Discourses of Mass Probation: From Managing Risk to Ending Human Warehousing in Michigan

Abstract

Over the past decade, some Western countries have begun to re-embrace the language of rehabilitation and calls for penal moderation. Risk logics—which undergirded the rise of mass incarceration in the U.S.—are now being repurposed to call for decarceration. Yet while risk played a key role in the transformation from modern to post-modern punishment, its development remains poorly understood. This article explores the discourses and practices of risk from the 1970s through to 2014 in one U.S. state (Michigan). The analyses focus on probation, the primary alternative to prison. The results show that risk discourses and practices emerged in the 1970s as a mode of resistance to the prison boom and have been adapted in each subsequent decade to address state governing crises.

Introduction

Over the last half-century, the Western world has been increasingly characterized as a “risk society” (Beck 1992) driven by the logic of neo-liberalism. The epicenter of this transformation has been the U.S. criminal justice system, which experienced a massive expansion in the severity of penal sanctions and a shift from the “penal-welfarist” model of punishment to a “culture of control” focused on managing risk (Garland 2001; Simon & Feeley 1992; though see O’Malley 2000; Hutchinson 2006). Despite this shift in the meta-narrative of punishment, scholars across the West have documented that rehabilitation persisted throughout the punitive turn, suggesting a “gap” between the discourses and practices of punishment (e.g. McNeill et al. 2009; Phelps 2011; Robinson 2008). More recently, scholars have begun to describe the emergence of new discourses of punishment, including the “cheap on crime” ethos (Aviram 2015) and the return of rehabilitation, now justified through the language of reducing risk (Green 2013; Klingele 2016; Rengifo, Stemen, & Amidon 2017; Robinson 2008; Steen, Lacock, & McKinzey 2012). These shifts have sparked debate about the potential and pitfalls of risk narratives to challenge the new punitiveness in the U.S. and beyond (e.g., Gottschalk 2015; Klingele 2016; O’Malley 2008; Robinson 2002; Seeds 2016).

While there is an extensive literature on the transition from penal-wefarism to the new penology (and critiques of that account), we know very little about the pre-history of risk logics in the criminal justice system and research on the contemporary role of risk-based reform is nascent. As Goodman, Page, and Phelps (2017) argue, looking only at snapshots of dramatic rupture moments in penal history distorts our vision of how and why discourses and practices change (see also Rubin 2017). In addition, as O’Malley (2008) has documented, risk is a malleable concept and complex governmental technology that can be harnessed for multiple and contradictory aims (see also Robinson 2002). The impact of policy developments can only be judged after evaluating how risk is deployed. Thus, without a full account of the pre-history, build-up, and contemporary shifts in risk management practices, we may misjudge its potential.

This article addresses this missing history, analyzing the discourses and practices of risk throughout the prison boom and recent reform efforts (~1974–2014). To trace risk discourses and policy, I focus on a critical, but under-researched, area of the criminal justice system—the expansion of mass probation. In the U.S., probation provides supervision in the community in lieu of imprisonment (in contrast to parole, which is community supervision after release from prison) and represents the largest form of criminal justice supervision. At its peak in 2007, 1 in every 61 adults (and 1 in 15 black men) was on probation (Phelps 2017). Designed and promoted as a more rehabilitative alternative to prison, the persistence of probation across the Western world during the punitive turn is puzzling (Robinson, McNeill, & Maruna 2013). As Robinson (2002) argues with regard to England and Wales, probation is thus “deeply implicated in the purported shift toward a ‘new’ or ‘postmodern’ penal rationality with risk management rather than normalization at its heart” (Robinson 2002: 5). However, due to the focus on prisons rather than probation in international punishment studies (Robinson 2016), we have no long-term historical studies of probation adapted through these different eras.

Given the variation across the U.S., any historical study of punishment must focus on individual states’ trajectories (Barker 2009). The case for this study is Michigan, a large Midwestern state with a prison population equivalent is roughly half the size of that for England and Wales (Walmsley 2016). Relying on archival documents, I trace how the state mobilized key discourses and policies to construct a plausible account of probation’s purpose and, by extension, corrections writ large. I focus in particular on the discourses and practices related to risk and rehabilitation (including programs and services, risk assessment tools, and revocation rates). I find that throughout this history, understandings of “risk” as applied to individuals in the criminal justice system remained surprisingly constant: risk represented the chance of future re-offending, particularly for violent offenses. So too did discourses about the department’s responsibility for managing risk and probation’s role as a cost-effective supervision strategy for “low risk” cases remain persistent themes.

Yet the ways in which the state deployed risk (and for what purposes) varied considerably—discourses of risks and costs and their associated practices were adapted to fit the unique conditions of each governing crisis du jour. Most surprisingly, these discourses of cost-risk calculations initially emerged not as a hand-maiden to mass incarceration, but rather as an attempt from criminal justice administrators to resist legislators’ “tough on crime” policies, with probation serving as a “test” of individuals’ suitability for freedom. By the 1990s, the logic of probation as a cost-effective alternative to prison had become entrenched, as probation facilitated long prison sentences for the “truly dangerous.” In recent years, risk discourses have been deployed to justify reforming “warehouse prisons,” with the state increasing diversion to probation and prison releases and reducing probation revocations to prison. The evolution suggests that risk logics are compatible with multiple, overlapping, and contradictory penal agendas. It also shows that bureaucratic actors can wield risk logics to modestly reduce the prison population, but risk alone is unlikely to radically disrupt the new penology vision.

