Idleness as work? How public defenders do their job by waiting

How can public defenders provide adequate legal representation when they are underbudgeted and overworked? Mexican public defenders represent 75-90% of those facing a judicial process (either through the entire process or at some points of it) and receive around one tenth of the budget allocated for their counterparts, prosecutors or fiscalías. Drawing from fieldwork conducted in Oaxaca city, Mexicali, and Tijuana, I find that the public defenders spend a considerable amount of time waiting: waiting in the office for their next case to be assigned, waiting to receive a file, or waiting for the judge to show up in court. Although superficially this waiting can be interpreted as idleness, I identify three ways in which defenders do their jobs during their down-time. First, they interchange key information about specific cases with other operators of the system and families and acquaintances of the accused, thus building legal strategies. Second, down time allows them to socialize with agents of the fiscalía, cops, and other courthouse workers, a crucial activity for the establishment of cordial interagency relations. Finally, they also use this down time to do what I call “forced encroachment,” meaning the undertaking of activities—such as professionalization of legal translators—that are not necessarily their responsibility but that are nevertheless crucial for the jobs. Waiting, in a way, allows public defenders to provide a legal defense in the absence of time and money to investigate, thus keeping the criminal justice system running.


Introduction
In 2008, a documentary film called Presunto Culpable (Presumed Guilty) narrated the case of a man who was accused, tried, and convicted of murder in Mexico city. The documentary showed the pathologies of the Mexican criminal justice system, where cases often rely purely on witness testimony, charges were easily fabricated, and lengthy, bureaucratic processes enabled the constant abuses suffered by those unlucky enough to encounter it. The documentary was the last of a series of efforts to visibilize and reform the Mexican justice system, and indeed, a few months after its release, the Congress was able to pass an ambitious reform that changed the system from written to adversarial. Among the changes implemented, the reform mandated the right to a quality defense through the creation of the Public Defender's Office (PDO). By providing free representation, the reform aimed at protecting the due process of those accused, thus avoiding the egregious abuses that had characterized the previous system. Despite the promising reform, however, Mexico still faces a carceral crisis: recent reports suggest that over 40% of prisons in Mexico are overcrowded ("Población en cárceles crece a ritmo récord," 2020) and violations to due process are still ubiquitous (Villalobos, 2017). Furthermore, PDOs are overworked and underfunded, taking upwards of 90% of the cases but receiving only a tenth of the funding enjoyed by their counterparts at the prosecutor's office. How can we make sense of the work performed by public defenders in such a strained system? This paper suggests one way in which public defenders operate even under dire circumstances. By looking at the day-to-day work conducted by public defenders in Mexico, I argue that the same conditions that enable the crisis of the criminal justice system-insufficient funding, overwork, and the neglect for proper defense council-also carve spaces for public defenders to fulfil their responsibilities. I develop this argument in three steps. First, I compare the organizational and material conditions of the states' Public Defenders' Office (PDO) and prosecutor's offices (fiscalías), looking at their place in the governmental organization, their assigned budgets, and the staffing of their offices. Despite the constitutional reform that (on paper) named defenders as equal adversaries to the fiscales, my comparison shows that PDOs are at an organizational and material disadvantage vis-à-vis their counterparts.
Secondly, I draw from fieldwork in the PDOs in Baja California and Oaxaca and find that this lack of resources-as well as the Fiscalías superior status in practice-does not create a frantic work pace for defenders but, paradoxically, results in a lot of down-time. That is, despite the workload, PDs in the three offices I visited spend a lot of time waiting: they wait in the office for their next case to be assigned, to receive a file, or for a hearing to start.
Although this waiting could be interpreted as idleness or incompetence, the third part of my argument proposes that waiting enables PDs to do at least part of their jobs in three ways. First, it allows defenders to provide legal representation, for example communicating with family members, or by covering for private councils who failed to show up to a hearing. Second, down time allows PDs to Idleness as work? How public defenders do their job by waiting socialize and build networks with workers of the prosecutor's office and courthouse, or agents of the investigative police, interchanging key information about specific cases. Thirdly, PDs also use some of their wait time doing what I call "forced encroachment," undertaking activities related to legal representation, but that are not their specific responsibility. I illustrate this last point by looking at how the PD office in Oaxaca, a state where 12% of the population does not speak Spanish (INEGI, 2020c), provides translation services for indigenous people.
