lower priority of attractiveness also meant that the costs of sourcing was reduced. Although weapons per se could be sourced (although not necessarily in this prison) or higher quality manufactured weapons could be purchased from skilled makers, most prisoners had the time, privacy, and resources to make their own weapons with low costs items acquired legitimately in the prison environment. Referring to types of low-status kitchen knives in the community, Brennan (2017) questioned whether the attractiveness of available weapons would ever become so low that this would prevent a weapon from being used. In prison, as long as a weapon has the potential to do harm, a lower threshold of attractiveness does not seem to exist.

The violent ‘social order’ of the prison described by these respondents is consistent with those described in the prison literature. In general, prisoners would prefer to avoid violence (Bottoms, 1999; Crewe, 2012) and this perspective increased with age (McCorkle, 1992; Ricciardelli, 2016) However, through mechanisms such as aggressive precaution against victimisation or group behavioural norms, involvement in violence was unavoidable for some. Prisoners who felt that violence was unavoidable also rationalised that weapon-carrying and use was necessary because the potential costs to personal safety – both proximal in terms of injury and distal in terms of showing vulnerability – of being unarmed far outweighed the potential costs of punishment by prison authorities or the moral costs of doing someone serious harm.

The study demonstrates some generalisability between community and prison use of weapons. For example, individuals employed similar processes to select available weapons by balancing affordability, attractiveness and accessibility. However, the hyperviolent nature of prison described by the respondents, increased surveillance, and highly limited availability of weapons per se mean that these domains interacted in different ways to those proposed by Sidebottom et al (2021). In the community, weapon availability is in part driven by being able to source a particular weapon which is suited to the context for which it is planned (Brennan, 2017). As weapons per se are generally unavailable in custody, the decision-making regarding availability turned more to the materials that could be used. Yet, this did not present much of a challenge. Weapons or the materials for making a weapon could be purchased from the prison's canteen list. That prisons have a high degree of control over what is available via canteen demonstrates how easily available weapons are for someone motivated to acquire one. Secondly, attractiveness also interacted with the location where the weapon would be used. Some weapons were more attractive for use in a cell compared to use on a prison wing landing. This created a potential limit on how successful an attack might be (Logan et al., 2021) and an interaction between weapon availability and attractiveness. How far the context of violence is its own distinct domain or whether it is a subset of attractiveness requires further examination. These exceptions aside, the 4A’s framework appears applicable to a custodial setting.

To summarise the preceding paragraphs in simpler terms: when a prisoner’s world was one of frequent and serious violence, access to weapons was necessary to manage the uncertainty of this environment. As long as the potential harm could be controlled and was suited to the space in which it would be used, prisoners were not particular about the weapons they chose. While prison staff, when competent, facilitated safety through their presence, in the absence of capable guardianship, there was little deterrent effect of prison rules on weapon-carrying and use. There also was little long-term benefit of confiscating weapons as their creation cost so little. The implications of this for prison safety and governance are clear. Just as in the community, if the prison cannot create a safe, predictable living space, weapons will be an inevitable feature of the general prison environment. The inverse of this is that, while programmes of confiscation or weapon amnesties may do little to reduce the availability of weapons in a prison, data on the recovery of weapons – their type and number – may provide value as an early warning of changes in the balanced social order of a prison environment and a systematic method for observing and interpreting weapon-related trends could be established at the wing or unit level.


Although the sample were one-third of the eligible population and were a rare group of ‘experts by experience’, the accounts of six prisoners from one local remand prison cannot generalise to other prisoners and prison settings. We employed in vivo coding as participants appeared to speak candidly, offering deep and novel insight into this complex phenomenon. However, the sensitive subject matter, identity of the interviewer as a member of staff and prisoner ‘code of silence’ may have precluded the participants from being completely open with their responses. An additional and notable limitation is that the sample was restricted to those who were known to the prison security department. Prisoners whose weapon-related behaviour was unknown to the prison may have provided alternative perspectives which could have changed the findings. We used a framework to guide our questions. While this appears to have been a successful way to capture the meaning and utility of weapons in this sample, its focus on decision-making limited the scope of our inquiry. For example, although gangs or post code affiliation was a significant security issue in the prison, we learned little about how gang involvement or the importation of violence from the community influenced weapon-related behaviours and because the availability of weapons per se was rare, we learned little about how these entered prison. Finally, as noted earlier, because the framework was derived from community settings, we also learned little about the meaning and utility of weapon use in different prison spaces. These represent valuable areas for future exploration.


