Prisons or Problem-Solving: Does the Public Support Specialty Courts?

ABSTRACT

Recently, “problem-solving” courts have been developed as an alternative to imprisonment. They are often called “specialty” courts because they process and divert into treatment programs offenders who are seen as different from the general criminal population, such as those with mental health or drug problems, those who are homeless or veterans, and those who engage in domestic violence. Based on a 2017 national survey of 1,000 respondents, the current study examines overall public support for rehabilitation as a goal of corrections and then focuses specifically on support for different types of specialty courts. The analysis reveals that the American public endorses not only the rehabilitative ideal but also the use of problem-solving courts. Further, with only minimal variation, strong support for these courts appears to exist regardless of political orientation and sociodemographic characteristics.

KEYWORDS:

• Problem-solving courts
• specialty courts
• drug courts
• public opinion
• rehabilitation

During the past three decades, “problem-solving” courts have proliferated in the United States (Boldt, 2017; Mitchell, 2011; Strong & Kyckelhahn, 2016)—with this innovation migrating to other societies as well (see, e.g., Bowen & Whitehead, 2016; Li & Liu, 2019; Nolan, 2012; Schaefer & Beriman, 2019). Also called “specialty” courts, the origins of this justice invention are usually traced to a drug court created in 1989 in Miami-Dade County, Florida, “first presided over by Judge Stanley Goldstein” (Nolan, 2012, p. 151; see also Berman & Feinblatt, 2001; Mitchell, 2011). Since that time, the number of U.S. drug courts has risen dramatically and currently is estimated at more than 3,000. In 2014, these tribunals served about 127,000 offenders (Marlowe, Hardin, & Fox, 2016). Equally important, the variety of these courts has steadily diversified, expanding in particular to focus on offenders with mental health challenges (Rossman et al., 2012; Sarteschi, Vaughn, & Kim, 2011). Today, the roster also includes community courts, domestic violence courts, fathering courts, gambling courts, gun courts, homeless courts, human trafficking courts, truancy courts, reentry courts, and veterans courts (Marlowe et al., 2016; National Center for State Courts, 2018)—with others undoubtedly to come (see, e.g., Steinberg, 2018). Veterans Treatment Courts have recently surged in popularity and now exceed 461 (Tsai, Finlay, Flatley, Kasprow, & Clark, 2018; see also Douds, Ahlin, Howard, & Stigerwalt, 2017; Douds & Hammer, 2019). Based on their Painting the Current Picture Survey, Marlowe et al. (2016) have calculated that “there was a grand total of 4,368 problem-solving courts in the United States as of December 31, 2014” (p. 9). Given the upward trajectory of the growth curve in these courts, the current number is likely much higher.

The problem-solving court movement was ushered in by the origin and growth of drug courts, which served as a model for subsequent variations (Berman & Feinblatt, 2001). By the late 1980s, the so-called War on Drugs had ensnared thousands of “low-level drug offenders” who “flooded the courts and crowded correctional facilities” (Mitchell, 2011, p. 848). “Many jurisdictions,” notes Mitchell (2011), “implemented drug courts out of sheer need” (p. 848). But need alone was insufficient to ensure its spread; the proposed model had to make sense—and it did, because it offered something for everyone. For liberals, the drug-involved escaped the harshness of a prison term and received effective treatment for their addiction. For conservatives, offenders in drug courts were closely supervised, held accountable, and ultimately punished for lack of program compliance. Costs were saved and public safety possibly enhanced. And for judges hamstrung by legislators mandating certain sanctions, they were granted discretion to use their expertise to ensure that the courts improved rather than damaged offenders (see also Winick & Wexler, 2002). As Mitchell (2011) puts it, “Compassionate but tough, drug courts were attractive to policymakers of all ilk” (p. 849). A strong financial incentive was offered as well: Between 1989 and 1997, the federal government allocated $80 million in funding (Mitchell, 2011). Five years after the Miami-Dade County court was invented, 472 drug courts were operating (Mitchell, 2011). Note that federal legislation also has designated tens of millions of dollars to encourage collaboration between state and local criminal justice and mental health agencies, including for the development of mental health courts (CSG Justice Center, 2016; Rossman et al., 2012).