I conclude by discussing how the actions of bureaucrats-in this case, public defenders-can shed light on the contradictions of the criminal justice system. Accounts of modern punishment note the turn towards a codified legal system that decreases uncertainty for those accused, including uncertainty about timing of processes and imprisonment (Foucault, 1995), and the importance of controlling time as a disciplining mechanism in industrial societies (Thompson, 1967). Yet, much of this literature has focused on organizing the time of those being disciplined, and not on those bureaucrats enforcing the disciplinary mechanisms. In this sense, the 2008 Mexican Constitutional reform-which aimed at (ostensibly) modernizing the Mexican criminal justice system-did not create a clockwork-like efficient machinery, for prisoners nor for bureaucrats. Instead, it allows for idle time that is unaccounted for but that, nevertheless, is used to keep the criminal justice running.

Public defenders in Mexico
The foundational 1917 Mexican Constitution stated the right to a defense but the fulfilment of this right has changed along the years. Originally, the Constitution stated that the defense could be fulfilled by the accused himself, a lawyer, or a "trusted person" (defensor de oficio). 1 In short, representation could be provided by someone with no legal training, affecting those who could not afford to hire a private lawyer. Furthermore, in an authoritarian context such as pre-2000 Mexico, this arrangement was taken advantage of by the regime: research shows that it was common practice to write up anyone who was accompanying the accused as a trusted person (friends or family members). 2 In this way, the state fulfilled this right without effectively allowing those in the system to have adequate legal representation (Magaloni Kerpel & Ibarra Olguín, 2008;Pásara, 2003).
To address this inequality, the Federal Institute of Public Defenders was created in 1998. This institute would provide legal representation and legal advice through asesores jurídicos or legal counselors that would advise citizens under certain conditions (for example, in instances where citizens needed help to exercise their rights). 3 Yet, this particular office was bound to fail from the beginning: while the law granted the Institute "technical and operational independence," it was actually created as an organ of the judiciary (Diario Oficial de la Federación, 1998a). That is, public defenders, their evaluation, career paths, and budget, depended directly on the judiciary. As such, it is no surprise that PDs themselves often characterize the role of the public defender previous to the 2008 reform as "paper pushers" to keep the courthouses running. 4 Furthermore, the institute created in 1998 lacked enough material resources. Given that most of the crimes actually fall under the state's jurisdiction, the law left the majority of those who encountered the criminal justice system without guarantee of legal representation. Even more importantly, the particular inadequacy mentioned above-allowing someone with no knowledge or education in the law to provide legal defense-remained. Empirically, defendants reported that their right to counsel had not been fulfilled: in a survey conducted in Mexican prisons in the early 2000s, approximately 70% of those surveyed reported they had not had access to lawyers while waiting for indictments (Magaloni Kerpel & Ibarra Olguín, 2008).
By the mid-2000s, it was clear that the criminal justice system in Mexico faced a crisis. Although most of this crisis was exemplified by the rampant impunity-less than 4%of reported crimes had resulted in anyone being brought to justice and less than 1% of crimes had resulted in someone being sentenced 5 -the lack of technical expertise by investigative police and the lack of affordable and effective defense are also mentioned as key points to be solved. 6 Partly due to these shortcomings, individuals charged with a crime often faced incredibly lengthy criminal processes: although the government does not (and has not) kept records on how long a process takes, the proportion of incarcerated people who are in preventative prison (charged but awaiting trial) has been high, showing that the criminal justice system cannot cope with those encountering it. 7 Delays mean that the Constitutional right of the accused to a timely trial and verdict are often violated.
Finally, in 2008, a comprehensive reform changed the criminal justice system and with it, the role and organization of the public defenders. 8 Most importantly, the reform changed the penal justice system from inquisitorial, where courts actively play a role in the investigation, to adversarial, where courts are an impartial arbiter between the prosecution and the defense. 9 The reform also introduced the possibility 1 3 Idleness as work? How public defenders do their job by waiting of alternative proceedings, so victims (and prosecutors) could seek non-punitive arrangements (monetary compensation, for example). These changes meant that defenders, either public or private, were now one of the three main actors of any criminal proceeding, and were now (on paper at least) an active counterpart of the prosecutor in the courtroom. 10 In addition, the reform also introduced presumption of innocence as a constitutional guarantee and emphasized the importance of due process, directly addressing the abusive conditions frequently encountered in the system.
But probably the most important change that impacted directly public defenders came from the introduction of the right to an adequate legal defense by a lawyer, 11 guaranteed by both the federal and state governments through the establishment of a "quality" office of public defenders or defensorías. In this way, those encountering the criminal justice system would be guaranteed a professional defense, thus ending the practice of the simulated legal defense. Of course, the constitutional reform, including establishment of the defensorías, could not be implemented overnight. Foreseeing this, legislators had stipulated that the new criminal justice system should be implemented by 2016. 12 Research on the new criminal justice system is scarce, partly because of its novelty and partly because of the lack of access to trustworthy data. Studies seem to suggest that the new system has changed some practices in the courts: judges are considered now to be more present and attentive to the hearings (a change attributed to the fact that hearings are now recorded), judicial decisions are better understood by those accused, and processes are faster because they allow non-punitive solutions (World Justice Project & Abogados con Cámara, 2018). In addition, those processed under the new system are less likely to report being tortured, suggesting that the new system indeed improves the due process (Silva-Forné & Padilla-Oñate, 2020).
