Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
1 Introduction: Sentencing Practices and the Sentencing Reform Movement The sentencing decision is the symbolic keystone of the criminal justice system. It is here that conflicts between the goals of equal justice under the law and individualized justice with punishment tailored to the of- fender are played out and here that the criminal law is interpreted and applied. So it is not surprising that, as crime increased and questions about the fairness and effectiveness of the criminal justice system grew more and more pressing in the United States in the early 1970s, re- formers turned to the courts and their sentencing practices, which one federal judge characterized as "lawless" (Franker, 1972~. Increased awareness of the pivotal role of sentencing in linking the criminal law and criminal sanctions has recently focused reform efforts on sentencing. These developments followed 50 years in which there had been little change in sentencing practices and institutions. When the National Commission on Reform of Federal Criminal Laws reported to Congress in 1970, its sentencing proposals to rationalize and simplify the then-ubiquitous systems of indeterminate sentencing differed little from those of the Model Penal Code developed in the 1950s. When the commission reported, "determinate," "presumptive," and "flat-time" sentencing had not yet been proposed. The U.S. Parole Commission's parole guidelines were several years away. With minor exceptions, sen- tencing was not on state legislative agendas. Sentencing guidelines were beyond the horizon. Since 1975, however, substantial changes have been introduced. Pa- role has been abolished in at least 10 jurisdictions, while parole guide- 39
40 RESEARCH ON SENTENCING: THE SEARCH FOR REFORM lines have been established in at least 9 others More than 30 states have passed mandatory minimum sentencing laws for selected offenses. By 1982, statewide sentencing guidelines were in effect or in advanced stages of development in 6 states, and local sentencing guidelines had been developed in more than 50 jurisdictions. In Alaska, plea bargaining has been abolished by the attorney general, and many local prosecutors have banned some or all forms of plea bargaining or have regulated it closely. The sentencing reform movement has forced a reconsideration of the sentencing process and the goals of criminal sanctions. For many years the term "sentencing" produced civics-book images of high-ceilinged courtrooms, robed judges, and abashed defendants. The existence of plea negotiations was usually not acknowledged, and so defendants were often required to pretend at sentencing that their confessions had re- sulted only from remorse or contrition. The hypocrisy was blatant: every- one in the courtroom knew that most guilty pleas were induced by prosecutorial concessions or assurances, but the illusion of autonomous judicial decision making was maintained. Similarly, questions about the conflicts between utilitarian and retributive sentencing goals and the tensions between an individualized offender-oriented approach and uni- form treatment of similar offenses were ignored. The claims of sup- porters of the system that the indeterminate sentence simultaneously was just and effective in incapacitating, rehabilitating, and deterring would-be offenders meshed neatly with the interests of criminal justice system personnel to maintain the status quo. By the late 1970s it was generally acknowledged that negotiated justice is the norm in most criminal courts, and there was a growing sense that neither fairness nor crime control had resulted from existing practices. And it was also rec- ognized that "sentencing" encompasses a variety of participants, proc- esses, and conflicting goals that influence a judge's sentence. ~ Sentencing is now understood as the allocation of punishment, and among the allocators are legislators, victims, police officers, prosecutors, defense counsel, judges (and occasionally juries), parole boards and examiners, and prison administrators. The decisions of criminal justice officials at arrest, prosecution, conviction, sentencing, and parole affect the nature and amount of punishment suffered by an offender. Addi- tional choices by criminal justice system officials also can affect the punishment of an offender: the bail-release decision; assignment to a diversion program; assignment to a particular prison; loss or award of "good time" time off a sentence for good behavior by prison au- thorities; and revocation of probation or parole. At any point in the
Introduction 41 process the decisions of victims, police officers, prosecutors, and judges can terminate official proceedings or affect the decisions of others at a subsequent stage. Each of those decisions takes place in the context provided by the legislature through the criminal laws that prohibit certain behaviors and establish minimum and maximum punishments for them. And throughout this process, the decisions are affected by the conflicting normative goals and institutional interests that characterize the system. In this chapter we first describe the variety of decision processes that, together, determine whether an offender is formally punished and, if so, how much. We next survey the origins of the processes described and the purposes they serve and review some of the philosophical con- troversies involved in the sentencing decision. We conclude with a brief survey of the origins, chronology, and manifestations of the current movement to change sentencing rules and institutions. THE PROCESSES THAT CONSTITUTE SENTENCING Any effort to "reform" or even to understand sentencing must take into account the existence of the many participants and decisions that to- gether constitute "sentencing" and the conflicting values, perspectives, and interests among them. This very complexity, however, frustrates efforts to change the criminal justice process in America. VICTIMS AND WITNESSES Victims initiate criminal justice action when they decide to complain to the police. They also, subsequently, affect the likelihood of conviction and punishment through their ability and willingness to cooperate with the prosecution. Victim and witness noncooperation is a major cause of charge dismissals in the United States (Institute for Law and Social Research, 1981; Vera Institute of Justice, 1977~. According to the Na- tional Crime Survey, 56 percent of violent crimes went unreported in 1978 (including 35 percent of robberies with injury), as did 75 percent of personal crimes of theft and 64 percent of household crimes (U.S. Department of Justice, 1980b). In general, the more serious the crime and the greater the likelihood that reporting the crime will produce some result, the higher the rate of reporting, and the more likely a victim is to cooperate with the prosecution. Victims have little direct effect on the actual sentences received by convicted offenders, because they rarely are consulted by the judge or the prosecutor during plea negotiations, at trial, or during a sentencing
42 RESEARCH ON SENTENCING THE SEARCH FOR REFORM hearing. However, their role- in activating the criminal justice system is very important, and the growing awareness of the frustration and neglect of victims of crime has contributed to increased attention to their con- cerns. POLICE Police decide whom to notice, to stop, to arrest, to book, and (in some jurisdictions) to charge. Police officers have the primary authority to decide who will not be pursued by the criminal justice system. Most police patrol work involves officers in "keeping the peace" or handling threats to public order (Bittner, 1970; Wilson, 1973~. For perpetrators of minor offenses involving public disorder, family violence, and small- scale drug trafficking, Feeley (1979) asserts that "the process tof going through misdemeanor court] is the punishment" (also see Alfini, 1981; Ryan, 1980/1981~. The exercise of discretion in the police decision to arrest largely dictates the outcome in these cases. The police also possess substantial autonomy in handling serious crimes of violence and inves- tigating organized illegal activities and large property loss or damage (see Manning, 1980; Rubenstein, 1974~. Police are relatively free to decide which complaints to follow up, with what diligence and resources, and to select their means of investigation, using informants, surveillance, undercover, and "sting" operations. Police decisions to file criminal charges are subject to review by prosecutors and judges, but police decisions~to disregard crimes or to pursue only informal remedies are not subject to any further review. PROSECUTORS Prosecutors establish priorities and determine the vigor with which var- ious kinds of cases will be pursued. In the 1970s, for example, many prosecutors ceased prosecuting marijuana possession cases; in effect, those prosecutors decriminalized marijuana use in their jurisdictions. Prosecutors also exercise substantial discretion over individual cases. Prosecutors decide what charges to file or, if the police file charges, what to dismiss. Like the decisions of police officers, prosecutors' de- cisions to release without arrest, or to arrest on only minor charges, are final. Charge dismissals or unilateral reductions are not subject to in- dependent review. Prosecutors also decide whether, when, and what to negotiate and whether to recommend a particular sentence to a judge or agree to a recommendation by defense counsel. The large majority
