House of Commons - Justice Committee

Cutting crime: the case for justice reinvestment - Justice Committee Contents

Memorandum submitted by Michael Marcus, Circuit Court Judge

I have passionately pursued for many years some of the same questions which confront this Committee.1 I believe that I have learned some of the most basic issues that must be addressed before we can expect meaningful improvement in the use and consequences of prisons in the many countries uncomfortably competing for leadership with the United States in imprisonment rates.2 I understand this Committee to be concerned, in part, with the implications of the Carter Report, and I will respond to those portions of the Terms of Reference where I believe I may have something to offer. But I learned in 20 years of practice as an attorney before taking the bench in 1990 that I should make my most important points first. This brief paper will endeavor to do so. The Carter Report's hoary recommendation of "purchasing and converting a suitable vessel into a prison ship"3 is symbolic of the need for a profound alteration in our approach to these problems.

With respect, the Report misses the same point ubiquitously missed by almost all prison reform work in the Western World for the last four or five decades: we can make no substantial progress without eschewing magical thinking and embracing the same tools required for meaningful progress in any social endeavour with measurable consequences: prescription of measurable objectives and rigorous application of rational thought solidly grounded in empiricism. Of course we can and should vet addiction treatment programmes for effectiveness and send many addicts to them instead of to prison. But the enormous problems of crimes and prisons will continue to frustrate our best intentions until and unless we abandon the magical thinking that "just deserts" is a sufficient objective of sentencing and that it actually promotes the objectives we apparently assign to it. In essence, we must displace the false religion that dominates criminal justice with behaviours that actually serve public safety and public values.

I now will attempt to outline the major points I hope to convey, providing some links to supporting material. I would welcome enthusiastically any request to provide more documentation or to continue the discussion.

1. SENTENCING COMMISSIONS: JUST DESERTS IS NOT ENOUGH

The Carter Report is obviously taken with the combination of sentencing guidelines and a sentencing commission. I've been deeply involved in this discussion in the United States, having testified before our own Sentencing Guidelines Council before Oregon adopted our guidelines in 1989, and having been involved as well in the American Law Institute's current project to revise the Model Penal Code to implement a guidelines/commission approach to sentencing. Guidelines have definite values.4 They allow an avenue by which to normalize sentencing so as to predict, and to some extent control, our use of prison resources. They accomplish some meaningful moderation of the alarming disparity sometimes achieved by unbridled—and unguided and uninformed—judicial discretion.

But guidelines in the US have at best reduced prison population growth rates, and they have repeatedly fallen to popular will motivated to increase sentences, reduce sentencing discretion, and create mandatory minimum sentences—all of which continue to swell prisons. Guidelines' accomplish consistency largely by pretending that quite disparate offenders or offenses are alike. But most importantly, by focusing on the symptom—prison rate escalation—they wholly ignore the disease. That disease is the magical thinking that just deserts - a "just sentence"—is a sufficient accomplishment, to the practical exclusion of responsible allocation of prison and programme resources according to real evidence of real risk and real opportunity for reduction in criminal behaviour. Historically, the surge in prison growth in recent decades was itself in part a reaction to failure of the magical thinking that merely sending offenders to programmes nominally relevant to their criminal behaviour—theft talk, alcohol or drug treatment, anger management, and the like—we would reduce their criminal behaviour. In both applications, magical thinking does not produce the results we desire just because we hope for or expect success. Like anything else that matters, we need to be rigorous, measure, and progress through the application of empirical experience. To the extent that any approach deviates from that principle, we are doomed to repeat the mistakes of the past and, increasingly, to suffer their consequences.

