When politics are contagious: Covid-19 and political resistance inside an immigration detention center

During the early stages of the Covid-19 pandemic, immigration detainees in the United States rose up to protests their forced confinement during a global pandemic, launching collective hunger strikes across separate facilities on a national scale. In this article, I utilize primary and secondary sources to examine the strike that occurred at the Otay Mesa Detention Center in San Diego in the Spring 2020. While acts of resistance are hardly a new phenomenon in immigration detention, the 2020 protests were unusually powerful because of their range and the pace at which they spread across facilities. Their power was a direct consequence of Covid-19, not only because the pandemic triggered the strikes, but because it introduced a common condition of vulnerability among the detained population, thus encouraging collective organization. The strikes and the pandemic showed a common form of expansion, which was acknowledged by the authorities themselves, as they adopted the same strategies of lockdown and quarantine to contain both phenomena. The history of this protest, along with those that erupted across carceral sites globally during this period, constitutes an important testimony to the political effects of the pandemic, and to the possibility of political resistance in detention.


Introduction
At the beginning of the Covid-19 pandemic, in between the spring and summer of 2020, immigration detainees 1 in the United States rose up to protest against their forced confinement during a global pandemic. As the virus ravaged through the detention system, large hunger strikes erupted to gather public attention and put pressure on immigration authorities. This was hardly exclusive to immigration detention, or to the United States. The beginning of the pandemic coincided with protests in carceral facilities all over the world, marking one of the most conflictual periods of prison politics in recent memory (see for example Heard, 2020;Jacquard, 2020;Milella, 2020;Vera, 2020).
With this article, I focus on one protest in particular: the hunger strike at the Otay Mesa Detention Center (OMDC) in San Diego, California. In April 2020, OMDC registered multiple strikes across separate units of the facility. The precise number of protesters remains unknown, but strikes occurred in at least 11 different units, likely comprising more than a hundred detainees in total (Morrissey, 2020a). Throughout the same period, strikes multiplied across the immigration detention system on a national scale, leading to fierce battles between the strikers and the authorities attempting to regain control of their facilities. This essay reconstructs the events to prevent their fading from collective memory, and it participates in the efforts to register protests in carceral sites during the first wave of the pandemic, and on a global scale (Hanan, 2021;Pattavina & Palmieri, 2020;Perilous, 2020).
But furthermore, I analyze the events at OMDC to examine the power of these protests, and to identify those factors that determined their strength. Previous strikes in detention were largely smaller affairs, often comprising only the individuals who launched them, or larger groups that could not, however, spark mobilizations across the country. Conversely, the 2020 protests ran through the detention system as a fire. Thus, my leading question is why these protests succeeded where others had failed, and how Covid-19 was responsible for their success. The virus did two things: first, it brought people together by virtue of a common vulnerability; second, it allowed the protest to replicate its pace and expansion, with the hunger strike following the pandemic and moving across the detention network as a contagion.
The link between the strike and the pandemic was acknowledged by the authorities themselves, as they adopted the same strategies of lockdown and quarantine to contain both phenomena. For Immigration and Customs Enforcement (ICE), the hunger strike was threatening due to its ability to travel, not unlike the pandemic. Both spread across facilities as a contagion, and ICE attempted to contain them by isolating the prisoners into their units and cells to prevent the contagion from growing. The same strategy spilled into the courtroom when detainees' advocates brought class action lawsuits to demand collective releases on humanitarian grounds. In court, ICE stressed how people suffered from different levels of vulnerability to the virus, and it attempted to sever the class actions to force the detainees to pursue their own individual cases. Whether in court or in detention, ICE's strategy remained consistent in its attempts to isolate people to prevent them from organizing together.
This article offers a comprehensive analysis of the events to unveil the (bio)political effects of the pandemic across US immigration detention. To account for the virus' role throughout this process, I conceptualize the protest, the detainees, and the virus itself as a force, through a framework that I borrow from the work of Friedrich Nietzsche (1989) and Gilles Deleuze (1983). Importantly, I do not present this essay as an opportunity to apply an overarching theory to a case study. Instead, I develop my theoretical analysis in reference to the empirical problem of constructing a theory that explains the dimension of this protest.
The article's structure is designed to reflect this. I begin with a description of my methods and the sources of my data. I proceed by discussing the conditions that existed at OMDC prior to the pandemic, and I follow with an analysis of the strike that highlights how Covid-19 created favorable conditions for the protest. In the fifth section, I direct the attention to the authorities managing immigration detention and OMDC, and to their strategies to contain the pandemic and the hunger strike at once. Finally, I conclude by discussing the legal litigation that followed the protest, when legal advocates brought a class action to demand the release of the people detained.

Methodology
To describe the events inside OMDC, I pull from my qualitative work and secondary sources, as well as my experience working as a volunteer for various detention advocacy and abolitionist organizations in California. This research project also underwent a rigorous university IRB ethics review, and all my data sources have been fully anonymized for privacy and confidentiality.
Secondary sources for this work include: media coverage of the hunger strike at OMDC by media outlets, and particularly the San Diego Union Tribune; the 2021 review of immigration detention facilities by the California Attorney General (California AG, 2021); and documents associated with the legal case of Alcantara v. Archambeault, which addressed conditions at OMDC during the pandemic. Alcantara was a class action lawsuit brought by immigration detainees at OMDC, and it constitutes the focus of the last section of the article.