Introduction 43 of convictions result from guilty pleas, most of which are the result of negotiations. Plea bargaining takes diverse forms. In horizontal charge bargains, a prosecutor agrees to drop several charges for an offense type if the defendant pleads guilty to the remaining charges (e.g., three burglary charges are dropped when the defendant pleads guilty to a fourth). In vertical charge bargains, a prosecutor agrees to drop the highest charge if the defendant pleads guilty to a less serious charge (e.g., a narcotics trafficking charge is dropped if the defendant pleads guilty to a narcotics possession charge, or a charge of armed robbery is dropped if the de- fendant pleads guilty to a charge of robbery). In sentence bargains, a prosecutor agrees that the defendant will receive a specific sentence in return for a guilty plea. In fact bargains, a prosecutor agrees not to introduce evidence of specific aggravating circumstances. Other plea bargaining variants involve prosecutorial agreements to recommend or not to oppose particular sentences or to dismiss charges in consideration of the defendant's cooperation in other prosecutions or investigations. Whatever form plea bargaining takes, the prosecutor and to a lesser extent the defense counsel often stand supreme. The judge sometimes has little choice but to ratify their decisions, and, constitutionally, pros- ecutors' plea-bargaining tactics are virtually immune from judicial re- view (Bordenkircher v. Hayes, 434 U.S. 357 1978. The criteria by which prosecutors screen and evaluate cases and al- locate their offices' limited resources are rarely the subject of public debate, but they pose difficult questions regarding priorities, policies, and goals. Should uniform policies or rules regulating plea bargaining be adopted? What should they be? How should such policies or rules balance considerations of the seriousness of the offense, the character- istics of the offender, the strength of evidence or the likelihood of winning a case, and its possible political repercussions? Should an office concentrate its resources on, and recommend incarceration for, chronic property offenders who may pose little physical danger to other people but who are likely to continue offending; on white-collar offenders or corrupt public officials whose nonviolent property offenses may involve large dollar losses to the public or affect confidence in the integrity of their government; or on violent offenders, particularly those who may have short prior records and who may be unlikely to repeat their of- fenses? Should considerations of whether an individual is likely to be deterred from further offending, incapacitated by incarceration, or re- habilitated by a particular sanction affect prosecutors' recommenda- tions, or should there be a uniform standard of punishment based on
44 RESEARCH ON SENTENCING THE SEARCH FOR REFORM only the seriousness of the offense? If the latter, how should a uniform standard be devised? JUDGES Judges impose sentences. They decide who goes to prison and who does not; they set the terms of nonincarcerative sentences; and (depending on whether there is a parole board and on the rules governing parole eligibility) they set minimum, maximum, or actual lengths of jail and prison terms. Where there is a parole release agency, the judge's critical decisions are who goes to prison and for what minimum and maximum terms; where there is no such agency, the judge's decision also deter- mines the actual amount of time served. Judges' decisions are affected by the diverse goals they pursue in sentencing in general and in any particular sentence. How might a judge sentence a person who is a chronic property offender compared with an offender convicted for the first time of assault? A judge who has utili- tarian goals might be more inclined to incarcerate the property offender on the basis of crime prevention concerns. Such a judge might ask whether a period of incarceration is likely to deter or rehabilitate either offender and assess the chances of recidivism in terms of the offender's prior record and personal characteristics. Or the judge might consider how many similar offenses might be averted by incarceration and weigh the cost of incapacitation against the cost of the crimes and the danger to public safety posed by the offender. A judge who has retributive goals would focus on the amount of harm done by the criminal acts and the offender's personal culpability in deciding on the sentence that is "deserved." Such a judge might give the assaulter a heavier sentence on the basis of offense seriousness. Judges' powers, however, are informally but importantly affected by the work of other court personnel. First, in jurisdictions in which sen- tence bargaining is common, often a judge's choice is whether to ratify the negotiated sentence. Second, where charge bargaining is prevalent, a judge usually accedes to proposed charge dismissals and may impose a sentence only within the constraints set by any statutory sentence provisions. Third, probation officers devote more time to investigation of the offender's circumstances and to consideration of the case than judges possibly can, and so they control the flow of information to judges. Probation officers are attached to most modern felony courts; presentence reports containing their recommendations are commonly provided to judges, and these recommendations are usually followed (Carter and Wilkins, 1967; Townsend et al., 1978~.
Introduction PAROLE BOARDS 45 Although parole boards have been abolished in some jurisdictions and in others they have lost their authority to determine release dates in the majority of states they retain control over parole release. Judges often set maximum sentences (and in some states minimums as well), but the maximum is often very long; parole boards decide who and when to release prior to sentence expiration; the conditions to which a parolee will be subject while on parole; when and why parole can be revoked; and when after revocation, if at all, an offender can be rereleased prior to the end of the maximum sentence. Parole revocation receives little attention from researchers or reformers, yet one of every five paroled prisoners is recommitted or otherwise returned to prison for violation of parole conditions within 3 years after an initial parole (Criminal Justice Research Center, 1980:668, hereafter cited as Sourcebook, 1980~. Parole boards traditionally make individualized release decisions, tak- ing account of a wide variety of offender characteristics. In establishing uniform criteria for releasing offenders, they, too, face the basic dilemma in criminal justice: How much emphasis should be placed on the seri- ousness of the conviction offense in attempting to follow the injunction to "treat like cases alike" and how much on the characteristics of the defendant, including prior record and employment status, in predicting whether the release constitutes a danger to the community? CORRECTIONS ADMINISTRATORS Corrections administrators affect the duration of imprisonment by the award, withdrawal, or denial of time off for good behavior and by their recommendations and reports to parole boards when a prisoner is being considered for early release. Corrections administrators also influence the quality of a prisoner's confinement through decisions about insti- tutional assignments and participation in various kinds of furlough pro- grams. Whether an inmate spends time in a maximum security prison, in a less restrictive minimum security facility, or in a group home in his or her hometown is almost entirely in the hands of corrections author- ities. Admission to a work, educational, or terminal furlough program is often akin to release from prison. EXECUTIVE CLEMENCY: COMMUTATIONS AND PARDONS Although pardons and similar executive release mechanisms once played a major part in prison releases (see Barnes and Teeters, 1959; Messinger, 1979), these ad hoc powers are no longer extensively used in most states.