This flaw in the mission and structure of sentencing commissions and guidelines is wholly shared with discretionary sentencing in the absence of guidelines, and completely eclipses the significance of heated debates concerning the value of judicial discretion and the correct locus of guideline authority in the judicial, legislative, or executive portion of government. The meaningful issue is whether sentences or guidelines are properly directed and responsibly crafted—and all branches of government have triumphed in their demonstration of a capacity to get it wholly wrong. Stated otherwise, the profound failure of guideline structures is that they have always tended to use typical judicial behaviour as a lodestar for normalization—without bothering to notice that the outcome of that typical behaviour is enormous recidivism. Missed, also, is that guidelines seek to capture in a matrix the collected judicial wisdom which made no meaningful attempt to direct sentencing at reducing recidivism (or serving any goal other than just deserts), and which was based on no effort whatever to accumulate any information upon which responsibly to base any such attempt.

So guidelines continue to allocate prison beds based on ordered just deserts rather than risk, need, and science. From a risk point, we have found that Oregon's guidelines get it wrong two-thirds of the time.5 This is not just an immediate waste of resources, but a substantial part of the problem. Empirically incorrect imprisonment—in the sense that we send many of the wrong people there—is rampant, and it is itself a major criminogenic factor for many offenders whose recidivism rate is increased by their experience. In other words, misusing prison is one of the major causes of prison growth. The great majority of inmates now in prisons were previously sentenced for prior crimes6—reflecting the reality that sentencing makes no responsible effort to reduce future crime. Unfortunately, we increase the criminality of many offenders by sending them to prison so that we eventually need prisons for them when we didn't in the first place. This factor escalates as prison overcrowding worsens the already counterproductive aspects of prison life. We contribute to victimization at the hands of several groups of offenders for whom our mindless distribution of sentences makes things worse:

— those who would have committed fewer crimes had we sent them to appropriate community based programmes instead of to prison;

— those who would have committed fewer crimes had we afforded them the programmes that we cannot afford because we misallocate resources to prisons;

— those whose prison terms were not long enough to serve public safety because we were diverting substantial prison resources from those that should to those who should not be in prison; and

— those in prison who would have committed fewer crimes after prison had we allocated resources to their rehabilitation and reintegration—instead of using those resources to house others for whom prison was not the best public safety response.

Guidelines, commissions, and appellate review can be a useful part of the solution, but they must first overcome these flaws. In brief, we need to use the guidance of this triumvirate not primarily to attempt to manipulate how sentencing affects prison rates, but to achieve evidence-based allocation of prison and other sentencing resources—within limits of proportionality and law to be sure, but to the ends of serving public safety and public values, with rigorous attention to actual impact on actual objectives. Similarly, we must be rigorous in measuring the performance of programmes that are supposed to reduce criminal behaviour, and in allocating them intelligently in and out of prison. We know a lot more now about what makes a programme work than we did when some prematurely reached the conclusion that "nothing works", one prominent genesis of the prison population explosion. See Responding to the Model Penal Code Sentencing Revisions: Tips for Early Adopters and Power Users, note 4.

2. RISK ASSESSMENT—ONE PIECE OF THE ANSWER

Perhaps the most valuable innovation in the United States guidelines evolution is the experience of the Virginia Criminal Sentencing Commission.7 To be sure, that commission, like all others to date, is profoundly impaired by the pervasive and suffocating effect of just deserts, with vast attention to "aggravation", "mitigation", and supposed past judicial wisdom and public vindictiveness. But Virginia, and to a growing extent, Missouri, have introduced risk assessment into the mix with very promising results. Virginia developed and validated an instrument, and first applied it to select the highest risk sex offenders for increased incarceration in pursuit of public safety. The politics of the result were sufficient to motivate the Virginia Legislature to direct the Commission to develop and implement a risk assessment instrument to divert a substantial portion of lower risk offenders away from prison and into community based sentences. Virginia's experience is that the result significantly impacted the demands on prison and reduced the recidivism rate of violent offenders.

Missouri's experience is newer, and consists of a risk-assessment based instrument to assist in allocating prison beds based on risk.8

To be sure, all existing sentencing commissions seem to concern themselves with monitoring and reporting on compliance by judges with sentencing guidelines—a worthwhile subsidiary goal, but only to the extent that the guidelines are themselves productive of something of social value. A thing not worth doing is not worth doing well. In my view, commissions should be primarily charged with monitoring how well sentencing achieves its social objectives of supporting public safety and public values,9 and only secondarily tasked with monitoring the effect of guidelines on sentencing behaviour and prison bed demand.

