Every Mind is a Different World: do Judges Truly Decide in Such Diverse Ways?^*

Abstract

In this study, we investigated judicial behavior during custody hearings in Brazil, a legal proceeding in which judges determine whether the arrested person should be kept in custody or released pending trial. Drawing on primary data obtained from hearings held between 2015 and 2016, our objective was to investigate the consistency of judges' decisions and discern whether these decisions were influenced by individual ideologies. We employed semi-structured interviews, which we contend to be a valuable research method, to evaluate the judges’ ideologies. Our findings revealed that the judge aligning with a punitivist stance in the interviews exhibited a tendency for issuing more pretrial detentions. Additionally, using two binomial regression models, we highlighted how these understandings serve as a factor that differentiates the decisions made by the two judges. Consequently, our findings confirm the hypothesis that ideology plays a pivotal role in comprehending the intricacies of the judicial decision-making process in Brazil.

Judicial behavior; attitudinal model; judicial discretion; custody hearings

We are technicians’, said a prosecutor working on custody hearings¹ in Brazil. He continued: ‘We know our actions are guided not by the use of handcuffs, but by the imperative to ensure the safety of the facilities and their employees’. To say that the law serves as a technical framework for managing judicial conflicts implies that emotions, feelings, or public opinion would be irrelevant to the outcome of cases. Furthermore, it implies that any judge, regardless of age, gender, or their university of origin, operating within any institutional setting, would reach a similar or closely conclusion when making technically informed decisions.

One of the judges who regularly collaborated with the aforementioned prosecutor in custody hearings expressed a notable concern about conducting hearings without resorting to handcuffs, [since] ‘It’s a rare occurrence for individuals to be here [at these hearings] who are not repeat offenders, dangerous, and with numerous prior criminal convictions’. For another judge, however, that is not the major problem, but rather that ‘Criminal Justice primarily relies on arrests in the act [and therefore] operates as justice system against poor and Black people’.

These statements stem from semi-structured interviews conducted with judges, prosecutors, and public defenders who were actively involved in custody hearings in the city of Belo Horizonte during 2015 and 2016. They signal two starkly contrasting perspectives: one judge sees custody hearings as involving potentially dangerous individuals, while the other perceives them as a reflection of the selectivity within the criminal justice system, disproportionately affecting Black and poor people. A central question emerges: when confronted with the same defendants, would these judges reach identical decisions? More broadly, do judges arrive at similar verdicts regardless of their personal worldviews or the socioeconomic background of the accused?

To comprehend the impact of judges’ ideologies, i.e., their worldviews and perceptions of their roles (as legal practitioners), the role of justice, and their relationship with society, this study stems from an attitudinal approach to judicial behavior to understand the verdicts in custody hearings. In Brazil, custody hearings involve the immediate presentation of individuals arrested in flagrante delicto before a judicial authority, during which the legality of the arrest is assessed as well as the feasibility of implementing alternative precautionary measures, such as electronic monitoring or nighttime curfew. Pretrial detention should only be considered as a last resort, since one of the primary objectives of the custody hearing is precisely to reduce the provisionally detained population (IDDD, 2019).

The hypothesis of our study posits that judges who hold a stricter viewpoint, who perceive imprisonment as a fitting response to criminal conflicts, are more inclined to opt for pretrial detention in custody hearings, especially when dealing with individuals whom the justice system deems ‘dangerous’, even in situations involving offenses and suspects that may not necessarily warrant pretrial detention. In contrast, judges aligned with a guarantist² viewpoint, i.e., emphasizing the protection of individual rights, would be more inclined to rule in favor of liberty over incarceration, emphasizing the equality of individuals under arrest.

As an exploratory study, our objective is not to conclusively validate or refute this hypothesis, but rather to refine it for more comprehensive analyses in future research. In pursuit of this goal, we seek to make a non-exhaustive comparison of the proportions of pretrial detentions ordered by two judges who carried out daily custody hearings from September 2015 to May 2016. The novelty of this study resides in its use of semi-structured interviews as a valuable insight for assessing the ideologies of legal practitioners. In other words, we will use qualitative data obtained through semi-structured interviews to inform our quantitative analyses and interpret the resulting statistical findings.

Hence, our goal is to pinpoint potential directions for exploring the influence of ideology on the decision-making process, with a particular focus on Brazil, where this area of research remains relatively uncharted (SILVA and VIEIRA, 2022). To structure our article, we have arranged the content as follows. The first section, offers a comprehensive review of the existing literature on judicial behavior. The second section, discusses the institutional framework of custody hearings in Brazil. The third section, outlines the methodology employed in this study. The fourth section, presents and analyzes the generated data; and finally, the fifth section, delivers our concluding remarks and considerations.

Models of Judicial Behavior

Numerous theories have been postulated to comprehend the behavior of judges. The formal legalist theory posits that a judge’s role primarily entails the mechanical application of the law to social conflicts. This viewpoint maintains that the decision-making process follows a flawless legal rationality, implying that, regardless of the presiding judge, the verdict would remain consistent, thus rendering the study of judicial behavior irrelevant (NEMACKECK, 2017). However, upon closer examination of judicial decisions, particularly within the domain of criminal justice, the concept of legality seems to operate as an imperfect mechanism, as individuals who are Black or have lower educational attainment are more likely to face incarceration or receive lengthier prison sentences (VARGAS, 2014). Even in qualitative analyses, and even though the legal practitioners we interviewed often framed the law as a technical tool for decision-making, being male and Black increases the likelihood of incarceration since this demographic profile is frequently perceived as ‘criminal’ (AZEVEDO and SINHORETTO, 2017). Therefore, concerning our research problem, it becomes evident that the legalistic model falls short in providing a comprehensive explanation for the decisions made during custody hearings (LAGES and RIBEIRO, 2019).

In the United States, where the study of judges' behavior has a long history, the formal legalist perspective gradually gave way to the legal realism movement, which sought to comprehend judicial practices beyond mere normative imperatives. A pioneering work in this context was Herman Pritchett’s study in 1940, which analyzed how Supreme Court judges appointed by President Roosevelt tended to make aligned decisions. His conclusion, drawn from the ideological alignment of the judges, was that the judges’ ideologies ultimately shape public law (NEMACHECK, 2017). Within this framework, the attitudinal theory, influenced by behaviorism (MAGALHÃES, 2020b), posits that judges make sincere decisions. From this perspective, judges apply what they perceive to be fair based on their ideological preferences (SEGAL, 2008). Consequently, the legal arguments presented in their decisions can be viewed as rationalizations of the judges’ underlying system of beliefs, values, and attitudes (RIBEIRO; ARGUELHES, 2013).

According to Segal (2008), judges do not operate in an institutional vacuum and are bound by the rules of the game, both formal and informal. As he puts it, “the likelihood of judges behaving consistently with the attitudinal model will depend on institutional incentives and disincentives for ideological behavior” (SEGAL, 2008, p. 25). The author contends that, unlike lower-court judges who might be inclined to align with the preferences of higher courts to advance in their careers, U.S. Supreme Court Justices enjoy a greater degree of decision-making autonomy. Having reached the pinnacle of their professional careers, these judges face minimal or no constraints on their decision-making freedom, enabling them to sincerely express their political preferences in their decisions. Supreme Court Justices exercise control over their agendas, are not beholden to electoral or political accountability, have no further career ambitions to pursue (as they have already reached the highest echelon of their profession), and their decisions are not subject to judicial appeal (SEGAL, 2008). Consequently, they have little incentive to act strategically and are more likely to act in accordance with their political preferences.