In response to this major court innovation, a fairly large body of literature has emerged assessing the effectiveness and legal consequences of drug and other specialty courts (see, e.g., Boldt, 2010, 2017; Marlowe et al., 2016; Mitchell, 2011). Despite this attention, to our knowledge, no national-level study has been conducted that assesses the degree to which the American public supports problem-solving courts. This omission is striking, given that public opinion toward virtually every correctional policy has been studied either by polling companies and/or by academic researchers (see, e.g., Cullen, Fisher, & Applegate, 2000; Thielo, 2017; Thielo, Cullen, Cohen, & Chouhy, 2016). In this context, the current project examines public support for five prominent types of problem-solving courts: drug, mental health, veterans, homeless, and domestic violence. The study also probes whether Americans show a high level of consensus in their views or whether social cleavages exist in endorsement of this reform. As a prelude to presenting this analysis, we first review issues central to the operation of specialty courts.

Problem-solving courts: principles, practice, and perceptions

Underlying principles

As Boldt (2014) notes, there is no coherent theory underlying problem-solving courts. Ironically, these courts were themselves, as noted above, a reform aimed at solving the problem of a system overrun by drug offenders and ill-equipped to handle special offender populations. Faced with pressing concerns, observes Boldt (2014), this movement “has been shaped not by a foundational theoretical perspective but by an essentially pragmatic set of instincts” (p. 1125). Motivated by such “atheoretical pragmatism” (p. 1126), he continues, “legal and human services professionals” have been “guided by the concrete cues of everyday experience and not by the abstract direction suggested by a top-down theory” (p. 1125). The result is that problem-solving courts manifest diversity both within and across types of courts as local officials strive to improve the quality and effectiveness of their responses to offenders.

Despite the lack of a unifying and directive theory (such as the risk–need–responsivity [RNR] model in correctional rehabilitation; see Bonta & Andrews, 2017), problem-solving courts in any given jurisdiction are not invented anew from a blank slate. A pragmatic approach involves, at least to a degree, an attempt to model local practices after other courts in existence. As a result, these problem-solving courts tend to share a set of common principles. These have been described in various terms, but we will summarize them by identifying five core principles (see also Berman & Feinblatt, 2001; Bowen & Whitehead, 2016; Mitchell, 2011; Wolf, 2007).

First, there is the principle of diversion from imprisonment. In an observation that pertains to all specialty tribunals, Steinberg (2018) notes that “the crux of the drug court model is the treatment alternative to incarceration” (p. 1587). Incarceration is held to be counterproductive because it is inordinately costly and fails to address the unique characteristics and criminogenic needs that bring a special population of offenders into the justice system. A generic punitive sanction thus is not responsive to the problems of these offenders and is likely to be ineffective (Bonta & Andrews, 2017; Cullen, Jonson, & Nagin, 2011). Treatment in the community, not punishment behind bars, is the most pragmatic policy option.

Second—and this is the key idea to the enterprise—there is the principle of problem-solving. In correctional rehabilitation, the RNR model argues that the criminogenic needs leading to recidivism tend to be general (Bonta & Andrews, 2017). Offenders are differentiated not by offense type or unique status but by risk level (the risk principle, which argues that intervention should target higher risk offenders). By contrast, problem-solving courts separate offenders by a characteristic that identifies them as having a special need—such as substance addiction, mental health troubles, or service in the military. This model assumes that these special populations are sufficiently alike to one another and different from other offenders that they merit being pulled out of the criminal justice system. Exclusion criteria usually bar the truly dangerous from such special consideration. Otherwise, however, they have the option of selecting into a court arranged to understand and respond to the unique challenge they face. Importantly, the very name of the court—such as a “drug court,” “mental health court,” or “veterans treatment court”—identifies the problem that all those under its umbrella share. The specialized nature of the court makes it capable of not generically sanctioning offenders under its jurisdiction but of solving their problems.

Third, there is the principle of individualized treatment. These courts are marked by the traditional rehabilitative ideal (Allen, 1981; Cullen & Gilbert, 1982; Rothman, 1980). Although offenders may share a common problem (such as drug addiction), the mixture of factors causing that problem is potentially unique to any given offender. As a result, interventions must be individualized. Accomplishing this goal means that judges must have wide discretion to fit the treatment to the individual offender rather than simply apply a standard sanction to the crime in question. Judges abandon their role as an arbiter ensuring that adversarial parties abide by appropriate legal procedures in favor of “therapeutic jurisprudence,” in which their abiding goal becomes saving troubled offenders (Boldt, 2014; see also Winick & Wexler, 2002). To design the most efficacious response, judges also forfeit any pretense of impersonality and impartiality; instead, they actively collaborate with criminal justice officials (e.g., prosecutors, probation officers), human service providers, and even community members (Nolan, 2012). Such collaboration is held to be an important ingredient in finding a solution to each offender’s problem (Boldt, 2014).