As for the primary sources, they include qualitative interviews and mail correspondence with immigration detainees. This article is part of a larger research project on immigration detention that was underway prior to the Covid-19 pandemic. Between October 2019 and March 2020, I collaborated with a detention visitation program in San Diego, and I conducted 15 in-person visitations with individuals who were detained at OMDC. The purpose of the visits was to connect detainees with legal advocacy groups, but also to offer human company and emotional support to those who desired it. During visits, my observations and conversations allowed me to gain a better understanding of the inner workings of the facility. As an activist, I also used visitations to collect anonymous information on alleged abuses, food quality, medical care and other common complaints.
Beside performing these tasks as a volunteer, I utilized the visits as an opportunity to initiate contact, and I explained to my visiters that I was interested in gathering data on the facility for research purposes. I would then initiate mail correspondence with those who were willing, with the understanding that I would have followed up with an interview after their release. 2 However, I was only able to conduct two qualitative interviews due to the Covid outbreak, which effectively shut down the facility and prevented me from recruiting further research participants. Nonetheless, the information uncovered through these interviews is incredibly rich and helps set the backdrop for the protests at OMDC.
The first interview concerned the state of OMDC prior to the pandemic, as the respondent was released before the 2020 outbreak (Personal Interview, 12/8/2020). Conversely, the respondent for the second interview was detained during the spring and summer of 2020, and he actively participated in the hunger strike inside his unit, which I refer to as Pod X (Personal Interview, 9/24/2020). This interview constitutes the main source for the description of the hunger strike as it unfolded in Pod X between April and June 2020.
In addition to the interviews, I utilized mail exchanges with immigration detainees to document conditions inside OMDC both prior to and during the pandemic. Furthermore, between 2019 and 2020 I collaborated with a pen pal program for people in detention, which sought to build a letter archive to document conditions inside OMDC and other facilities. 3 During the pandemic, detainees sent several letters to this organization with the purpose of bringing attention to their situation and to the hunger strikes. I utilized these letters to obtain a more comprehensive understanding of the protests, and as a cross reference to confirm and corroborate the data that I gathered from interviews.
Finally, in the spring and summer of 2020 I volunteered for a national hotline that people in detention could call to obtain support and report abuses. These calls confirmed that the behavior of the authorities inside OMDC was not exclusive to this facility, as detainees in other detention centers reported similar issues.
In total, the article relies on a variety of both primary source material and secondary sources. While the description of the internal dynamics of the hunger strike relies largely on my interview with a former striker, the dates and most of the details have been confirmed by separate letters, media reports, and court documents. All the data from the qualitative work has been anonymized, and I avoid references to individuals' characteristics to prevent any identification of the people that I spoke with.

Agency in detention: political agency and the non-human
In the last 15 years, a vast literature has emerged focusing on the possibility for political resistance in detention, where scholars have conceptualized resistance as an instrument to claim a political identity against a system that deprives its prisoners of their agency and voice (Conlon, 2013;McGregor, 2011). Theoretically, these contributions borrow from philosophers who theorize "the political" as a matter of human agency, and as a status that one reaches through action or language, like Hannah Arendt (1998), Jacques Rancière (1999), and Engin Isin (2008). The first two authors in particular have been very influential in the literature (see Fiske, 2016;Nyers, 2008;Vicaro, 2015). Furthermore, scholars have borrowed from Giorgio Agamben and his theorization of bare life (1998), often using the aforementioned philosophers to find the possibility for politics in bare life (Owens, 2009;Edkins & Pin-Fat, 2005; also Martin et al., 2020, pp. 753-54).
With some relevant exceptions (for example Campesi, 2015;Fischer, 2015;Conlon, 2013), there are three recurring themes in this literature. First, the political is conceptualized as discourse, and it is conceived as an act to be analyzed for its symbolic meaning, or as a performance (Cox & Minahan, 2004;Pellander & Horsti, 2018). Hunger strikes and acts of self-mutilation are treated as form of "speaking through the body" by those who have been deprived of a voice (Puggioni, 2014). Second, the political act is an event that qualifies the actor as a political being, forcing an acknowledgement as such by a human other, and therefore, a dialogic act (see Isin, 2008). Third, these contributions aim at enhancing the detainees' ability to rebel, and to criticize those readings that depict them as passive subjects. For this reason, there is a tendency to think of agency in terms of intentionality and self-determination. While this is not always the case (for example see Abrahamsson & Dányi, 2019), this conceptualization of agency is explicit in certain theoretical sources (Arendt, 1998), and it often constitutes a logical result of the conceptualization of political action in discursive terms.
Such a dialogic framework, however, is less suited to the present analysis, as here I seek to account for the power of a protest more than its discursive meaning. From a dialogic perspective, the act of resistance is only relevant to the extent that it constructs an antagonistic identity, and the non-discursive dynamics that affect its range and strength are of less interest to the analysis. Furthermore, under this framework every act of resistance possesses the same value as a performance or a speech act, and thus, the distinction between individual or collective strikes is of less relevance. Coherently, the political actor that launches the protest is always treated as an individual, and even in the case of a group we would consider it one group, speaking with one voice, because the mutual relations among the members possess a lesser centrality than the antagonistic relation between the group and the carceral system. For this reason, a dialogic framework would fail to identify those material circumstances that affected the range of the 2020 hunger strikes, and which distinguished it from smaller protests that had previously erupted at OMDC or other facilities. Importantly, this is not a critique of the literature, which remains cogent and efficacious to analyze resistance from its own discursive perspective. However, the same framework is less useful for the question guiding the present work.
