Printer Friendly

Procedural rights at sentencing.


In determining which constitutional procedural rights apply at sentencing, courts have distinguished between mandatory and discretionary sentencing systems. For mandatory systems--systems that limit sentencing factors and specify particular punishments based on particular facts--defendants enjoy important rights including the right to a jury, the right to proof beyond a reasonable doubt, the right to notice of potential sentencing aggravators, and the right not to be sentenced based on ex post facto laws. By contrast, for discretionary systems--systems that leave the determination of sentencing factors and how much punishment to impose based on particular facts to the judge's discretion--defendants do not enjoy these protections. This Article challenges this discrepancy. It argues that, given the rationales underlying each of these rights, there is equal reason to apply these rights in discretionary sentencing systems as in mandatory ones. As it explains, procedural rights regulate the means by which facts are found and the manner in which courts use those facts, and consequently are critical to discretionary systems. Just as in mandatory sentencing systems, judges in discretionary systems must make factual findings to determine the appropriate sentence to impose. This Article argues that the various justifications for providing fewer procedures in discretionary schemes are based on misconceptions about the nature of discretion at sentencing and inaccurate historical analysis.


In recent years, the Supreme Court has greatly expanded the scope of constitutional protections at sentencing. At the beginning of the twentieth century, courts enforced few, if any, of the criminal procedural rights prescribed by the Constitution during a sentencing proceeding. They applied those rights only at trial to determine the guilt or innocence of a defendant. But in the last half-century, courts have enforced a number of procedural protections at sentencing. (1) Most notably, in a series of recent cases, the Supreme Court has said that defendants at sentencing have the right to have facts that increase a sentence found by a jury, (2) the right to have those facts proven beyond a reasonable doubt, (3) the right to notice of the facts that may increase their sentence, (4) and the right not to be sentenced under retroactive laws that are harsher than the ones in effect at the time of the commission of the offense. (5)

But the Court has limited the application of these procedural protections to mandatory sentencing systems--systems that specify a particular punishment based on particular facts. By contrast, defendants do not enjoy these procedural protections in discretionary sentencing systems--systems that leave essentially all decisions affecting the amount of punishment to the judge's discretion. In discretionary schemes, judges, not juries, may find the facts that affect the sentence; the government need not prove those facts beyond a reasonable doubt; defendants are not entitled to notice of the considerations that may increase their sentences; and their sentences may be increased under retroactive laws.

This Article argues that it is inappropriate to treat procedural rights differently depending on judicial sentencing discretion. As it explains, if one examines the rationales underlying each of these rights, it is apparent that the rights should apply equally in discretionary and mandatory sentencing schemes. There is no principled reason to require the observance of these procedures in mandatory systems, but not in discretionary systems. Indeed, the rationales behind many of these rights suggest that their enforcement is more important in discretionary systems.

Imposing different procedural requirements for mandatory and discretionary sentencing schemes has also led to confusion. No sentencing system in effect today is completely mandatory or completely discretionary; instead, they are hybrid mandatory and discretionary schemes. (6) The consequence has been uncertainty over precisely how much discretion a sentencing system must afford judges before a defendant is no longer entitled to procedural rights. Adding to the confusion, the Court has applied different tests to determine whether a sentencing scheme is mandatory or discretionary for different doctrines. For example, the Supreme Court has treated the now-advisory Federal Sentencing Guidelines system as discretionary in deciding whether to enforce the jury trial right, the proof beyond a reasonable doubt requirement, and the due process right to notice; but it treated the same system as mandatory in determining whether to apply the Ex Post Facto Clause. (7) Looking to the rationale for each right in order to determine its applicability at sentencing avoids that confusion, and it provides a more logical method to determine which rights ought to apply at sentencing.

This Article proceeds in four parts. Part I gives a brief historical overview of sentencing in the American criminal justice system, including the shift from discretionary to mandatory sentencing, the recognition of various procedural rights at sentencing, and the Court's differentiation between discretionary and mandatory sentencing systems in the wake of the recent Supreme Court cases Blakely v. Washington (8) and United States v. Booker. (9) Part II provides a nuanced account of judicial discretion at sentencing, including a discussion of how discretionary sentencing systems differ from mandatory schemes.

Part III builds the positive case for applying the same procedural protections in discretionary as in mandatory sentencing schemes. It does so by focusing on four procedural rights--the right to a jury, the right to have sentencing facts proven beyond a reasonable doubt, the due process right to notice, and the applicability of the Ex Post Facto Clause--that have been treated differently in mandatory and discretionary systems. It discusses the rationale underlying each of the rights, and it explains why, in light of those rationales, conditioning the enforceability of each right on the type of sentencing system is unsupportable.

Part IV addresses the three main justifications that have been given for extending procedural protections to mandatory but not discretionary sentencing schemes, and it explains why those arguments are unconvincing. First, it refutes the various claims that there is something inherent in the nature of judicial discretion that makes procedural protections unnecessary. Second, Part IV addresses the argument that additional procedural protections are necessary only in mandatory schemes because defendants have an expectation not to receive a sentence increase in a mandatory scheme without findings of aggravating facts. It explains that, because judges under discretionary schemes must make rational distinctions among defendants, they cannot sentence all defendants to the maximum allowable sentence. Accordingly, defendants in discretionary schemes also have an expectation not to receive high sentences without additional fact findings. Finally, Part IV challenges the Court's history-based conclusion that mandatory sentencing laws effectively created new crimes and, accordingly, trigger all of the procedural protections that apply at trial. It explains that the Court's conclusion is not supported by the historical evidence and that, even if it were supported by history, the Court's decisions are not justified because historical practice has generally not informed the development of procedural law at sentencing.


For nearly a hundred years, beginning in the late nineteenth century, discretionary sentencing was the norm in the United States. (10) In a typical discretionary sentencing system, after a defendant had been convicted, the judge conducted a separate sentencing hearing at which she imposed a sentence based on her assessment of the offender and the circumstances under which the crime was committed. (11) The judge's discretion was limited by the statutory sentencing range--that is, she could not impose a sentence above the statutory maximum sentence for the crime of conviction, nor a sentence below any applicable statutory minimum sentence--but those ranges were ordinarily quite broad. In selecting a sentence within the statutory range, judges often considered a wide range of factors including the defendant's criminal history, employment history, family ties, educational level, military service, charitable activities, and age; harm caused by the criminal act; and the defendant's motive. (12)

Historically, discretionary sentencing systems gave significant authority to sentencing judges. Whether to consider any of these factors and the weight to accord to various factors were decisions left almost entirely to the discretion of individual judges. (13) Indeed, the perception was that sentencing courts could impose sentence for any reason, or no reason at all. (14) But in reality there were real limits on the discretion of sentencing courts. Even in systems conferring the broadest discretion, sentencing judges could not base sentences on false information (15) or on an improper factor, such as a defendant's race, gender, or religion. (16) Moreover, sentencing judges in discretionary schemes could not impose the statutory maximum punishment based solely on the crime of conviction, but were required to individualize sentences based on the facts or circumstances of the defendant's offense and personal background. (17)

Enforcement of these limitations, however, varied from jurisdiction to jurisdiction. Some state jurisdictions permitted appellate review of sentencing, albeit under deferential standards. (18) Other systems, however, provided little appellate review. Review of sentencing was particularly limited in the federal system. An 1891 statute essentially removed jurisdiction to review sentencing decisions from the appellate courts, (19) leading the judiciary to conclude that "once it is determined that a sentence is within the limitations set forth in the statute under which it is imposed, appellate review is at an end." (20) Further limiting appellate review in the federal system was the fact that sentencing courts were not required to provide the reasons for the sentences that they imposed. (21) When lower courts did not explain their sentences, appellate courts had little or no ability to determine whether the sentencing judges exercised their discretion in an appropriate way.

Discretionary sentencing systems faced significant criticism by the mid-twentieth century. Chief among those criticisms was that the vast discretion left to sentencing judges, combined with the virtual unreviewability of the sentences, led to disparate sentences for similarly situated defendants. (22) Discretionary systems not only conferred on judges the flexibility to account for factual differences among offenders, they also gave judges the authority to decide what sorts of differences ought to matter, and how much they ought to matter. Because different judges had different ideas about the appropriate punishment for a particular offense and how to weigh various sentencing factors, similar offenders often received different sentences. (23) For example, a widely reported sentencing experiment in the Second Circuit revealed that "[w]here one judge sentenced a defendant to three years, another judge chose twenty years." (24)

The perceived disparity in sentencing led to disenchantment with discretionary sentencing. Some critics believed that judicial sentencing discretion often resulted in unacceptably low sentences, while others were more concerned that sentencing decisions were being influenced by defendants' race and class. (25) A number of jurisdictions across the country decided to impose limitations on judicial sentencing discretion. (26) States adopted sentencing regimes that ranged across a spectrum of specificity and varied in terms of the amount of authority retained by trial court judges. (27)

The most widely studied structured sentencing system, the Federal Sentencing Guidelines, combined a high level of specificity with strict limitations on trial court variation. For example, before the adoption of the Federal Sentencing Guidelines, if a defendant violated a federal anti-fraud law carrying a possible punishment of up to 20 years of imprisonment, a judge could have imposed a sentence as low as no incarceration to a sentence as high as 20 years in prison based on whatever factors the judge thought relevant. (28) The Guidelines limited that sentencing range based on the facts of the particular case before the sentencing judge. For example, for an offender with no previous convictions who defrauded his business partners out of $175,000 using a scheme that required some planning, the Guidelines required judges to impose a sentence within the range of 15 to 21 months. (29) The judge would be permitted to impose a sentence lower than 15 months or higher than 21 months only if a few previously specified conditions existed, (30) or if the case involved circumstances not adequately taken into consideration by the United States Sentencing Commission. (31)

The rise of structured sentencing systems, like the federal system, that severely restricted a sentencing judge's authority--systems that we refer to as mandatory sentencing systems--ameliorated problems of perceived sentencing arbitrariness and inequality. (32) But these mandatory systems spawned new criticisms. In particular, because particular factual findings often triggered mandatory punishment enhancements, and because those enhancements were far more visible than sentencing practices in discretionary systems, new attention was paid to the process through which those factual findings were made. (33)

A number of legal commentators began to raise questions about the need for procedural protections at sentencing, given that judges were required to impose specific punishments based on their factual findings. (34) Arguments of these sorts gained traction in the courts. Although the trend in favor of recognizing more procedural sentencing rights predates the adoption of mandatory sentencing systems, (35) it seems to have grown stronger in the last quarter of the twentieth century--i.e., after the proliferation of mandatory sentencing systems. (36)

The connection between procedural rights and mandatory sentencing regimes was made explicit in a series of cases beginning in 2000. In Apprendi v. New Jersey, (37) the Supreme Court deemed unconstitutional a statutory sentencing enhancement that provided for an increase in the maximum sentence for the unlawful possession of a firearm if the sentencing judge found that the defendant possessed the firearm to intimidate someone because of their race. The Apprendi Court held that, other than the fact of prior conviction, any fact that increases the penalty for a crime beyond the prescribed statutory maximum must be submitted to a jury and proved beyond a reasonable doubt. (38) The subsequent decision in Blakely v. Washington made clear that Apprendi's holding applied to all mandatory sentencing guideline regimes. (39) The Blakely Court explained that "the 'statutory maximum' for Apprendi purposes is the maximum sentence a judge may impose solely on the basis of the facts reflected in the jury verdict or admitted by the defendant." (40) Thus, if a mandatory sentencing regime limits a sentencing judge's discretion to a range narrower than the statutory range, and if a sentencing court may sentence above that range only if the judge makes a particular finding, then the finding must be submitted to a jury and proven beyond a reasonable doubt. To do otherwise, the Court stated, would violate the Sixth Amendment. (41)

One term after its decision in Blakely, the Court addressed the constitutionality of the Federal Sentencing Guidelines in United States v. Booker. (42) While finding that the Federal Sentencing Guidelines suffered from the same constitutional infirmity as the state guidelines in Blakely, the Booker Court held that the Sixth Amendment problem could be cured, not only by sending aggravating factual findings to a jury, but also by restoring the discretion of sentencing judges. Specifically, five justices elected to remedy the constitutional defect of the Federal Sentencing Guidelines by excising the provision of the Sentencing Reform Act that made the Guidelines mandatory. (43) According to the remedial majority in Booker, making the Guidelines advisory, rather than mandatory, avoids the constitutional problem identified in Apprendi and Blakely because a factual finding is no longer required to sentence above the Guideline's range. (44)

Notably, in the wake of Booker, the Supreme Court has retained a number of procedural features of the previously mandatory federal system. Those procedures "impose a series of requirements on sentencing courts that cabin the exercise of [judicial] discretion." (45) Yet, the Supreme Court has never articulated how much discretion judges must retain so that the jury right and the proof beyond a reasonable doubt requirement are not triggered.

More generally, while the Supreme Court has often sharply distinguished between mandatory and discretionary sentencing regimes--as if mandatory and discretionary are binary categories--as explained more fully below, sentencing systems are better conceived of as falling across a spectrum. (46) That is to say, all sentencing systems contain some mandatory and some discretionary features.

