Introduction

On a mild autumn morning in 2017, we followed the day-to-day life of security at Rivertown (fictitious name), a shopping mall located on the west side of the city of São Paulo, in Brazil. This was one of the first developments of its kind to open in the city back in the 1980’s, when the country’s private security sector was also being formalized and expanded, and the number of enclosed leisure and living spaces guarded by private security guards was beginning to grow (^{Caldeira 2000}; ^{Cubas 2017}; ^{Durão, Robb-Larkins e Fischmann 2021}; ^{Robb-Larkins e Durão 2022}).¹ Since then, shopping malls have been fundamental to urban life and are considered safe spaces for consumption in the country (Durão, Robb-Larkins e Fischmann 2021; Robb-Larkins 2023).² Today there are 639 malls throughout Brazil. Rivertown is among the 15 largest and most frequented malls, with an average daily attendance of 60,000 people.³ We accompanied security guard Serra on his daily security work at the mall.

Next to Serra we observed the flow of people at the foot of the escalator, as usual. We noticed that he and the other security guards were smiling and nodding to anyone who made eye contact with them.⁴ Serra tells us: “The important thing is that customers feel good in the mall; we are part of that environment and we provide that well-being”. To this end, the presence of this security guard close to the customer must be gentle, discreet and friendly, but always attentive to the smallest details of what might be “out of place”, as the guards tell us. On the other hand, the security team that works in the parking lot and has closer contact with the urban environment from the outside is heavily armed with 38-caliber revolvers and batons. They are supposed to prevent the “outside evil from getting in and escalating any problems”, we are told. The security guards who work inside the mall are hired directly; they have what is known as an “organic contract”. Those outside are outsourced and are employees of a large multinational private security company.

The security that takes place inside the mall is designed to support customers, particularly people who are looking for information. It was in this spirit that Serra took the lead in guiding a man who was having difficulty manipulating the touchscreen to find information about the location of a store, a guidance that was always accompanied by extreme kindness. This is one of many cases that show how security guards are confused with the mall’s material and environmental indicators of comfort: the carefully landscaped, tree-lined atriums, the small rest areas equipped with padded armchairs, and the terrace with its panoramic view of the great metropolis. Serra is in all these places, keeping watch, sometimes camouflaging himself, but most of the time visible and available to the clients, also known in the jargon as “our guests”. The uniforms of these professionals reinforce the image of care and protection, oscillating between “administrative” and “tactical” (as they say in this environment), with hats inspired by the custodian helmets of the British police (^{Reiner 1985}) and a predominance of the sober color black. As the security manager we interviewed explained: “We need to convey peace of mind to our customers, but also impose a certain respect without physical intervention.” In addition, the security guards say: “Not only do we have to deal with the customers, but here we have to deal with slackers and child beggars.”

At a certain moment, Serra stops. With him, we stare at a black man who appears to be in his thirties, dressed in sweatpants and a black t-shirt, carrying a motorcycle helmet. He is sitting on a wooden bench, unaware of our presence. The reason for the watchful eye is that the customer is in front of a jewelry store, “suspiciously handling his cell phone” - as our interlocutor points out. When the man gets up, the guard discreetly follows him and warns his colleagues at the electronic monitoring command base on the radio: “Here’s a code 2” - in other words, the identification of a suspicious person - “walking towards the jewelry store”. The transmission of codes on the radio is an essential tool for security that operates unarmed, so that customers don’t know what is being communicated. So the man is now a monitored suspect. We ask Serra what he notices is strange about the behavior of the man being observed. Serra hastens, a little nervously, to say: “No, it’s not because he’s black, he could be white, but he seems restless and he’s looking at the jewelry store a lot”. A few minutes later, a woman who appears to be the “suspec’s” wife appears. They both go to a counter to deposit a coupon for the “consumer month” draw in March. Serra assumes that it was a false alarm but insists: “we don’t always get it right, but we have to read the signs”.

