Security from and at the Margins

Over the last decade anthropology has struggled to find an operative definition for the concept of security, either explicitly assuming a versatility that accounts for the diversity of actors and experiences it convokes, or through formulations that foreground the state or agents with power of decision and intervention as its key players. Yet, as Daniel Goldstein's work pointed out more than a decade ago, “for the people and societies that anthropologists study [. . .] issues of security and insecurity are critical matters with which ethnographic subjects must contend as they attempt to forge a life in a complex, conflictive, and often violent and dangerous social and political-economic milieu” (2010: 489).

I agree with Halbmayer and Naucke's (in this issue) claim that a large part of the anthropology of security research is centered on policies, technological devices, and security apparatuses utilized by state (or sometimes para-state and para-military) entities for different purposes and in different domains; whether in border control, to monitor migratory fluxes, for policing suburban districts, in the private security market, or the use of biometric data and DNA profiling (e.g., Albro et al. 2012; Fassin 2013; Maguire et al. 2018; Sausdal 2020). Furthermore, we find instances of this trend in studies dealing with problems such as the criminalization of immigrant populations (what Katja Franko [2019] has designated as The Crimmigrant Other) or the challenges and limits posed by new technological developments, such as DNA criminal databases, whose status as an infallible evidence to identify authors of crime thus renders it the ultimate tool for punishing crime and enhancing security (Ball et al. 2014; Gledhill 2018; Lyon 2015, 2018; Machado and Prainsack 2016; Marx 2016; Monahan and Wood 2018).

There are plenty of useful examples that lead us to think on whether anthropologists committed to the problematization and questioning of security-related issues are not reproducing hegemonic discourses and arguments that have become socially stabilized. My argument is that even when focusing on actors whose power is recognized or legitimated—whether decision-making or action-taking power—it can still be that anthropologists are leaving out the everyday practices of security as experienced by ordinary people that could somehow correspond to a need to achieve safety. This absence and invisibility (that a substantial part of the anthropological production on security excludes) of ordinary people that aim to experience safety, rather than security, can be overcome by considering those who are one of its principal agents and actors: crime offenders.

In the multi-sited ethnography I carried out in 12 male and female prison facilities in Portugal, notions of marginalization (in the sense of social exclusion), marginality (in the sense of deviance or deviant behavior), and margin (in the topographical sense) were constantly being brought to the fore by the narratives and interactions of incarcerated individuals. Portuguese prison population is comprised largely of people in situations of marginalization, social exclusion, vulnerability, precarity, and uncertainty at various levels (Cunha 2020; Gomes 2018; Frois 2020). It can be characterized as a population with limited formal education, living in precarious conditions in areas situated in the peripheries of major urban centers, strongly reliant on state welfare benefits and steeped in contexts pervaded by violence and problems of ethnic-racial discrimination. In their condition as incarcerated people, they are granted a set of rights (shelter, nourishment, work, education, and healthcare). Within walls, though, “order and security”—in the formula referred to me by a chief prison officer—is directed toward the protection and safeguarding, first and foremost of the institution's staff and secondly of the community living outside its walls (by minimizing the risk of escape, for instance). Only then, and as far as it does not compromise the former, is the safety and protection of inmates considered.

Just as outlying urban areas are territorially marginal vis-à-vis productive and economic urban areas, so too prisons can be considered as occupying a doubly marginal territory, removed as it usually is from major urban centers. From this perspective, prisons can be compared to “fortified enclaves” (Caldeira 2001) or “gated communities” (Low 2003), whose design is subordinated to questions of security—high fences, gates, alarms, and video surveillance systems, et cetera. But contrary to these, whose concern is to protect its inhabitants from outside risks and to make them feel secure while inside, the security apparatus in place in prisons is designed to contain the risks to security posed by the communities from within (see also Ivasiuc [2019] on “nomad camps”).

My proposal is that focusing on security from the margins and marginalized actors allows us to incorporate different dimensions and implications of security and safety: physical, environmental, or territorial; involving sociality and interactions; intersecting variables of gender, class, and race.