Perhaps more importantly, the Court has not articulated a consistent principle to explain why certain procedural rights are available in mandatory sentencing systems, but not discretionary systems. In some cases, the Court has relied on the expectations of defendants to impose particular sentences under mandatory schemes; (47) in others they have rejected that approach and instead relied on historical practice. (48) But in still others, the Court has eschewed heavy reliance on history to justify its different treatment of mandatory and discretionary sentencing systems. (49)


When holding that certain procedural protections must be afforded at sentencing in mandatory sentencing systems, the Supreme Court often reaffirms the authority of judges to impose sentences without providing those protections in discretionary systems. (50) This Article challenges the practice of providing more procedural protections in mandatory sentencing systems than in discretionary systems. To understand why the presence of judicial discretion ought not affect a defendant's procedural sentencing rights, it is first necessary to define discretion and explain how it works at sentencing. This Section does just that.

A. Defining Discretion

Discretion, at its core, means the "power of free decision; individual judgment; undirected choice." (51) There are two ways that a trial court may have the power to act according to its judgment, and they are often not carefully distinguished. (52) First, the law may explicitly confer the authority on a judge to use her judgment in choosing between possible outcomes, rather than prescribing a particular outcome. We call this type of discretion "explicit discretion." Explicit discretion is often discussed in terms of flexible legal standards, as opposed to rigid legal rules. (53) Rules prescribe an outcome based on particular facts; by contrast, standards prescribe factors for a judge to consider in determining an outcome. (54) For example, a rule would outlaw driving over 60 mph, while a standard would outlaw driving unreasonably fast. (55) The application of standards involves discretion. A standard provides guiding principles for the judge, but a decision cannot be made without an exercise of personal judgment in which the judge balances those factors against one another.

The degree of explicit discretion can vary significantly. The law may grant vast explicit discretion; at its extreme, the law may empower judges to make decisions "where 'there is no law to apply,'" (56) in which case the decision is left to the judge's "individual sense of what is right and just." (57) Or the law may narrow a judge's discretion by limiting the factors that a judge may consider in making a decision. For example, 28 U.S.C. [section] 1367 enumerates several reasons that a trial court may opt to dismiss a state-law claim falling within supplemental jurisdiction and dismissals for other reasons are forbidden. (58)

The second type of discretion involves the reviewability of judicial decisions. (59) The law may effectively confer discretion on a judge by insulating his decisions from appellate review. We call this type of discretion "de facto discretion," because as a practical matter it results in the judge having the power to act according to his judgment. As with explicit discretion, there is a range of de facto discretion. The law may confer complete de facto discretion on a trial court by forbidding all appellate review, (60) or it may confer no de facto discretion by directing de novo appellate review. (61) Between those two extremes is a range of appellate standards of varying degrees of deference. (62)

Although they are distinct, explicit discretion and de facto discretion are often intertwined. Appellate courts regularly refuse to scrutinize trial court decisions closely because the law confers explicit discretion on the trial court. (63) But the two types of discretion do not always go hand in hand. Appellate courts sometimes closely review trial court decisions that are the product of explicit discretion, (64) and other times give scant review where the law prescribes a clear rule. (65)

B. Discretion in Sentencing

The use of the term "discretion" in the sentencing context usually refers to explicit discretion. So-called discretionary sentencing systems do not prescribe particular sentences based on particular factual findings; instead, they empower sentencing judges to exercise their judgment in imposing sentence. To be sure, judges in discretionary systems also usually have more de facto discretion than judges in mandatory systems. (66) But the existence of this de facto discretion is not the basis on which sentencing schemes have been deemed discretionary.

In discretionary sentencing schemes, judges make three separate determinations: (1) policy decisions, (2) factual decisions, and (3) decisions applying policy decisions to particular facts. (67) Policy decisions are those that dictate what considerations should affect punishment. Some of those decisions are at a high level of generality, such as whether to prioritize incapacitation and other crime control goals above retributive concerns like culpability. But most policy decisions involve translating these general goals into tangible rules by making choices about which particular sentencing factors ought to increase or decrease a sentence. For example, a judge who prioritizes retributivism might shorten sentences for youthful offenders because of their decreased culpability, while a judge who prioritizes incapacitation might impose longer sentences because young people have higher recidivism rates. (68)

Factual determinations are the means by which a judge determines whether to apply a particular policy to an offender. They are the primary method of distinguishing between offenders. (69) The differences in the background of two offenders who committed the same crime and the circumstances of their crime justify imposing different sentences on them. If one defendant merely brandished a gun in a robbery, and another defendant actually fired the gun during the robbery, the difference in those facts justifies a longer sentence for the second offender.

Although many facts can be objectively ascertained--such as determining what happened in the past--other factual findings contain a normative component. (70) These normative judgments are oftentimes mixed up with questions of historical fact. For example, in self-defense cases, juries are often asked to decide whether the defendant had a reasonable belief that the victim was about to harm her. (71) What the defendant believed is a historical fact; whether that belief was reasonable is a normative judgment.

The third decision judges make in discretionary schemes is how to apply the sentencing policies to the particular facts. To return to the example of age: if a judge concludes that a defendant is young and that, as a matter of policy, young offenders should receive reduced sentences, the judge still must determine by how much to decrease the sentence for the particular defendant. In other words, once a judge determines that a relevant sentencing factor is present, the judge must also determine the weight to afford to that factor. (72) As with policy decisions, in discretionary systems, the application of facts to the applicable policy rules is performed by judges in the context of individual sentencing proceedings.

Judges play a smaller role in mandatory sentencing schemes. Unlike in discretionary schemes, judges in mandatory systems do not make policy determinations. Instead, policy decisions are made ex ante in rules prescribed by the legislature or a sentencing commission. Similarly, judges in mandatory systems often do not determine the weight to afford each sentencing factor in a particular case. Instead, the legislature or a sentencing commission specifies ex ante the amount by which a particular fact increases or decreases a sentence.

By contrast, the judge's role in fact finding--both the objective and normative sort--is the same in both mandatory and discretionary sentencing systems. In both systems, judges must make factual findings to determine what sentence to impose. (73) In mandatory systems, the judge finds facts to determine whether to apply policies articulated in ex ante rules. (74) Similarly, in discretionary systems, the judge must finds facts to determine whether to apply the policy rules that she has adopted in the particular case. (75)

In short, the difference between mandatory and discretionary systems lies in policy and application decisions. Discretionary systems commit policy and application decisions to the discretion of trial judges, who make those decisions in the context of individual sentencing, (76) while mandatory systems remove discretion over policy and application decisions from judges; those decisions are instead embodied in generally applicable rules.

The control over policy and application decisions in discretionary schemes provides judges with greater flexibility to tailor the punishment to each offender, thus allowing them to impose sentences that better achieve the goals of punishment. (77) Mandatory schemes do not allow similar tailoring because they specify ex ante sentencing factors and the weight to be accorded to those factors. Those ex ante rules often fail to account for salient differences that arise in particular cases because legislatures and sentencing commissions are not omniscient. (78)

Consider a judge faced with a defendant who has been convicted of using a controlled substance. During the sentencing proceeding, the defendant presents evidence that he had become addicted to narcotics because, when serving in the armed forces, he had been subject to numerous experiments by the government in which various controlled substances had been administered to him. A judge in a discretionary system is free to take this evidence into account when imposing sentence. But a judge in a mandatory system would be prohibited from reducing the defendant's sentence if that system had a rule deeming the reason for using controlled substances irrelevant. (79) Thus, a judge in a discretionary system could individualize the sentence for this offender, while a judge in a mandatory system could not.

Although mandatory and discretionary sentencing systems in theory have substantial differences, in practice, sentencing systems cannot be neatly divided into those two categories. All sentencing systems have some mandatory features and some discretionary features. The major distinction between mandatory and discretionary systems is the amount of control judges retain over policy and application decisions. Yet judges in mandatory sentencing systems retain some limited power over policy and application decisions. (80) Most notably, even in the most rigid mandatory systems, judges retain authority to fashion their own sentences in unusual circumstances. (81) And in discretionary sentencing systems there are often substantive limitations on how judges make policy and application decisions. (82) It accordingly is understandable why both courts (83) and commentators (84) have expressed confusion over where to draw the line between mandatory and discretionary systems.

Blakely v. Washington provides a vivid example of a sentencing regime straddling this line. (85) There, Washington law set a "standard" sentencing range for most crimes but permitted judges to sentence above that range if an aggravating factor was present. (86) The statute provided a list of aggravating factors, but judges were not limited to that list of factors nor were they obligated to increase a sentence based on a finding of one of the aggravating factors. Individual sentencing judges retained discretion to increase sentences based on facts that had not been identified by the statute, as well as the discretion not to increase a sentence if an aggravating fact was present. (87) In other words, judges in Washington retained significant authority over policy and application decisions--precisely those features which are the hallmark of discretionary sentencing--yet the Supreme Court concluded that Washington's sentencing scheme was mandatory. (88)


As the previous sections note, the amount of discretion a judge retains over policy and application decisions at sentencing in particular, whether the courts label a sentencing system a mandatory system or a discretionary system--affects the number of procedural rights that defendants in that system will receive at sentencing. The Supreme Court has held that only those defendants who are sentenced in a mandatory system are entitled to the right to a jury to determine sentencing facts, the right to have sentencing facts proven beyond a reasonable doubt, the right to notice of which facts and other considerations will inform the judges' sentencing decisions, and the right not to be sentenced under ex post facto laws. (89)

Basic principles of constitutional interpretation dictate that courts ought to examine the text and purpose of a constitutional provision in determining the scope of its enforcement. (90) Thus, in deciding to enforce a constitutional right in a mandatory sentencing system, but not a discretionary system, one would expect to see the Court explain how the mandatory nature of a sentencing system is relevant to the text or purpose of that constitutional provision. While the Court's procedural sentencing rights cases have occasionally contained such analysis, (91) they often do not. (92) Instead, courts have offered other arguments for why procedural rights do not apply in discretionary schemes--arguments that we address in Part IV.

This Section provides foundational constitutional analysis for each of these four sentencing rights, and it explains why these procedural rights should not be enforced differently in discretionary and mandatory schemes. Nothing in the text of the relevant constitutional provisions distinguishes between discretionary and mandatory sentencing systems. Moreover, the reasons underlying each of these rights do not support more rigorous enforcement in mandatory rather than discretionary sentencing systems. To the contrary, the reason behind each of these rights suggest that these rights should be enforced equally, if not more strongly, in discretionary sentencing systems than in mandatory ones.

A. Sixth Amendment Right to a Jury

The Sixth Amendment provides that "[i] n all criminal prosecutions, the accused shall enjoy the right to a ... trial, by ... jury." (93) This right to a jury trial reflects the belief that judges cannot be trusted to make determinations resulting in a criminal conviction and the ensuing punishment. (94) Instead of allowing judges to make that determination, the task is assigned to a group of peers. This jury acts as a check on the potential abuses of the government. (95) The jury trial right also provides a means of injecting community values into criminal determinations. (96) Most judges come from similar economic backgrounds, have similar educations, and travel in the same social circles. (97) They therefore often do not hold values held by many in the community. Exacerbating this disconnect with the community are the life tenure and salary protections afforded to many judges. Although valuable for shielding judges from retaliation for unpopular decisions, these protections render judges less responsive to popular opinions. (98) The jury right addresses this situation by assigning important decisions to a group of nongovernment actors. (99)

The Supreme Court has sharply distinguished between mandatory and discretionary sentencing determinations in enforcing the right to a jury. For factual findings in mandatory systems, defendants have a robust jury right. The Court has held that, under the Sixth Amendment, "[o]ther than the fact of a prior conviction," any fact that increases the sentencing range for a crime--either by increasing the maximum (100) or the minimum sentence (101) that the defendant faces--must be submitted to a jury. Because mandatory systems alter the range of punishment based on factual findings, the Sixth Amendment requires that those findings be made by a jury. (102)

By contrast, the Court has held that the jury trial right does not limit a judge's ability to find facts in a discretionary sentencing system. (103) It has stated that only those factual findings that alter the range of punishment within which a judge may sentence trigger the jury right. (104) Factfinding made to select a sentence from within that range does not. (105) Thus, when the law confers discretion on a judge to pick a sentence within a particular range, current doctrine permits a judge to make factual findings to guide her discretion; those findings need not be submitted to a jury.

The disparate treatment of the jury right in discretionary and mandatory systems is unwarranted. To start, even under the Court's test, factual findings in discretionary schemes should be treated the same as factual findings in mandatory schemes. That is because a judge in a discretionary scheme cannot impose the maximum sentence within the range without making a factual finding.