Security on the move

This security, which segregates, differentiates and often covertly discriminates, has a spatial expression. The infrastructural imposition of security in São Paulo’s urban space is impressive. Walls, gates, guardhouses and technological systems for controlling access and monitoring private and public spaces proliferate in the urban landscape in Brazil and Latin America (^{Cardoso 2014}; ^{Durão 2023}; ^{Peron and Alvarez 2019}).⁵ Today, there are few streets in middle-class residential neighborhoods that are not monitored by public-private consortia. Residential condominiums dominate, defining urban landscapes considered “modern”, representing certain cultural values such as privacy, exclusivity and security (^{Herzog 2015}: 2; see also ^{Duany, Plater-Zyberk and Speck 2000}; ^{Blakely and Snyder 1999}). Millions of square meters of urban space are occupied with grandiose enclosed shopping malls, which are not only a form of consumption for the middle classes - the booming commodified modality of security -, but have become a political and moral good (^{Goold, Loader and Thumala 2010}). Social segregation is forbidden in these spaces; racial discrimination is a crime.⁶ But separations, identification and the production of suspicion involving black, brown and ethnic minority people are actions that are permanently reinforced by these infrastructures of control, and are increasingly based on the confusion between security and care, as we have argued in this text and others (^{Durão, Robb-Larkins and Fischmann 2021}; ^{Robb-Larkins and Durão 2023}). As such, spatial segregation by class or race cannot be openly expressed and practiced under penalty of prosecution, but it can be observed ethnographically in residential and commercial contexts, with common characteristics and nuances, based on what we argue is hospitality security.

The security atmospheres created for residential condominiums and built-up commercial and leisure spaces share some fundamental characteristics. Firstly, the production of suspicion is generally carried out with a certain delicacy so that passers-by don’t notice the work of a socio-technical apparatus that monitors them, but also monitors the guards and inhibits them from confronting citizens. Secondly, the material objective of security is to provide environments where the provision of surveillance services is perceived as being integrated into the structure of the spaces. Everything is done to create a feeling of receptivity and care for property and interpersonal relationships for those who frequent the places, especially for those who pay for the experience, whether as residents or customers. Particularly important is the concern for the urban order and cleanliness of the spaces, which generally also involves the attention of the security guards. For example, when security guards see garbage lying on the floor or equipment out of place, they immediately take the initiative to put it right. Thirdly, as we’ll see in the text, all of this depends on carefully planned daily micro-actions in the interaction between security workers and customers. The security guard thus ceases to be a mere inhibitor and preventer of criminal action, as provided for in the sector’s 1983 law, and, in a way, becomes part of the consumption of the place.

What characterizes security in this mall, as in others, is the absence of obstacles to people’s entry - such as gates, turnstiles, biometric identification or any other equipment that can identify who is entering beforehand. In this way, the attentive circulation of security guards, their willingness to be sought out to inform or support customers and a closed electronic monitoring system behind the scenes that guides the work of identifying suspects, are essential aspects of security management. As we have shown in previous texts, in the case of residential agglomerations to be protected, the concierge is the determining infrastructural figure in property security - this reception area prepared for control, identification and hospitality at the entrance to buildings (^{Durão 2023}; ^{Durão, Robb-Larkins and Argentin 2024}). The human element of security at the gates are the security guards, but they can also be doormen, access controllers and informal workers, often reserving the work of women for day shifts, and men, with or without weapons, for night shifts. In this sense, fixed posts and localized security control is what characterizes work at the gates of condominiums and buildings, with the integration of more complex technological defense systems as the wealth of residents increases (cf.^{Durão, Robb-Larkins and Argentin 2024}).

Lacking control of who is who - unlike gatekeepers who manipulate databases with identity records - security guards in shopping malls deal with an enormous flow of people, and therefore face a series of complex situations on a daily basis that often go unnoticed by customers. The apparent lack of control of this passer-by population is compensated for by direct management, daily orientations, shift change meetings, the monitoring system and, fundamentally, the distanced scrutiny of the guards as they move around the commercial space. In these environments, private security is produced in movement. It is a spatial movement, in that there are no fixed points for identification and direct monitoring.

In addition, these enclosed commercial spaces have dynamics associated with the very idea of property security, which is intended to be presented at all times as welcoming, similar to what has been created for hotel environments (^{Clifton 2022}; ^{Lucca and Costa 2016}; ^{Silva and Mesquita 2018}). This is a model that has been increasingly advocated among managers, entrepreneurs and security actors, especially after several serious incidents involving security guards in commercial areas that have been the subject of global media coverage and controversial trials (^{Durão, Argentin and Vituri 2024}). A more recent example was the murder of a black customer in one of Carrefour’s supermarkets at the hands of two security guards in the store’s parking garage (^{Durão and Paes 2021}; ^{Robb-Larkins 2023}). Critical events like this provoke new types of anxiety, attention and care in private security managers and actors when carrying out what they generically call an “operation” - that is, setting up a large and complex security system with shifts and duty rosters of 12 hours of work for 36 hours of rest, 24 hours a day, and more or less organized standard procedures.