In contrast to the attitudinal theory, the strategic theory posits that judicial decisions are not the result of sincere behavior but rather involve sophisticated decision-making. Judges, when making decisions, would carefully weigh various interests, which may encompass professional, ideological, or personal variables. In other words, judges operate within the confines of both formal and informal rules and are compelled to adopt strategic behavior (EPSTEIN and KNIGHT, 2013). Consequently, even if their aim is to advance their political preferences (which may not always be their top priority), judges would, to avoid retaliation or sanctions, adopt a sophisticated behavior to ensure that their decisions cannot be easily questioned, thereby reducing the likelihood of being reviewed by colleagues or higher courts.

In summary, both attitudinal and strategic theories challenge the legalist model, which posits that judges are mere enforcers of the law with no room for discretion. According to Segal (2008), the attitudinal model can explain the decisions of the U.S. Supreme Court because the institutional framework does not impose sufficient constraints on the ‘Justices’ to make decisions in a sophisticated manner.

Within the context of this study, the hypothesis is that in Brazil, lower-court judges, particularly in custody hearings, make sincere decisions, i.e., in line with their personal preferences, irrespective of institutional constraints. As we will delve into in the next section, the institutional arrangement in Brazil, including both formal and informal rules, does not provide sufficient incentives for judges to engage in sophisticated decision-making behavior. Consequently, their personal worldviews play a pivotal role in the decision-making process. In addition to their ideologies, criminal judges make decisions in a context marked by inequality. On one side, those who make the decisions are part of the country’s elite, while on the other side, those who are arrested typically belong to the most economically disadvantaged segments of the population.

At this point, as this article delves into judicial decisions within the realm of criminal justice, it is essential to emphasize the significance of studies on sentencing patterns, which evaluate factors beyond the behavior of judges. As highlighted by Raupp (2015), studies on sentencing, which primarily evolved with a focus on criminal decisions, gained prominence in the United States during the 1960s. Their primary objective was to understand the impact of extralegal factors on the decision-making process, including variables such as the social class, gender, and race of individuals brought to court. As a result, these analyses incorporate traditional variables associated with judicial behavior, which focus on the judge, along with additional dimensions related to the legal intricacies of the case and the characteristics of both the suspect and the victim of the crime (RIBEIRO et al., 2022).

The findings of these studies vary, with some suggesting a direct influence of extralegal factors (ADORNO, 1996), while others reinforce the importance of legal criteria, such as the severity of the crime and recidivism (CANO et al., 2010). As a result, despite the clear overrepresentation of poor and Black individuals in the criminal justice system, these studies raise questions about whether judicial institutions actively contribute to this selectivity or merely perpetuate existing social inequalities (LAGES and RIBEIRO, 2019). Due to the often inconclusive results in prior research, there is a growing need to incorporate environmental variables into the analysis of extralegal factors, such as organizational culture, the level of bureaucratization within institutions, and the individual characteristics of judges (ULMER, 2019). In light of these considerations, our study analyzes the judicial behavior of judges in custody hearings while taking into account the mediating role of extralegal variables.

The institutional framework of custody hearings: are there constraints on releasing individuals arrested in flagrante delicto?

The management of justice by legal practitioners can display considerable variation, especially when formal regulations provide them with wide-ranging decision-making discretion (ULMER, 2019). In Brazil, studies seeking to comprehend how judges arrive at their decisions, whether from a strategic or ideological perspective, have predominantly focused on the Federal Supreme Court (STF in the Portuguese acronym). And the behavior of its judges (OLIVEIRA, 2012, 2012; RIBEIRO and ARGUELHES, 2013; SILVA and VIEIRA, 2022), with relatively few works dedicated to understanding the judicial behavior of lower courts.

Castro (2012), in an analysis of the impact of judges’ professional backgrounds on their decisions, found that legal professionals are more likely to make decisions favoring the defendant, while those from the Public Prosecutor's Office tend to decide in the opposite manner. Focusing specifically on criminal justice, Magalhães (2020a) examined sentences delivered in São Paulo between 2013 and 2019 to investigate whether punitivism, the gender of the judge, and the gender of the defendant serve as explanatory variables for judicial decisions. Castro’s findings suggest that female judges tend to be more punitive compared to their male counterparts, but female defendants are less likely to be convicted than male defendants. Additionally, understanding how a judge tends to decide in a particular type of crime – whether with a more guarantist or punitive stance – predicts that judge’s decisions in other cases. Once again, attitudinal theory appears to provide greater explanatory power for understanding the decision-making process in state-level lower courts in Brazil.

As previously discussed, there are notable differences in the American context (SEGAL, 2008). First, the legislation in the USA tends to be much more specific about what lower court judges can or cannot do. Even in cases with a broader margin of discretion, the institutional framework is designed to constrain judges’ ability to act solely in accordance with their preferences (ULMER, 2019). In other words, in contrast to the Brazilian context (MAGALHÃES, 2020a; CASTRO, 2012), there is less room for lower court judges in the USA to make decisions based on their political preferences.

The USA operates under the Common Law system, which implies a greater degree of consensus between sentences, since decisions by higher courts serve as binding precedents for lower courts. On the contrary, Brazil operates under the Civil Law system, which places a strong emphasis on the principle of freely motivated conviction. This principle allows judges to actively seek evidence until they are convinced of the narrative presented in the case, aiming to establish the “real truth” (MENDES, 2012). Therefore, judges in Brazil have the discretion to set aside evidence they ‘do not deem appropriate’, based on their own judgment, to support their decisions. In practice, as Mendes (2012) has argued, this element introduces considerable subjectivity into Brazilian judicial decisions and allows for rulings that may deviate from interpretations made by higher courts in similar cases.

Moreover, the institutional context for lower-court judges in Brazil does not provide strong incentives for the adoption of strategic behavior (MAGALHÃES, 2020b). Despite the possibility of overturning decisions through appeals to higher courts, cases are assigned to judges through random allocation, making it challenging for judges to anticipate their actions and engage in sophisticated behavior (CASTRO, 2012). Furthermore, these judges are not politically appointed (as they are selected through civil service examinations rather than by popular vote) and they do not make collective decisions (as their rulings are rendered individually) (RAMOS and CASTRO, 2019). Consequently, lower court judges are not encouraged to act strategically.

Nonetheless, we should add an important caveat. While lower-court judges in Brazil make individual decisions without the need for negotiation, the organizational context of trial courts has garnered the attention of analysts. In the United States, some studies have highlighted that judges form close relationships with prosecutors and defense attorneys over numerous trials, resulting in negotiations on case resolutions (GALANTER, 1974). When testing this hypothesis in the Brazilian context, Ribeiro et al. (2022) discovered that when deciding on pretrial detention, judges tend to align with the prosecution's arguments, given the prosecutor’s constant presence in the organizational context. Conversely, the defense attorney, being a highly variable actor and thus unable to develop close relationships with judges, faces a reduced likelihood of securing favorable outcomes for their clients.