Entering the Prison: A Brief Note on Methods

Whether stemming from my prior work on surveillance technologies (Frois 2011, 2013) or from the lack of security approaches not focused on state actors, while investigating security and safety within the prison, I came across a spectrum of unexpected meanings and experiences. I was considering individuals that employed security practices that could be contextualized within a vernacular approach to security, as I aimed to delve into “how security or insecurity is perceived and lived by individuals in their daily lives” (Baker and Lekunze 2019: 208). In other words, imprisoned people produced discourses where the concept of security became dynamic and fluid, demanding attention to shifting circumstances and individual wills, even when the speaker uttering the word was oblivious to these nuances. Security and safety were equated simultaneously (in reference both to the context and timeframe of their criminal activity and during imprisonment) with notions of protection, well-being, comfort, absence of risk, and absence of uncertainty.

Initially, I planned to spend two years doing fieldwork in three male prisons and one female prison in the greater Lisbon area, the Portuguese capital. However, this investigation turned out to be a long-term fieldwork from 2015 to 2019, resumed in 2022 until 2023, covering prisons throughout the country, involving the inmate population, prison officers, correctional treatment staff, and wardens (Frois 2020). The methodology was similar in all places, although the duration of my stay was variable. As a rule, I spent two to four weeks in each place, sometimes this period was extended to two to four months, yet in other cases the option was to make monthly visits throughout one year—each lasting one week.

Invariably, the first meeting was with the warden, the head of the prison officers, and someone from the correctional treatment staff. The purpose of this preliminary meeting was to present the general objectives of the study and to select the participants, considering ages, profession, origin, nationality, ethnicity, and crime typologies. The choice was based on my interlocutors’ assessment of the inmates’ willingness and interest in talking to an outsider about their past experiences and prison life, as well as ensuring my work did not interfere with inmate's activities such as educational or work activities.

This procedure did not exclude the prospect of including other interlocutors in the course of fieldwork, especially considering that in male prisons my interaction occurred outside the common spaces of the inmates’ wings and was limited to a room in the administrative area. In women's prisons there was greater freedom, and with the exception of cells, I had full access and could accompany inmates in the courtyard, cafeteria, common room, and workspaces. All interviews were conducted without the presence of prison officers or members of the correctional treatment staff, and recording was authorized as long as in compliance with principles of anonymity, privacy, and confidentiality.

At no time was I asked about the information I collected. Neither was I ever asked to provide a copy of the material prior to publication or presentation in public or academic forums. This contributed to my general perception, already substantiated through discussions with other colleagues doing work in similar settings, that the absence of control or censorship was due mostly to the fact that my institutional interlocutors did not expect me to reveal any problems or issues that had not already been publicly acknowledged by the Ministry of Justice, the Portuguese entity that manages the penitentiary apparatus. Admittedly, I was not looking for “hot” topics—prison officers’ violence, corruption, drug trafficking, et cetera—and it is clear that not everything has been said to me, as it is equally obvious that some men and women have omitted information.

In the following pages I discuss how elements such as overcrowding, infrastructure degradation, shortage of human, and material resources have an impact on the security and safety experienced by inmates in prison. From the outside, the prison represents a place that grants security to the outside—the opposite of the gated communities I mentioned before. As experienced from within, on the contrary, security-related issues unfold into analogous concepts that are closer to notions of safety and unsafety, protection and unprotection. It should be emphasized that in the personal narratives of inmates, prison officers, and correctional treatment staff alike, security also emerged in connection with the economic, community, environmental, physical dimensions, and interaction with the welfare state in solving associated problems (Frois 2017). Therefore, conceptions of security, safety, and well-being are intertwined in often complex and variable ways with experiences of risk, vulnerability, or exclusion. This led me, on the one hand, to once again consider the importance of thinking about security from the margins—and the different forms margins can assume—and, on the other hand, to explore the everyday practices of security and the strive for safety as they are negotiated on a daily basis.

Prison: Whose Security?