46 LEGISLATURES RESEARCH ON SENTENCING: THE SEARCH FOR REFORM Legislative influence in sentencing is first and last: it is first because a legislature constructs and can always alter the basic statutory framework that other officials are charged to carry out; it is last because most punishments prescribed by law are not self-executing but can be realized only through other officials. If those officials behave inconsistently with the law, there is little a legislature can do. Even such seemingly au- thoritative laws as those calling for mandatory minimum sentences can be effected only through others; if prosecutors and judges choose to circumvent the law, mandatory terms will not be imposed. Sometimes a legislature's punishment decisions are definitive: for example, if in- carceration is eliminated from the sanctions available for marijuana possession, the remaining punishment decisions are of less consequence than before; if marijuana use were legalized, punishment would no longer be applicable. Sometimes statutes are drafted so broadly that they provide little guidance in individual cases. For example, the max- imum prison terms authorized for most offenses- 5 or 10 or 25 years- are so much longer than the sentences typically imposed or served that the legislative decision has little significance for the operation of the system. THE SYSTEM AS A WHOLE The operations of this complex system of criminal justice, with its net- work of multiple, overlapping, and interconnecting discretions and con- flicting goals, are not easily altered; like the operation of any complex system, they are influenced by powerful forces of tradition, institutional convenience, scarcity of resources, and self-interest. Officials who wish to circumvent or undermine a new law can usually find ways to do so; legislative changes are impositions from outside and are often resisted. A mandatory minimum sentence law, for example, can easily be avoided if the prosecutor dismisses or never files charges. A determinate sen- tencing law or sentencing guidelines can be evaded by artful charge bargaining. Parole guidelines can be evaded by lawyers who regard the applicable parole release date as an upper limit and arrange for the defendant to plead guilty to an offense bearing a maximum sentence less than that specified in the guidelines. Such reactions are foreseeable. The staffs of prosecutors' offices and the courts have institutional goals and personal interests to serve and limited resources to expend. Sometimes their personal views of justice
Introduction 47 and injustice may not easily accept legislative solutions to the crime problem. Since new laws are seldom accompanied by appropriation of funds sufficient to permit literal and wholehearted compliance with them, something must give, and that something is often compliance. The complexity of the system also often confounds reform initiatives by merely shifting the locus of decision-making power from one agency to another. The California determinate sentencing law, for instance, eliminated parole release decisions for most prisoners, but in doing so it simply shifted power over release from the parole agency to the judge and beyond the judge to the prosecutor. Although there is no evidence that prosecutors are better situated or qualified than judges or parole boards to make sentencing decisions, the greater predictability of sen- tences under the new law afforded prosecutors increased influence on sentences by means of their charging and charge dismissal decisions. Illinois's determinate sentencing law abolished parole release and al- lowed day-for-day credits for good behavior, but the prisoner receives no vested right to earned good time. And since good time can be with- drawn for misconduct, it is prison guards and officials, not the parole board, the judge, or the legislature, who effectively determine when prisoners are released. The criminal justice system's complexity makes it difficult to predict the effects of change. In later chapters, we discuss the implications of this complexity for conducting research on the determinants of sen- tencing (Chapter 2), for thinking about how to structure and implement new sentencing strategies (Chapter 3), for evaluating the impact of new sentencing systems (Chapter 4), and for anticipating and structuring the effect of changes in sentencing on prisons (Chapter 5~. In the rest of this chapter, we explore the often-conflicting goals of criminal justice sanctions; we briefly review the evolution of present American sen- tencing institutions, comparing them with their European counterparts; and we use that as background for describing the origins of the sen- tencing reform movement. TtIE GOALS OF CRIMINAL SANCTIONS Sentencing in general or the methods and consequences of change in specific sentencing practices cannot be considered without regard for the purposes and goals of the sanctioning process. Whether the allo- cation of punishment is efficient, just, or effective cannot be assessed without specifying the criteria by which to judge the outcomes. The criteria might include the expeditious disposition of cases, the reduction
48 RESEARCH ON SENTENCING THE SEARCH FOR REFORM of crime, and the rendering of morally perfect justice, however mea- sured. Such issues have moral and philosophical implications far beyond the panel's mandate or competence. While we have made no effort to resolve differences about the philosophy of punishment, we have at- tempted to be sensitive to those differences. We also suggest how various philosophical premises might differentially affect sentencing structures and the formulation of sentencing policy. The goals and purposes of punishment that are most often asserted are of two sorts: normative and functional. . NORMATIVE GOALS It is often stated that the normative goals of punishment are the utili- tarian ones of rehabilitation, incapacitation, and deterrence and the retributive one of imposing deserved punishment as an end in itself. Rehabilitation refers to the reform of the offender, and it can include special deterrence, which is the inhibiting effect of the sanction on the future behavior of the offender. Incapacitation refers to the effect of isolating identified offenders from society, thereby preventing them from committing further crimes Deterrence refers primarily to general de- terrence, which is the inhibiting effect of sanctions on others. To state these goals, however, obscures more than it enlightens. These diverse goals often conflict and, depending on their relative priority, may argue for different dispositions in particular cases. The English philosopher H. L. A. Hart (1968) provides a useful frame- work for consideration of the normative goals of punishment. Observing that debates about the philosophy of punishment are often unnecessarily confused, he proposed that debaters devote separate attention to the three distinct questions: · The general justifying aimWhat is the general justification of the social institution of punishment? · The question of liability Who is to be punished? · The question of amount How much? Hart's framework usefully isolates issues for discussion and demon- strates the potential coherence of punishment philosophies that have more than one purpose. For example, one can reasonably claim the utilitarian goal of crime prevention as the general justification of pun- ishment and still insist that retributive considerations require that pun- ishment be limited to conscious offenders and that the amount of punishment be closely proportioned to the offender's moral culpability.
Introduction 49 Thus one can consistently accept utilitarian prevention as the social justification of punishment and at the same time argue that moral con- siderations forbid the imposition of exemplary punishments. Alterna- tively, one can invoke retributive considerations to argue that liability to punishment should depend on an individual's conscious offending while the amount of punishment need not be closely proportioned to culpability but can instead be adjusted to reflect rehabilitative needs, deterrent and incapacitative considerations, and so on. Developments in philosophy over the last 20 years have importantly influenced the criminal law and sentencing. In both moral and political philosophy there has been a resurgence of interest in contractarian the- ories, often called "rights theories" (e.g., Dworkin, 1977; Gewirth, 1978; Nozick, 1973; Rawls, 1971), which pose several questions as their central concerns: What rights do individuals have? What is the source of those rights? When and under what circumstances may rights be disregarded or overridden? This new interest in rights conflicts with the older utilitarian premises of the substantive criminal law and the insti- tutions of the criminal justice system. A punishment philosophy based primarily on concern for rights is what Hart called "backward-looking," interested primarily in the moral quality of the offender's acts and the punishment that the offender deserves for them; a utilitarian punishment philosophy is "forward-looking," primarily concerned with the effects of punishment. In Hart's terms, indeterminate sentencing and its institutions are de- cidedly forward-looking. Thus, one rationale of parole is that people will remain incarcerated until they are rehabilitated. Parole release pol- icies have typically been influenced by recidivism rates and the aim of keeping in prison longer those offenders who are expected to commit additional crimes. Criminal codes commonly provide maximum sen- tences that are designed to permit substantial scope for their discre- tionary reduction by the parole board and to permit judges to indivi- dualize sentences. Both the Model Penal Code (American Law Institute, 1962) and the Study Draft of the National Commission on Reform of Federal Criminal Laws (1970) are avowedly rehabilitative in their prem- ises. Few of these practices or premises follow from a backward-looking punishment philosophy. One influential manifestation of the development of rights theories is the just deserts theory articulated by Andrew von Hirsch, who argues that the justification of punishment in individual cases rests on the of- fender's moral culpability and that the amount of punishment must be proportional to that culpability rather than being determined by utili-