Thus, in Brazil, which is familiar with the anti-racist debate, one of the dominant and growing trends in the private protection industry in Brazilian cities is what we call hospitality security (cf.^{Durão, Robb-Larkins and Fischmann 2021}; ^{Robb-Larkins and Durão 2023}). Hospitality security - the modality that serves to protect property and also produces racialized and white class care - is a consumer good and is also a distinctive element that aims to separate spaces without clearly affirming political and economic segregation. This type of protection generates a nuanced dynamic of security and interaction, a greater concern with the management of suspicion and, consequently, produces violent actions that are more cordial and less explicit, if not disguised. Unlike community policing, for example, hospitality security serves the interests of customers and business owners, and not necessarily citizens.

In this article we intend to explain how hospitality security operates the dynamics and layers of social suspicion in shopping malls, and how violence and social contradictions are produced and concealed that also affect the security workers themselves, who operate this security model. Through ethnography, we are interested in describing and analyzing how what we call “ante-cases” are produced. In other words, how security guards organize themselves to prevent police-type events or occurrences from happening. This is a form of security that acts on the basis of the discretion of its operators, but which feeds suspicions directed at certain people and social groups - including those who operate it. One of the most central aspects is the work of keeping bodies that seem out of place under control. For this reason, we will focus on the daily and tactical work of the security guards and their supervision in maintaining social and racial orders in the mall. We will see that it is precisely the confusion between security and care for some, and remote control for others, that creates an environment where violent actions need to be hidden from public view. But, as we will see, there are several problems with this type of security. Our choice is to detail ethnographically and, in these contradictions, how “environments of peace” are produced.

We present data from long-term ethnographic research. From 2016 to roughly 2022, we have been conducting research in the field of urban private security.⁷ Between 2018 and 2019 we carried out field research over six months in the so-called security “operation” at the Rivertown shopping center. At the time, we interviewed 40 security guards, including organic and outsourced security guards, and made long visits to follow their activities closely and on a daily basis, circulating and walking alongside them, without interfering in their work interactions, but trying to find out more about situational decisions and negotiations. This methodology has been conceptualized as shadowing ethnography, an ethnography committed to following more deeply and even literally the movement of the field and the field in motion: the decisions, interactions, work, meanings of the flow of everyday life (^{Czarniawska 2007}; ^{Engstrom 2010}; ^{Mcdonald 2005}). This approach is essential to understanding the “orders of interaction”, to use ^{Goffman’s notion (2019}),- in hospitality security, where the more subtle dimensions of security practices involving “social sensitivity” and potential conflict actually take place. The questions we ask and return to at the end of the text are: How do we envision forms of security on the premise that they are produced as non-violent security? How can we conduct ethnography and analyze private policing that presents itself as discreet services? Where, in an urban scenario of predominantly reactive and heavily armed policing responses, can the emergence of new security spaces be situated as hospitality?

Social approaches in hospitality

Every time a security guard cordially asks a person in a shopping mall if they need help, in order to dissuade them from committing an illegal act, they are making what is known in this environment as a “social approach” (^{Durão, Robb-Larkins and Fischmann 2021}: 139). This interaction can precede the confirmation of suspicion. It is an active and tactical production of a non-fact. The security guard notices something that he thinks is wrong with a person’s behavior, approaches them and, depending on their reaction, confirms or disproves his previous intuition, the so-called “sensitivity of the guard”. There are many variants that determine this final confirmation, and they are not always related to fragrant behavior. It is also said to be a social approach, now in an ironic tone, when it is necessary to approach someone who has actually stolen a product from a store in the mall. Even so, the social approach in shopping malls tends to be preventative. In other words, it implies that the guard is present and that, in the interaction, he can show the initiative to prevent a theft. In practice, however, “if you don’t catch someone in the act, you can never be sure of their intentions,” one guard told us.

The notion of suspect that guides the social approach is, in theory, broad; it relates to what the security guards call “suspicious behavior”. When a person stares in search of cameras, walks very slowly and looks around, or frequently turns away from the guards, these are behaviors that are considered suspicious. “As a security guard you have to be suspicious of everything and everyone” - Gilson tells us - “sometimes a guy dressed as a doctor is the one who’s going to commit the crime. It’s the opportunity that makes the thief, so a well-dressed, well-off guy, if he sees an opportunity to do well, can take it and commit a crime”. But in practice, there are preferred suspects. The black body, as we often see, is “intuitively” considered a “code 2”. The racialized bodies of Latin American immigrants are a kind of “universal suspect”, the undesirable poor immigrant of the shopping mall, referred to in the professional lexicon as “tango lima”. A third category of suspects are the “minors”, as the security guards say, referring to children and teenagers who go to the mall to ask customers for food and money.