Instead, here the political agency of the hunger strikers must be conceptualized in less humanistic and intentionalistic terms, so as to focus on those material components that affected their ability to act and the power of their protest. This problem resonates with current debates in migration studies concerning the agency of migrants caught in the politics of border regimes, where scholars have criticized the oversimplified debates that depict migrants as either passive victims of structural factors, or as heroic subjects whose actions are the sole result of their own autonomous decisions (Aru, 2021;Mainwaring, 2016;Squire, 2017). Against this dichotomy, migration scholars have focused on alternative conceptualizations of agency to stress its relational character, where the migrants' ability to act and react against migration policies is produced by the interplay of different forces, and where the constraints to their mobility are not simply an obstacle, but a constitutive factor for agency itself (Aru, 2021;Wheatley & Gomberg-Muñoz, 2016).
This focus on agency as a dynamic relation of power is more suited to the present work because it directs attention to the non-human factors that influence the humans' ability to act. From this perspective, political agency is not a solely human affair, in as much as a process bringing into relation human and non-human beings (see Braidotti, 2013). When looking at the protest through these lenses, the focus is not exclusive to what the hunger strikers were able to communicate, but also on their ability to organize together to do so, and on those non-human factors that helped them reaching their goal.
For this reason, the present analysis begins with identifying those changes that occurred at OMDC once Covid-19 entered the facility. While the beginning of the pandemic preceded the declaration of the strike, it constituted the beginning of a shift that made the strike itself possible. The virus altered conditions at OMDC by introducing a common vulnerability among the population, thus encouraging individuals to associate together politically. Covid-19 did not simply work as a cause of the strike, but as a constitutive component of the political agency of the strikers, thus assuming a biopolitical significance across the events. The very biological characteristics of the virus undermined the preexisting regime, which was designed precisely to prevent the detained population from organizing collectively.
More specifically, the US detention system is organized and managed to prevent political unrest and discourage collective organization. It achieves this by punishing those who attempt to rebel, but also by dividing the population into different statuses and legal classes, which discourage collective efforts. More broadly, the entire immigration machine functions through statuses that regulate the paths of noncitizens in the United States (see Eagly, 2013Eagly, , pp. 1137Eagly, -1139. To move through these paths one transitions between statuses, either above, toward residence and citizenship, or below, toward deportation (Martin, 2005). People in detention are always in a "detainable" status, meaning a legal status that allows the authorities to detain them while their removal proceedings are carried out. 4 But while all detainable, they are not all equal. Their possibilities for release depend on their status, and on their ability to take advantage of the legal pathways that are associated with their legal class (see Asad, 2019;Ryo, 2016). 5 Within such a system, collective organization is not particularly helpful due to the highly varied situation of people, who navigate it within different, vertical paths. To engage in politics inside detention is a disciplinary infraction that can be used against them, as it is treated as proof of dangerousness and lack of discipline. Those who try to organize are punished with solitary confinement, and they are regularly transferred across units so that they don't succeed. Additionally, the population in detention is multinational, and people speak different languages, practice different religions, and come from different cultures. Their treatment also varies depending on nationality, race, gender, and sexuality, and research and detainees' testimonies indicate that conditions of detention are highly unequal depending on these factors (see Franco, Patler, & Reiter, 2020;Reema, 2022). 6 These obstacles do not make collective organization impossible, but they do complicate it, and they require time to overcome. Time however is limited, as people are constantly transferred, deported, or released, therefore maintaining a high turnover inside detention facilities (Hiemstra, 2019). For collective organization to spark within this system, there needs to be an event that can "bend" these vertical statuses and legal paths, so as to expand them horizontally and allow individuals coming together. There needs to be a common condition embracing several people who can consequently act as a group.
We might be tempted to think of Covid-19 as a danger for the detainees, a danger of such intensity to spur them into action. However, immigration detainees are normally facing risks just as great as Covid, if not more. They are all in removal proceedings, and many of them may lose their lives if they are deported. For example, a hunger striker who spoke to me after being deported to his country spent several months in hiding upon his return. When we spoke, he couldn't come out of his hiding for fear that members of a local gang would identify him and kill him. Far from exceptional, these risks are common for immigration detainees, and many of them would rather take their chances with Covid than being deported. This is also true for those who spent long periods of their lives in the United States, possibly since they were children, and for whom deportation signifies the banishment from their families and country (see Zilberg, 2004).
The reason for Covid's disruption is not just the risk in itself, even though it matters, but its horizontal form of expansion. Suddenly, everybody has a common status of vulnerability due to the virus, and to speak of one's case is to speak of them all. Obviously, the risk is not shared equally. People suffer greater or lesser risk depending on their medical condition or their age, but the virus acts as an equalizer nonetheless, leveling differences and presenting a common problem that encourages collective organization.