But the major point of this subsection is that any improvement must embrace and not eschew risk assessment as an ingredient of the solution. There are voices contending that risk assessment is imperfect because productive of "false positives", and unfair because either "punishing for future conduct" or invoking static variables over which offenders have no control. Although I advocate empirical vigor in all of this and make no exception for risk assessment—we must be mindful of its flaws and vigilant in pursuing its continued improvement— none of these arguments have ultimate merit. They miss the point that it is risk that we are managing, not future conduct that we punish. They support sentencing based on ephemeral rather than empirical bases which, if measured by fairness or public safety, is far more productive of false positives (and false negatives). Deeming an offender a high risk of producing harm is not "false" simply because he does not actually commit that new crime; the issue is whether his risk was accurately assessed. As long as the resulting sentence is neither disproportionate in its severity nor unlawful, there is nothing unfair or inappropriate in allocating prison time and space to offenders who are, based on our best efforts at assessment, more likely than others to do more harm. To insist that we ignore this difference is to punish many who do not need it (and thus to be unfair to them) and to incapacitate insufficiently those who do need it—which is unfair to their avoidable victims.10

Ultimately, the opponents of risk assessment must and do reject public safety as the purpose of incapacitation, for if public safety is its purpose, pursuit of effective risk assessment is a necessary corollary.11 It makes no sense to deride risk assessment as imperfect or unfair (or as "preventive incarceration") when the default these opponents defend is overwhelmingly less informed, less careful, less analytical, and routinely productive of high recidivism rates.12

3. ACCOMPLISHING CHANGE

This part of the piece assumes agreement that any meaningful solution requires a profound change in the very culture of sentencing, that the objective is a rational allocation of all sentencing resources within the limits of proportionality and law, based on risk, resource, and priority. The question that remains is how to move a long-entrenched and archaic liturgy13 into the 21st century to join other social institutions in an evidence-based, performance measured, accountable engine by which to pursue rational social objectives. I am not much of a politician, although I have successfully run in a contested election for the office I hold, and have often testified to our state Legislature in support of efforts related to this one.14 I have a few thoughts on the politics of this undertaking that may be useful, and a good deal to report on our strategies within the judicial department for accomplishing this change.

Politics of effective sentencing. What I have to offer is limited. Most is recited in Justitia's Bandage: Blind Sentencing.15 First, we must understand that one obstacle to "reform" is a misconception that the public is an enormous obstacle. Policy makers constantly overestimate the public's punitivism, and underestimate its support for effective efforts at rehabilitation.16 This is not foreign to the United Kingdom. The Halliday Report concluded:

When asked unprompted what the purpose of sentencing should be, the most common response is that it should aim to stop re-offending, reduce crime or create a safer community. Next most frequently mentioned are deterrence and rehabilitation. Very few spontaneously refer to punishment or incapacitation.

The next concept is that there is much room for bridging ideological divides by agreeing that severity vs. leniency is not at stake—that what matters is the rational and efficient distribution of existing resources so as to accomplish most efficiently crime reduction and whatever other public purposes are to be pursued with sentencing. There is, as stated, enormously underestimated public consensus that reducing recidivism is the major purpose of criminal law and sentences, that rehabilitation can serve these goals for some, and that incapacitation is necessary for others. The public, for example, is easily persuaded to divert drug users to treatment in lieu of prison, itself a relatively easy but limited remedy to misuse of prison.

The exclusive pursuit of "just deserts" has been an enormously destructive but effective excuse for not making any responsible effort to meet these public expectations. It has also enabled the public fallacy that severity and effectiveness are directly proportional for all offenders. Accepting accountability for outcomes and for appropriate allocation of resources measured by their actual impact on these goals has enormous potential for support along the vast majority of the continuum of public attitudes towards crime and punishment.