Intuitive suspects

We spent several days with supervisor Geraldo, a 40-year-old white man who has been working at the mall for eight years, in the monitoring center. Hospitality security needs to avoid the spectacle of more overt approaches and therefore depends on the support of those at the monitoring center. The center has the important role of monitoring the suspect from a distance and contacting the security guards closest to the suspect’s location. They must then be alert and, if necessary, carry out tactical social approaches, which are considered a friendly deterrent and are the most commonly used in shopping centers. The center can also identify what it considers to be a potential suspect and activate a security guard who, while moving around the mall, must remain alert, but at a distance and without any contact with the person. This was the case initially described where the potential suspect was not supposed to be aware of the guard’s suspicion.

Let’s describe another situation: on a September morning when we are with Geraldo, no security guard is activated, but the supervisor is paying attention to the screens showing images from various locations in the mall. At one point, Geraldo zooms in on an image and asks supervisor Mendes, who is on the floor, to approach the selected location by radio. The image from one of the cameras at the ground floor entrance shows a black man wearing a red t-shirt, shorts and a cap, walking towards the main door. Geraldo seems to have no doubts: “We need to monitor,” he says. But he also asks Mendes if he can call the guard responsible for the entrance, i.e. warn him about the presence of “code 2”. Mendes looks at us and seems intimidated about giving the order in our presence. He makes a joke: “Just because he’s black? No, it’s not like that! The guard has to watch his behavior,” he says as he laughs and looks in our direction. Geraldo seems to know what this means; he knows his friend: “I’m going to call Ferreira” - he exclaims as a way of confirming that he has analyzed the behavior. With the radio in his hands, he transmits: “Ferreira, code 2 in Tango [code for the first floor], no social approach for now”. Geraldo warns Mendes of the presence of a suspect on the first floor but advises the guard to avoid the social approach. This is an exemplary case of how suspicion operates by intuition.

Universal suspects

We’re following Silvia, a 40-year-old white security guard born in Paraíba who moved to São Paulo 10 years ago. We’re on the second floor in front of a toy and children’s goods store. It’s 30 minutes before 6 pm and our interlocutor is looking more extroverted and livelier as the end of working hours approaches. But suddenly her relaxation fades and she inadvertently say she has spotted a suspicious person. It was a woman walking on the other side of the corridor, 20 meters in front of us. The woman appears to be about 30 years old, wearing light blue jeans, open-toed shoes, straight black hair and tied up, holding a small child by the hand. The suspect seems not to have noticed that she is being watched and enters a cosmetics store.

Silvia says sarcastically that she has no doubt: “It’s a tango-lima”. Our interlocutor then reports the case to the monitoring center and asks for attention. She may have to make a social approach to the person. The suspicious woman leaves the store, where she stayed for five minutes, and goes into another jewelry store. Silvia waits patiently, even though she knows her shift is about to end. We sense that she wants to show us her service. The suspect leaves the store. Silvia walks towards her and asks: “Good afternoon, are you looking for a store? How can I help you?” The suspect seems nervous at being approached by the guard and hastens to refuse help, going on her way. Silvia reports on the radio to the guard who is in the area where the woman and child are heading. Silvia sums it up: “When the tango-lima come into the mall and they’re alone or with children, they always come to steal, that’s for sure, there’s no mistake here.” Tango lima is a term in the phonetic alphabet that means “Latin traits” (in Portuguese, “traços latinos”). Self-evidently, tango lima refers to Latin American immigrants with demarcated ethnic traits, who are also often mistaken for indigenous Brazilians. This is the standard example of the “universal” certainties of suspicion in private security.

Child beggars

In the food square, security guard Sandro, a 48-year-old black man, walks towards a child he has spotted asking mall customers for money and food. In this case, our interlocutor seems to take extra care in his approach. He walks very calmly, skirting the restaurants around the food court, avoiding taking the fastest route. Sandro approaches the boy, who is no more than 10 years old, and tells him: “This activity is forbidden in the mall, you have to stop”. The child starts shouting and hitting the guard’s legs. Instigating a reaction in the security guard, the boy shouts “come here and stop me, come here”. Sandro, for his part, receives the blows without getting impatient. The whole situation exposes the security guard to the customers, who begin to look in both their directions. But it’s the mall managers’ orders. “We can’t lay a hand on them,” Sandro tells us later. All he has to do in the situation is receive the blows, listen to the screams, monitor the minor and radio to the monitoring center that he needs the support of a more experienced supervisor to help him with the approach.