In this context, the political should not be associated with a choice to act that is taken by a human actor acting independently. Such a humanistic focus 7 risks confusing the scenario that is forcefully enacted by the detention system with a reality that exists a priori, 8 because individuals are only acting alone when the system can successfully separate them and individualize them. Individual actions testify to the strength of the system, which prevents single actors from affecting others into joining them. On the contrary, the power of the protest consists in its breaking through individual differences to spark collective organization, thus defying the boundaries separating each individual actor from others. In this sense, Covid-19 did not simply cause the protest by giving a reason to strike, but it also undermined the preexisting regime by creating the possibility for larger resistance.

The beginning of the protest: the hunger strike as a force
In March 2020, US authorities began to acknowledge the threat posed by Covid-19 to incarcerated individuals, and preventive measures were taken across the country (Federal Bureau of Prisons, 2020). The immigration detention system was no exception, and things inside OMDC began to change. On the 13th, all visitations to detention centers were suspended, and Immigration and Customs Enforcement (ICE), which is the federal agency tasked with managing immigration detention in the United States, initiated transfers in-between facilities to reduce overcrowding. The population at OMDC decreased, also because new admissions into the facility were put on hold. Still, in April the ICE section of the center 9 was detaining around 700 people across separate units (Alcantara v. Archambeault, 2020a, *2). Meanwhile, a sentiment of fear spread across the population. In the month that preceded the strike, detainees began realizing the danger, as authorities did not appear able to stop the virus and neither they seemed to care. They were unable to conduct testing, it was impossible to respect social distancing, and furthermore, as soon as people began to fall sick they were simply transferred to a "medical pod" where they did not receive adequate medical attention (California AG, 2021, pp. 140-142).
On March 30, an employee at the facility tested positive for Covid-19 (Morrissey & Lopez-Villafaña, 2020). In the following days more people began showing symptoms, and by April 15 twenty-seven detainees had tested positive, making OMDC the detention center with the highest number of infections in the country. By the end of April that number had more than tripled, with 98 confirmed cases (Alcantara v. Archambeault, 2020a, *8). Finally, on May 6 OMDC registered the first Covid-19 related death in the national detention system, that of an El Salvadoran man named Carlos Ernesto Escobar Mejía (Morrissey, 2020b). But already by mid-April the medical pod was full, and the situation out of control. It was in this period that detainees across the center decided to go on strike (Rivlin-Nader, 2020).
The first hunger strikes were declared in early April, and by the 17th there were active strikers in at least 11 units of the detention center (Morrissey, 2020a). ICE had already put the units on lockdown in an effort to prevent contagions, and detainees could have no contacts with people outside of their pods. Consequently, the hunger strikes began separately. As discussed, my knowledge of the strike's internal development relies on one of my interviews, and it is specific to the unit where my respondent was detained, which I refer to as Pod X (Personal Interview, 9/24/2020). In early April, detainees in Pod X began meeting in the common area of their unit to debate possible courses of action. These were difficult discussions among people who did not always share a common language, and the meetings required work from interpreters to translate. Rivalries or distances due to cultural reasons had to be set apart, as people acknowledged that this could only work through cooperation. Finally, on April 17 they decided to go on strike. 10 The decision to strike was initially motivated by the specific conditions in Pod X, but the detainees were further encouraged after realizing that hunger strikes were already being declared in other units and elsewhere in the country (O'Connor, 2020). This information could have come from the televised news, or through phone calls and letters from family members, supporters, and lawyers. Knowledge of these developments facilitated the formation of a network encompassing separate units and facilities, and the strikers were attracted by the possibility of joining a growing movement, as this would have resulted in a more powerful protest than they could accomplish individually. In this sense, each striker's decision to join was affected by others who had already done so, and by the awareness that new participants would have encouraged others to follow, both within their own unit and elsewhere.
Human agency in this case does not consist in an inherent ability to act that is possessed by each striker independently. Instead, agency is produced through affection, so that our focus should switch from the strikers themselves to the mutual relations among them (see Pile, 2010;also Clough, 2012). It is their ability to communicate with each other that fosters political agency, just like their ability to infect each other allows the pandemic to expand. This is a process of contagion, where the strike spreads across bodies and facilities thanks to the people's ability to transmit it. Hence its formidable pace, with the protest spreading across the two coasts in a matter of days, just like the pandemic.
When looking at the strike in these terms, it is possible to identify those factors that determine its strength. The strike is a virtual possibility whose power depends on its potential to spread. It possesses a specific temporality and spatiality because it consists not solely of those who are currently striking, but most importantly of their power to encourage others to follow in the immediate future and elsewhere. The authorities who fight it will similarly operate in space and time to prevent it from growing, and to keep it contained to the facilities and cells to which it has already spread. For the authorities, the most pressing issue 7 "Humanism" here does not merely translate as human-centered, but it identifies any philosophical approach that posits the human as a subject that relates to others while maintaining its own integrity and identity as an individual human. 8 Notably, while the literature concerning the political in detention has primarily focused on human actors, there are scholars working in carceral geography who have expanded their focus to animal studies and human/non-human relations inside prison gates (Moran et al., 2021;Morin, 2018). 9 OMDC also detains individuals for the United States Marshal Services in a separate area of the detention center. In this paper, I solely focus on the ICE section of the facility. 10 The strikers would alternate periods of starvation ranging from a few days to a couple of weeks to periods where they would go back to eating. This allowed them to reduce the risk of dying or weakening themselves excessively in the case of an infection.

