The well-documented over-representation of people with serious mental illnesses in the criminal legal system has led to a proliferation of interventions at the national, state, and local levels over the past 20 years (Bonfine et al., 2020). There are now multiple “off ramps” for people with serious mental illnesses at various intercepts of the criminal legal system that are designed to prevent further penetration into the system. An increasing number of police agencies have Crisis Intervention Team (CIT) programs that develop partnerships between law enforcement and mental health services and train officers to recognize and safely respond to people experiencing a mental health crisis (Watson et al., 2017); more than 500 counties across the country have joined the Stepping Up Initiative to reduce the number of people with mental illnesses in jails; and the expansion of specialty courts, pretrial diversion programs, and tailored supervision models is well underway in many jurisdictions (Bonfine et al., 2020). As further evidence of the effort to create interventions that prevent criminal legal system involvement, communities are increasingly focused on investments at “Intercept 0”—the community services and crisis response systems that can intervene before any law enforcement contact or arrest occurs (Abreau et al., 2017).

Even with such reform efforts, people with serious mental illnesses continue to have disproportionate contact at all stages of the criminal legal system and experience worse outcomes, including a higher likelihood of incarceration for non-violent misdemeanors and longer lengths of stay in jail (Hall et al., 2019; Kaba et al., 2015). Criminal justice and public health scholars thus continue to conduct research unpacking why these disparities persist. Some of this scholarship has focused on police decision-making (Engel & Silver, 2001) or their “decision pathways” (Lum, 2011) as gatekeepers to the criminal legal system. More recently, building on the Sequential Intercept Model (Griffin et al., 2015; Munetz & Griffin, 2006), researchers have focused on how decisions occur across intercepts of the criminal legal system (Comartin et al., 2021) and within multiple, dynamic contexts (Wood et al., 2022). In our recent study of misdemeanor arrests among people with serious mental illnesses in four jurisdictions in the United States (Atlanta, Chicago, New York City, and Philadelphia), we conducted systems mapping exercises with criminal legal and behavioral health stakeholders that aimed to visualize the pathway of misdemeanor case processing in each of the sites. The results (reported on in Wood et al., 2022) demonstrate that each of the sites experienced shared contexts that shaped how, when, and where they intervene with people with mental illnesses: (1) the law and policy environment; (2) location of the behavior; (3) expectations of stakeholders; (4) knowledge of mental illnesses among criminal legal professionals; and (5) access to community resources.

Our purpose in this article is to examine the values—or drawing from Thacher—the “value pluralism” (Thacher, 2001) that shapes decision-making and animates the above decision-making contexts. This paper focuses on the orientations that each agency brings to the table with respect to criminal legal system involvement for people with serious mental illnesses. Two intersecting strands of literature inform our work: (1) general systems theory and the concept of “loose coupling” (Hagan, 1989), and (2) the focal concerns framework. With respect to general systems theory, scholars have debated the extent to which the criminal legal system is a “system” at all, given that agencies are organized sequentially, lack integration, and exhibit within-agency differences across different locations. However, Bernard et al., (2005) argue that general systems theory can be fruitfully applied to the criminal legal system, particularly if the system is considered as “loosely coupled.” A “loosely coupled” system is comprised of multiple, autonomous bureaucracies with low levels of interdependency with each other. Agencies may have different rules and mandates and be oriented toward different goals or focal concerns. This is evident in the case of the criminal legal system where multiple agencies often appear to be uncoordinated and incompatible even as they are united around at least one common goal of turning offenders into non-offenders or “closing cases so that they stay closed” (Bernard et al., 2005: 208). The concept of “loose coupling” is valuable for explaining interorganizational relationships across the criminal legal system. In combination with the focal concerns framework, it helps illuminate how there can appear to be little agreement on goals or how stakeholders should pursue them within the criminal legal system.

The focal concerns framework has become a dominant framework to explain disparities in criminal legal system sentencing by race, gender, and age. Originally developed by Steffensmeier and colleagues (Steffensmeier & Demuth, 2000; Steffensmeier et al., 1998) to explain judicial sentencing decisions, the focal concerns framework argues that judges’ sentencing decisions are based on three “focal concerns”: (1) blameworthiness (and the desire to impose a sentence that is “just” in relation to the severity of the crime); (2) protection of the community (related to preventing offending through deterrence or incapacitation); and (3) practical constraints (i.e., extra-legal factors related to organizational and individual considerations that may influence judicial decision-making). Further, the focal concerns perspective suggests that judges develop a “perceptual shorthand” (Hawkins, 1981) in considering these factors, since their time is limited and they rarely have complete information; this shorthand is often linked, perhaps implicitly, to race, gender, and age attributes.