In the past, with the increase of cases of presence of “child beggars”, as they say, at the Rivertown mall, the security manager got, in specific cases, to trigger the Guardianship Council - body responsible for ensuring the rights of children and adolescents.⁸ The manager said that the guardianship counselor of the region was in the mall sometimes take out the minors and contact families, but later he would have been prevented by order of the Public Prosecutor’s Office because they are commercial spaces. The manager regrets: “If the minor enters the mall to ask for money or food, if he managed to enter, it is because the security of the ground failed to prevent his entry”. As in other situations, Sandro knows that you have to wait for a moment of respite from the child’s screaming, try to talk quietly or stay quiet while reinforcements arrive. The poor “minors” are thus explicitly undesirable elements in the mall and one of the focus of attention of security. They also represent one of the greatest challenges to the practice of hospitality in shopping mall security. With them, all care in treatment is little. The beggar children are also those who highlight the limits of action in private security and the desire for punishment by the state. Perhaps not by chance, many of the police officers and vigilantes whose work we have followed over the years are in favor of lowering the age of imprisonment.

Failure in hospitality security

It is six o’clock on a Sunday in March 2017. We go to the monitoring center, located in the underground armored room. It is the time of daily meeting of the teams of the shift from 6am to 6pm. This is a space for discussion, often uncomfortable, of revealing details and failures in operation. The room is full with the totality of 17 vigilantes - five are outsourced and the remaining 12 are from the organic team. The supervisor Galter takes the initiative to speak, agitated, standing, putting his hands on the table. It is noticeable that he and Serra cross their eyes. “I would like an explanation of what happened in the morning,” says Galter, in a cynical tone, “I received a phone call from one of the outsourced employees who told me that he was pursued by an employee of the mall while accompanying his wife to exchange the gift he bought him days ago”. Galter makes a point of paraphrasing the outrage of the vigilante who was mistaken for a suspect: “He told me this: I was mistaken for a vagabond!”.⁹ Serra realizes that the criticism is directed at him, he becomes blushed and ashamed. Our interlocutor understood that the supervisor was referring to the false alarm code 2 reported that morning, an event we reported here in the vignette. Serra did not realize that the alleged suspect was a co-worker; they never crossed paths on the same shift. The shame of Serra clearly derived from the fact that it was another vigilante, a problem that would not be placed if occurring with a client. In this case it would be a “false suspicion” as they say, that can happen and not come to intimidate anyone clearly. Although not openly recognized in the middle, confusion between black clients and suspects is not uncommon. This suspicion is dissimilated in the premise that in hospitality security “everyone is suspicious”.

Galter does not give up and says that he reported the case to the internal management of security and expects a position “coming from above”. Later, in a cell phone message, the supervisor Galter comes to alert one of the researchers to the seriousness of this case. The supervisor tells us that what happened may be reason enough for the vigilante confused with a suspect to start a lawsuit against the company or the mall and win the case. Galter explains: “In these cases it is necessary that there will be a punishment. But you have to wait to see what should be less costly for the mall. It may be more damaging to bear the labor costs of a dismissal, since the case does not fit in with a dismissal for just cause and the mall has to pay for the termination. Perhaps the best is to make a pecuniary agreement with the vigilante framed as a suspect so that he does not start a judicial process”.

Over the following weeks, the vigilante Serra remained astonished at his work. Although we had a private conversation with the supervisor, everyone knew, including and especially Serra, that by Galter’s tone of concern he was likely to be held responsible. Serra is one of the newest and most recently hired agents at the mall. With a little more than two years in the organic post, that is, hired directly by the mall, it has one of the most stable working conditions when compared to outsourced guards. Serra was a third-party vigilante in residential condominiums for 10 years. He told us not to miss the work of access control in luxurious condominiums in Jardins Paulista. At the time, to supplement his income of 1500 reais monthly, he was engaged in arranging informal and sporadic services of all kinds, which he performed extra-shift. With the contract in the mall, Serra stopped accumulating hours of extra work and gained relative stability. He earns 2000 reais per month net and, like most of his colleagues, continues to work overtime at the mall. This arrangement showed its precariousness before the event.

The ethnographic cases expose the internal pressures and contradictions of hospitality security, the tension between care, control, suspicion. However, there is an additional risk that the last report reveals. Those who collaborate to maintain social orders, urban orders and white environments, fragrant and clean, with their constant little gestures of zeal and cordiality with head nods and open smiles, also run the risk of passing as suspects in the same place. The vigilantes can be “out of place” when they dare to frequent as consumers these spaces that they maintain. The expectation of the proper functioning of hospitality security falls on the figure of the vigilante and other protection professionals especially if they are black or people of color, but rarely implies the mall itself and its models of security. In other words, hospitality security accelerates the inequalities within the working group itself by increasing the possibility that the guards are the target of the suspicion they feed.