E. Asoni
is not to stop the strikers from starving, but to act on those material conditions that allow the strike to travel, because those who are currently striking are always more than their sum, as what matters is their potential for growth. The strike is led by the desire to escape detention, but there is a fundamental difference between seeking release through this process or through the pathways that are offered by immigration law. In the latter option, the detainee seeks a transition into a different status that would lead to a release on bond or parole. Borrowing from Deleuze and Guattari (1987), we may describe this legal option as a "reproduction by filiation," where a subject mutates individually through sudden transitions in between static moments. Conversely, the collective strike corresponds to a "reproduction by contagion." This is a process of becoming that disrupts a fixed being by bringing it into relation with others, so to lose its individual subjectivity and become many: "The difference is that contagion, epidemic, involves terms that are entirely heterogeneous: for example, a human being, an animal, and a bacterium, a virus, a molecule, a microorganism" (242).
Simply by gathering, and by cultivating new relations among themselves, the detainees spark a process that makes the label of "detainee" amply insufficient to describe them. They are a multiplicity. The multiplicity is not the sum of the individual strikers, and neither a mass that could be counted and governed on a macro scale (Deleuze & Guattari, 1987, p. 8). Why is this relevant? Because the existing regulations of the detention center, and immigration law itself, can only govern subjects who are either individual "aliens" or populations. By forming a multiplicity, the strikers are becoming ungovernable, as they defy scales. This puts the detention system into a crisis.
Is the multiplicity striking? Perhaps, the problem with this question is already evident. The strike constitutes a process involving people coming and remaining together to starve. To interpret it as the act of an actor would be just as absurd as considering the pandemic the act of a subject that is Covid-19. Quoting Nietzsche, that would be akin to considering the lightning as producing the flash in a storm. The two are not separable: 'there is no "being" behind doing, effecting, becoming [emphasis added]; "the doer" is merely a fiction added to the deed-the deed is everything' (Nietzsche, 1989, p. 45; also see Esposito, 2015, pp. 182-184;Feldman, 1991, pp. 3-4). The pandemic is a process whose existence is conditioned on the bodies that spread it, and on a space that is sufficiently open to allow it (see van Loon, 2005). The mobilization of strikers resembles it because it is a becoming-political of the pandemic itself, and furthermore a becoming-pandemic of the humans detained, who borrow the virus' strength.
I conceptualize the strike as a force to highlight its ability to overcome the individualities of the single participants. This ability is the source of its power, and the protest is a biopolitical force whose strength is conditional on how many people it can affect into joining. The affection is only possible when people get close, when they share, when they touch each other (Arenas, 2015). This political pandemic has a greater range than the detention center, it travels within institutions as soon as somebody brings it in through a phone call, or through a segment on the news. It bends space, shrinking distances and moving fast.

To break up the strike: counterstrategies within the detention center
To interpret the strike as a force is not only useful under a theoretical perspective, but also necessary to grasp the behavior of the authorities that reacted against it. For those managing detention, and particularly ICE, the strike and the pandemic posed a significant threat due to their ability to spread at a pace that was impossible to govern. ICE manages its detention network as a circuit of different facilities, where each site functions as a node that is interconnected with others (see Gill et al., 2018;also Brooks & Best, 2021). All sorts of goods, people, capital, and information circulate across the routes connecting each node, but ICE must be able to recognize "bad" circulations and lockdown the nodes that are producing them (Tazzioli, 2020;Foucault, 2009, pp. 325-326). What rendered the pandemic so threatening was ICE's inability to stop it from circulating and infecting the entire network.
Covid-19 connects people and owes its power to the fact that people are connected. ICE's power comes from the opposite: it relies on its ability to separate people to control them. In this sense, the hunger strike-pandemic and ICE owe their strength to opposite spatial organizations. The conflict between the two unfolds across this trajectory, and it constitutes a struggle between two forces who battle for their own existence. This is not an overstatement. Had the detainees been able to obtain mass releases, ICE would have been left with empty detention centers, which are not economically manageable. This would cause the private contractors to go bankrupt, and worse, it could reveal that immigration can be managed without relying on a detention system of this magnitude. We should appreciate how threatening this scenario is for ICE, particularly at a time when calls for its abolition have become mainstream (Ember & Herndon, 2018).
In the previous sections, I began the analysis by illustrating the change in human and political relations inside OMDC before and after the pandemic. Here, I do the same by describing how ICE changed its approach toward political unrest once the protest began.
Importantly, ICE does not manage OMDC by itself. It is a privately run facility owned by CoreCivic, which is a private contractor operating in the detention and prison system in the United States (Hiemstra & Conlon, 2017). ICE charges CoreCivic with the day-to-day management of the center, which includes disciplining violations. In my experience talking with detainees at OMDC, I repeatedly heard that the two organizations are not always on the same page, and that there are often tensions among the two. In general, ICE agents appear to be wary of CoreCivic's officers unnecessarily escalating situations, as this may result in unwanted public or legal attention to the detention center (Personal Interview, 12/8/2020). It is most likely for this reason that ICE takes care of hunger strikes directly. Whenever a detainee stops eating, CoreCivic is under strict instructions to report to ICE, and not to take matters into its own hands (US Immigration and Customs Enforcement, 2019, pp. 108-109). Once the agency is informed, they will generally wait 48 h before sending an agent to talk with the hunger striker. During my visits, and prior to the pandemic, I met two men who underwent this procedure on separate occasions. In both cases, an ICE agent visited the men and kindly and emphatically convinced them that stopping the strike was in their best interest. Both men gave up the protest after the meeting, but if they had not, ICE would have sent them to solitary confinement and eventually obtained a court order to force feed them (Stevens, 2019).
