Legal Studies Research Paper Series Paper No. 1368

The Promises and Perils of EvidenceBased Corrections
Cecelia Klingele
This paper can be downloaded without charge from the
Social Science Research Network Electronic Paper Collection at:

http://ssrn.com/abstract=2697240

ND SSRN.DOCX (DO NOT DELETE)

11/30/2015 11:20 AM

THE PROMISES AND PERILS OF EVIDENCE-BASED
CORRECTIONS
Cecelia Klingele*
Public beliefs about the best way to respond to crime change over time,
and have been doing so at a rapid pace in recent years. After more than
forty years of ever more severe penal policies, the punitive sentiment that
fueled the growth of mass incarceration in the United States appears to be
softening. Across the country, prison growth has slowed and, in some places,
has even reversed. Many new laws and policies have enabled this change.
The most prominent of these implement or reflect what have been called “evidence-based practices” designed to reduce prison populations and their associated fiscal and human costs. These practices—which broadly include
the use of actuarial risk assessment tools, the development of deterrencebased sanctioning programs, and the adoption of new supervision techniques—are based on criminological research about “what works” to reduce convicted individuals’ odds of committing future crimes.
Because evidence-based practices focus on reducing crime and recidivism, they are usually promoted as progressive tools for making the criminal
justice system more humane. And while many have the potential to do just
that, evidence-based practices are not inherently benign with respect to their
effect on mass incarceration and the breadth of the penal state. In their reliance on aggregate data and classification, many such practices have as
much in common with the “new penology” that enabled mass incarceration
as with the neorehabilitationism they are ordinarily thought to represent.
Without denying the contribution that such practices are making to current reform efforts, this Article seeks to highlight the unintended ways in
which evidence-based tools could be used to expand, rather than reduce,
state correctional control over justice-involved individuals. It explains what
evidence-based practices are, why they have gained traction, and how they

* Assistant Professor of Law, University of Wisconsin Law School.

Many thanks to Richard
Garnett, Meghan Ryan, Francis Cullen, Keith Findley, Howard Erlanger, Carissa Hessick,
Randy Kozel, Mark McKenna, Kevin Reitz, Ed Rhine, Ron Corbett, Nicole Garnett, Jordan
Woods, Daniel McConkie, Jason Krieg, Roger Alford, and Stephen Smith, for their comments
on various iterations of this paper. Thanks also to University of Wisconsin Law School students Zachary Carlson, Tip McGuire, and Matthew Hefti for their able editorial assistance.
101

ND SSRN.DOCX (DO NOT DELETE)

11/30/2015 11:20 AM

NOTRE DAME LAW REVIEW

102

[VOL. XX:N

fit into existing paradigms for understanding the role of the criminal justice
system in the lives of those subject to its control. Finally, it calls on policymakers and practitioners to implement these practices in ways that ensure
they are used to improve the quality and fairness of the criminal justice system and not to reinforce the institutional constructs that have sustained the
growth of the penal state.
INTRODUCTION
The criminal justice system has long been in the business of trying to
prevent crime by controlling the behavior of known past offenders. Methods
of control have varied over time, from execution to banishment to forms of
“rehabilitation” ranging from mentoring and job counseling to forced psychosurgery. Always, system actors have justified their methods by reference
to a mix of values and science, which change over time.
In recent years, the American conversation about punishment is again
changing, and with it the forty year trend of ever-increasing correctional populations. Every year from 1970 to 2008 saw an increase in the number of
convicted people under the control of the penal state, whether on probation,
in jail, or in prison. 1 Beginning in 2008, however, the United States saw five
consecutive years of reductions in the total number of people confined in
state and local correctional institutions, and in those serving terms of community supervision on probation and parole. 2 While those national statistics
mask significant regional variations, they speak to a notable shift in the way
punishment is being imposed and executed in the United States today.
The recent reduction in the U.S. prison population has been facilitated
by laws and policies designed to stem the growth of custodial populations.
These include the decriminalization of some drug and traffic offenses, 3 re-

LAURA MARUSCHAK, LAUREN GLAZE & THOMAS BONCZAR, BUREAU OF JUSTICE STAADULTS ON PROBATION IN THE UNITED STATES, 1977–2012 (2013),
http://www.bjs.gov/index.cfm?ty=pbdetail&iid=2026.
2 LAUREN E. GLAZE & ERINN J. HERBERMAN, BUREAU OF JUSTICE STATISTICS, CORRECTIONAL POPULATIONS IN THE UNITED STATES, 2012 2 (2013), http://www.bjs.gov/content/pub/pdf/cpus12.pdf. In 2013, the downward trend altered slightly, with a slight increase
in state prison populations and modest decreases in the jail and community supervision populations. LAUREN E. GLAZE & DANIELLE KAEBLE, BUREAU OF JUSTICE STATISTICS, CORRECTIONAL POPULATIONS IN THE UNITED STATES, 2013 2 (2014), http://www.bjs.gov/content/pub/pdf/cpus13.pdf.
3 See, e.g., NAT’L ASSOC. OF STATE BUDGET OFFICERS, STATE SPENDING FOR CORRECTIONS: LONG-TERM TRENDS AND RECENT CRIMINAL JUSTICE POLICY REFORMS 5 (2013).
1

TISTICS,

ND SSRN.DOCX (DO NOT DELETE)

201X]

11/30/2015 11:20 AM

103

peal of mandatory sentencing provisions for many lower level drug offenses, 4 increases in pre- and post-charge diversion programs, 5 and the expansion of early release mechanisms, such as “good time” credit. 6 In addition, recent years have seen a surge in the popularity of new correctional
techniques, loosely classified as “evidence-based practices,” that courts,
community supervision agencies, and correctional institutions are rapidly
adopting in their efforts to deliver more targeted (and less expensive) services to individuals under state control. These practices include the use of
actuarial risk and need assessment instruments, motivational interviewing
and counseling techniques, deterrence-based sanction programs, and incentives to probationers and parolees for successful compliance with court orders. 7
These new policies and practices have been promulgated at every level
of government through both grassroots efforts and organized coalitions of
established nonprofits seeking systemic criminal justice reform. 8 In an effort
to capitalize on the opportunity for reform provided by historically low crime
rates 9 and the 2009 United States financial crisis, 10 proponents of these new
policies aim to solve many problems at once. They want to reduce the number of people behind bars, improve the fairness of sentencing and supervision, decrease the financial cost of punishment, reduce recidivism, and improve public safety. While reform efforts have taken many forms, many of
the most influential recent efforts have been spearheaded by the National
Institute of Corrections and by the Justice Reinvestment Initiative (JRI), a
joint public-private coalition of the U.S. Department of Justice, the Pew
Charitable Trust, the Center for State Governments, and the Vera Institute of
See, e.g., RAM SUBRAMANIAN & RUTH DELANEY, VERA INST. OF JUSTICE, PLAYBOOK
(2014).
5 See, e.g., NANCY G. LAVIGNE ET AL., URBAN INST., JUSTICE REINVESTMENT INITIATIVE STATE ASSESSMENT REPORT 6, 10, 23 (2014); ALISON LAWRENCE, NAT’L CONFERENCE OF
STATE LEGISLATURES, TRENDS IN SENTENCING AND CORRECTIONS: STATE LEGISLATION
(2013).
6 See, e.g., LAVIGNE ET AL., supra note 5, at 2, 20.
7 ROGER K. WARREN, CRIME & JUSTICE INST. & NAT’L INST. OF CORRECTIONS, EVIDENCE-BASED PRACTICE TO REDUCE RECIDIVISM: IMPLICATIONS FOR STATE JUDICIARIES xi–xii,
xiv, 48 (2007), https://s3.amazonaws.com/static.nicic.gov/Library/023358.pdf; PEW CTR. ON
THE STATES, RISK/NEEDS ASSESSMENT 101: SCIENCE REVEALS NEW TOOLS TO MANAGE OFFENDERS 6 (2011) (discussing new legislation in Arkansas, Kentucky, New Hampshire, and
South Carolina mandating the use of risk and needs assessment tools).
8 See infra Part II.A.
9 See FED. BUREAU OF INVESTIGATION, CRIME IN THE UNITED STATES: 2011,
http://www.fbi.gov/about-us/cjis/ucr/crime-in-the-u.s/2011/crime-in-the-u.s.-2011/tables/table-1 (last visited Mar. 5, 2015).
10 See generally NAT’L ASS’N OF STATE BUDGET OFFICERS, STATE SPENDING FOR CORRECTIONS: LONG-TERM TRENDS AND RECENT CRIMINAL JUSTICE POLICY REFORMS (2013),
http://www.nasbo.org/sites/default/files/pdf/State%20Spending%20for%20Corrections.pdf.
4

FOR CHANGE? STATES RECONSIDER MANDATORY SENTENCES

ND SSRN.DOCX (DO NOT DELETE)

104

11/30/2015 11:20 AM

NOTRE DAME LAW REVIEW

[VOL. XX:N

Justice. JRI’s advocacy has reached 34 states,11 and involves millions of
dollars in public and private expenditures. 12
Despite the massive scale of these national efforts to change correctional practices, relatively little attention has been paid by legal scholars to
the substance of the practices being labeled as “evidence based” outside the
context of sentencing, 13 or to their implications for the practical and theoretical functioning of the criminal justice system more broadly. Although scholars and policymakers have reached a broad consensus that mass incarceration has come at too high a price, 14 the legal mechanisms by which overly
punitive policies should be undone is a matter that has been largely undertheorized. Methods matter.

11 LAVIGNE ET AL., supra note 5, at 57–123 (2014) (discussing JRI work in Arkansas,
Delaware, Georgia, Hawaii, Kansas, Kentucky, Louisiana, Missouri, New Hampshire, North
Carolina, Ohio, Oklahoma, Oregon, Pennsylvania, South Carolina, South Dakota, and West
Virginia); JAMES AUSTIN ET AL., ENDING MASS INCARCERATION: CHARTING A NEW JUSTICE
REINVESTMENT 1 (2013) (adding Alabama, Arizona, Connecticut, Indiana, Rhode Island,
Michigan, Nevada, Texas, Vermont, and Wisconsin to the list of states that participated in
precursors to JRI); BUREAU OF JUSTICE ASSISTANCE, JRI Sites, https://www.bja.gov/programs/justicereinvestment/jri_sites.html (last visited Sept. 7, 2015) (mapping 24 currentlyinvolved states, including Idaho, Mississippi, Nebraska, Oklahoma, Oregon, Utah, and Washington).
12 In 2014, Congress allocated $27.5 million to the Justice Reinvestment Initiative. See
Consolidated Appropriations Act, 2014, Pub. L. No. 113-76, 128 Stat. 5, 63 (2014).
13
An emerging body of literature has begun to assess and critique the use of actuarial
risk assessment instruments at sentencing, raising both methodological and equity concerns
about the accuracy and fairness of these predictive tools in the context of criminal sentencing.
See, e.g., Dawinder S. Sidhu, Moneyball Sentencing, 56 B.C. L. REV. 671 (2015); Melissa
Hamilton, Back to the Future: The Influence of Criminal History on Risk Assessments, 20
BERKELEY J. CRIM. L. 75 (2015); Sonja B. Starr, Evidence-Based Sentencing and the Scientific
Rationalization of Discrimination, 66 STAN. L. REV. 803, 821 (2014); J.C. Oleson, Risk in
Sentencing: Constitutionally Suspect Variables and Evidence-Based Sentencing, 64 SMU L.
REV. 1329, 1340-47 (2011). Much less attention has been paid to the use of risk assessments
and other types of “evidence-based practices” in the execution of sentence, whether by probation and parole officers, or by institutional correctional officials.
14 A large and influential body of literature has catalogued the severity of punishment
in the United States and detailed the ways in which current sentencing and correctional policies disproportionately damage the poor and racial minorities, often in ways that endure across
generations. For a small sampling, see generally TODD R. CLEAR, IMPRISONING COMMUNITIES:
HOW MASS INCARCERATION MAKES DISADVANTAGED NEIGHBORHOODS WORSE (2007) [hereinafter IMPRISONING COMMUNITIES]; MARIE GOTTSCHALK, THE PRISON AND THE GALLOWS:
THE POLITICS OF MASS INCARCERATION IN AMERICA (2006) [hereinafter THE PRISON AND THE
GALLOWS]; JEFF MANZA & CHRISTOPHER UGGEN, LOCKED OUT: FELON DISENFRANCHISEMENT
AND AMERICAN DEMOCRACY (2006); MICHAEL TONRY, PUNISHING RACE: A CONTINUING
AMERICAN DILEMMA (2011) [hereinafter PUNISHING RACE]; INVISIBLE PUNISHMENT: THE COLLATERAL CONSEQUENCES OF MASS IMPRISONMENT (Marc Mauer & Meda Chesney-Lind eds.,
2002).

ND SSRN.DOCX (DO NOT DELETE)

201X]

11/30/2015 11:20 AM

105

This Article responds to a gap in current legal literature by examining
the proliferation of “evidence-based practices” in correctional settings—particularly in the context of community corrections—and exploring the ways
in which these practices and the risk management framework they embrace
fit into existing conceptual frameworks for understanding the criminal justice system. Although most proponents of evidence-based correctional practices frame them as rehabilitative tools designed to reduce the use of incarceration and make correctional interventions more modest and humane,
these tools are capable of doing the very opposite. Actuarial risk assessment
instruments, electronic monitoring and other forms of surveillance for high
risk populations, and even cognitive-behavioral interventions designed to increase compliance with conditions of supervision, can all be used to expand
and enforce the scope of state control over the lives of people entangled in
the justice system. Unless such tools are implemented with conscious attention to their limits and with appreciation for their potential for abuse, these
new practices have the potential to thwart long-term efforts to decrease mass
incarceration by inadvertently expanding the scope of state control over the
lives of justice-involved individuals and their communities. This Article is
not intended to derail efforts to bolster criminal justice decision-making (and
decrease bias) through the use of better data, research, or programs. It is,
however, a call for reflection about the limits and potential misuses of popular evidence-based correctional practices. It is also a call for practitioners
and policymakers to monitor the implementation of evidence-based practices
to ensure consistency between the ways they are being used and the purposes
they are intended to advance.
Part I very briefly recounts the escalation of punishment, and several of
the tools that enabled it, emphasizing the contributions of what Feeley and
Simon have dubbed the “New Penology,” which prioritized control of the
underclass through mass surveillance and use of the police power. Part I also
explores recent changes that are now driving states to reconsider their commitment to sustaining high rates of incarceration. Part II examines the growing popularity of evidence-based correctional practices as a way to reduce
overreliance on incarceration as a response to crime. It describes the institutional structures through which evidence-based correctional practices have
been widely promulgated, explores the reasons why they have gained so
much traction among lawmakers and policy advocates, and provides a basic
explanation of a few of the most popular practices being implemented in the
field. Part III places these new practices into a larger conceptual framework.
Without denying that many evidence-based practices arise out of a neorehabilitative tradition that seeks to make criminal justice more humane, it also
observes that many evidence-based correctional practices are embedded with
features of the control-orientated culture they are designed to disrupt. Part
IV contemplates the future of evidence-based practices as a tool for reducing

ND SSRN.DOCX (DO NOT DELETE)

11/30/2015 11:20 AM

NOTRE DAME LAW REVIEW

106

[VOL. XX:N

reliance on incarceration. It concludes that while advocates and policymakers should not reject the potential of these practices to improve the quality
and effectiveness of correctional interventions, they must be equally alert to
their potential for coercion and abuse. Jurisdictions embracing evidencebased practices should therefore consciously monitor such practices to ensure they are being used in ways that reduce the reach of the penal state,
rather than facilitate its growth.
I.

THE PATH TO MODERN CORRECTIONAL REFORM

The story of modern American sentencing and punishment trends has
been told often, and is well known to many. 15 Nonetheless, because changing ideas about punishment are central to understanding the promises and
perils of current reform practices, a brief summary of how we came to the
present moment is instructive.
A. The Rise of the Penal State
Americans weren’t always “tough on crime”—at least not openly.
From the end of the 19th Century through the early 1970s, the prevailing penal philosophy was the progressive Rehabilitative Ideal, in which “[t]he
sanctions of the criminal law were seen as providing opportunities for modifying the behavior of offenders in the interests of both social defense and
the happiness, health, and satisfactions of the individual offender.” 16 In the
rehabilitative paradigm, correctional intervention was a means of healing the
soul-sick—a use of state power that found its justification in bettering the
individual subject to correctional control as a means of restoring him to full
participation in “the law-abiding community.” 17 The instrumental mechanisms by which rehabilitation was achieved were numerous, but all relied
heavily on emerging social and medical science.
Practitioners of the day were confident in their ability to accurately
identify those offenders at risk of re-offense and optimistic about their ability

15 See, e.g., BERT USEEM & ANNE MORRISON PIEHL, PRISON STATE: THE CHALLENGE OF
MASS INCARCERATION (2008); THE PRISON AND THE GALLOWS, supra note 13; MARC MAUER,
RACE TO INCARCERATE (1999); Francis T. Cullen, Rehabilitation: Beyond Nothing Works, 42
CRIME & JUST. 299 (2013); Adam Gopnik, The Caging of America, THE NEW YORKER, Jan.
30, 2012.
16 Francis A. Allen, Central Problems of American Criminal Justice, 75 MICH. L. REV.
813, 821 (1977).
17 This phrase, coined by Professor Kevin Reitz, summarizes not only the traditional
aims of rehabilitation but the continuing goals of sentencing, as echoed throughout the revised
sentencing provisions of Model Penal Code. See, e.g., MODEL PENAL CODE §§ 6.02A, 6.04(7),
6.04(16), 6.0(8)(3)(b)(ii) (AM. LAW INST., Discussion Draft No. 4, 2012).