The primary justification for discretionary sentencing schemes is that they enable judges to tailor sentences to each offender. (106) Imposing the maximum sentence on every offender violates that norm. Different offenders must be treated differently. A defendant in a discretionary system should receive the maximum only if there are reasons to impose the maximum on him. Factual findings are what enable a judge to distinguish between two offenders. A judge is justified in imposing different sentences on two individuals who committed the same crime only because there are relevant factual differences between the two. (107) Indeed, it would be an abuse of discretion for a judge to sentence all defendants convicted of a particular crime to the maximum statutory sentence because she believes all people who commit that crime deserve significant punishment, or because she believes that the legislature set the punishment range too low. (108)

To be sure, unlike judges in mandatory schemes, judges in discretionary schemes make policy and application decisions in addition to factual findings, (109) and those nonfactual decisions may affect the sentences imposed. (110) But that does not diminish the significance of factual findings in discretionary schemes. It is only through factual findings that a judge can determine which sentencing policy to apply and how to apply it to a particular defendant. Factfinding is just as consequential in discretionary systems as in mandatory ones.

What is more, the reasons underlying the jury right apply more strongly to discretionary sentencing than mandatory sentencing. There is greater need for protecting the two values underlying the jury trial--preventing government abuses and injecting community values--in a discretionary sentencing system than in a mandatory one. First, a mandatory scheme presents less opportunity for judicial abuse than a discretionary scheme. Under a mandatory sentencing system, the law establishes sentencing ranges ex ante based on specific factors. (111) The sentencing judge thus has less ability to manipulate a sentence based on inappropriate considerations.

Consider the following hypothetical. Two states enact laws prohibiting burglary. The first state's statute provides up to 5 years of imprisonment for the offense, unless the burglary occurred at night, in which case up to 20 years of imprisonment may be imposed. Under that statute, a judge cannot impose a sentence higher than 5 years without a finding that the defendant committed the burglary at night. The second state's statute provides up to 20 years of imprisonment for burglary, and it leaves to the judge to determine what facts and circumstances ought to result in a sentence approaching 20 years. In the first system (a mandatory system), a judge may not impose a higher sentence without a specific factual finding, thus limiting the judge's ability to impose a high sentence based on bias or whim. (112) In the second system (a discretionary system), judges have the power to pick sentences based on their overall assessment of the offender and the offense. (113) This gestalt approach virtually invites sentences based on inappropriate factors. (114)

Concerns of these sorts are what prompted Congress and various states to adopt mandatory sentencing systems. Under discretionary regimes, sentences varied wildly depending on the values and prejudices of the judge imposing the sentence, and the temperament of the judge at the time he imposed the sentence. (115) Legislators adopted mandatory sentencing rules to reduce these unjustified disparities by prescribing narrow sentencing ranges based on particular factual circumstances. (116)

Second, sentences imposed under a mandatory scheme are more likely to reflect community values than sentences under discretionary systems. Sentences imposed under mandatory schemes are the product of determinations by the legislature (or its agent) of the appropriate punishments to impose for particular acts. Because legislators are elected, mandatory schemes ostensibly reflect community values; at the least, legislators may be held democratically accountable for schemes that diverge from community values. (117) By contrast, in discretionary systems, sentences reflect only the sentencing judge's assessment of what punishment is appropriate. Although some judges face elections similar to legislators, most have substantially more job security. They accordingly are less accountable and less likely to represent community values than legislators. Jury participation accordingly is more important in discretionary schemes.

B. Due Process Right to Proof Beyond a Reasonable Doubt

Under the Due Process Clause, a defendant may be convicted of a crime only upon proof beyond a reasonable doubt "of every fact necessary to constitute the crime with which he is charged." (118) The Court has given two reasons for this requirement. The first is to reduce the risk of erroneous convictions. (119) Because juries must decide based on incomplete information, there is always a risk that they will make erroneous factual conclusions. The proof beyond a reasonable doubt standard reduces the risk that those errors harm defendants by requiring juries to resolve doubts in the defendants' favor. (120) Although the high standard also increases the risk of erroneous acquittals, erroneous acquittals are preferable to erroneous convictions because the consequences to the accused of an erroneous conviction--the deprivation of life, liberty, or property, and the stigmatization from the conviction--are more grave than allowing the guilty to walk free. (121) The second reason for the requirement is to promote trust in government and preserve its legitimacy. Society is unlikely to support a government that imposes criminal penalties on those whom society perceives might be innocent--not only because members of society may feel that the government has abused its power, but also because members of society may fear that they will be next. (122) The proof beyond a reasonable doubt standard minimizes that risk.

Based on the same rationale for extending the jury right to factual findings in mandatory schemes, the Supreme Court held in Apprendi that the proof beyond a reasonable doubt standard applies to factual findings under mandatory schemes. (123) A court may increase a sentence under a mandatory scheme based on a factual finding only if that fact is proven beyond a reasonable doubt. (124) By contrast, the Court has refused to impose that burden of proof for factual findings under discretionary sentencing schemes. Instead, the Court has stated that a preponderance standard satisfies due process in discretionary systems, and it has suggested that an even lower burden of proof may suffice. (125)

Apprendi's test does not provide a sound basis for distinguishing between mandatory and discretionary sentencing schemes. More important, the reasons underlying the proof beyond a reasonable doubt requirement--to prevent erroneous factual determinations that harm the defendant and to protect the legitimacy of sentencing--do not support distinguishing between mandatory and discretionary schemes.

As the previous sections explain, factual findings play an integral role in sentencing under both mandatory and discretionary schemes. (126) A heightened standard of proof for sentencing facts under a mandatory scheme reduces the risk of offenders erroneously spending more time in prison. A court may impose a higher sentence based on a particular fact only if that fact is proven beyond a reasonable doubt. That standard ensures that a defendant does not receive a sentence increase if there are any doubts whether that fact is true. That reasoning applies equally to discretionary sentencing schemes. In a discretionary system, a judge imposes a sentence based on her assessment of the offender and the offense. But that assessment cannot be based simply on a gut feeling, but instead must be the product of a series of findings about the nature of the offense and offender. (127) A heightened burden of proof increases the burden on the government to produce reliable sentencing evidence, thus reducing the risk that the judge's conclusions are incorrect. (128) Indeed, there are many examples from discretionary sentencing systems where defendants were sentenced to harsh penalties based on erroneous factual conclusions. (129)

One might argue that a heightened burden of proof is unnecessary at sentencing because, unlike with erroneous convictions, forcing a guilty defendant to spend a longer amount of time in prison is less problematic than subjecting an innocent person to conviction and punishment. In other words, doubts at sentencing, unlike at the guilt phase, ought not always be resolved in the defendant's favor. Or one might argue that proof beyond a reasonable doubt interferes with the states' ability to make sentencing decisions efficiently. While these arguments might support having a lower burden of proof for facts at sentencing generally, they do not support having a higher burden of proof only in mandatory sentencing systems. That is because the conviction versus amount of punishment distinction and the efficiency argument are unaffected by the only difference between mandatory and discretionary systems--namely, whether judges have authority over policy and application decisions.

Nor is there reason to think that a higher burden of proof is more important to the legitimacy of mandatory sentencing than to the legitimacy of discretionary schemes. The threat to legitimacy from a low burden of proof is that society may perceive that the government may deprive individuals of their liberty based on untrue facts. Because factual findings play a critical role in determining sentences in both sentencing schemes, that legitimacy concern applies no less to discretionary schemes than to mandatory ones.

C. Due Process Right to Notice

Due process requires that, when the government seeks to deprive an individual of life, liberty, or property, it must provide that individual with adequate notice of the reason for that deprivation. (130) A good deal of confusion has resulted from failing to distinguish between two separate types of notice that due process requires. (131) The first is that the law must give notice to the public of what conduct is prohibited and the consequences for performing that conduct. (132) This type of notice--which we refer to as "ex ante notice"--both allows members of the public to conform their behavior to the law and provides guidance to the government to prevent arbitrary and discriminatory enforcement of the law. (133)

The second type of notice--which we refer to as "adversarial notice"--applies during adjudication. When the government seeks to enforce a law against an individual, it must provide a defendant with notice of the allegations against him and an opportunity to respond to those allegations. (134) These requirements of notice and opportunity to be heard help to ensure that the defendant is not being punished except as authorized by the law. (135) A hearing allows the defendant to challenge the evidence against him, to point out flaws in the government's arguments, and to launch other attacks on the government's case. (136) Adversarial notice is critical to ensure that the defendant's opportunity to respond to the charges against him is meaningful. (137) It allows the defendant time to prepare a defense that responds to the allegations against him and to gather any evidence to support his position. (138)

In Burns v. United States, the Supreme Court strongly indicated that defendants have a due process right to notice of information that may be used in sentencing under mandatory schemes. (139) Bums involved the imposition of a sentence under the then-mandatory Federal Sentencing Guidelines. The Court held that a court cannot impose a sentence above the range specified by the relevant Guideline based on a factual finding unless the defendant is first given notice of the basis for that increase. (140) It explained that notice was necessary to ensure "focused, adversarial development of the factual and legal issues relevant to determining the appropriate" sentence. (141) Although the Court based its conclusion on an interpretation of the Sentencing Reform Act and Federal Rule of Criminal Procedure 32 instead of the Due Process Clause, (142) it explained that a contrary interpretation would raise a "serious question" under the Due Process Clause. (143) As others have noted, this canon of constitutional avoidance--adopting a particular interpretation of a statute to avoid a potential constitutional concern--is tantamount to a constitutional conclusion. (144) The canon results in an interpretation based on potential constitutional concerns instead of based on what Congress intended. (145)

The Court has also made clear that the due process right to notice discussed in Bums does not extend to discretionary schemes. In Irizarry v. United States, the Court explained that the only reason that due process may require notice of a potential enhancement under a mandatory scheme is that those guidelines created an expectation that the defendant's sentence could be increased under only very limited circumstances. (146) In discretionary schemes, the Court reasoned, defendants do not have a comparable expectation and accordingly do not have a right to notice of potential grounds for enhancements. (147) In other words, when a mandatory scheme identifies limited sentencing factors as relevant, defendants must receive notice if a judge is going to consider any other factors; but because discretionary schemes do not limit the consideration of sentencing factors, defendants need not receive prior notice of which factors the judge intends to consider.

The upshot of Bums and Irizarry is that whether a defendant has a right to adversarial notice of a sentencing consideration turns on the degree of ex ante notice he has received regarding sentencing considerations. When the law specifies particular mandatory sentencing enhancements, the sentencing court must provide notice if it is considering increasing a defendant's sentence for any additional reason. But when the law leaves all enhancements to the discretion of the judge, the sentencing judge need not give the defendant any hint of which considerations might result in a higher sentence.

It is unclear why the right to adversarial notice should depend on the provision of ex ante notice. The two types of notice perform different functions. Ex ante notice informs the public of what conduct may be punished and how much punishment may be imposed. By contrast, adversarial notice ensures that the defendant has a meaningful opportunity to challenge the government's efforts to seek punishment for particular acts. The level of specificity in generally applicable laws regarding how punishment is calculated should not affect the amount of information an individual defendant will receive about how those laws will be applied to her. Certainly, no one would think that the clarity with which the law defines a crime should affect the specificity of notice an indictment must provide of the factual basis for a charge under that law.

To the extent that ex ante and adversarial notice should be linked, the less of one ought to result in more of the other. A motivating principle behind both ex ante and adversarial notice is to prevent the government from acting arbitrarily against an individual. Ex ante notice accomplishes that goal by specifying the grounds on which the government may seek punishment, and adversarial notice does so by providing a defendant with a meaningful opportunity to challenge the basis on which the government seeks to impose punishment. When a law provides less ex ante notice, it constrains the government less. (148) That weaker constraint may be offset to some degree (though not completely) by recognizing a stronger right to adversarial notice. (149)

More generally, the Court's focus on expectations in distinguishing discretionary and mandatory schemes is misplaced. The Due Process Clause does not protect against defeated expectations; it protects against the deprivation of life, liberty, or property without adequate process. (150) The type of process required should depend on the gravity of the deprivation and the usefulness of providing that procedure. (151) These considerations suggest that notice should be required no less in discretionary than in mandatory schemes.

Whether a sentencing scheme is mandatory or discretionary does not change the deprivation of liberty resulting from sentencing findings. The deprivation of liberty resulting from increasing a sentence by five years based on a judge's determination in a discretionary system that the offender used a firearm is the same as the deprivation resulting from an increase of five years based on a finding of a firearm under a mandatory scheme. Both result in five additional years of imprisonment.

Providing notice of the grounds for enhancement is equally important in discretionary and mandatory schemes. In both systems, sentencing decisions turn on factual findings about the circumstances of the crime--such as harm to the victim, the defendant's motive, and the danger the crime created for others--and about the characteristics of the defendant--such as his age, previous convictions, and family status. (152) Providing notice that these facts may matter to the sentence allows the defendant to gather evidence and otherwise mount a meaningful response to the government's allegations.