The mistakes and failures in private work contexts in Brazil, unlike public employment contexts, are paid with layoffs and easy rehiring of people, which generates an impressive turnover. As the vigilantes often say: “For each one of us who is ‘off’, have a thousand with vigilante wallet available in the market to hire”. The private security model, based on this human structure, highly fragile and disposable, is sold as professional, effective, and an example of modernity. Its contradictions are only exposed when they harm the image of establishments and the security industry (^{Robb-Larkins 2023}). All the logistics that operate behind the vigilante hide both the immense precariousness of surveillance workers and the structural racism imprinted on their gestures and imaginaries (cf.^{Durão, Argentin and Vituri 2024}).

The maintenance of condominiums and commercial establishments depends on this immense mass of black and poor workers. Thus, in hospitality security, the vigilante must know “what is his place”, where social positions are naturalized. The position of this worker is subject to an immense panoply of employers and consumers of his service. This is the “place of the black man” that Lélia Gonzalez and Hazenbalg speak of, which imputes to the subordinate group the responsibility for its unequal position (cf.^{Gonzalez and Hasenbalg 2022} [1979]). It is no wonder that security workers not only emphasize racism in their actions but also become the only ones responsible for the violent and racist structural incidents they incur in. This type of security is not simply a local private security modality without violence and created to operate in commercial and residential environments. Hospitality security is responsible for maintaining social and patrimonial orders in all spaces where it operates. Thus, security is part of the job of keeping cities growing unevenly and racially segregated.

This securitarian tendency, in some places, becomes more shadowy with the covert presence of public security agents scattered throughout all aspects of private security. In the malls, for many years there were subcontracted police officers, whom it was agreed to call “white hand” (^{Durão and Argentin 2023}; ^{Durão, Robb Larkins and Argentin 2024}). Today hospitality security in the malls is dependent on poor and black workers. These are the ones that guarantee the protection of people and social groups with power, money or status in relation to suspects. They are the ones who care about maintaining the social, racial and aesthetic order of the places so that customers can move around carefree. Behind the scenes, inequality and racism are part of the way these workers operate.

Hopitality security in the city

As we have been describing so far, hospitality security is a particular business modality of private security because it is linked to the way in which private property is managed in Brazilian cities, and more particularly in the movement of people and goods and the logistics of the spaces to be guarded. It is a socio-technical structure, but also a moral economy that sustains “conviviality in inequality” (^{Durão 2023}: 03). This particular type of security results from five interrelated factors that we have detailed.

First, the disclosure of crime scenes and violence is a constant in Brazilian society. Since the 1980’s, public discourse on crime, amplified by the media and, more recently, social networks and digital environments, has been based on the promotion of images associated with violent, ubiquitous and predatory crime (^{Caldeira 2000}). In everyday life, it is observable the exchange of stories about violence, robberies, thefts as a way to order the world in which one lives - what Caldeira coined as “talk of crime” (^{Caldeira 2000}: 27). This urban violence, amplified by the cases that are told and heard, generates a feeling of permanent insecurity, not always associated with the real growth of crime. The private security markets are based on this scenario of imagined ubiquitous insecurity and affirm themselves as an activity of national interest and complementary to public safety (^{Durão, Robb-Larkins and Argentin 2024}). Private security organizes the city experience, creating niches that are configured as the antithesis of scenarios seen as chaotic. The southeastern region, in particular, concentrates most of this market, with 46% of all workforce (^{Brazilian Yearbook of Public Security 2024}).¹⁰

Secondly, the transformations in the residential and commercial infrastructure of cities have led, among other things, to the creation of unique identities of buildings that are treated as heritage entities to be protected. For example, in São Paulo and other cities, the horizontal closed condominiums and the building condominiums have their own names - such as “Porto Seguro”, “Prima Sofia”, “Edifício Tejo”, “Edifício Amazonas”, etc. - probably related to family stories, tastes of builders or imaginaries of owners. In condominium neighborhoods with high-end construction, the names may refer to large European cities and the beaches of the Mediterranean, thus associating idealized images of paradises of well-being and a fantasy of belonging to the “first world” (^{Melgaço 2012}). Also the malls follow this logic. When they do not take the name of the neighborhoods in which they are located, the shopping malls make mention of the “Gardens”, the “Parks”, the “Courtyards” - as consumption oases in the cities (^{Caldeira 2000}). Buildings are thus treated as material personalities, symbols of social distinction and spaces for personal and collective investment of residents or business consortia that have a value to preserve in a very dynamic real estate market.