From Rehabilitation to Risk (and Back)?

Across diffuse social domains, scholars have argued that many Western countries have moved toward a “risk society” (Beck 1992). This is particularly apparent in the U.S. criminal justice system, which experienced a massive expansion in punitive policy in the wake of increases in serious violent crime in the 1960s (Garland 2001). As imprisonment rates exploded in the U.S. and expanded more modestly in some European countries, scholars debated whether we were entering a unique “neo-liberal” or “post-modern” period of punishment, or “culture of control,” defined by an obsessive drive to identify, monitor, and contain risky individuals and dangerous situations (Garland 2001; Hallsworth 2002; Pratt 1997). These accounts share a focus on increasing punitiveness, discourses of individual responsibility and citizens as the “customers” of justice, and a retrenchment of therapeutic or rehabilitative aims. Feeley and Simon (1992) describe this transformation as the new penology, arguing that the criminal justice system shifted from rehabilitating criminal offenders to managing public safety risks, regulating “levels of deviance” rather than seeking to “intervene or respond to individual deviants or social malformations” (1992: 452).

This managerial perspective was assisted by new technologies, including drug testing, electronic monitoring, half-way houses, and the expansion of imprisonment which allowed for long-term “containment” of social refuse. Perhaps the most crucial of these new technologies was the development of risk assessment tools, which promised a “scientific” approach to measuring the probability of recidivism in lieu of subjective assessments of “dangerousness” (Hannah-Moffat 2013). Adapted from actuarial tools developed by the insurance industry, risk assessment tools first entered the criminal justice system to select promising parole candidates, gaining mainstream prominence in the 1990s and later adopted for a variety of different decision points, including bail and pre-trial detention, civil commitment, criminal sentencing, prison security level, and community supervision intensity (Harcourt 2007; Hannah-Moffat 2013; Simon 2005). Key questions, however, remain about their implementation and validity (Taxman 2016; Prins & Reich 2017). In addition, ambiguities persist about the overlap between recidivism risk and dangerousness, since property offenses tend to have the highest repeat conviction rates (Grattet, Petersilia, Lin, & Beckman 2009).

The new penology and related perspectives have been sharply critiqued for their characterization of a clean break from penal welfarism to risk management, with scholars arguing that penal practices are instead “braided” (Hutchinson 2006) or “layered” (Rubin 2016), with a “governmentality gap” between broad-scale discourses and local practices (McNeill et al. 2009; see also Cheliotis 2006; Goodman 2012; Hallsworth 2002; Hutchinson 2006; McCorkle & Crank 1996; Miller 2001; O’Malley 2000; Phelps 2011; Robinson 2008). Maurutto and Hannah-Moffat (2006) explain this gap through the concept of assemblages, arguing that “new penal technologies combine, merge and continually reassemble risk with other logics in response to various institutional agendas” (439). As Robinson (2008) finds in the Anglo-Welsh context, rather than theorizing the death of rehabilitation, scholars ought to consider its evolution, positing that late-modern rehabilitation gained legitimacy by articulating utilitarian, managerial, and expressive purposes.

From this perspective, the (re-)emergence of “ “penal optimism” (Green 2013) in the late 2000s and 2010s should not be surprising. Maurutto and Hannah-Moffat (2006) show, for example, how risk assessment tools came to incorporate a dynamic view of individuals “risks” and “needs,” creating space for treatment interventions designed to reduce risk. Under this framework, individuals are responsible for their own saving and any state intervention in this project is justified through public safety (reducing “risk”) rather than individual transformation (Bosworth 2007; Goodman 2012; Gottschalk 2015; Hannah-Moffat 2013; Klingele 2016; Robinson 2008; Steen, Lacock, & McKinzey 2012). However, in a notable shift since the 2000s, the state has become a more active participant in this process, reframing high revocation rates as departmental failure rather than solely supervisees’ individual responsibility (Robinson 2002; Klingele 2016; Rengifo, Stemen, & Amidon 2017).

The Persistence of Probation

Although much of the research on the hardening of the penal state focuses on imprisonment, probation—the “ideal penal type” of the Progressive model (Simon 2013)—is crucial to understanding this shift (Robinson 2002; Robinson, McNeill, & Maruna 2013). Probation sits precisely on the fulcrum point between rehabilitative and punitive logics, embracing both a social work and law enforcement perspective. Feeley and Simon (1992) propose that probation and parole persisted by adapting to the new penology, reconfigured as low-cost management solution for growing populations. Rather than rehabilitate their charges, probation and parole officers became compliance monitors, incarcerating those who posed too great a risk to the community (Simon 1993). Consistent with this narrative, corrections departments in both the US and UK increasingly adopted many of the “tough” innovations associated with the new penology, including electronic monitoring, frequent drug testing, partnerships with the police, and “intensive” supervision programs, making compliance more difficult and increasing revocation rates (Robinson, McNeill, & Maruna 2012). Together with a massively expanding population, this led to probation increasingly serving as a “net-widener” that expanded the total number under supervision and increased prison growth (Aebi, Delgrande, & Marguet 2015; Klingele 2013; Phelps 2013).