During the pandemic, however, ICE did not attempt to reach anybody individually, and the agency made no effort to negotiate with detainees. The change in strategy corresponded to the acknowledgement that the collective strike constituted a different threat. In this case, the most immediate goal was not to break the individual resilience of the single strikers, but to prevent the strike from traveling across the circuit. Similarly, the fear of the pandemic was not due to the strength of each single infection, but to the possibility of having too many infections at once, which would have rendered the situation unmanageable. That this was the case can be inferred from the agency's behavior during the pandemic.
First, in mid-March ICE began conducting mass transfers across facilities. More than a hundred detainees left OMDC on March 15, and they were transferred to the detention center of Joe Corley, in Texas. Obviously, the transfers were not advisable from a medical perspective as they could have spread the contagion, but it eventually became clear that ICE had a different goal: they wanted to reduce overcrowding and empty out units that could be turned into medical pods to detain the sick. In fact, one of the triggers to the strike in Pod X was the notice of an incoming transfer to a separate unit that had already registered positive cases. Clearly, the logic of the transfer was not to prevent contagions, but to facilitate the management of the incoming outbreaks by reserving an entire unit to the custody of sick prisoners.
Second, all units were put on lockdown, thus preventing people in different pods from visiting the same common areas where they could have spread the contagion. However, to lockdown the pods had an additional advantage: it prevented people from contacting individuals outside of their units. But just like the pandemic, the hunger strike could not be stopped by mere walls. Detainees in Pod X heard about protests in other units through the same channels that conveyed news of developments in other facilities. The virtual channels of communication permitted a horizontal expansion that connected the various nodes across the circuit and reduced isolation and distances (Marshall et al., 2017). Therefore, ICE adjusted its strategy to the new threat.
The adjustments were designed to close the channels of communication in between the different nodes, so as to increase their isolation. The most delicate were phone calls and TV news, because the information would flow two ways. The protests and lawsuits outside would rely on information coming from the inside, while the detainees would rely on information from supporters and the media to link their struggles with others. Thus, to break up this connection could isolate people outside as well, so that their protest and the strike would not become one. This was achieved through retaliation against detainees speaking to the media or to activist groups. In June, ICE went as far as formally prohibiting any phone or letter contact with the activist group Otay Mesa Detention Resistance (Morrissey, 2020c). This strategy was applied nationwide, and supporters countered it by holding demonstrations in front of detention centers, honking horns and chanting so that the detainees could hear them (Kutz, 2020).
Furthermore, people needed to be stopped from organizing inside their pods as well. To do this, CoreCivic established a full lockdown for the strikers, so that they could not leave their cells and access the common area of their units. This was particularly effective in Pod X, because the people there were classified as mid or high risk, and their cells had only two beds for security reasons. Additionally, CoreCivic sent some strikers to solitary confinement in the segregation unit, where they were fully isolated from the rest and without access to phone calls.
When looking at this strategy, I direct attention to how its goal is not to bring the number of strikers or contagions down to zero. Rather than curing the sick or forcing individual strikers to eat, ICE focused on preventing the strike and the pandemic from growing to a point where they would become unmanageable. This design is coherent with ICE's structural characteristics as a force. ICE did not fear the resilience of individual strikers or the gravity of the single infections. Instead, the threat stemmed from each person's ability to transform into a node that could facilitate the "bad" circulation across the circuit. The strategy targeted this very ability by reaching a certain level of isolation within and between facilities, thus acting on the connections between individuals and not the individual themselves (also see MacKinnon, 2011).

Representing multiplicities: the class action and its commonality requirement
In this last section, I consider how the protest spilled into the courtroom when detainees' advocates fought for the release of the detained population. The battle in court resembled the conflict inside detention. The advocates pursued class actions lawsuits under the argument that detainees suffered from a generalized risk of infection, thus presenting a common interest which required a collective release as a unified solution. Conversely, ICE stressed how each person suffered from different levels of risk due to their age or medical conditions, thus urging the court to sever their cases and force them to pursue their own individual petitions. The conflict unfolded across the same trajectory discussed above, as it became a matter of establishing the nature of the detainees as either a multiplicity or a sum of individuals with separate interests.
The legal battle is yet another field across the conflict inside detention, and one that is crucial for its outcome. It should not be treated as external to the strike, as it affected its dimension and range like the other factors discussed above. My analysis here resonates with the work of legal geographers, and particularly with those scholars who conceive of law and space as mutually constitutive of each other (for example Bennett & Layard, 2015;Braverman et al., 2014;Delaney, 2010). When detainees at OMDC first gathered to decide whether to strike, they knew that their demands had to be shaped in a form that could be legally feasible. To demand a collective release was an exceptional request unlikely to succeed, and they needed to balance this request with the break from the existing regulations implied by their demand. A letter sent to the local media on April 17 exemplifies how they worked around the problem.