The focal concerns framework has more recently been extended to apply to prosecutors (Spohn et al., 2001), policing (Ishoy and Dabney, 2018), and the juvenile justice system (Bishop et al., 2010; Ericson & Eckberg, 2016), with some variation in how each of these stakeholders understand and apply the three primary focal concerns based on their position within the criminal legal system. For example, Ishoy and Dabney (2018) argue that street-level police officers are more focused on short-term focal concerns because of the nature of their shift-work. Meanwhile, in their work on juvenile justice system processing, Bishop et al., (2010) argue that the social welfare orientation of the juvenile justice system has a bearing on focal concerns, forcing it to balance accountability more consciously with rehabilitation.

In this paper, we extend our work on the contexts shaping misdemeanor system decision-making (Wood et al., 2022) by examining the plurality of values guiding interventions among people with serious mental illnesses. We follow the proposal by Bishop et al., (2010:214) to integrate the focal concerns framework with the “loose coupling” perspective from general systems theory to ask, “Who are the decision makers at each processing stage? And ‘What orientations do they bring to the table?’” Specifically, we seek to understand whether stakeholders at different intercepts (police, prosecutors, public defenders, judges, and behavioral health service providers) share a common goal of reducing system contact for people with serious mental illnesses and how they articulate the “best approach” to do so.

Methods

Data Collection

This study conducted focus groups and in-depth interviews in four sites to obtain a rich understanding of how specific misdemeanor charges are used and processed in relation to people with serious mental illnesses in each of these locations. The four sites included Atlanta, Chicago, New York City (specifically, the borough of Manhattan), and Philadelphia. Misdemeanor charges of interest were selected based on prior analysis of administrative data in Chicago, Georgia, and New York, demonstrating that certain misdemeanor charges are overrepresented among people with serious mental illnesses (Compton et al., 2023; Compton et al., 2022), as well as follow-up discussions with stakeholders about the most common charges they see when encountering people with serious mental illnesses who are arrested. All cities examined criminal trespass and petit larceny/retail theft/shoplifting. Additional misdemeanor charges of interest varied by site and included the following: disorderly conduct (Atlanta), willful obstruction of a law enforcement officer (Atlanta), 3rd degree/simple assault (New York, Philadelphia, Chicago) and domestic battery (Chicago).

Building on the relationships developed from the study’s earlier systems mapping exercises, the research teams recruited a range of criminal legal and behavioral health stakeholders to join site-specific focus groups and in-depth interviews between December 2020 and April 2021. The study was reviewed and approved by the relevant Institutional Review Boards. A total of 14 focus groups and 3 interviews were conducted. Focus groups ranged from 4 to 11 participants. Interviews included one or two participants and were held in place of focus groups in Philadelphia due to participant availability. Focus groups and interviews were held via Zoom except for two held in-person with Chicago and Philadelphia police officers. Table 1 shows the location, stakeholder group, and number of participants for each of the study’s focus groups/interviews. Participants in Atlanta represented the city of Atlanta as well as Fulton County. In New York City, court operations are borough/county-specific, and all participants were from the borough of Manhattan. Mental health service providers participating in focus groups in Atlanta, Chicago, and Manhattan included both direct care providers and clinical supervisors and program leadership.

Table 1 Focus Group/Interview Participants across Four Study Sites
Full size table

A semi-structured interview protocol was developed with four “layers” of questions and adapted for each stakeholder group. The first three layers were focused on asking a series of questions for each of the charges, and associated behaviors of interest in a site. Questions prompted participants to articulate prototypical examples of behaviors related to the misdemeanor charges of interest, describe the options they have when handling such a case for a person with a serious mental illness and how they decide to proceed, assess what underlying problems drive case-specific behavior, discuss the intended outcomes when arresting or charging someone with the particular misdemeanor charge, and detail what different demands or pressures they face from other system stakeholders and impacted groups when handling these cases. Participants were probed to articulate whether and how COVID-19 or local policy changes (e.g., bail reform) impacted their options and decision-making. In the fourth and final layer, the interviewer or focus group moderators asked participants a series of broader questions about serious mental illnesses and misdemeanor case processing, including what change to the system they would introduce to improve outcomes for people with serious mental illnesses. The question guide was semi-structured in nature, allowing participants to generate narratives/stories to illustrate their perspectives. The four phases of the exercise were designed to occur sequentially, though facilitators worked to ensure that there was an opportunity for conversations and narratives to emerge organically. For example, participants sometimes referred back to comments about other charges or associated behaviors when discussing differences in how they handled case processing.