Regarding the formal structure of custody hearings, the law defines pretrial detention as a measure of last resort, to be employed only in exceptional cases, when less severe alternatives prove ineffective, such as electronic monitoring, nighttime home curfew, and restrictions on specific places or associations (LAGES and RIBEIRO, 2019). It is important to note that the notion of ‘exceptional’ is somewhat flexible: the legal grounds provided in the law for ordering pretrial detention are broad and may allow for the inclusion of political and moral considerations (AZEVEDO et al., 2017). Consequently, the interpretation of ‘exceptional cases’ can vary depending on legal practitioners, their practices, values, and worldviews (RIBEIRO et al., 2022). This discretionary element highlights the potential for differences in judicial practices, as legal practitioners have significant leeway to interpret the necessity of pretrial detention to varying degrees.

From an informal standpoint, despite the legal principle that freedom should be the norm rather than the exception, prior research on custody hearings suggests a reluctance to release individuals. Judges tend to view the decision to grant freedom not as a right but as an opportunity to bring about a ‘life change’ for those in custody, a privilege to be granted only when the judge genuinely perceives this prospect (ABREU, 2018). In essence, freedom becomes the exceptional outcome. Judges, in this context, are apprehensive that by not ordering pretrial detention, they might be labeled as soft on crime by public opinion (ALVES and MOREIRA, 2022). Thus, despite the reluctance to detain individuals (as the law prescribes freedom as the rule), there is also a reluctance to release individuals from custody and perpetuate the notion that ‘the police arrest, and the justice system releases’ (JESUS et al., 2018).

Furthermore, the decision-making process regarding who to arrest is mediated by various stereotypes associated with the identity of the criminal, including considerations of social class and race (ADORNO, 1996; ALVES and MOREIRA, 2022; GARCIA et al., 2022). As Misse (2010) has shown, not all individuals engaging in behavior classified as criminal are effectively labeled as ‘criminals’. Instead, this label depends not only on their actions but also on specific personal characteristics such as poverty, skin color, and lifestyle. Those who are socially perceived as suspects are more frequently targeted by the police and are less likely to be granted the ‘opportunity for freedom’ during custody hearings (AZEVEDO et al., 2022; KULLER, 2016; LAGES and RIBEIRO, 2019). Ramos et al. (2022) has shown that specific characteristics, such as being male, young, and Black, are considered suspicious even before any signs of criminal activity. As a result, individuals with these profiles attract the attention of the police, leading to their apprehension, despite the absence of any unlawful behavior.

Furthermore, another crucial aspect to comprehend the decision-making context is that these hearings primarily involve arrests in the act, associated with everyday crimes in major cities. The majority of suspects are apprehended by the military police for offenses such as robbery, theft, and drug trafficking (AZEVEDO et al., 2017; IDDD, 2017; 2019; LAGES and RIBEIRO, 2019). Individuals arrested under these circumstances form a relatively uniform demographic: predominantly young, Black males with low income and limited education. These individuals are judged in hearings that usually receive minimal or no attention from public opinion, as they are often perceived as the ‘typical criminals’ (ALVES and MOREIRA, 2022; AZEVEDO et al., 2022; GARCIA et al., 2022; KULLER, 2016). While judges may feel public pressure to prioritize public security, cases heard in these courts typically do not attract media coverage or the scrutiny of higher courts. This scenario enhances the likelihood of judges acting in alignment with their political preferences. Furthermore, the judges presiding custody hearings in Brazil are not constrained by legal precedents (in contrast to the Common Law system), affording them more decision-making latitude. Lastly, they are not subject to public scrutiny, as they are not elected (as is the case in most lower courts in the USA).

As a result, custody hearings provide a valuable opportunity to observe the variations in the behavior of legal practitioners. This is not only because the law allows for a certain degree of discretion, but also due to the absence of strong incentives for the adoption of strategic behavior. Consequently, we stem from the hypothesis that this institutional framework allows for a wide spectrum of decisions during custody hearings, allowing judges to align their verdicts with their political leanings, particularly in terms of their position on the criminal justice spectrum: whether they lean towards a more punitive or a more guarantist approach. We analyze this spectrum within the context of the legal field, where, as discussed above, a prevailing culture of incarceration primarily targets men, youth, and people of color (a punitive culture), rather than reserving imprisonment for genuinely exceptional crimes and circumstances (a guarantist culture).

Data and methodology

The empirical basis of this study was laid during the period of 2015 and 2016, through the observation of 825 custody hearings in the city of Belo Horizonte. To ensure the internal validity of the data, we established a prospective sample of hearings to be monitored. Taking into account the average daily caseload, our research team sought to monitor at least 14% of the hearings held each week. Our researchers attended the Lafayete Court every day, where the procedural rites occurred, including weekends and holidays, across various shifts, in order to encompass a broad spectrum of potential scenarios.

We gathered and documented the information by using two separate forms. The first form was filled out during the hearings and contains information related to the parties' requests and the procedural dynamics, such as the duration of the hearing and the questions posed. The second form was based on a documentary analysis, with a specific focus on the Social Defense Event Record (the local equivalent of a Police Record in Minas Gerais), the Report of Arrest in Flagrante Delicto, and the outcome of the hearing. This second form was used to compile information about the individuals involved, as well as the ultimate decision reached during the hearings.

Out of the 825 monitored hearings, 442 resulted in pretrial detention (53.6%). Freedom, with or without precautionary measures, was granted in 378 cases (45.8%), and in five cases, the arrest was considered illegal (0.6%). Throughout our six-month research period, two judges, two public defenders, and a prosecutor were designated as permanent practitioners in custody hearings, making them the professionals most frequently involved in these proceedings. In addition to the permanent participants, there were substitute judges, prosecutors, and defenders who typically handled hearings during weekends and holidays, as indicated in the table below.

In this study, we have taken measures to protect the real identity of the legal practitioners by using fictitious nicknames throughout the text. These nicknames have been standardized as predominantly male names due to the male-dominated nature of legal professions, especially within the judiciary (CNJ, 2018). Since the gender of the judges is discussed in the literature as an explanatory variable for judicial decisions (MAGALHÃES, 2020b), is it worth mentioning that both judges analyzed in this research are of the same gender.

As shown in Table 01, the permanent judges, prosecutors, and public defenders presided over a significant majority of the hearings, while the substitute practitioners had a relatively minor role. Furthermore, in 36% of the observed custody hearings, all three key legal practitioners (judges, prosecutors, and defenders) were permanent fixtures. In addition, the defenders consisted of both public (Carlos and José) and private lawyers, but the diversity and quantity of the latter make it impractical to provide details about each of them.

Table 01
Judges, Prosecutors, and Defenders participating in custody hearings in Belo Horizonte - (2015-2016)

Legal practitioner		N	%
Judge	Cézar	225	27.27
	Paulo	379	45.94
	Substitute judge	221	26.79
	Total	825	100
Prosecutor	Ricardo	558	67.64
	Substitute Prosecutor	267	32.36
	Total	825	100
Defender	Carlos	122	14.79
	José	296	35.88
	Substitute defender	232	28.12
	Private lawyer	175	21.21
	Total	825	100

Source: Elaborated by the authors based on research data.

The professionals who most actively participated in the custody hearings – judges Paulo and Cézar, defenders José and Carlos, and prosecutor Ricardo – were interviewed by our research team at the conclusion of the monitoring period, spanning from April to July 2016. These interviews provided valuable insights, allowing the researchers to map and compare the professionals’ worldviews, as well as their perspectives on the criminal justice system, its clientele, and its effectiveness.