Yet, the empirical literature 13 on public defenders under this new system is still quite new. 14 Suárez Ávila et al. (2018) study Mexican PDOs to evaluate the implementation of the professional career service, and conclude that the offices are training their lawyers, but that their careers lack the security and monetary payment comparable to their counterparts at the prosecutor's office. Ang and Blajer de la Garza (2020) find that PDOs decrease reported violations to due process but that this effect is even greater for people who reported having been defended by a private lawyer.

3
To make sense of this puzzling finding, the authors suggest that perhaps the establishment of a high-quality PDO shape the interactions that all accused have with prosecutors, police, and judges, regardless of whether they had private or public representation.
This paper seeks to contribute to this nascent literature by theorizing how public defenders work in the new criminal justice system. Empirically, the paper relies on fieldwork 15 that included 26 semi-structured interviews conducted in three Public Defender Offices: two in the state of Baja California (offices of Mexicali and Tijuana) and one in the city of Oaxaca. The case selection follows a most-different research design (Anckar, 2008;Slater & Ziblatt, 2009), where cases differ on values of possible alternative explanations but where the outcome of interest, in this case, the similar dynamics found in the overworked and underfunded Public Defender Offices, remains constant. 16 Specifically, Oaxaca and Baja California have different political trajectories-Baja California was governed by the opposition conservative party PAN from 1989 to 2019, whereas Oaxaca has mostly elected executives from former authoritarian catch-all party PRI 17 -meaning that the ideological leaning of governors or the experience of states with democratic alternation cannot explain the similarity between the offices.
Although both states are similar in population size (Oaxaca has a population of 4,132,148 and Baja California of 3,769,020), inhabitants of these states differ in socioeconomic characteristics: half of Oaxaca's population (51%) lives in rural areas, and one fifth lives in extreme poverty whereas Baja California's population is urban (94%) and only 1.5% lives in extreme poverty. 18 The population of these states also encounter different carceral arrangements: in 2020, Baja California reported space to house 16,500 inmates (76% occupancy rate) whereas Oaxaca reported space for only 4,000 inmates (91.5% occupied). 19 Baja California has the highest rate of incarcerated population in the country-363 inmates per 100,000 inhabitants-whereas Oaxaca is the 26th out of 32 states with 79 inmates per 100,000 inhabitants. Again, these differences show that the similarities in dynamics cannot be accounted for by population at large or carceral capacity.
In both cases, I contacted the directors of the PDOs via email, introducing myself and asking for permission to visit the offices in order to understand the administrative and bureaucratic mechanism that enabled the protection of vulnerable populations. Upon my visit, I first met with the directors of the PDOs to familiarize myself with the day-to-day allocation of tasks, and to strategize how to approach workers with minimal disruption. I approached the defenders in the office, and asked for 15 I visited the Mexicali PD office and courthouses from July 1st to July 3 rd , 2019, and Tijuana on July 4th and 5 th, 2019. My visit to the Oaxaca office was from December 16 to December 19, 2019. IRB number for this project: IRB19-0843. 16 As Anckar argues, this approach is well-suited for small-N studies (such as this one) where keeping alternative factors constant is impossible due to the sample size. 17 From 2010 to 2016, the Oaxacan governor was from party Movimiento Ciudadano, but he had been priísta. 18 All statistics were based on the 2020 Census (INEGI, 2021b an anonymized interview about their work at the PDO 20 and provided them with printed protocols of data privacy and my contact information. In total, I conducted 26 semi-structured interviews across the three offices: 15 public defenders, 6 PDO workers that had some supervisory capacity, and 5 support staff (1 administrative assistant, 2 drivers, and 2 social workers), and only one person declined the interview. In each case, I began by asking for a description of the work they performed at the PDO, how long they had been working there, and why or how did they decide to work there. I also inquired about the work environment within the PDO and with other people they might be in contact with because of their job, and asked them what they identified were the main obstacles for doing their job. The interviews varied in length, ranging from 10 min to an hour depending on how busy they were-some defenders were only available to talk to me between one appointment and the next, whereas others asked me to accompany them to grab lunch, which resulted in longer conversations. Besides the interviews, I accompanied defenders in their daily tasks whenever possible, for example by attending public courthouse proceedings. To protect the well-being and anonymity of my respondents, their answers were not recorded. I rely instead on my extensive note taking during the interviews.