"Evidence Based Sentencing" has something of a burden in overcoming a reputation for leniency. Perhaps it must be renamed, but whatever it is called, the trick is to apply it not just to justify programmes for lower risk offenders, but across the entire spectrum of offenses and offenders, within limits of proportionality and resource, so that we act as if we are actually part of the modern world of empiricism, as opposed to participants in ancient rites founded upon magical thinking. This means recognizing the large cohort of offenders for whom less intervention is demonstrably better for public safety than more; wisely vetting programmes for their actual usefulness in reducing the criminogenic aspects of cohorts of offenders as demonstrated by reduction in recidivism; ensuring that we assign programmes and custody resources based on our very best efforts to do that which works on those on whom it works,18 and substantially lengthening the terms of violent recidivist offenders against whom incapacitation is indeed the morally justified and empirically required social protection.

Strategies for changing sentencing. It is here that I finally reach our unique product, Multnomah County's Sentencing Support tools. I reach them here because their primary purpose is not to dictate sentences, but to assist in changing the culture of sentencing by encouraging and informing the discussion about what is most likely to reduce a given offender's future criminal conduct. The tools are best described through the web link in endnote 1 (and I'm happy to demonstrate them when we meet). In brief, they display recidivism outcomes correlated with sentencing elements (such as jail terms, forms of supervision, treatment programmes, alternative programmes, and prison) imposed for similar offenders for similar crimes. The user enters a case number (or other identifier for an individual offender), chooses a crime for which a sentence is contemplated, and promptly receives a display of sentencing elements used for such offenders for such crimes. The elements are displayed by bars in the order of the frequency of that element's imposition for such offenders for such crimes. The height of each bar in the chart is reflective of the proportion of offenders who received that sentencing element who were free of recidivism measured—by default—as freedom from a conviction for a similar offense within three years of sentencing. Users can modify what is meant by a "similar offence," a "similar offender," or "recidivism," and promptly receive a new display based on the altered variables. The tools are designed to be convenient and quick enough to use in court during a sentencing hearing; I can typically produce a result in less than a minute including printing copies for counsel. The court provides a dedicated computer for the use of attorneys. The prosecutor's office has the tools in-house, as does the largest of the indigent defense firms.

The tools are not designed to display causal relationships between dispositions and results, but to group meaningful displays of historical outcomes—to encourage a pursuit of responsible predictions. They are far from perfect, but far better than what usually drives sentencing decisions.

Sentencing support tools are but one device for attempting to move sentencing toward best practices. Others are described in several articles,19 but here is a short description of the ideas we have pursued:

— Proclaiming in law and in resolutions that sentencing must pursue public safety through validated responses that demonstrably reduce recidivism.20

— Constructing and maintaining database practices and applications that provide information to support data-driven sentencing decisions.21

— Integrating risk and need assessment into sentencing analysis.22

— Developing evidence-based practices in collaboration between courts and probation departments.23

— Requiring that authors of pre-sentence reports for the courts explore, analyze, and report on what is most likely to reduce future crime—including the availability of effective programmes in and out of custody.24

— Including evidence-based practices, stage of change analysis, motivational interviewing, and practical sentencing in judicial education curricula.25

— Providing on-line practical references for smart sentencing to judges and advocates.26

— Attempting to move the culture of plea-negotiation towards evidence-based practices.27

— Adopting court performance measures that actually assess how effective sentencing is at achieving public safety or any other purpose.28

4. THE CARTER REPORT

I am, of course, in no position to do the budget projections and analysis necessary to give this Committee precise answers to questions about the costs associated with the Carter Report recommendations, or related probation and programmatic efforts. I can with confidence, however, offer some propositions that I expect to withstand analysis and the test of time. These propositions are equally applicable in the US—particularly in California and Oregon, which have debated responding to crime and prison problems with some mix of much more prison and some more "treatment."29

First, whether or not it is unavoidable to include some expansion of the supply of prison beds in the short run, expanding prison capacity in relation to the general population can only exacerbate the problems that generate the prison crisis. Until and unless we learn to use prison more wisely—by far—than we do now, adding prison beds will promote increase in crime and the need for yet more prison beds. We are effectively addicted to the essentially mindless use of prisons. As with any addiction, feeding the habit is no cure.