Third, especially for those with greater purchasing power, there is a valuation of what we can call “patrimonial governance”. This governance, located in buildings and condominiums, is manifested in the maintenance routines of spaces. This conservation is structural and material - with various improvements and investments in the beautification and technologies of condominiums and buildings -, but it is also the effect of the numerous gestures of daily care. Part of the care are, for example, the services of tree pruning, the maintenance of spaces always clean and pleasant, the treatment of swimming pools and a constant monitoring on what may be out of place made between residents and facilitated by the creation of Whatsapp groups. This constant attention projected on the buildings is produced in opposition to the public spaces of the historic centers of cities, often considered ungoverned, “messy”, “dirty”, “abandoned”. In the most gentrified regions of the city, most apartments are sold with the guarantee of ensuring 24-hour concierge as a guarantee of tranquility. In the shopping malls there are always visible guards on the various floors, and when activity ceases night teams are maintained. In this cluster of privatizations of protection, even those who cannot or want to pay for a complete system sold by large companies create their own wealth care system. For a value below the one practiced in the formal security market, there is the work of so-called “street guards”. The guards roam with motorcycles or personal cars and are sitting on benches, sometimes near fences or guardhouses that improvise gatehouses. Their role is to watch the residences, anticipate thefts and maintain the perception that the space is not only being watched but being taken care of.¹¹

Fourth, there is in Brazil a massive supply of security professionals and an immense presence of other facilities (general maintenance services) that include, very particularly, the concierge workers, cleaning and other services of attention to the patrimonial order. The growth and transversal implementation of electronic control and monitoring technologies in the country did not mean a decrease in the labor force of support services and outsourced private security. This is largely due to the low wage level, which can vary between one and two minimum wages, equivalent to little more than 200 and 400 euros per month. The various services of patrimonial care, which from this point of view include hospitality security, thus become a broad labor market for the poorest citizens with little education. In contrast to these well cared for and assisted worlds, security workers, mostly “brown” and black, come from the poor suburbs of big cities where lack of security prevails (^{Durão 2023}).¹²

Finally, we observed a certain convergence between safety and care in images and public speeches. Public security actors in Brazil continue to be identified with an ostensible, armed, “muscular” and reactive performance, most often associated with military aesthetics. Going in another direction, hospitality security is produced from a different ethos, cultivating an alternative aesthetic appearance, with more friendly, “executive” and “clean” uniforms. From the 2000’s, at the same time as the speeches and legislation that favored and valued care jobs and professions - mainly carried out by women (^{Debert and Félix 2024}; ^{Guimarães, Hirata and Sugita 2011}) - , there is a huge growth of the private security industry.¹³ The appreciation of care for private property and the “clients” of these spaces grows together with the affirmation of care as a value in Brazilian society. Hospitality security becomes part of the socialities and shared modes of coexistence in urban life. Being subject to access control or monitoring, receiving guidance or support, attention or care at the entrance of a building are common and intricate everyday experiences. All these urban experiences have become commonplace for the Brazilian middle classes, especially for those who circulate between the protected environment of the residential condominium, the mall and the work performed in a building with restricted access. In this sense, the patrimonial care, which is at the base of the offer of the services of the hospitality security, is like an insurance of life. Middle-class citizens who want to “live well” maintain the control and protection services in their condominiums, as they pay for a health insurance or private school for their children.

But they also pay for safety not only to attend protected environments, but to be able to circulate in care environments, fragrant, bright, new, modern and, at the limit, socially differentiated. Anything that affects the immaculate image of the commercial space of a mall should be avoided.

Conclusion

In countries where everyday life is crossed by many forms of violence and social racism, such as Brazil, subtle ethnographic approaches of these orders of interaction are scarce. The starting point of Brazilian and Latin American literature has been the realization of panoramas and general diagnoses of the functioning of public security, emphasizing the insecurity and participation of these actors in the vital forces of the so-called violent democracy (^{Caldeira 2000}; ^{Goldstein 2010}). Police violence and the lack of social justice in Brazil are thus often highlighted in the explanation of the country’s structural problems. The socio-political studies on the public security system, criminal justice and urban violence thus focus on the historical denunciation of violent excesses and failures of the rule of law by police or private agents (cf.^{Adorno 1995}; ^{Adorno and Nery 2013}; ^{Caldeira 2000}; ^{Durão and Robb-Larkins 2020}; ^{Durão and Paes 2021}; ^{Lopes 2011}, ²⁰¹⁴, ²⁰¹⁸; ^{Muniz, Caruso and Freitas 2017}; ^{Zaluar 1999}; ^{Zanetic 2009}, ²⁰¹⁰, ²⁰¹²). In this perspective, security agents, both public and private, reproduce and strengthen social discrimination, being no more than operators of these violence (^{Huggins, Haritos-Fatouros and Zimbardo 2002}).