50 RESEARCH ON SENTENCING: THE SEARCH FOR REFORM tartan considerations. One important corollary of a punishment juris- prudence that emphasizes desert is that equality in sentences imposed becomes a concern of the highest priority. Utilitarian considerations such as deterrent or incapacitative effects or the defendant's alleged need for rehabilitation cannot be invoked to justify unequal sentences in individual cases (see Coffee, 1978; Singer, 1979; von Hirsch, 1981~. From this perspective, sentencing "disparity," a term implying the ab- sence of uniformity, proportionality, or both, came to be seen as a primary source of injustice. While some supporters of the retributive goals of sentencing, including von Hirsch, have advocated uniform sen- tences that would generally decrease the severity of punishment and reserve the use of prison as a sanction largely for violent offenders, a retributive perspective does not necessarily imply any particular level of sanction severity. Indeed, other advocates of sentencing based on retributive goals believe that, for offenders to receive just sanctions proportionate to their crimes, sentences should be more severe and certain as well as more consistent. These are not simply theoretical arguments. Debates about the goals of sanctions and the problem of disparity have echoed throughout leg- islative chambers and judicial conferences across the country. Legisla- tors and public officials have adopted policies that express support for punishment regimes premised on just deserts and similar notions. Sec- tion 1170 of the California Penal Code, for example, begins: "The Leg- islature finds and declares that the purpose of imprisonment for crime is punishment." Section 2 of the act authorizing the establishment of the Oregon parole guidelines provides that the ranges of prison sentences contained in the parole guidelines (1977:Ch. 372, Sec. 2~: shall be designed to achieve the following objectives: (a) Punishment which is commensurate with the seriousness of the prisoner's criminal conduct; and (b) to the extent not inconsistent with paragraph (a) of this sub- section "deterrence and incapacitation]. Thus Oregon law explicitly subordinates utilitarian considerations of crime prevention to the achievement of commensurate punishment. The Minnesota Sentencing Guidelines Commission (which was established in 1978 by the legislature to design sentencing guidelines to structure judges' decisions) considered various optionslabeled just deserts, modified just deserts, incapacitation, and modified incapacitation in deciding which types of cases should go to prison. The commission
Introduction 51 adopted the modified just deserts option. Illinois's determinate sen- tencing law seeks to achieve greater proportionality between an of- fender's culpability and the sanction by increasing certain sentences through a separate schedule of "extended terms" for crimes involving "exceptionally brutal or heinous behavior indicative of wanton cruelty" (Illinois Revised Statutes, 1977:Ch. 38, Sec. 1005-8-1~. In addition, many states have adopted mandatory minimum sentencing laws that reflect, in part, concern with ensuring more certain and severe sanctions for certain serious offenses. Thus concern with equality, proportionality, and the reduction of disparity in sentencing has animated many reform efforts and has significantly affected policy debate and decisions in many jurisdictions. FUNCTIONAL GOALS David Rothman's (1980) book on the development of various criminal justice and mental health institutions in this century is entitled Con- science and Convenience: conscience because the creation and dispersion of these institutions resulted in part from the efforts of benevolent re- formers; convenience because the reformers' individual treatment ethos legitimated administrators' possession and exercise of enormous, sel- dom-reviewed powers over their patients, prisoners, and clients. Roth- man's thesis is that these "progressive" institutions survived and retained their credibility for so long because of this congruence between reform- ers' visions and administrators' needs. One need not be a cynic to accept Rothman's broad thesis. It requires no conspiracy theories to recognize that the day-to-day operation of institutions is often substantially determined by the needs of the people who manage them. Individuals operating inside institutions are moti- vated by diverse mixtures of personal, institutional, professional, and altruistic considerations. Efforts to change institutions and their oper- ations are likely to founder unless one considers the functional goals of the people who operate them. A substantial literature on the operations of criminal courts has come into being in the last 15 years (e.g., Eisenstein and Jacob, 1977; Levin, 1977), and it suggests some of the functional goals that motivate the lawyers, judges, and others who operate the courts: to achieve just results by the participants' standardsin individual cases; to maintain an acceptable guilty-plea rate in order to process cases expeditiously; to maintain amicable relations with the other participants in the process;
52 RESEARCH ON SENTENCING: THE SEARCH FOR REFORM to operate efficiently within material and personnel restraints.) The sentencing reforms of the last decade have challenged the long-standing equilibrium between the formal goals of punishment and the functional goals of those who operate the criminal justice system. The tensions between normative goals and institutional interests have been resolved in various ways in other countries and at other times in the United States. AMERICAN SENTENCING IN COMPARATIVE AND HISTORICAL PERSPECTIVE CONTEMPORARY AMERICAN SENTENCING The characteristic of the modern American criminal justice system that distinguishes it from earlier times and from other countries today is the existence of three independent centers of sentencing authority pros- ecutors, trial judges, and parole boards. The plea-barga~ned guilty plea which results from prosecutors' offices is the most usual outcome of criminal cases in America; in many jurisdictions even a judge's influence over plea bargaining is limited. At the next stage of the process, judges have sole authority to decide who goes to prison (or jail), subject to negotiated guilty-plea constraints and mandatory sentencing laws and for all convictions following trials. Then, within the constraints of max- imum and minimum sentences set by the judge, parole boards (in those jurisdictions that retain parole release) have authority to decide when prisoners are released. By law and tradition, each of these three decision makers is organizationally and politically separate. Appellate courts, the independent forum for review of administrative and judicial decisions in other contexts, have traditionally deferred to the decisions of pros- ecutors, trial judges, and parole boards. The U.S. Supreme Court has affirmed that most prosecutorial charg- ing and plea-bargaining decisions are not subject to judicial review (Bor- denkircher v. Hayes, 434 U.S. 357 t19783) and that the U.S. Parole Commission's release decisions do not present judicially cognizable sub- iSome historians and social theorists offer various hypotheses about the latent social functions of punishment. Specifically, these theorists argue that changes in the forms of social control, including the penal system, are influenced by changes in the social structure, such as: shifts in labor market conditions; the need to avoid disruptions caused by un- employed, underemployed, and unemployable people; and the need to channel workers' discontent and maintain existing power relations. These issues go far beyond our focus on the criminal justice system and are not considered in this report.