ND SSRN.DOCX (DO NOT DELETE)

201X]

11/30/2015 11:20 AM

107

to cure some while incapacitating the truly “defective.” 18 Rehabilitative
techniques ranged from those that were overtly benign, such as vocational
training and basic education, to those that decidedly were not, such as psychosurgery, forced sterilization, and physically intrusive “behavior modification” programs. 19 Despite the sometimes dramatic abuses that paraded under the guise of scientific intervention, proponents of the Rehabilitative Ideal
sincerely believed that a cure for criminality was attainable, desirable, and
more humane than a penal system designed merely to punish. 20
This Rehabilitative Ideal dominated penal philosophy and practice
through the first half of the 20th Century. Beginning in the late 1960s, however, a confluence of developments led to its collapse. 21 First, a series of
influential new studies undermined confidence that rehabilitation programs

18 Even those committed to rehabilitation realized that not every individual could be
“corrected.” Hereditary criminologists argued that some people were destined to a life of
crime. “‘[E]very imbecile’” was viewed as “‘a potential criminal, needing only the proper
environment and opportunity for the development and expression of his criminal tendencies. . . . From a biological standpoint . . . [he was] an inferior human being.’” Michael Willrich, The Two Percent Solution: Eugenic Jurisprudence and the Socialization of American
Law, 1900–1930, 16 LAW & HIST. REV. 63, 85 (1998) (quoting JAMES W. TRENT, JR., INVENTING THE FEEBLE MIND: A HISTORY OF MENTAL RETARDATION IN THE UNITED STATES 161
(1995)). For these hopeless cases, the best institutional response was thought to be a lengthy
period of incapacitation, and when possible, sterilization to prevent the spread of criminality
to future generations. Id. See also VICTORIA F. NOURSE, IN RECKLESS HANDS: SKINNER V.
OKLAHOMA AND THE NEAR TRIUMPH OF AMERICAN EUGENICS (2008).
19 See, e.g., James J. Gobert, Psychosurgery, Conditioning, and the Prisoner’s Right to
Refuse “Rehabilitation,” 61 VA. L. REV. 155, 161 (1975) (“A growing number of neurologists
maintain that violent behavior is a product of brain dysfunction, either acquired or genetic.
Since psychotherapy does not treat brain dysfunction, it is unable to alter deviant behavior in
these cases. The answer, the neurologists say, is psychosurgery. . . . By cutting faulty circuiting systems in the brain, psychosurgeons believe they can control disturbed emotional patterns.”) (footnotes omitted); Stanley J. Dirks, Note, Aversion Therapy: Its Limited Potential
for Use in the Correctional Setting, 26 STAN. L. REV. 1327, 1327–29 (1974) (describing aversion therapy—a process in which a prisoner is induced to imagine engaging in deviant behavior and is then given “a nausea-creating drug, an electric shock, a nauseous verbal description
that the patient is instructed to imagine, or a paralyzing drug”—being used to treat “alcoholism, heroin addiction, smoking, homosexuality, exhibitionism, voyeurism, pedophilia, transvestism, overeating, psychotic firesetting, and shoplifting”) (footnotes omitted).
20 See, e.g., Charlton T. Lewis, The Indeterminate Sentence, 9 YALE L.J. 17, 20 (1899)
(“The principle of the reformatory sentence, in its completeness, implies the conversion of the
prison into an institution combining the means and aims of hospital, school and church, for
the healing and culture of body, mind and will. . . . [I]t is to be held in view as the standard
by which our partial and tentative reforms must be measured; and just in the degree that it is
approached will the possible beneficence of the principle be realized.”).
21 For a lengthier discussion of the factors that led to the collapse of the Rehabilitative
Ideal, see FRANCIS A. ALLEN, THE DECLINE OF THE REHABILITATIVE IDEAL: PENAL POLICY AND
SOCIAL PURPOSE (1981).

ND SSRN.DOCX (DO NOT DELETE)

108

11/30/2015 11:20 AM

NOTRE DAME LAW REVIEW

[VOL. XX:N

worked. 22 At the same time, critics began to attack the decisions of paroling
and other correctional officials as arbitrary and illegitimate, subject to no
public oversight and unaccountable to any legislative or judicial authority. 23
Finally—and perhaps most importantly—observers as diverse as Michel
Foucault and the American Friends Service Committee challenged the notion
that the rehabilitative state provided progressive and benevolent assistance
to the downtrodden. Instead, they asserted, rehabilitation had become a
cover for class warfare. By imposing elite values on the underclass, rehabilitative program providers, correctional officials, and parole decisionmakers
forced the poor to conform to privileged white values and behavior, and
sanctioned any deviation from those upper-class norms with imprisonment. 24
In response to these critiques (which occurred alongside concerns about rising crime rates 25), states across the country began to change their sentencing
and correctional practices in several ways.
First, a number of reform efforts attempted to replace rehabilitation with
pure punishment. Many proponents of these changes saw them as more humane than the rehabilitative system of earlier decades. They argued that
punishing people for wrongdoing—rather than trying to change who they
were—would reduce disparities in sentencing and prevent the state from becoming overly-involved in the lives of offenders. 26 In fact, the change in

22 Most famous of these was Robert Martinson’s 1974 report finding that, “[w]ith few
and isolated exceptions, the rehabilitative efforts that have been reported so far have had no
appreciable effect on recidivism.” Robert Martinson, What Works?—Questions and Answers
About Prison Reform, 35 PUB. INT. 22, 25 (1974) (emphasis omitted).
23 See Marvin E. Frankel, Lawlessness in Sentencing, 41 U. CIN. L. REV. 1, 16 (1972)
(“[P]arole boards, subject to no precise criteria and offering no explicit clues as to why particular decisions go as they do, exercise secretly the power to decide within broad ranges the
actual number of years of confinement. . . . Decisions based upon secret reasons bear no credentials of care or legitimacy.”).
24 MICHEL FOUCAULT, DISCIPLINE AND PUNISH: THE BIRTH OF THE PRISON 18 (Alan Sheridan trans., Vintage Books 2d ed. 1995) (1977); AM. FRIENDS SERV. COMM., STRUGGLE FOR
JUSTICE 85 (1971) (“An important force in the reform movement was the mixture of hatred,
fear, and revulsion that white, middle-class, Protestant reformers felt toward lower-class persons. . . . These difficult feelings were disguised as humanitarian concern for the ‘health’ of
threatening subculture members. Imprisonment dressed up as treatment was a particularly
suitable response for reformers’ complicated and inconsistent feelings.”).
25 The U.S. Crime Index rose steadily from 1960 to 1991, with only a few short-lived
exceptions. LEONARD A. MAROWITZ, CRIMINAL JUSTICE STATISTICS CTR., WHY DID THE
CRIME RATE DECREASE THROUGH 1999? (AND WHY MIGHT IT DECREASE OR INCREASE IN 2000
AND BEYOND?), at 3 (2000).
26 Dhammika Dharmapala, Nuno Garoupa & Joanna M. Shepherd, Legislatures,
Judges, and Parole Boards: The Allocation of Discretion Under Determinate Sentencing, 62
FLA. L. REV. 1037, 1044–45 (2010) (“Liberals believed that the reforms, by restricting discretion of judges and parole boards, would reduce sentencing discrimination and sentence-length
disparity. On the other hand, conservatives believed that determinate sentencing reforms

ND SSRN.DOCX (DO NOT DELETE)

201X]

11/30/2015 11:20 AM

109

penal philosophy away from rehabilitation and toward retribution brought
with it a hardening of sentencing and correctional policies at every stage of
the criminal justice process. In the decades that followed the collapse of the
Rehabilitative Ideal, punishment became not only more predictable, but
more harsh. 27 On the front end of sentencing, the number of crimes increased
as lawmakers criminalized conduct previously deemed merely antisocial or
ill advised. 28 Maximum penalties for crimes increased, 29 and mandatory
minimum sentences, penalty enhancements for repeat offenders, and terms
of lifetime supervision all became tools for ensuring that criminal offenders
were held to account for their infractions. 30 During this same period,
changes also occurred on the back end of the sentencing process. Prisonbased educational, vocational, and rehabilitative programs decreased.31 (After all, if nothing worked, then prison programs did not deserve to receive
would result in more certain and more severe sentences that would reduce crime.”) (footnote
omitted); PAMALA L. GRISET, DETERMINATE SENTENCING: THE PROMISE AND THE REALITY OF
RETRIBUTIVE JUSTICE 184 (1991) (noting that the California Prisoners Union supported the
adoption of determinate sentencing in California “because of its perceived equity and fairness”). See also David Dolinko, Three Mistakes of Retributivism, 39 UCLA L. REV. 1623
(1992) (providing and dismissing retributivists’ claims to moral advantage over advocates of
other influential punishment theories); Richard S. Frase, Sentencing Principles in Theory and
Practice, 22 CRIME & JUST. 363 (1997) (outlining the purposes, principles and emergence of
the theory of limiting retributivism).
27
This harshness included enhanced penalties for repeat offenders, as well as longer
sentences in some instances, brought about by the adoption of more uniform sentencing guidelines adopted in some states, see Dhammika Dharmapalaa, Nuno Garoupaa, & Joanna M.
Shepherd, Legislature, Judges, and Parole Boards: The Allocation of Discretion Under Determinate Sentencing, 62 Fla. L. Rev. 1037, 1049, n.45, 70 (2010) (discussing increased sentence lengths under state guidelines), and by decreased opportunities for both discretionary
and mandatory parole release, see, e.g., Anne Yantus, Sentence Creep: Increasing Penalties
in Michigan and the Need for Sentencing Reform, 47 U. MICH. J.L. REFORM 645, 691 (2014)
(reporting that in Michigan “[i]n 1991, only 16.5 percent of prisoners were not paroled on
their earliest release dates, while in 2003 nearly thirty-five percent of prisoners were serving
past the first parole eligibility date”).
28 Marc A. Levin, At the State Level, So-Called Crimes are Here, There, Everywhere,
28 CRIM. JUST. 4 (2013) (noting that “[i]n Texas, lawmakers have created over 1,700 criminal
offenses, including 11 felonies relating to harvesting and handling oysters”).
29 In Wisconsin, the maximum term of imprisonment for a Class B felony increased
from twenty years imprisonment to forty years in 1994 and then to sixty years in 1999. Compare WIS. STAT. ANN § 939.50(3)(b) (West 1990), with WIS. STAT. ANN § 939.50(3)(b) (West
1994), and with WIS. STAT. ANN § 939.50(3)(b) (West 1999).
30 Michael Vitiello, Three Strikes: Can We Return to Rationality?, 87 J. CRIM. L. &
CRIMINOLOGY 395, 400, 425 (1997).
31 DIANA BRAZZELL ET AL., URBAN INST., FROM THE CLASSROOM TO THE COMMUNITY:
EXPLORING THE ROLE OF EDUCATION DURING INCARCERATION AND REENTRY 10–12 (2009),
https://www.prisonlegalnews.org/media/publications/urban_institute_class_to_community_education_role_in_reentry_2009.pdf; T.A. Ryan, Correctional Education: Past is Prologue to the Future, 46 J. CORRECTIONAL EDUC. 60, 60–62 (1995).

ND SSRN.DOCX (DO NOT DELETE)

110

11/30/2015 11:20 AM

NOTRE DAME LAW REVIEW

[VOL. XX:N

taxpayer dollars. 32) Legal mechanisms for softening sentence lengths, such
as discretionary parole and sentence credit for good behavior, were also restricted or abolished in many jurisdictions. 33 By the end of the 1990s, 84
percent of states had adopted determinate sentencing laws that severely limited the ability of prisoners to seek discretionary parole release. 34
While retributive policies were gaining traction, a second set of changes
was also occurring in response to the collapse of the Rehabilitative Ideal. All
critics of rehabilitation accepted the implausibility of “curing” criminality,
but not all agreed that punishment alone was a sufficient response. After all,
if a past criminal offender could not be disabused of his propensity to offend
through treatment or re-education, and if the conditions in which his offense
occurred could not be easily remedied, then the risk of future offending remained. To those concerned about the potential for crime and disorder posed
by unrehabilitated individuals, the primary governmental concern was
providing for public safety. With intervention in the lives of individual offenders now deemed fruitless, the response became a bureaucracy around
risk containment.
As crime rates rose and the size of the American penal state began to
grow from the 1970s into the 1990s, new tools were needed for managing
the growing number of people subject to state control. Specific crime and
punishment policies adopted during this period had the effect of controlling
poor communities through broad and aggressive use of policing, prosecution,
confinement, and community supervision for those deemed risks to public
safety. 35 This “new penology,” as Malcom Feeley and Jonathon Simon
32 Cf. DAVID FARABEE, RETHINKING REHABILITATION: WHY CAN’T WE REFORM OUR
CRIMINALS? 25 (2005) (suggesting that most prison-based correctional programs were ineffective and wasteful). Even if programs had no effect on reducing recidivism—a fact that
remains hotly contested—such programs still served to fill empty time and provide some solace and stimulation to those serving sentences of incarceration. And, while reduced in number
and size, limited programs continued to operate in nearly all American prisons. DAVID GARLAND, THE CULTURE OF CONTROL: CRIME AND SOCIAL ORDER IN CONTEMPORARY SOCIETY 170
(2001) (reporting on a 1995 survey by the U.S. Department of Justice finding that 97 percent
of prisons offered counseling, 90 percent offered drug treatment, and 60 percent offered employment counseling or skills classes).
33 Fueled by federal funds designed to promote “truth in sentencing,” many states abandoned or severely restricted the use of indeterminate sentencing. 42 U.S.C. § 13704 (2012)
(making prison-building grants available to states that adopted determinate sentencing practices).
34 WILLIAM J. SABOL ET AL., URBAN INST., THE INFLUENCES OF TRUTH-IN-SENTENCING
REFORMS ON CHANGES IN STATES’ SENTENCING PRACTICES AND PRISON POPULATIONS 7
(2002).
35 See generally Malcolm M. Feeley & Jonathan Simon, The New Penology: Notes on
the Emerging Strategy of Corrections and its Implications, 30 CRIMINOLOGY 449 (1992). See
also GARLAND, supra note 30 (providing a similar framework for analyzing the exertion of
mass social control in both the United Kingdom and the United States in the name of public

ND SSRN.DOCX (DO NOT DELETE)

11/30/2015 11:20 AM

201X]

111

termed it, was really a managerial strategy that emphasized risk over culpability, and relied heavily on aggregate data to identify and respond to perceived threats to public safety. In contrast to the “old penology,”
[T]he new penology is markedly less concerned with responsibility, fault,
moral sensibility, diagnosis, or intervention and treatment of the individual offender. Rather, it is concerned with techniques to identify, classify,
and manage groupings sorted by dangerousness. The task is managerial,
36
not transformative.

Rather than focusing on the reasonableness of individual behavior, the
new penology focused on administration. Its focus was not on punishment
at all, but on “identifying and managing unruly groups.” 37 “Its goal [was]
not to eliminate crime but to make it tolerable through systemic coordination.” 38
The tools of the new penology were tools of classification and containment: day reporting centers, drug testing, electronic monitoring, and risk assessment instruments, used to divide individuals by their statistical likelihood of engaging in future criminal activity. 39 While the surveillance and
supervision that characterized the new penology were not intended to rehabilitate individuals convicted of crime, they nonetheless managed to entangle
those individuals and their families in a net of state-mandated social control.
Though not intended to “cure,” programmatic and administrative interventions were used for the purpose of controlling behavior when possible; when
such community-based management techniques failed, confinement was often the result.
The new methods of responding to crime discussed above—both retributive and managerial—brought with them a change in the rhetoric of criminal justice. Probation officers, who in gentler times spoke of “assisting” their

safety). Somewhat ironically, the term “new penology” is not new at all, and was used by
rehabilitation advocates at the turn of the 20th Century to describe the move they advocated
from a punishment-based model to the rehabilitative treatment model discussed supra, Part I.
Cullen, supra note 14, at 310–11.
36 Feeley & Simon, supra note 33, at 452 (citation omitted).
37 Id. at 455.
38 Id.
39 Id. at 455–56; JONATHAN SIMON, POOR DISCIPLINE: PAROLE AND THE SOCIAL CONTROL OF THE UNDERCLASS, 1890–1900, at 169–89 (1993).

ND SSRN.DOCX (DO NOT DELETE)

112

11/30/2015 11:20 AM

NOTRE DAME LAW REVIEW

[VOL. XX:N

“charges,” began to talk about “managing” the “offenders” on their caseloads. 40 Politicians used executions as campaign fodder, 41 supported legislation imposing life sentences for nonhomicide offenses, 42 and, predicting a
wave of “juvenile superpredators,” 43 urged states to lower the age at which
children could be tried as adults. 44
The result of these changes was unprecedented growth in the U.S.
prison population. From 1970 to 2010, the number of U.S. prisoners skyrocketed from 196,429 to more than 1.5 million, 45 while the number of people confined in local jails increased at roughly the same rate.46 By the turn
of the century, roughly one of every thirty-five adults in the United States
was under some form of correctional control, and nearly one in one hundred
was behind bars. 47
B. The Fiscal Crisis and the Changing Language of Correctional Reform
Growth in the correctional population meant growth in the correctional
arm of state governments. Between 1977 and 1995, spending on incarceration increased 823% (compared to an increase of 374% for higher education
40 See generally Joseph E. Kennedy, Monstrous Offenders and the Search for Solidarity
Through Modern Punishment, 51 HASTINGS L.J. 829 (2000) (discussing the emphasis on rhetoric that has accompanied severity in punishment). But see Michelle S. Phelps, Rehabilitation
in the Punitive Era: The Gap Between Rhetoric and Reality in U.S. Prison Programs, 45 LAW
& SOC’Y REV. 33 (2011) (acknowledging the clear change in rhetoric but arguing that the
actions taken by corrections officials did not match the rhetoric).
41 While campaigning for the presidency, Bill Clinton, then Governor of Arkansas, famously “suspended his New Hampshire primary campaign to fly home to Arkansas and oversee the execution of a braindamaged inmate convicted of murdering a police officer.” David
B. Holian, HE’S STEALING MY ISSUES! Clinton’s Crime Rhetoric and the Dynamics of
Issue Ownership, 26 POL. BEHAV. 95, 96 (2004). That public statement of support for capital
punishment was viewed by many observers as an important step in recapturing the Democratic
Party’s credibility with the public on crime issues.
42 See Michael Vitiello, supra note 28 (arguing that politicians were fully aware that
even nonviolent offenders could be sentenced to life imprisonment under California’s Three
Strikes law).
43 John J. DiIulio, Jr., The Coming of the Super-Predators, WKLY. STANDARD, Nov. 27,
1995,
at
23–28,
http://www.weeklystandard.com/Content/Protected/Articles/000/000/007/011vsbrv.asp.
44 See generally Alison Powers, Note, Cruel and Unusual Punishment: Mandatory Sentencing of Juveniles Tried as Adults Without the Possibility of Youth as a Mitigating Factor,
62 RUTGERS L. REV. 241 (2009).
45 JAMES AUSTIN ET AL., UNLOCKING AMERICA: WHY AND HOW TO REDUCE AMERICA’S
PRISON POPULATION 1 (2007).
46 See JOHN SCHMITT ET AL., CTR. FOR ECON. & POL’Y RESEARCH, THE HIGH BUDGETARY COST OF INCARCERATION 7 (2010) (showing increase in jail population from 119,671 in
1960 to 785,556 in 2008).
47 GLAZE & HERBERMAN, supra note 2.