How can we measure the ideology of judges?

The ideology of judges can be challenging to measure directly (MAGALHÃES, 2020b). One common approach is to use the political party of the person who appointed the judge as a proxy for their political preference (PERLIN and SANTOS, 2019). However, this method is not suitable for our study, which concentrates on lower courts where judges are civil servants and are not appointed through political processes. Additionally, this proxy is indirect, relying on the ideology of a third party rather than the judges themselves.

In a study investigating the impact of ideology on the actions of U.S. Justices, Segal (1989) employed an alternative method. He utilized newspaper reports on the ideological stance of judges and compared these reports to the liberalism-versus-conservatism spectrum based on the Americans for Democratic Action (ADA) scale. This methodology enabled the author to juxtapose the ideologies of judges with those of other authorities in various spheres of power, highlighting the significance of the ideological variable in judicial decisions. Martin and Quinn (2002) took a different approach and developed a dynamic ideological measure, based on the voting patterns of judges. This measure allows for comparisons over time for the same judge, capturing shifts in their opinion or the influence of different perspectives on voting behavior, extending beyond the traditional liberal-conservative spectrum to cover various thematic areas.

In the Brazilian context, Magalhães (2020a) applied a similar strategy to examine sentencing in drug trafficking cases. The author used the judges’ decisions in other crime categories (e.g., violence against women) as an indicator of their ideological position on the punitive-guarantist spectrum. This approach proves beneficial since lower-court judges seldom have their positions reported in newspapers, which makes it difficult to adopt Segal’s (2008) methodology. However, it comes with the drawback of endogeneity, where the decision is explained by prior outcomes, creating a self-reinforcing cycle.

In this study, we advocate for the use of semi-structured interviews as a tool to measure the ideology of judges and subsequently draw inferences about its impact on the decision-making process. Interviews offer the advantage of being conducted through personal interactions with professionals, providing a direct method for evaluating ideological perspectives. Moreover, they serve as an external source of information about the verdicts, mitigating issues related to the endogeneity of the phenomenon. This method is particularly valuable as it allows for a nuanced understanding, uncovering contradictions and enabling the scaling of ideological positions, which can be challenging with quantitative data alone. Consequently, we believe this methodology is well-suited for assessing the ideologies of court members within a specific court, especially when dealing with a limited number of judges. Additionally, our study also involved interviews with prosecutors and public defenders directly involved in custody hearings.

The use of interviews, however, is contingent on the researcher’s direct interaction with legal practitioners, which may not always be feasible. It also demands a significant amount of time, limiting the practical ability to expand the study sample. Besides sample size constraints, interviews can potentially elicit strategic behavior from interviewees, where they may not express their genuine views but instead provide responses aligned with their institutional roles. While controlling for this factor can be challenging, the semi-structured interview format allows for more fluid conversations, facilitating back-and-forth discussions on specific topics to assess the consistency of the arguments presented by the interviewee (RIBEIRO and VILAROUCA, 2012).

On a broader scale, this study represents an initial effort to evaluate the effectiveness of interviews as a method for discerning judges’ ideologies. For the organization of the qualitative data analysis, the interview responses were compiled in an Excel spreadsheet. The data was structured into rows corresponding to the interviewees and columns containing the statements made by the legal practitioners. These statements were then categorized as either more guarantist, aligning with the concept of a minimal penal state, or more punitive, endorsing the idea of a maximal penal state.

According to Salo de Carvalho (2008), the notion of guarantism has its roots in Enlightenment liberal logic, underpinned by a contractual foundation, and aims to maximize individual freedoms, secularize and humanize punishment, and perceives imprisonment as an extreme measure. Advocates of guarantism argue that an excessive focus on punishment can lead to injustice, violate fundamental rights, and perpetuate cycles of crime. In contrast, punitivism emphasizes punishment as the primary response to crimes, especially those committed as a ‘survival strategy’. In this perspective, imposing severe penalties is deemed necessary to deter individuals from engaging in criminal activities and safeguard society from the dangers posed by released offenders. These concepts represent the two extremes of a spectrum and should be understood here as ideal types for framing the discourse of the legal practitioners whose interviews we have analyzed in this article.

Data analysis

Based on the dataset derived from the interviews with both judges, we conducted a comparative analysis, classifying one judge as more punitive and the other as more guarantist. Following this classification, we performed a statistical analysis to ascertain whether the judge categorized as punitive issued more detention orders than the judge categorized as more guarantist – in other words, whether he was more inclined to restrict the freedom of those in custody. For this analysis, we used the Rstudio software to conduct a Pearson's chi-square test, which assesses whether the distribution of a categorical variable is random or not in relation to the distribution of another variable. In simpler terms, it examines whether the two social phenomena (the verdicts and the ideologies of the practitioners) are statistically associated.

Following this analysis, we developed two binomial logistic regression models: one to assess Cézar’s decisions and the other to examine Paulo’s decisions. The purpose of these models was to determine whether ideology remains a significant factor, even when considering the characteristics of the suspects brought to the custody hearings. The decision to incorporate the profile of the individuals in custody into the model was influenced by the observed homogeneity among the suspects presented at the custody hearings (refer to Table 02). The prevailing demographic profile of individuals consisted of Black men, aged up to 29 years, with education up to primary school. In fact, a large part of cases featured individuals fitting this description, colloquially referred to as ‘suspicious elements’ (RAMOS and MUSUMECI, 2004).

Table 02
Socioeconomic and offense profile of suspects in custody hearings

Characteristics	N	%
Male	742	90%
Black	643	78%
Aged between 18 and 29	558	68%
Education up to middle school level	481	58%
Use of a weapon when committing the crime	224	27%
Prior criminal record	779	94%

Source: Elaborated by the authors based on research data.

The custody hearing, while intended to be a platform for legal practitioners to interact with individuals in custody and possibly improve their defense prospects, can also be susceptible to biases and moral judgments (TOLEDO, 2020). Often, it fails to ensure the active participation and voices of the individuals in custody (GISI et al., 2022). In our model, we investigated whether the presence of the ‘suspicious element’ (male, Black, young, with a low educational level) profile influenced the likelihood of a pretrial detention order depending on the judge presiding the case.

To evaluate the legal dimensions, we examined two variables that, according to legal precedent, could be considered by the judge as reference points: the use of a firearm in the crime (suggestive of violence and a potential threat to life) and the existence of a criminal record (indicating a history of criminal involvement and an increased likelihood of reoffending). It is essential to note that this variable was established based on any pre-arrest police record prior to the custody hearing, without differentiation between cases of legal recidivism and active arrest warrants.

Lastly, we investigated whether the configuration of the legal practitioners had any impact on the verdicts. Given that the same prosecutors, defenders, and judges were involved in multiple cases, we sought to evaluate if discernible patterns emerged based on the specific prosecutor and defender involved in each case. Consequently, hearings involving both permanent and substitute prosecutors and defenders were also incorporated into the models.

Different judges, same decisions?