Finally, some of the statements I discuss here are from the event "Feria x la Justicia," an event organized by USAID that brought together several criminal justice workers, policymakers, scholars, and politicians with the purpose of discussing the new criminal justice system. 21 The event was open to the public, and the comments reproduced here have also been anonymized.

The least favorite child
In one of my interviews, a public defender said that in Mexico "the fiscalía (prosecutor) and the defensoría (defender's office) are both children of the same father, but the fiscalía is el consentido (the favorite one)." 22 I find this metaphor illustrative of the dynamics of the adversarial criminal justice system. The 2008 reform placed the defender on equal footing than the prosecutor, but the former is not on par with the latter either organizationally-meaning that their place in the criminal justice system is subordinated to politicians in a way that the prosecutors' are not-or in terms of budget or resources. To be clear, this inequality does not mean that defenders are subordinates of fiscalías, nor that they view themselves as such. Defenders see themselves as part of the criminal justice system, often referring to themselves as "operators of the system" (operadores del sistema), a term also used by prosecutors, judges and court secretaries 23 to talk about themselves and about other workers of the criminal justice system. Yet, defenders were also constantly distancing themselves from the other operators, for example by claiming that "outside of the courtroom we might be friends, but inside of the courtroom we can really fight." 24 Instead, being "the least favorite child" means that defenders have less access to resources than their counterparts, an important feature of the context in which defenders do their constitutionally mandated job.
To illustrate organizational inequality, take the example of the two offices I visited. In Oaxaca, the PDO is a "public and decentralized organism associated with the Legal Counsel Office," 25 whereas the Fiscalía is an autonomous organ (Constitución Política Del Estado Libre y Soberano de Oaxaca, 1922) Both of these figures grant budgetary and technical autonomy 26 but not political autonomy: the Defensoría's budget and internal rules are made and enforced by a governing body (junta de gobierno) headed by the chief of the Legal Counsel Office, a position filled at will by the state's Governor. The director of the Defensoría-a position also filled at will by the Executive-is part of the governing body and has a voice, but cannot vote. 27 In contrast, the Fiscalía is headed by the Fiscal General, a position jointly filled by the executive and the legislative. The position of Fiscal General lasts for 7 years and is only removable under certain conditions.
In Baja California, the organizational inequality is even more pronounced: the Defensoría is a directorate, meaning that they directly depend on the Ministry of Government. It enjoys no autonomy of budget or operation, even on paper. 28 The Fiscalía in Baja California, similar to Oaxaca, is an autonomous body headed by the Fiscal General who is nominated by the Executive and approved by the legislature, and whose tenure lasts for 6 years. In this way, Defensorías are subordinated to the Executive, whereas the Fiscalías enjoy more autonomy.
In addition, Defensorías also receive less budget. Figure 1 shows the exercised budget in 2019 by each Defensoría and Fiscalía 29 in Mexican pesos per capita. In each case, the Fiscalía exercised a considerably larger budget than its Defensoría counterpart: in states like Guanajuato or San Luis Potosí, Fiscalías received around 7.5 times more money than PDs, but in cases like Tamaulipas or Veracruz, they received 60 times more than Defensorías. At the extreme, Hidalgo's Fiscalía's received 350 pesos for every peso given to the Defensoría.

3
Idleness as work? How public defenders do their job by waiting Fiscalías and Defensorías are equally mismatched when it comes to staffing. Figure 2 shows the total number of workers 30 in each of these offices per 100,000 inhabitants in 2019. The inequality is, again, striking: Oaxaca has 9.2 Fiscalía employees per PDO employee-not far from Sinaloa, the state with the lowest ratio of Fiscalía-to-PDO workers (6.8)-and in Baja California this ratio is 17.8. The worst imbalance takes place in Veracruz where there are 40 Fiscalía employees per PDO employee.
The inequality between these two offices is striking even after considering that the public defenders' office is not the adversary in every single case or hearing, but only in those where the defendant cannot-or does not want to-secure private counsel. Yet, the data shows that a majority of those in prison rely on public defenders: a 2016 survey showed that nationally, 82.3% of individuals that had already been sentenced had relied on a public defender at some point in their process . Therefore, even accounting for cases overseen by public defenders, fiscalías are still considerably better funded and staffed than defensorías. Being the 'least favorite child' means that PDs and fiscales are equals in a court of law, but that the former are at a disadvantage vis-a-vis the latter: PDOs are subordinated to the local Executive in a way that fiscalías are not, and they have considerably less material and staffing resources.