Data Analysis

All focus group and interview discussions were audio-recorded via Zoom or digital recording device and transcribed verbatim for analysis. Transcripts were de-identified and identifying pieces of information were removed and/or anonymized. The research team used a thematic analysis approach (Braun & Clarke, 2006). An initial codebook was developed based on notes and memos and the common themes and topics that emerged during the focus groups and interviews. The iterative process of codebook development was informed by a collective reading and discussion of the site transcripts and notes that were taken during the focus groups and interviews. A set of 23 codes was developed for use in Dedoose, a qualitative analysis software. Sub-codes or ‘child codes’ were developed for parent codes where appropriate and were modified for each site, such as instances where unique language was used to refer to different charges or system policies and procedures. Certain codes were aimed at describing features of a case or situation discussed, such as charge. Other codes were designed to capture aspects of a system’s functioning, such as changes to the system. Some codes captured aspects of decision-making in terms of processes and practices (e.g., gathering information about SMI, communication between stakeholders), or factors influencing decision-making (e.g., external factors such as program options). Another set of codes captured normative statements about participants’ roles or the system generally, such as beliefs about the system or moral judgments or beliefs about one’s role. The codebook also contained codes related to the experiences of people with lived experience, such as impact of CLS involvement.

Weekly research team meetings facilitated the process of expanding and refining the set of parent codes, and where relevant, their constituent sub-codes. Ultimately a codebook was finalized that included the code name, definition, and an illustrative quote from each site. Each site team then divided up and assigned transcripts to a primary coder whose job was to read through site transcripts and apply the codes in the form of a directed content analysis using Dedoose. To enhance reliability in the coding process, a sample of transcripts were blindly double-coded. Any coding discrepancies between researchers were resolved in consensus discussion. The research team then reviewed code reports and generated memos isolating the findings pertaining to various codes to facilitate further analysis. During the process of writing this article, the authors distilled the primary themes around which to structure the presentation of results. The data reported on here are drawn primarily from the focus groups with criminal legal system stakeholders and service providers. We compare the perspectives of police, prosecutors, defenders and mental health service providers (the final two of which are considered together because of their overlap). The perspectives of people with lived experience from the focus group in Manhattan are being considered separately as part of a forthcoming analysis that also includes data from 50 individual interviews with people with lived experience of mental illness and criminal legal involvement.

Results

The narratives generated across the four sites illuminate a series of tensions regarding how criminal legal system stakeholders process people with serious mental illnesses through the misdemeanor system. As the data below will make clear, criminal legal and behavioral health agencies approach the problem of overrepresentation with unique orientations that create points of alignment as well as ones of stark divergence. Three primary themes emerge: (1) the shared commitment across criminal legal system agencies (police, defenders, prosecutors, and judges) to “getting people out” of the system; (2) disagreement about the best approach to make good on that commitment; and (3) the limits of existing criminal legal system tools as well as larger social systems to meaningfully tackle the situations that put people with serious mental illnesses at risk for criminal legal system involvement.

The Shared Goal of “Getting People Out”

Police, defenders, prosecutors, judges, and mental health service providers all reflected a common goal of reducing the disproportionate contact that people with serious mental illnesses have with the misdemeanor criminal legal system. Although this goal may have taken on additional urgency during the COVID-19 pandemic with associated public health priorities, stakeholders at all sites described their agency’s commitment as preceding the pandemic and being part of longer-term shifts in attitudes about crime and punishment. For police, this goal was largely pursued by implementing strategies to avoid making an arrest for a misdemeanor offense. As the initial point of contact in a longer chain of decision-making, police noted that their first resort is rarely to make an arrest for low-level issues like criminal trespassing or disorderly conduct. Rather, their work is primarily short-term in nature, focused on resolving cases quickly and satisfying external stakeholders who exert pressure on them to take action. As a result, clearing public space, or “moving people along” to satisfy complainants, is often their initial strategy. Two central concerns motivate this strategy. First, in cases like criminal trespassing or disorderly conduct, officers are often responding at the request of business owners or other stakeholders who own property and want the person removed from their space. As police supervisors in Atlanta noted, since it is rare for there to be harm to a person or property in these cases, most business owners want nothing more than a warning to be issued so that the person does not come back to their business. This was reflected by one lieutenant who noted, “overwhelmingly our first option, our first move is not to arrest people for criminal trespass.” An officer in Chicago echoed this noting, “they generally just want the person away from them, away from their property… so optimally, we remove them from the situation, the complainant is happy.” Second, police officers described focusing on these short-term strategies of removal because they were not always convinced that alternative strategies would be more effective. For example, while noting that removing a person with a serious mental illness “doesn’t solve the problem,” a Chicago police officer also suggested that taking the person to the hospital was generally ineffective “because they’re usually released immediately and put back where they were before.” Meanwhile, making an arrest was also not seen as particularly effective given the likelihood that many of these misdemeanor cases would not be prosecuted. “I don’t have much faith that that would go beyond the initial court date, unless it was a serious crime” a sergeant in Chicago said. This theme was also reflected by police officers in Philadelphia, with one noting there were “no consequences to your [a person’s] actions at all.”