Second, we must commit to the primary objective of using our resources across the entire spectrum of offenders to accomplish public safety within the limits of proportionality, risk, and priority. In order to accomplish the primacy of the pursuit of public safety, we must displace the facade of just deserts as an adequate measure of our sentencing performance. There are legitimate objectives lurking with "just deserts," but they, too, are capable of identification and measurement—at least to an enormously improved extent as compared with existing practices. Thus, we should do our best to quantify those bundled objectives and to determine the actual extent to which they are served by sentencing dispositions. On occasion, but rarely, they will justify deviation from the sentence best crafted to serve public safety.30

Third, our commitment to actual outcomes must include relying on the best research and data in making policy-level and individual sentencing decisions. The issue of the reliability of the evidence for "evidence based practices" is quite real. There are some clear essentials:

— Research is essential, but so is actual performance measurement based on the impact on recidivism, for comparable cohorts, of all of the dispositions practically and legally available for those cohorts—from and including doing nothing substantial—which is demonstrably the most effective response to a large portion of the least serious categories of offenders, through community based programmes and alternatives, to prison—including the length and circumstances of incarceration, which in turn includes prison programming and reintegration efforts.

— Because the meaning of research and data is so easily lost on those who collect and dwell on it, and manipulated by those who have an interest in outcomes, the highest calling of sentencing commissions is not monitoring compliance with guidelines or sentencing protocols, but (1) vetting research and data for constant validation and improvement, and (2) devising and deploying strategies for achieving evidence-based practices and desired public safety results both at policy level and individual sentencing decisions.

— Whatever the deficits in available research and data, sentencing that attempts to make the best use of that research and data is overwhelmingly more likely to achieve public safety and to promote public values than current sentencing behaviours—which are predominantly dependent upon the fallacy that just deserts is a sufficient accomplishment.

Finally, the long term costs and cost savings of responses to the crime and prison crisis will overwhelming depend not upon the immediate amount of any investment, but upon the extent to which we can improve our ability to deploy all correctional assets efficiently to reduce recidivism. Recidivism is the major driver of these expenses, as well as of the expense to society of avoidable victimizations.

CONCLUSION

The shelves of public and private collections, and their more recent electronic storage equivalents, are filled with decades of well-intended and presumably thoughtful studies by advisory groups and committees (I've served on several)—most of which have had little impact on our behaviours and are long forgotten. It is respectfully submitted that this Committee's work is likely to suffer the same fate unless it responds on the level of these suggestions—not necessarily by adopting all or even any of them, but at least by vigorously pursuing the causes of the crisis rather than merely its symptoms, and by boldly confronting the need for profound, systematic alterations to the culture and practice of criminal justice.

April 2008

REFERENCES1 Much of this pursuit is evidenced on or through the web site I have maintained for many years, smartsentencing.com. This site describes our sentencing support tools, our legislative efforts to promote evidence-based sentencing, and the various articles I have published in an attempt to promote the changes I believe necessary to take us all out of our cruel morass. The site also has an eMail link. I endeavour to answer every question put to me about this effort; the site has a "frequently asked question" page for the same purpose.

2 The most current report puts the U.S. at the very top of this terrible pile, with England and Wales at about 16th place—just behind Romania. One in 100: Behind Bars in America 2008 at 35, Table A-7 (Pew Charitable Trust 2008). While the US is way ahead in the proportion of its population in prison, the prison growth rate in England and Wales well justifies grave concern—and eases, somewhat, the irony of the notion that we in the U.S. have something to offer. I believe some of us do have much to offer, but our ideas are either obviously wrong or quite right but well out of the mainstream of actual practice. I think I can make the case for the latter.

3 Carter Report at 28. The rotting prison hulks in the Thames came immediately to my mind. See What Are We up to and Why—or If We're Doing More Harm than Good, Why Rush?, presented at the West Central Wardens & Superintendents Association "Round-up on the Oregon Trail," Pendleton, Oregon, June 1997. To be fair, using a suitable vessel to supplement prison space as a temporary solution to overcrowding is hardly irrational, and the prison ship is but a minuscule part of Lord Carter's recommendations.