Yet, as described above, a critical literature argues that the transition to the new penology was more pronounced in rhetoric than reality. McCorkle and Crank (1996) argue that departments adopted “organizational ceremonies … to demonstrate to a political constituency that they could relieve prison crowding, control crime, and provide adequate punishment to offenders,” (3) while changing little about the daily routines of probation supervision. Ethnographic accounts across countries show that parole and probation officers in the 1990s through the 2010s resisted the push to become “waste managers,” and instead continued to articulate rehabilitative logics (or a social work orientation) and selectively implemented actuarial tools to reinforce their own judgement (in the U.S., see Bayens, Manske, Smykla 1998; Lemert 1993; Lynch 2000; Rudes 2012; Werth 2016; for European countries, see Bullock 2011; Hannah-Moffat, Maurutto, & Turnbull 2009; Robinson 2002). However, in the U.S., massively expanding caseloads throughout this period meant that resources were stretched thin, particularly for treatment-oriented services, and revocation rates increased precipitously (Caplow & Simon 1999; Lynch 2000). More recent work suggests that some parole officers have continued to resist new management logics, in this case subverting reforms designed to improve reentry outcomes (Rengifo, Stemen, & Amidon 2017).

Case Selection, Data, and Methodology¹

U.S. states vary dramatically in terms of political cultures, demographics, crime rates, and the scale and form of punishment. Thus, any historical analysis of probation in the U.S. context beyond a quantitative gloss must focus on individual states. I use the case of Michigan, a relatively large Midwestern state that has been on the forefront of both the penal build-up in the 1970s through 1990s and in downsizing in the 2000s and 2010. By 2000, Michigan’s adult prison population was one of the 10 largest in absolute size, with the 12^th highest incarceration rate (Beck & Harrison 2001); probation rates similarly in the state were among the 10^th highest among U.S. states in 2000 (Glaze 2001). Yet Michigan is also praised for leading the U.S. prison downsizing trend (Greene & Mauer 2010). Compared to Western European counterparts, both the rate and absolute size of the prison populations is astonishing even after recent reductions; in 2015, Michigan’s prison population (roughly 43,000 in state adult prisons) was roughly half of the imprisoned population in all of England and Wales and more than five times as large as the prison population of Scotland (Walmsley 2016). Figure 1 summarizes the growth in prison, felony probation, and parole populations in Michigan throughout these years.² Figure 1 only includes probationers under supervision for felony-level offenses; in Michigan, probation supervision for misdemeanants is administered by local agencies. The case study therefore focuses on felony probation.

Figure 1.

Probation, Parole and Prison Totals in Michigan, 1977–2014

The data are all printed Michigan Department of Corrections (hereafter MDOC) materials produced during the study period, including annual reports, statistical reports, newsletters, and press releases. These materials were produced for both public and internal consumption; for example, annual reports were intended primarily to summarize the work of the department to the state legislature, while internal newsletters were addressed primarily to employees. Nevertheless, the department’s tone was surprisingly consistent across document types, as shown in the analyses. Except in special cases (e.g. letters from the governor or director), the specific author or writer of each piece is not identified. The analyses begin in the mid-1970s because this is the earliest publication date for many of the relevant archival materials (most notably the MDOC’s annual report) and loosely corresponds to the start of the penal build-up. I follow this trajectory through to contemporary reforms (ending in 2014).

Documents were culled from three archives (University of Michigan Law School, State of Michigan Archives, and an internal library at the MDOC offices). According to the extant records, my archives include all public documents produced by the department during the study period. I first combed and organized the archival material, scanning copies of all documents (except routine press releases, e.g. parole board hearing notices). Working with research assistants, I then created a spreadsheet to summarize this digital library of reports, newsletters, and press releases, creating a unique record for every story. This process generated over 550 articles that related to probation in some way (see Appendix Table A).

I used qualitative content analysis to trace change and continuity in these institutional discourses. Reviewing the materials, it was quickly apparent that the themes of risk and public safety, cost-efficiency, and probation as a “tough” sanction were ubiquitous. I re-read the entire corpus of materials with these themes in mind, analyzing how each discourse was deployed over time. Rather than a genealogy (or “history of the present”), I apply an archeological approach, tracing the discourses and practices sequentially in each period (Garland 2014). The analyses also trace macro-level trends in probation policies, as gleaned from official reports and statistics, which provide a birds-eye perspective on outcomes (e.g. sentencing trends, staffing levels, new technologies of control, and revocation rates) independent of official MDOC rhetoric.

Although much of the argument that follows focuses on the continuity in probation in Michigan over time, I break the analyses into three loose time periods for organizational purposes. I adopt the periodization scheme presented in Campbell and Schoenfeld (2013), beginning in the “contestation” period (roughly the mid-1970s to the end of the 1980s). During this period, “high crime politics” developed and states rapidly expanded their prison rolls, generating a series of political and bureaucratic struggles. Their next period, “reconstruction,” which runs from the early 1990s up to the start of the 2000s, was dominated by a solidification of this new penal order as states continued to toughen criminal justice policy. To characterize the period of reform that followed (loosely 2000–2014), I adopt Seeds’ (2016) “late mass incarceration” terminology. Table 1 summarizes the results for each period.

Table 1.