This is not to say that police avoid making arrests entirely in misdemeanor cases. Police noted that they face external pressures from a variety of stakeholders to take an enforcement approach in certain circumstances. In misdemeanor shoplifting cases, officers face complaints from business owners who want to know that police officers have “taken action.” A lieutenant in Atlanta described this:

[What can get] a lot of police departments in trouble is a business owner calling and saying, ‘I called the police. I pay taxes. Somebody stole from me, and they caught the person, I wanted to press charges and the officer let him go’…Police officers have pressure on them from [these] outside factors.

In cases involving bodily harm to a complainant, like simple assault, officers may similarly feel that their hands are tied if one person wants to pursue criminal charges. Officers may try to mediate between parties first but in some cases “you’ve got people demanding, ‘Hey, I want that person arrested’” [Atlanta lieutenant]. In other cases, especially those involving assaults on strangers or the use of a weapon, police officers consistently agreed that there was a bold line where it would be necessary to make an arrest in the interest of public safety. Although officers from all cities acknowledged that most people who are in mental health crisis are not violent, one Chicago officer also gave an example of what differentiates this type of case from the more common encounters detailed above:

It’s one thing, too, just to walk around muttering, threatening people under their breath. I don’t think we’re really arresting people for that. When you rise to the level of physically attacking someone, especially if you know you’re just a stranger to each other, that’s different.

When an arrest does occur, police transfer decision-making authority to the criminal court system, where stakeholders have a somewhat narrower range of options to bring a case to resolution. All stakeholders involved in the prosecution and adjudication of these cases agreed that rehabilitation and/or care and support is the goal for people with serious mental illnesses who are arrested on misdemeanor charges. Public defenders across all sites were, not surprisingly, oriented toward getting their clients access to help. Defenders believed that getting their clients access to resources that meet underlying needs may prevent future offending. For example, some defenders in Manhattan suggested that the right programming and treatment may be useful for breaking a cycle of criminal legal system involvement for someone with multiple petit larceny (e.g., minor theft) charges:

You would hope that they get that intervention that obviously breaks the cycle. So, whether it’s drug treatment or whether it’s counseling for mental health issues, or something as basic as trying to get the person into the shelter system or a means of legitimate financial support. I think that’s the strategy for the most part because those are all the areas that basically bring people, unfortunately, to that situation in life where they are basically boosting [i.e., shoplifting] for purposes of getting through the day.

As this quote suggests, public defenders and the mental health service providers with whom they collaborate were concerned with not only providing health services and treatment to people with serious mental illnesses but also with ensuring that underlying material needs and social determinants that often drive system involvement were being met (e.g., adequate income, housing). They viewed their clients’ problems not so much as individualized behavior that needs to be course-corrected, but as driven by the larger social context in which their clients find themselves. The central goal for defenders then, as one reflected in Philadelphia, is to avoid getting their client “sucked up into a system that is sort of unforgiving.”

Prosecutors across all cities were similarly focused on the goal of treatment, with the hope of intervening in the lives of defendants with serious mental illnesses to break a cycle of misdemeanor offending. They acknowledged that a punitive approach to misdemeanor justice for people with serious mental illnesses was not effective in addressing the underlying causes of their behavior. One Assistant District Attorney in Philadelphia described this in the following way:

Well, I mean, unfortunately, the traditional way of dealing with something like that, which is you know arrest, charge, conviction, sentence… doesn’t do anything to help the situation of the seriously mentally ill person. It’s not going to improve life for anybody.

Even so, prosecutors maintained a belief that criminal legal system involvement could catalyze a process of recovery through mandated diversion programs that could both reduce recidivism and help solve the problems of defendants. “I have seen firsthand that we can bring people back from the edge and a lot of us… are advocating for these options wherever we see them so that we can bring people to a place of health and safety,” said one Assistant District Attorney in Manhattan. Indeed, although prosecutors recognized that some critics think rehabilitation is the terrain of “bleeding heart liberals,” they likewise recognized that solutions rooted in diversion and rehabilitation were common-sense alternatives compared to the high fiscal and personal costs of incarcerating someone on a low-level offense. This calculus is discussed further in the following section.