Having started fieldwork in prison facilities where security issues were notorious (Frois 2016) helped me to realize that these were settings where inmates’ sense of personal safety was largely determined by the specific material and social conditions of the prison facility they were allocated to. As such, I realized that I should not speak about “the prison” but rather about “prisons”, and the question of security—along with several factors—was a determinant element in everyday life in confinement.

Thus, for instance, in one of the prisons, intended mainly for men aged between 18 and 25 years, convicted of crimes with sentences ranging from 9 to 15 years in prison (the maximum in Portugal being 25 years of imprisonment), the physical security of inmates and prison officers was constantly at stake. Being young adults convicted of crimes involving the use of weapons and violence raised the general awareness (of prison officers and inmates alike) of the imminent threat to overall security and personal safety, given the ease with which a verbal dispute could quickly escalate to physical assaults among inmates or between inmates and prison officers.

Violence and confrontation were ubiquitous in the histories and trajectories of these men, both before they were sentenced to prison and during their time there. Their life histories described criminal track records that began in adolescence, whether on their own or in the context of a group or gang. A large majority had at some point in their childhood been admitted to “boarding schools”, as they called them, meaning foster homes and institutions for children and youths in the care of the state. Other inmates had already served sentences in juvenile detention centers, resulting from being criminalized in their youth (Bartlett and Ricciardeli 2023; Biondi et al. 2018). In several circumstances the relationship with the family of origin was almost non-existent, since many of them had been rejected by both nuclear and extended family, in contexts of economic precariousness, alcoholism, or domestic violence.

In this prison, most men claimed to come from the “bairro”, social environments with similar characteristics to those of their fellow inmates, which provided a kind of social collective comprised of persons with a common background and experience: youths who had dropped out of school relied on their group of friends as their main socializing circle. I realized that in this particular prison, the inmate population comprised men from the same neighborhoods and from the same areas of origin. It was clearly what Manuela Cunha (2008) had described as the phenomenon whereby the neighborhood is brought into the prison, just as, conversely, the prison is already “found inside the neighborhood” (see also Boe 2016; Comfort 2008).³

This points out the marginalization and marginalized features of crime offenders and inmates, this time as the continuity of a social collective maintained outside and inside the prison (Weegels et al. 2020). The intricacy of what was being replicated within prison walls largely comprised the stereotypical logic of the street gang. Once more, we are facing notions of masculinity and virility that, as Fassin observes, “is often imported from outside the prison, and is manifested through the performance whereby an inmate stages his masculine status in the presence of other men” (2017: 175 and following).

In the opinion of prison officers, this fusion between ghetto and prison, or put differently, between familiarity and incarceration, was harmful. On the one hand, young adults seemed to accept deprivation of liberty almost all too readily, reflecting the prevailing sense of feeling “at home” in a setting whose familiarity seemed to be compounded by their being surrounded by outside acquaintances or associates or even friends. The same group dynamics are reproduced inside the prison and are equally founded on violence, on the conquest of “respect”. The assertion of a masculinity that supports the cohesion of the social group served both as a means of personal protection and of strengthening an individual's standing as an agent of security and order within this particular collective (Biondi 2017; Godoi 2017). On the other hand, the negative consequences of a coextension between ghetto and prison were evident in the daily lives for some of these men. Unsettled disputes were now resolved within walls, replicating the organized groups established outside the prison. One of the inmates, who came from a different geographic area and thus had no friendships or frequented the same neighborhoods as the other inmates, requested to be transferred to another prison, alleging personal safety issues and explaining that “anyone who's not from the neighborhood and has no friends [here] is screwed, gets beat up, is robbed daily.”

At the time I was doing fieldwork here, the infirmary actually served as a security wing for men who were considered to be in danger within the common population, particularly regarding their physical integrity and personal safety. One of the men told me about his case. He asked another inmate for an ounce of “borrowed” tobacco, which he would have to return in 15 days, adding a pack of rolling papers. He was hoping his wife would send him money or tobacco to pay off the debt. The money, however, did not come. He not only failed to comply with the agreement but found himself needing to borrow some more tobacco. He explained the situation to the other inmate, asking for further credit for a new ounce of tobacco. The terms for this new loan involved returning four ounces worth. If he didn't make this payment, the consequence was clear: “they'd have to find another form of payment,” meaning he would take a beating or the debt would be collected directly from his family.