Penal Goals, Discourses, and Practices in Three Periods

	Contestation (~1974–1990)	Reconstruction (~1991–2001)	Late Mass Incarceration (~2002–2014)
Primary Penal Goal	Establish justification for corrections after rehabilitation’s collapse	Incapacitate dangerous subjects to ensure public safety	Downsize prisons while guarding public safety
Deployment of Risk	Testing Risk: Probation as test of risk (i.e. dangerousness) in the community as alternative to prison building	Reserving Beds: Probation as diversion for low-risk to reserve prison beds for incapacitating “truly dangerous”	Supporting Success: Probation as diversion for low-risk; Risk used to target services and reduce revocations
Probation Practice Innovations	Probationer compliance programs; Development of risk assessment tools	Boot camp, electronic monitoring, supervision fees, and collaborations with police; Local development of progressive services	Tailored supervision plans and services based on risk-need assessments; Graduated sanctions for violations

Mass Probation: Risk, Costs, and “Tough Enough” Supervision

Contestation (~1974–1990)

As in states across the country, the 1960s and 1970s in Michigan were a period of tumult or “destabilization” (Cambell & Schoenfeld 2013). In the wake of city riots, prison protests, and the start of the decline in American manufacturing, crime rates doubled as white residents fled to the suburbs in Michigan cities, creating a series of governmental crises. Once the manufacturing center of the country, Detroit was the epicenter of this racialized post-Fordist collapse that would end with the city declaring bankruptcy in 2013 (Sugrue 2014). This local context, coupled with the national conversation around the futility of “treating” criminality, all fed an emergent “law and order” ethos as legislators quickly began to “crack down” on crime, especially for drug and violent offenses and those with prior convictions. As early as the mid-1970s, Governor William G. Milliken was calling for the criminal justice system to “get tough,” particularly for violent and repeat offenses and crimes involving guns and narcotics.³ By 1978, legislators began to curtail good time credits (that allowed for early prison release) and enacted a draconian drug law modeled after New York’s Rockefeller laws.⁴ As these policies piled up, the state rapidly built new prisons (growing from 9 in 1980 to 28 by 1989),⁵ and yet could not keep pace with demand, prompting a series of crowding crises, riots, and judicial interventions.

Throughout these struggles the state was deeply fractured, with the MDOC aggressively fighting the emergent tough-on-crime ethos of the legislature, governor, and public (Rubin & Phelps, 2017). The opening statement from Director Perry Johnson in the 1974 Annual Report highlights the tactics the MDOC would mobilize throughout the contestation period. Johnson, a Canadian immigrant and former prison counselor who rose through the ranks to serve as a warden and then MDOC Director, was a staunch defender of moderate and progressive punishment. The statement begins:

During the past few years, the Michigan Department of Corrections has become more acutely aware of public dissatisfaction with the criminal justice system’s ability to make substantial inroads against crime… the public feels now it has been deceived; that a commitment to prison should have meant an individual would never again commit a crime because he would be rehabilitated. Instead, much of the citizenry now believes that parolees and ex-prisoners are a major part of the problem.⁶

He continues that this “feeling of betrayal” has led to two reactions: a desire to either abolish prisons or expand prisons for incapacitation. Relying on recidivism statistics that show “one parolee in 100 will commit a very serious crime involving death or injury to the victim,” the Director argues that neither the abolitionist agenda (which would offer “no protection”) nor the emergent “law and order” ethos (which would “lock up all 100 persons longer”) is a just, efficient, or effective response to the problem of crime. As an alternative vision, Johnson articulates the department’s “single and overriding justification for being” as “protection of the public” via risk management: “the greatest failure of the corrections system is not its inability to rehabilitate or to deter, but its failure to realize its full potential in testing and screening out the possibly dangerous.” Probation (and, to a lesser extent, parole) was framed as the “best” (i.e. most cost-effective) “testing and sorting” mechanism, allowing the department to cheaply supervise people in the community (and revoke to prison anyone unable to comply).

To mold their operations to match this vision, the department pursued three central reforms. First, the MDOC strove to increase the percent of felony cases sentenced to probation instead of prison—a priority that was documented as early as their 1972 “10-Point Plan.”⁷ Throughout the 1970s, Director Johnson fought against punitive sentencing changes, arguing that prison build-ups were a result of an “emotional response” by judges⁸ and “hasty or piecemeal changes in the state’s penal code” from legislators.⁹ The department argued that community sanctions were the only way to “really do the job of protecting the public without bankrupting it.”¹⁰ In 1988, the MDOC won a decisive victory in these battles with the passage of the Community Corrections Act (MI 511), which provided economic incentives to counties to divert people convicted of non-violent felonies to probation. Following the Act’s passage, the percent of felons sentenced to prison in the state began to decline, falling from 37 to 33 percent from 1988 to 1992.¹¹

Second, MODC leaders worked to present probation as a “tough” sanction—that is, a legitimate alternative to imprisonment. The 1976 Annual Report, for example, includes: “It … is a mistake to think of probation as nothing but a form of leniency, with prison as the only real punishment,” since probation can be “a significant sanction in its own right” and is “more difficult for some offenders than short terms of incarceration.”¹² MDOC administrators sought to increase resources for field services so that officers could spend more time supervising probationers and parolees. At the start of the 1970s, the department admitted that staff spent most of their time writing Pre-Sentence Investigation Reports (which provide information to the courts) rather than supervision.¹³ Throughout the late 1970s, the department pursued a series of federal grants and appropriations from the legislature that allowed them to hire additional field staff. ¹⁴ Through these various initiatives, the MDOC reached a standard caseload size of 75 probationers and parolees per officer by 1979.¹⁵ Through the next decade, the department struggled to keep pace with the increase in the numbers of probationers; by 1986, the average caseload would climb up to 95 cases per officer due to increasing probation and parole populations.¹⁶

Third, the MDOC began innovating new probation features and programs that fit the new penal regime. In the mid- to late-1970s, the department developed pilot programs that established performance contracts for probationers, provided fiscal incentives to counties that reduced the rate of revocations, and built community corrections centers and resident homes to provide an intermediate punishment for probation violators (an precursor to contemporary day reporting centers and half-way houses).¹⁷ Together, these programs gave the department concrete examples to show how they were improving probation, producing “greater public safety, improved rehabilitative opportunity and dollar savings.”¹⁸ Indeed, such examples comprised much of the text related to probation in the department’s annual reports and internal newsletters.