Alternative Approaches to “Getting People Out”

Even as stakeholders were aligned in their commitment to reducing the disproportionate contact that people with serious mental illnesses have with the criminal legal system—and even though they never described a retributive impulse as guiding their decision-making—there was significant disagreement about how to best make good on this commitment. Our focus groups make clear that “rehabilitation” means markedly different things for different stakeholders depending on both the practical constraints they face in their jobs and the broader orientation and values of their agencies.

Police – Providing Temporary Remedies

Police officers approached the problem of how to get people with mental illnesses out of the criminal legal system from two perspectives—what they feel they can do as “the front-line triage” [Chicago police sergeant] and what they wish could happen further downstream in a better resourced system. As the initial point of contact in the criminal legal system, police officers shared that their options are fairly limited to taking people to the hospital or making an arrest. The current constellation of services in most cities led most officers to conclude that transporting people to the hospital is their best option when they encounter a person with mental illness. As one Chicago officer shared, “Optimally, get them to the hospital, remove them from the situation, complainant is happy, they receive the treatment and help that they want if they proceed to go forward with it.” Another Chicago sergeant expressed, “I don’t want them to go to jail. I just want to them to go to the hospital.” At the same time, officers in Atlanta shared that they may default to arresting someone because it is more expedient. One lieutenant reflected, “The reality is, a lot of times it’s easier to arrest someone than to do anything else. It’s easier to lock somebody [up], to write a ticket, book them into the jail, than it is to do other things.” He later went on to clarify that the time it takes people to get processed at the hospital means extensive wait times for officers, who end up justifying an arrest as a means to more expedient treatment. “A lot of times, folks adapt their frame of mind: ‘Okay, you can get mental treatment at the jail.’”

When police officers rely on the hospital, it has more to do with the practical exigencies of their work than a belief that taking someone to the hospital is the best way to get them out of a cycle of criminal legal involvement. Officers in Atlanta, Chicago, and Philadelphia all spoke about the hospital as a “revolving door” that had minimal impact on cycles of recidivism and did little to change outcomes for people with serious mental illnesses. One Chicago police sergeant framed the core issue:

There are absolutely zero resources that I’m aware of other than a hospital emergency room where these people can get any sort of treatment. So basically, it’s a revolving door between the police interacting with them because citizens are calling, going to an emergency room, not getting adequate treatment, and then [they] come right back to the street where they run into police again.

Officers wanted more intensive options for people, especially for individuals they encounter repeatedly. This included calls for facilities that provide ongoing care in Atlanta and Philadelphia, as well as suggestions for more mandated treatment or involuntary, civil treatment in Philadelphia and Chicago. One sergeant in Chicago spoke about the benefits of court-mandated treatment: “That’s one advantage I see of the criminal justice system. It’s either go to jail or you can go to the hospital and get treatment.” The challenge, from police officers’ perspective, is that they are only the first responders in a much longer chain of events that they rarely have a view into. “We’re a little part of this whole,” said the Chicago sergeant. “We get all the responsibility, but we are a tiny little part.”

Prosecutors – Leveraging the System to Activate Treatment

As described above, prosecutors across all cities were oriented toward rehabilitative goals for low-level misdemeanors, arguing that retribution is generally not an appropriate justification for such cases. This was especially true in cases like criminal trespassing and certain instances of retail theft, where there is no motivated complainant or where charges are “so minor” that prosecutors and judges don’t see “societal vengeance” [Atlanta prosecutor] as a necessary goal. Rather, prosecutors promoted diversion as a way to “make people get the healthcare they neglected to get before” [Manhattan Criminal Court Supervisor] and, in so doing, reduce repeated system contact and incarceration. Prosecutors utilize “accountability courts” and “problem-solving courts” that provide treatment and strategies to address underlying mental health issues they perceive as driving offending behavior; such strategies aim for “corrective behavior change.” An Assistant District Attorney in Philadelphia similarly described using treatment courts for “behavioral correction”:

We won’t have to actually prosecute them in trial and have them found guilty. We can keep them in the court system, and that’s I think one of the big benefits in cases like that. It’s not just they, you know, dismissed the charges or let’s go to trial. We’re kind of that middle ground and say listen, I want to keep this here for a while, have them engage in treatment, come back to you and you know, at the end of the day, we all get what we want where hopefully this person will be in a better place and will continue to take their medication and at the same time not have a criminal record as a result.