Faced with this prospect, he knew his physical security was at risk; he felt unsafe. He had lost the respect of the other members of the group and of the prison population on the whole, and so he was endangered. Since the cells in this prison are individual, the solution he found was to remain in his cell permanently, asking the prison officer to not open the cell door every morning. He never went out, neither to the courtyard nor even for meals, which were brought to his cell. All he could do was wait for the situation to solve itself somehow, even if it implied being locked up in the cell 24 hours a day until his creditor was released or transferred.

The prison officers felt powerless to put an end to this way of conflict resolution. By the moment they became aware—when inmates finally came to them asking for “protection”—they were already facing a situation of imminent physical danger. They recognized that the lack of sufficient manpower to ensure the safety of everyone within the prison space increased the risk. As one officer in a prison with similar characteristics (in another part of the country) explained:

Imagine this: there are only one or two guards available for an entire wing. A man gets out of his cell, goes for a coffee and leaves the cell door open because he has no key. When he comes back, he doesn't have his sneakers, he doesn't have his cigarettes, he doesn't have his television. Then the conflicts start, there's a fight between them. If the officer is alone at the other end of the building, he cannot see what is happening, and even if he realizes there is a fight, he has to wait for support from other colleagues. Officers are not supposed to be the jailers who merely open and close doors, and they're not there to judge or serve as arbiters either. But that explains the insecurity situations that they go through here. The officer doesn't work miracles, he can't be everywhere at once.

Here we have a clear example of the ambiguity of the senses and meanings of security to which I alluded earlier. On one hand, there is the state security apparatus in operation: the prison as a space of punishment, the prison guards as actors of state security, with their inherent material and human resources. On the other hand, this is the lived experience of unsafety by imprisoned men, which goes beyond merely physical security. In other words, we are faced with a paradoxical situation that only ethnography can reveal: the security apparatus does not necessarily equate the experience of safety. This important nuance must have implications for how the anthropology of security can and should find the operative concepts with which it works in the study of everyday security.

In yet another prison, with severe overcrowding and infrastructure degradation, the sense of physical security, of being safe, seemed to depend crucially on an element as specific as the wing the person was allocated to, which in turn depended on a notion as imponderable as what the inmates describe as “luck”. If an inmate was “lucky”, as they themselves put it, he would be allocated to a refurbished wing, where approximately one hundred men were housed in individual cells, with toilets and showers, where he could have the cell open without fear of being robbed or assaulted. As a rule, this wing was populated by men in a more advanced age-group or from socio-economic strata above the prison population average. Conversely, if he was “unlucky”, he might find himself placed in a wing that was in all respects identical—in its size and number of cells—but where the showers were shared, as were the cells (usually three men to a bunkbed) which, combined with the other common spaces, made for a constant movement and interaction between inmates, and all the potential for situations of conflict this entailed. One of my interviewees described his experience:

In the first wing where I was placed it was a mess, things were broken everywhere: toilets and sinks broken, electrical outlets torn out, there was nothing that was not damaged. I quickly began to understand how to behave and move in that space; like, “If I go through there, I'll be robbed, I'll lose the sneakers.” That place looks very big, but it boils down to two corridors. In every other cell you'll find illegal cell phones, drugs, tobacco, cleaning services, sex, anything you can think of. After a while you get used to it. Luckily, I was transferred to this wing where everything is calmer, there's no running, there's no shouting, everything works better.

This reveals that the prison administration may have a differential formula when allocating inmates to different wings: those with more resources (even if symbolic) are deemed deserving of better living conditions than those who do not possess the same resources. It means, therefore, that the “prison climates” (Martin et al. 2014) is subject to variables that inmates are unaware of, even though they all start off in the same condition as criminalized individuals.