ND SSRN.DOCX (DO NOT DELETE)

201X]

11/30/2015 11:20 AM

113

spending during the same period). 48 By 2010, state and the federal governments spent approximately $80 billion on corrections annually. 49 Many
items drove these costs, including prison buildings, basic programming, and
medical care, especially for the aged and infirm. 50 Another significant portion of the growing cost was the bureaucracy required to sustain the penal
state. 51 More convicted individuals meant the need for more probation officers, prison guards, middle managers, administrative hearing officers, and
associated support staff needed to monitor compliance with terms of conditional release, maintain discipline in institutions, and keep paperwork in
good order. In many ways, this growth in the penal state was both a function
and cause of the further entrenchment of the new penology. Faced with
crushing “caseload pressures” at every stage of the criminal justice process,
system actors institutionalized practices and structures that allowed them to
track and manage large numbers, albeit at significant expense.
If the experiment in what has been called the “Punishment Impera52
tive” had produced fairer results than had the rehabilitative state, its high
price tag might have been tolerable. But the dramatic expansion of the penal
state came at a high human cost. This can be seen most clearly with respect
to the effects of mass incarceration. While imprisonment is meant to punish
convicted individuals by depriving them of liberty, the deprivations that attend imprisonment go far beyond restrictions on autonomy. 53 Given the so-

48 LOÏC WACQUANT, PUNISHING THE POOR: THE NEOLIBERAL GOVERNMENT OF SOCIAL
INSECURITY 158 (2009).
49 U.S. DEP’T OF JUSTICE, SMART ON CRIME: REFORMING THE CRIMINAL JUSTICE SYSTEM
FOR THE 21ST CENTURY 1 (2013). Incarcerating millions requires new facilities with associated staffing costs and overhead expenses, along with basic expenses for food, clothing, and
hygiene. CHRISTIAN HENRICHSON & RUTH DELANEY, VERA INST. OF JUSTICE, THE PRICE OF
PRISONS: WHAT INCARCERATION COSTS TAXPAYERS 5–6 (2012).
50 HENRICHSON & DELANEY, supra note 47, at 6. A study examining prison health care
spending found that the number of state and federal prisoners age 55 or older increased 94
percent from 2001 to 2008, from 40,200 to 77,800. PEW CHARITABLE TRS., MANAGING
PRISON HEALTH CARE SPENDING 8 (2013). The same report noted that prisoners suffer from
a “higher incidence of mental illness and chronic and infectious diseases, such as AIDS and
hepatitis C, than the general population,” contributing to the cost of their care. Id.
51 CHRISTINE S. SCOTT-HAYWARD, VERA INST. OF JUSTICE, THE FISCAL CRISIS IN CORRECTIONS: RETHINKING POLICIES AND PRACTICES 2 (2009) (“[s]taffing typically accounts for
75 to 80 percent of corrections budgets”).
52 TODD R. CLEAR & NATASHA A. FROST, THE PUNISHMENT IMPERATIVE: THE RISE AND
FAILURE OF MASS INCARCERATION IN AMERICA (2014).
53 Although it is generally agreed that, as a matter of principle, individuals are sent to
prison “as punishment, not for punishment,” the lived experience is quite different. See generally CRAIG HANEY, REFORMING PUNISHMENT: PSYCHOLOGICAL LIMITS TO THE PAINS OF IMPRISONMENT (2006). Collateral punishments come in many forms: missing births and deaths
of loved ones, worrying about personal safety, and confronting the desolation brought about

ND SSRN.DOCX (DO NOT DELETE)

114

11/30/2015 11:20 AM

NOTRE DAME LAW REVIEW

[VOL. XX:N

cial disruption, isolation, and substandard conditions that define the experience of imprisonment in America today, it is no surprise that people who are
incarcerated are at a higher risk of being re-incarcerated in the future. While
many factors drive re-imprisonment rates, reliable estimates suggest that
within five years of release, three-fourths of prisoners will be re-arrested. 54
Half of released prisoners will return to prison or jail within that same time
frame, either as a result of new criminal activity or a violation of community
supervision conditions. 55 Whether by design or default, it is clear that many
individuals serving time behind bars are failing to find effective rehabilitation behind prison walls. 56
Communities are also negatively affected by mass imprisonment. A
host of formal and informal collateral consequences—including disenfranchisement, deportation, exclusion from public housing, and limitations on
employment licensing—await those who have been incarcerated, making
them less productive parents and citizens when they return home. 57 In places
with disproportionately high rates of incarceration, traditional family structures are weakened, democratic power is diluted, and neighborhoods are destabilized: 58

by vast swaths of empty time. For a thorough discussion of the subjectively punitive aspects
of modern imprisonment, see generally ROBERT A. FERGUSON, INFERNO: AN ANATOMY OF
AMERICAN PUNISHMENT (2014).
54 MATTHEW R. DUROSE ET AL., BUREAU OF JUSTICE STATISTICS, RECIDIVISM OF PRISONERS RELEASED IN 30 STATES IN 2005: PATTERNS FROM 2005 TO 2010 1, 7 (2014). It is difficult
to say to what degree re-arrest is a sign of new criminal activity versus targeting of former
prisoners by the police for closer scrutiny and suspicion. Both factors are likely at play and
demonstrate how the effects of incarceration long outlast the court-imposed sentence.
55 Id. at 15. The violations that may justify revocation from probation or parole include
behavior ranging from engaging in new criminal activity and substance abuse to missing an
appointment or taking a trip without prior permission from a community supervision officer.
See Cecelia Klingele, Rethinking the Use of Community Supervision 103 J. CRIM. L. & CRIMINOLOGY 1015, 1030–41 (2013).
56 It could be argued that what the current system lacks in rehabilitation it makes up for
in deterrence and incapacitation: after all, crime rates have reached historic lows. There is
broad consensus, however, in the academic community that while growth in incarceration
may have accounted for a fraction of the reduction in crime seen during the 1990s and
throughout the 21st Century, the scale of imprisonment greatly outpaced its positive deterrent
and incapacitative effects. See generally, e.g., OLIVER ROEDER ET AL., BRENNAN CTR. FOR
JUSTIce, WHAT CAUSED THE CRIME DECLINE? (2015);
FRANKLIN E. ZIMRING, THE GREAT AMERICAN CRIME DECLINE (2007).
57 See generally MARGARET COLGATE LOVE, JENNY ROBERTS & CECELIA KLINGELE,
COLLATERAL CONSEQUENCES OF CRIMINAL CONVICTION: LAW, POLICY, AND PRACTICE (2013)
(describing the broad range of legally authorized collateral consequences).
58 See generally CLEAR, supra note 13. See also JEFF MANZA & CHRISTOPHER UGGEN,
LOCKED OUT: FELON DISENFRANCHISEMENT AND AMERICAN DEMOCRACY 157–63 (2006);
Pamela S. Karlan, Convictions and Doubts: Retribution, Representation, and the Debate Over

ND SSRN.DOCX (DO NOT DELETE)

11/30/2015 11:20 AM

201X]

115

By leaving a community bereft of siblings, husbands, and fathers, as well
as potential spouses, economically-contributing actors, and role models,
long-term incarceration of large numbers of principally male adults
erodes a community’s ability to maintain the informal social controls
59
serving as the first line of defense against crime and discord.

As prison populations soared, the negative consequences of mass incarceration did not go unnoticed. While the politicians pushed a “tough on
crime” agenda, critics decried America’s growing addiction to incarceration. 60 Scholars and reformers alike challenged the wisdom of the Reagan
Era War on Drugs, and pointed with concern to widening racial disparities
in incarceration. 61 But although advocates appealed to lawmakers and the
public with statistics and narrative descriptions of the effects of punitive drug
policies on minority communities, their normative critiques about American
crime policy had no discernable impact on sentencing practices. Incarceration rates continued to rise.
The turning point in the conversation about mass incarceration came
around the turn of the century when Jeremy Travis and other social scientists
and reform advocates began documenting the challenges faced by people
“re-entering” society from prison in the areas of housing, employment, and
family life, and connecting these challenges to prisoners’ high rates of recidivism. 62 Advocates used these newly-developing narratives to persuade

Felon Disenfranchisement, 56 STAN. L. REV. 1147, 1148–49 (2004); Marc Mauer, Mass Imprisonment and the Disappearing Voters, in INVISIBLE PUNISHMENT: THE COLLATERAL CONSEQUENCES OF MASS IMPRISONMENT 50, 57 (Marc Mauer & Meda Chesney-Lind, eds. 2002).
59 Paul J. Larkin, Jr., Crack Cocaine, Congressional Inaction, and Equal Protection, 37
HARV. J. L. & PUB. POL’Y 241, 290 (2014).
60 See generally Patrick A. Langan, America’s Soaring Prison Population, 251 SCIENCE
1568 (1991); Alfred Blumstein, Prison Populations: A System Out of Control? 10 CRIME &
JUST. 231 (1988).
61 See, e.g., Alfred Blumstein & Allen J. Beck, Population Growth in U.S. Prisons,
1980–1996, 26 CRIME & JUST. 17, 20 (1999) (“It is widely recognized that the ‘drug war’ has
contributed to a major growth in the number of people imprisoned for drug offenses.”); Robert
J. Sampson & Janet L. Lauritsen, Racial and Ethnic Disparities in Crime and Criminal Justice
in the United States 21 CRIME & JUST. 311 (1997). In addition, policy advocacy groups such
as Families Against Mandatory Minimums and the Sentencing Project pressed policymakers
for years about the ways in which harsh drug sentencing laws were disproportionately affecting minority communities. See, e.g., MARC MAUER, RACE TO INCARCERATE 24, 124–
25(1999).
62 See generally, e.g., FAYE S. TAXMAN ET AL., BUREAU OF GOVERNMENTAL RESEARCH,
FROM PRISON SAFETY TO PUBLIC SAFETY: INNOVATIONS IN OFFENDER REENTRY (2002) (presenting a conceptual model of the offender reentry process); JEREMY TRAVIS, NAT’L INST. OF
JUSTICE, BUT THEY ALL COME BACK: RETHINKING PRISONER REENTRY (2000) (describing the
challenges facing returning prisoners); JEREMY TRAVIS ET AL., URBAN INST., FAMILIES LEFT
BEHIND: THE HIDDEN COSTS OF INCARCERATION AND REENTRY (2d ed. 2005) (discussing the
negative effects of incarceration on prisoners’ children and extended families); JEREMY

ND SSRN.DOCX (DO NOT DELETE)

116

11/30/2015 11:20 AM

NOTRE DAME LAW REVIEW

[VOL. XX:N

Congress that the government had a role to play in easing the transition from
prison to community. 63 In 2008, Congress passed the Second Chance Act,
which authorized hundreds of millions of dollars in funding for programs
and research designed to improve outcomes for people leaving custody. 64
On its face, the Reentry Movement had nothing to say about the growth and
size of America’s prison population. Nevertheless, it provided a lens through
which policymakers became educated about the costs of incarceration.65 As
a result, when the financial crisis of 2009 hit several years later, draining
government coffers, 66 policymakers were already positioned to question
whether the status quo was worth preserving.
II. THE PROMOTION OF EVIDENCE-BASED CORRECTIONAL PRACTICES
What is old often becomes new again. The same can be said of rehabilitation and its role in the criminal justice system. Although the latter part
of the 20th Century was characterized by rapid and steady expansion of penal
populations, mass incarceration has always had its critics. Their criticism
approached the problem from many angles: challenges to the futility of the
war on drugs, attacks on profiteering by private prison corporations, and condemnation of the racial and income inequalities that have continued to characterize those subject to correctional control. Still others argued that rehabilitation had been rejected too hastily, with an inadequate appreciation for
the ways in which appropriately designed and executed interventions could
improve the lives of those within the correctional system. 67

TRAVIS ET AL., URBAN INST., FROM PRISON TO HOME: THE DIMENSIONS AND CONSEQUENCES
PRISONER REENTRY (2001); James Austin, Prisoner Reentry: Current Trends, Practices,
and Issues, 47 CRIME & DELINQUENCY 314 (2001); John Hagan & Juleigh Petty Coleman,
Returning Captives of the American War on Drugs: Issues of Community and Family Reentry,
47 CRIME & DELINQUENCY 352 (2001); Theodore Hammett et al., Health-Related Issues in
Prisoner Reentry, 47 CRIME & DELINQUENCY 390 (2001); Bruce Western et al., The Labor
Market Consequences of Incarceration, 47 CRIME & DELINQUENCY 410 (2001).
63 With the support of a bipartisan coalition of advocacy groups ranging from “George
Soros’s Open Society Institute [to] Chuck Colson’s Prison Fellowship.” Jeremy Travis, Reflections on the Reentry Movement, 20 FED. SENT’G REP. 84, 84 (2007).
64 Second Chance Act of 2007, Pub. L. No. 110-199, 122 Stat. 657 (2008). Congress
was pushed to pass the legislation by a bipartisan coalition of advocacy groups ranging from
“George Soros’s Open Society Institute [to] Chuck Colson’s Prison Fellowship.” Travis, supra note 61, at 84.
65 Cecelia Klingele, The Early Demise of Early Release, 114 W. VA. L. REV. 415, 424
(2012).
66 For a discussion of the effects of the financial crisis on state governments’ correctional spending, see SCOTT-HAYWARD, supra note 49, at 4.
67 See Francis T. Cullen & Paul Gendreau, Assessing Correctional Rehabilitation: Policy, Practice, and Prospects, in 3 POLICIES, PROCESSES, AND DECISIONS OF THE CRIMINAL JUSTICE SYSTEM 109, 124–31 (2000).
OF

ND SSRN.DOCX (DO NOT DELETE)

201X]

11/30/2015 11:20 AM

117

These latter reformers, many of whom were criminologists by training,
began to carefully document correctional programs and practices that were
shown to reduce future offending. Over time, organizations such as the National Institute of Corrections began to re-introduce the idea that correctional
programming—which had continued to be offered throughout the later decades of the 21st century, albeit in reduced and often haphazard ways—could
be used effectively to reduce recidivism and aid prisoners in successful reentry. With its emphasis on the importance of using research data to identify
and evaluate successful programs, this general approach to correctional intervention came to be known as “evidenced-based” practice.
The influence of evidence-based correctional practices on current criminal justice reform efforts is hard to overstate. In the span of a single decade,
correctional agencies throughout the country have moved from a position of
skepticism with respect to rehabilitative interventions to a full-on embrace
of practices that promise to reduce risk of re-offending by convicted persons—often in a non-custodial setting. In jurisdictions across the country,
probation officers now discuss their contacts with clients in terms of “dosage;” 68 magistrates and correctional officers routinely employ actuarial risk
assessment instruments in deciding whether to grant bail, how often to require reporting, and whether to grant parole; 69 and judges increasingly refer
to defendants’ “criminogenic needs” when imposing sentence. 70 What happened?
The following sections explain in greater detail what is meant by evidence-based correctional practice, and discuss the basic mechanisms by
which they are being adopted by jurisdictions around the country.

68 Cf. MADELINE M. CARTER & HON. RICHARD J. SANKOVITZ, CTR. FOR EFFECTIVE PUB.
POL’Y, DOSAGE PROBATION: RETHINKING THE STRUCTURE OF PROBATION SENTENCES, (2014).
69 See, e.g., ARK. CODE ANN. § 16-93-615 (a)(1)(B) (2015) (“The determination . . .
shall be made by reviewing information such as the result of the risk-needs assessment to
inform the decision of whether to release a person on parole by quantifying that person’s risk
to reoffend, and if parole is granted, this information shall be used to set conditions for supervision.”); N.H. REV. STAT. ANN. § 504-A:15(I) (2011) (requiring that “[e]very person placed
on probation or parole . . . be assessed by the department of corrections, using a valid and
objective risk assessment tool, to determine that person’s risk of recidivating” and that the
results be used to determine the length of active supervision); VT. STAT. ANN. tit. 13,
§ 7554c(a)(1) (2015) (“The objective of a pretrial risk assessment is to provide information
to the Court for the purpose of determining whether a person presents a risk of nonappearance
or a threat to public safety so the Court can make an appropriate order concerning bail and
conditions of pretrial release.”).
70 Cf. PAMELA M. CASEY ET AL., NAT’L CTR. FOR STATE COURTS, USING OFFENDER RISK
AND NEEDS ASSESSMENT INFORMATION AT SENTENCING: GUIDANCE FOR COURTS FROM A NATIONAL WORKING GROUP (2011).

ND SSRN.DOCX (DO NOT DELETE)

118

11/30/2015 11:20 AM

NOTRE DAME LAW REVIEW

[VOL. XX:N

A. Neorehabilitation and Evidence-Based Corrections
Even during the height of the “tough on crime” era, rehabilitative programs did not disappear from the criminal justice system entirely. Prisons
continued to employ psychologists, drug counselors, and teachers, albeit on
a scale that failed to meet demand. 71 Drug treatment, vocational training,
and secondary education remained honored components of probation orders
and parole conditions. Throughout this period, service providers themselves,
along with advocates of rehabilitation, sought to validate the importance of
these interventions, confident that they worked “not simply [as] a matter of
‘doing good’ for offenders but also of protecting public safety.” 72
Although apologists believed rehabilitative programs and practices had
intrinsic value as a means of affirming human dignity and promoting equality
for the marginalized, 73 they also realized that policymakers wanted proof that
these programs were worthy of investment, particularly in light of the skepticism that had arisen in the 1970s about the efficacy of correctional interventions. Persuading safety and accountability-conscious decisionmakers of
the value of rehabilitation meant offering objective evidence that correctional programs could, in fact, reduce crime in a cost-effective manner. That
task would require the careful collection and analysis of data—a practice
largely foreign to the criminal justice system. 74
In an essay published in 1998, Lawrence Sherman, writing about the
importance of data collection and analysis in policing, pointed to a model for
criminal justice reform: Evidence-Based Medicine. 75 Championed by Dr.
David Sackett in the early 1990s, evidence-based medicine is an approach to
71 Cullen, supra note 14, at 330–331. See also Brandon K. Applegate et al., Public
Support for Correctional Treatment: The Continuing Appeal of the Rehabilitative Ideal, 77
PRISON J. 237 (1997) (discussing the persistence of public support for rehabilitative goals in
criminal justice).
72 Cullen & Gendreau, supra note 65, at 161.
73 See generally FRANCIS T. CULLEN & KAREN E. GILBERT, REAFFIRMING REHABILITATION 247–53 (1982); Cullen, supra note 14.
74 The failure of the criminal justice system to routinely collect, analyze, and disseminate information about its programs and basic operations has been decried by scholars of
every subfield of criminal justice for the greater part of the past century. See, e.g., PRESIDENT’S COMM’N ON LAW ENF’T & ADMIN. OF JUSTICE, THE CHALLENGE OF CRIME IN A FREE
SOCIETY 273 (1967) (“Few domestic social problems more seriously threaten our welfare or
exact a greater toll on our resources [than crime]. But society has relied primarily on traditional answers and has looked almost exclusively to common sense and hunch for needed
changes.”). See also Part III.B.2., infra.
75 LAWRENCE W. SHERMAN, POLICE FOUND., EVIDENCE-BASED POLICING (1998); Evidence-Based Medicine Working Group, Evidence-Based Medicine: A New Approach to
Teaching the Practice of Medicine, 268 [J]AMA 2420, 2420–21 (1992). See also Jeffrey A.
Claridge & Timothy C. Fabian, History and Development of Evidence-Based Medicine, 29
WORLD J. SURGERY 547 (2005).