In this context of inequality, how do PDs do their jobs? I address this question in the next section, firstly, by arguing that the public defenders' material limitations and their unequal relation with prosecutors create a lot of downtime where seemingly nothing happens. Importantly, I do not claim that fiscales never wait, nor that they wait more or less than PDs; since I did not conduct any fieldwork with fiscales, I do not know if waiting happens. However, my previous discussion suggests that if fiscales also wait, this happens for different reasons than the PDOs (after all, fiscales are better funded and they exert more control over the timing of hearings). Secondly, drawing from my fieldwork I note that, in these seemingly idle moments, public defenders are, in fact, working: they strategize, network, and undertake activities that are not in their purview but that must be done to guarantee an adequate defense (what I call "forced encroachment). 30 Sources for the Defensorías staff: INEGI (2020b). Source for Fiscalías staff: INEGI (2020a).The States of Coahuila, Chiapas, Chihuahua, Durango, Guerrero, and Quintana Roo are missing from the figure due to lack of data on personnel in the Public Defender's Office. The data prevents us from only looking at prosecutors in the Fiscalías, but we do have access to the number of defenders. In the case of Baja California, the Fiscalía reported 184 defenders (out of 257 workers, 72% of total) and in the case of Oaxaca, 177 of the 298 workers are reported as defenders (60%). An important part of the resource inequality can also be seen in the amount of resources that were given to the different actors as they transitioned into the new criminal justice system: whereas 16 states reported investment in training of new prosecutors, only 5 invested in training public defenders (México Evalúa, 2018).

Idleness… as work?
Noting that public defenders are overworked and underfunded is relevant, but not particularly novel. 31 In this section, I argue that public defenders engage in a particular type of overwork: one in which they spend a lot of their time waiting, seemingly doing nothing. I develop this argument by explaining how limited resources and inequality within different parts of the criminal justice system help create an environment in which defenders spend their time waiting. I then identify three ways in which defenders use that time to work: by strategizing on specific cases, by communicating and socializing with other workers of the criminal justice system, and by engaging in activities that are not their responsibility but that are necessary for the PDs to fulfil their duty, a practice that I call "forced encroachment." In the three offices I visited, the most important job that the defender's office had to do 32 was staffing court hearings, 33 a task where difficulties arise and most opportunities for idle time take place. 32 Not all defenders are equally pressed to staff hearings: some (a minority) of PDs are focused on appeals and other processes that require more in-office paperwork than making appearances at the courthouse. 33 In Mexicali, for example, a PD commented that if they were not present for a hearing, then the tribunal could file a complaint against them. Pablo, Mexicali (July 2nd, 2019).
What are these difficulties and why do they happen? Partly, such difficulties could be attributed to the unexpected nature of the criminal justice system: given that part of the defender's job is to make sure the accused has been informed of their rights, the PD must guarantee a prompt meeting with anyone who comes in contact with the accused. Thus, the three offices I visited had set up a system where one or two defenders are on-call 24 h, often waiting near the courthouse, revising timetables, and staffing hearings when needed. This system allows the PD office to make sure that individuals in unplanned hearings, for example, those who were caught red-handed and brought soon after for an arraignment hearing, are appropriately represented.
Partly, however, the difficulty in controlling their time stems from the uncertainty in who will be representing those in the system. That is, every hearing necessitates the three parts, prosecutor, defender, and judge, to be present. But whereas prosecutors and judges have to be in the courtroom-which allows them more certainty over where they are supposed to be, those accused can be represented by a public or private defender. This would not be an issue, of course, if PDs knew ex-ante whether their legal advice and representation was necessary, but this seems not to be the case: as one of the PDs in the Justice Fair event stated, she said she was sometimes sent to a hearing where "you arrive and there is already another defender that was summoned for the same case, and then it turns out that the accused wanted a private defender." 34 This assertion that was met with nods of agreement by other defenders, suggesting that this was a common occurrence.
Staffing court proceedings was an issue in Mexicali: two defenders mentioned that sometimes hearings are scheduled and marked to be attended by a private defender, only to find out right as the hearing is about to start that the private attorney had not shown up. In those cases, a public defender often had to step in at the last minute-failing to do so might result in either a reprimand by the courts or the postponement of the hearing, which, as a defender said, would mostly damage those accused. 35 In Oaxaca, on the other hand, defenders did not identify any scheduling issues, and a supervisor even said "There is always a way to address the staffing of hearings." 36 However, as I discuss in the subsequent sections, I witnessed postponed or cancelled court hearings in the three offices I visited.
Finally, the lack of control over their schedule also responds to the fact that they are, so to speak, at the mercy of the prosecutors: Fiscalías are the institutions that determine the workload not only of the PDOs but of the courthouses as well. The superiority of the Fiscalía, already visible in resources and political autonomy, can also be observed in the process by which cases are attended to and hearings are scheduled. To illustrate the degree to which prosecutors control the process, I use the next short anecdote of one of the first investigations in which a judicial agent (from the investigative police) from Chihuahua participated. 37 This agent had found 34 Rosa, Feria x la Justicia (June 17, 2019). 35 Guillermo (July 3rd, 2019). 36 Diego, Oaxaca city (December 17, 2019). 37 Story shared in a conversation at the Feria x la Justicia.