The MDOC also worked to develop the kinds of risk assessment tools that would allow them to predict dangerousness—i.e. identify Johnson’s “one in 100” cases that required incapacitation. By 1979, the department would boast that they had “the most sophisticated and accurate risk prediction system in the country … to better protect the citizens of this state and to make the best use of our limited prison bed space.” ¹⁹ In 1982, the MDOC received federal funds from the National Institute of Corrections to implement a new assessment system for classifying probationers and parolees to supervision levels (an innovation that would be rediscovered in the 2000s reforms).²⁰ Despite planning a two-year implementation phase, by 1986, the MDOC was still seeking a reliable risk assessment model, contracting out to a researcher to develop a tool that would “predict a probationer’s behavior” and guide decisions about supervision levels.²¹

By the mid-1980s, the shift toward the new penology model was already evident in other respects as well, as the department began to experiment with the rhetoric and practices that would become dominant in the 1990s. In 1984, a series of prison scandals, overcrowding crises, and the death of a police officer at the hands of a parolee eventually pushed Johnson to resign the directorship. He was replaced by Robert Brown Jr, the MDOC’s first African-American director. Brown had a less progressive and more managerial approach, using his “Vantage Point” columns in the annual reports to highlight topics like employee management²² and efficient drug testing protocols.²³ Gone was the rallying against state officials; in its place was an increasingly distanced (and periodically sneering) tone toward “offenders.”²⁴ Community supervision policies increasingly moved toward a “law enforcement” (as opposed to social services) model, with field agents approved in 1986 to carry handguns.²⁵ Most poignantly, this “tougher” perspective can be seen in the increasing number of probation violators sent to prison, which increased from a total of 859 in 1970 to 1,510 by 1977, then declined for a few years, and increased again more rapidly in the late 1980s (reaching 2,374 by 1990).²⁶ This translated into an increase in the revocation rate from 49 to 55 per 1,000 probationers between 1980 and 1990.²⁷

Reconstruction (~1991–2001)

By 1992, Michigan would report that it had the highest rate of felony supervision in the Great Lakes states and was among the five highest jurisdictions nationally.²⁸ As in other states, this growth continued into the reconstruction period. Under the leadership of Republican Governor John Engler (who served from 1991 to 2003), Michigan ratcheted up penalties for many crimes and adopted tough mandatory minimum sentencing and truth-in-sentencing provisions (Clark 1995; Smith 2016). Despite a decline in felony cases sentenced to prison,²⁹ this legislation, together with declining parole approval rates,³⁰ exacerbated overcrowding even in the face of continued prison-building (the total number of Michigan prisons rose to 42 by 2000).³¹

In this period, the MDOC’s public rhetoric came to closely resemble the new penology (or post-modern, neo-liberal) model, adopting discourses of managing bundles of risk, framing the public as the “customers” of the MDOC, and stressing supervisees’ “personal responsibility.”³² Language of rehabilitation or the treatment potential of supervision was mostly suppressed, with the MDOC officials focusing instead on the retributive and incapacitative functions of punishment. With probation effectively branded as a cost-efficient “alternative” to prison, it was now promoted to reserve prison beds for the long-term incapacitation of “dangerous” felons. In addition, with the development of new risk assessment tools in the contestation period, the MDOC leaders now interpreted probation not as a “test,” but rather supervision for “low-risk offenders.” The “dangerous” or “high-risk” were a familiar group—violent and chronic offenders.

Rather than fighting against other state officials, the MDOC’s rhetoric in this period was now aligned with the political ethos. The 1993 MDOC annual report begins with a letter from Governor Engler, where he states “my most critical responsibility” is “public protection—the safety of each and every family.” To accomplish this goal, the state needed to “punish certain types of low-level felons on the local level, rather than in prison” in order to “have more prison space for Michigan’s violent and chronic offenders.”³³ In the following year, MDOC Director Ken McGinnis would support the Governor’s sentencing reform package, arguing that “Lower level, nonviolent offenders must be punished, but the use of scarce and expensive prison beds … is not the answer” and concluding, “with wise use of existing prison resources, I am confident the initiative can accomplish its primary goal—improvement of public safety—by insuring appropriate punishment for violent criminals.”³⁴ In 1998, Bill Martin (a former state police trooper turned politician) took over the MDOC directorship, continuing the party line that the MDOC “best served” the citizens of Michigan when “it maximizes public protection and minimizes public cost,” reserving prison beds for “violent and chronic offenders” and supporting “alternatives to prison.”³⁵

As in the prior period, the MDOC pursued three tracks to enact and confirm this vision: encouraging courts to sentence felons to probation, increasing the field services budget to hire additional officers, and developing new probation technologies. The first track built off the success of the 1988 Community Corrections Act, which incentivized counties to sentence fewer felons to prison. In addition, as noted below, the department worked to increase field services staff, explicitly arguing that such enhancements would build the legitimacy of probation with the judiciary.³⁶ Throughout this period, the prison commitment rate for felonies declined from 33 percent in 1992 to 26 in 2001,³⁷ an achievement the MDOC frequently took credit for and celebrated. In 1997, for example, the MDOC would announce that they had “retired” the “monster who was swallowing Michigan state government” by controlling prison growth and costs.³⁸