ND SSRN.DOCX (DO NOT DELETE)

201X]

11/30/2015 11:20 AM

119

patient care that requires doctors to root treatment decisions in scientificallyvalidated clinical studies and peer-reviewed reports. 76 Advocates of evidence-based medicine encourage doctors to think of themselves as researchers, whose practice is rooted in findings from scientifically-validated clinical
studies, rather than as healers who treatment decisions are based on ad hoc
observations, peer opinions, or unfounded local traditions. 77 Evidence-based
medicine grew quickly in popularity, and by the turn of the century had become the standard for training doctors and approaching patient decisionmaking. 78
As technology increased capacity for aggregating and disseminating information to professionals, the popularity of evidence-based approaches
spread throughout and beyond scientific disciplines. The evidence-based approach pioneered in medicine quickly translated to other fields requiring
clinical judgment, such as nursing and psychology, 79 and then later to the
social sciences and other structured fields of inquiry, including education.80
By the time Sherman brought the “evidence-based” label to police
work, policing itself had already been transformed by the collection and analysis of data. 81 Problem-oriented methods of policing—including situational
crime prevention—searched for patterns that predicted criminal offenses,
and sought to disrupt crime by altering incentives and hardening targets.
These policing practices—the forerunners of today’s data mining and hot
spot policing 82—demonstrated the power of relying on data over intuition to
76 Claridge & Fabian, supra note 73, at 547, 552; Evidence-Based Medicine Working
Group, supra note 73, at 2423.
77 Evidence-Based Medicine Working Group, supra note 73, at 2420–22.
78 See John Tucker, A Novel Approach to Determining Best Medical Practices: Looking
at the Evidence, 10 HOUS. J. HEALTH L. & POL’Y 147, 180 (2010) (describing how medical
school curricula at many universities now includes evidence-based medicine within the six
competencies that students must achieve before being licensed to practice).
79 See, e.g., Kirk Heilbrun et al., Standards of Practice and Care in Forensic Mental
Health Assessment: Legal, Professional, and Principles-Based Considerations, 14 PSYCHOL.
PUB. POL’Y & L. 1, 5 (2008) (noting the trend towards evidence-based practices in psychology
and other health care professions).
80 A quick search of the term “evidence-based” in the database JSTOR reveals articles
discussing evidence-based education, business, medicine, nursing, health policymaking, management, social work, and conservation.
81 See, e.g., 5 ROUTINE ACTIVITY AND RATIONAL CHOICE: ADVANCES IN CRIMINOLOGICAL THEORY (Ronald V. Clarke & Marcus Felson, eds., 1993); Kenneth Chelst, An Algorithm
for Deploying a Crime Directed (Tactical) Patrol Force, 24 MGMT. SCI. 1314 (1978); Lawrence W. Sherman, Attacking Crime: Police and Crime Control, 15 CRIME & JUST. 159, 176–
81 (1992) (discussing proactive strategies of policing that require assessment of data regarding offenders, crime targets, and crime locations).
82 See generally Natalia Lazzati & Amilcar Menichini, Hot Spot Policing: A Study of
Place-Based Strategies to Crime Prevention (Mar. 4, 2014) (unpublished manuscript); Colleen
McCue, Connecting the Dots: Data Mining and Predictive Analytics in Law Enforcement and

ND SSRN.DOCX (DO NOT DELETE)

120

11/30/2015 11:20 AM

NOTRE DAME LAW REVIEW

[VOL. XX:N

spot and reduce crime. 83 The success of these data-driven policing methods
opened up the possibility that a greater focus on data might lead to more
efficient uses of other criminal justice resources, as well.
Early advocates of “evidence-based corrections” were primarily criminologists, like Francis Cullen and Paul Gendreau, who saw evidence-based
correctional practices as a framework for revitalizing the Rehabilitative Ideal
and affirming the value of criminological research. They understood that the
criminal justice system, unlike medicine, was inherently subject to a multitude of unscientific pressures and considerations. “[C]orrections will never
be the exclusive domain of ‘what works,’” they properly noted; “policy decisions will reflect fundamental cultural values, organizational resources,
and political realities—among other factors.” 84 Even so, Cullen and Gendreau hoped that evidence-based approaches would encourage criminal justice decisionmakers to exercise discretion not “based merely on custom or
common sense but on [their] research knowledge about what is the ‘best bet’
to reduce offender recidivism.” 85 They were confident that reliance on such
data would lead to a fresh embrace of rehabilitative interventions and tools,
and ultimately to a system that was more humane than the ever-expansive
penal state. 86 Cullen and Gendreau were also convinced that criminology
was well positioned as a field to collect and analyze data, providing increasingly reliable “evidence” upon which future reforms could build.
B. What Evidence? And Which Practices?
The practitioners and criminologists who advocated for a new rehabilitative model of corrections were right to be concerned about ad hoc decisionmaking. Although “evidence” has always played an important role in
Intelligence Analysis, 70 POLICE CHIEF (2003), http://www.policechiefmagazine.org/magazine/index.cfm?fuseaction=display_arch&article_id=121&issue_id=102003.
83 See generally Anthony A. Braga & Brenda J. Bond, Policing Crime and Disorder
Hot Spots: A Randomized Controlled Trial, 46 CRIMINOLOGY 577 (2008) (reporting significant reductions in crime and disorder calls for service from hot spot policing with no significant displacement of crime); David Weisburd et al., Is Problem-Oriented Policing Effective
in Reducing Crime and Disorder?, 9 CRIMINOLOGY & PUB. POL’Y 139 (2010) (using metaanalysis to conclude that problem-oriented policing techniques yield at least modest reductions in crime and disorder). Notably, however, those same data-driven techniques have also
opened up a conversation about the ways in which aggressive monitoring of information can
impinge on the privacy and collective sense of safety felt by heavily monitored neighborhoods. CITE
84 Francis T. Cullen & Paul Gendreau, Assessing Correctional Rehabilitation: Policy,
Practice, and Prospects, in 3 CRIMINAL JUSTICE 2000: POLICIES, PROCESSES AND DECISIONS
OF THE CRIMINAL JUSTICE SYSTEM
109, 158 (Julie Horney, ed. 2000),
https://www.ncjrs.gov/criminal_justice2000/vol_3/03d.pdf.
85 Id.
86 Insert Cullen quote here & reference to the What Works lit.

ND SSRN.DOCX (DO NOT DELETE)

201X]

11/30/2015 11:20 AM

121

criminal justice, from crime scene investigation through trial and sentencing,
what the word means changes depending on the context in which it is used.
At trial, evidence can be testimonial, physical, or scientific, and expert or
lay. It can be probative, dispositive, or irrelevant. In the adjudicative stages
of criminal proceedings, rules of evidence and constitutional due process
protections govern the kind of evidence upon which decisionmakers can rely,
and how much weight can be given to different kinds of evidence in various
circumstances. Those same rules and protections do not apply, and traditionally have not been applied, to correctional decisionmaking.
Although some correctional decisions (such as whether to revoke parole) are accompanied by due process protections that require a modicum of
evidence, many decisions are made without reliance on tested facts of any
kind. Whether to require mental health counseling; which drug treatment
program to order; what housing to approve; how long supervision should
last . . . all of these decisions, while of utmost importance to the people being
sanctioned, have traditionally been made ad hoc, in response to a judge’s
intuitions or a probation officer’s habitual practice. As a result, a host of
programmatic interventions have been imposed on defendants over the years
that were later shown to be ineffective or even counter-productive. 87
In the context of evidence-based correctional practice, “evidence” is
broadly defined as “findings from empirically sound social science research” 88—a definitional choice that makes it easy to see why social scientists have been among its strongest promoters. Evidence-based practice, by
extension, is any correctional practice or intervention whose effectiveness at
achieving its stated goal is supported by “empirically sound” research of
some kind. Advocates of evidence-based correctional practice contrast reliance on such research findings with reliance on hunches, instincts, or best
guesses about “what works” in corrections—approaches they suggest have
defined criminal justice interventions in the past.
Even among those correctional practices that qualify as “evidencebased” under this standard, there is a wide range in the quality of evidence
that supports various interventions. Within the hierarchy of “evidence,”
findings derived from double-blind controlled studies are considered the
most desirable, while shared anecdotal observations are considered the most
87 These include the abusive rehabilitative practices of the early 20th Century discussed
supra, Part I.A, along with more modern interventions, such as boot camps. See, e.g., DALE
G. PARENT, NAT’L INST. OF JUST., CORRECTIONAL BOOT CAMPS: LESSONS FROM A DECADE OF
RESEARCH 1 (2003), https://www.ncjrs.gov/pdffiles1/nij/197018.pdf (finding that although
boot camps “had positive effects on the attitudes, perceptions, behavior, and skills of inmates
during their confinement” with “limited exceptions, these positive changes did not translate
into reduced recidivism”).
88 CTR. FOR EFFECTIVE PUB. POL’Y ET AL., A FRAMEWORK FOR EVIDENCE-BASED DECIMAKING IN LOCAL CRIMINAL JUSTICE SYSTEMS 7 (3d ed. 2010),
SION
http://www.cepp.com/documents/EBDM%20Framework.pdf.

ND SSRN.DOCX (DO NOT DELETE)

122

11/30/2015 11:20 AM

NOTRE DAME LAW REVIEW

[VOL. XX:N

suspect. 89 Technically speaking, any practice supported by reference to any
kind of information may be dubbed “evidence-based;” however, the term is
ordinarily reserved for those correctional interventions that have been subjected to formal assessment and have been shown to demonstrate positive
outcomes. While any practice that relies on accumulated knowledge can be
labeled “evidence-based” in one sense, the research support gradient (figure
1, below) is used by proponents of evidence-based practice to encourages
system actors to promote the best-tested interventions available, and to develop additional data about new and existing programs by subjecting them
to evaluation using control groups and replication studies whenever possible. 90
FIGURE 1
RESEARCH SUPPORT GRADIENT
GOLD
• Experimental/control research design with controls for attrition
• Significant sustained reductions in recidivism obtained
• Multiple site replications
• Preponderance of all evidence supports effectiveness
SILVER
• Quasi-experimental control research with appropriate statistical
controls for comparison group
• Significant sustained reductions in recidivism obtained
• Multiple site replications
• Preponderance of all evidence supports effectiveness
BRONZE
• Matched comparison group without complete statistical controls
• Significant sustained reductions in recidivism obtained
• Multiple site replications
• Preponderance of all evidence supports effectiveness
IRON
• Conflicting findings and/or inadequate research designs
DIRT
• Silver and Gold research showing negative outcomes

89 See Suzette Glasner-Edwards & Richard Rawson, Evidence-Based Practices in Addiction Treatment: Review and Recommendations for Public Policy, 97 HEALTH POL’Y 93, 95
tbl.1 (2010).
90 BRAD BOGUE ET AL., CRIME & JUSTICE INST., IMPLEMENTING EVIDENCE-BASED PRACTICE IN COMMUNITY CORRECTIONS: THE PRINCIPLES OF EFFECTIVE INTERVENTION 17 (2004)
[hereinafter PRINCIPLES OF EFFECTIVE INTERVENTION].

ND SSRN.DOCX (DO NOT DELETE)

11/30/2015 11:20 AM

201X]

123

Excerpted from Principles of Effective Intervention, supra note 88, at 17.

The prospect of wading through literature on human behavior, psychology, and medicine to locate practices that are supported by sound research is
a daunting task for most criminal justice agencies, many of whom do not
employ analysts or other formally trained social scientists. System actors
themselves are often given little training in statistical methods and many do
not possess degrees in fields that would permit easy comprehension of the
type of social science literature on which evidence-based correctional practices are based. 91 While awareness of the research gradient may promote
better criminal justice data collection and analysis in the future, the fact remains that very few correctional practices in use today can meet the gold
standard, or even the bronze. 92
Since the Wickersham Commission of the 1930s, system actors and administrators have lamented the lack of readily available data about the operation of the criminal justice system, 93 and with good reason. The criminal
justice system lags behind most other government agencies when it comes
to data tracking, for a very simple reason: the “system” is not a system at all.
Instead, it is a loose affiliation among independent law enforcement agencies, individual counties, local jails, and state prisons. Computer databases
are often incompatible among agencies, even within the same county. Police
records are not accessible to courts or corrections, and as a result it is hard to
91 See, e.g., 13 ALASKA ADMIN. CODE tit. 13, § 85.210 (2015) (setting the basic employment educational standard for all probation, parole, and correctional officers as a high school
diploma or its equivalent); GA. COMP. R. & REGS. 503-1-.21(b) (2015) (requiring probation
officers at the time of appointment to have completed a “standard two-year college course of
study or 90 quarter hours or 60 semester hours from an accredited institution or have four
years of law enforcement experience as a certified [sic] peace officer or jurisdictional equivalent”); 12 N.C. ADMIN. CODE 09G .0204(b) (2015) (requiring probation and parole officers
to possess a bachelor’s degree in any field).
92 The reasons for this dearth of quantitative information are many. First, a lack of
access to good data has limited the ability of system administrators to test the effectiveness of
various interventions with anything approaching scientific rigor. Perhaps as a result, criminal
justice programs have traditionally been evaluated more by “feel” than by reference to quantifiable proof, with the “success” of programs measured by their popularity with administrators, participants, and the public, rather than on the degree to which they achieve their stated
goals. An infamous example of this phenomenon is the Drug Abuse Resistance Education
(DARE) program, which began in 1983 as a tool for educating school children on the dangers
of drug abuse and continues to bring police officers into schools to teach children about drugs,
despite the fact that formal evaluations have consistently shown that children who complete
the program use drugs at the same rate as children not exposed to the program. See MARJORIE
E. KANOF, U.S. GOV’T ACCOUNTABILITY OFF., GAO-03-172R, YOUTH ILLICIT DRUG USE PREVENTION: DARE LONG-TERM EVALUATIONS AND FEDERAL EFFORTS TO IDENTIFY EFFECTIVE
PROGRAMS 5 (2003), http://www.gao.gov/new.items/d03172r.pdf.
93 See Michael Tonry, Evidence, Ideology, and Politics in the Making of American
Criminal Justice Policy, 42 CRIME & JUST. 1, 1 (2013).

ND SSRN.DOCX (DO NOT DELETE)

124

11/30/2015 11:20 AM

NOTRE DAME LAW REVIEW

[VOL. XX:N

know who is being sentenced to what, much less whether the sentences imposed are effective at preventing recidivism or aiding in the process of desistance from crime. Moreover, many agencies track only the most basic
information about crimes and offenders, and fail to engage in any systematic
review of the effectiveness of various formal interventions on the behaviors
they seek to alter.
With the exception of law enforcement agencies 94—some of which employ crime analysts and many of which carefully track information relevant
to crimes and suspected offenders—most criminal justice agencies lack the
ability to track and analyze data in sophisticated ways. When efforts are
made to assess the effectiveness of criminal justice programs, budget cuts,
personnel changes, and changing agency priorities make it difficult for programs to remain stable long enough for reliable results to be collected. When
studies are conducted—often by program administrators themselves since
few agencies fund trained researchers—it is often difficult to know which of
many possible components of a program is responsible for its success or failure. While some of these challenges are common to other settings in which
social scientists work (such as schools, for example), many have observed
that the criminal justice system provides unique challenges for those wishing
to develop a body of reliable knowledge about “what works” in the correctional context. 95
Perhaps as a result of the limited data currently available, promoters of
evidence-based correctional practices have derived from the relatively small
body of relevant research literature a number of “core principles” of evidence-based practice in the field of sentencing and corrections. These principles include using actuarial risk prediction instruments to assess individual
risks and needs; 96 using behavior management techniques, including rewards
94 Unlike courts and corrections, policing has been heavily influenced by research for
many decades. See id. at 3–4 (naming John Eck, Herman Goldstein, George Kelling, Stephen
Mastrofsky, Mark H. Moore, Lawrence W. Sherman, Wesley G. Skogan, Michael E. Smith,
David Weisburd, and James Q. Wilson as scholars whose research has had a significant impact on police practices).
95 See generally Alexander Volokh, Do Faith-Based Prisons Work?, 63 ALA. L. REV.
43 (2011) (discussing the problem of selection bias that arises in voluntary programs/treatment settings).
96 See PRINCIPLES OF EFFECTIVE INTERVENTION, supra note 88, at 3 (“Assessing offenders in a reliable and valid manner is a prerequisite for the effective management (i.e.: supervision and treatment) of offenders. Timely, relevant measures of offender risk and need at
the individual and [population] levels are essential for the implementation of numerous principles of best practice in corrections . . . . Screening and assessment tools that focus on dynamic and static risk factors, profile criminogenic needs, and have been validated on similar
populations are preferred.”). The selection of a risk instrument is a matter of some debate in
the field, and the popularity of specific instruments (which, importantly, are not duplicative
of one another) varies tremendously from one jurisdiction to another, and even from on
agency to another within the same jurisdiction. Federal courts and probation officers use the

ND SSRN.DOCX (DO NOT DELETE)

201X]