Fourth, there is the principle of accountability. The traditional rehabilitative ideal has a failsafe mechanism for those who do not respond to treatment and threaten public safety—a custodial placement, often for a lengthy period of time (Rothman, 1980). In a similar way, notes Boldt (2010), problem-solving courts are marked not only by the offer of a “healing balm” but also by the threat of a “tomahawk” (p. 45). These courts were not marketed as form of bleeding-heart liberalism but as tough love—a factor making them politically viable (Boldt, 2010). In exchange for problem-solving treatment, offenders are informed that they will be monitored (e.g., drug tests) and experience graduated sanctions for noncompliance with court-ordered programmatic participation and conditions of behavioral conduct. Failure to complete the designated program during their time in the court could lead to a return to traditional court or to imprisonment. Note that coercion—the tomahawk hanging over their heads—is seen as an important tool to motivate offenders to take the necessary steps to address their problem. The obvious difficulty is that with offenders who are terminated from the program, punishment is likely to trump any further concern for their problem and its treatment (see Rothman, 1980).

Fifth, there is the principle of effectiveness. The core of pragmatism is that the courts “work” better than the alternative. “Specialized problem-solving courts,” observes Boldt (2014), “are said to work. They save money, they reduce recidivism, and they save lives” (p. 1129). Put another way, problem-solving courts must be able to solve problems! Otherwise, their legitimacy is in question. Notably, this innovation aligns with the “what works” movement in corrections and the appeal of evidence-based interventions (Cullen & Gendreau, 2001; MacKenzie, 2006).

Courts in practice

The primary policy issue is whether, when put into practice, problem-solving courts are in fact effective—especially in reducing offender recidivism. Overall, the evidence seems promising (Marlowe et al., 2016). Evaluations on drug courts are most plentiful. Systematic reviews and meta-analyses of this literature show that, overall, drug courts appear to be lower in recidivism versus standard sanctions (Logan & Link, 2019; Marlowe, 2011; Marlowe et al., 2016; Mitchell, 2011; Mitchell, Wilson, Eggers, & MacKenzie, 2012; Shaffer, 2011). The picture for juveniles is less clear, with results ranging from slightly positive to null (Stein, Homan, & DeBerard, 2015; Sullivan, Blair, Latessa, & Sullivan, 2016; Tanner-Smith, Lipsey, & Wilson, 2016). Mental health courts have also been evaluated with some frequency, with meta-analyses showing positive outcomes (Cross, 2011; Lowder, Rade, & Desmarais, 2018; Sarteschi et al., 2011). It should be noted, however, that the meta-analyses report considerable heterogeneity in effect sizes. Unlocking the “black box” to reveal what produces such variation remains at a beginning stage (see, e.g., Shaffer, 2011). Further, research on other types of problem-solving courts, though occasionally promising (Tsai et al., 2018), remains in short supply (Marlowe et al., 2016).

Similar to other criminal justice innovations (e.g., boot camps, intensive supervision programs, Project HOPE), problem-solving courts have been adopted widely based on limited empirical evidence (see Cullen & Jonson, 2017; Cullen, Pratt, Turanovic, & Butler, 2018). As a pragmatic enterprise, these courts are driven by local concerns and interests—and not by criminological theory or rigorous research (Boldt, 2014). Only in the most generous meaning of the phrase can problem-solving courts thus be called “evidence based.” Yes, some positive results exist—certainly enough for careful experimentation—but hubris too often has trumped humility. All types of problem-solving courts have been put into place prior to evidence demonstrating their effectiveness (Douds et al., 2017; Kulig & Butler, 2019; Marlowe et al., 2016). Further, even the fairly extensive empirical literature on drug courts lacks enough specificity to define the components of an efficacious, replicable model applicable to all jurisdictions and generalizable to other types of specialty tribunals (see, e.g., Shaffer, 2011). It is disquieting that within criminal justice, programs affecting the lives of offenders—and their potential victims—are implemented based on such shaky grounds. If pharmaceutical companies engaged in such practices, they would be accused of marketing untested products and end up in their own “drug court” (see Cullen, Manchak, & Duriez, 2014).