In another line of continuity, the MDOC materials continued to emphasize the “tough” nature of probation, now adopting the frame of personal responsibility. In a 1992 report, for example, the MDOC listed as one of its goals: “Improve the accountability of offenders by providing incentives and opportunities for positive changes in behavior and by providing legitimate sanctions for unacceptable behavior. This will mean setting specific performance objectives for each offender.”³⁹ They also moved to a “zero tolerance” policy for violations, writing in a 1996 newsletter: “Noncompliance with DOC rules is unacceptable, and offenders who don’t abide by the rules are being punished for their behavior.” ⁴⁰

To make this rhetoric a reality, the MDOC again focused on decreasing caseload sizes, which had grown as the number of field agents increased more slowly than the parole and probation populations.⁴¹ Between 1991 and 1999, the department increased the number of field officers by 62 percent, reducing the average caseload⁴² but still falling short of the department’s goal of reaching 90 cases per officer.⁴³

In addition, the reconstruction period saw a number of new probation technologies and programs to support the department’s goal of enhancing “public protection by more ‘in your face’ monitoring.”⁴⁴ Banner achievements in this period included the development of electronic monitoring programs,⁴⁵ a centralized computer system (the Offender Management Network Information or OMNI) to track cases,⁴⁶ and monthly supervision fees for probationers and parolees.⁴⁷ The 1990s ushered in “zero tolerance” drug testing, special sex offender programs, and the “Night Hawk Program,” a collaboration with law enforcement that allowed field officers to make unscheduled visits to supervisees’ homes at night^.48 The department also expanded day reporting and technical rule violator centers (to supervise and sanction probationers and parolees)⁴⁹ and the boot camp program (“Special Alternative Incarceration” or SAI), which housed both probationers and lower-level prisoners and traded intense military-style supervision for shorter sentences.⁵⁰

These innovations corresponded with a rise in revocation rates. The number of prison admissions from probation violations increased from just under 2,000 per year in 1992 to a peak of 4,230 in 2002. These trends reflected a fluctuating revocation rate, which declined in the early 1990s, but then ratcheted up to a peak of 69 prison admissions per 1,000 probationers in 2001.⁵¹ The MDOC proudly took credit for these trends, explaining that this expansion was a consequence of sending fewer felons to prison initially (thus increasing the average probationer risk profile) and providing more intensive supervision. A 1997 newsletter declared: “Department officials don’t view this as a ‘good news, bad news’ scenario, but rather as a ‘win, win’ situation in which the public is better protected while the growth of the state prison system has been slowed.”⁵²

However, somewhat paradoxically, during this same decade the Community Corrections Act provided funding to counties to improve community supervision, in part by expanding treatment-oriented approaches. The department awarded grants to counties that established Community Corrections Advisory Boards to improve local supervision programs for felony probationers. Throughout the 2000s (and on into the 2010s), the Office of Community Corrections appropriated roughly $13 million annually for counties’ “comprehensive plans and services.”⁵³ These funds went to support diverse efforts, many progressive in orientation, including group counseling and mental health support, assessment and case management, substance abuse treatment, and employment programs. Notably, these more treatment-oriented local approaches were not highlighted in the state’s annual reports and newsletters during this period.

Late Mass Incarceration (~2002–2014)

By the early 2000s, Michigan was facing the start of a series of budget crises (which would culminate in the great recession) and corrections spending and populations had ballooned. In addition, although Detroit continues to lead the country in murders, violent crime has been declining since the 1990s and by the mid-2010s stood lower than before the prison boom (Friedman, Fortier, & Cullen 2015). With Democrat Jennifer Granholm elected Governor in 2003, policy-makers coalesced around the conclusion that the state’s current level of incarceration was unsustainable and unnecessary. The state began an ambitious reform campaign, changing sentencing laws to eliminate mandatory minimums for drug violations and moderating sentencing guidelines and good time laws.⁵⁴ These efforts continued (with some retrenchment) when Granholm was succeeded by Republican Rick Snyder in 2011.

The department’s reform rhetoric in this period of late mass incarceration disavowed the earlier tough on crime policies, returning to a (revised) vision of rehabilitation. This new emphasis was branded under the slogan of “offender success” and the concept of “reentry,” with services justified through the language of public safety and cost-effectiveness. By 2009, Director Patricia Caruso would confidently announce: “We continue to have the support of Governor Granholm, legislators, community partners and interest groups who care about offender success and who realize that human warehousing is a poor strategy for successful offender re-entry policy.⁵⁵ In 2014, Governor Snyder’s pick for MDOC Director, Daniel Heyns, would announce the new motto: “95% of the prisoners under our jurisdiction will return to our communities as fellow citizens … We ‘Help Make Things Right’ by embracing our Department’s multifaceted purpose to create a safer Michigan while also working to rehabilitate the offenders who are under our jurisdiction, increasing their chances of success in our communities so that they will not recidivate and revictimize.”⁵⁶

Probation remained central to this effort as a “cost-effective” alternative to prison for “low-risk offenders.” In a 2007 newsletter reporting on the department’s work with the Council of State Governments on devising a reform plan, administrators again emphasized that community corrections were a critical piece of “safely continuing the down-sizing corrections” so that the MDOC could continue to “focus on saving expensive prison beds for our most serious and violent offenders.”⁵⁷ Yet in an important shift, the high probation and parole revocation rates defended as a “win-win” outcome in the 1990s instead became framed as a sign of department failure rather than simply individuals’ risk. Instead of conceptualizing risk as a static trait, the MDOC came to frame risk as malleable and open to intervention, using risk discourses as a way to justify parceling out limited resources.