During fieldwork with men, when I asked the inmates if prison was a secure place, they were invariably confounded by the question, unable to ascertain exactly what I meant by the word “security”. They often responded by asking, “secure how?” or “safe for whom?”, followed by replies such as, “not for inmates, that's for sure!” This perplexed response to a question I had believed to be straightforward, was one of the things that prompted me to reflect upon the very notions of security I had construed prior to my fieldwork. Especially in the light of the conflicting meanings attached to it, depending on whether one looked at prisons from an outside or inside perspective, but also depending on whether considered from an institutional/conceptual perspective, or from an individual/lived experience.

Moreover, and returning to the question of the margins and the processes of marginalization, the life trajectory described by my interlocutors revealed that they had lived in conditions of insecurity and vulnerability form much of their lives, namely insecurity regarding physical economic and environmental violence—the same applying to female populations, as I will discuss in the next section.

Violence, crime and conflict are inscribed in criminalized people’ experience as a legitimate means to uphold shared values such as respect from peers (Bourgois 2003). It is also as an instrumental skill—a learning naturally carried into the prison setting, where it also regulates interactions with other inmates while imprisoned (corresponding to what Irwin and Cressey [1962] defined as the “importation model”). To be safe, to feel secure, they had to resort to acts that endangered the physical and material security of others. In this regard, it is as if the concepts of security and insecurity lose their fixed conceptual dichotomy and become interchangeable, or at least ambivalent, and thus apt to produce what Erik Bähre (2015), has termed the “sociality of violence”: when each individual can be simultaneously an agent and a victim of violence.

In the prison of Coimbra, one of the officers described a situation (among many) that called attention to yet another aspect of security within prison, but this time as a place that besides punishing, also becomes a place of care for inmates. This point is also raised by Luisa Schneider (2023) who highlights that individuals experiencing homelessness are employing prisons in unanticipated ways: including seeking refuge from violence, ensuring survival, accessing social or medical assistance, improving their opportunities, or finding housing solutions. Prison facilities are essentially conceived as spaces that must serve both their purpose to confine and punish, and their aim to rehabilitate and capacitate, or in other words, as spaces of temporary reclusion that are ultimately guided toward eventual release.

This prison was occupied mainly by men over the age of 30, sentenced to over ten years of imprisonment and generally having low academic qualifications whose professions related to industry or technical skills such as mechanics, electricians, etc. The crimes for which they were convicted were designated by the officers as “serious crimes”: murder, attempted murder, robbery with violence and use of a weapon. The fact that they were older inmates, who avoided conflict, combined with the length of sentences being served created the kind of stable environment that is propitious to establish sustained programs of paid labor occupations, which included a wide variety of crafts and technical work: restoration of furniture and books, production and repairing of shoes, repairing of air conditioning systems and refrigeration appliances. This prison facility resembled a well-oiled machine whose rhythms and routines recalled well known comparisons of the prison with the monastery, the school and the factory, as Foucault (1995) had done, or the characteristics of the total institution, as proposed by Erving Goffman (1961). Now, when inmates are just over 30 when they are imprisoned and may remain there for 10, 15, 20 years, the major difficulty lies precisely in the moment of release to the outside world. This is not a minor issue and was indeed frequently mentioned by my interlocutors. According to one of the officers, when the court intervenes and condemns them to prison, it promotes their disintegration as a person:

An individual left here a few days ago, but he should never have spent a day out of jail. Of course, the law doesn't allow him to stay here, but look: he's been stuck here all along, for years. When he came in, he was young, he killed a person and served his time to the full extent of the sentence. He was released, which is to say he was abandoned. He was a normal person here, he worked, he was functional. Soon he was back here again, he's only been out there a few months, committed another crime and was convicted. That was inevitable. Another example: there is this inmate who is up for temporary leave, they want to let him go home for four days, but he himself does not want to go because he has nowhere to go. It is not easy.