This text allows us to affirm that there is a lack of complementary reflection to the consensus notion that security in democracy has risen against population groups. Considering how security is sold, bought and operated subtly as a service of “care” for certain mainly white social groups, we propose to broaden the critical look. We envisage the structures of organization of patrimonial security and the production of social suspicion in the fine gestures of vigilantes in their workplaces. As the text attests, it was necessary to dive into the social dynamics of a mode of security - such as hospitality - that becomes a good among others valued by the middle classes. This is done not only to ensure the “sense of security”, as was already exemplarily worked by ^{Robb-Larkins (2023)}, but also to hide from the eyes of these same middle classes and elite’s violence that are considered to be present and represent poor urban spaces.

Immersing yourself in this security that is part of a project of socio-structural whiteness implies accompanying the flow and movement of the production of ante-cases, or all the work prior to the occurrence of occurrences. This destabilizes the tendency to problematize only what in the professional jargon of security are called “occurrences” very present in the literature on policing. The occurrences are the result of actions of police interest in various stages of development. An occurrence is said to occur when a theft is taking place or has already taken place. Also characterized as a system of occurrences are the written records after the fact. These actions of police life have always interested researchers. The spectacle of detention and the ostensible persecutions, the securitarian search for occurrences is what exalts the ostensible police work and its operatives, but also the ethnographic descriptions themselves (^{Durão 2017}; ^{Fassin 2017}; ^{Manning 1978}). The few texts that were dedicated to explain how the routine at the tip is characterized by many moments of inaction and without the promised stimuli of “fight against crime” focus on the description of boredom as something that is devalued or even detested by police (^{Durão 2008}; ^{Fassin 2013}, ²⁰¹⁷).

In hospitality security, police operations should be diverted from the imaginary of customers who want to feel “safe”, that is, looking around and seeing everything “in order”. As the manager of Rivertown told us: “A shooting in this mall would be your death; it would never exist commercially again” (^{Durão, Robb-Larkins and Fischmann 2021}: 155). Thus, the work prior to the event in hospitality security is fundamental and based on sensations and maintaining the idea of care with the order of relatively closed environments and the well-being of people seen as clients. Tactical managers of security such as hospitality are the watchdogs in their daily micro-gestures. They are the ones who in Brazil are hired, either organically or outsourced by the establishments to operate this security model. Although such discreet gestures have less appeal to the ethnographic view of the security scholar, they are decisive if we want to understand securitized societies. It is through the monitoring methodologies of vigilante work that we understand that the discreet elements of private security are not just ornaments with no effect on global security. On the contrary, these are central elements in the security plan of cities, especially structurally violent and unequal cities of the global south, such as Brazil. The spaces for consumption and urban circulation “in peace” depend on these professionals.

This care is always precarious, and risks are taken, as security is dynamic and occurs on the move. As per the ethnographic example that opens this text, some security flaws such as hospitality are crucial and pay dearly. The ethnographic cases selected for this article expose the pressures and internal ambivalences of hospitality security - that tension is always present between acting with care and acting with control/suspicion. Vigilantes need to learn how to disguise violence in their treatment, using the “social approach”. Sometimes they interact cordially with the suspects in the act of committing an illegal act; others avoid even interacting, trying not to be noticed while watching. All this is discussed in the lectures. It is effectively a security strategy designed for shopping malls and also developed in the monthly meetings of all the mall security managers of Abrasce, the Brazilian Association of Shopping Centers. To paraphrase the vigilante Nilton, “it is the production of the bubble”.

The work in the production of suspects is, as we have seen, an essential care for the production of orders and pre-cases manipulated by the vigilantes. We describe situations and people who are considered intuitive suspects (usually black men), universal (tango-lima) and undesirable (minor beggars). Security codes are created for these groups and individuals. But the point is that they, and society in general, cannot know openly that these are the preferred suspects against whom the shopping mall mounts its daily security operation. It is on them that the watchful eye of the guards falls, with the support of electronic monitoring cameras and radio communication system. Thus, we conclude that hospitality security is an urban and moral value, but it is also and fundamentally a labor skill of the vigilantes. It is part of the work of keeping cities growing unevenly, with discrete and non-spectacular racialized separations. This is how citizens and the different social groups classified by the security system are kept in their place - including those who operate these systems.