3
Idleness as work? How public defenders do their job by waiting a dead man in a public bathroom, with a wound in his head. He and his colleague investigated and secured a suspect. He shares that when the suspect was "just about to confess, I went to talk to the MP (agente del ministerio público or prosecutor). Turns out that the MP had already closed the case and filed it as an accident. Thus, there was no crime to investigate." 38 This anecdote shows that prosecutors are, indeed "the gate-keepers of the judiciary" (Rios-Fígueroa, 2012, p.197), enjoying the power of even determining which acts are even considered a crime.
The uncertainty surrounding the day-to-day activities of PDs work, as well as its de facto reactivity to the prosecutor's office means that defenders spend a lot of time waiting either on call or outside the courthouse, just to see if they are needed. They also wait when the hearing is delayed-which happens often as earlier hearings take setting back all the schedule of the courthouse-or when the judge or prosecutor show up late, or if the accused' arrival to the hearing runs late.
The lack of resources also creates inefficient scheduling, carving even more down time for defenders even in the face of an overwhelming amount of cases and hearings. Take, for example, the fact that PDs in most states cannot request lab or forensic tests, as the labs are often under the prosecutor's jurisdiction. Thus, if the PDs require a test, they have three options: request family members of the accused to hire an independent specialist, request special funding for a specific test, or request prosecutors to request tests. The first two options are usually financially unfeasible, people facing a criminal accusation are overwhelmingly poor, particularly those who cannot afford a private lawyer and PDOs are underfunded. Therefore, PDs often have to communicate with the fiscalías for such requests, a communication that was portrayed as cordial and friendly in the three offices I visited. When the fiscalías grant this, the defender often has to go and pick up the results, which often means accommodating to the schedule of the prosecutors and, yes, waiting for them.
I witnessed an instance of the consequences of lack of resources while Abril, a public defender in Oaxaca city, received a WhatsApp message from an administrative assistant from the Fiscalía, announcing that some photocopies of a file that she had ordered had arrived. We were mid-conversation, but she asked me if I had ever been to the Fiscalía and, after hearing that I had not, she told me to walk with her. The PDO in Oaxaca city is particular: it is part of a l3 building complex inaugurated in 2010 called Ciudad Judicial (Judicial city). This complex houses the Public Defenders Office, the Fiscalía, a series of courtrooms, and other government offices. The prosecutors building was a short walk away-probably 10 min-and once we arrived, Abril checked in with the administrative assistant that had contacted her. The assistant requested the copies, and after around 45 min, they finally arrive. 39 That is, this simple action of getting photocopies took a little more than an hour of a public defender's time, most of which was spent just… waiting.
Thus, the lack of resources and the fact that PDs are at the mercy of the courts' and Fiscalías' scheduling mean that PDs cannot use this time investigating cases, or visiting inmates. Instead, PDs spend a lot of time checking their phones, or socializing with court employees or family members of those they represent. Superficially, this might be seen as idling. Yet, in the next sections, I illustrate at least three ways in which defenders take advantage of this down time: by communicating with family members and those accused to build the best strategy possible, by building social networks with other criminal justice offices and workers, and by creating programs and undertaking activities that are technically not their responsibility, but that they must do in order to provide legal representation to anyone who requires it.
Importantly, I make no judgement regarding the usefulness or desirability of the activities identified in the next section. Put differently, I do not claim that PDs would be more productive if their schedule was perfectly structured and they had no idle time, nor that they would be more effective if they were accomplishing other tasks while waiting. Instead, my claim is that the public defender's status in the criminal justice system creates pockets of unexpected down time and that such time is filled with tasks directly relevant to their job. Put differently, PDs can-and often doinvestigate and build strong legal defenses, but their status as 'least favorite child' can force them to do what they can while waiting for other people.

Strategizing
While in Oaxaca city, I visited the Ixcotel courthouse (a courthouse outside of the Ciudad de la Justicia compound). The first hearing I attended started at 10am sharp: the judge and the defender were there but the fiscal was nowhere to be found. The workers of the courthouse tried to find the fiscal in charge, and while this happened, the judge left the courtroom. During this time, the defender was in the hallway, talking to some people that had not been previously in the courtroom with us. Later, I found out that they were family members of inmates that were up for parole: around that time (Christmas), the state of Oaxaca hosts an event to release inmates who qualified for either parole or alternative agreements. 40 The defender used this time to inform family members what to expect from the hearings that were scheduled later that day.