Introduction 53 stantive issues (U.S. v. Addonizio, 442 U.S. 178 [19793~. Although prac- tical considerations and deference to administrative expertise are some- times invoked as reasons for this hands-off approach, the fundamental explanation is rooted in the basic concept of separation of powers: prosecutors and parole boards are in the executive branch of government and hence not subject to certain kinds of judicial review. (Of course, that both are executive branch agencies does not mean that their pro- cesses or policies are coordinated: the two agencies have different origins, different rationales, and different constituencies.) Prosecutors Public prosecution in the United States is locally organized and highly political. Since chief prosecutors (or district attorneys) are usually elected local officials and are often ambitious politicians anxious for higher office or judgeships, their political terms of reference are primarily local and largely insulated from external controls. Serving political as well as managerial functions, chief prosecutors are in a position to affect policy through internal administrative procedures. Chief prosecutors can es- tablish supervisory, monitoring, and record-keeping systems to ensure that assistant prosecutors comply with their policies. This is possible because prosecutors' offices are usually small organizations; even in the largest cities, the professional staffs number in the few hundreds. Fur- thermore, assistant prosecutors are often young lawyers who view their entry-level, low-paying jobs as temporary apprenticeships on the way to private practice or political careers. Hence, they are anxious to dem- onstrate that they are team players and gain favorable recommendations from the chief prosecutor. Efforts to achieve more uniform statewide sentencing practices must win the support of chief prosecutors, accept the fact that local chief prosecutors can defy state policies when they wish to do so, or develop other means to restrict the prosecutors' powers. Parole Boards Parole boards are state agencies; their members are typically appointed by the governor (or, at the federal level, the President), sometimes with the advice and consent of the legislature. The relevant political consti- tuencies are at the state level, notably the governor, the legislature, and the press. Although local controversies occasionally reverberate in state capitals, a parole board, much more than a prosecutor's office, can distance itself from day-to-day politics. Parole boards set policies that apply to all state prisoners. Because hearing examiners are dependent
54 RESEARCH ON SENTENCING: THE SEARCH FOR REFORM on their superiors for assignments, performance evaluations, and pro- motions, parole boards are in a position to establish criteria for release decisions and to establish administrative systems to monitor compliance by hearing examiners. Trial Judges As a formal matter, a trial judge's sentencing decisions are constrained only by conscience and by the usually wide range of sentences authorized by the applicable criminal statutes. Trial judges, unlike parole hearing examiners and assistant prosecutors, are almost immune from organi- zational controls: their salaries are fixed by law; their terms are usually long; impeachment is a difficult and rarely invoked procedure. Only in periods immediately preceding a campaign for reelection or retention need judges pay particular heed to public opinion or to the political ramifications of their decisions. In the United States, judges value their independence and are not easily regulated. No general right of sentence appeal exists under federal law, and there is reason to doubt whether meaningful review is available in those states that have established systems of appellate sentence review (Sam- uelson, 1977; Zeisel and Diamond, 1977~. The Eighth Amendment of the U.S. Constitution, as currently interpreted, does not empower ap- pellate judges in most cases to decide whether lawful sentences imposed by trial judges are excessively severe (Estelle v. Rummel, 445 U.S. 263 t19803; Hutto v. Davis, 445 U.S. 947 t19803~. Consequently, appellate review of sentences is available only when a legislature, state supreme court, or state constitution has expressly established such a system. As a practical matter, judges are subject to organizational constraints. Judges in administrative positions can place some pressures on their colleagues by threatening to assign them to unpopular courts or dockets. Trial judges may be constrained by bargains negotiated by counsel. Charge bargains may reduce the maximum sentence allowable to a level below that which the judge believes appropriate. (In most jurisdictions, for example, a negotiated misdemeanor plea to a felony charge will make a state prison sentence impossible.) Sentence bargains require that a judge accept the sentence negotiated or let the defendant withdraw a guilty plea; although a judge is not required to accept sentence bar- gains, a proposal that both the prosecutor and the defense counsel believe to be appropriate is unlikely to be rejected. And, although a trial judge's sentencing decisions are seldom subject to meaningful re- view by appellate judges, the intended length of a prison sentence is subject to reconsideration by the parole board.
Introduction 55 An additional practical consideration for judges, particularly when sentencing an offender convicted of a heinous crime, is the increased public pressure that results from the judge's heightened visibility through media coverage. The indirect effect on judicial behavior of both peer pressure and prevailing community standards of justice can be seen in differences in the sentences imposed for similar offenses by judges who ride a circuit and sentence cases in more than one district (Gibson, 1978b) EUROPEAN SENTENCING SYSTEMS The U.S. criminal justice system is the product of U.S. history. Despite diversity among European criminal justice systems, three features of the U.S. sentencing process distinguish it from many of those in Europe: the importance of plea bargaining and the prominence of the prosecutor; the inflation of prison sentence lengths in anticipation of their later reduction by parole boards; and the absence of meaningful appellate sentence review. A common theme in all these features is the relatively greater detachment from politics in European sentencing. Plea Bargaining In comparison with U.S. practice, plea bargaining is not as important a feature of the criminal process in England, France, the Scandinavian countries, or west Germany (Andenaes, 1983; Jackson, 1972; Thomas, 1979; Weigend, 1980~.2 Defendants who plead guilty receive leniency in some of these countries, but this leniency is available to all who plead guilty, is modest in amount, and does not result from the negotiations of lawyers. Although the organization of public prosecution varies from country to country, a common feature is that public prosecutors in Western Europe are not elected officials. In France and West Germany, for example, public prosecutors are appointed officials and are career civil servants. In England, there are no public prosecutors except for the Director of Public Prosecutions (a central government official) and his 2 there have in recent years been assertions that something akin to plea bargaining exists in several European criminal justice systems (see, for example, Baldwin and McConville t1977] concerning England and Goldstein and Marcus t1977] concerning France, Italy, and West Germany). To the extent that plea-bargaining analogues exist in those countries, they are substantially less visible than in the United States.
56 RESEARCH ON SENTENCING: THE SEARCH FOR REFORM small staff, who prosecute the few most serious cases; the function of public prosecution is largely performed by police solicitors and private lawyers who are appointed on a case-by-case basis to represent the Crown. It is uncertain why plea bargaining is of limited importance in Western European legal systems. Smaller case loads, stronger professional norms against negotiation, removal of the prosecutor from local politics, and the organizational structure of prosecutors' offices have been suggested as contributing factors. Parole By 1930 every U.S. state had created a parole board (Rothman, 1980), and parole subsequently became the primary form of prison release. In 1976, for example, more than 70 percent of persons released from U.S. prisons were released on parole (Sourcebook, 1980:661~. By contrast, release on parole remains the exception, not the norm, in most European countries. The English Parole Board, for example, was not created until 1967, and most prisoners in England are not released on parole, because sentences imposed are much shorter than in America (seldom longer than 2-3 years); remission of sentence (time off for good behavior) reduces maximum sentences by a third; and prisoners are not eligible for parole until they have served one-third of the term imposed (Jackson, 1972~. In Norway, prisoners are eligible for parole only after serving one-half the sentence imposed, and sentences seldom exceed 2 years. For a 2-year sentence, remission shortens the time served by 8 months, and so parole release would reduce that sentence by at most another 4 months (Andenaes, 1983~. In West Germany, where local panels of judges have authority to release prisoners early, the scope of parole release is similarly narrow (Weigend, 1983~. The lesser reliance on parole authorities to determine the lengths of prison sentences and to shorten the nominal sentences declared by judges has at least two important consequences in Europe: the judge is much more the central figure in sentencing, and sentences that are imposed are very close to the sentences actually served. European judges are seldom presented with negotiated proposals for disposition of cases, and they are not as constrained as American judges by personal, institu- tional, and work-group considerations to delegate or share their power. Moreover, they have no need to increase the sentences they impose to offset the amounts by which parole boards will routinely shorten them. Thus the sentences imposed in Europe are shorter than those in the
Introduction 57 United States; however, there is conflicting literature on whether prison sentences actually served in various European countries are shorter than those actually served in the United States (see Advisory Council on the Penal System, 1978:Appendix C). Appellate Sentence Review Appellate review of sentences, which is rare in America, is common in Europe. In England and Norway, appellate judges have established case law standards for sentences, and aggrieved defendants can seek review of the appropriateness of their sentences. This establishes meaningful constraints on trial judges' decisions and provides a mechanism for re- ducing the incidence and extent of unwanted disparities in sentences. It is possible that the failure of appellate sentence review to take hold in the United States resulted in part from the widespread adoption of parole. When parole boards exercised authority over release, judges' sentences were of secondary importance: if parole boards could release a prisoner when they saw fit, case law standards for nominal sentences would have little real meaning. Moreover, if sentences were too severe, parole boards could rectify matters. Thus, in a sense, parole boards became mechanisms for review of sentences in the United States and appellate courts did not. Plea bargaining also has provided a practical impediment to appellate sentence review. Defendants who plead guilty in connection with a sentence bargain are not well situated later to object to the sentence received. Although charge bargains need not severely limit judges' sentencing options, the perception that the defendant has voluntarily pled guilty knowing that he is vulnerable to any lawful sen- tence may also have impeded the development of appellate sentence review. THE DEVEOPMENT OF AMERICAN SENTENCING STRUCTURES AND EFFORTS TO REFORM THEM The ideological bases and institutional structures of sentencing in Amer- ica have changed substantially since colonial days. These changes reflect and parallel a series of reforms throughout the criminal justice system that have contributed to its contemporary form. The main characteristics of sentencing goals and practices in several periods are briefly noted here to illustrate the relatively recent origin of current sentencing prac- tices.