11/30/2015 11:20 AM

125

for good behavior and swift sanctions for bad behavior, to “motivate[e] . . .
change;” 97 and engaging pro-social community members and resources to
help influence and structure the lives of convicted individuals. 98 These guiding principles are intended to provide a framework for agencies as they work
to adopt more specific evidence-based interventions in assessment and treatment. 99
Post Conviction Risk Assessment specially designed for use in the federal system. See Christopher T. Lowenkamp et al., The Federal Post Conviction Risk Assessment (PCRA): A Construction and Validation Study, 10 PSYCHOL. SERVS. 87 (2013). Other actuarial risk prediction
tools in widespread use include COMPAS, PACT, LS/CMI, the YASI, the Level of Service
Inventory-Revised (LSI-R), the Psychopathy Checklist-Revised (PCL-R), the Static-99, the
Violence Risk Appraisal Guide (VRAG), and the Historical, Clinical, Risk Management-20
(HCR-20). See CHRISTOPHER BAIRD, NAT’L COUNCIL ON CRIME & DELINQUENCY, A QUESTION OF EVIDENCE: A CRITIQUE OF RISK ASSESSMENT MODELS USED IN THE JUSTICE SYSTEM 3
(2009); Seena Fazel et al., Use of Risk Assessment Instruments to Predict Violence and Antisocial Behaviour in 73 Samples Involving 24,827 People: Systematic Review and Meta-Analysis, BMJ (2012), http://bmj.com/content/345/bmj.e4692. Instruments used for special subpopulations include the Sex Offender Risk Appraisal Guide (SORAG), Sexual Violence Risk20 (SVR-20), the Spousal Assault Risk Assessment (SARA), and the Structured Assessment
of Violence Risk in Youth (SAVRY). Id.
97 PRINCIPLES OF EFFECTIVE INTERVENTION, supra note 88, at 4 (“Staff should relate to
offenders in interpersonally sensitive and constructive ways to enhance intrinsic motivation
in offenders. Behavioral change is an inside job; for lasting change to occur, a level of intrinsic motivation is needed. Motivation to change is dynamic and the probability that change
may occur is strongly influenced by interpersonal interactions . . . .”).
98 See id. at 6 (“Realign and actively engage pro-social supports for offenders in their
communities. Research indicates that many successful interventions with extreme populations (e.g., inner city substance abusers, homeless, dual diagnosed) actively recruit and use
family members, spouses, and supportive others in the offender’s immediate environment to
positively reinforce desired new behaviors.”).
99 Among the more specific correctional interventions and policies that have been promoted as evidence-based are programs that divert substance abusers into drug and alcohol
treatment, see ALISON LAWRENCE & DONNA LYONS, NAT’L CONFERENCE OF STATE LEGISLATURES, CRIME BRIEF: JUSTICE REINVESTMENT 3 (2013), http://www.ncsl.org/Documents/CJ/July2013CrimeBrief.pdf (reporting that Kentucky’s use of justice reinvestment to
work towards rehabilitation of substance abusers. In General Assembly reinvested savings of
nearly $6.8 million in new substance abuse treatment programs and provided almost $9 million through fiscal year 2014 for local correctional facilities and programs), changes in supervision practices and revocation policies that emphasize swift and certain (but usually short
and sometimes noncustodial) responses to rule violations, see, e.g., Mark A.R. Kleiman,
Smart on Crime, 28 DEMOCRACY 51, 60 (2013); PEW CTR. ON THE STATES, THE IMPACT OF
HAWAII’S HOPE PROGRAM ON DRUG USE, CRIME AND RECIDIVISM 1 (2010) (finding that participants in swift and certain program were “55 percent less likely to be arrested for a new
crime, 72 percent less likely to use drugs, 61 percent less likely to skip appointments with
their supervisory officer and 53 percent less likely to have their probation revoked” than nonparticipants), and the use of “motivational interviewing” techniques by probation officers to
promote pro-social behavior change in individuals under supervision. See COUNCIL OF STATE
GOV’TS JUSTICE CTR, JUSTICE REINVESTMENT IN IDAHO: ANALYSIS & POLICY FRAMEWORK 18

ND SSRN.DOCX (DO NOT DELETE)

126

11/30/2015 11:20 AM

NOTRE DAME LAW REVIEW

[VOL. XX:N

A good example of the way in which reformers hope to see “evidence”
and data change correctional practice can be found in the use of risk and need
assessment tools. Predicting the risk that a convicted person will commit
future crimes and thereby endanger the community has long been an important piece of correctional decisionmaking. Judges weigh risk when deciding on a sentence length and when selecting between community-based
and custody-based sanctions. Prison officials consider it when making security classification decisions, and paroling officials rely on it when deciding
whom to release from prison, and under what conditions.
But free will means that human behavior is not easily predictable, and
studies have shown that criminal justice system actors are not particularly
omniscient when it comes to predicting who is—and is not—most likely to
criminally re-offend. Hunches about “risk” are often rooted in misinformation and subconscious biases about race, class, and culture that often bear
only passing resemblance to actual dangerousness.100
Against this backdrop, proponents of evidence-based approaches to correctional risk management have argued that statistical prediction methods
outperform human intuition in identifying those at greatest risk of re-offense. 101 Moreover, when risk profiles are augmented with information
about a defendant’s “criminogenic needs”—that is, the deficiencies most
strongly correlated with risk of future criminality—correctional officials can
tailor sentencing conditions to target for intervention people most likely to
benefit from correctional programming. 102
While tools for managing and classifying risks posed by criminal offenders have been in use for more than a century, 103 in the early years of the
21st Century, advocates of evidence-based practices began more forcefully
asserting that better data analysis practices had enabled these tools to evolve
over time, making them fairer and more reliable.104 They argued that using

(2014), https://www.bja.gov/Publications/CSG-IdahoJusticeReinvestment.pdf (recommending that all current and new community correctional officers be trained in core correctional
practices including motivational interviewing, by the end of 2016).
100 See, e.g., Shima Baradaran, Race, Prediction, and Discretion, 81 GEO. WASH. L. REV.
157 (2013) (finding that despite popular predictions to the contrary, drug possession bears no
relation to violent crime).
101 See Tracey L. Treger, One Jury Indivisible: A Group Dynamics Approach to Voir Dire,
68 CHI.-KENT L. REV. 549, 564–65 (1992) (citing Michael J. Saks, The Limits of Scientific Jury
Selection: Ethical and Empirical, 17 JURIMETRICS J. 8–9, nn. 20–21 (1976)).
102 Ctr. on Sentencing & Corrections & Vera Inst. of Justice, The Potential of Community

Corrections to Improve Communities and Reduce Incarceration, 26 FED. SENT’G REP. 128,
135 (2013).
103 See, e.g., John Monahan & Jennifer L. Skeem, Risk Redux: The Resurgence of Risk
Assessment in Criminal Sentencing, 26 FED. SENT’G REP. 158, 158 (2014).
104 See D. A. Andrews et al., The Recent Past and Near Future of Risk and/or Need
Assessment, 52 CRIME & DELINQUENCY 7, 8 (2006) (“[T]heoretical, empirical, and applied

ND SSRN.DOCX (DO NOT DELETE)

11/30/2015 11:20 AM

201X]

127

the results of these assessments, along with better data about the kinds of
programmatic interventions that work best with specific kinds of people
(opiate users; individuals with co-occurring substance abuse and mental
health issues; women; domestic abusers; sex offenders; etc.), would reduce
the chance that people who pose a low risk of re-offense will be sent to
prison, and raise the chance that court-ordered correctional programs will
target areas of need that actually correspond to individual levels of dangerousness. 105 These are changes, they asserted, that could make the criminal
justice system simultaneously more effective and less punitive.
It is hardly surprising that criminologists and rehabilitation-minded reformers, naturally eager to promote practices they viewed as both reliable
and benign, would embrace not only the use of risk assessment tools, but
other similarly “scientific” interventions that research suggested would reduce recidivism. The key obstacle to implementing these evidence-based
practices lay with policymakers and practitioners who had embraced decades
of punitive policies and who saw rehabilitation as a failed experiment. But
here, too, advocates saw an evidence-based approach as a promising framework for opening dialogue.
Data, with its promise of impartiality, predictability, and rationality, can
be a powerful unifier in modern America, and the rhetoric of evidence-based
practice met an especially receptive audience in the world of sentencing and
corrections, where decisionmakers have long struggled to avoid decisions
about punishment that often feel unanchored or even arbitrary. Having identified the adoption of “evidence-based” correctional principles and practices
as the best hope for improving the quality and fairness of the criminal justice
system, advocates just needed a vehicle for delivering their message to policymakers.
C. Translating Theory Into Ground-Level Reform
Two organizational entities deserve much of the credit for connecting
the research findings of criminologists with correctional officials and other
criminal justice system actors capable of implementing evidence-based correctional practices: the National Institute of Corrections and the Justice Reinvestment Initiative. Since the turn of the millennium both have played key
roles in disseminating information about evidence-based correctional practices to those in the field, persuading them of the usefulness of such practices
and providing the technical assistance needed to train system actors and implement new laws and policies.

progress within the psychology of criminal conduct . . . has been nothing less than revolutionary.”).
105 See, e.g., Ctr. on Sentencing & Corrections & Vera Inst. of Justice, supra note 100,
at 134–35.

ND SSRN.DOCX (DO NOT DELETE)

128

11/30/2015 11:20 AM

NOTRE DAME LAW REVIEW

[VOL. XX:N

The National Institute of Corrections (NIC), an agency of the Department of Justice’s U.S. Bureau of Prisons, has spread information about evidence-based correctional practices in a wide variety of ways. Partnering with
groups such as the Center for Effective Public Policy and the Justice Management Institute, NIC has produced written resources for correctional agencies that set forth principles for implementing evidence-based correctional
practices at the local level; 106 provided online and in-person training on specific evidence-based correctional practices; 107 given technical assistance to
sites implementing evidence-based decisionmaking models; 108 and even developed its own popular evidence-based cognitive-behavioral program
called “Thinking for a Change.” 109
The Justice Reinvestment Initiative (JRI) is a separate public-private
partnership between the U.S. Department of Justice’s Bureau of Justice Assistance and the Pew Charitable Trusts, with involvement from the Council
of State Governments and the Vera Institute of Justice. JRI was launched in
the early 2000s with a threefold goal: (1) to analyze state data, recommending ways to reduce prison population and “generate savings for reinvestment
in local high incarceration communities;” (2) to “[e]ngage development experts to identify and steer investment opportunities;” and (3) to “[o]rganize
demand by affected communities, advocates and institutions for neighborhood reinvestment.” 110 In 2004, JRI began offering technical assistance to
states interested in reducing their prison populations. 111 In selected states,
researchers examined available data to identify the causes of correctional

See, e.g., MEGHAN GUEVARA & ENVER SOLOMON, NAT’L INST. OF CORRS., IMPLEEVIDENCE-BASED POLICY AND PRACTICE IN COMMUNITY CORRECTIONS (2d ed.
2009); RALPH C. SERIN, NAT’L INST. OF CORRS., EVIDENCE-BASED PRACTICE: PRINCIPLES FOR
ENHANCING CORRECTIONAL RESULTS IN PRISONS (2005).
107 The Institute offers a wealth of workshops, including training in motivational interviewing and counseling and risk classification within jails. See, e.g., Event Catalog, NAT’L
INST. OF CORRS., http://nicic.gov/training/.
108 See, Evidence-Based Decision Making, NAT’L INST. OF CORRS, http://nicic.gov/ebdm
(describing on the ground assistance to localities in Indiana, Virginia, and Wisconsin).
109 JACK BUSH ET AL., NAT’L INST. OF CORRS., THINKING FOR A CHANGE: INTEGRATED
COGNITIVE BEHAVIOR CHANGE PROGRAM, V (VERSION 3.1 2011)
110 AUSTIN ET AL., supra note 11, at 6. See also Susan B. Tucker & Eric Cadora, Justice
Reinvestment, 3 IDEAS FOR AN OPEN SOC’Y 2 (2003).
111 The Center was assisted in its efforts by the Pew Charitable Trusts, the Department
of Justice’s Bureau of Justice Assistance, and the Vera Institute of Justice. See LAVIGNE ET
AL., supra note 5, at 6. In 2010, Congress increased funding for these efforts through appropriations in the 2010 Omnibus Consolidated Appropriations Act under the rather cumbersome
title “Criminal Justice Improvement and Recidivism Reduction through State, Local, and
Tribal Justice Reinvestment.” Id. at 6 n.12 & 125.
106

MENTING

ND SSRN.DOCX (DO NOT DELETE)

201X]

11/30/2015 11:20 AM

129

costs and population levels, and assisted the state in developing “policy solutions that target correctional population and cost drivers.” 112 In theory, as
the policy are implemented and savings realized, a portion is to be “reinvested in evidence-based efforts to support additional public safety improvements.” 113
The attraction of this assistance for legislators and criminal justice system administrators has been primarily financial. JRI promises to save states
money—a lot of money. As of 2014, in the eight states where JRI-inspired
reforms had been in place for more than one year, projected savings ranged
“from $7.7 million (over 5 years) to $875 million (over 11 years). Total
projected savings amount to as much as $4.6 billion.” 114 These promised
savings were predicted to come primarily in the form of “averted operating
costs as a result of incarcerating a smaller population and averted
construction costs as a result of not having to build new facilities to
incarcerate larger justice system populations.” 115 And such savings do not
factor in the “reinvestment” piece of justice reinvestment, which suggests
that states take some portion of the savings they realize from reforms and
invest them in resources designed to prevent re-offending. 116 Even so, for
cash-strapped states, the promise of large-scale savings is a significant enticement.
Although each state that works with JRI receives an individualized assessment and report on the local dynamics of correctional spending, 117 reports reveal common causes for prison growth in most states. The leading
drivers of prison population growth are increasing numbers of jail and prison
sentences (as opposed to sentences of community supervision); longer sentences; fewer releases through discretionary parole; parole processing delays; and high rates of revocation from both probation and parole. 118
Because the drivers of prison growth tend to be similar across jurisdictions, so too are the solutions offered. Law and policy changes frequently
promoted by JRI have included—not surprisingly—the adoption and use of
112 NANCY G. LAVIGNE ET AL., URBAN INST., THE JUSTICE REINVESTMENT INITIATIVE:
EXPERIENCES FROM THE STATES 1 (2013). In order to receive support from JRI, states were
(and still are) required to demonstrate a bipartisan, interbranch desire for assistance by forming a team of “elected and appointed state and local officials to work with researchers and
criminal justice policy experts” supplied by JRI. Id.
113 Id.
114 LAVIGNE ET AL., supra note 5, at 3 (2014).
115 Id.
116 Id. Neither do these projections necessarily comport with real savings: early results
have shown somewhat disappointing results. See MARIE GOTTSCHALK, CAUGHT: THE PRISON
STATE AND THE LOCKDOWN OF AMERICAN POLITICS 107–08 (2015) (discussing Pennsylvania’s
mixed results with justice reinvestment).
117 See LAVIGNE ET AL., supra note 5, at 57–123 (describing the work of JRI in 17 states).
118 See LAVIGNE ET AL., supra note 110, at 2.

ND SSRN.DOCX (DO NOT DELETE)

130

11/30/2015 11:20 AM

NOTRE DAME LAW REVIEW

[VOL. XX:N

risk and needs assessments. 119 States have also been encouraged to expand
their use of “problem-solving courts focuse[d] on arrestees with substance
abuse and mental health disorders,” to adopt “intermediate and graduated
sanctions [to] establish swift and certain responses, such as short jail stays,
for parole and probation technical violators,” to expand parole, “good time,”
and earned credit to shorten sentences and reward program participation and
compliance for those in prison and on community supervision, 120 to increase
community-based drug treatment; to reduce penalties for criminal offenses
(particularly mandatory minimum sentences), 121 and to require post-release
supervision for all prisoners. 122 Finally, JRI has promoted the development
of “[a]ccountability measures” for criminal justice agencies, such as “mandatory data reporting, annual reports of criminal justice performance
measures, and upgrades and integration of data.” 123
Although advocates of criminal justice reform—including proponents
of evidence-based practices—are often deeply concerned over the ways in
which mass incarceration has crippled communities and impaired the life
prospects of former criminals, the current language of both NCI and JRI emphasizes the financial benefits of reform over its moral ones.124 The reason
for this is both pragmatic and political. On a practical level, administrators
are easier to reach and educate than community members, and better positioned to make policy-level changes. Moreover, in an era of deep
partisanship, the virtue of frugality is one things on which politicians and the
119 See id. Uses for these instruments are many, and include “inform[ing] decisions
about detention, incarceration, and release conditions as well as the allocation of supervision
and treatment resources.” Id.
120 For a discussion of the perils of increasing opportunities for prison release through
the use of sentence credit, see Cecelia Klingele, The Early Demise of Early Release, 114 W.
VA. L. REV. 415, 446–50 (2012); Cecelia Klingele, Changing the Sentence Without Hiding
the Truth: Judicial Sentence Modification as a Promising Method of Early Release, 52 WM.
& MARY L. REV. 465, 488–91 (2010).
121 Some of the most promising criminal justice reforms are those aimed at decriminalizing minor conduct and reducing the inflation of maximum penalties that has occurred in
recent decades. Although such efforts are worthy of discussion, this Article focuses on laws
and penal practices that are focused on sentencing and the execution of sentences.
122 LAVIGNE ET AL., supra note 110, at 2–3.
123 Id. at 3.
124 In its promotional literature, for example, JRI advertises that “[j]ustice reinvestment
is a data-driven approach to improve public safety, reduce corrections and related criminal
justice spending, and reinvest savings in strategies that can decrease crime and reduce recidivism.” COUNCIL OF STATE GOV’TS, JUSTICE REINVESTMENT, http://csgjusticecenter.org/jr
(last visited Sept. 22, 2015). See also JRI One Pager, BUREAU OF JUSTICE. ASSISTANCE,
https://www.bja.gov/Programs/JRIonepager.pdf (last visited Sept. 22, 2015) (“Justice Reinvestment is a data-driven approach to reduce spending on corrections and reinvest identified
savings in evidence-based strategies designed to increase public safety and hold offenders
accountable.”).

ND SSRN.DOCX (DO NOT DELETE)

201X]

11/30/2015 11:20 AM

131

public can agree. Unlike arguments for change grounded in principles of
racial equality or proportionality, which have failed to carry the day in past
decades, reformers now want to bring about change by simply asking policymakers to follow the data and save money in the process. By framing
reform in pragmatic terms, proponents hope to bring people of different ideologies to the same table and, in doing so, open up possibilities for change
that seemed impossible only a decade earlier. 125
In many ways, that is exactly what has happened. Since its inception in
2002, well over half of U.S. states have received some form of assistance
from JRI, making it a national leader in the conversation about reducing mass
incarceration. 126 NIC has worked both directly and indirectly with many
more, through both its physical and online presence. Despite the influence
these agencies are having on the practices of state and local correctional
agencies, not all reformers have been comfortable with the approach being
taken by evidence-based proponents.
JRI, in particular, has come under attack for the way in which it has
framed its reform efforts. In 2013, a coalition of scholars and advocates,
many of whom strongly supported early JRI efforts, published a critique of
the goals and strategies being used in JRI’s work with the states. 127 The
authors of the report, titled Ending Mass Incarceration: Charting a New Justice Reinvestment, claimed that the Initiative had lost its moorings by failing
to use its political and financial leverage to mount an all-out attack on the
penal state. 128 For these critics, the purpose of reform is the dismantling of
mass incarceration and the build-up of impoverished neighborhoods through

125 The unanimous passage of recent legislation in Idaho illustrates this phenomenon.
Cf. OFF. OF GOV. C.L. “BUTCH” OTTER, 14:021, GOVERNOR OTTER SIGNS JUSTICE REINVESTBILL
(Mar.
19,
2014),
https://gov.
idaho.gov/mediacenMENT
ter/press/pr2014/3%20Mar/pr_021.html (quoting Idaho Rep. Rich Wills as saying, “‘[t]he
process enabled all sides to agree on the major drivers of growth in Idaho’s correction system. . . . . After we found consensus on the sources of the problem, the question turned to the
best way to increase safety and lower spending. The answer we came up with is the policy
framework codified in this bill.’”).
126 See AUSTIN ET. AL., supra note 11, at 1 n.1 (adding Alabama, Arizona, Connecticut,
Indiana, Michigan, Nevada, Rhode Island, Texas, Vermont, and Wisconsin to the list of states
that participated in precursors to JRI); LAVIGNE ET AL., supra note 5, at 55–124 (2014) (discussing JRI work in Arkansas, Delaware, Georgia, Hawaii, Kansas, Kentucky, Louisiana,
Missouri, New Hampshire, North Carolina, Ohio, Oklahoma, Oregon, Pennsylvania, South
Carolina, South Dakota, and West Virginia); BUREAU OF JUSTICE ASSISTANCE, JUSTICE REININITIATIVE: JRI SITES,
https://www.bja.gov/programs/justicereinvestVESTMENT
ment/jri_sites.html (last visited Sept. 23, 2015) (mapping 24 currently-involved states, including Idaho, Mississippi, Nebraska, Oklahoma, Oregon, Utah, and Washington).
127 See generally AUSTIN ET. AL., supra note 124.
128 See id. at 16.