Advocates also do not consider conflicts inherent in the problem-solving model (Boldt, 2010, 2017). In the rehabilitative ideal, which is at the heart of these courts, a tension exists between treatment and control—between the desire to do good and save offenders versus a willingness to punish anyone who proves incorrigible (Cullen & Gilbert, 1982; Rothman, 1980). In specialty courts, participation and thus access to a treatment response depends on offenders confessing their guilt. This admission frees judges of their arbiter status in an adversarial process to instead assume, in essence, the lead role in a collaborative treatment team (Boldt, 2017). One difficulty is that judges are not required to have any training in criminology or in correctional rehabilitation, yet exuberantly exercise wide discretion in individualizing treatment within their court (Nolan, 2012). It is also not always clear whether providers of human services programming have behavioral change expertise and whether jurists can assess such capabilities.

A major policy challenge arises when participants fail to complete the course of treatment mandated by the specialty court. For example, on average, noncompletion rates hover near 50% for offenders in both adult and juvenile drug courts (Mitchell, 2011; Stein, DeBerard, & Homan, 2013). It might be possible to learn from failure to devise more effective programming (Douds & Hammer, 2019; see also Smith, Labrecque, Smith, & Latessa, 2012). Most often, however, judges respond to offender noncompliance by relinquishing their treatment persona in favor of the exercise of legal control (e.g., returning offenders to traditional courts, revoking their probation). One difficulty, as Boldt (2017) notes, is that this “augmented punishment” is being “ordered by a decision-maker whose capacity for formal fairness has been compromised by problem-solving informality” (p. 298). Another difficulty, also demarcated by Boldt (2014), is that the punishment of troubled defendants (e.g., suffering from drug addiction or mental illness) assumes that “these individuals are capable of making morally significant choices, either to adhere to program requirements or to depart from express rules and clear obligations” (p. 1151). It is as though a hospital is turning away its sickest patients and blaming them for failing to choose to get well.

Favoring punishment over continued treatment for noncompleters may have unanticipated consequences. In their meta-analysis of how drug courts influence the use of incarceration, Sevigny, Fuleihan, and Ferdik (2013) show that participants are less likely to experience imprisonment (“incidence”) because of lower recidivism rates. Despite this positive finding, drug courts do not decrease prison use overall because noncompleters receive longer terms behind bars than if they had opted for processing as usual. As a result, observe Sevigny et al. (2013), these courts “do not appear to reduce the aggregate, near-term burden placed on correctional resources” (p. 416).

Other limitations can be documented that speak to the challenge of translating the ideal of the problem-solving court into real world practice (see, e.g., Boldt, 2017). We will list just one more issue: racial and ethnic disparities. Analyzing drug court data from 2014, Marlowe et al. (2016) report that two-thirds (67%) of participants are White, whereas only 17% are African American and 10% Hispanic. These figures underrepresent the proportion of minorities for “arrestee, probation, and incarcerated populations” (Marlowe et al., 2016, p. 8). In some courts, completion rates for African Americans and Hispanics are much lower than for other participants. “Drug courts have an affirmative obligation,” conclude Marlowe et al. (2016), “to examine the reasons for these disparities and institute remedial measures to correct the problem” (p. 8; see also Boldt, 2017).

Public perceptions

The very proliferation of problem-solving courts and the lack of opposition to them suggest that the concept of specialty tribunals resonates with underlying views about corrections. Although the current study represents the first direct investigation of citizens’ support for problem-oriented courts, four aspects of public opinion are relevant to this claim. Recall that these courts are rooted in the rehabilitative ideal, are promoted as an alternative to incarceration, and offer a mixture of treatment and control. Research reveals findings consistent with these components of specialty courts.

First, since the mid-1990s, punitive sentiments in the United States have declined (Enns, 2014; Pickett, 2019; Ramirez, 2013a, 2013b). For example, between 1994 and 2017, support for capital punishment declined from 80% to 55% (Pickett, 2019), whereas between 1994 and 2012, support for harsher courts fell from 85% to 62% (Ramirez, 2013a). Pressure to “get tough” on crime has lessened, creating opportunities for reform (Cohen, 2017; Cohen, Cullen, & Thielo, 2019; Petersilia & Cullen, 2015). Second, the embrace of rehabilitation as a correctional goal has been strong for decades and remains so today (Cullen, 2013; Cullen et al., 2000; Thielo, 2017). Third, whether in Red States or Blue States, the public is open to community alternatives to imprisonment, especially for nonviolent offenders (Sundt, Cullen, Thielo, & Jonson, 2015; Thielo et al., 2016). Fourth, Americans tend to favor a mixture of rehabilitation and punishment in a correctional response to offenders (Cullen et al., 2000). Recently, this has been referred to as a “balanced” approach to sanctioning (see, e.g., Mears, Pickett, & Mancini, 2015).