A similar situation was described in a female prison in yet a whole other setting. According to a prison officer:

There's a girl in here who's got a five-year sentence. On her first leave she got drunk, when she returned, she lost the job she had here and stopped attending school. Then she was granted one last leave because she was about to complete five years, and she freaked out, got unbalanced, was making trouble with everyone. I asked her “What's happening?” and she said to me: “I was aware of the street, it's very difficult. I'm afraid to go out into the street, I'm afraid to go free, I'm already used to this.” She has no mother, no father, her husband died, her children were put in an institution, and she has nothing left. But she can't stay here forever, can she?

These narratives showed that uncertainty and insecurity regarding the future or even daily life outside prison corresponded to another interpretation of the meaning of security, in which material conditions, access to housing, the labor market, food and healthcare were considered. In a sense, it was almost as if we were returning to the beginning of these people's trajectories. Low education levels and skills, doing unqualified work with precarious employment and an overall low socio-economic status, subject to social exclusion, both before and after incarceration, represented the materialization of insecurity within the broader society. The problem of marginalization becomes crucial again for my reflection on the ambiguities surrounding physical security and personal safety. In fact, some of the men and women in prison belonged to a collective that existed within a broader normative system that classified them as marginal. Inside prison, however, they encountered the model of “good life” (as a normative ideal) that they had coveted before and during the practice of crime.

Security, Vulnerability and “Crimes of Subsistence”: When State Security Fails

This last example introduced me to yet another dimension of insecurity, as I now focus on the issue more centered on gender roles (Gentry et al. 2019; Hoogensen and Stuvøy 2006). While with men the experience of insecurity was almost invariably associated and described as relating to material issues and situations of confrontation and physical threat, in the case of women, as I will now discuss, its meaning points to other aspects that are more accurately associated with notions of vernacular security regarding safety, protection, and well-being.

Throughout my research, I did fieldwork in two female prisons, one on the outskirts of the city of Lisbon (Tires), with capacity to accommodate six hundred women and another in the south of the country (Odemira), which housed approximately 50 inmates. During my initial visits to Tires (first in 2015 and later in 2017), initially nothing seemed to be different from other prisons with the same capacity, except for the wide outdoor spaces, with pine areas and small gardens. There were two large pavilions that housed most of the imprisoned women, some smaller pavilions adapted for occupational activities and a therapeutic unit for people with drug addictions.

As I was led through the various spaces, however, I was intrigued by a ground-floor building, with typical prison architecture of the late 1970s in Portugal. As I came to find out, this was the so-called House for Mothers (Casa das Mães), whose courtyard served as a kindergarten used by inmates who had small children in their care. Children ranged from one to three years old, infants and toddlers still learning to walk. As in any nursery or kindergarten, when visitors approached—this day prison officers and the warden, who knew them by name—they would instantly come over, eager to interact.

The possibility of infant care in prisons is a right inscribed in Portuguese law and immediately raised a host of issues that I had not anticipated. It opened up a whole new set of questions regarding the practices, experiences, and meanings of security, demonstrating once again that the conceptual definitions developed by the anthropology of security were inadequate as analytical tools to interpret the specificities that fieldwork faced me with. Thus, while the gender differentiation of inmates provided an obvious starting point, all the implications this might have on the notion of security proved less obvious, as the question of motherhood soon revealed. The premise can be simple enough: women commit crimes and are convicted; the law enables women who are arrested to care for their children within a specific age-range and for a limited period. This means that in certain situations imprisoned women who are mothers continue to take responsibility for child-rearing during part of their imprisonment. This may be interrupted either because their sentence is served and they leave with their children or because the children are placed with relatives or in foster homes after the legal period allowed by law and the women remain imprisoned.

It is obvious that not all imprisoned women have offspring or had children with them in prison, but as I came to realize, this dimension was fundamental to understand their criminal trajectory and the impact of incarceration. As I discuss elsewhere (Frois 2017), while most women imprisoned in Portugal are not different from the male population in terms of their low socio-economic status, low education levels and unqualified jobs, they are often the only caregivers of the children or the nuclear family (Gomes and Duarte 2018). Moreover, while in the case of men, prison sentences include a wide variety of crime typologies, the overwhelming majority of crimes committed by women are limited to a few categories, most notably crimes connected to drug trafficking or activities that fit into what is commonly referred to by prison officers as “crimes of subsistence”.