ND SSRN.DOCX (DO NOT DELETE)

132

11/30/2015 11:20 AM

NOTRE DAME LAW REVIEW

[VOL. XX:N

a broader attack on policing, prosecution, and sentencing laws and practices—not the streamlining of correctional agencies. 129
Whatever the wisdom of its chosen strategy, it is difficult to overstate
the influence that JRI, NIC, and similar state and locally-initiated efforts
have on the spread of evidence-based correctional practices. 130 Many states
have now passed legislation requiring judges and correctional agencies to
adopt specific evidence-based correctional practices—risk assessment, in
particular. Several states have passed legislation that now requires judges be
provided with risk assessment and recidivism data at sentencing, 131 and
many more have passed laws that require the use of risk and needs assessments by correctional agencies. 132 Incorporating the principle that individuals at low risk of re-offense should not be subject to significant intervention, 133 some new laws require that medium- and high-risk individuals be
given priority in gaining access to correctional treatment programs, 134 while
other legislation, drawing on psychological findings that suggest rewards are
more motivating than punishment, requires courts and correctional agencies
to create incentives for individuals to successfully complete their sentence
requirements. 135
As efforts to hasten the spread of evidence-based correctional practices
have accelerated, the goals of doing so have undeniably become more openly

See id. at 17–19.
In recent years, the Federal Bureau of Prisons and U.S. Office of Probation and Pretrial Services have undergone a similar evidence-based transformation, adopting standardized
risk assessment tools (and recalibrating supervision and services according to their results),
and instituting data-gathering requirements and outcome assessments for many contracted
programs. CITE.
131 See, e.g., ARIZ. REV. STAT. ANN. § 6-201.01(J)(2) (2014); IDAHO CODE § 19-2517
(2014); LA. STAT. ANN. § 15:326(A) (2014); 42 PA. CONS. STAT. § 2154.7 (2010).
132 See, e.g., DEL. CODE ANN. tit. 11, § 4321(b)(2) (2012); 730 ILL. COMP. STAT. 190/10
(2010); KY. REV. STAT. ANN. § 3(a) (West 2011); OHIO REV. CODE ANN. § 2301.30(D)(1)
(West 2011); OKLA. STAT. ANN. tit. 22, § 988.17 (West 1999).
133 Studies have found that “low risk” individuals who are required to engage in significant formal interventions (such as treatment programs or frequent visits with their probation
or parole agent) have higher rates of recidivism than those who are more or less left alone.
The reasons for this difference are not entirely clear, but likely involve a combination of the
fact that those being supervised more get caught more often, and the negative effects of workforce disruption and the poor social influences that occur when lower risk individuals are
required to attend programs with higher risk individuals. Cf. CHRISTOPHER T. LOWENKAMP &
EDWARD J. LATESSA, UNDERSTANDING THE RISK PRINCIPLE: HOW AND WHY CORRECTIONAL
INTERVENTIONS CAN HARM LOW-RISK OFFENDERS, in NAT’L INST. OF CORRECTIONS, TOPICS
IN COMMUNITY CORRECTIONS 3, 3–8 (2004).
134 See, e.g., 2012 Ga. Laws 902 (limiting drug court participation to medium and high
risk defendants); IDAHO CODE § 19-2524(2)(d) (2014).
135 See, e.g., IDAHO CODE § 19-2601(5) (2014).
129
130

ND SSRN.DOCX (DO NOT DELETE)

11/30/2015 11:20 AM

201X]

133

modest. Cost containment and population stabilization outweigh decarceration and neighborhood revitalization as the focus of technical assistance and
advocacy efforts. 136 Nevertheless, in a short time, through the efforts of organizations like NIC and JRI, the language and tools of correctional practice
have rapidly changed, with agencies across the country openly embracing
the goal of reducing prison populations, with the implementation of evidence-based correctional practices as their primary strategy for doing so.
III.

COMPETING PARADIGMS

To date, efforts to persuade states to adopt evidence-based practices
have been much more technical than theoretical. This omission is intentional: policymakers from across the political spectrum have adopted evidence-based correctional practices because they promise financial savings,
increased efficiency, and “scientifically-proven” results—not necessarily because they believe current correctional practices are morally unjustified. To
obtain buy-in from practitioners and policymakers, reformers have directed
their resources to gathering new data, providing up-to-date, reliable assessments of existing correctional programs, and developing new programs that
draw on the findings of the limited available social science research. As the
popularity of evidence-based practices demonstrates, if implementation of
these new practices is the metric of success, then their efforts have succeeded. The problem is that the cost of maintaining buy-in from a broad
range of policymakers has been neglect of a deeper conversation about the
goals of the correctional system, and the uses to which new evidence-based
tools will be put.
These details matter. Scholars have written at length about the values—
from punishing the guilty to entrenching white privilege—that have been
used to justify and enable the growth of the penal state in the second half of
the 20th Century. 137 Fewer have explored the values that might justify reintroducing rehabilitation through evidence-based practices as a legitimate
mechanism for reversing what many now consider to be the overly-harsh
consequences of the “punishment imperative” 138 that dominated late 20th
Century sentencing and correctional practices.

136 See AUSTIN ET AL., supra note 11, at 9–11. In its work with the states, JRI recommendations now propose that only a small portion of the money saved from decarceration be
earmarked for crime reduction, with most—if not all—the rest used to build up formal community corrections and community-based treatment programs. See id.
137 See, e.g., GOTTSCHALK, supra note 114; MICHELLE ALEXANDER, THE NEW JIM CROW:
MASS INCARCERATION IN THE AGE OF COLORBLINDNESS (2010); Norval Morris, The Future of
Imprisonment: Toward a Punitive Philosophy, 72 MICH. L. REV. 1161 (1974).
138 See TODD R. CLEAR & NATASHA A. FROST, THE PUNISHMENT IMPERATIVE: THE RISE
AND FAILURE OF MASS INCARCERATION IN AMERICA (2013).

ND SSRN.DOCX (DO NOT DELETE)

134

11/30/2015 11:20 AM

NOTRE DAME LAW REVIEW

[VOL. XX:N

Being clear about the purposes for which evidence-based practices
should be used is essential to avoiding their abuse. Much like the treatment
programs and other interventions that comprised America’s attempts at correctional rehabilitation in the first half of the 20th Century, evidence-based
correctional practices can take many forms and be used in many ways, not
all of which are benign. 139 Without a clear normative framework to guide
the use of these new tools, there is a danger that could be used to further the
scope and scale of the penal state, rather than to reduce its reach and soften
its impact. To better understand why this is so, it is helpful to examine in
some detail the competing ways in which support for evidence-based correctional practices can be understood.
A. Neorehabilitation and its Goals
The first, and superficially most obvious, way to view evidence-based
correctional practices is as a form of neorehabilitationism. Neorehabilitationism seeks to reintroduce, or rehabilitate if you will, rehabilitation as an
animating principle of criminal justice. 140 Its advocates press for programs
and interventions that are designed to help justice-involved individuals attain
stability and autonomy, and end involvement in crime and the criminal justice system.
Neorehabilitationism itself has several distinctive strains, which sometimes operate in unacknowledged tension with one another. The first is humanitarian. It is rooted deeply in the normative belief that the punitive practices that have characterized modern penal practice are de-humanizing and
unjust. This strain of neorehabilitationism sees rehabilitative interventions
as affirming the dignity of justice-involved individuals—and in doing so,
affirming the legitimacy of the state.141 Advocates of this form of neorehabilitationism argue that trying to strip the penal system of rehabilitative aspirations was a recipe for disaster, for unless the justice system embraces the
goal of improving the lot of those subject to punishment, nothing tempers
the retributive impulses that run high in human nature. 142 (The growth in
139 Cf. Michael Tonry, Legal and Ethical Issues in the Prediction of Recidivism, 26 FED.
SENT’G REP. 167, 167 (2014) (observing that there is “a collective amnesia about what was
learned about the use of prediction in the 1970s when widespread support for indeterminate
sentencing collapsed. Basing decisions about individuals’ liberty and autonomy on calculations of risk raises fundamental normative and ethical issues that were once taken seriously
but are no longer often acknowledged or discussed.”).
140 See Jessica M. Eaglin, Against Neorehabilitation, 66 SMU L. REV. 189, 193 (2013);
Melissa Hamilton, Risk-Needs Assessment: Constitutional and Ethical Challenges, 52 AM.
CRIM. L. REV. 231, 233 (2015).
141 See generally FRANCIS T. CULLEN & KAREN E. GILBERT, REAFFIRMING REHABILITATION (1982).
142 See id. at 257.

ND SSRN.DOCX (DO NOT DELETE)

201X]

11/30/2015 11:20 AM

135

imprisonment that followed the publication of their work strengthens this
thesis.) In this view, rehabilitative programs and interventions signal that
everyone is capable of betterment, given the right opportunities and circumstances.
From this perspective, rehabilitative efforts are desirable even if they
are not tremendously successful in terms of lasting treatment effects, because
they demonstrate independent respect for the humanity of those in the system
and acknowledge the disadvantage that often defines their existence. 143 This
theme has been echoed not only by some proponents of evidence-based practices, but also by advocates of restorative justice, drug treatment, and specialized courts. 144 Humanitarian neorehabilitationists emphasize the way in
which rehabilitation as a philosophy softens what is an otherwise harsh and
unforgiving justice system that disproportionately punishes the poor, the deficient, and the abused.
A second strain of neorehabilitationism is more scientific than humanitarian. Scientific neorehabilitionism is primarily focused on the question of
how to stop criminal behavior at the individual level. It seeks to identify the
mechanisms by which behavioral change can be effectively manipulated
through formal intervention, to embed those mechanisms in formal correctional programs, and to assist agencies in replicating effective programs so
that justice-involved individuals can be given appropriate treatment for their
perceived deficiencies. Simply put, this form of neorehabilitationism is
more focused on the how of rehabilitation than on the why.
At its core, scientific neorehabilitationism embraces much the same
principles that animated early 20th Century rehabilitionionism, including a
belief that science can identify and “cure” many of the problems associated
with criminal offending, whether through medical, cognitive, or social interventions. 145 Ever since Martinson’s famous study left politicians claiming
that “nothing worked” to rehabilitate prisoners, criminologists have worked
to build a case that Martinson was wrong. Beginning with Ted Palmer’s reexamination of Martinson’s own data (which debunked Martinson’s conclusions), 146 rehabilitation apologists have built up a small but sound body of
work demonstrating that some correctional interventions do have aggregate
143 See, e.g., U.S. Cath. Conf., Responsibility, Rehabilitation, and Restoration: A Catholic Perspective on Crime and Criminal Justice 21–25 (2000); Francis T. Cullen, Rehabilitation: Beyond Nothing Works, 42 Crime & Just. 299 (2013).
144 See, e.g., Michael L. Perlin, 'The Judge, He Cast His Robe Aside': Mental Health
Courts, Dignity and Due Process, 3 J. MENTAL HEALTH L. & POL’Y 1, 21 (2013); Amanda
Ploch, Note, Why Dignity Matters: Dignity and the Right (or Not) to Rehabilitation from International and National Perspectives, 44 N.Y.U. J. INT’L L. & POL’Y 887, 894-95 (2012)
145
For a discussion of the ways in which scientific neorehabilitationism has seen a resurgence in the sentencing context, see generally Meghan J. Ryan, Science and the New Rehabilitation, 3 VA. J. CRIM. L. 261 (2015).
146 Ted Palmer, Martinson Revisited, 12 J. RES. CRIME & DELINQ. 133 (1975).

ND SSRN.DOCX (DO NOT DELETE)

136

11/30/2015 11:20 AM

NOTRE DAME LAW REVIEW

[VOL. XX:N

modest-to-significant success at reducing criminal and antisocial behaviors
for individuals with particular characteristics, such as substance abuse problems or mental health disorders. 147
Scientific neorehabilitationism provides the “evidence” behind evidence-based practices. Its proponents are committed to identifying effective
interventions for reducing criminal offending, and to implementing them in
the field. Representative of the scientific approach to rehabilitation is Roger
Warren, former director of the National Center for State Courts, and author
of several leading manuals and articles on evidence-based practice provides
a good example of this approach to rehabilitation: 148
[T]he optimism that characterized the “rehabilitative ideal” during the
first three quarters of the twentieth century was obviously naïve. We did
not then possess the practical knowledge or tools to be able to effect
meaningful change in offender behavior. Today, however, unlike thirty
years ago, we know—based on meticulous meta-analyses of rigorously
conducted scientific research—that unlike incarceration the right kinds
of rehabilitation and treatment programs carefully targeted at specific
crime-related risk factors among medium- to high-risk offenders can reduce offender recidivism by conservative estimates of 10 to 20 per149
cent.

Although scientific neorehabilitationists may be (and often are) animated by humanitarian concerns, they need not be—and it is this point that
is often unappreciated by opponents of mass incarceration. Mary Fan has
suggested that unlike classical rehabilitationism, neorehabilitationism is fundamentally pragmatic. She writes:
We are well past the time of starry-eyed and egalitarian hope for the redemption of all. The rationale of rehabilitation is being redefined away
from the interest of the prisoner in redemption, an ideal that has lost its
political and moral power to stitch together a broad-based social consensus because of fractures in worldviews of what we should value normatively. Instead rehabilitative pragmatism is centered on the public inter-

147 See, e.g., Cullen & Gilbert, supra note 139, at 170; Francis T. Cullen, The Twelve
People Who Saved Rehabilitation: How the Science of Criminology Made a Difference, 43
CRIMINOLOGY 1, 3 (2005); WHAT WORKS: REDUCING REOFFENDING: GUIDELINES FROM RESEARCH AND PRACTICE (JAMES MCGUIRE, ed. 1995); Palmer, supra note 143, at 142.
148 See generally Roger K. Warren, Evidence-Based Sentencing: The Application of
Principles of Evidence-Based Practice to State Sentencing Practice and Policy, 43 U.S.F. L.
REV. 585 (2009); ROGER K. WARREN, U.S. DEP’T OF JUSTICE., EVIDENCE-BASED PRACTICE TO
REDUCE RECIDIVISM: IMPLICATIONS FOR STATE JUDICIARIES (2008), http://works.bepress.com/cgi/viewcontent.cgi?article=1000&context=roger_warren.
149 Roger K. Warren, The Most Promising Way Forward: Incorporating Evidence-Based
Practice into State Sentencing and Corrections Policies, 20 FED. SENT’G REP. 322, 323
(2008).

ND SSRN.DOCX (DO NOT DELETE)

11/30/2015 11:20 AM

201X]

137

est in safety and reducing costs in the most cost-effective manner. Rehabilitation pragmatism is cautious and selective, with a greater reliance on
scientific data in selecting participants who are more apt to succeed and
most in need of intervention in a system that must practice triage because
150
of chronic overload.

While admitting that this view of rehabilitation is “harder-edged” than
a more humanitarian version of rehabilitation, the “pragmatic”—what I call
“scientific”—approach is to be favored because, Fan asserts, “[a]n emphasis
on evidence of efficacy is a more widely appealing idiom in a time of ascendant scientism that has supplanted normative, moral, or religious ideals
that formerly helped give the rehabilitative ideal added appeal.” 151
From this perspective, evidence-based interventions are not intended to
“cure” criminality for the sake of the offender, but for the sake of the public
benefit that intervention yields. 152 Much like the quarantined treatment that
contains a tuberculosis outbreak, the treatment is administered for “us,” not
for “them.” In this articulation of neorehabilitationism, we hear echoes of
earlier debates on the methods and purposes of rehabilitative intervention.
The criticisms that led to the collapse of the Rehabilitative Ideal, Version 1.0 153 did not turn merely on the scientific reliability of the correctional
programs in use during the early 20th Century, but also on the uses to which
those programmatic interventions were being put. While neorehabilitationists have responded to criticisms about the effectiveness of rehabilitation by
promoting the adoption of evidence-based practices, they have largely overlooked critiques about the ways in which rehabilitative practices can be used
to manipulate, marginalize, and harm those it purports to cure. They ignore
these concerns at their peril.
Already, new voices and seasoned ones are beginning to challenge neorehabilitationism, and their complaints echo old refrains. Although some

150 Mary D. Fan, Beyond Budget-Cut Criminal Justice: The Future of Penal Law, 90
N.C. L. REV. 581, 637 (2012) (footnote omitted).
151 Id. at 637–38.
152
Meghan Ryan has characterized the difference between the traditional Rehabilitative
Ideal and the “New Rehabilitation” as a shift in focus from one centered on offender character to one centered on offender behavior: “Whereas early rehabilitative efforts focused on
removing the offender from his corrupt surroundings and treating his character through religious and vocational training, modern understandings of rehabilitation focus on the offender's behavior by placing primary importance on the offender's reintegration into society. . . .
Although modern commentators may refer to character change, it is most often with the aim
of improving society through offender reintegration. This notion is emphasized through
commentators’ primary method of determining whether rehabilitation has been achieved:
recidivism. This measures offenders’ effects on society rather than necessarily measuring
any change within the offenders themselves.” Meghan J. Ryan, Science and the New Rehabilitation, 3 VA. J. CRIM. L. 261, 327-28 (2015) (emphasis added).
153 See supra Part I.A.