Finally, at 10.40am, the fiscal arrived, but now the judge needed to be called. While the judge arrived, the defender had a short conversation with the fiscal, reviewing and confirming what was about to happen: they would request the judge for reparation of damages instead of prison time. Both parties then talked to the victim to ask if she agreed: since arrangements like these necessitate the consent of the victim, it is likely that this solution had already been discussed with the victim, and that this conversation reiterated what was about to happen. Indeed, at 10.55almost one hour after the scheduled time, the judge arrived and the agreement was 1 3 Idleness as work? How public defenders do their job by waiting formalized. Thus, the delay of the two other workers at the courthouse-fiscal and judge-allowed the PD to inform family members on other cases, and to confirm with the fiscal and victim on their hearing case that the arrangement was still valid.
Similarly, public defenders in Mexicali took advantage of this down time between hearings to catch up with their work. On one occasion, a defender took advantage of a delayed hearing to run back to the office, which is in a different wing of the same building as the courthouse, and get files (and read them) for a hearing that she unexpectedly had to attend. In short, defenders use this down time to prepare (themselves and others) for hearings they need to attend.

Network building
Public defenders also use down-time to network and communicate with workers of the fiscalía and the courthouse. For example, while Abril was waiting to get the photocopies in Oaxaca city, she started a conversation with the assistant-later she told me she needed to "clear some scheduling issues" regarding another case and that the assistant and her were really good friends (while waiting, she also made a quick phone call from a relative of one of her cases, thus strategizing). In a similar way, the previous example at the Ixcotel courthouse, the defender used the time before the judge came to discuss details of the arrangement, but also to catch up with the Fiscal about how had they been and what were they up to.
PDs also use this down time to socialize among them, something that might seem unimportant but it is not: one of the indirect changes that the 2008 reform brought about was the creation of the PD as an organization with a specific purpose separated from the courts and the judicial system overall. That is, as explained in previous sections, before the 2008 reform, public defenders viewed their work as 'paper pushers,' merely making sure that the system kept running. As one of the defenders who has joined the PDO in 2005, a few years before the reform, told me: I entered the system not knowing anything. That is how one starts, seeking support among MPs… because before [the reform], we [defenders] didn't know each other, we were assigned to one courthouse, so my coworkers were the MP, they were the ones I worked with. 41 Thus, the reform not only changed the PD's specific role on the courthouse, but also the organization's internal dynamic. In this context, it is relevant to underscore that using this wait time to socialize with other PDs is not merely "hanging out with friends," but instead it creates opportunities for mentorship and learning from peers within the institution. In Mexicali, for example, a defender approached a supervisor I was interviewing and asked him a question regarding how to handle a "delicate" case (victim was acquainted with a judge). In Oaxaca, a defender approached a supervisor outside the courthouse to ask what to do if a relative of an inmate had not shown up-this was a case where parole was a possibility and they needed the relative to testify that the inmate would have a place to live.
This socialization, of course, extends beyond the professional: One of the PDs I interviewed introduced me to their spouse, an agent of the judicial (investigative) police. They met while they were both assigned to the same case (opposite sides, of course), and after meeting at work and in the courthouse, they started a romantic relationship. Unprompted, the PD explained that "being with another worker of the criminal justice system (operador del sistema) is easier, because they understand the difficulties of the job: the schedule, the things you see…" 42 Similarly, Guillermo, a PD supervisor in Mexicali told me that "there is an increase in couples where both belong to the system, because they understand each other." 43 Both of these examples show that down time is used by defenders to socialize and build networks. This socialization need not be only about work, but as Oberfield discusses, it is crucial to understand the rules of behavior that bureaucrats create for themselves and for the organization (Oberfield, 2014).

Forced encroachment
Public defenders in Mexico also use their down time to undertake activities that do not fall under their purview, but that are crucial for them to do their job. The clearest example of this pertains providing adequate defense for those individuals who might not speak Spanish, a pervasive problem in the state of Oaxaca, home to over 1.2 million people who identify as indigenous, and where 12% of the population does not speak Spanish. This is a problem for the entire justice system: since very few fiscales, judges, or PDs speak indigenous languages, cases involving non-Spanish speakers tend to take longer and present a lot of logistical difficulties.
Following federal legislation, guaranteeing access to a process in their language seems to be the responsibility of the courts (Marcos Escobar, 2012). Oaxaca's article 16 of the Constitution states that: […] Authorities will make sure that, preferably [sic], the public ministries o fiscales, the judges and magistrates are speakers of the native tongue or, absent this, that they [indigenous people or afro-mexicans facing trial] have access to a bilingual translator or interpreter, a defender who is technical and adequate, and/or juridical advisor….
Thus, although the law provides no clear mandate on whose responsibility it is to provide translation services, it does suggest that the public defenders are there to provide technical and adequate defense, not translation. Yet, public defenders are the ones that have tackled this issue: the Oaxaca office approaches college students in public universities who speak indigenous languages and invite them to do social 1 3 Idleness as work? How public defenders do their job by waiting service (a requirement for all bachelor degrees in Mexico) in the PDO by serving as translators.