In this context, two studies on public sanctioning preferences are particularly relevant—one regarding justice-involved veterans (Atkin-Plunk & Sloas, 2018) and one regarding drug offenders (Sloas & Atkin-Plunk, 2018). Based on a 2016 survey of 575 undergraduate students attending a university located in the South, Atkin-Plunk and Sloas examined which of the following options the respondents favored in the sentencing of nonviolent and violent veteran and drug offenders: (1) only or mostly punishment, (2) only or mostly rehabilitation, or (3) the “balanced” approach of both equally. In two separate articles, the researchers report findings similar for both types of veteran and drug offenders (Atkin-Plunk & Sloas, 2018; Sloas & Atkin-Plunk, 2018): For those in court for nonviolent offenses, more than 70% of the sample favored a rehabilitation-only response, with 4% or less preferring a punishment-only response. For violent offenses, the balanced approach was most preferred for both veterans and the drug-involved, with about 45% of the sample choosing this option (the second choice for veterans was rehabilitation at 27.7% and for drug offenders was punishment at 42.2%). Although the respondents were not asked specifically about problem-solving courts, the researchers recognized that these results have clear policy implications, given that such courts “embrace the balanced justice ideal” (Atkin-Plunk & Sloas, 2018, p. 12). Taken together with the general public opinion literature, Atkin-Plunk and Sloas’s studies thus suggest that the public is likely to endorse the use of problem-solving courts for different types of specialty populations.

Methods

Sample

The data for this study are drawn from a national survey of 1,000 respondents conducted between March 3 and 7, 2017 by YouGov America, Inc. YouGov employs opt-in internet survey methodology to conduct marketing, political, and current affairs polls—an approach that is increasingly standard in contemporary social science and criminological research. Essentially, potential respondents opt in or agree to complete surveys in exchange for earning points toward vouchers for “big brand” gift cards, such as Amazon, Macy’s, Old Navy, and Walmart (YouGov, 2018). YouGov has successfully attracted a panel of more than 2 million participants in the United States. For any given project, the company uses a sophisticated propensity score matching process to create a sample matched to respondents in other nationally representative surveys. The data are then weighted to increase the representativeness of the responses (for more information on this method, see McManus, Cullen, Jonson, Burton, & Burton, in press; Thielo, 2017).

In addition to questions that “customers” (such as our research team) include on a survey, YouGov requires all participants to answer a “basic battery of questions—socio and political profile items when they join the panel,” known as the “Core Profile Items” (YouGov, 2017, p. 1). This information is updated regularly. Along with the responses to the designed survey, these items are included in the SPSS file transmitted to the customer following the study’s completion. Selected Core Profile Items were used in the current project’s multivariate analysis.

Measures of public opinion

As a prelude to examining support for problem-solving courts, the current project first assessed global support for rehabilitation. This analysis is intended to gauge the extent to which the contemporary American public embraces offender treatment as a correctional policy. This issue was addressed in two ways.

First, for more than 50 years, surveys have assessed the goals of corrections. Starting with a 1967 Harris Poll, the following question—or a close variation—has been asked repeatedly (see Cullen et al., 2000; Cullen, Pealer, Fisher, Applegate, & Santana, 2002; Flanagan & Caulfield, 1984; Harris, 1968): “What do you think should be the main emphasis in most prisons—punishing the individual convicted of a crime, trying to rehabilitate the individual so that he or she might return to society as a productive citizen, or protecting society from future crimes he or she might commit?” The respondents were asked this question and given four possible responses: Punishing the individual, Trying to rehabilitate the individual, Protecting society, and Not sure. This question was then repeated, except that the respondents were asked: “Now, what do you think should be the second most important emphasis in most prisons?”

Second, drawing on previous research (e.g., Cullen et al., 2002), we include the responses to two statements assessing support for rehabilitation in the community and for chronic offenders: (1) It is a good idea to provide treatment for offenders who are supervised by the courts and live in the community; and (2) Rehabilitation programs should be available even for offenders who have been involved in a lot of crime in their lives. To rate these items, the respondents used a Likert scale ranging from 1 (strongly agree) to 6 (strongly disagree).

The core of this study consists of the items that examine support for problem-solving or specialty courts. Note that the survey used the phrase “problem-oriented” rather than “problem-solving.” This wording choice seems inconsequential, given that the nature of the courts is clearly described. The respondents were presented with this question stem:

Recently, a number of places have started what are known as “problem-oriented” (or specialty) courts. These courts deal with a specific kind of offender—such as someone using drugs. The goal is to try to treat the underlying problem (e.g., drug addiction) by placing an offender in a rehabilitation program in the community rather than sending them to prison for punishment. We would like to know the extent to which you would support or oppose this kind of problem-oriented court for each of the types of offenders listed below.