58 Colonial America RESEARCH ON SENTENCING: THE SEARCH FOR REFORM Prior to the American revolution, American criminal justice practices resembled those in Britain. The criminal law was harsh. In a society with limited communications, no police (only a voluntary nightwatch system), few jails, and rudimentary record-keeping, crime control efforts were designed principally to deter would-be offenders by imposing pun- ishments so terrible that few would dare break the law. More than 350 offenses were punishable by capital punishment (Hartung, 1952), and less serious offenses were subject to sentences of corporal punishment, fines, or banishment. Penal incarceration was a rarely applied sanction, although sometimes offenders were held in jail to prevent flight pending trial or until their execution. 1790-1820: Reform of the Law The founding of the republic and the era of political ferment that fol- lowed it led to major reforms in American sentencing ideas and practices that increasingly diverged from those in Europe. The principal one was a movement away from capital punishment to imprisonment a mea- sured time of exclusion from society for criminal offenses. This change had both ideological and practical support. The number of capital crimes was greatly reduced, in part because such harsh punishment had failed to deter crime and had resulted, instead, in juries that more and more frequently refused to convict. At the same time, Enlightenment ideas, particularly those of the philosopher Beccaria, gained favor. He argued that greater certainty of punishment, rather than an emphasis on se- verity, would more effectively deter crime. Many legislatures came to the conclusion that wholesale reliance on capital punishment was self- defeating. More lenient codes prescribing fixed periods of imprisonment were recognized as both more humane and appropriate for a new nation with a populist government. They also represented a more promising form of crime control, since they offered the prospect of reforming criminals, not merely punishing them. By 1820 many state legislatures had drafted new criminal codes that prescribed fixed sentences to match the seriousness of the offense. Sen- tences were long: 40 years for murder, 20 years for arson, and 10 years for burglary were common (Rothman, 1981~. Yet, in contrast to capital punishment, imprisonment seemed more humane and more certain, and therefore more effective.
Introduction 1820-1900: Emphasis on Prisons 59 Although reformers at the beginning of the nineteenth century had hoped to find an answer to the crime problem through sentencing reform and criminal codes, for much of the rest of the century attention was shifted from sentences and the likelihood of their imposition to the penitentiary as nineteenth century prisons were called and the ex- perience of punishment as the critical mechanism of crime control. By midcentury penitentiaries had been built in many states. The rea- sons for and significance of this development have been variously in- terpreted. It has been argued that they signified the triumph of Enlight- enment ideals and the rejection of inhumane forms of corporal and capital punishment (McKelvey, 1977~; an indictment of a disordered society by Jacksonian reformers nostalgic for a stable but vanished co- lonial society (Rothman, 1971~; and a precursor, or metaphor, for a surveillant disciplinary society (Foucault, 1978~. Whatever their origins, prisons were built and prisoners were sentenced to long terms of incar- ceration fixed by judges. Faith in the rehabilitative potential of the penitentiary affected sen- tencing procedures in several ways. Initially it reinforced the legitimacy of uniform fixed dispositions based exclusively on the crime itself: the penitentiary was viewed as a panacea for all types of deviant behavior. In addition, the promise of offender reform led to a shift in emphasis from the traditional principles of deterrence and retribution to concern with rehabilitation. Reformers expected imprisonment not only to dis- suade would-be and sentenced offenders from pursuing criminal acts, but also to alter the offenders themselves. Until about 1850 there was no sense of conflict among the purposes of sentencing, since reformers viewed the penitentiary and its regimen as simultaneously deterring offenders from further criminality, incapacitating them, and rehabili- tating them through fixed sentences of long duration. All agreed that sentences should be lengthy: for the conservatives, to deter offenders; for reformers, to allow time for rehabilitation to occur. By the 1860s, when penitentiaries were only a few decades old, their defects had become clear, and a new wave of reformers set out to improve the institutions and save prisoners from them (Rothman, 1980~. To cope with the crowding, brutality, and disorder of prisons, "good time" was introduced, giving wardens a mechanism of control other than corporal punishment; the use of governors' pardoning powers greatly increased; and probation programs first appeared, to keep minor of- fenders out of institutions. More important, from the standpoint of sentencing, was the arrival of indeterminate sentences under which
60 RESEARCH ON SENTENCING: THE SEARCH FOR REFORM corrections officials could decide when prisoners would be released based on their rehabilitation initiated at New York State's Elmira Refor- matory when that institution opened in 1877. 1900-1970: The Rehabilitative Model Among the accomplishments of the Progressive era reformers between 1900 and 1930 was the all-but-universal adoption of indeterminate sen- tencing based on rehabilitation of the offender and the creation of parole boards with discretion over release decisions. Reformers asserted that the causes of criminal behavior were different for each offender, and therefore they sought to individualize criminal justice procedures. The medical model prevailed as the offender came to be viewed as sicker in need of treatment and the prescription had to be tailored to each offender's illness. Thus the determinants of sentences were shifted away from the offense to the offender from what he had done to who he was. And a decision about the offender's "cure" could only be made by a professional after treatment, not at the time of commitment.3 Just as historians differ in their accounts of the origins of the prisons, they also disagree on the reasons for the creation of the rehabilitative penal system with its vast discretions and minimal accountability. Roth- man (1980) believes modern institutions resulted from a congruence between the humanitarian impulses of benevolent reformers and the instrumental convenience that rehabilitative discretions afforded offi- cials; other historians disagree with this explanation (see, e.g., Mennel, 1973; Platt, 1977; Walker, 1979, 1980~. There is agreement, however, that modern American criminal justice practices are of very recent or- . . gln. Between 1930 and 1970 there were changes in criminal law and pro- cedure, but these had little impact on the ways criminal offenders were handled. The American Law Institute worked on the Model Penal Code throughout the l950s and completed its work in 1962; in the following years many state legislatures adopted derivative criminal codes. The Model Penal Code and the proposed criminal code of the National Commission on Reform of Federal Criminal Laws (1970) exemplify the stability of the attitudes of the legal establishment toward the criminal justice system between 1930 and 1970. Both codes had avowedly re- habilitative outlooks; both perpetuated the indeterminate sentence and 3 During this period, too, juvenile courts were established with an even stronger emphasis on rehabilitation, and the scope and claims of probation were expanded.