Prosecutors admitted there are challenges to using diversion and programming as a primary approach. First, there may be tension with complaining witnesses who want a harsher response. This was discussed by prosecutors in Chicago with respect to repeat trespassing cases. “Sometimes you get a little bit of pushback from victims and complaining witnesses who maybe don’t understand that… putting [the defendant] in touch with the mental health program is going to get them services that they need to stop offending rather than just sending them to jail. And victims oftentimes just focus on the jail time or conviction.” Second, there may be individuals who do not want to engage in treatment, who refuse to take an offer of diversion, or who repeatedly get arrested. In these cases, prosecutors may feel compelled to seek jail time for the defendant on the basis of incapacitation or, alternatively, to adjourn the charges in contemplation of dismissal because they worry about the detrimental effects of keeping a case open (e.g., failure to appear in court and related consequences). Nonetheless, prosecutors in all cities described their preference for seeking diversion on the basis that they believed it could connect someone with needed services, reduce recidivism, and clear cases. “I think it kind of benefits the system in the long run,” noted a criminal court supervisor in Manhattan. “Because then we have less cases to deal with because hopefully, they’re not getting rearrested and it’s less work for us too.”

Defenders and Service Providers – Protecting the Right to Self-Determination

In contrast to the general approach articulated by prosecutors—where diversion programming is pursued to both connect people with services and enforce court compliance—public defenders and service providers described an approach to rehabilitation rooted in minimizing court interaction and connecting clients with tailored services that align with client choice and self-determination. Defenders across all cities articulated how the requirement to do what is best for their client often means rejecting an offer of diversionary programming for low-level misdemeanors. In their view, mandated treatment and diversion agreements often set up requirements that defendants with serious mental illnesses, particularly those also facing homelessness or comorbid substance use, do not have the resources or structures to effectively meet. One Chicago defender described their client’s thinking about mandated treatment; “that’s a lot hanging over me. I just want time served.” An Atlanta defender similarly shared the calculus, “[Do you want] time served, or do you want to get a year’s probation on the condition you don’t ever get high again?” Thus, as much as they want to help their clients access resources, they tend to reject mandated treatment programs they believe will set their clients up for failure. One Manhattan defender explained this as follows:

I mean, we’d like to help very much. But [you do] whatever is ultimately going to be best for the client in the moment and having less court interaction—if it’s a matter of that versus time served—it’s always going to be time served that wins the day.

Defenders may still connect their clients with services, but they do so outside of the court’s mandate. A Chicago defender shared that their best-case scenario is one where “the case is dismissed, and our client gets some help.”

Defenders also described how their defense strategy is guided by the choices a client wants to make about their own case. As one Manhattan defender put it, people with lived experience of a serious mental illness, “especially people who are charged with [petit larceny] a lot, they know what’s best for them and they’re able to accurately say what works for them and what doesn’t.” Another defender from Manhattan shared the following:

At the end of the day, whether it was programming or not, it really has to come down to the client’s choice… It’s not for me to try to convince, persuade, or cajole or do whatever it is, thinking that I’m doing the best for the client, that’s not my role, that’s not my function.

For this reason, when defenders do pursue treatment, it is generally done with the framework of self-determination.

Some Limits of Existing Criminal Legal Tools for “fixing the person”Footnote 1

As the above section makes clear, criminal legal and mental health stakeholders bring different orientations to the agenda of reducing criminal legal contact among people with serious mental illnesses. They all recognized, however, that existing approaches within the criminal legal system present a series of seemingly intractable challenges. Two challenges emerged as particularly salient among the many listed by stakeholders. First, as articulated primarily by public defenders and mental health service providers, using the criminal legal system to connect people with treatment inevitably involves net-widening as more people with serious mental illnesses get entangled in the system and face a host of collateral consequences. This can happen even very early in case processing, as Manhattan service providers shared when describing the challenges with New York City’s Supervised Release program, which provides court-ordered programming for individuals with pending cases in Criminal or Supreme Court (as an alternative to remand, setting money bail, or releasing on own recognizance (ROR)). “Supervised release is the new ROR,” one commented in discussing how it is often being used instead of ROR now. Another noted how Supervised Release can create a “domino effect after the arraignment” in terms of the complications it can create with terms of compliance. Chicago service providers described similar challenges that arise when people with serious mental illnesses fail to comply with onerous court conditions and are therefore incarcerated for non-compliance.