Five different types of specialty populations, corresponding to major problem-solving courts, were rated: (1) offenders who are addicted to drugs; (2) offenders who are mentally ill; (3) offenders who are veterans; (4) offenders who are homeless; and (5) offenders who engage in domestic violence. To express whether they endorsed a specialty court for each of these kinds of offenders, the respondents chose one of the following answers: strongly support, support, somewhat support, somewhat oppose, oppose, strongly oppose.

Sources of public opinion

The major purpose of this project was to assess the level of support for problem-solving courts. Still, a collateral issue is whether cleavages in such support exist along social or political lines or whether there is consensus in how Americans view this policy reform. To assess this issue, we combined into a single scale the five items that measured public opinion about courts for drug, mentally ill, veteran, homeless, and domestic violence offenders. Because the responses to each item ranged from 1 (strongly agree) to 6 (strongly disagree), possible scores extended from 5 to 30. The scale was coded such that high scores indicated more support for problem-solving courts. The Cronbach’s alpha for this 5-item scale was .789.

To assess possible attitudinal divisions, we included four standard sociodemographic variables: sex (1 = male, 0 = female), race (1 = White, 0 = other), age (measured as a continuous variable), and education (highest level completed, ranging from 1 = no high school degree to 6 = postgraduate degree). Given the salient role politics has traditionally played in criminal justice policy, two additional measures were included: party affiliation (1 = Republican, 0 = other) and political ideology (1 = conservative, and 0 = other). Informed by previous research (Haner, Cullen, Jonson, Burton, & Kulig, 2019), these measures were dichotomized to avoid loss of cases for those who answered not sure. This response is an option included in the YouGov core item for these variables, and it was chosen by 56 respondents for party affiliation and by 80 respondents for political ideology.

Finally, multicollinearity was assessed in two ways—examining VIFs and bivariate correlations among the independent variables. VIF values above 10 and correlations above .80 are indicative of multicollinearity (Weisburd & Britt, 2014). Both of these assessments revealed that multicollinearity was not a problem in the model, as VIFs ranged between 1.01 and 1.37 and no two independent variables shared a correlation above .484. All other OLS regression assumptions were met in the model.

Results

As can be seen in Table 1, the respondents believe that correctional sanctions should accomplish multiple goals, with a priority placed on rehabilitation and protecting society. For our purposes, it is instructive that nearly two-thirds of the sample selected rehabilitation as their preferred main emphasis (38.0%) or second most important emphasis (26.9%). These findings are consistent with the data reported in Table 2. Nearly 9 in 10 respondents (88.7%) agreed that it was a “good idea to provide treatment for offenders who are supervised by the courts and live in the community”—precisely what problem-solving courts do. More than 7 in 10 respondents (72.4%) even favor treatment programs for more chronic offenders—those “involved in a lot of crime in their lives.”

Table 1. Public support for emphasis (or goals) of prisons, by main and second most important emphasis (percentages reported).

Emphasis of prison should be:	Main emphasis	Second most important emphasis	Total
1. Punishment—punishing the individual convicted of a crime	15.5	24.3	39.8
2. Rehabilitation—trying to rehabilitate the individual so that he or she might return to society as a productive citizen	38.9	26.9	65.8
3. Protecting society—protecting society from future crimes he or she might commit?	38.0	40.1	78.1
4. Not sure	7.6	8.7	16.3

Table 2. Public support for rehabilitation.

Items	Percent agree
1. It is a good idea to provide treatment for offenders who are who are supervised by the courts and live in the community	88.7
2. Rehabilitation programs should be available even for offenders who have been involved in a lot of crime in their lives	72.4

Note: Agree includes respondents answering 1 = strongly agree, 2 = agree, 3 = somewhat agree (versus 4 = somewhat disagree, 5 = disagree, and 6 = strongly disagree).