Overall, this study develops the literature on weapon decision-making, extending it into a novel setting and addresses a significant gap in the prison research literature about the meaning and utility of weapons in a custodial setting. Respondents described a hyperviolent milieu for some in which access to weapons was essential and wherein the official consequences of weapon-carrying were outweighed by the potential costs of victimisation. Weapons served a variety of purposes for prisoners. At the individual level, they reduced the uncertainty of a hyperviolent environment and they were used to construct and manage a violent identity as an aggressive precaution against victimisation. The use of a community-derived framework for understanding weapon-carrying translated well into a prison environment and offers support for the synthesis of community and prison models of violence.


1 Weapon type categories were Knife/ Stabbing implement, Hitting implement, Glass/bottle, Firearm, Stones and Other.

2 As with violence in the community, this is likely to be a gross underestimate of the total count of violence in prison (Schenk and Fremouw, 2012). Also, as weapon-involved incidents are more likely to be reported (Brennan, 2011), the stated proportion of prison violence that involves weapons is likely to be an overestimate.

3 Also known as a ‘trusty system’, some prisoners operated as enforcers of prison rules in exchange for privileges. This enforcement was often violent and was banned in 1980.

4 A formal disciplinary procedure

5 The act of throwing boiling water over a person. This may include the addition to sugar to cause more lasting burns by facilitating a higher boiling point (Mawhinney et al., 2022); also known as ‘napalming’.

6 The act of throwing faeces and urine over someone; also known as ‘gassing’ (Shalev, 2009) or ‘chunking’ (Mears, 2006). It is unlikely that this list is exhaustive.

7 This observation was not verified

8 Not literally buying weapons per se, but items such as razor blades and tin cans that can be easily adapted or used as weapons.


Bottoms, A.E., 1999. Interpersonal violence and social order in prisons. In: M. Tonry and J. Petersilia, eds. 1999. Prisons: Crime and Justice – A Review of Research, vol. 29. London: University of Chicago Press.

Braun, V. and Clarke, V., 2021. Thematic analysis. In: H. Cooper et al., eds. 2021. APA handbook of research methods in psychology, Vol. 2. Research designs: Quantitative, qualitative, neuropsychological, and biological. Washington, DC: American Psychological Association, pp.57–71.

Brennan, I.R., 2011. In vino silentium? Individual, situational, and alcohol-related factors in reporting violence to the police. Violence and Victims, 26(2), pp.191-207.

Brennan, I.R. and Moore, S.C., 2009. Weapons and violence: A review of theory and research. Aggression and Violent Behavior, 14(3), pp.215-225.

Brennan, I.R., Moore, S.C. and Shepherd, J.P., 2006. Non-firearm weapon use and injury severity: priorities for prevention. Injury Prevention, 12, pp.395–399.

Brennan, I.R., Moore, S.C. and Shepherd, J.P., 2010. Aggression and attitudes to time and risk in weapon-using violent offenders. Psychiatry Research, 178(3), pp.536-539.

Brennan, I.R., 2017. High stakes: The role of weapons in offender decision-making. In: W. Bernasco, H. Elffers and J.-L. van Gelder, eds. 2017. Oxford Handbook of Offender Decision Making. Oxford: Oxford University Press, pp.421–444.

Brennan, I., 2018. Understanding weapon violence in the UK: Thirteen unanswered questions. [online] Available at: [Accessed 19 January 2024].

Brennan, I.R., 2019. Weapon-carrying and the reduction of violent harm. The British Journal of Criminology, 59(3), pp.571-593.

Brennan, I.R., 2021. Violence, worry and trust in emergence of weapon-carrying. European Journal of Criminology. [online] Available at: [Accessed 19 January 2024].

Bronfenbrenner, U., 1979. The ecology of human development: Experiments by nature and design. Cambridge, MA: Harvard University Press.

Bullock, K. et al., 2023. Police practitioner views on the challenges of analysing and responding to knife crime. Crime Science, 12(1), pp.1-11.

Colvin, M., 1992. The penitentiary in crisis: From accommodation to riot in New Mexico. Albany, NY: SUNY Press.

Crewe, B., 2009. The Prisoner Society: Power, Adaptation, and Social Life in an English Prison. Oxford: Oxford University Press.

Crouch, B.M. and Marquart, J.W., 1989. An appeal to justice: Litigated reform of Texas prisons. Austin, TX: University of Texas Press.

Davies, P., 1982. Violence in prisons. In: P. Feldman, ed. 1982. Developments in Study of Criminal Behaviour, vol. 2. Chichester: Wiley.

Dijkstra, J.K. et al., 2010. Influence and selection processes in weapon carrying during adolescence: The roles of status, aggression, and vulnerability. Criminology: An Interdisciplinary Journal, 48(1), pp.187–220.

Dilulio, J., 1987. Governing prisons. New York: Free Press.

Fleisher, M., 1989. Warehousing violence. Newbury Park, CA: Sage.