The principle of the rule of law, grounded in a government bound by laws rather than individual discretion, hinges on the idea that decisions in similar cases should be consistent, irrespective of the decision-maker. However, the two judges interviewed openly asserted that each judge decides in their own unique manner, reinforcing the common adage that ‘to each their own judgment’. In their view, this disparity arises because each judge can arrive at their own conclusions regarding the necessity of pretrial detention and who should be deprived of their liberty. They believe this explains why substitute judges often make significantly different decisions compared to permanent judges. In their own words:

The way I see it, perhaps... there's a certain difference, as the judges are consistently the same during the week, we tend to have a more or less standard decision for each case. That doesn’t mean it's pre-determined, but there's a kind of consensus regarding each crime or the situation of each person charged, you understand? I think this pattern varies a bit on weekends because those judges aren’t regularly in custody hearings. But this isn't anything new or absurd because each judge thinks as they want, right? They decide however they wish and are persuaded by one argument or another that might be different from the others, right? So... I believe that is the only difference. (Judge Paulo[3] (2016) – emphasis added)

I have a jurisdictional stance, right? The same goes for [Judge Paulo]. So, the decisions made during the week are more... they align with my perspective, my position. On weekends, we have completely different judges, substitute judges. Some judges are more rigorous, more aligned with law and order, and thus, there'll be many pretrial arrests. On the other hand, other judges are more liberal, and so... there'll be many provisional freedoms! There’s a whole and complete difference. The same goes for defenders and prosecutors. (Judge Cézar (2016) – emphasis added)

The judges seem to perceive decision-making disparities as a natural aspect of their roles, contrasting with the perspective of the prosecutor who sees these discrepancies as a problem that could potentially be addressed through judicial training programs. In his words:

Researcher: Do you believe there is a need for participation [in a training course]?

Prosecutor: Yes, and that includes standardizing actions because... a person in custody shouldn’t be left to the chance of encountering a more lenient or stricter judge or prosecutor. I believe that in this initial stage, where we’re not assessing the actual merit of the case but rather the necessity of... the... the possibility of being detained or released until trial, there should be a more consistent approach. It should be... consistent for all scenarios that fall under each criminal statute. (Prosecutor Ricardo, 2016)

While acknowledging the potential discrepancies in decisions, the prosecutor sees himself as a technical professional, akin to judges and public defenders. He believes that their actions can be standardized through legal knowledge. In a similar vein, one of the public defenders sees custody hearings as a prime opportunity to prevent unjust arrests. However, in practice, he has observed that the case’s outcome often hinges on the presiding judge:

Researcher – What are your thoughts on the Custody Hearings Project?

Defender: t's been great to learn about this initiative, particularly in terms of preventing abuses that occur during arrests in flagrante delicto. So, my initial impression has been highly positive. I believe it can significantly reduce the impacts of wrongful arrests. (Defender Carlos, 2016)

Researcher – And has it changed? Your perception?

Defender: Somewhat, because in some cases, I see that many individuals are detained simply because they can't afford bail, for instance. And I think this completely contradicts the innovative intention of the custody hearing. [...] I believe it depends on the professional – whether they allow themselves to be affected [by the detainee’s situation] or not. If they remain closed off, it won’t... it won’t influence them. If they are open to seeing the individual not just as another case number but as a human being with their own circumstances... trying to understand why they are on this path, in theory a life of crime – because we never know for sure, everything will be further investigated – they... they can be empathetic, and this would be very positive. However, some professionals are not receptive to this and view [the detained individuals] as mere case numbers. (Defender Carlos, 2016)

Based on the interviews, we find that legal practitioners, in general, perceive the variability in decision-making as stemming from differences in the judges’ positioning, acknowledging the influence of the judges’ ideologies (attitudinal model) on the decision-making process. The judges interviewed perceive this variability as a natural and inherent aspect of the act of judging. In contrast, the prosecutor views it as a problem and suggests the need for training programs to ‘standardize decisions’. The public defender underscored the ‘lack of agency’ on the defense's part, as, in practice, the outcome hinges on the stance of the presiding judge – whether they lean towards a more punitive approach or whether they ‘grasp the reality of the situation’. Therefore, it becomes crucial to analyze the ideological stance of each of these judges.

Through the interviewees' statements, we were able to discern, to varying degrees, different value judgments, allowing us to categorize one judge as more punitive and another as more guarantist. It is important to note that this classification was made comparatively and has no external validity for other members of the judiciary. Furthermore, a judge considered guarantist in this context might be seen as a punitivist judge in another situation.

In terms of crime control and the level of ‘strictness’ applied to those detained in custody hearings, Paulo advocated for a more stringent legal system, as he deems the existing laws too lenient towards criminals:

We can’t handle the rapidly rising crime rates, right? That's why I believe our justice system is ultimately inefficient. The criminal enforcement law is highly lenient; there's no way to curb crime if someone receives a twelve-year sentence and is back on the streets after just two years. The concept of serving only one-sixth of the sentence, with regime progression, is unfeasible. Nobody is held accountable for their actions, and the time spent in prison is exceptionally short. Although the sentences are lengthy, the effective time served is extremely brief. I believe these are the main problems. (Judge Paulo, 2016)

The mere occurrence or absence of a custody hearing does not inherently lead to an increase in the prison population. The prison population is on the rise due to what I believe is the unfortunate lack of an effective deterrent for criminal behavior. Committing crimes has become too easy, too commonplace, and even when there is a penalty, it's often brief... Unfortunately, in our country today, crime seems to pay. That’s why the prison population keeps growing. (Judge Paulo, 2016)

Judge Paulo tends to lean towards a punitive approach, prioritizing punishment over potential benefits provided by criminal law. On the other hand, Judge Cézar demonstrates a higher degree of concern for the rights of detained individuals. He emphasizes the importance of understanding the conditions of the detention facility before ordering an arrest, warranting awareness of where the detainee will be sent:

When I decide on pretrial detention, I have to do it with a clear understanding of where I'm sending the person, right? What exactly are the conditions there? So… it's not just about a judge inside an office anymore, as they say, right? […] It’s about realizing that you also become somewhat responsible for this, don’t you? So, I believe we need to address and improve the prison system, which is a terrible, horrible thing. [silence] The Custody Hearing isn’t about simply releasing someone; it was meant to spark a discussion: Are we using incarceration to help resolve the issue of crime and violence? We’re not! It’s clear that we're not! (Judge Cézar, 2016)

[...] I believe we need to focus more on preventing prisoners from returning after the Custody Hearing, you see? And to achieve this, it’s crucial that we invest in precautionary measures, in the diligent application of these precautions. (Judge Cézar, 2016)

Judge Cézar emphasizes that imprisonment is not the solution to violence, highlighting the crucial role of custody hearings in preventing individuals from entering the prison system when pretrial detention is unnecessary. He advocates for improving precautionary measures to offer individuals opportunities to distance themselves from a criminal lifestyle. Comparatively, we categorize Judge Cézar as more aligned with a guarantist approach than Judge Paulo. While one views prison as a solution to criminal conflicts, the other questions the necessity of imprisonment and shows concern for the rights of those detained by the State and their prospects for rehabilitation. Following this categorization, we conducted a Pearson’s chi-square test, crossing the verdict variable (categorized as imprisonment or release) with the judge variable (categorized as Judge Paulo or Cézar). The test revealed a statistically significant association between the type of verdict and the judge issuing it, as shown in Table 03 below.