The key findings are presented in Table 3. Three conclusions merit attention. First, the respondents clearly supported problem-solving courts across all five special offender populations. When the three support categories are combined (i.e., strongly support, support, somewhat support), the percentage of overall support is more than 80% for drug, mentally ill, veteran, and homeless offenders. Sixty percent of the sample even supports specialty courts for those engaging in domestic violence. Second, support for courts is especially high for offenders with military service. More than 4 in 10 respondents (40.5%) “strongly supported” veteran treatment courts, and nearly 9 in 10 (89.5%) supported this idea to at least some degree. Third, support was least pronounced for domestic violence offenders—a group that rarely inspires sympathy. But as noted, a majority of the public appears open to processing these offenders through a problem-solving court, assuming perhaps that they require specialized treatment.

Table 3. Public support for problem-solving (specialty) courts by offense (percentages reported).

Type of court	Strongly support	Support	Somewhat support	Somewhat oppose	Oppose	Strongly oppose
1. Offenders who are addicted to drugs	31.0	26.2	23.0	10.2	5.5	4.1
2. Offenders who are mentally ill	31.7	25.6	24.6	10.1	5.1	2.9
3. Offenders who are veterans	40.5	27.2	21.8	6.7	2.2	1.6
4. Offenders who are homeless	32.6	29.0	27.0	7.7	2.2	1.5
5. Offenders who engage in domestic violence	14.0	18.9	27.1	19.4	11.9	8.7

Finally, the multivariate analysis reported in Table 4 explores sources of public support for problem-solving courts. Males and those with more education are significantly more likely, and Republicans significantly less likely, to favor this reform. Still, two findings place these differences into context. First, the adjusted R² is negligible, with only 2.6% of the variance in support for problem-solving courts explained by the demographic and political variables in the model. Given that these factors have negligible effects, this finding suggests that there is substantial consensus in the public’s support of problem-solving courts. Second, this conclusion can be demonstrated by support for such courts among even Republicans—who are less likely to endorse this reform. Thus, cross-tabulations show that the percentage of Republicans expressing support for specialty courts is as follows: 78.9% for drug offenders, 80.8% for mentally ill offenders, 91.0% for veteran offenders, 85.8% for homeless offenders, and 53.2% for domestic violence offenders. No meaningful political divide on this policy appears to exist.

Table 4. Sources of public support for problem-solving courts.

Variables	b	β	p-value
Male	.580	.065	.040
White	.408	.043	.195
Age	–.001	–.005	.889
Education	.305	.104	.001
Republicanism	–1.115	–.106	.004
Conservatism	–.405	–.043	.240
R^2		.032
Adjusted R^2		.026

Abbreviations: b = unstandardized regression coefficient; β = standardized regression coefficient.

Discussion

The genius of problem-solving courts is that this reform embraces the model of the rehabilitative ideal and individualized justice. These ideas were instrumental in justifying the creation of the modern correctional system, including indeterminate sentencing, probation and parole, treatment programs, and a separate juvenile justice system (Cullen & Gilbert, 1982; Rothman, 1980). Based on positivist criminology, this approach starts with the common-sense observation that individuals go into crime for a reason—perhaps bad parents, perhaps bad peers, or perhaps a bad psychology. Punishing the crime rather than treating the person ignores this stubborn reality—that behavior has causes and will not change unless these factors are targeted for intervention (see also Bonta & Andrews, 2017). Once it is agreed that offenders’ conduct is not freely chosen but caused in some way, it is a short step to the conclusion that such criminogenic “problems” must be “solved.” Setting up courts to take on this challenge seems only logical. However, whether problems are best solved through a general therapeutic court or by specialized courts is a complicated matter. Dividing up offenders by their most obvious visible trait—for example, their drug addiction, mental symptoms, or military deployment—assumes that offenders within such categories are homogenous and have more in common with one another than they do with those treated in other courts (see Pratt & Turanovic, 2019).

Whether explicitly or implicitly, public opinion aligns with the logic underlying courts invented to address the needs of specialized offender populations. Our 2017 national survey suggests, first of all, that support for treating offenders remains strong (see also Cullen et al., 2000, 2002). Punitive sentiments still exist, and Americans want the system to punish and protect. But in conjunction with these preferences, the public also expects the correctional system to “correct”—to save offenders from a life in crime. It seems clear that a high percentage of the citizenry believes that problem-solving courts comprise a sensible way of achieving this goal. In fact, for most types of courts, there is a general consensus spanning sociodemographic groups and political partisanship that specialized courts warrant support.

This global conclusion, however, likely will be specified by subsequent research. In particular, three issues merit further inquiry. First, our data are too nonspecific to illuminate which type of offender will be allowed into problem-solving courts. This issue relates to whether a specialty court, as many do now, have exclusionary criteria. The key divide is likely between violent and nonviolent offenders. The difficulty is that assessing risk level is a complex technical task that goes beyond the current offense an offender may have committed (see Bonta & Andrews, 2017).