Introduction 61 the parole board; both granted immense discretion to judges and parole boards. Although the Model Penal Code also dealt with some matters of criminal procedure, the major procedural changes of the 1960s em- anated from the Warren Court, which issued a series of opinions that applied to the states most of the criminal procedure provisions the Bill of Rights imposed on the federal system. THE CURRENT SENTENCING REFORM MOVEMENT Since 1970 numerous sweeping reforms of the criminal justice system have been undertaken. After nearly 40 years of stability, the indeter- minate sentencing system has been abruptly rejected in state after state. Between 1975 and January 1982, 11 states abolished parole release for the majority of offenders,4 17 states established administrative rules for release decisions (e.g., parole guidelines),5 more than 30 states passed mandatory minimum sentence laws, and, in almost every state, judges experimented with guidelines to structure their own sentencing deci- sions. The history of sentencing reform in the 1970s is yet to be written; time must pass before historians will be able to understand this frenzy of activity. However, we note several recent developments that either influenced recent sentencing initiatives or were themselves symptomatic of the same social forces that caused those initiatives. Prison Uprisings and the Civil Rights Movement Prison uprisings in the late 1960s, at the Tombs in New York City and state prisons in Florida' Indiana, New York (Attica), and elsewhere, demonstrated several things: prisoners were deeply discontented; they were disproportionately black and brown; rehabilitation rhetoric was, in many prisons, no more than rhetoric. The civil rights movement had reached inside prison walls by the mid-1960s. A large number of suc- cessful prisoners' rights cases in the federal courts gave prisoners an opportunity to be heard outside the prison, and they were listened to. The first influential book calling for rejection of the indeterminate sen- 4Alaska, Arizona, California, Colorado, Connecticut, Illinois, Indiana, Maine (which was first), Minnesota, New Mexico, and North Carolina. s Florida, Georgia, Hawaii, Louisiana, Maryland, Michigan, Minnesota, New York, Ohio, Oregon, Rhode Island, South Carolina, Washington, West Virginia, and Wisconsin (Uniform Parole Reports, 1980~.
62 RESEARCH ON SENTENCING: THE SEARCH FOR REFORM fencing system, Struggle for Justice (American Friends Service Com- mittee, 1971), began by quoting the demands of a group of prisoners in the Tombs. Prisoners, their complaints, and what goes on inside prisons were major catalysts of modern sentencing reform (see, for example, Zimring, 1983~. The Philosophy of Rights and Control of Discretion Modern criminal law and the indeterminate sentencing system took shape in an era when utilitarianism was the philosophical perspective most commonly brought to bear on public policy matters. The Model Penal Code and the more than 30 state codes emulating it were avowedly utilitarian in premises and outlook. Retribution was expressly de- nounced, and rehabilitation was endorsed as the primary goal of pun- ishment. Williams in England (e.g., 1961) and Wechsler in the United States, among the foremost scholars of the criminal law, were both utilitarians primarily concerned with crime prevention as the foremost goal of the criminal law. The indeterminate sentencing system overtly focused primarily on offenders and their amenability to treatment rather than on their offenses. Until the 1960s few principled objections were raised to indeterminacy, to the rehabilitative ideal, and to the primacy of utilitarianism in the philosophy of punishment (but see Allen, 1959; Hart, 1968~. Although it was anticipated by Hart's Punishment and Responsibility (1968), the recent challenge to utilitarianism was exemplified by Rawls's A Theory of Justice (1971), which was followed by a series of powerful antiutilitarian books (e.g., Dworkin, 1978; Gewirth, 1978; Nozick, 1974~. Grossly oversimplified, the primary complaint of antiutilitarians was that utilitarianism does not adequately address justice for individuals. The advocates of rights theories are primarily concerned with the rights of individuals and the constraints that those rights place on the assertion of state power. These theories require that criminal responsibility should be predicated on moral culpability (which is not necessarily a require- ment of a utilitarian jurisprudence) and that punishment should be pri- marily retributive in aim and proportional in amount to an offender's culpability. Parallel trends appeared in writings on sentencing reforms. Von Hirsch's Doing Justice (1976) endorsed retribution, or just deserts, and proposed that sentencing be guided by detailed sentencing criteria relating largely to a defendant's moral culpability (see also The Twentieth Century Fund, 1976~. The indeterminate sentencing system with its vast range
Introduction 63 of discretion conflicted with rights theories. The principal theoretical rival of retribution, the modified utilitarianism of Norval Morris (1974) and the American Bar Association Criminal Justice Standards Project (1980), gives greater weight to crime preventive strategies in sentencing while still insisting that retributive concerns establish meaningful limits on the amount of punishment that can be imposed in individual cases. These philosophical developments, along with the distrust for authority characteristic of the 1960s and 1970s, gave important support to pro- posals for changing the extensive discretion of judges and parole ad- ministrators in deciding who went to prison and how long they stayed there. Demand for Accountability Throughout the legal system in the 1960s and 1970s, there was a move- ment for increased accountability in official decision making. Judicial decisions in many contexts required that public officials indicate reasons for decisions and give adversely affected individuals an opportunity to defend themselves and to dispute material allegations or evidence. Prison administrators, for example, began to be required to publish their dis- ciplinary rules and to give prisoners an opportunity to defend themselves against rule violation charges (Wolff v. McDonnell, 418 U.S. 539 t19743~. Other cases established procedural requirements to be observed before offenders could have probation revoked (Gagnon v. Scarpelli, 411 U.S. 778 t19734), have parole revoked (Morrissey v. Brewer, 408 U.S. 471 t1972~), or, in some states, be denied parole (Greenholtz v. Inmates, 442 U.S. 1 t1979~. Parole came under vigorous attack on the grounds that parole release decisions lacked standards and, hence, accountability (Davis, 1969, 1976~. These attacks, coupled with the general movement for increased ac- countability in official decision making, led to studies of whether parole boards followed implicit criteria in parole release decisions and whether those criteria could be expressed in decision rules. A major long-term project demonstrated the feasibility of detailed published criteria for parole release decisions (Gottfredson et al., 1978~. The U.S. Parole Commission adopted parole guidelines based on that research in 1974, and several state parole boards soon followed. The research team that had developed the first parole guidelines later explored the feasibility of using that same method to develop sentencing guidelines for judges (Wilkins et al., 1978~. Empirically derived sentencing guidelines projects have since been undertaken in more than 50 jurisdictions.