On the surface, this program looks like it takes too much time: workers of the PDO have to actively seek students, something that takes time and resources. Participants are not necessarily studying law or even social sciences-during my fieldwork I met two of these students, one who was studying biology and one was studying to become a veterinarian-so defenders need to train them the intricacies of the legal system so that they are able to translate and explain the charges and process to those facing an accusation. In addition, PDOs provide some emotional containment to translators, who, as defenders explained, were often receivers and communicators of bad news. 44 This specific job of training these translators is often done outside of the courthouse, while defenders wait, or on the long and frequent trips they take to visit the prisons.
Public defenders know that they need this program: absent translators, PDs will have a hard time fulfilling their duties of guaranteeing respect to due process since they will be unable to explain charges and rights to those accused. Furthermore, if courts or fiscales fail to provide a process that can be understood by the accused they can request a postponement until translation is secured. But PDs know that doing this might mean failing to provide legal protection and, as a result, they are effectively forced to provide translators.
Similarly, three defenders in Oaxaca mentioned that they were learning of cases and ways in which to respect the juridical norms of indigenous communities that overlap with state judicial codes, a phenomenon called juridical pluralism. One of the defenders said that googling information on juridical pluralism was a "good way to pass time while on duty" (Diego, December 17, 2019). Again, this might seem like idleness-learning about something that is not directly linked to the job, but that it is necessary for a meaningful provision of proper defense.

Discussion and concluding remarks
As mentioned in the introduction, current studies of punishment modern criminal justice systems emphasize the role that time management and accountability has for a range of outcomes from disciplining those incarcerated (Malacrida, 2005;Wahidin, 2006) to providing certainty of the process (Domingo, 1999;Pal, 2020) and guaranteeing expedited justice (Zepeda Lecuona, 2008). Yet, the phenomena identified here suggests that the 2008 modernizing reform of the Mexican criminal justice system-through all its new codes and reorganization (see Shirk 2016)-did not create the expected rigid institutional order with predictable timings. Instead, it resulted in a system that, through its very organization, produces unpredictable moments of waiting for (at least some of) its bureaucrats. In these moments, public defenders engage in activities that enable them to provide a legal defense even in the direst of situations.
This finding strongly resonates with Auyero's (2012) study of Argentinian public service provision, where he suggests that "everyday political domination is what happens when nothing apparently happens, when people 'just wait'." (Auyero, 2012 p. 19, see also Cohen 2018). But what is at stake in the waiting of the public defenders is not merely the budgetary or operational superiority of one bureaucracy over another one: it is also what-and crucially who-those bureaucracies represent. That is, existing literature has pointed out that the criminal justice system is now used to handle a variety of social problems such as homelessness (Gonzalez et al., 2018), mental health issues (Torrey et al., 1998), lack of discipline in the classroom (Welch, 2018), and overall poverty (Wacquant, 2009), and how this practices further reify the social disparities. This paper also contributes to this literature by claiming that the working conditions of the public defenders-those who protect poor and vulnerable populations-is also plagued with waiting times enabled by the organization of the criminal justice system. I conclude by noting that the working and funding conditions discussed in this paper are not particular to Mexican public defenders: they are notably similar to those reported in the United States. For example, it is reported that, in the US, up to 80% of felony defendants are assisted by public counsel, whereas in Mexico the estimate for that number ranges from 75 to 90% (Azaola & Bergman, 2009). PDs are also reported to have very little resources, meaning that they can rarely investigate, and are overwhelmed with clients, meaning that they cannot take the time to mount a defense. As a result, PDs in the US are incentivized to convince clients to take plea agreements, a common practice that has been linked to an increase in false convictions.
The argument that I lay out here, I believe, points to a similar conclusion. This idleness-as-work practices allows public defenders to provide an adequate (at least on paper) legal defense even in the direst of situations. The lack of resources and control over their own schedule means that PDs must use the time they have in tasks like talking to family members, making sure that the agreements with their counterparts are still valid, and training translators, instead of interviewing witnesses or investigating. Ironically, however, precisely by doing their best with what they have they keep the criminal justice system running, enabling the reproduction of social inequality not by failing to do their job but, paradoxically, by doing it even under harsh conditions.
Funding The author received financial support for this research from the Stigler Center, Booth School of Business at the University of Chicago.
Data availability Notes for interviews are available upon reasonable request.

Declarations
This research followed applicable laws and institutional guidelines. Research involved human subjects and was conducted under University of Chicago's IRB number IRB19-0843. Verbal informed consent was obtained prior to interviews, and interviews were anonymized. No deception was used.