Second, the argument that treatment is the best way to “stop the offending” rests on the tenuous assumption that the treatment defendants receive is quality treatment targeted to their needs and that treatment alone is sufficient. This assumption was challenged even by prosecutors, with one Philadelphia Assistant District Attorney reflecting that the quality of court-ordered service provision varies, from “fabulous” to “more dehumanizing and more degrading than incarceration would be.” Further, many available treatment programs through the criminal legal system do not take a holistic approach to the needs of people with serious mental illnesses who are entangled in the system—or cannot because of the general lack of resources available. Prosecutors recognized that their reach was limited because diverting someone to treatment didn’t necessarily mean that the person would get more immediate needs met like housing, jobs, access to resources, and community supports. Defenders and mental health providers working within the criminal legal system were likewise frustrated that solutions were not readily available. As one community support worker in Chicago noted, “I’m not a magician. I don’t suddenly have housing and treatment tomorrow. Sometimes they hear you’re a mental health caseworker and they think that you can fix everything instantly or have resources suddenly that don’t exist.” Thus, it is not only that mental health treatment alone is not enough to reduce the involvement of people with serious mental illnesses in the criminal legal system, but also that broader solutions targeting social determinants of health are not readily available.

Discussion

These findings extend our work on system-level decision-making (Wood et al., 2022). They also further work by Bishop et al., (2010) in demonstrating how the integration of the “loose coupling” and focal concerns perspectives can be useful in thinking about who is making decisions at each stage of the criminal legal system and what orientations they bring to the table. Several points of discussion are noteworthy. Our findings make concrete how the criminal legal system is indeed a “system” but a loosely coupled one. Across the four study sites, we found a shared commitment to reducing the entanglement of people with serious mental illnesses in the criminal legal system, an objective that sits within the larger goal of the criminal legal system to “close cases that stay closed.” But this is a challenging goal, particularly when it must be squared with the desire to also satisfy victims and the public. As our stakeholders demonstrate, there may be multiple, overlapping, and even contradictory ideas about how to achieve this common goal and it can appear that the various agency professionals are incapable of finding common ground (Bernard et al., 2005). The result is often adversarial relationships (as one Manhattan Criminal Court Supervisor reflected: “Why is that a fight we have to have with everybody to pair someone with services that might change their life?”). The concept of loose coupling informs us that extra-legal characteristics like race are more likely to impact decision-making in loosely coupled stages of the criminal legal system where there are diverse organizational players with different goals and values (Bishop et al., 2010; Ericson & Eckberg, 2016). Bishop et al. (2010) found that in early stages of juvenile justice processing (e.g., intake), decision-making frequently lacked internal consistency—influenced by multiple players who allowed an array of contextual variables to come into play. Seen from this perspective, mental illness might also be understood as an extra-legal characteristic that influences how stakeholders make decisions, with different understandings of mental illness and its relationship to underlying offenses driving different goals and approaches across different agencies.

The focal concerns framework further elucidates how a loosely coupled system brings different decision makers with different orientations and concerns to the table. As originally framed with respect to judicial decision making, three primary focal concerns are central within the criminal legal system: blameworthiness, protection of the community, and practical constraints (Steffensmeier et al., 1998). Our data suggest that in misdemeanor cases involving people with mental illnesses, police, prosecutors, and defenders all grappled with the best approach for protecting the community and with the practical considerations they should consider when handling these types of cases. Police and prosecutors, for example, were careful to note the different kinds of misdemeanor assault, describing how assaults on strangers crossed a line and would require a more punitive approach. Meanwhile, defenders were skeptical of treatment mandates because they worried that a client with serious mental illness would violate the terms of the mandate and then be further ensnared in the system. The strategies criminal legal system professionals choose to resolve cases reflect how they balance the concerns of their stakeholders as well as their own understanding of how a particular course of action in the criminal legal system (e.g., diversion to treatment, dropped charges) will affect an offender in the long term.

Our study also pushes the concept of blameworthiness in the focal concerns literature. Sentencing research generally demonstrates that seriousness of the offense (alongside crime wrongfulness and harmfulness, defined in various ways) is the most important factor influencing criminal legal stakeholders’ views of blameworthiness (Steffensmeier et al., 1998). Given the low-level misdemeanor charges that were the focus of this study, it is not surprising that ideas of blameworthiness were muted. Police recognized that it was not always in the interests of either complainants or the subjects of complaints to arrest people with serious mental illnesses. And prosecutors conveyed that retribution is generally not appropriate when someone is arrested on misdemeanor charges. However, given that blameworthiness can also be influenced by a subject’s personal factors, it is worth considering how the fact of mental illness could increase or mitigate perceived blameworthiness. Our data suggest that both can occur. On the one hand, defenders described their concern that the fact of mental illness led to further entrenchment in the system through mandated treatment or supervised release that could result in myriad negative, collateral consequences. Studies demonstrate that youth with mental health problems are more likely to be sent to correctional confinement (White, 2016) and that adults with serious mental illnesses who are arrested on misdemeanor charges are more likely to receive a jail sentence after controlling for other case characteristics (Hall et al., 2019). On the other hand, stakeholders across the criminal legal system in our study shared the perspective that it is a priority to divert people with mental illnesses out of the criminal legal system and to pursue treatment options. This approach suggests that stakeholders may feel that people with mental illnesses may have less culpability for their criminal legal system involvement or that a social welfare approach is more appropriate in these types of cases. Future research should further investigate how perceptions of mental illness influence decision making and whether such decisions are being made because of negative stereotypes associated with mental illness such as dangerousness and unpredictability (Link et al., 1999) or because of a therapeutic mission.