In other words, the profile of imprisoned women is also that of vulnerable individuals, living in situations of structural poverty and social exclusion, with family backgrounds characterized by physical and verbal violence (of which they themselves were victims and sometimes also authors). Criminalized actions are often practiced as a last resource (Frois 2018; Moore 2020) either to overcome material insecurity, or to protect against family and community unsafety. The most common strategies to overcome these difficulties was to either become involved in drug trafficking (whether dealing in the street or as a courier) or carry out thefts that can produce quick cash returns.

One of the women imprisoned in Odemira, of Roma ethnicity, was a repeat offender for drug trafficking, and this was the third time she had been arrested. She explained that:

I don't have the fourth grade, I can't read well, I have a criminal record and there are restaurants that don't accept people like me [referring to her ethnicity], it's very complicated. I had no salary, only state allowance, and then I was dealing drugs, but now I think the opposite, it is not only the drug that gives food to us. I can work, I can ask for welfare benefits, because they [state agencies] give help, food. I don't want to leave my kids and I don't want to be stuck in here all the time either.

The prison officers discussed the situation of this woman and others in her family pointing to the relevance of ethnic origin as an aggravating factor to finding a job. Even if while in prison they attended schooling activities (learning to read and write being the most common) or professional training courses, there was little chance of reversing their living conditions outside prison. Although the vast majority of inmates, regardless of race and ethnicity, have to face the obstacles of education and qualification in their access to the labor market, this is compounded in specific ethnic minorities, as they have difficulties in working also in unqualified jobs, such as house cleaning or in the hospitality sector.

Furthermore, in Portugal, women are the main interlocutors dealing with state entities that grant financial help and social housing, to find support to maintain the minimum subsistence for themselves, their children, and their homes (Matos 2020; Narotzky 2020). The expression “crimes of subsistence” was very suitable to explain the motivations of crime—that is, to find money to buy food, clothes, medicines—and also the needs of their daily lives. Welfare state entities help to ensure a certain degree of economic security but are almost always deficient to cover basic needs. Drug trafficking, in the neighborhood or between countries or cities, was a paid activity that was accessible for those without formal requirements, such as education, professional experience, or with scheduling or other limitations. Outside the formal labor market, if they received some kind of state support, for most female inmates, drug trafficking represented a source of income whose risks were considered to be largely compensated by the prospect of being able to ensure livelihoods—pay the bills, provide for children, feel secure.

As I have discussed in the case of men, the porousness between neighborhood and prison was as much an enabler of socialization, as it could pose the risk of conflict and threat. For women, some positive aspects were prevalent. There were many situations in which several women from the same family or neighborhood were imprisoned together (Hutton and Moran 2019). This familiarity effectively counteracted the loneliness of incarceration, namely the separation from children. In short, allowed to maintain some degree of “normality” brought by closeness, even intimacy with fellow inmates, whether friends, relatives or neighbors.

In the Tires prison there was a separate section of the large pavilions, where approximately two hundred women had their routines, whether they were chatting, working, attending school, or in the courtyard. This section was a small house with a gate and barbed wire, which looked more like a bungalow and not part of a prison. One of the prison officers explained that it was a unit intended for women who the correctional treatment staff considered needed more proximate attention, enjoying routines that could not be guaranteed in the pavilions. In this space were allocated women who had presented physical and psychological frailties, regardless of age or the crimes for which they were convicted. Many of them were being treated for depression, sometimes prior to confinement. In their words, they said that in this “house” they found a space of tranquility, well-being, or what we could define as a sense of physical security and personal safety where they were being taken care of (see also Bucerius et al. 2020). There were no cells, but rooms for two people. All women worked, participated in courses, and at the end of the day they were together. It was not idyllic, but there was no noise, no bars and locks, no crowds. Care was a feature provided by the prison to afford a sense of safety.