ND SSRN.DOCX (DO NOT DELETE)

138

11/30/2015 11:20 AM

NOTRE DAME LAW REVIEW

[VOL. XX:N

criticism has centered on the reliability of evidence-based practices (risk assessment tools and specialty courts, in particular), 154 a separate strain renews
concerns about the potential for superficially benign interventions to be used
in abusive ways. Scholars like Jessica Eaglin and Michael Tonry have derided the control-focused tone of scientific rehabilitationism, 155 implying
that it is just a new iteration of old models of social control. 156 Others, such
as Sonja Starr and Bernard Harcourt, have raised concerns about the disparate racial impacts of evidence-based tools like risk assessments. 157 Their
allegations merit further consideration.
B. The New Penology and Evidence-Based Tools
Understanding the ways in which practices intended to reduce the use
of incarceration might inadvertently reinforce the size and scope of the penal
state requires an examination of how the new penology has functioned in the
modern era, and why reformers today are vulnerable to overlooking how evidence-based practices might be co-opted by the bureaucratic needs of the
criminal justice system to reinforce state power in ways that are far from
benign.
As discussed above in Part I.A., the “new penology” is a way of understanding how criminal justice system actors approach and carry out their
work. It is distinguishable from both rehabilitation and retribution, and is

154 See, e.g BERNARD E. HARCOURT, AGAINST PREDICTION: PROFILING, POLICING, AND
PUNISHING IN AN ACTUARIAL AGE 239 (2007); Eric J. Miller, Drugs, Courts, and the New
Penology, 20 STAN. L. & POL’Y REV. 417 (2009).
155 In the views of these skeptics, the reform framework developed by JRI is an intentional effort by criminal justice stakeholders with an investment in the status quo to appear
humane and solicitous without undermining the paradigm of control that enables the differential punishment of people according to race and class. See, e.g., Gerald P. López, How
Mainstream Reformers Design Ambitious Reentry Programs Doomed to Fail and Destined to
Reinforce Targeted Mass Incarceration and Social Control, 11 HASTINGS RACE & POVERTY
L.J. 1, 71 (2014) (alleging that critics who dare to challenge the reigning reforms face retaliation in the form of denied grant applications and ostracization from mainstream scholarship
and professional opportunities).
156 See, e.g., Eaglin, supra note 138, at 222. (“The limitations of the neorehabilitative
model are inherent because this particular form of rehabilitation, over-emphasizing evidencebased programming and predictive tools, has its origin in the same theory that created total
incapacitation.”). See also López, supra note 151, at 101 (“I am among those who consider
the prevailing approach to criminal justice—targeted mass incarceration and social control—
wrong. . . . And I am among an apparently much smaller group of people who consider the
vision of reentry articulated by the Reentry Policy Council likely doomed by its own inability
or unwillingness to expose these biases.”).
157 Sonja B. Starr, Evidence-Based Sentencing and the Scientific Rationalization of Discrimination, 66 STAN. L. REV. 803, 821 (2014); Harcourt, supra note 150, at 41.

ND SSRN.DOCX (DO NOT DELETE)

201X]

11/30/2015 11:20 AM

139

characterized by its focus not on punishment, but on “identifying and managing unruly groups.” 158 In the new penology the efficient administration of
the criminal justice system takes priority over the propriety of any person’s
behavior: “Its goal is not to eliminate crime but to make it tolerable through
systemic coordination.” 159
In the new penology, those under state control are neither hated nor
cared for—they are simply managed. Through aggressive monitoring followed by prosecution of petty and nonviolent offenses, tools of the new penology have kept troublesome people (often young men, especially young
men of color) on a short leash. Rules of community supervision (curfews,
drug testing, and reporting requirements especially) and custodial sentences
allow system actors to incapacitate those deemed “risky” without regard for
the quantum of punishment they deserve. The result is the creation of an
underclass of invisible people—managed as “waste” and unworthy of individualized consideration. 160
The detachment that characterizes the new penology is both a cause and
function of the volume of people within the criminal justice system. Faced
with crushing caseload pressures at every stage of the criminal justice process, system actors have institutionalized practices and structures that allow
them to track and manage large numbers of people efficiently. 161 Even when
these actors view themselves as connected to their work and the people under
their supervision, 162 the tools they use for allocating their limited time and
resources are often quite impersonal. 163 Some scholars have suggested that
“the new penology” is not an intentionally malignant effort to control the
poor, but rather that “managerialism is a phenomenon that is largely explicable in terms of the dynamics of organizational growth and the new possibilities for control generated by advances in information technology” have

Feeley & Simon, supra note 33, at 455.
Id.
See id. at 469–70.
Forty states reported an aggregate total of more than 15 million pending criminal
cases in 2013. See Dataviewer, COURT STATISTICS PROJECT (R. LaFountain, et al., eds. 2015),
www.courtstatistics.org (last updated Feb. 12, 2015).
162 See generally Mona Lynch, Waste Managers? The New Penology, Crime Fighting,
and Parole Agent Identity, 32 L. & SOC’Y REV. 839 (1998).
163 In many places, for example, risk assessment results dictate contact hours, program
assignments, and standardized conditions of supervision. Deterrence-based correctional programs allow courts to rapidly process individuals who violate their supervision conditions by
imposing pre-ordained graduated sanctions—with a kind word, perhaps, but without the need
for time and resource intensive consideration of the individual circumstances. Cf. MARK A.
R. KLEIMAN, WHEN BRUTE FORCE FAILS: HOW TO HAVE LESS CRIME AND LESS PUNISHMENT
41 (2009) (“Since H.O.P.E. [a deterrence-based correctional program] is much less expensive
and much less time-consuming for the judge and the judge’s staff, it can—where drug courts
cannot—be expanded to mass scale. . . .”).
158
159
160
161

ND SSRN.DOCX (DO NOT DELETE)

140

11/30/2015 11:20 AM

NOTRE DAME LAW REVIEW

[VOL. XX:N

allowed. 164 Whether intended or not, the tools of the new penology, from
hot spot policing to GPS tracking, have undeniably imposed significant restraints on those subject to them. 165 Given that fact, it is worth asking
whether evidence-based practices might also be used as tools of less-thanbenign social control.
At first blush, the evidence-based practices being promoted by neorehabilitationists seem clearly distinguishable from elements of the new penology described by Feeley and Simon. After all, the fundamental feature of the
new penology is its lack of concern for individuals, and evidence-based practices seem to require the very opposite. While both the new penology and
neorehabilitationism favor the use of predictive risk instruments, neorehabilitation advocates promote the use of tools like risk assessments for the purpose of tailoring interventions to match specific individuals’ identified criminogenic needs. 166 The new penology, by contrast, uses risk prediction solely
for the purpose of channeling high risk offenders into more secure forms of
incapacitation, without regard for individual characteristics. Evidence-based
risk and needs instruments at least nominally rely not only on static factors
such as age at first arrest and criminal history (which are predictive but unchangeable), but also on dynamic factors, such as employment and educational status, social influences, and level of community engagement—all of
which are individualized and potentially responsive to correctional intervention. 167 Beyond risk assessment, other evidence-based correctional practices
are even more clearly focused on individual characteristics: motivational interviewing techniques work to build a relationship of trust between the su-

164 Jonathan Simon & Malcom M. Feeley, The Form and Limits of the New Penology, in
PUNISHMENT AND SOCIAL CONTROL 75, 76 (Thomas G. Blomberg & Stanley Cohen, eds., enlarged 2d ed. 2003) (quoting David Garland, Penal Modernism and Postmodernism, in PUNISHMENT AND SOCIAL CONTROL 181, 201 (Thomas G. Blomberg & Stanley Cohen, eds., 1st
ed. 1995)).
165 See, e.g., Kimberlé W. Crenshaw, From Private Violence to Mass Incarceration:
Thinking Intersectionally About Women, Race, and Social Control, 59 UCLA L. REV. 1418,
1441–43 (2012) (discussing intrusiveness of state surveillance and intervention in the space
and lives of black mothers); Brett G. Stoudt, et. al., Growing Up Policed in the Age of Aggressive Policing Policies, 56 N.Y.L. SCH. L. REV. 1331 (2011/2012) (discussing experiences
of minority youth in heavily surveilled areas of New York City).
166 See Christopher T. Lowencamp & Edward J. Latessa, Understanding the Risk Principle: How and Why Correctional Interventions can Harm Low-Risk Offenders, in TOPICS IN
COMMUNITY CORRECTIONS 3, 3 (U.S. Dep’t of Justice, Nat’l Inst. of Corr. 2004).
167 John Monahan & Jennifer L. Skeem, Risk Redux: The Resurgence of Risk Assessment
in Criminal Sanctioning, 26 FED. SENT’G REP. 158, 160-61 (2014).

ND SSRN.DOCX (DO NOT DELETE)

201X]

11/30/2015 11:20 AM

141

pervising agent and probationer in order to improve the probationer’s commitment to change, 168 and specialty courts often provide personalized responses to relapses and other violations. 169
At the same time, a thin line separates the humanitarian neorehabilitationist’s use of aggregate statistical information to help select an appropriate
treatment program for a drug-addicted probationer from the neopenologist’s
reliance on the same aggregate information for the purpose of disinterested
control. In many ways, the very term rehabilitation, with its connotations of
concern for the welfare of the marginalized, provides a dangerous veneer
that makes observers less keen to possible abuses of “rehabilitative” tools.
In 1970, the American Friends Service Committee in 1970 warned:
“[R]ehabilitation has introduced a new form of brutality, more subtle and
elusive. That rehabilitation is less disturbing to the deliverers who, consequently, have spread it among a much larger number of persons is also
true.” 170 There is danger that the same criticism might one day be leveled
against the evidence-based practices neorehabilitationists are now promulgating as a solution to the problem of mass incarceration. These dangers
divide into three categories: the danger of forgetting the past, of overselling
the present, and of misidentifying the purpose of the correctional enterprise.
1. The Danger of Forgetting
Memory fades quickly, and for most system actors extends only as far
back as their own training. As a result, it is easy to forget that some of the
scientific tools in which neorehabilitationists place so much stock are not
that far removed from the now-discredited science that rehabilitationists
promulgated less than a century ago. 171 While new techniques of risk prediction may look more sophisticated than last century’s phrenology, the truth
is that our ability to predict future human behavior remains mightily

BRADFORD BOGUE & ANJALI NANDI, U.S. DEP’T OF JUSTICE, MOTIVATIONAL INTER(2012).
169 Eric Miller has previously connected drug courts—a subset of the many types of specialty courts that now seek to address specific classes of individuals and criminal cases—to
both neorehabilitation and the new penology. He explains: “[d]rug courts represent a combination of the managerial and responsibilization aspects of the adaptive strategy, while maintaining the old penology emphasis on individualization and rehabilitation . . . . The success
of the drug court has been to rework the old penology of intervention and treatment into what
might be called ‘neorehabilitation,’ using supervision and incapacitation as a form of risk
management to train individuals as responsible members of society or send the incorrigible to
jail or prison.” Miller, supra note 150, at 440–41.
170 AM. FRIENDS SERV. COMM., supra note 23, at 96.
171 See Tonry, supra note 137, at 167.
168

VIEWING IN CORRECTIONS 3

ND SSRN.DOCX (DO NOT DELETE)

142

11/30/2015 11:20 AM

NOTRE DAME LAW REVIEW

[VOL. XX:N

flawed. 172 Scientific inquiry should of course play a role in the justice system, helping us learn how to motivate lasting change and treat neurological
deficits and psychological illnesses that can lead to criminal behavior. At
the same time, advocates should approach the task of implementing new
practices with a generous dose of humility, realizing that similar attempts
have been made before, with every bit as much certainty in the state of scientific knowledge and with almost uniformly unsatisfactory results. 173 In
promoting evidence-based practices, it is essential that advocates of evidence-based tools remember and teach the lessons of history: that past certainty has often been misplaced and that the “help” offered by the criminal
justice system has often been used to harm individuals and communities in
significant and lasting ways. 174
2. The Danger of Overselling
Closely related to the danger of forgetting the past is the danger of overselling the present state of knowledge. Advocates, caught up in the force of
their own rhetoric and eager to take advantage of shifting sensibilities about
punishment, have sometimes gone too far in describing the power of evidence-based practices to revolutionize the criminal justice system at this moment in time. In the main, they have been reticent to acknowledge the paucity of reliable evidence that now exists, 175 and the limits of the interventions
about which we do possess evidence. Unless criminal justice system actors
are made fully aware of the limits of the tools they are being asked to implement, they are likely to misuse them.
Once again, risk assessment tools provide a good example of evidencebased practices that have been promulgated with insufficient attention to
their limitations. Most risk instruments in widespread use today have been
See Starr, supra note 153, at 842; HARCOURT, supra note 150, at 2–3.
Microscopic hair analysis, odontology, arson blaze patterns, and even Shaken Baby
Syndrome are among the once-popular techniques for gathering and analyzing evidence that
have been discredited or called into serious question over the past twenty years. See Mary A.
Bush et al., Statistical Evidence for the Similarity of the Human Dentition, 56 J. FORENSIC SCI.
118 (2011); Brandon L. Garrett & Peter J. Neufeld, Invalid Forensic Science Testimony and
Wrongful Convictions, 95 VA. L. REV. 1 (2009); Cassandra Ann Jenecke, Note, Shaken Baby
Syndrome, Wrongful Convictions, and the Dangers of Aversion to Changing Science in Criminal Law, 48 U.S.F. L. REV. 147 (2013).
174 See Part I.A., supra. Cf. United States v. McLaurin, 731 F.3d 258, 259, 262 (2d Cir.
2013) (holding that requirement that sex offender “participate in an approved program of sex
offender evaluation and treatment, which may include . , , plethysmograph examinations” violated substantive due process on the ground that “‘[t]here is a line at which the government
must stop. Penile plethysmography testing crosses it.’” Id. at 259, 262 (quoting Judgment at
4, United States v. McLaurin, No. 2:11-cr-00013-wks (D.Vt. Aug. 22, 2012); United States
v. Weber, 451 F.3d 552, 571 (9th Cir. 2006)).
175 See Part II.B, supra.
172
173

ND SSRN.DOCX (DO NOT DELETE)

201X]

11/30/2015 11:20 AM

143

subjected to scientific validation and have been found to be more accurate at
predicting “risk” than clinical judgment alone. 176 Even so, these instruments
are far from failsafe. As an initial matter, risk is a squishy concept and its
variations (low, medium, and high) are subject to all manner of manipulation. The risk prediction instruments in use today can predict a person’s
statistical risk of re-arrest and re-conviction across the general population,
but few tools differentiate carefully between the specific kinds of conduct
for which a person is at risk of being caught. 177 Moreover, to retain their
accuracy, risk instruments must be constantly re-normed for changing populations and sub-populations. As a result, a prediction may at any given time
be more or less accurate with respect to any particular individual.178 Not
infrequently, advocates of evidence-based practice have pressed correctional
agencies to adopt risk instruments without first ensuring they have built capacity for maintaining those instruments or providing adequate warning to
local system actors about the need for constant monitoring of their continued
accuracy. 179
And accuracy is no small matter. Reliably predicting future human behavior is impossible in any individual case, and remains challenging even
when assessing aggregate risk. Although the predictive value of actuarial risk
assessment instruments has improved over time, commenters have noted the
difficulty of predicting with any degree of helpful reliability the risk that an
individual being supervised by the state will engage in the type of criminal
176 See, e.g., Daniel J. Neller & Richard I. Frederick, Classification Accuracy of Actuarial Risk Assessment Instruments, 31 BEHAV. SCI. & L. 141, 141 (2013); John Monahan, A
Jurisprudence of Risk Assessment: Forecasting Harm Among Prisoners, Predators, and Patients, 92 VA. L. REV. 391, 408, 427 (2006) (affirming the superiority of risk tools to clinical
judgment but suggesting their uses should be limited in criminal justice contexts). But see
Starr, supra note 153 (questioning the accuracy of risk tools at predicting individual behavior).
177 Being at high risk of a bar fight may be less serious than being at low (but not no)
risk of murdering. Modern tools, however, are notoriously incapable of making such distinctions. (The one notable exception is the category of sex offender risk assessment instruments,
which are ordinarily focused solely on the risk of sexual re-offense.)
178 See BAIRD, supra note 94, at 3–6 (2009); James Austin, How Much Risk Can We
Take? The Misuse of Risk Assessment in Corrections, 70 FED. PROBATION 58, 59 (2006).
179 Without the internal capacity to ensure long-term reliability, criminal justice agencies
are left to either continue using outdated instruments or pay outside research agencies to supply them with regularly updated instruments, often normed against national populations.
Texas is a good example of this choice. Until 2015, the state used the same unaltered risk
instrument, whose reliability was highly suspect and to which practitioners in the field gave
little credence. In 2015, it adopted a new instrument with assistance from researchers at the
University of Cincinnati and Sam Houston State University. See The Texas Risk Assessment
System: A New Direction In Supervision Planning, 22 CRIM. JUST. CONNECTIONS 1, 1-2
(2015),
https://www.tdcj.state.tx.us/connections/JanFeb2015/Images/JanFeb2015_agency_TRASS.pdf.

ND SSRN.DOCX (DO NOT DELETE)

144

11/30/2015 11:20 AM

NOTRE DAME LAW REVIEW

[VOL. XX:N

behavior that would justify greater intrusions on liberty. 180 Moreover, even
when an instrument is accurate in its statistical predictions, other concerns
may outweigh the utility of such information. A significant body of literature
has found that risk assessment tools disproportionately classify minorities
and the poor as higher risk, often due to factors outside their control, such as
familial background and education, potentially subjecting them to harsher
treatment throughout the penal system. 181 As a result, reliance on risk assessment tools at sentencing and in correctional decision-making remains
highly controversial as a normative matter.182 Furthermore, practitioners frequently misunderstand their proper uses, using them as ways to predict future
behavior with scientific certainty, rather than as tools for better understanding specific individuals and their propensities. 183 Despite knowing these limitations, reformers have pressed for the use of risk assessments throughout
the sentencing and correctional systems, not only as tools to augment clinical
judgment, but often as a substitute for it. 184 By overselling the accuracy and
utility of risk assessment tools, reformers risk contributing to their misuse in
ways that run counter to “evidence” on the limits of these tools.
Being honest about the limits of our knowledge about evidence-based
practice can be difficult, especially for reformers wishing make a quick and
noticeable impact on sentencing and correctional practices. While this impulse is understandable, lasting institutional change is usually slow and
measured for good reason. Unless advocates of new reforms take greater
care to be honest about the limits of current knowledge in the field, they risk

180 See generally Melissa Hamilton, Public Safety, Individual Liberty, and Suspect Science: Future Dangerousness Assessments and Sex Offender Laws, 83 TEMP. L. REV. 697
(2011); CHRISTOPHER BAIRD, A QUESTION OF EVIDENCE: A CRITIQUE OF RISK ASSESSMENT
MODELS USED IN THE JUSTICE SYSTEM, NAT’L COUNCIL ON CRIME & DELINQUENCY 7 (2009),
available at http://faculty.uml.edu/jbyrne/44.203/NCCD%20Baird%20on%20Risk.pdf
(“Nearly all of the literature on popular risk models refers to their demonstrated validity and
reliability. In actuality, there is little information available that supports model reliability,
and much of what is available either addresses the wrong issue (internal consistency) or provides inadequate tests of inter-rater reliability”); Richard Rogers, The Uncritical Acceptance
of Risk Assessment in Forensic Practice, 24 L. & Hum. Behavior 595 (2000).
181 See, e.g., Kelly Hannah-Moffat, Actuarial Sentencing: An “Unsettled” Proposition,
30 JUST. Q. 270, 283 (2013).
182 See generally Melissa Hamilton, Risk-Needs Assessment: Constitutional and Ethical
Challenges, 52 AM. CRIM. L. REV. 231 (2015); see also Starr, supra note 153, at 870–71;
BAIRD, supra note 174, at 3–6; Bernard E. Harcourt, Risk as a Proxy for Race 2 (John M. Olin
Law & Economics, Working Paper No. 535 & Public Law and Legal Theory Working Paper,
No. 323, 2010), http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1677654.
183 Austin, supra note 174, at 59 (2006); Hamilton, supra note 180, at 753–54 (describing cases in which the result of risk assessments have been misinterpreted by courts and even
by expert witnesses).
184 See, e.g., GA. ACT 709 § 2-1 (2012) (denying certain kinds of drug treatment to individuals with low risk scores regardless of personal needs and characteristics).