Indeed, adversarial notice is more important in discretionary schemes because, unlike in mandatory systems, judges have the power to make policy determinations about which facts should affect a sentence and the amount by which they should affect a sentence. (153) Those determinations--whether to modify a sentence based on a particular factual finding and if so, how much--are not set in stone, and the prosecutor and defendant each have an opportunity to present arguments about the matter. Affording the defendant notice is critical for the defendant to be able to formulate a persuasive argument about which facts should matter and the amount of weight the judge should give those facts. (154)

To be sure, providing notice in a discretionary system would generate some costs. But the costs would not be appreciably greater than those incurred by providing notice in a mandatory system. The only cost would be requiring judges to identify the factors that may influence their sentencing decision before conducting a hearing. Judges face the same task in mandatory schemes; the only difference is that, while mandatory schemes specify the possible grounds that a judge may pick from, discretionary schemes give judges more leeway in determining what may affect their decisions. That is hardly a heavy burden (155)--after all, the judge must make that determination at some point--and it is certainly not heavy enough to deprive defendants of a full opportunity to challenge the bases for their sentence.

D. Ex Post Facto Clause

The Constitution prohibits the federal and state governments from enacting ex post facto laws. (156) An ex post facto law is a criminal law that has a retroactive effect. (157) As the Supreme Court has explained, the prohibition forbids not only a law that criminalizes conduct that was not illegal when committed, but also a law that "inflicts a greater punishment, than the law annexed to the crime, when committed." (158) Moreover, the Court has said, a retroactive law need not guarantee a heightened punishment to run afoul of the clause; instead, a law violates the Ex Post Facto Clause merely if it "creates a significant risk" of increasing a defendant's punishment. (159) Although the prohibition on ex post facto laws seems better understood as a substantive rule because it bars the courts from using a particular substantive rule in imposing punishment, it is often characterized as a procedural right. (160) We accordingly address it here.

There are three reasons for the prohibition on ex post facto laws. First, the prohibition ensures that individuals have ex ante notice of what is illegal and the consequences of engaging in that illegal conduct. Ex post facto laws fail to provide such notice because they change the consequences of engaging in activity after an individual has already chosen to engage in that activity. (161) An individual accordingly has no opportunity to conform her behavior to the law. Likewise, a law that retroactively increases a penalty for conduct that is already illegal fails to provide an individual with the information necessary to determine whether violating the law is worth the possible consequences. (162)

Second, the prohibition protects against arbitrary and vindictive legislation. (163) Through retroactive criminal laws, a legislature may punish disfavored individuals by criminalizing or increasing punishment for conduct that has already occurred. (164) For example, one could easily imagine that, without the Ex Post Facto Clause, Congress would have responded to public anger at Enron and other business scandals with increased criminal penalties for the past transgression of white collar offenders. (165)

Third, the prohibition promotes fairness by requiring the government to abide by the laws it has established governing the circumstances under which it can deprive a person of his or her liberty or life. (166) It avoids the sense of unfairness and illegitimacy that would arise if the government could change the rules of the game after the players have begun to play.

In Miller v. Florida, the Supreme Court held that the prohibition extends to mandatory sentencing guidelines. (167) Thus, when the government identifies new sentencing factors on which to increase a sentence or seeks to increase the penalty for previously identified sentencing factors, it may only do so prospectively. The Ex Post Facto Clause forbids the retroactive application of those mandatory sentencing guidelines.

By contrast, the Court has not adopted a blanket prohibition on retroactively increasing sentencing recommendations under more discretionary schemes. It has explained that the test for whether the Ex Post Facto Clause prohibits a retroactive change in the law is whether such a change "creates a significant risk" of increasing the punishment for the crime. (168) Therefore, whether a change in sentencing law may be applied retroactively depends on the amount of discretion courts have not to apply that changed law.

When a judge has only limited discretion not to impose a new law, the Ex Post Facto Clause may prohibit the retroactive application of that new sentencing law. Thus, in Peugh v. United States, the Court held that a retroactive increase to the now-advisory Federal Sentencing Guidelines violated the ex post facto prohibition. (169) It explained that, although the Guidelines are not binding, district court judges are required to begin their sentencing analysis with a calculation

from the Guidelines, (170) resulting in sentences that are "anchored by the Guidelines." (171) Accordingly, the Court concluded, a change to the Guidelines created a significant risk of an increase in punishment. (172)

By contrast, the Court has said that when a court has substantial discretion not to apply a change to sentencing law, the Ex Post Facto Clause may not prohibit the retroactive application of that factor. (173) Thus, under a so-called purely voluntary sentencing guidelines system--that is, a system that articulates policy and application rules, but does not limit the authority of judges to disregard those rules (174)--the Ex Post Facto Clause does not prohibit the judge from retroactively applying a new rule. To be sure, the Court has suggested that, even if there is not a substantial risk of increased punishment, an ex post facto challenge could be successful if the exercise of discretion under a voluntary guidelines system actually results in a higher sentence because of the new rule. (175) But meeting that standard is often difficult because it requires being able to identify each factor affecting the judge's exercise of discretion and the weight that the judge placed on each factor (176)--determinations that are difficult to quantify and are often not recorded.

The different treatment of ex post facto laws in discretionary and mandatory systems is unwarranted. To start, the substantial risk test does not provide a sound basis for distinguishing between discretionary and mandatory sentencing schemes. Even under a completely discretionary scheme, such as purely voluntary sentencing guidelines, legislative identification of possible sentencing factors and the recommended punishment for those factors will have an effect on sentences. (177) Although not obligated to follow the guidelines, some judges may do so, (178) if for no other reason than following the legislative guidance reduces the judge's burden in exercising his discretion. Retroactive changes to those guidelines which increase a sentence therefore present a risk that a court will indeed impose that higher sentence. Even if the risk is not substantial in all cases, there is a substantial chance that the guidelines will influence courts in at least some cases. (179)

More important, the Ex Post Facto Clause prohibits laws that are designed to inflict a punishment greater than what was prescribed under the law when the crime was committed. (180) Any law that calls for retroactive increases in punishment meets this definition. Even when a judge is not obliged to follow the new law, it is still an official statement setting forth substantive standards on what sentence is "appropriate" and its purpose is to affect the punishment imposed.

That a judge may have discretion not to follow a law does not make it any less law. (181) Judges often have discretion not to enforce a particular regulation. They may, for example, refuse to award an injunction enforcing a law if they conclude that the injunction would be against the public interest. (182) Similarly, prosecutors ordinarily have plenary discretion not to bring a prosecution for a violation of a criminal statute. Neither form of discretion deprives the law of its status as law, nor does conferring discretion on a judge not to follow a particular guideline mean that guideline is not law. Whether an enactment constitutes law depends on the process by which it is enacted. (183)

Indeed, consistent with this view, older cases hold that the Ex Post Facto Clause prohibits changes to the substantive standards by which punishment is calculated; they did not ask whether those changes presented a substantial risk of increased punishment. (184) The substantial risk test was designed to determine whether changes to a procedure triggered the Ex Post Facto Clause. In Beazell v. Ohio, the case from which the substantial risk test derives, the Court considered whether a statute changing the requirement that defendants jointly charged with the same felony be tried separately to one where they would be tried jointly violated the Ex Post Facto Clause. (185) In finding no violation, the Court explained that the change was only procedural and, unlike changes to substantive law, whether a change to a procedural law violates the Ex Post Facto Clause is a question of "degree." (186) Subsequent cases confirm that the substantial risk test was to apply only to changes in procedures that might result in an increase of a defendant's punishment--such as a change in the rules of evidence or restricting the frequency of a parole hearing (187)--and not to changes to a substantive standard.

Prohibiting retroactive increases to all sentencing laws, even purely voluntary sentencing guidelines, also promotes the reasons underlying the ex post facto prohibition--providing notice, limiting vindictive government action, and avoiding a sense of unfairness. Even when a sentencing law is advisory, that law notifies both sentencing judges and the public about the considerations that the government believes should affect sentencing. To be sure, retroactive changes to mandatory laws may raise greater notice problems than retroactive changes to laws in a discretionary system. Mandatory sentencing laws purport to inform the defendant of the precise consequences of his actions. Changing a mandatory law after the defendant commits the crime upsets his expectations. By contrast, the defendant in a discretionary system knows at the time he commits the crime that the judge is free to disregard the sentencing laws, and thus has less reason to complain if he receives a higher sentence based on a revised law. (188) But advisory sentencing laws still provide a notice function. They communicate the government's position on what circumstances deserve what amount of punishment, (189) and may well be the only guidance that the public has about what to expect at sentencing.

Retroactive application of voluntary laws also raises concerns about vindictive punishment. Outraged regulators may promulgate laws in the hope of persuading judges to increase the sentences of disfavored individuals who committed their wrongs in the past. That a judge may not follow the guideline does not reduce its vindictive purpose. (190) And in any event, judges may be more likely to rely on laws that are the product of public outrage, either because they share the same sentiment that gave rise to the new laws (191) or because they fear retribution from the public for not following the laws. (192) In either situation, the law provides a basis for the judge to legitimate her decision.

For similar reasons, applying sentencing laws retroactively raises a risk of perceived unfairness and illegitimacy. By enforcing a new sentencing rule, the government communicates that it no longer plays by the rules in effect at the time that the defendant committed his offense. The impetus behind changing those laws is to change the sentences that judges impose. Of course, a judge in a discretionary system is not obliged to follow either the old or the new sentencing laws. But the laws still reflect the position of the government, and unless the laws are an entirely empty exercise, they have some effect on sentencing judges. For example, even advisory sentencing laws are bound to anchor a judge's sentencing determination, and applying new laws retroactively changes that anchoring point. (193)


There are three commonly made arguments for treating procedural rights differently in mandatory and discretionary sentencing schemes. First, recognizing procedural rights will unduly interfere with judicial sentencing discretion. Second, defendants have settled expectations about how sentencing factors will be enforced in mandatory schemes and therefore more protections are necessary. Third, sentencing factors under mandatory schemes are actually elements of crimes while in discretionary schemes they are not. This Part addresses each argument of those arguments in turn.

A. Interference with Judicial Discretion

The primary objection to recognizing procedural rights in discretionary sentencing schemes is that procedural protections are incompatible with, or unnecessary in, discretionary systems. The argument has three variations: First, that procedural rights will interfere with judges' ability to use their discretion to individualize sentences; second, that procedural protections are less important in discretionary systems because discretion allows a judge to adjust sentences to account for the lack of procedures; and third, that sentencing under a discretionary system involves value judgments, which are unlikely to be affected by procedural protections.

1. Individualized Sentences

The primary objection to recognizing procedural rights in discretionary sentencing schemes is that procedural protections interfere with a judge's ability to individualize sentences. (194) The argument runs as follows: The principal benefit of discretionary schemes is that they allow judges to tailor the punishment for each individual offender based on the offender's background and personal characteristics and on the circumstances of the offense. To determine the best sentence for an individual offender, a judge must have unfettered access to information about the offense and the offender, as well as the ability to use that information without restriction. (195) Applying procedural rights at sentencing unduly impairs a judge's ability to gather information about the offender and offense and to impose sentence based on that information.

This argument, however, is overly broad. Access to information is not an end in and of itself at sentencing; rather, the reason for broad access to information is to allow a judge to impose a more accurate sentence. Although procedural rights do impair a court's access and ability to use information in sentencing, the reason that many (though not all) procedural rights do so is to enhance the accuracy of the information provided to the court. (196) Consider the right to notice. One of the reasons for the right is to improve the accuracy of decisions. Notice provides a defendant with the opportunity to gather evidence and marshal arguments to present to the court; that additional information allows a court to make a more informed decision. Another example is the right to confront witnesses--another right that courts have refused to recognize in discretionary systems, (197) but that a few have recognized in mandatory systems. (198) There is no doubt that the right impairs a court's ability to rely on hearsay at sentencing. (199) But the reason for the right is to increase accuracy. It reflects a constitutional judgment that confrontation is "how reliability can best be determined." (200)

Moreover, the fear that procedural rights will interfere with a judge's ability to make accurate factual determinations in sentencing is not a sensible ground for recognizing different procedural rights in mandatory and discretionary sentencing schemes. Mandatory and discretionary sentencing schemes depend equally on factual findings, and in both systems those decisions are committed to the sentencing judge. (201) The only difference is that mandatory systems tell a judge which facts are relevant and how to calculate punishment based on the factual findings that she makes, while discretionary systems leave those policy and application decisions to the trial judge. Thus, to the extent procedural rights impair accurate factfinding, that impairment is no different in mandatory and discretionary systems.

To be sure, discretionary sentencing schemes may entail more factual findings than mandatory ones. Mandatory schemes identify ex ante a limited set of facts that are relevant to sentencing, while discretionary schemes give judges broad authority in determining which facts should affect a sentence. But the number of factual findings that a judge may be empowered to make is irrelevant to the desirability of procedures to control how those findings are made. Accordingly, there is no less an interest in observing the procedural protections that affect factual findings in discretionary systems than in mandatory ones. Both systems depend on the availability and accuracy of information for those factual findings.