UNITED IMMIGRANTS OF THE OTAY MESA DETENTION CENTER
On April 17 we joined the hunger strike of our brothers in other units of this detention center [ …] we fear getting infected by Covid, as this detention center already has the highest number of contagions in the entire American nation.
Because of this, we ask the highest boss of ICE, the Attorney General, the Inspector General, and all the Federal Courts of the District and the Supreme Court to please release us on bond or parole so that we can continue our immigration proceedings with our families, because if we keep staying here many lives will be lost because of  we are human beings that are asking for an opportunity in this great nation.
We are fathers, sons, brothers, and grandfathers.
Please, we ask you that you let us stay with our families in these difficult times.
Thank you.

GOD BLESS AMERICA [in English in the original] 11
Someone who is unfamiliar with US immigration law could confuse this text for a demand that is solely written to convey a sense of humanity and dignity against an oppressive system. But this is a document authored by people familiar with the regulations of detention, and it shows. Specifically, the United Immigrants are using certain characterizations that are meaningful under immigration law. First, they claim to have family members in the United States who are suffering from their detention. Second, they show devotion to the country. Both arguments carry weight in bond proceedings, where a detainee can ask to be released on bail, and they are also factors to consider when granting a cancellation of removal in an Immigration Court. 12 Third, they express the desire to continue their immigration proceedings at home with their families. This means they have no intention to flee, and that they are willing to remain at a precise domicile while their proceedings unfold. All this is doable under ICE's regulations.
In other words, the strikers are not fighting outside the law, and much less against it (see Bailey, 2009;Reiter, 2014). The United Immigrants are well embedded in the law, and their possibilities for constructing a discursive identity are partly determined by the existing legal system. They are not writing on a blank page, as they must follow certain rules for their text to be considered. This apparently diminishes the insurrectional character of their protest, and that is precisely their intention.
To bring this claim into court is a lawyer's job, and the class action is the best instrument to do so. A class action lawsuit allows the petitioners to present themselves as a group, as an exception from the normal rules 11 The translation from Spanish is mine. This letter was sent to an activist group that I worked with, but the authors also sent it to the San Diego Union Tribune, which published a few quotes (Morrissey, 2020a). 12 Cancellation of removal is enshrined by INA §240 A(b)1. regulating standing in court, which restricts it to single individuals acting on their own exclusive interest. At the national level, the first class action lawsuit concerning immigration detainees during the pandemic was Fraihat v. ICE, which was filed in April 2020. The plaintiffs in Fraihat were detained in separate facilities, and they certified two classes that included every detainee with specific risk factors that made them vulnerable to a Covid-19 infection. 13 They argued that medical conditions inside detention were so poor that to be detained in those conditions raised to the level of punishment, thus violating their constitutional rights. Therefore, they demanded the court to issue a preliminary injunction forcing ICE to find immediate solutions.
Notably, the plaintiffs in Fraihat did not ask for a court-mandated release because that would not have been realistic. From a legal standpoint, ICE is entitled to a high level of discretion in the management of detention, and for a court to order a release there must be a clear assessment that any alternative solution would fail to prevent a violation of the plaintiffs' rights. Thus, a release could be demanded, but only for a class that is narrow enough to justify it. In this case, and as argued by ICE, the classes included people who were detained for different reasons, with different levels of vulnerability, and in separate detention centers. If the class is too broad, as in this case, the plaintiffs may only demand a solution that is as general as the class itself. Thus, the court ordered ICE to run proper medical screening, and to consider releasing the detainees who were particularly vulnerable, while leaving ample discretion in this matter. 14 After Fraihat, the American Civil Liberties Union (ACLU) filed a class action lawsuit for detainees at OMDC in May 2020, this time asking for a release. In this case, the problem was how to balance two contrasting goals: first, to certify a class as inclusive as possible to represent a larger share of the detained population; second, to certify a class that was narrow enough to justify the request for release. As expected, ICE pushed against this process by arguing that the class was too large to show a sufficiently common interest. ICE argued that the class members possessed different levels of medical risk to an infection, and that releases needed to be assessed individually, and according to the specific interests of each person.
Here, there is a glaring continuity between ICE's behavior inside detention and its legal arguments in court. In both instances, ICE aimed at the same goal: to break the multiplicity apart. But while its strategies converged in their purpose, they relied on opposite descriptions of the pandemic to achieve it. In detention, ICE treated the pandemic as a force spreading across the facility. When looking at the contagion through these lenses, what is significant is the virus' ability to travel across bodies, with each body becoming a node that allows the contagion to accelerate and grow. This ability to circulate and travel can be quantified, and it corresponds to the contagion, or infection rate. Conversely, in the courtroom ICE focused on the gravity of each single infection onto the bodies that were being affected. This is another aspect of the contagion, which expresses the gravity for the individual who falls sick, and it can be quantified as well. Based on their medical conditions, people may be assigned a value that describes the risk of long-term consequences, or even death. This value individualizes the humanvirus encounter, as it predicts the changes that the virus provokes to a specific body, instead of the body's ability to spread the contagion.