Table 03
Cross between presiding judge and type of decision (imprisonment or release) and presiding judge

Judge	Pretrial detention				Total
	No		Yes
	N	%	N	%	N	%
Substitute Judge	120	54.1%	101	45.9%	221	100.0%
Paulo	142	37.5%	237	62.5%	379	100.0%
Cézar	121	54.0%	104	46.0%	225	100.0%
Total	383	46.4%	442	53.6%	825	100.0%

Source: Elaborated by the authors based on research data.
• Note: Chi-square: 14.62278; GL=1; p<0.001.

Judge Cézar statistically issued less detention orders compared to Judge Paulo. Specifically, the former opts for imprisonment in 46% of cases, while the latter issues detention orders in over half of the cases (62%), despite the law prescribing the exceptional nature of pretrial detention.

These quantitative findings align with the qualitative data: Judge Paulo, categorized as more ‘strict’ in the interviews, tends to issue more pretrial detention orders compared to Judge Cézar, who presents himself as more open to ‘understanding the reality of the prisoner’ (Table 04).

Table 04
Chi-square test between the detention orders issued by Judge Paulo and Judge Cezar in custody hearings in Belo Horizonte

Judge	Prisons enacted	Percentage of imprisonment	Total decisions	Chi-square	Confidence interval	p value
Cézar	104	46.0%	225	14.62	-0.25 - -0.08	p < 0.001
Paulo	237	62.5%	379

Source: Elaborated by the authors based on research data.

Lastly, we created two binomial logistic regression models, one for each judge, with the dependent variable being the issuance of a pretrial detention order. The independent variables included the defendant’s profile as a ‘suspicious element’ – male, young, low education, and Black – reflecting the stereotype associated with the idea of a ‘criminal’ (MISSE, 2010; RAMOS and MUSUMECI, 2004). This aimed to explore whether the characteristics of individuals arrested in flagrante delicto serve as a differentiating factor in the judges’ decisions. We found a notable contrast in the judges’ perspectives: Judge Paulo tends to view those arrested in flagrante delicto as dangerous, while Judge Cézar emphasizes the selectivity of the criminal justice system.

How can I conduct a hearing without handcuffs when, in ninety percent of the cases, we're dealing with repeat offenders or crimes involving weapons, do you understand? These are dangerous individuals, aren’t they? I believe that custody hearings… the resolution tends to portray those who come through here as people without the slightest danger, almost as victims of police violence... and, most of the time, that’s not the case. (Judge Paulo, 2016)

The criminal justice system is incredibly selective, isn’t it? Unfortunately… it creates significant stratification… and when people say that the justice system works against poor people or Black people… unfortunately that’s what we see in the Custody Hearings. So, I believe that having a more precise and sophisticated intelligence service, a more meticulous investigative approach, could help us prevent this situation. (Judge Cézar, 2016)

In addition to profile variables, we incorporated dimensions to account for legal factors, such as the use of a firearm (indicative of offense severity) and the presence of a criminal record (indicative of prior criminal history and potential danger). Moreover, to assess the hearing’s informal arrangement, we also included in our model the presence of permanent prosecutors and defenders, and not their substitute counterparts, to explore the impact of these ‘repeat actors’ on the decision-making process.

Table 05 shows that the suspect’s profile interferes differently in the decision of the two judges. The model does not reveal a statistically significant association (p<0,005) between Cézar’s decisions and the detainee’s profile. Conversely, when Paulo is the presiding judge, Individuals with low educational attainment are more likely to be incarcerated. This observation holds even when accounting for legal variables and the arrangement of practitioners in the hearing. This finding suggests that the more punitive judge is not oblivious to the defendant’s profile, reinforcing the idea that criminal justice decisions can exacerbate social biases.

Table 05
Results of the binomial logistic regression models estimating the odds ratios of pretrial detention by the presiding judge

	Judge Cézar			Judge Paulo
Variable	B	Sig.	Exp(B)	B	Sig.	Exp(B)
Permanent Prosecutor	0.638	0.269	1.892	0.285	0.392	1.330
Permanent Defender	0.491	0.488	1.635	-0.477	0.126	0.621
Male	0.804	0.202	2.236	0.775	0.110	2.170
Black	-0.437	0.330	0.646	-0.260	0.370	0.771
Aged between 18 and 29	0.227	0.585	1.255	0.475	0.107	1.608
Education up to middle school level	0.613	0.177	1.846	0.818	0.030	2.266
Use of a firearm	1.166	0.006	3.210	1.887	0.000	6.598
Presence of criminal record	1.497	0.195	4.469	3.131	0.006	22.903
Constant	-4.075	0.007	0.017	-4.418	0.001	0.012
Number of cases	136			267
Nagelkerke Pseudo R-square	0.105			0.152

Source: Elaborated by the authors based on research data.

Still according to Table 05, we find that neither of the two judges felt compelled to modify their behavior as a result of negotiations with the permanent prosecutor or either of the two permanent public defenders. Neither of these variables exhibited statistical significance. Therefore, the notion that recurrent participants in the trial process would have more negotiation leeway does not appear to align with the reality in the custody hearings we analyzed. In other words, while Segal (2008) identifies substantial decision-making autonomy among Supreme Court Justices in the U.S. context, the binomial logistic regression models suggest a similar scenario in Brazil among lower-court judges.

As for the legal dimensions, the two variables affect the decisions of the two judges differently. For judge Cézar, the ‘use of a firearm’ triples the likelihood of imprisonment, indicating how the severity of the crime emerges as an important factor in his decision. On the other hand, this model did not show statistical significance for the presence of a criminal record.

For Judge Paulo, the legal variables have a more pronounced impact. The ‘use of a firearm’ increases the likelihood of imprisonment by 6.5 times. It’s worth noting that during his interview, Judge Paulo placed great emphasis on the ‘violence’ associated with the crimes presented during custody hearings. Even though firearms were present in less than a third of the cases (27%), he stressed that the majority of these cases involved crimes committed with weapons by ‘dangerous’ individuals. In turn, the presence of criminal records increases the likelihood of imprisonment by 22.9 times. This high odds ratio indicates the significance of criminal records as a decisive factor for imprisonment in Judge Paulo's decisions, implying in significant disadvantages for the suspect.

Our examination of custody hearings in Belo Horizonte found that judges operate within an environment characterized by broad discretion in formal rules. Additionally, informal rules do not seem to provide strong incentives for the adoption of sophisticated behavior. While our data does not conclusively determine whether decisions are sincere or the result of strategic behavior, it strongly suggests that ideology is a key explanatory variable in the decision-making process at the lower court level. This underscores the importance of ideology in comprehending judicial decisions, not only at the level of the Supreme Court, where Justices are traditionally seen as having the role of innovating the law, but also within the realm of lower-court judges in Brazil.

Final considerations

In this study, we sought to contribute to the debate on judicial behavior in Brazil. While the influence of ideology on the decision-making processes of Supreme Court Judges has been extensively discussed at the national level, it remains relatively unexplored in the context of judges in other legal bodies. Therefore, the analysis presented represents an initial effort to investigate whether the ideology of legal professionals, as gauged through semi-structured interviews, indeed plays a pivotal role in the decision-making process in custody hearings. Additionally, we explored methods for identifying and quantifying this ideology through statistical models, aiming to understand the variables associated with more rigorous decisions, such as pretrial detention.