It is instructive that in the current study, 6 in 10 respondents showed a willingness to permit specialty courts for domestic violence offenders. Given this finding, would the public be open to expanding specialty courts to include a wider range of traditional street offenders, such as robbers and burglars? One consideration might be the extent to which these courts mandated a higher level of supervision, such as offenders wearing monitors. Recall the research discussed previously showing that for violent drug and veteran offenders, the public favored a “balanced” sanction—one that emphasized both rehabilitation and punishment (Atkin-Plunk & Sloas, 2018; Sloas & Atkin-Plunk, 2018). The point is thus how far the American public will go—and under what conditions—in permitting the problem-solving model to be applied to offenders. Might they even endorse an entire court system in which all offenders are categorized and then sent into a court designed to address their specific offenses and criminogenic needs?

Second, Veterans Treatment Courts (VTCs), which are expanding rapidly (Tsai et al., 2018), are particularly favored by the American public. As noted, nearly 9 in 10 of the respondents supported specialty courts for veterans, with 4 in 10 doing so “strongly.” In their survey, Atkin-Plunk and Sloas (2018) asked—for both violent and nonviolent crimes—whether “veteran offenders deserve to be provided access to rehabilitation programs?” On a 5-point scale ranging from 1 (definitely not) to 5 (definitely), they report mean scores of 4.40 for nonviolent offenders and 4.02 for violent offenders. It is thus likely that the public believes that military service should be rewarded with a more compassionate correctional response—or, as Atkin-Plunk and Sloas (2018, p. 6) term it, that such “offenders deserve treatment.” The next step would be to disaggregate the global status of veterans to see if attitudes are conditioned by this heterogeneity. For example, is military service in general seen as a mitigating legal factor or does the public judge those deployed to war zones more deserving of treatment than those who do not see combat? Support for VTCs may also be predicated on the image that veteran offenders are suffering from post-traumatic stress disorder brought on by exposure to disquieting battlefield experiences. Do sanctioning preferences vary by whether veteran offenders are diagnosed with PTSD? Further, being a veteran is a social status, not a criminological status. Veterans may suffer from drug addiction or mental illness and they may be homeless or abuse their family members. Is there a point at which the public supports diverting veterans, whose military service is in the past, to specialty courts designed to solve the specific problems that they are now manifesting (e.g., a drug court)?

Third, related to this discussion is the issue of attribution—what the public believes are the causes of the behavior in question (see, e.g., Cullen, Clark, Cullen, & Mathers, 1985; Unnever, Cochran, Cullen, & Applegate, 2010). Again, specialty courts are predicated on the assumption that offenders diverted to these tribunals have problems that are integral to their criminal participation. It is this view—that the choice of crime is bounded and not fully free—that underlies the problem-solving model. For each type of court, research could probe profitably how public conceptions about the criminogenic needs of the specialty population shape their endorsement of the model and preferences for intervention. For example, does the public attribute crimes by mentally ill offenders to their psychological disorder or to other risk factors (Link, Cullen, Agnew, & Link, 2016)? Or do Americans see substance abuse as a disease that compromises free choice, and does this vary by the type of drug involved?

Finally, the clear support for problem-solving courts comes with a hidden danger: Public opinion is unlikely to be a barrier to the further spread of such courts. If anything, policymakers can expect acquiescence to, if not enthusiastic endorsement of, the concept of courts targeting offender populations that can claim to be “special” in some way. If these offenders are portrayed as vulnerable and victims of circumstances beyond their control—such as those addicted to opiates or young women being sexually trafficked—then the persuasive logic underlying the problem-solving model will be difficult to resist (see Kulig & Butler, 2019). Advocates of specialty courts will, of course, see a receptive public as a positive development—as one less hurdle to surmount in the pursuit of legal reform. Still, a measure of caution is in order.

Just because problem-oriented courts can be implemented in the current context does not mean that they should be. As noted, the empirical support for specialty courts, though promising in some areas, is in short supply and insufficient to define clearly best practices for each type of tribunal. Further, when implemented, problem-solving courts have raised sticky justice issues and have experienced unanticipated consequences, especially with regard to those who fail to complete the court-ordered program. In the end, if the public is willing to give permission to undertake a legal innovation, reformers have a responsibility to embrace evidence-based practices and to be fully aware of how their intervention impacts those offenders under their stewardship.

Disclosure statement

No potential conflict of interest was reported by the authors.