As in the two earlier periods, the MDOC officials worked to enact this vision of probation by continuing to nudge the judiciary. For example, one of the first bullet-points in the 2003 “Five-Year Plan to Control Prison Growth” was to expand probation for low-level felons.⁵⁸ Continuing to use the Community Corrections Act to incentivize diversion, Michigan further decreased the percent of felons sentenced to prison (from 26 percent in 2001 to 22 in 2014).⁵⁹ Even bigger declines were seen in the percent of “straddle cells” in the sentencing grid (or intermediate severity cases) sentenced to prison (Greene & Mauer 2010).

Second, the department continued their focus on framing probation as a legitimate (i.e. “tough”) sanction. When announcing a special week to celebrate field agents in 2007, department leaders described field agents as “unsung heroes” who hold “offenders accountable for the wrong they have committed to our families, friends and communities” and make “our communities a safer place to live” by supervising “the nearly five million adults under community supervision in areas that most people would never enter.”⁶⁰ Throughout the budget contractions, the MDOC continued to invest heavily in community corrections,⁶¹ hiring more than 150 new field officers in 2009 and 2010 alone.⁶² However, community corrections remained an overall small share of the total MDOC budget, with Field Operations and the Office of Community Corrections marshalling just 12% of the department’s expenditures (or $231 million) in 2010.⁶³

Third, the MDOC overhauled community supervision protocols to match the new reform rhetoric. Even as early as 2001, department officials would report that “Increased emphasis is being placed throughout the state on treatment and cognitive behavioral-based substance abuse, education and employment programming which large-scale research studies have shown reduce recidivism among higher-risk cases.”⁶⁴ By 2014, a local field office supervisor would report in the newsletter that “a probation officer was once seen as someone whose job was to track, catch and send offenders back to jail. Today … their job is seen as helping individuals make the transition, as much as possible, from being involved in the criminal justice system to being a productive, well-adjusted member of society.”⁶⁵ The most widely touted (and expensive) of these efforts was the Michigan Prisoner Reentry Initiative (MPRI), which began as a pilot in the early 2000s and expanded statewide in 2009, reaching an increased total of $57 million in the 2010 Governor’s budget even as other corrections categories were cut.⁶⁶ This program implemented a range of reforms, including expanding parole release, building links between parole agents and local service providers, developing individualized re-entry plans, and adopting intermediate sanctions for responding to parole violations (see also Greene & Mauer 2010). After Director Heyns’ appointment in 2011, some of the reforms for increasing parole releases were scaled back, ⁶⁷ but the work of MPRI (now branded as the “Michigan Offender Success Model”) remains central to the MDOC vision.⁶⁸

The MDOC also implemented many of the same kinds of changes in probation, although they were buried under the tremendous press around parole. These efforts, developed with technical assistance from the Council of State Governments,⁶⁹ largely followed the blueprint of MPRI, including new risk assessment tools, case management plans, and intermediate sanctions.⁷⁰ The MDOC also implemented kiosk technology that allowed probationers deemed low-risk to report electronically and allowed agents to focus their attention on smaller caseloads.⁷¹ In 2012, the legislature enacted Senate Bill 1141, which created the Probation Swift and Sure Sanction Act, a pilot program intended to model the HOPE (Hawaii Opportunity Probation with Enforcement) program.⁷² Collectively, these reforms corresponded with a noticeable drop in revocations. Between 2001 and 2014, the probation revocation rate declined from 69 to 52, driving down prison admission totals.⁷³

At the same time as the department celebrated their role in facilitating “offender success,” their public materials continued to highlight the aggressive policing of public safety risks, continuing many of the technologies of control that expanded in the 1990s. The most publicized of these efforts was the “crack down” on absconders—or “apprehending violent fugitives” ⁷⁴ —in large part through the “NightHawk” collaboration with local police. A 2000 newsletter report on the program argued: “It’s hard to forget you’re on probation when your agent and the cops show up at your house at 8 p.m some evening … It’s hard to forget when you know you could look up from a beer at the neighborhood tavern and see them all bearing down on you.”⁷⁵ The MDOC also continued many of the other innovations of the reconstruction period, including electronic monitoring and intensive supervision for sex offenders.

In some cases, the punitive innovations of the 1990s were revamped with the new orientation. For example, the department’s SAI program shifted from a boot-camp model to a risk-needs and cognitive skills training approach vetted by criminologist James Austin.⁷⁶ Thus, just as the development of the new penology rhetoric and practices was accompanied by a seemingly paradoxical increase in funding for treatment-oriented services, so too has the return of rehabilitation at MDOC been accompanied by the continued embracement of tough discourses and a “crack-down” on absconders.

Conclusion

The case study demonstrates that discourses of risk were central to justifying the persistence of probation (and, more broadly, the corrections department writ-large) throughout the past half-century in Michigan. These discourses were surprisingly consistent: risky subjects are those with multiple convictions and histories of violence; it is the responsibility of the state to manage risk; and probation is a cost-effective alternative to prisons. Yet the deployment of risk discourses varied across the periods, with the MDOC responding to different governing crises and positing a different role for probation vis-à-vis the management of risk, as summarized in Table 1. Thus, while the basic structure of discourses was surprisingly consistent across time, risk language was adapted in each period to define and respond to a different set of bureaucratic concerns.