Our observations suggest that interviewees acknowledge the variability in decision-making depending on the ideological leanings of the judge – whether they are more hardline or more lenient. While the interviewees did not unanimously agree on whether this influence is problematic, there appears to be a general agreement that ideology plays a significant role in decisions, a notion shared not only among the judges but also among the prosecutors and defenders. Furthermore, we noted that the judge exhibiting a stricter attitude in the interviews issued a higher proportion of pretrial detention orders compared to the judge categorized as more lenient. This finding underscores the significance of considering ideology as a crucial variable for comprehending judicial decisions at the lower court level.

The binomial logistic regression models facilitated a comparative analysis of the decision-making processes employed by the two judges. For either judge, the presence of a permanent prosecutor or a permanent public defender did not change the outcome. Additionally, for both judges the variable ‘use of firearms’ increased the likelihood of pretrial detention. Specifically for Paulo, the variable ‘presence of criminal records’ is highly associated with harsher outcomes. Furthermore, while the detainee's profile did not influence Judge Cézar's decisions, we found that, for Judge Paulo, people with low education levels were more likely to remain in custody.

Additionally, as we emphasize the crucial role of ideology in decision-making, we contend that semi-structured interviews offer a valuable research method for juxtaposing judicial decisions with judges' ideological leanings. This approach enables direct engagement with legal practitioners, fostering a nuanced understanding of both the consistencies and divergences in the interviewees' perspectives. However, extracting ideology from qualitative data requires a thorough analysis of each judge's standpoint, which may constrain the feasibility of conducting research with an extensive sample size.

While this study sought to highlight disparities in decision-making among the interviewed judges, further research is necessary to delve deeper into the analysis of the decision-making processes of lower-court judges. Our study findings cannot be extrapolated to the entire judiciary, underscoring the importance of enhancing the representativeness of future research. Moreover, there is a need to control for additional factors that may influence the decision-making process, such as the specific nature of the crime under consideration, public opinion, the judge’s educational background, and even the gender of the decision-maker. These variables can wield substantial influence over judicial decisions and should be factored into future studies. Only through more comprehensive and in-depth studies can we achieve a more well-rounded and informed comprehension of the justice system and identify potential areas for improvement.

References

• ABREU, João Victor Freitas Duarte (2018), Quando é preciso soltar: os dilemas morais dos magistrados ao conceder o alvará de soltura numa Central de Audiências de Custódia. Paper presented at 42º Encontro Anual da ANPOCS.
• ADORNO, Sérgio (1996), Racismo, criminalidade violenta e justiça penal: réus brancos e negros em perspectiva comparativa. Revista Estudos Históricos Vol. 09, Nº 18, pp. 283-300.
• ALVES, Júlia Somberg and MOREIRA, Lisandra Espíndula (2022), Enquadro e banco dos réus: racismo e sistema de justiça. Psicologia & Sociedade Vol. 34, pp. 01-15.
• AZEVEDO, Rodrigo Ghiringhelli de and SINHORETTO, Jacqueline (2017), O sistema de justiça criminal na perspectiva da antropologia e da sociologia. BIB-Revista Brasileira de Informação Bibliográfica em Ciências Sociais Nº 84, pp. 188-215.
• AZEVEDO, Rodrigo Ghiringhelli de; SINHORETTO, Jacqueline, and SILVESTRE, Giane (2022), Encarceramento e desencarceramento no Brasil: a audiência de custódia como espaço de disputa. Sociologias Vol. 24, Nº 59, pp. 264-294.
• CANO, Ignacio; RIBEIRO, Ludmila, and MEIRELES, Elisabet (2010), Race and crime and criminal justice in Brazil. In: Race, crime and criminal justice: international perspectives. Edited by KALUNTA-CRUPTON, Anita. New York: Palgrave Macmillan. pp. 207-241.
• CARVALHO, Salo de (2008 ), Penas e garantias: a crise do direito e do processo penal, o garantismo jurídico, as teorias da pena, os sistemas de execução, a lei de execução penal, os conflitos carcerários, os direitos (de resistência) dos presos. Rio de Janeiro: Lumen Juris. 307 pp..
• CASTRO, Alexandre Sami de (2012), Ensaios sobre o poder judiciário no Brasil. Doctoral thesis Escola de Pós-Graduação em Economia. Fundação Getúlio Vargas.
• CONSELHO NACIONAL DE JUSTIÇA -CNJ (2018), Perfil sociodemográfico dos magistrados brasileiros. Brasília: CNJ. 291 pp..
• DEFENDER, Carlos (2016), Entrevista concedida a Sara Prado. Centro de estudos em criminalidade e Segurança Pública. Belo Horizonte, março de 2016. Duração: 45m12s.
• EPSTEIN, Lee and KNIGHT, Jack (2013), Reconsidering judicial preferences. Annual Review of Political Science Vol. 16, pp. 11- 31.
• GALANTER, Marc (1974), Why the 'Haves' come out ahead: speculations on the limits of legal change. Law & Society Review Vol. 09, Nº 01, pp. 165-230.
• GARCIA, Ludmila Abrante; PADILHA, Weslen Santana; LA RUBIA, Lecymar Paraguassu Batista Santos, and VILLANOVA, Victoria Cristina Gonçalves (2022), Pés descalços e descamisados: uma degustação do processo punitivo ou um retrato social? Ciência & Saúde Coletiva. Vol. 27, Nº 12, pp. 4365-4369.
• GISI, Bruna; SILVESTRE, Giane; GARCÍA-SÁNCHEZ, Efrain; CRUZ, Fernanda Novaes, and JESUS, Maria Gorete Marques de (2022), The exercise of authority during interactions in custody hearings in São Paulo (Brazil): building legitimacy through exclusion. Punishment & Society Vol. 25, Nº 03, pp.0 1-19.
• IDDD - INSTITUTO DE DEFESA DO DIREITO DE DEFESA (2019), O fim da liberdade: a urgência de recuperar o sentido e a efetividade das audiências de custódia. São Paulo: IDDD. 150 pp..
• IDDD - INSTITUTO DE DEFESA DO DIREITO DE DEFESA (2017), Audiência de custódia: panorama nacional. São Paulo: IDDD. 87 pp..
• JESUS, Maria Gorete Marques de; RUOTTI, Caren, and ALVES, Renato (2018), 'A gente prende, a audiência de custódia solta': narrativas policiais sobre as audiências de custódia e a crença na prisão. Revista Brasileira de Segurança Pública Vol. 12, Nº 01, pp. 152-172.
• JUDGE CÉZAR (2016), Entrevista concedida a Sara Prado. Centro de estudos em criminalidade e Segurança Pública. Belo Horizonte, março de 2016. Duração: 1h37m10s.
• JUDGE PAULO (2016), Entrevista concedida a Sara Prado. Centro de estudos em criminalidade e Segurança Pública. Belo Horizonte, março de 2016. Duração: 29m59s.
• KULLER, Laís Bóas Figueiredo (2016), Audiências de custódia: um ponto de inflexão no sistema de justiça criminal? Master's dissertation. Ciências Humanas e Sociais. Universidade Federal do ABC.
• LAGES, Lívia Bastos and RIBEIRO, Ludmila Mendonça Lopes (2019), Os determinantes da prisão preventiva em audiência de custódia: reforço de estereótipos sociais? Revista Direito GV Vol. 15, Nº 03, pp. 01-35.
• MAGALHÃES, Lucas Fernandes de (2020a), Condenados pela ideologia? Analisando o efeito de fatores extrajurídicos sobre sentenças criminais na Justiça Estadual de São Paulo. Paper presented at 44º Encontro da ANPOCS.
• MAGALHÃES, Lucas Fernandes de (2020b), Preferências políticas importam? Uma análise das sentenças criminais proferidas pelos juízes do estado de São Paulo. Master's dissertation Departamento de Ciência Política. Programa de Pós-Graduação em Ciência Política. Universidade Federal de Minas Gerais.
• MARTIN, Andrew D. and QUINN, Kevin M. (2002), Dynamic ideal point estimate via Markov Chain Monte Carlo for the U. S. Supreme Court, 1953-1999. Political Analysis Vol. 10, Nº 02, pp.134-153..
• MENDES, Regina (2012), Verdade real e livre convencimento. DILEMAS: Revista de Estudos de Conflito e Controle Social Vol. 05, Nº 03, pp.447-482.
• MISSE, Michel (2010), Crime, sujeito e sujeição criminal: aspectos de uma contribuição analítica sobre a categoria bandido. Lua Nova Vol. 79, pp. 15-38.
• NEMACHECK, Christine L. (2017), Appointing Supreme Court justices. In: The Oxford handbook of U.S. judicial behavior. Edited by EPSTEIN, Lee and LINDQUIST, Stefanie A.. New York: Oxford University Press. pp. 29-46.
• OLIVEIRA, Fabiana Luci de (2012), Supremo relator: processo decisório e mudanças na composição do STF nos governos FHC e Lula. Revista Brasileira de Ciências Sociais Vol. 27, Nº 80, pp. 89-115.
• PERLIN, Giovana and SANTOS, Manoel Leonardo (ed) (2019), Presidencialismo de coalizão em movimento. Brasília: Câmara dos Deputados. 482 pp..
• PROSECUTOR RICARDO (2016), Entrevista concedida a Sara Prado. Centro de estudos em criminalidade e Segurança Pública. Belo Horizonte, março de 2016. Duração: 42m48s.
• RAMOS, Marcelo Maciel and CASTRO, Felipe Araújo (2019), Aristocracia judicial brasileira: privilégios, habitus e cumplicidade estrutural. Revista Direito GV Vol. 15, Nº 02, pp. 01-36.
• RAMOS, Silvia and MUSUMECI, Leonarda (2004), ‘Elemento suspeito’. Abordagem policial e discriminação na cidade do Rio de Janeiro. Boletim Segurança e Cidadania Vol. 03, Nº 08, pp. 01-16.
• RAMOS, Silvia; SILVA, Pedro Paulo da: SILVA, Itamar, FRANCISCO, Diego (2022), Negro trauma: racismo e abordagem policial no Rio de Janeiro. Rio de Janeiro: CESeC. 60 pp..
• RAUPP, Marianne (2015), As pesquisas sobre o 'sentencing': disparidade, punição e vocabulários de motivos. Revista de Estudos Empíricos em Direito Vol. 02, Nº 02, pp. 174-191.
• RIBEIRO, Leandro Molhano and ARGUELHES, Diego Werneck (2013), Preferências, estratégias e motivações: pressupostos institucionais de teorias sobre comportamento judicial e sua transposição para o caso brasileiro. Revista Direito e Práxis Vol. 04, Nº 07, pp. 85-121.
• RIBEIRO, Ludmila Mendonça Lopes; DINIZ, Alexandre Magno Alves, and LAGES, Lívia Bastos (2022), Decision-making in an inquisitorial system: lessons from Brazil. Law & Society Review. Vol. 56, Nº 01, pp. 101-121.
• RIBEIRO, Ludmila Mendonça Lopes and VILAROUCA, Márcio Grijó (2012), Quando devo fazer pesquisa por meio de entrevistas, e como fazer. In: Metodologia jurídica: um roteiro prático para trabalhos de conclusão de curso. Edited by QUEIROZ, Rafael Mafei Rabelo and FEFERBAUM, Marina. São Paulo: Saraiva. Pp. 228-253.
• SEGAL, Jeffrey. A. (2008), Judicial behavior. In: The Oxford handbook of law and politics Edited by WHITTINGTON, Keith E.; KELEMEN, R. Daniel, and CALDEIRA, Gregory A.. Oxford: Oxford University Press. pp. 19-33.
• SEGAL, Jeffrey A. and COVER, Albert D. (1989), Ideological values and the votes of U. S. Supreme Court Justices. American Political Science Review Vol 83, Nº 02, pp. 557- 565.
• SILVA, Diogo Bacha e and VIEIRA, José Ribas (2022), Os itinerários da politização do Supremo Tribunal Federal: do ativismo ao populismo judicial. Sequência Vol. 43, Nº 91, pp. 01-34.
• TOLEDO, Fabio Lopes (2020), 'O flagrante ganha voz'?: os significados da presença da pessoa presa nas audiências de custódia no Estado de São Paulo. São Paulo: IBCCRIM.164 pp..
• ULMER, Jeffery T. (2019), Criminal Courts as inhabited institutions: making sense of difference and similarity in sentencing. Crime and Justice Vol. 48, Nº 01, p. 483-522.
• VARGAS, Joana Domingues (2014), Fluxo do sistema de justiça criminal. Crime, polícia e justiça no Brasil. Contexto. Vol. 01, pp. 412-423.