ND SSRN.DOCX (DO NOT DELETE)

201X]

11/30/2015 11:20 AM

145

enabling abuse of the tools they are promoting, and losing credibility when
the limits of those tools are eventually discovered.
3. The Danger of Misframing
The final danger ties back to the strategy by which evidence-based practices have been so rapidly implemented, and that is advocates’ willingness
to emphasize scientific reliability over moral desirability, thereby circumventing potential obstacles created by normative disagreements among policymakers. 185 While many practices supported by data are also morally desirable, there is a danger in justifying the use of particular evidence-based
practices by reference to efficacy alone. As an example, consider the issue
of procedural justice.
A large body of sociological research supports the intuitive proposition
that individuals view criminal justice officials as more legitimate, and are
more likely to comply with their directives, when those officials act in ways
that demonstrate respect and impartiality. 186 Similarly, research has shown
that when people receive praise for their successful progress they become
more motivated to change than when they are merely sanctioned or reprimanded for their failures. 187 While it is true that these findings have been
confirmed by experimental research, framing these behaviors as “correctional practices” rather than basic decency risks inviting system actors to use
fairness, respect, and praise as tools of control and behavioral manipulation.
By including in the body of “evidence-based practices” habits that should be
dictated by conscience more than science, proponents of evidence-based
practices risk unintentionally reinforcing the use of these inherently valuable
behaviors as expressions of a bureaucratic neopenology, rather than as tools
for ameliorating the harms caused by the expansion of the penal state. And
those risks are real.
Labeling values and moral principles like fairness and kindness “evidence-based” is problematic, even if true. While data that supports the use
of procedural justice to reduce recidivism can reinforce the importance of
those principles, the values that underlie procedural justice should be promoted and rewarded on their own terms. Being fair and treating prisoners,
probationers, and parolees with respect is important not because it induces
compliance with state mandates, but because it is a fundamentally just and
appropriate way for state actors to interact with citizens under state control.
By failing to properly frame the reason for behaving in accordance with fundamental values, proponents of evidence-based correctional practices risk
See Part II.C., supra.
See generally TOM R. TYLER, WHY PEOPLE OBEY THE LAW (2006).
See, e.g., Judy Cameron et al., Achievement-Based Rewards and Intrinsic Motivation:
A Test of Cognitive Mediators, 97 J. EDUC. PSYCHOL. 641, 654 (2005).
185
186
187

ND SSRN.DOCX (DO NOT DELETE)

146

11/30/2015 11:20 AM

NOTRE DAME LAW REVIEW

[VOL. XX:N

turning positive behaviors into tools of coercion. Failure to properly frame
these issues is not consequence-free.
Christopher Lowenkamp, a leader in evidence-based practice research,
has written that “despite the widespread dissemination of ‘Evidence-Based
Practices’ and the ‘What Works!’ literature,” correctional agents have persistently remained focused on asking questions and gathering information
that “pertain[s] solely to the requirements of supervision (e.g., drug testing,
contact with law enforcement, gathering restitution payments, address
changes, and the like).” 188 This limited interaction, he suggests, severely
curtails the ability of the officer to motivate and support behavioral
change. 189 He concludes that there is a need for criminal justice actors to
focus more on the people in front of them:
When we fail to acknowledge [the complexities inherent in any human
being’s life], the view of the offender as an authentic and autonomous
person, with his own intentions and initiatives, is lost. . . . Understanding
the offender’s world requires taking a risk—not grave risk, but risk none190
theless, as doing so is certain to cut against the grain of the status quo.

For Lowenkamp and other humanitarian neorehabilitationists, valuing
the person matters, both because it affirms inherent dignity and because it
can be expected to lead to better criminal justice outcomes: fewer violations,
fewer sanctions, and less future criminal activity. For both humanitarian and
pragmatic neorehabilitationists, that is a win. It is also an evidence-based
practice—one that should be promoted, not simply because it is anchored in
research, but also because it is the right thing to do.
Articulating the values behind evidence-based practices is essential if
they are to avoid becoming instruments of control. Bernard Harcourt has
articulated well how, in the absence of conscious reflection, reliance on predictive tools can shape our beliefs about the purposes of punishment:
The use of predictive methods has begun to distort our carceral imagination, to mold our notions of justice, without our full acquiescence—without deliberation, almost subconsciously or subliminally. Today . . . [w]e

188 Christopher T. Lowenkamp et al., When a Person Isn’t a Data Point: Making Evidence-Based Practice Work, 76 FED. PROBATION 11, 15 (2012).
189 Id. at 15.
190 Id. at 17 (internal quotations omitted) (quoting BAS VOGELVANG, A COMMUNICATION
MODEL FOR OFFENDER SUPERVISION: EIGHT STEPS TO MAKE SENSE OF SCIENCE IN A STREETLEVEL DIALOGUE (2012)). Somewhat defensively, Lowenkamp is quick to assert that such
relationships would not require the revision of fraternization rules. Id. “We are not calling
for any ‘hug-a-thug’ programs that compromise the authority or integrity of the agency. Likewise we are not calling for any approval of criminal behavior. We are, however, calling for
an understanding of that behavior and a willingness to see the person as a person, separate
from the behavior they may have engaged in.” Id.

ND SSRN.DOCX (DO NOT DELETE)

11/30/2015 11:20 AM

201X]

147

have come to associate the prediction of future criminality with just punishment . . . . But the fact is, we have chosen this conception of just punishment . . . . We choose it against a rehabilitative model and as against a
more strictly retributivist model. Or rather, it chose us. Remarkably,
what triggered the shift in our conception of just punishment from notions
of reform and rehabilitation to notions of risk assessment . . . is the production of technical knowledge: our progress in techniques of predicting
criminality is what fueled our jurisprudential conception of just punish191
ment.

His point is well-taken, and equally applicable to other evidence-based
practices. 192 If those in the criminal justice system do not consciously articulate and guard the values that animate their use of state power, then the tools
they use take on a life of their own, imposing bureaucratic values, like efficiency and risk aversion, in place of the moral principles that have long justified the exercise of penal power. 193
There is no question that this is a moment of opportunity in which the
conversation about penal policy has opened up in new ways. 194 How advocates frame the purposes of reform and the methods by which those purposes
are to be achieved will likely mean the difference between sustained change
in the form of stronger communities, decreased imprisonment, and lower supervision rates and the further entrenchment of the penal state.
CONCLUSION: THE FUTURE OF EVIDENCE-BASED PRACTICES
It bears repeating that there is much about evidence-based correctional
practices and efforts to disseminate them that deserves praise. Judges who
use risk assessment tools to check their unconscious biases and ensure that
“low risk” defendants are not being over-punished shield real people from
the criminogenic influences of prison life. Probation officers who use techniques of motivational interviewing to engage with their clients and invest
in their success increase the likelihood that those clients will desist from

HARCOURT, supra note 150, at 31–32.
Cf. Cecelia Klingele, What Are We Hoping For? Defining Purpose in DeterrenceBased Correctional Programs, 99 MINN. L. REV. 1631, 1647 (making a similar case against
the misuse of deterrence-based correctional programs).
193 Cite case law on the centrality of moral blameworthiness in punishment; Kant;
194 See Michael Tonry, Evidence, Ideology, and Politics, in 42 CRIME & JUSTICE IN
AMERICA, 1975-2025, 1, 7 (“Carole Weiss . . . showed that in any place and time, boundaries
exist beyond which change is not possible or even politically imaginable. Public opinion
pollster Daniel Yankelovich (1991) extended the notion to explore the ‘boundaries of public
permission’ outside of which policy changes are unlikely, but within which change is possible
if advocates and public officials are prepared to invest the necessary effort.”).
191
192

ND SSRN.DOCX (DO NOT DELETE)

148

11/30/2015 11:20 AM

NOTRE DAME LAW REVIEW

[VOL. XX:N

crime in the future. 195 Agencies that collect better information about the effects of current programs ensure the wise stewardship of limited public resources and improve our knowledge about how best to help justice-involved
individuals exit the criminal justice system. These are desirable outcomes
that demonstrate how using evidence-based practices can help improve the
fairness and effectiveness of sentencing and correctional practices.
Conversely, when probation officers use risk assessments and motivational interviewing as ways to classify and de-personalize “offenders” and
manipulate them into performing dictated actions, they widen the distance
between themselves and those under their supervision, decreasing the authenticity and legitimacy of supervision in the eyes of the individuals under
state control. 196 When judges use deterrence-based correctional programs to
sanction individuals for relapsing into addiction without concern for triggering stressors, they reduce the legitimacy of the system. 197 Treating people
as subjects to be controlled through techniques of psychological coercion
may be effective at achieving short-term compliance with court orders, but
it perpetuates a belief in the “offender” as “other,” and by doing so reinforces
the idea that the state’s role is to control dangerous populations. That is not
a recipe for reducing the size of the penal state, or diminishing its destructiveness.
Given the current scale of mass incarceration, it is not surprising that
evidence-based practices have thus far been linked to modest-but-real reductions in correctional populations. After all, with nearly one in one hundred
Americans under correctional control, there is plenty of low hanging fruit to
pluck. Even so, the introduction of evidence-based correctional practices
has so far done no more than “nibbl[e] at the edges” of the problem of mass
incarceration. 198 A 2013 study 199 comparing incarceration rates of early JRI
195
196

Cf. Faye S. Taxman, 7 Keys to “Make EBPs Stick”: Lessons from the Field, 77 FED.
PROBATION 76, 76 (2013) (“The evidence-based supervision model . . . is landing onto an organizational landscape where the ‘culture of control’ has existed for over 30 years. To successfully place RNR supervision within these existing organizations, with their mimic massincarceration policies and practices (i.e. punitive, severity, etc.), organizations need to address
the systematic issues that have thrived and existed for the last 30 years—and that present
barriers for the new innovation or refined probation practices to thrive and exist.”).
197 See Klingele, supra note 187, at 1649, 1658, 1660.
198 Michael Tonry, Making Peace, Not a Desert, 10 CRIMINOLOGY & PUB. POL’Y 637,
638 (2011). See also id. at 637 (“For the past 40 years, most advocates for humane criminal
justice policies have made the fundamental mistake of arguing disingenuously. Instead of
arguing that unduly harsh policies are unjust, and should be repealed or modified for that
reason, they much more often argued that policies—which they believed to be unjust —should
be changed because they are ineffective or too costly. . . . This is a mistake.”).
199 AUSTIN ET. AL., supra note 11, at 14, 16. This study was conducted by a distinguished
cohort of scholars and researchers, many of whom were involved in the initial stages of Justice
Reinvestment in 2002-2005, and whom had then offered favorable assessments of its promise.

ND SSRN.DOCX (DO NOT DELETE)

201X]

11/30/2015 11:20 AM

149

states with non-participating states found no significant difference between
the jurisdictions in terms of either admissions or lengths of stay—the two
determinants of prison population size. 200
Defenders of the current approach to reform emphasize that any decrease in prison growth is better than a continuation of the soaring rates of
custody that have defined the past forty years. In this view, incremental improvement—or even stabilization—means more net justice than an approach
that makes reformers feel morally superior but does nothing to alleviate the
human suffering caused by unnecessarily harsh penalties. 201
If the worst that could come from the current approach to reform was a
modest but real reduction in the punitiveness of American penal policy, it
might be forgiven for its lack of ambition. In fact, however, as the preceding
sections have demonstrated, many evidence-based correctional practices
have the ability to be used in ways that might strengthen, rather than undermine, the foundations of the penal state. From excessive drug court requirements 202 to state efforts to remedy risks created by a defendant’s “family
criminality” or “social isolation,” 203 these practices have the ability to expand the control-oriented mentality that allowed for the growth of the criminal justice system over the past century.
Study authors included James Austin, Todd Clear, Malcolm Young, Judith Greene, and representatives from the Justice Mapping Center, the ACLU, the Sentencing Project, and the
Open Society Foundations.
200 See generally Todd R. Clear & James Austin, Reducing Mass Incarceration: Implications of the Iron Law of Prison Populations, 3 HARV. L. & POL’Y REV. 307, 312 (2009)
(“[T]he size of a prison population is completely determined by two factors: how many people
go to prison and how long they stay.”). The researchers attributed this finding to the fact that
JRI had not persuaded legislatures to tackle sentences for individuals “convicted of violent or
sex crimes, drug sales, and second/third felonies” because doing so was not politically feasible. AUSTIN ET. AL., supra note 11, at 16, 18. With respect to this omission, the researchers
observed, “[i]t is insufficient to say that elected officials will not consider these changes without helping them understand that unless length of stay is addressed, prison populations will
remain much as they are today.” Id.
201 Even critics concede that the work of Justice Reinvestment in promoting evidencebased practices has value regardless of its actual effect on prison population size: “JRI has
played a major role in educating state legislators and public officials about the bloated and
expensive correctional system, persuading them to undertake reforms not previously on the
table. Considering the country’s four-decade addiction to mass incarceration and harsh punishment, the general refusal to acknowledge its failures and the monumental resistance to
change, JRI’s most enduring contribution to date may be its having created a space and a
mindset among state officials to seriously entertain the possibility of lowering prison populations.” AUSTIN ET AL., supra note 11, at 1.
202 Miller, supra note 150, at 441.
203 Many risks and needs assessment tools identify “criminogenic needs” of offenders,
some of which reflect differences in class, culture, and experience that are relevant to risk,
but not to deserved punishment. Need categories reported on the COMPAS assessment tools,
for example, include “family criminality,” “socialization failure,” “criminal personality,” and

ND SSRN.DOCX (DO NOT DELETE)

150

11/30/2015 11:20 AM

NOTRE DAME LAW REVIEW

[VOL. XX:N

The rapid spread of evidence-based correctional practices has been attributed in part to their pragmatic, bottom-line, hard data rhetoric. Talk of
data and efficiencies and actuarial tools is cool and detached, and can rise
above some of the heated partisan rancor that has so long defined and complicated conversations about criminal justice. The problem is that de-personalization is just that. It divorces even those implementing reform from
confronting the underlying reason why reform is necessary: not because
prison is costly, but because prisons are filled with too many people locked
in cages for years at a time, not infrequently for crimes that only a few short
decades ago would have gone unpunished or drawn a substantially less severe sentence. That is an uncomfortable truth. By talking about money and
data, many reformers hope to avoid these hard conversations and jump
straight to solving the perceived problems of an overly harsh and insufficiently rehabilitative criminal justice system. But there are no shortcuts to
culture change. Fundamentally, underneath the talk of money and evidence
is a belief on the part of most policymakers that too many people are being
punished too harshly.
Both common experience and “evidence” suggest that the answer is a
correctional system that responds to the concerns, needs, and antisocial propensities of actual people—not aggregate stereotypes or depersonalized
“risks.” In this model, evidence-based correctional practices are important
because they enable system actors to identify interventions that may assist
individuals in “making good.” 204 Always, though, the data about what works
in the aggregate must be made subservient to the needs and responsiveness
of the individual.
Without explicit discussion of the normative purposes of correctional
intervention, evidence-based practices—or any correctional practices, for
that matter—become ends unto themselves. When that occurs, evidencebased correctional “reforms” quickly become indistinguishable from the new
penology they seek to disrupt.
Scholars, policymakers, and practitioners should recognize the potential of evidence-based practices to improve the quality and effectiveness of
correctional interventions, while remaining equally alert to their potential for
coercion and abuse. There are many ways this can be done—all of which
are subjects worthy of greater analysis and future study. Examples include
discussion groups within probation and parole agencies (or in the context of
larger criminal justice working groups) that create space for those in the field
to air concerns about specific ways in which a focus on tools and metrics
“social adjustment problems.” See generally NORTHPOINTE, PRACTITIONERS GUIDE TO COMPAS
(2012),
http://www.northpointeinc.com/files/technical_documents/FieldGuide2_081412.pdf.
204 Cf. SHADD MARUNA, MAKING GOOD: HOW EX-CONVICTS REFORM AND REBUILD
THEIR LIVES 85–108 (2001).

ND SSRN.DOCX (DO NOT DELETE)

201X]

11/30/2015 11:20 AM

151

may be obscuring larger system goals. 205 Acknowledging the need for such
conversations is itself a check on the potential abuse of evidence-based correctional tools. They also include efforts like those of the Robina Institute’s
Probation Revocation and Parole Release projects (with which this author is
affiliated), that team academic researchers with local practitioners to explore
agency culture around areas of common concern and identify areas for improvement in the delivery of justice. 206
There is no better time to undertake such efforts. As groups like the
NIC and JRI equip system actors with new tools and train them in new skills,
scholars, policymakers, and practitioners must help place those tools into a
larger framework that challenges the habits of mind and practice that enabled
mass incarceration in the first instance. “In the end, law and legal institutions—especially concerning issues as emotionally laden as crime and punishment—are based on values, not on cost–benefit analyses and effectiveness studies.” 207
Significant and sustained reductions in prison populations will only
happen when we believe collectively that the scale of the penal state is the
right thing to do. When used wisely and with caution, evidence-based correctional practices can help in these efforts. To ensure that these reforms
meet their potential, however, we must continually monitor whether they are
being used as tools to reduce the reach of the penal state, or to facilitate its
growth.

205 Although most efforts to engage in such conversations are ad hoc, a small body of
literature on criminal justice councils and work groups describes how such spaces might be
created. See, e.g., M. ELAINE NUGENT-BORAKOVE & MAREA BEEMAN, JUSTICE MGMT. INST.,
FOSTERING AND SUSTAINING CRIMINAL JUSTICE SYSTEM REFORM: THE POTENTIAL OF CRIMINAL JUSTICE COORDINATING COUNCILS (2013), http://69.195.124.207/~jmijust1/wp-content/uploads/2014/04/Fostering-and-Sustaining-CJ-Reform.pdf.
206 See ROBINA INST. OF CRIMINAL LAW & CRIMINAL JUSTICE, Parole Release & Revocation Project, http://www.robinainstitute.org/parole-release-revocation-project/ (last visited
Sept. 30, 2015); Probation Revocation Project, ROBINA INSTITUTE OF CRIMINAL LAW & CRIMINAL JUSTICE, http://www.robinainstitute.org/probation-revocation-project/.
207 Tonry, supra note 193, at 640.