Indeed, there is good reason to think that procedural protections are more important in discretionary than in mandatory schemes because of the greater judicial authority in discretionary schemes over policy and application decisions. Determining which facts should matter to sentencing and how much they should matter requires the judge to exercise her judgment, and there is a risk that a judge will make unjustifiable or erroneous policy and application judgments. That risk is not present in mandatory schemes because they prescribe the facts relevant to sentencing and punishment for those facts. Several procedures--most notably notice and a hearing--reduce that risk by allowing the defendant to provide the judge with more information about how to exercise her judgment. (202) Thus, if anything, when judges have discretion about policy and application decisions, it may be more important to provide procedural rights for defendants.

2. Discounting Weak Evidence

Another argument for fewer procedures in discretionary systems is that judges in those systems may use their discretion to offset the lack of procedural protections. Judges in mandatory systems must adjust sentences based on factual findings. By contrast, judges in discretionary systems need not adjust a sentence based on a particular factual finding but instead may decline to do so if they conclude that a higher sentence is inappropriate. (203) Moreover, judges in discretionary systems may adjust the weight they place on a particular fact based on the probability that they believe the fact to be true. (204) For example, a judge who concludes that it is only 30% likely that a defendant brandished a gun during the commission of an offense may proportionately discount the increase in sentence he would have otherwise imposed based on the brandishing of the gun. (205)

But these two differences do not establish that procedural protections are less important in discretionary regimes. Consider the first difference--that judges are not required to adjust a sentence based on a factual finding. That a judge is free to identify which facts to consider at sentencing has no bearing on the process the judge follows in making factual findings. Once a judge decides that a particular fact is relevant to sentencing, that fact will influence sentencing no less than a fact will in a mandatory scheme. Consider a judge in a discretionary system who thinks the amount of money a defendant stole in a robbery is an important fact to consider at sentencing, and a judge sentencing in a mandatory system that dictates that the amount of money affects the sentence. In both systems, the amount of money will affect the sentence. The possibility that the judge in the discretionary system could have, but did not, disregard the amount is irrelevant. Once the judge decides to consider that fact, it will bear on the sentence. Therefore, the same procedures should apply.

To be sure, procedural protections are sometimes less important in discretionary systems than in mandatory systems. To return to the previous example, if the judge in the discretionary system states that she believes the defendant stole $10,000, but that she is not considering the amount of money stolen in selecting a sentence, then the procedures used for calculating the $10,000 figure are less important--just as there is generally less concern about policing factual determinations that have no bearing on an outcome. (206) But when a fact is material to sentencing, it is equally important regardless whether its materiality derives from a judge's discretion or an ex ante rule.

Nor does the second difference--that judges in discretionary systems may adjust the weight they place on a particular sentencing fact based on the probability that the fact is actually true--provide a sound basis for recognizing fewer procedural rights in discretionary systems. (207) For one thing, while judges in discretionary systems may adjust the weight of certain sentencing facts based on the probability that those facts are true, they are under no obligation to do so. If they opt not to discount the sentence based on weak proof, there is equal reason to recognize the procedures that are applied in mandatory sentencing schemes. For another, judges rarely have the information necessary to assess with any accuracy the margin by which a fact has not been proven. It is for this reason that judges do not apply quantitative standards for burdens of proof, but instead apply qualitative standards, such as whether there is a reasonable doubt or whether a fact it is more likely than not to be true. (208) Accurate discounting requires finer gradations, which are more difficult--if not impossible--to make. That extra level of uncertainty introduces an additional layer of judicial subjectivity, suggesting that procedural protections are even more important.

3. Sentencing as Value Judgment

A third argument against the recognition of procedural rights in discretionary sentencing systems is that those procedures are unnecessary because sentences imposed under a discretionary system involve value judgments, which are unlikely to be improved through procedural protections. (209) This argument is sometimes framed as a claim that judges in discretionary systems are making judgments about a defendant's rehabilitative prospects, which necessarily require reliance on intuition or best estimates. (210) But the fact that a sentence may involve value judgments is not a sound reason to dispense with procedural rights that are recognized in mandatory schemes. The procedural rights recognized in mandatory sentencing schemes apply to factual findings, and factual findings play a very important role in sentencing under discretionary schemes as well. (211)

More important, to the extent that sentencing in discretionary schemes involves value judgments, procedures are useful to ensure that judges make justifiable value judgments. Discretion is not tantamount to unbounded decisionmaking authority. Discretion does not mean that choices are left to the court's inclination. Judges in discretionary systems face specific limitations on what may inform their sentencing decisions. They cannot sentence above the statutory maximum, they cannot rely on false information in imposing sentence, (212) and they cannot vary sentences based on the offender's race, gender, or religion. (213) Moreover, beyond these specific constraints, the exercise of discretion must be the product of "[rational] judgment ... guided by sound legal principles." (214) As Judge Friendly put it, even in a system that does not impose explicit constraints, "there is still the implicit command that the judge shall exercise his power reasonably." (215)

Adhering to procedural protections increases the likelihood that courts obey these legal limits on their discretion. The right to notice, for example, forces a judge to articulate those facts and factors she believes are relevant in a particular sentencing proceeding, and it provides the defendant with the opportunity to prepare his case so that the court is better informed about the legal justifications for imposing a particular sentence. In a similar vein, one reason for giving a decision to a jury is to avoid leaving the decision to a judge who might abuse her power. Likewise, the right to proof beyond a reasonable doubt reduces the effect of bias; it tilts the scales toward the defendant so that a court's predisposition against a defendant has less impact. Even the prohibition on ex post facto laws serves the goal of reducing the opportunity for the arbitrary exercise of power by requiring the government to play by the rules as they were at the time that the offense was committed. (216)

Finally, to the extent that the value judgment in a sentence is important because of the message it sends to the offender, victims, and broader community, (217) a lack of procedures may undermine the validity of that message. Observing fair procedures, such as those prescribed in the Constitution, creates an aura of fairness and legitimacy for even adverse determinations. (218)

B. Defense Expectations

A different argument for providing more procedural protections in mandatory sentencing systems than in discretionary ones turns on the expectations of the defendant. According to the Supreme Court, when a guilty verdict authorizes a sentencing range, a defendant expects to receive a sentence within that range. (219) If a law provides that the sentencing maximum may be increased only upon additional factual findings, then the defendant is entitled to be sentenced at or below the original maximum sentence unless various procedural protections are observed in making the additional factual findings. Thus, for example, the Court stated in Blakely that " [i]n a system that says the judge may punish burglary with 10 to 40 years, every burglar knows he is risking 40 years in jail." (220) By contrast, the Court explained, "[i]n a system that punishes burglary with a 10-year sentence, with another 30 added for use of a gun, the burglar who enters a home unarmed is entitled to no more than a 10-year sentence." (221) The maximum penalty authorized by the facts underlying the verdict sets the ceiling on the punishment that the burglar expects to receive for his offense, and this limitation ensures that "the defendant 'will never get more punishment than he bargained for when he did the crime.'" (222)

This expectation argument is not coherent. A defendant's expectations depend on the way that a law is written. If a legislature writes a law that imposes conditions under which various sentences may be imposed, those conditions are built into the defendant's expectations. (223) For example, if a legislature creates a sentencing range of 10 to 40 years for arson but specifically states that the defendant may receive a sentence above 20 years only if the judge makes a factual finding, every arsonist knows that he faces up to 40 years of imprisonment for his crimes and that he will receive above 20 years in the event the judge makes a factual finding. His expectation is that the sentence he will receive depends on the facts found by the judge, because that is how the law is written.

Another weakness of the expectation theory is that it cannot explain the Court's decisions. For example, the expectation theory does not justify the Court's recent decision in Alleyne v. United States, which requires procedural protections for findings triggering mandatory minimum sentences. (224) Even when a judge does not make a factual finding that triggers the mandatory minimum, she may impose a sentence equivalent to what would have been the mandatory minimum because it is within the original range. For instance, if burglary carries a penalty of 10 to 40 years, and imposes a mandatory minimum of 15 years upon a finding that the offender used a firearm in the commission of the offense, ajudge could still impose a 15-year sentence even if she did not find that the defendant used a firearm. (225) It is accordingly unsurprising that the Court did not rely on the expectation theory in extending procedural protections to mandatory minimums. (226)

Moreover, the expectation theory does not provide a basis to treat discretionary and mandatory sentencing schemes differently. That is because defendants also have expectations in discretionary systems. By prescribing a sentencing range of X to Y years for the commission of a crime, a legislature creates an expectation that not everyone who commits the crime will receive the maximum penalty of Y. The existence of a range implies that different offenders will be treated differently; indeed, it is this ability to individualize sentences that is the primary justification for the discretionary sentencing schemes. (227) A defendant in a discretionary system therefore will receive the maximum sentence only if there are sufficient reasons to impose the maximum. (228) A discretionary system thus creates an expectation in the offender that he will receive less than the maximum unless the judge makes some finding supporting a maximum sentence.

To be sure, unlike mandatory systems, which identify ex ante the facts that will trigger a higher maximum, discretionary systems do not specify what facts a judge must find to impose the maximum sentence. But judges in discretionary systems must have some identifiable reason to impose the maximum sentence. A discretionary system does not create in an offender the expectation that he will receive no higher than a particular sentence without a finding. Instead, it creates an expectation that the offender will receive less than a particular amount except on a finding. But that difference should not matter. The important point is that the defendant expects not to receive a particular sentence without a finding. (229) Only the "worst" offenders should expect to receive the highest sentence, and a court can determine which offenders are the "worst" only through making factual determinations about the offender and her crime. Put simply, a defendant's expectations are not materially different in mandatory and discretionary sentencing systems.

C. Mandatory Sentencing Factors as Offense Elements

The third argument for recognizing more procedural protections in mandatory sentencing schemes is based on the conclusion that facts that trigger mandatory sentencing laws constitute elements of a criminal offense instead of merely sentencing factors. (230) Under this theory, all the procedural rights applicable to determinations of guilt at criminal trials extend to determinations of facts under mandatory sentencing schemes. By contrast, factual determinations that merely inform a judge's discretion at sentencing do not constitute elements of an offense and consequently need not receive the procedural protections applicable to trials.

The Court has based its conclusion that mandatory sentencing factors are equivalent to elements of a new offense on history. That historical argument, however, has evolved over time. In its original form, the Court relied on its assessment that judges had little or no discretion in imposing penalties for criminal offenses at the Founding. (231) Instead, a criminal statute specified a particular punishment to be imposed on all defendants convicted of the offense, irrespective of aggravating and mitigating factors. (232) More recently, the Court has recognized that some offenses at the Founding did authorize sentencing ranges. (233) It accordingly has adopted a new historical argument, stating that facts that modified those ranges constitute elements of the crime but that facts affecting the sentence within a range do not. (234)

As the evolving nature of the Court's analysis suggests, the Court's historical argument is not sound. Indeed, the Court's characterization of the history of judicial discretion at sentencing has varied wildly in recent years. (235) Other recent cases have stated that, at the Founding, judges had broad discretion at sentencing, (236) and it has relied on that view of history to support its conclusion that procedural rights do not apply to discretionary sentencing systems. (237)

The Court's reliance on history to apportion procedural rights is also problematic because the state of judicial discretion at sentencing in the American colonies and immediately after the ratification of the Constitution simply is not clear. (238) Whether a judge had discretion at sentencing often appears to have depended on the type of offense. Felonies usually carried specific sentences. Sentencing for felonies thus was not a separate phase of the trial; instead, it was part of the trial phase aimed at finding innocence or guilt. (239) By contrast, judges did usually have discretion in sentencing for misdemeanors. For those crimes, judges could choose among various penalties, such as fines, imprisonment, or corporal punishment. (240) They also had discretion to set the amount of the punishment, such as the number of lashes or the years of imprisonment. (241)

Most important, substantial evidence suggests that the Court's historical account in support of equating certain facts with elements is wrong. A number of commentators have undertaken exhaustive histories exploring the link between those facts that affect available punishment ranges and elements of criminal offenses. (242) Their findings demonstrate that not all facts that changed the range of punishment were treated as elements. For example, Jonathan Mitchell's extensive survey of nineteenth-century cases demonstrates that courts repeatedly distinguished between "elements" and facts that increased the maximum allowable sentence; it also demonstrates that the jury right was traditionally understood to extend beyond elements to other facts that could affect punishment. (243) History therefore does not provide a firm foundation for providing more procedural protections in mandatory sentencing systems than in discretionary ones.