• 1
  [Translator’s note] In Brazil, custody hearings, known as ‘Audiências de Custódia’ (in Portuguese), are legal proceedings conducted shortly after an individual’s arrest. These hearings serve as a mechanism to review the legality of the arrest, assess the conditions of the detainee, and determine whether the arrested person should be kept in custody (pretrial detention) or released pending trial. During the custody hearing, a judge does not assess the evidence or consider the nature of the alleged offense. Instead, their focus lies solely on evaluating factors such as flight risk and potential danger to society in order to impose pretrial detention.
• 2
  [Translator’s Note] A ‘guarantist’ perspective in the context of judicial decisions and legal theory refers to a viewpoint that places a strong emphasis on protecting individual rights, due process, and ensuring that individuals are not unjustly deprived of their liberty. Guarantism is often associated with an approach that leans towards granting pretrial release and reserving pretrial detention for cases where it is deemed absolutely necessary, to protect society or prevent a suspect from fleeing. In other words, it prioritizes the presumption of innocence and liberty over the use of incarceration.
  Conversely, a ‘punitive’ perspective, often contrasted with guarantism, leans more towards imposing stricter penalties, including pretrial detention, as a means of addressing criminal issues and protecting the interests of society. These terms are often used in discussions related to criminal justice and the role of judges in making decisions, particularly regarding pretrial detention and bail. Judges who follow a guarantist approach are more likely to prioritize individual rights and liberty, while those with a punitive stance may prioritize societal safety and the potential risk posed by the defendant.
• 3
  Interviewees are identified by fictitious names.
• *
  This article was presented at the 13th Meeting of the Brazilian Political Science Association (ABCP), which took place from September 19 to September 23, 2022, in an online format, and was named as the best paper presented in the subfield of “Politics, Law and Judiciary”.
• Data replication: https://doi.org/10.7910/DVN/1TMCZ8

Data availability

Data replication: https://doi.org/10.7910/DVN/1TMCZ8

Publication Dates

• Publication in this collection
  22 July 2024
• Date of issue
  June 2024

History

• Received
  09 Nov 2022
• Accepted
  24 Aug 2023