Goffman, E., 1961. Encounters: Two studies in the sociology of interaction. Indianapolis, IN: Bobbs-Merrill.

Harcourt, B.E., 2006. Language of the gun: Youth, crime, and public policy. Chicago, IL: University of Chicago Press.

Harding, S., 2020. Getting to the point? Reframing narratives on knife crime. Youth Justice, 20(1-2), pp.31-49.

Holligan, C., McLean, R. and Deuchar, R., 2017. Weapon-carrying among young men in Glasgow: street scripts and signals in uncertain social spaces. Critical Criminology, 25, pp.137-151.

House of Commons Library, 2023. Knife crime statistics: England and Wales. [online] Available at: [Accessed 19 January 2024].

Irwin, J. and Cressey, D.R., 1962. Thieves, convicts and the inmate culture. Social Problems, 10, pp.142-155.

Lincoln, J.M. et al., 2006. Inmate-made weapons in prison facilities: assessing the injury risk. Injury Prevention, 12(3), pp.195-198.

Logan, M.K. et al., 2021. Constraints to malevolent innovation in terrorist attacks. Psychology of Aesthetics, Creativity, and the Arts, 17(1), pp.91–105.

Löffler, D., McGraw, J.J. and Johannsen, N.N., 2019. Weapons in and as history: on the ontogenerative function of materialized preemption and intelligence in weapons technology. Identities: Journal for Politics, Gender and Culture, 16(1-2), pp.68-77.

Lowman, J., 1986. Images of discipline in prisons. In: N. Boyd, ed. 1986. The Social Dimensions of Law. Englewood Cliffs, NJ: Prentice-Hall.

Marfleet, N., 2008. Why carry a weapon?: A study of knife crime amongst 15-17-year-old males in London. London: Howard League for Penal Reform.

Mawhinney, J.A. et al., 2022. Does ‘Prison Napalm’ work? Measuring the cooling temperature of sugar solution burns in a porcine model. Burns, 48(5), pp.1209-1212.

Mears, D., 2006. Evaluating the effectiveness of supermax prisons. Washington, D.C: Urban Institute. [online] Available at: [Accessed 19 January 2024].

Ministry of Justice, 2023. Assaults in prison custody 2000 to 2022. [online] Available at: [Accessed 19 January 2024].

Muggah, R., 2006. Emerging from the shadow of war: A critical perspective on DDR and weapons reduction in the post-conflict period. Contemporary Security Policy, 27:1, pp.190-205.

O'Donnell, I. and Edgar, K., 1999. Fear in prison. The Prison Journal, 79(1), pp.90-99.

Office For National Statistics, 2023a. Crime in England and Wales: year ending December 2022. [online] Available at: [Accessed 19 January 2024].

Office For National Statistics, 2023b. Homicide in England and Wales: year ending March 2022. [online] Available at: [Accessed 19 January 2024].

Palasinski, M., 2013. Security, respect and culture in British teenagers' discourses of knife‐carrying. Safer Communities, 12(2), pp.71-78.

Penchansky, R. and Thomas, J.W., 1981. The concept of access: definition and relationship to consumer satisfaction. Medical Care, pp.127-140.

Pyrooz, D.C. et al., 2016. Taking Stock of the Relationship Between Gang Membership and Offending: A Meta-Analysis. Criminal Justice and Behavior, 43(3), pp.365-397.

Reid, E., 2023. ‘Trap Life’: The psychosocial underpinnings of street crime in inner-city London. The British Journal of Criminology, 63(1), pp.168-183.

Ricciardelli, R., 2014. Coping Strategies: Investigating How Male Prisoners Manage the Threat of Victimization in Federal Prisons. The Prison Journal, 94(4), pp.411-434.

Rocheleau, A.M., 2015. Ways of coping and involvement in prison violence. International Journal of Offender Therapy and Comparative Criminology, 59(4), pp.359-383.

Shalev, S., 2009. Supermax: Controlling Risk Through Solitary Confinement. Cullompton: Willan.

Sidebottom, A., Brennan, I., Agar, I., Ashby, M., Bullock, K., Hales, G. and Tilley, N., 2021. Knife Crime: A Problem Solving Guide. Ryton: College of Policing.

Siegel, M., Ross, C.S. and King, C., 2013. The relationship between gun ownership and firearm homicide rates in the United States, 1981–2010. American Journal of Public Health, 103, pp.2098-2105. Available at:

Traynor, P. R., 2016. Closing the ‘security gap’: Young people, ‘street life’ and knife crime. Doctoral dissertation. University of Leeds.

United Nations Office on Drugs and Crime, 2019. Global Study on Homicide. United Nations.

No comments here
Why not start the discussion?