64 RESEARCH ON SENTENCING: THE SEARCH FOR REFORM Rejection of Rehabilitation After three-quarters of a century of intellectual hegemony, the reha- bilitative ideal began to crumble in the 1960s (see, e.g., Allen, 1959) and collapsed in the 1970s, primarily because of two objections. First, there was substantial evidence that rehabilitative programs do not de- monstrably and substantially reduce the later criminality of their clients (Brody, 1976; Greenberg, 1975; Lipton et al., 1975; Robison and Smith, 1971~. In reviewing the most comprehensive of the reviews, by Lipton et al., a National Research Council panel concluded that the authors were "reasonably accurate and fair in their appraisal of the rehabilitation literature," but it noted two significant limitations (Sechrest et al., 1979:5- 6~: . . . first, inferences about the integrity of the treatments analyzed were uncertain and the interventions involved were generally weak; second, there are suggestions to be found concerning successful rehabilitation efforts that qualify the Lipton et al.] conclusion that "nothing works." Nonetheless, the uncritical support that rehabilitative programs received before 1970 has been displaced by a deep skepticism. If rehabilitative programs didn't "work," the claim that prisoners could be released when they were rehabilitated lost much of its credibility. Second, there were objections to the rehabilitative ideal because the extensive discretion characterizing sentencing and corrections programs were often abused. This was a powerful criticism in a period of wide- spread distrust of authority and acute sensitivity to the reality even to the appearance- of racial discrimination and arbitrary decisions. Disparity and Discrimination No doubt influenced by prison uprisings, rights theories, increasing em- phasis on accountability, and decreasing emphasis on rehabilitation (as well as by the widespread availability of computers for social science research, which made elaborate multivariate analyses possible), re- searchers undertook many statistical simulation studies to determine whether there was substantial evidence of disparity and racial and class discrimination in sentencing. The findings on discrimination were mixed (see Hagan, 1974; Hagan and Bumiller, Volume II), but on disparity they were striking: the research could account for only a small amount of the variation in sentences imposed by judges (Diamond and Zeisel, 1975; Institute for Law and Social Research, 1981; Partridge and Eld-
Introduction 65 ridge, 1974; Rich et al., 1981; Tiffany et al., 1975). While such disparities were not particularly troublesome when the rehabilitative ideal pre- scribed sentence variation based on offender characteristics, without that justification the evidence of substantial, unexplained, and often apparently unwarranted disparities in sentencing became a primary ra- tionale for proposals to structure, confine, and monitor the discretion of trial judges and parole boards. Crime Control Official rates of reported and recorded crime have increased almost steadily since the early 1960s, and there have been numerous criticisms of the effectiveness of the criminal justice system (van den Haag, 1975; Wilson, 1975~. Accordingly, increased attention and federal research funding were given to nonrehabilitative sentencing strategies like in- capacitation and deterrence and to projects concerned with "career criminals." (Federal funds were used to establish "career criminal" pros- ecution units across the United States.) With this heightened emphasis on crime control, people on the political right joined people on the political left, concerned about discrimination, disparity, and accounta- bility, in a combined assault on the institutions of the indeterminate sentencing system. The indeterminate sentencing system that had been all-but-universally supported through most of the 1960s had few defenders left by the late 1970s. By then, a broad consensus in favor of change had formed among the political left and right, law enforcement agencies and prisoners' groups, and reformers and criminal justice systems officials. There was rather less agreement on what should replace indeterminate sentencing. Unlike previous waves of reform, the current movement is characterized by a cacophony of voices disagreeing over the purposes and justification for determinate sentences and over whose discretion should be curtailed. The Role of Social Science Research in Sentencing Changes Social science research tends to percolate into the policy arena and subtly alter the ways policy makers and citizens think about issues. The results of sentencing research have followed this pattern. Sentencing policy changes have been influenced by social science research findings and have themselves precipitated a substantial body of research (see Weiss, 1981~. On the subject of rehabilitation, research pulling together the many assessments of rehabilitation programs tended to confirm what some
66 RESEARCH ON SENTENCING: THE SEARCH FOR REFORM observers had suspected about those programs and removed the basic justifications for indeterminate sentencing policies. While research was not the driving force behind the effort to reduce unwarranted disparity in sentencing, the availability of data confirming the impression of wide- spread disparities and suggesting the existence of discrimination by race and socioeconomic status contributed to the quest for policies to limit judicial discretion. Similarly, as the goals of sentencing shifted toward deterrence and incapacitation, researchers sought to assess the magni- tude of the deterrent and incapacitative effects of various sanctioning policies on crime rates, as well as to assess the career criminal programs designed to implement those goals. The effect of research on the development of sentencing guidelines has been direct and instrumental. The tradition of research on predictors of parole success, which dates back to the 1920s, was stimulated in the 1960s by the availability of multivariate statistical techniques facilitating better identification of the predictors of success. Corrections authorities and researchers worked cooperatively to develop improved prediction instruments to help parole authorities structure discretionary release procedures. After successfully demonstrating the feasibility of the U.S. Parole Commission's guidelines, researchers applied the same tech- niques to modeling the factors associated with judicial sentencing de- cisions. At a time when judicial decision making was under strong attack, "descriptive" guidelines that would articulate and rationalize existing sentencing policies had an appeal that led to their widespread dissem- ination and adoption. More recently, social science modeling methods and data on past sentencing practice were used by the Minnesota Sen- tencing Guidelines Commission in developing "prescriptive" guidelines that explicitly altered existing policies and practices. In sum, research on sentencing has contributed to the general dis- cussion of sentencing policy in several ways: it challenged prevailing doctrines and assumptions; documented emerging beliefs and thereby gave them added impetus; specified the nature and extent of bias in the system; strengthened the case for change; provided a technology for individual decision making; legitimated alternative rationales for pun- ishment; encouraged the search for alternative policies while providing ammunition for a critique of these options; and provided a conceptual language for the policy discourse. Several groups have had important roles in the diffusion of research into the policy arena. Legal scholars, blue-ribbon commissions, and crusading or popular authors have all drawn on social science research to support policy recommendations. Federal agencies, particularly the Law Enforcement Assistance Administration and the National Institute
Introduction 67 of Justice (NIJ), have supported empirical research on sentencing. In the late 1970s, NIJ made sentencing a priority area for research funding. It has both funded and disseminated the results of some policy-relevant research and evaluation studies, which include the development and testing of various kinds of descriptive sentencing guidelines and assess- ments of the impact of determinate sentencing laws. SCOPE OF THIS REPORT The policy and research developments in sentencing in the past decade; the variety of proposals for changing sentencing practices that are pend- ing in Congress, state legislatures, and administrative agencies; and the newly emerging data on the impact of recently adopted policy innova- tions suggested the need for an interim review of empirical findings about what may be termed the sentencing reform movement. This report is designed to meet that need. It reviews the findings and methodologies of several bodies of sentencing research, points to the ways social science research has informed policy making, and suggests future avenues of inquiry and improved methods for research and for formulating sen- tencing policy. The historical, comparative, and descriptive review presented in this chapter provides some perspective for what follows. Chapter 2 reviews the methods and findings of empirical research on the determinants of sentences, with particular attention to research on discrimination and disparity. Chapter 3 considers the development and formulation of sen- tencing policy. It reviews the variety of approaches taken to develop systems of structured discretion for greater evenhandedness in sentenc- ing decisions. Chapter 4 reviews the evaluation literature that has at- tempted to determine the effects of various innovations on sentencing outcomes and officials' behavior. Chapter 5 considers the relationship between sentencing policies and prison populations. It examines the implications of changes in sentencing practices on the size, conditions, and management of prison populations, as well as the problems of pro- jecting and controlling the size of those populations. Chapter 6 sets out directions for future research. It is important to make clear what is not included in this volume and the reasons for these omissions. Although we recognize the importance of the crime control effects of sentencing, we do not attempt to account for the effects of sentencing on offender rehabilitation or on deterrence and incapacitation; these subjects have recently been considered by other National Research Council panels (Blumstein et al., 1978; Martin et al., 1981; Sechrest et al., 1979~. Limitations of time and expertise
68 RESEARCH ON SENTENCING: THE SEARCH FOR REFORM and the absence of appropriate research led us to exclude several other subjects from its broad conception of sentencing. For example, we did not examine police arrest and charging practices, bail-setting and pretrial release policies, or the role of public opinion in the establishment of sentencing policies. Nor did we examine research or policy concerning the sentencing of juveniles, even though young offenders are dispro- portionately arrested for serious offenses against persons. We deter- mined that despite the overlap of the juvenile and criminal justice sys- tems, legal and organizational issues raised by an inquiry into the former system would detract from a more intensive focus on the latter. Finally, although we did devote some attention to programs that provide alter- natives to incarceration, we have not surveyed those programs or ex- haustively reviewed the relevant evaluation literature.