4 For purposes of this discussion, all of the points are supported at greater length in particular in two articles on the subject, of which I have reprints that I'd be happy to provide when we meet in person: Responding to the Model Penal Code Sentencing Revisions: Tips for Early Adopters and Power Users, 17 S Cal Interdiscipl L J 68 (2007) (includes A Harm-Reduction Sentencing Code); and Comments on the Model Penal Code: Sentencing Preliminary Draft No. 1, 30 American Journal of Criminal Law 135 (2003).

5 See Responding to the Model Penal Code Sentencing Revisions: Tips for Early Adopters and Power Users, supra note 4, 17 S Cal Interdiscipl L J at 76 & n 30.

6 Eg, Bureau of Justice Statistics Criminal Offenders Statistics, United States Department of Justice.

7 Virginia's web site provides the annual and subject reports that describe the risk assessment experience in detail, particularly those of the initial validation and application of risk assessment to sex offenders, describing the implementation of risk assessment to divert low risk offenders from prison, and describing the result: fewer lower risk offenders in prison, longer terms for higher risk offenders, and lower recidivism rates for higher risk offenders.

8 The Missouri Sentencing Advisory Commission implemented a risk assessment-based "System of Recommended Sentencing" in 2006, and described its impact on reducing the flow of incoming prisoners most recently in a 2007 Biennial Report.

9 Eg, Responding to the Model Penal Code Sentencing Revisions: Tips for Early Adopters and Power Users, supra note 4.

10 See generally, Christopher Slobogin, The Civilization of the Criminal Law, 58 Vanderbilt Law Rev 118 (2005), and authorities cited. [The link takes you to an abstract; the entire article may be downloaded from any of the links below the abstract]. For a more extended examination of risk assessment, see Marcus, Blakely, Booker, and the Future of Sentencing, 17 Federal Sentencing Reporter, 243 (2005); Marcus, Limiting Retributivism: Revisions to Model Penal Code Sentencing Provisions, 29 Whittier Law Review 295, 317-29 (2007).

11 If public safety is not the objective, it makes more sense and is more humane at least to offer caning as an alternative to prison terms. If public safety is out of the equation, caning would probably satisfy any public need for retribution short of capital cases; it would solve the prison overcrowding problem; it would save enormous sums that might be devoted instead to programmes that actually work to reduce criminal behaviour, and to social services that divert potential offenders from crime altogether: high school completion, parenting education, and interdisciplinary multisystemic intervention.

12 Comments on the Model Penal Code: Sentencing Preliminary Draft No. 1, supra note 4, at 145-56 & n 44, cited in The Civilization of the Criminal Law, supra note 10, at 22.

13 Marcus, Archaic Sentencing Liturgy Sacrifices Public Safety: What's Wrong and How We Can Fix It, 16 Federal Sentencing Reporter 76 (2003).

14 We are in the midst of a sweeping modernization of court technology. Our Chief Justice asked me and our local Trial Court Administrator to make the case to our Legislature that improving our performance of our mission—including public safety and community well being—is the reason to avoid delay in funding notwithstanding an economic downturn. We obtained the authorization. (I've put these clips on YouTube at least temporarily to facilitate sharing them; I hope to find a more suitable mechanism—and disclaim any embarrassing unintended associations with content of obvious irrelevance.)

15 Marcus, Justitia's Bandage: Blind Sentencing, International Journal of Punishment and Sentencing (2005).

16 Princeton Survey Research Associates International for the National Center for State Courts, The NCSC Sentencing Attitudes Survey: A Report on the Findings (July 2006) at 36; Peter D. Hart Research Associates, Inc., Changing Public Attitudes Toward the Criminal Justice System (The Open Society Institute, February 2002); Belden, Russonello & Stewart, Optimism, Pessimism, and Jailhouse Redemption: American Attitudes on Crime, Punishment, and Over-incarceration (Washington, DC 2001); Cutting Correctly—New Priso N Policies for Times of Fiscal Crisis (Center on Juvenile and Criminal Justice, 2002), US Department of Justice, National Institute of Corrections, Promoting Pubic Safety Using Effective Interventions, Section 1 (February 2001), citing, eg, B K Applegate and F T Cullen, and B S Fisher, Public Support for Correctional Treatment: The Continuing Appeal of the Rehabilitative Ideal, 77 Prison Journal 237-58 (1997).