Even if the Court's historical argument were correct, the Court has often not followed history in defining constitutional rights at sentencing. (244) In recent decades, the Court has recognized many constitutional limitations on sentencing that had not been previously established. (245) For example, there is no historical foundation for a right to notice of possible enhancements under a mandatory sentencing regime because mandatory guidelines did not exist at common law. (246) Nor is there a historical foundation for the Court's holding that the Sixth Amendment requires the appointment of counsel for indigents, (247) as well as the effective assistance of counsel, (248) at sentencing, because neither the right to appointed counsel nor the right to effective assistance was recognized at the Founding. Likewise, history does not support the present requirement that the death penalty be imposed only upon additional findings of aggravation; (249) historically, death was automatic for many felonies. (250) Indeed, the Court at one time acknowledged that there was no firm historical basis for its conclusion that the Sixth Amendment and Due Process Clause together require that a jury find beyond a reasonable doubt any fact that increases the maximum possible punishment. (251)


The inconsistent enforcement of procedural rights in mandatory and discretionary sentencing schemes is unwarranted. Courts have justified the practice on the grounds that procedural protections are inconsistent with the goals of discretionary systems, that mandatory systems create stronger expectations in defendants that need greater protection, and that historically defendants in mandatory schemes received greater protections. None of these arguments are satisfying. Enforcing procedural protections in discretionary systems would impair the ability to impose sentence no more than they do in mandatory schemes and would enhance many of the determinations made in discretionary systems. Discretionary systems create strong expectations for defendants, and history does not support different treatment in mandatory and discretionary schemes.

Instead of turning on whether the judge has discretion in imposing sentence, the determination whether a procedural right should be required at sentencing ought to depend on the reason for the procedure. Different procedural rights serve different purposes, and some seek to achieve multiple goals. For example, the right to confrontation increases the reliability of evidence against the defendant in court, while the right to proof beyond a reasonable doubt ensures that errors in assessing the evidence are resolved in the defendant's favor and protects government legitimacy. Thus, it may be more important to enforce confrontation rights at sentencing than to require that sentencing facts be proven beyond a reasonable doubt.

Our goal here is not to determine which procedural rights should apply at sentencing. That is a large and complicated question whose resolution is beyond the scope of this Article. (252) Instead, our point is that each procedural right ought to be examined on its own terms to determine whether it should be enforced at sentencing. That a judge has discretion at sentencing rarely, if ever, provides a sound reason for not affording a procedural right and often is a reason to provide more procedural protections. Courts in the last few decades have begun to recognize the need for inquiring into whether various procedural rights should extend to sentencing. (253) But they still regularly answer those questions based solely on whether the sentencing system is discretionary.

Shifting the inquiry to the reasons underlying the right instead of focusing on discretion would result in a more sensible assessment of which procedural rights should be extended to sentencing. It would also avoid the confusion over how much sentencing discretion judges must possess in order to avoid the enforcement of procedural rights. It may also increase legitimacy. The current focus on discretion has led several jurisdictions to increase judicial discretion over sentencing decisions precisely to avoid the procedural protections recognized by the Supreme Court in mandatory systems. (254)

In addition to shifting the focus of sentencing rights questions to the reasons underlying various rights, this Article is intended to cause courts and commentators to think more carefully about the nature of judicial sentencing discretion. A more nuanced discussion about how judicial discretion plays out in the sentencing process may lead not only to better decisions about the importance of procedural rights at sentencing, it may also help inform the national discussion about how best to reform sentencing practices in light of the Blakely and Booker decisions.

(1) E.g., Glover v. United States, 531 U.S. 198, 203-204 (2001) (right to effective assistance of counsel); Mitchell v. United States, 526 U.S. 314, 321-22 (1999) (privilege against self-incrimination); Mempa v. Rhay, 389 U.S. 128, 137 (1967) (right to counsel); see also Alan C. Michaels, Trial Rights at Sentencing, 81 N.C. L. Rev. 1771 (2003) (providing a comprehensive view of procedural rights at sentencing).

(2) Apprendi v. New Jersey, 530 U.S. 466, 490 (2000).

(3) Id.

(4) Burns v. United States, 501 U.S. 129, 138 (1991).

(5) Peugh v. United States, 133 S. Ct. 2072, 2085 (2013).

(6) See infra notes 81-90 and accompanying text.

(7) Compare Irizarry v. United States, 553 U.S. 708, (2008), and United States v. Booker, 543 U.S. 220, 245 (2005), with Peugh v. United States, 133 S. Ct. 2072, 2086 (2013).

(8) 542 U.S. 296 (2004).

(9) 543 U.S. 220 (2005).

(10) Carissa Byrne Hessick, Motive's Role in Criminal Punishment, 80 S. Cal. L. Rev. 89, 131 & n.183 (2006).

(11) Carissa Byrne Hessick & F. Andrew Hessick, Recognizing Constitutional Rights at Sentencing, 99 Cal. L. Rev. 47, 52 (2011) [hereinafter Hessick & Hessick, Constitutional Rights].

(12) See Kate Stith & Jose A. Cabranes, Fear of Judging 22 (1998).

(13) See id. at 14-15, 23 (recounting the broad discretion judges historically possessed regarding the identification and assessment of sentencing factors); Hessick & Hessick, Constitutional Rights, supra note 11, at 51-57 (same).

(14) See Pierce O'Donnell et al., Towards a Just and Effective Sentencing System 2-3 (1977) (noting that "there is no requirement that the sentence have any rational basis whatsoever"); Douglas A. Berman & Stephanos Bibas, Making Sentencing Sensible, 4 Ohio St. J. Crim. L. 37, 61 (2006) ("From the nineteenth through the mid-twentieth century, sentencing judges enjoyed almost unfettered discretion.... Sentencing judges were free to select any sentence within the range for any reason or no reason at all.").

(15) Townsend v. Burke, 334 U.S. 736, 740-41 (1948); see also United States v. Tucker, 404 U.S. 443, 447 (1972) (reaffirming Townsend).

(16) See United States v. Main, 598 F.2d 1086, 1094 (7th Cir. 1979) ("'[O]nce it is determined that a sentence is within the limitations set forth in the statute under which it is imposed, appellate review is at an end (footnote and citations omitted),' unless the sentencing judge relied on improper or unreliable information in exercising his or her discretion, or failed to exercise any discretion at all, in imposing sentence." (citations omitted)); see also Hessick & Hessick, Constitutional Rights, supra note 11, at 54-56 (collecting cases identifying improper sentencing factors).

(17) See, e.g., United States v. Hartford, 489 F.2d 652, 655 (5th Cir. 1974); United States v. Thompson, 483 F.2d 527, 529 (3d Cir. 1973); Woosley v. United States, 478 F.2d 139, 143 (8th Cir. 1973); United States v. Daniels, 446 F.2d 967, 970-71 (6th Cir. 1971); United States v. McCoy, 429 F.2d 739, 743 (D.C. Cir. 1970). Although federal appellate courts of this era generally declined to review sentences, see infra text accompanying notes 19-20, the courts in these cases justified their intervention on the theory that, when the sentencing court does not actually exercise its discretion in imposing sentence, then "the appellate court does not usurp the discretion vested in the district court when it reviews the sentence." United States v. Bates, 852 F.2d 212, 220 (7th Cir. 1988).

(18) See, e.g., State v. Kunz, 259 A.2d 895, 901-902 (N.J. 1969); Montalto v. State, 199 N.E. 198, 200 (Ohio Ct. App. 1935); see also Comment, Appellate Review of Primary Sentencing Decisions: A Connecticut Case Study, 69 Yale L.J. 1453, 1453 (1960) (noting that eleven states provided some form of review of criminal sentences).

(19) Stith & Cabranes, supra note 12, at 197 n.3; Symposium, Appellate Review of Sentences, 32 F.R.D. 249, 259 (1962) (statement of Irving R. Kaufman, Judge, U.S. Court of Appeals for the Second Circuit).

(20) Dorszynski v. United States, 418 U.S. 424, 431 (1974); see also Koon v. United States, 518 U.S. 81, 96 (1996) ("Before the Guidelines system, a federal criminal sentence within statutory limits was, for all practical purposes, not reviewable on appeal.").

(21) O'Donnell et al., supra note 14, at 2-3 (noting that due process protections, including a statement of reasons for government action, that apply to agency action appear not to govern federal sentencing).

(22) Stith & Cabranes, supra note 12, at 31-34.

(23) See S. Rep. No. 98-225, at 74-75 (1983), reprinted in 1984 U.S.C.C.A.N. 3182, 3257-58.

(24) Stephen Breyer, Federal Sentencing Guidelines Revisited, 11 Fed. Sent'g Rep. 180, 180 (1999).

(25) See, e.g., Stith & Cabranes, supra note 12, at 38-48.

(26) Notably, a significant number of states retained discretionary sentencing systems. See Stephanos Bibas & Susan Klein, The Sixth Amendment and Criminal Sentencing, 30 Cardozo L. Rev. 775, 785 & n.44 (2008) (noting that, as of 2004, seventeen states "granted judges unfettered discretion to sentence within a range").

(27) See id. at 785-86; Richard S. Frase, Sentencing Guidelines in the States: Lessons for States and Federal Reformers, 6 Fed. Sent'g Rep. 123, 123 (1993).

(28) 18 U.S.C. [section] 1341 (2012) (stating that a person convicted of fraud under this statute "shall be fined under this title or imprisoned not more than 20 years, or both"); see also United States v. Tucker, 404 U.S. 443, 446 (1972) ("[A] trial judge in the federal judicial system generally has wide discretion in determining what sentence to impose.").

(29) See Stith & Cabranes, supra note 12, at 192-93 app. D (giving the example of Martin Miller).

(30) See U.S. Sentencing Guidelines Manual [section] 5K1.1-5K2.24 (2004).

(31) 18 U.S.C. [section] 3553(b)(1) (2012). For an account of how trial judges' authority to impose sentences outside of the Guidelines was limited, see Amy Baron-Evans & Kate Stith, Booker Rules, 160 U. Pa. L. Rev. 1631, 1646-57 (2012).

(32) Notably, some commentators claimed that arbitrariness and inequality remained in the new systems, and others began to express concern that these structured sentencing systems gave too much power to prosecutors and/or treated dissimilar offenders the same. See, e.g., Daniel J. Freed, Federal Sentencing in the Wake of Guidelines: Unacceptable Limits on the Discretion of Sentencers, 101 Yale L.J. 1681 (1992); Charles J. Ogletree, Jr., The Death of Discretion? Reflections on the Federal Sentencing Guidelines, 101 Harv. L. Rev. 1938 (1988); Kate Stith, Two Fronts for Sentencing Reform, 20 Fed. Sent'g Rep. 343 (2008); see also Deborah Young, Fact-Finding at Federal Sentencing: Why the Guidelines Should Meet the Rules, 79 Cornell L. Rev. 299, 303 n.16 (1994) (collecting sources).

(33) See Gerald Leonard & Christine Dieter, Punishment Without Conviction: Controlling the Use of Unconvicted Conduct in Federal Sentencing, 17 Berkeley J. Crim. L. 260, 261 (2012) (noting that certain sentencing practices had long "occurred quietly in trial courts, where no law governed the practice and no appellate review disciplined the outcomes," and "only began to provoke sustained outrage among legal scholars after the advent of the U.S. Sentencing Guidelines" when they became "open and obvious").

(34) E.g., Sara Sun Beale, Procedural Issues Raised by Guidelines Sentencing: The Constitutional Significance of the "Elements of the Sentence", 35 Wm. & Mary L. Rev. 147 (1993); Susan N. Herman, The Tail That Wagged the Dog: Bifurcated Fact-finding Under the Federal Sentencing Guidelines, 66 S. Cal. L. Rev. 289 (1992); Elizabeth T. Lear, Is Conviction Irrelevant?, 40 UCLA L. Rev. 1179 (1993); Young, supra note 32; Stephanie C. Slatkin, Note, The Standard of Proof at Sentencing Hearings Under the Federal Sentencing Guidelines: Why the Preponderance of the Evidence Standard Is Constitutionally Inadequate, 1997 U. ILL. L. Rev. 583.

(35) For example, the Supreme Court recognized the right to counsel at sentencing in 1967. Mempa v. Rhay, 389 U.S. 128, 137 (1967).

(36) See, e.g., Glover v. United States, 531 U.S. 198, 203 (2001) (recognizing the right to effective assistance of counsel at sentencing); Mitchell v. United States, 526 U.S. 314, 330 (1999) (recognizing the privilege against self-incrimination at sentencing in 1999). Lower court decisions affirming various procedural rights at sentencing also appear to be more common after the rise of mandatory sentencing. See Michaels, supra note 1, at 1816 & n.194 (collecting cases on the due process right to discovery); id. at 1828-29 (collecting cases on the speedy trial right).

(37) 530 U.S. 466 (2000).

(38) Id. at 490.

(39) 542 U.S. 296 (2004).

(40) Id. at 303.

(41) Id. at 305.

(42) 543 U.S. 220 (2005).

(43) Id. at 259.

(44) Id.

(45) Peugh V. United States, 133 S. Ct. 2072, 2084 (2013).

(46) See infra notes 81-90 and accompanying text.

(47) See Harris v. United States, 536 U.S. 545 (2002). The expectation argument rests on the premise that mandatory schemes create certain expectations for defendants. According to the Court, when a guilty verdict authorizes a sentencing range, a defendant expects to receive a sentence within that range; if a law provides that the maximum sentence may be increased only upon additional factual findings, that limit on increasing the maximum creates an entitlement in the defendant that a higher sentence not be imposed on the defendant without sufficient procedural protections. Id. at 562 (plurality opinion) (noting that a defendant cannot predict precisely what sentence he will receive, he can only predict the "heaviest punishment" that he may face if he is convicted). A version of the expectation argument can also be found in Blakely v. Washington, 542 U.S. 296, 309 (2004).