By describing the pandemic through the individual risk factor, the agency was able to utilize the pandemic itself as an instrument to classify the detainees according to their vulnerability. Thus, while inside detention the virus operated as an equalizer, in court it had the opposite effect. To pursue its class action, the ACLU was forced to construct a smaller class than the entire population, and this only included detainees who were 45 years or older, or with specific medical conditions. On May 1st, the court found a compromise between their request and ICE's position, and it established that the age requirement must be raised to 60 years (Alcantara v. Archambeault, 2020a, *7). This class comprised 134 individuals, 15 but ICE refused to release 34 of its members because of their criminal records (some of them were in Pod X). The ACLU filed for a new court order, but they were eventually unsuccessful, because after a month from the previous decision the court established that the conditions in the center had changed enough to guarantee the minimum safety, so that ICE did not have to release aliens "with a demonstrated criminal history" (Alcantara v. Archambeault, 2020b, *1078).
As the case progressed, the detainees were progressively excluded from the struggle for their release. While their lawyers fought for making the classes as inclusive as possible, the necessary compromises prevented them from representing too many people at once. Through this process, the strikers were depicted as a sum of agentless bodies, and their requests and their actions ignored. Instead, they could only access the court as vulnerable subjects, and that is, biomedical objects whose interest only extended to the biological preservation of their individual health (see Dadusc & Mudu, 2020, pp. 8-9;Feldman & Ticktin, 2010, pp. 89-127). Furthermore, the juridical representation of the events ignored the political significance of the contagion, which was depicted as a solely medical and personal affair, thus obscuring the solidarity and unity that had been triggered by the pandemic. On the contrary, the possibility for demanding the release of some was justified precisely by the exclusion of others, because the more people were excluded the greater the exceptionality of the class.
In the end, the medical classification replicated the same process of separation and isolation that ICE carried inside OMDC. There, the case had tremendous effects. People began being released, but many were left behind. Those who remained inside had trouble maintaining the strike in the face of these releases, which confirmed that their fate was going to be decided by others (Personal Interview, 9/24/2020). Ultimately, the releases, the hunger, and the pressure by CoreCivic and ICE led to a sense of hopelessness and resignation. People began abandoning the strike in May, and the last strikers in Pod X gave up in early June.

Conclusions: the politics of Covid-19
In reading this article, people familiar with immigration detention in the United States may have noticed how the 2020 protests partly resembled previous episodes of resistance in detention. This was not the first protest across detention centers, and the authorities' reaction was not surprising, as it constituted an adjustment that did not entirely break from prior strategies to quell unrest. And yet, the speed at which this protest expanded, its capacity for involving people of different nationalities and languages, and ultimately its magnitude, which reverberated outside detention as well, calls for a greater attention to Covid-19 as something more than a deadly crisis.
The article analyzed the hunger strike to account for its power, which I have linked to the pandemic itself. I directed attention to the coming together of human and non-human beings as the locus of political agency, an agency that is more than human, and which bore enough strength to undermine ICE's ability to quell resistance. The resulting conflict was fought over the people's ability to connect with each other, with ICE attempting to sever the links between individuals to prevent the pandemic and the strike from spreading further. My theoretical framework was designed to account for this phenomenon. I utilized the 13 The first class included individuals possessing at least one of the "risk factors" that the Center for Disease Control had identified as possible vulnerabilities to Covid-19. The second class included detainees who possessed specific disabilities that placed them at high risk in the case of an infection. 14 More specifically, after Fraihat detainees could file for humanitarian parole to ask for release due to their medical conditions. However, this was a discretionary procedure and while ICE was bound to consider the parole applications there was no obligation on the agency to release the applicants. 15 One of the 134 class members was Carlos Escobar-Mejía, who died on May 3 before he could be released (Alcantara v. Archambeault, 2020b, *1076).
Nietzschean and Deleuzian concept of force to show how the power of the protest lay in its potential for growth, and thus, on the strike's ability to travel and spread through detention as a contagion, just like the pandemic.
This reading of the political does not undermine the sacrifice and efforts of the humans who participated in the strike. Instead, my analysis clarifies that the possibility for political resistance is dependent on specific material conditions. Political agency is here understood as a virtual possibility, which can strengthen when conditions are ripe. Far from merely expressing the autonomous decision of a subject, the strike blurred the boundary between the actor and the act, the human and the non-human, the political and the biological, thus resulting in a biopolitical force that turned Covid-19 into an instrument to attack the detention system from the ground up.
As a final note, my purpose with this essay was twofold. Firstly, to demonstrate how a non-humanistic reading of the political can wield insights when considering the strength of a protest. Secondly, to offer a detailed account of one instance in the global moment of insurrection in carceral sites at the beginning of the Covid-19 pandemic. At that time, the protests were largely disregarded by the public debate, and this damaged the ability to account for what happened and register it as an historic event. But while outside society did not pay attention, incarcerated individuals all over the world gave rise to spectacular protests, and carceral sites witnessed a mobilization of prisoners like few others in recent memory. In many ways, this was a flame that was smothered quickly. However, it remains an event of enormous significance, which testified to prisoners' ability to rebel and organize collectively. The 2020 protests are part of the history of the pandemic itself, and they should be remembered as such in present and future efforts to come to terms with the consequences of Covid-19.