Throughout these incarnations, much of probation supervision in practice remained the same, consistent with literature on the “gap” between institutional discourses and practices (McNeill et al. 2009). Probation and parole officers continued to manage very large caseloads, as the state and local counties experimented with (and often recycled old versions of) various treatment- and punitive-oriented programs like day-reporting centers, boot camps, probationer contracts or individualized case management plans, and substance abuse treatment. Yet these programs only made it to the front pages of official materials when they were consistent with the dominant narrative in punishment. The department heavily publicized (often small-scale) pilot programs and innovations that fit the new rhetoric. As McCorkle and Crank (1996) argue, the development of these kinds of costly programs are “presented as representative of the type of control community corrections could provide over offenders” to “convince its institutional audience of the credibility of its crime-control potential” and promulgate the “myth of control” (16-7).

Yet these programs were not simply “organizational ceremonies” devoid of meaningful consequences. Perhaps most critically, the continued insistence that probation was a better alternative to prison for many cases lined up with persistent declines in the percent of felons in Michigan sentenced to prison. And the shift to the new penology managerial logics in the 1990s, with its “in your face” style of aggressive supervision, corresponded with a spike in probation revocation rates. These trends drove much of the prison churn in the early 2000s that motivated reform in the late 2000s and early 2010s. Put simply, official discourses identify the dominant governance strategy at the time, which is refracted in the developments the state chooses to highlight in official materials and (to varying degrees) changing practices on the ground. Rather than a tight (or non-existent) coupling between discourses and practices, dominant institutional narratives relate haphazardly to on-the-ground changes and macro-level supervision outcomes. In addition, as other scholars have demonstrated, much of this gap is explained by the continued power of front-line staff and managers to interpret and enact these new mandates (Cheliotis 2006; McNeill et al. 2009; Rengifo, Stemen, & Amidon 2017; Robinson 2002; Rudes 2012). Thus, we must study institutional discourses and practices in tandem, rather than seperately.

What does this history suggest about the move from penal modernism to post-modernism and the utility (or futility) of risk narratives in corralling the penal state? First, joining a large literature, the results suggest that this shift was more partial and incomplete than is often suggested—and that many of the post-modern “innovations” discovered in the 1980s and 1990s were in fact rediscoveries of old ideas under new guise. Most surprising from the literature on the new penology is that the language of risk emerged not as a hand-maiden to mass incarceration, but as a strategy of resistance to the tough on crime political momentum. Ultimately, that discourse proved an ineffective sparring companion: rather than prevent mass incarceration, the logic of probation as alternative produced both mass incarceration and mass probation.

This failed history of resistance suggests that scholars and are right to be critical of reform discourses that pit the people we are “mad at” verses the “truly dangerous” (Gottschalk 2015). Yet it is overly simplistic and inaccurate to suggest that the logic of risk will inevitably lead to negative outcomes (Prins & Reich 2017). As O’Malley (2008) and others have argued, risk could be “consistent with programs of social justice” (453) and, in particular, challenging the harms of the penal state. Indeed, the risk narratives of the 2010s about the “revolving door” of corrections (and its high fiscal costs) have been crucial to increasing reentry services, reducing supervision burdens for low-risk cases, and driving down revocation rates. These reforms may be reaching their limits, but they have propelled meaningful reductions in punitive excess.

Certainly, a more robust critique around justice, equality, and penal moderation will need to expand to continue these declines to bring the U.S. in line with the punishment rates of Western European countries. Part of this movement has to be a reorientation around the concept of dangerousness, an animating focus of penal practices across time and place (Bottoms 1977; Seeds 2015; Pratt 1997). As noted above, while the “high-risk” designation is deployed politically to refer to those individuals who have committed serious, repeat, violent offenses, in practice risk assessment tools often measure the predicted probability of recidivism for all crimes. The risk of recidivism is distinct from fears of violent harm; for example, those convicted of homicide and sex offenders—two of the most politically toxic categories of prisoners—show the lowest re-incarceration rates (Langan & Levin 2002). This, while the distinction between low/high risk is often presented as self-evident, its meaning in practice is context-dependent. As local jurisdictions, states, and countries continue to experiment with criminal justice reform, scholars can better map out the potentials and pitfalls of risk logics (and their competitors) for challenging the penal state. In addition, future researchers might map institutional discourses and practices surrounding risk and probation in other states and countries to compare and contrast with the Michigan case and contribute to the budding socio-historical understanding of mass probation.

Appendix A. Archival Materials: Source Titles, Years, and Number by Period

Annual Reports	Statistical Reports	Press Releases	Newsletters
Contestation (~1974–1990)
Dimensions (1974–1978) Annual Report (1985–1989)	Criminal Statistics (1974–1974) Statistical Presentation (1978–1984)	1974–1990	Frontline (1972–1976) Dialogues (1975–1984) Celebrate (1979–1986) F.Y.I (1980–1990) Signal (1988–1990)
Reconstruction (~1991–2001)
Annual Report (1993, 1997–2001)	Statistical Report (1991–2001)	1991–2001	Signal (1991–1999) F.Y.I. (1992–2001)
Late Mass Incarceration (~2002–2014)
Annual Report (2002–2011)	Statistical Report (2001–2014)	2002–2014	F.Y.I. (2002–2014)