(48) Alleyne v. United States, 133 S. Ct. 2151 (2013) (overturning Harris and rejecting by implication Harris's reliance on the expectation theory and relying on historical arguments instead).

(49) Jones v. United States, 526 U.S. 227, 244 (1999) (noting uncertainty over whether the Framers would have understood the Sixth Amendment as requiring the approach taken by the modern Court).

(50) Apprendi v. New Jersey, 530 U.S. 466, 481 (2000) ("[N]othing in this history suggests that it is impermissible for judges to exercise discretion--taking into consideration various factors relating both to offense and offender--in imposing a judgment within the range prescribed by statute."); see also United States v. Booker, 543 U.S. 220, 233 (2005) ("We have never doubted the authority of a judge to exercise broad discretion in imposing a sentence within a statutory range."); Blakely v. Washington, 542 U.S. 296, 308-309 (2004) ("[The Sixth Amendment] limits judicial power only to the extent that the claimed judicial power infringes on the province of the jury. Indeterminate sentencing does not do so. It increases judicial discretion, to be sure, but not at the expense of the jury's traditional function of finding the facts essential to lawful imposition of the penalty.").

(51) 1 Webster's New International Dictionary of the English Language 745 (2d ed. 1949); see also Black's Law Dictionary 534 (9th ed. 2009) (defining discretion as "individual judgment; the power of free decision-making"); 3 Oxford English Dictionary 435 (Dr. James A. H. Murray ed. 1897) (defining discretion as "liberty or power of deciding ... according to one's own judgment").

(52) Henry J. Friendly, Indiscretion About Discretion, 31 Emory L.J. 747, 754-55 (1982); Maurice Rosenberg, Judicial Discretion of the Trial Court, Viewed from Above, 22 Syracuse L. Rev. 635, 636-43 (1971). Robert Post identifies a third way in which the term discretion is often used: the extent to which a decision maker is required to use particular procedures in the decision process--in particular, procedures that enhance reviewability. Robert C. Post, The Management of Speech: Discretion and Rights, 1984 Sup. Ct. Rev. 169, 207ndash;08, 218-21. Because our exploration of discretion at sentencing focuses on whether the presence of discretion eliminates the need for certain procedures, we do not include this definition of sentencing in the discussion.

(53) See, e.g., Roscoe Pound, Discretion, Dispensation and Mitigation: The Problem of the Individual Special Case, 35 N.Y.U. L. Rev. 925, 930 (1960) (discussing "discretion guided by the analogy of principles of law as starting points for reasoned determination [s]"). One of the major distinctions between rules and standards is that rules specify the content of the law ex ante, while standards leave the precise scope of the law to be clarified ex post. See Louis Kaplow, Rules Versus Standards: An Economic Analysis, 42 Duke L.J. 557, 559-61 (1992).

(54) Cass R. Sunstein, Problems with Rules, 83 Calif. L. Rev. 953, 964-65 (1995).

(55) See Kaplow, supra note 54, at 559-61.

(56) Friendly, supra note 53, at 765 (quoting Citizens to Pres. Overton Park, Inc. v. Volpe, 401 U.S. 402, 410 (1971)). While Pound stated that there were many such situations, Pound, supra note 54, at 929, Friendly stated that "[v]ery few cases, however, fall into this category," Friendly, supra note 53, at 765. There are a number of examples of broad grants of explicit discretion in administrative law, where agency action is not subject to judicial review "where 'agency action is committed to agency discretion by law.'" Overton Park, 401 U.S. at 410 (quoting Administrative Procedure Act [section] 701, 5 U.S.C. [section] 701 (1964)). Unreviewable agency action--that is, an action that is "committed to agency discretion"--exists "in those rare instances where 'statutes are drawn in such broad terms that in a given case there is no law to apply.' " Overton Park, 401 U.S. at 410 (quoting S. Rep. No. 79-752, at 212 (1945)).

(57) Pound, supra note 54, at 929. He gives the example of "proceedings for custody of children where compelling consideration cannot be reduced to rules, judicial determination must be left, to no small extent, to the disciplined but no less personal feeling of the judge for what justice demands." Id.

(58) See Exec. Software N. Am., Inc. v. United States Dist. Court, 24 F.3d 1545, 1562 (9th Cir. 1994) (holding that once a case is properly removed, 28 U.S.C. [section] 1367(c) "provides the exclusive means by which supplemental jurisdiction over pendent claims may be declined").

(59) Friendly, supra note 53, at 754; Rosenberg, supra note 53, at 637.

(60) See, e.g., 28 U.S.C. [section] 1447(d) (2012) (generally forbidding appellate review of orders remanding to state court cases removed to federal court).

(61) De novo review is generally reserved for questions of law. Harry T. Edwards & Linda A. Elliott, Federal Courts Standards of Review 3 (2007). Courts also employ de novo review in certain limited contexts to factual findings, see Bose Corp. v. Consumers Union of U.S., Inc., 466 U.S. 485, 508 n.27 (1984), and applications of law to fact, see Ornelas v. United States, 517 U.S. 690, 697 (1996).

(62) See generally Edwards & Elliott, supra note 62, at 19-22, 91-94 (discussing intermediate standards of review such as clear error, abuse of discretion, and plain error).

(63) See Friendly, supra note 53, at 765.

(64) See, e.g., Ornelas, 517 U.S. at 697 (reviewing determinations of probable cause de novo).

(65) For example, 28 U.S.C. [section] 1447(c) requires district courts to remand cases removed from state court if they determine that they lack subject-matter jurisdiction over the case, but [section] 1447(d) prohibits appellate review of those remands. 28 U.S.C. [section] 1447(d).

(66) See supra notes 22-23 and accompanying text.

(67) Our typology borrows from Paul H. Robinson and Barbara A. Spellman. See Paul H. Robinson & Barbara A. Spellman, Sentencing Decisions: Matching the Decisionmaker to the Decision Nature, 105 Colum. L. Rev. 1124 (2005). Robinson and Spellman identify six separate decisions. Id. at 1128-32.

(68) See F. Andrew Hessick & Carissa Byrne Hessick, The Non-Redelegation Doctrine, 55 Wm. & Mary L. Rev. 163, 203 (2013) [hereinafter Hessick & Hessick, Non-Redelegation].

(69) See Frank O. Bowman, III, Debacle: How the Supreme Court Has Mangled American Sentencing Law and How h Might Yet Be Mended, 77 U. Chi. L. Rev. 367, 371 (2010).

(70) See Robinson & Spellman, supra note 68, at 1130; see also Kyron Huigens, Solving the Williams Puzzle, 105 Colum. L. Rev. 1048, 1060-63 (2005).

(71) See Huigens, supra note 71, at 1060-61.

(72) Arguably, the application of policy to facts is itself a policy determination, for a judge applying a policy to a defendant makes a determination about the policy that should apply to this defendant. But because judges ordinarily do not conceive of application of policy to facts in this way, we treat it separately.

(73) Of course, factfinding itself involves some degree of discretion because reasonable people may differ on whether the burden of proof is satisfied. See Frank O. Bowman, III, Fear of Law: Thoughts on Fear of Judging and the State of the Federal Sentencing Guidelines, 44 St. Louis U. L.J. 299, 338 (2000); Hessick & Hessick, Non-Redelegation, supra note 69, at 180. But that discretion is inevitably less than the broad explicit discretion with respect to policy and application decisions that judges in discretionary systems enjoy.

(74) After Blakely and Booker, the judge is no longer authorized to make at least some of these factual findings. Facts that increase the maximum punishment, United States v. Booker, 543 U.S. 220, 232 (2005), and facts that trigger mandatory minimums, Alleyne v. United States, 133 S. Ct. 2151, 2158-63 (2013), must be found by juries.

(75) For cases indicating that judges were required to make factual findings in fully discretionary systems, see supra note 17 and accompanying text. But cf. Jonathan F. Mitchell, Apprendi's Domain, 2006 Sup. Ct. Rev. 297, 345 (stating that discretionary sentencing regimes "allow judges to base a defendant's sentence on factual determinations that were not submitted to a jury, but do not require them to do so").

(76) The degree of discretion possessed by trial judges over policy and application decisions varies across discretionary sentencing systems. Some sentencing systems do not leave the broad policy determinations to individual judges while other systems essentially commit those broad decisions to the discretion of individual judges. The articulation of sentencing rules also varies across systems. Some systems leave the articulation of all sentencing rules to the discretion of individual judges. Other systems identify relevant sentencing factors, but leave the weight of those factors to individual judges.

(77) Nancy Gertner, Supporting Advisory Guidelines, 3 Harv. L. & Pol'y Rev. 261, 263 (2009) ("Judges supposedly served as experts who knew how best to individualize sentences to reflect the goals of punishment, including rehabilitation and deterrence."); Meghan J. Ryan, The Missing fury: The Neglected Role of Juries in Eighth Amendment Punishments Clause Determinations, 64 Fla. L. Rev. 549, 584 (2012) ("[B]ecause the goals of retribution, deterrence, incapacitation, and rehabilitation are at the heart of sentencing, individualization is essential to accurately concentrate punishment on these specific goals.").

(78) Cf. Kaplow, supra note 54, at 622 ("When legislators leave the details of law to courts ... individuals may be left with little guidance for years or decades."). As Louis Kaplow explains, because complex rules are more costly to promulgate than complex standards, jurisdictions ordinarily promulgate either simple rules or complex standards. See id. at 588-93.

(79) To be sure, most sentencing systems permit at least some flexibility for judges to reduce sentences based on such extraordinary circumstances. See infra note 81. But that reflects only that all systems have reserved some explicit discretion for judges over policy and application decisions.

(80) See supra notes 76-79 and accompanying text. For example, under the Federal Sentencing Guidelines, judges were sometimes permitted to sentence based on policy considerations that were not spelled out in the Guidelines. See supra text accompanying notes 28-31. Similarly, under the Federal Sentencing Guidelines, judges had near-unfettered discretion to select a particular sentence within the range prescribed by the Guidelines--that is, they retained some discretion over application. U.S. Sentencing Guidelines Manual [section] 5C1.1 (a) (2004).

(81) E.g., 18 U.S.C. [section] 3553(b)(1) (2012); Ariz. Rev. Stat. Ann. [section] 13-701 (E)(6) (2014).

(82) For example, in the post-Booker advisory guideline system, judicial discretion over sentencing policy is limited by 18 U.S.C. [section] 3553(a), which lists the 7 factors that judges must consider in imposing a sentence. One of those seven factors is the Sentencing Guidelines, which trial courts must use as the starting point in their sentencing analysis. See Gall v. United States, 552 U.S. 38, 49, 50 n.6 (2007).

(83) See, e.g., People v. Black, 113 P.3d 534, 548 (Cal. 2005) (holding that a jury trial right did not apply to California's sentencing system because the "discretion available to a California judge ... appears comparable to the level of discretion that the high court has chosen to permit federal judges in post- Hooker sentencing"). That decision was reversed by the Supreme Court in Cunningham v. California, which disagreed with the California Supreme Court's conclusion that the discretion afforded California judges was sufficiently similar to that afforded federal judges after Booker. See 549 U.S. 270 (2007).

(84) See, e.g., Stephanos Bibas et al., Policing Politics at Sentencing, 103 Nw. U. L. Rev. 1371, 1372 (2009) ("Booker left unclear exactly how loose appellate review must be to satisfy the Sixth Amendment.").

(85) 542 U.S. 296 (2004).

(86) Id. at 299.

(87) See id.

(88) The Blakely Court reasoned that it was "immaterial" that a judge could increase a sentence based on "any aggravating fact," as opposed to only certain enumerated facts, because "it remains the case that the jury's verdict alone does not authorize the sentence." Id. at 305. In other words, the Blakely Court concluded that the necessity of any factual finding was what rendered the system mandatory. But as explained above, factual findings are also required in discretionary systems. See supra notes 70-76 and accompanying text.

(89) These procedural rights are not the only ones that receive different treatment in discretionary and mandatory schemes, see infra notes 199-200 and accompanying text, but they illustrate the phenomenon.

(90) Stephen Breyer, Active Liberty 7-8 (2005) (noting all judges use these, and other, basic interpretive tools but place differing levels of emphasis on each).

(91) E.g., Irizarry v. United States, 553 U.S. 708, 713-14 (2008).

(92) E.g., United States v. Booker, 543 U.S. 220, 233 (2005) (stating that the Court has "never doubted" that the right to a jury trial of sentencing facts does not apply in discretionary systems).

(93) U.S. Const, amend. VI. Although the right refers to "all criminal prosecutions," the Court has held that it generally applies only for those offenses that carry a possible sentence of more than six months of imprisonment. Baldwin v. New York, 399 U.S. 66, 73-74 (1970).