17 Making Punishments Work: Report of a Review of the Sentencing Framework for England and Wales (Home Office 2001) App 5 at 108. The British Sentencing Advisory Council commissioned public opinion polls to assist in recommending sentencing guidelines to the Court of Appeal because it saw such polls as a better measure of what it takes to maintain "public confidence" than press coverage of notorious cases. Unfortunately, the Council stacked the deck by focusing on public attitudes toward factors of aggravation and mitigation and particular variations in targeted crimes. Sentencing Advisory Panel, Annual Report (2003) at 25. The United States Sentencing Commission adopted a similar approach in commissioning a study of how Americans would sentence federal crimes. Peter H Rossi & Richard A Berk, National Sample Survey: Public Opinion on Sentencing Federal Crimes: Executive Summary (1997). Asking the public to hold forth on what punishment "fits the crime" is like asking the public for an opinion on religion. Carefully constructed opinion polls, however, can be a useful part of a process for assessing what is really necessary to serve such issues as public trust and confidence in courts and the government, and what actually serves the function of supporting those public purposes hidden behind the liturgical mask of "just deserts." See Responding to the Model Penal Code Sentencing Revisions: Tips for Early Adopters and Power Users, supra note 4.

18 The exhaustive 1998 "Maryland Study" of the effectiveness of programmes concluded:

et al

Lawrence W Sherman, Denise C Gottfredson, Doris L MacKenzie, John Eck, Peter Reuter, and Shawn D Bushway, Preventing Crime: What Works, What Doesn't, What's Promising (National Institute of Justice 1998). The quote is from the unpaginated report under the heading "4. Rehabilitation and Treatment."

19 Nation-wide, the most recent treatments of these types of attempts at injecting evidence-based sentencing into the culture of criminal justice are Roger K Warren, Evidence-Based Practices and State Sentencing Policy: Ten Policy Initiatives to Reduce Recidivism, 82 Indiana L J 1307 (2007); and Marcus, Smart Sentencing: Public Safety, Public Trust and Confidence Through Evidence-Based Dispositions, in Future Trends in State Courts 2006 at 56, (Nat'l Ctr For State Courts 2006). See also Justitia's Bandage: Blind Sentencing, supra note 15, at 20-28. Devising such strategies is the mission of the National Institute of Corrections's National Advisory Committee on Evidence Based Decision Making for Local Court Systems (of which I am a member) that last met in November, 2007. NIC has recently released a request for proposals for a Cooperative Agreement: Evidence Based Decision Making for Local Criminal Justice Systems.

20 Our examples are collected here: Legislative, Judicial, and Criminal Justice Commission Materials. The purposes recited in the Criminal Justice Act 2003 are typical of those in most US States, and reflect the same purposes adopted by the 1962 Model Penal Code. Section 142(1) of the Criminal Justice Act surely contains the crucial ingredients for these purposes: "the reduction of crime," "the reform and rehabilitation of offenders," and "the protection of the public." The problem has usually been that these are but banners for the justification of punishment rather than actual goals responsibly pursued. Moreover, the unprioritized list of purposes gives no actual direction to those who argue or decide sentencing issues—judging by the reduction of the whole exercise in most cases to ephemeral and unaccountable musings about aggravation and mitigation, the liturgy of just deserts. See, eg, Justitia's Bandage: Blind Sentencing, supra note 15, at 10-11, and authorities cited.