Thinking about safety and care led to a new insight to reflect upon the concept of security from the perspective of marginalized actors and from the standpoint of a place that is set apart and relegated to the margins of society (see also Ugelvik 2016). While physical and material security were a concern in prison everyday life both for men and women, in each case their implications point in significantly different directions. For women, the experiences of physical and symbolic violence were impacted by their condition as mothers, wives, and housewives, which in turn reflects upon their experience and impact of incarceration, namely at the emotional level (Crewe et al. 2017; Moore and Scranton 2014). Their daily life in prison was characterized by the management of anguish, loneliness, or the feeling of despair by the separation of their children and other relatives. In some cases, though, this was counterbalanced with the well-being and sense of opportunity that protected them from violence, provided working opportunities and time to envisage their future—however unfeasible these plans might turn out to be. Vocational training sessions and other occupational activities such as workshops, in addition to providing practical guidance and skills for the exercise of a profession, were also intended to operate at the level of physical and psychological well-being (Das 2008; Merry 2008). While for men, some of these activities included workshops directed at issues such as “anger management” or “generating social pathways”, for women the emphasis was domestic violence and forms of protection in case of risk, as well as actions intended to promote their personal autonomy and improve their self-esteem.

In Conclusion

To conclude, I take up the question posed by Pedersen and Holbraad in the Introduction to Times of Security: “What does being secure, and insecurity, look like for the people we study, and how might this compare to how anthropologists themselves, as well as other social scientists and commentators, imagine security when they broach it analytically?” (2013: 1).

In Portuguese, there is no clear distinction between safety and security. Consequently, a semantic slippage, which also occurs in the meanings of the English words, is more challenging to identify. This semantic slippage carries ideological implications I aimed to explore. Specifically, the fact that the security apparatus of the state and major public institutions does not guarantee safety for individuals who find themselves outside the socially hegemonic conditions. Those who lack hegemony in society will sooner or later discover that their safety is not dependent on, and it can be threatened by, the security apparatus. I tried to show that this paradox is compounded by an element of ambivalence when we consider the context of the prison: while some individuals are supposed to be restrained by the public security apparatus for being unsafe to the “public good,” they end up discovering a relative safety inside. That said, different agents are in different positions in their relationship with this semantic slippage, with this semantic trick ensuring the validation of the values and ways of life that the State and major public institutions (public hegemony) in Portugal seek to uphold.

Going back to my opening discussion on prisons, crime offenders, and prison inmates as subjects through which the notions of marginalization, marginality, and margins are conflated, I tend to follow Thomas Hylland Eriksen's proposal when he states that “asking for the ways in which people under different circumstances strive for security, and conversely identifying the factors that render them insecure may offer a promising framework for future anthropological research” (Eriksen et al. 2010: 2). Ideas of everyday and vernacular security permeated the discussion, which focused on both inmate's histories and trajectories in the present and their expectations for the future. This is a complex reality that calls for ongoing efforts to study and reflect upon security, inequalities, vulnerability, and social exclusion. By understanding the interconnectedness of individual lived experiences, institutional narratives, and cultural and structural relationships within the realms of law, illegal practices, and society, we can foster a more nuanced understanding that accounts for the diverse voices and realities of those who experience insecurity in its myriad forms.

This opens new avenues for a more comprehensive understanding of the versatile and ambivalent meanings attached to the concept of security and the potential of an anthropological approach. It foregrounds criminalized persons as ordinary people who practiced extraordinary actions under specific circumstances giving them a prominence they have lacked. It is a perspective that may potentially broaden the scope of our analytical interpretation. Understanding security from the margins, and specifically from the perspective of a place that is territorially marginal while at the same time occupies a central place in state intervention on the political, judicial and penitentiary dimensions, allows us to incorporate different dimensions and implications of security: physical, environmental or territorial; involving interactions, socialities and encounters with the other; intersecting variables of gender, class and race. The disruptive and creative potential evinced by the margins through the trajectories of incarcerated people and the everyday experiences of confinement, provide us with a textured and dense depiction of security and safety that is otherwise difficult to grasp fully.