The concept of blameworthiness in turn raises important questions about how stakeholders across agencies understand the relationship between mental illnesses and offending behavior. To the extent that criminal legal system stakeholders see untreated (or inadequately treated) mental illness as the direct cause of offending behavior—and appropriate treatment as capable of reducing problematic behavior—they are likely to see diversion to treatment as the appropriate course of action for people with serious mental illnesses. Indeed, scholars have described how the criminalization hypothesis led to “first-generation” mental health and criminal legal interventions united by the philosophy that criminal legal involvement of people with serious mental illnesses will be reduced primarily by providing these individuals with mental health treatment (Epperson et al., 2011, 2014). But this may miss larger questions about what approaches work for whom and under what circumstances. A growing body of research suggests that the relationship between mental illness and crime is confounded by other factors such as poverty, homelessness, and substance abuse—that people with serious mental illnesses encounter the criminal legal system for many of the same reasons as people without serious mental illnesses (Draine et al., 2002; Fisher et al., 2006; Epperson et al., 2014). In this sense, a narrow view of treatment solutions that are disconnected from larger systems of social support is likely to fall short in meeting the needs of this population or to meaningfully reduce the prevalence of people with serious mental illnesses in the criminal legal system.

The findings of this study suggest that we must also look across stages of the criminal legal system and consider strategies to align stakeholders with such differing values and goals. One possibility, as we develop elsewhere (Wood et al., 2022), is to bring together stakeholders across points in the system to participate in scenario-based exercises where they walk through the processing of “typical” misdemeanor cases involving people with mental illnesses to understand each actor’s decision-making contexts and the focal concerns that influence their perspectives and decisions. Acknowledging the value pluralism (Thacher, 2001) inherent in the criminal legal system and providing space for decision-makers to articulate the orientations of their agencies may create opportunities to identify new points of intervention and strategies that align values and advance justice and equity for people with serious mental illnesses.

This study is not without limitations. First, although the sample size is appropriate for a qualitative study of this type, the study is limited in only examining the experiences of stakeholders in four cities (and with a relatively small number of representatives for each stakeholder group in each city). Atlanta, Chicago, New York, and Philadelphia are all large cities with robust mental health service systems and relatively advanced approaches to tackling the overrepresentation of people with mental illnesses entangled in the misdemeanor criminal legal system. The perspectives here may therefore be more reflective of the competing perspectives in large, urban areas and miss important differences that shape other settings, including those in suburban and rural communities. Second, the original study design did not include focus groups with people with lived experience of mental illnesses given its focus on system processing and professional decision-making. Although a focus group was added in Manhattan to hear directly from people with experiences of mental illness and criminal legal system involvement, further in-depth work in this area is needed. An upcoming analysis looking at the direct experiences of people with mental illnesses as they interact with criminal legal system professionals will make use of this study’s data alongside interview data from a separate study with 50 people with mental illnesses and histories of arrest for misdemeanor charges. Third, this study does not adequately address how serious mental illnesses and race intersect in shaping the focal concerns of various criminal legal system stakeholders, especially given that the focal concerns literature directly engages with how concepts of blameworthiness, protection of the community, and practical constraints account for racial disparities in the criminal legal system. Additional work is needed that probes further into the ways in which structural racism influences decision-making about people with mental illnesses in the criminal legal system (Vinson & Dennis, 2021). Finally, this study took place during COVID-19, which may have affected case processing decisions in ways that we did not fully account for in this study. However, participants were asked to describe typical case processing versus those made in response to COVID-19. Future research should examine whether more durable changes were made to criminal legal system case processing as a result of the pandemic.

Future scholarship is needed to elaborate on the findings of this study and consider the processing of cases and people within the criminal legal system from a systems-level perspective. There are multiple initiatives across the country that aim to reduce criminal legal system contact among people with mental illnesses. Too often, however, these initiatives are siloed within discrete stages of the criminal legal system or designed by one set of stakeholders with little consideration for how adjacent stakeholders will react. Research that examines interagency collaborations and makes explicit the values and goals that different agencies bring to the table would be helpful in illuminating what might be required to implement interventions that meaningfully reduce the number of people with mental illnesses entangled in the criminal legal system.