21 Multnomah County's Sentencing Support tools however modest, are still the most highly developed in the world as far as I can tell. There is a recent analog for probation officers in the District of Columbia. Oregon's ongoing court technology modernization project is committed to expanding sentencing support tools state-wide, and also to release, probation, juvenile, and family law matters. Note that there is a more basic level at which technology can support smarter sentencing: simply providing convenient access to information such as prior criminal history, related cases and holds, performance on supervision, and available programmes in and out of custody.

22 As noted, Virginia (note 7, supra) and Missouri (note 8, supra) have accomplished this in their guidelines; Oregon has attempted to do so but has yet to succeed. We have certainly regularly incorporated risk assessment into presentence investigations.

23 Probation departments are often steeped on the literature of criminology and evidence-based practices; they are familiar with needs and risk assessment instruments. In the past, they have been allowed and encouraged to leave those concepts outside when they enter our temples of denunciation, and to argue deserts instead of science at probation hearings. Yet, by their training, at least many probation officers and their managers are generally receptive to changing the role of probation officers. In my experience, many are enthusiastic about becoming the courts' experts in what works and in what is available, and advocates for smarter sentencing. This is perhaps the most effective of our innovations to date.

24 We began this initiative in 2002 by adding a checkbox calling for such information to our local form of order for a presentence investigation. The Oregon Legislature expanded this requirement state-wide in 2005. The change has been dramatic and positive in the value of presentence investigations, but limited due to the relative rarity of such investigations since the advent of sentencing guidelines in 1989. Since these reports are prepared within our probation department, however, the change has spread to probation reports—which also regularly include risk and needs assessments, stage of change analysis, and useful information about the actual availability of programmes and their likely success. So far, however, only presentence investigations routinely include sentencing support tool results.

25 It took years, but we finally succeeded in getting a "practical sentencing"session on the curriculum of an annual trial judges' conference.

26 Oregon maintains an on-line Criminal Benchbook; Chapter 16, Sentencing, begins with some 30 pages of practical suggestions for achieving effective sentencing. [The page numbers and chapter heads are internal hyperlinks; the book itself is some 1,100 pages long.]

27 Plea bargains account for the vast majority of sentences in the US. This is apparently now true in the UK as well. Any change that matters must change plea bargaining. We are making initial attempts locally to address this aspect of the problem.

28 The "court performance measure" movement has so far almost entirely avoided measuring anything of value apart from speed. See generally Courtools On-line (National Center for State Courts). Oregon has slowly begun developing tools that address recidivism around treatment courts. The issue is, of course, controversial among judges who think only they get to do the judging, but I expect to expand this measure to sentencing in general within the next two to four years.

29 California's 2007 Assembly Bill 900, subject of a recent symposium at the University of San Francisco Law School in which I participated, is much of what the Carter Report prescribes—many more beds, some more treatment. A measure seeking adoption through the ballot process, California's Nonviolent Offender Rehabilitation Act of 2008, would laudably focus on providing treatment to the substantial portion of the criminal justice population subject to drug addiction. In Oregon, our November 2008 ballot will see a competition between proposals opting for a very great addition to our prison capacity and one offering a slightly less drastic addition coupled with a substantial increase in treatment resources.

30 The legitimate purposes of sentencing other than public safety are probably limited to those that respond to a legitimate need of a victim, prevent vigilantism or private retribution, maintain respect for legitimate authority, or enhance respect for the persons, property, or rights of others. In short, the purposes of punishment other than public safety are those that serve public values. Even if this list is not exhaustive, the critical piece is to identify purposes other than safety and to ensure that they impact sentencing through reasoned and evidence-based analysis, particularly in the rare instance when they might actually conflict with public safety objectives. Examples of the latter are the social drinker who causes a death, the sex abuser whose victim needs a tangible identification of blame with the offender, or the criminal behaviour that is actually based on a cost/benefit analysis, such as an industrial polluter. See Responding to the Model Penal Code Sentencing Revisions: Tips for Early Adopters and Power Users, supra note 4, 17 S Cal Interdiscipl L J at 78-83, 114-15.



© Parliamentary copyright 2010	Prepared 14 January 2010