(94) See United States v. Dougherty, 473 F.2d 1113, 1132 (D.C. Cir. 1972).

(95) See Foucha v. Louisiana, 504 U.S. 71, 80 (1992) (describing "[f]reedom from bodily restraint" based on "arbitrary governmental action" as "the core of ... the Due Process Clause"); Hessick & Hessick, Constitutional Rights, supra note 11, at 94.

(96) See Laura I. Appleman, The Lost Meaning of the Jury Trial Right, 84 Ind. L.J. 397, 405-40 (2009).

(97) Richard H. Fallon, Jr., The Core of an Uneasy Case for Judicial Review, 121 Harv. L. Rev. 1693, 1697 (2008) ("Virtually without exception, judges and Justices are well-educated members of the upper or upper-middle classes who have been socialized to accept professional norms.").

(98) Emerson H. Tiller & Frank B. Cross, A Modest Proposal for Improving American Justice, 99 Colum. L. Rev. 215, 215 (1999) (arguing that the life tenure of federal judges makes them politically unaccountable).

(99) See generally Vikram David Amar, Jury Service as Political Participation Akin to Voting, 80 Cornell L. Rev. 203 (1995) (discussing how the jury injects popularism into judicial decisions). It is for this reason that punitive damage assessments and many other determinations based on value judgments are assigned to juries. See George L. Priest, Justifying the Civil Jury, in Verdict 103, 106-107 (Robert E. Litan ed., 1993).

(100) Apprendi v. New Jersey, 530 U.S. 466, 490 (2000). Thus, in Apprendi, the Court held that the Sixth Amendment prohibited a judge from making factual findings that would increase a maximum possible sentence from ten years to twenty years. Id.

(101) Alleyne v. United States, 133 S. Ct. 2151, 2163 (2013).

(102) See supra text accompanying notes 37-45.

(103) E.g., Alleyne, 133 S. Ct., at 2161 n.2, 2163; United States v. Booker, 543 U.S. 220, 233 (2005).

(104) E.g., Alleyne, 133 S. Ct., 2161 n.2, 2163.

(105) See id. at 2163 ("[Establishing what punishment is available by law and setting a specific punishment within the bounds that the law has prescribed are two different things." (quoting Apprendi, 530 U.S. at 519 (Thomas, J., concurring) (internal quotation marks omitted)).

(106) Williams v. New York, 337 U.S. 241, 247-51 (1949).

(107) See supra notes 69-71 and accompanying text.

(108) See supra note 17 and accompanying text.

(109) See supra Section II.B; see also Robinson & Spellman, supra note 68, at 1129-32 (cataloguing the various distinct decisions that are required to impose a sentence in a particular case).

(110) See Hessick & Hessick, Non-Redelegation, supra note 69, at 202-03 (explaining how different policy decisions could result in the imposition of different sentences based on the same facts). Indeed, far from providing a basis to disregard the jury right for factual findings in discretionary schemes, Apprendi's logic that the jury right applies to findings resulting in an increased sentencing range suggests the right should extend in discretionary schemes to policy determinations when those determinations themselves provide the justification for increasing sentencing ranges in discretionary schemes. See generally Adriaan Lanni, Jury Sentencing in Noncapital Cases: An Idea Whose Time Has Come (Again)?, 108 Yale L.J. 1775 (1999) (arguing that jury sentencing may be the most effective means of achieving sentencing goals). Assigning such judgments to juries would also be consistent with the practice in civil cases. In civil cases for damages, there is often a wide range of possible damages that may be awarded. How much is awarded depends on a variety of considerations, and reasonable people may disagree on the appropriate amount to award. The amount of discretion is especially high in cases in which punitive damages are sought. That the Supreme Court has made it clear that these determinations are to be made by juries compels a similar conclusion. See Pac. Mut. Life Ins. Co. v. Haslip, 499 U.S. 1 (1991). In Haslip, the Supreme Court upheld a large punitive damages verdict against an insurance company for acting in bad faith. Id. at 15-16 ("Under the traditional common-law approach, the amount of the punitive award is initially determined by a jury instructed to consider the gravity of the wrong and the need to deter similar wrongful conduct.").

(111) But cf. supra note 80 (noting that judges in mandatory systems retain some discretion over policy and application decisions).

(112) Of course, a judge may inject bias through factual findings. See Hessick & Hessick, Non-Redelegation, supra note 69, at 179. But the opportunity for injecting bias is less than when the judge is granted discretion by law.

(113) Cf. supra note 14.

(114) See New York v. United States, 342 U.S. 882, 884 (1951) (per curiam) (Douglas, J., dissenting) ("Absolute discretion ... marks the beginning of the end of liberty.").

(115) See, e.g., Marvin E. Frankel, Criminal Sentences 21 (1972); Judicial Conference for the Second Circuit, Symposium, Appellate Review of Sentences, 32 F.R.D. 249, 265-66 (1962) (statement of Simon E. Sobeloff, Chief Judge of the United States Court of Appeals for the Fourth Circuit).

(116) Koon v. United States, 518 U.S. 81, 113 (1996).

(117) See Hessick & Hessick, Non-Redelegation, supra note 69, at 207-208. But see Carissa Byrne Hessick, Mandatory Minimums and Popular Punitiveness, 2011 Cardozo L. Rev. de novo 23, 24-27 (noting a number of information deficits that lead legislatures to adopt more severe sentencing policies than those that would be preferred by citizens in individual cases).

(118) In re Winship, 397 U.S. 358, 364 (1970). Although the Court did not deem the beyond a reasonable doubt standard a constitutional requirement until 1970, that standard was commonly used since as early as 1798. Jon O. Newman, Beyond "Reasonable Doubt", 68 N.Y.U. L. Rev. 979, 981-82 (1993).

(119) In re Winship, 397 U.S. at 362-63.

(120) See Newman, supra note 120, at 984.

(121) In re Winship, 397 U.S. at 363-64; id. at 372 (Harlan, J., concurring) ("[I]t is far worse to convict an innocent man than to let a guilty man go free.").

(122) In re Winship, 397 U.S. at 364 (majority opinion).

(123) Apprendi v. New Jersey, 530 U.S. 466, 490 (2000).

(124) Id.

(125) United States v. O'Brien, 560 U.S. 218, 224 (2010); McMillan v. Pennsylvania, 477 U.S. 79, 91 (1986) ("Sentencing courts ... traditionally heard evidence and found facts without any prescribed burden of proof at all."). That said, circuit courts have generally concluded that due process imposes a preponderance standard at sentencing. See Kinder v. United States, 504 U.S. 946, 948 (1992) (White, J., dissenting from denial of certiorari) (collecting cases). Although the Court has not imposed a burden of proof, it has said that a sentence may not be imposed based on information that is materially false or incorrect. See United States v. Tucker, 404 U.S. 443, 447 (1972).

(126) See supra Section II.B, III A.

(127) One criticism of discretionary sentencing is that some decisionmakers rely only on intuition at the expense of thoughtful consideration of particular facts and policies. See Kenneth Culp Davis, Discretionary Justice 5 (1969) ("[M]ost discretionary decisions are intuitive, and responses to influences often tend to crowd out thinking about values."). To be sure, even an assessment that is based on gut feeling is still the product of factual findings. Gut feelings are essentially a series of unarticulated, if not subconscious, factual conclusions.

(128) Counterintuitively, those who argue against procedural protections at sentencing often do so by contending that such protections will result in less accurate and reliable sentences. See infra subsection IV.A.1.

(129) See Stephen A. Fennell & William N. Hall, Due Process at Sentencing: An Empirical and Legal Analysis of the Disclosure of Presentence Reports in Federal Courts, 93 Harv. L. Rev. 1613, 1628-30 (1980) (recounting such cases).

(130) See Smith v. Goguen, 415 U.S. 566, 572 & n.8 (1974).

(131) Cf. Paul H. Robinson, Fair Notice and Fair Adjudication: Two Kinds of Legality, 154 U. Pa. L. Rev. 335, 336 (2005) ("[T]he traditional legality principle analysis actually conflates two distinct issues: one relating to the ex ante need for fair notice, the other to the ex post concern for fair adjudication.").

(132) See Goguen, 415 U.S. at 572-73.

(133) See, e.g., Papachristou v. City of Jacksonville, 405 U.S. 156, 168 (1972).

(134) See Mullane v. Cent. Hanover Bank & Trust Co., 339 U.S. 306, 313 (1950) ("[T]he Due Process Clause ... require [s] that deprivation of life, liberty or property by adjudication be preceded by notice and opportunity for hearing appropriate to the nature of the case."); In re Oliver, 333 U.S. 257, 273 (1948) ("A person's right to reasonable notice of a charge against him, and an opportunity to be heard in his defense--a right to his day in court--are basic in our system of jurisprudence.").

(135) Joint Anti-Fascist Refugee Comm. v. McGrath, 341 U.S. 123, 171-72 (1951) (Frankfurter, J., concurring).

(136) Id.

(137) Henry J. Friendly, "Some Kind of Hearing", 123 U. Pa. L. Rev. 1267, 1280-81 (1975) ("It is ... fundamental that notice be given and that it be timely and clearly inform the individual of the proposed action and the grounds for it.").

(138) Id. The amount of process necessary is not set but may vary depending on the interest of the individual, the cost of providing more process, and the likelihood that additional process will increase the accuracy of the determination. See Mathews v. Eldridge, 424 U.S. 319, 334, 347-49 (1976).

(139) 501 U.S. 129, 138 (1991).

(140) Id. at 134.

(141) Id.

(142) After Williams, Federal Rule of Criminal Procedure 32 was modified to provide for both disclosure of the information contained in the presentence report, as well as an opportunity for the defendant to respond. See Fennell & Hall, supra note 131, at 1632-35 (describing the fight over disclosure and the federal rules).

(143) Burns, 501 U.S at 138.

(144) Henry J. Friendly, Benchmarks 210-11 (1967); Frederick Schauer, Ashwander Revisited, 1995 Sup. Ct. Rev. 71, 83.

(145) Schauer, supra note 146, at 83.

(146) 553 U.S. 708, 713-14 (2008). Even before Irizarry, lower courts had consistently held that offenders have no right to notice of information that may be used at sentencing under discretionary schemes. See Fennell & Hall, supra note 131, at 1631 n.86 (collecting cases). Those decisions derive from Williams v. New York. 337 U.S. 241 (1949). There, New York law granted judges broad discretion to impose sentence based on a number of different factors. Id. at 245. Williams challenged his sentence on, inter alia, the ground that he had not been provided notice of the information on which his sentence was based. Id. Although the Williams Court did not address the notice issue, Note, Procedural Due Process at Judicial Sentencing for Felony, 81 Harv. L. Rev. 821, 827 (1968), courts have relied on Williams as establishing that defendants have no right to the disclosure of sentencing information in discretionary schemes, Fennell & Hall, supra note 131, at 1631 n.86 (collecting cases). Although defendants ordinarily do not have a right to notice in discretionary schemes, the Court has concluded that juveniles do have a due process right of notice regarding the information that will form the basis of a sentencing decision, Kent v. United States, 383 U.S. 541, 560 (1966) (finding a due process right to disclosure and a right to be heard in juvenile transfer proceedings), as do defendants facing capital punishment, Gardner v. Florida, 430 U.S. 349, 362 (1977) (plurality opinion) (finding a due process violation where a capital defendant was sentenced based on information in a presentence report not disclosed to defense counsel).

(147) Irizarry, 553 U.S. at 713-14.

(148) Our purpose here is not to argue that systems that confer complete sentencing discretion on trial judges are unconstitutional. But it is worth noting that those systems do raise notice problems comparable to those presented by unconstitutionally vague criminal statutes. Their failure to specify particular sentencing factors leaves individuals unable to alter their behavior to minimize legal consequences, and it gives the government unbridled discretion. Cf. Hessick & Hessick, Constitutional Rights, supra note 11, at 72.

(149) The refusal to require notice in discretionary schemes is also at odds with the notice doctrine in capital sentencing as developed in Lankford, v. Idaho, 500 U.S. 110 (1991). Lankford involved an Idaho law that allowed a judge to impose the death penalty even when the prosecutor does not request it. Based on that law, the judge imposed death on Lankford after conducting a sentencing hearing, even though Idaho state prosecutors did not seek a death sentence against Lankford and none of the arguments at the hearing focused on whether Lankford should receive capital punishment. Id. at 115-17. Although the judge had discretion whether to impose death, the Lankford Court held that a defendant's right to due process was violated when, at the time of the sentencing hearing, the defendant does not have notice that the judge might sentence the defendant to death. Id. at 122-24.

(150) U.S. Const, amend. V, XIV [section] 1.

(151) See Mathews v. Eldridge, 424 U.S. 319, 334 (1976).

(152) See supra Section II.B.

(153) See Irizarry v. United States, 553 U.S. 708, 721 (2008) (Breyer, J., dissenting) ("[I]f Booker expanded the number of grounds on which a district court may impose a non-Guidelines sentence, that would seem to be an additional argument in favor of, not against, giving the parties notice of the district court's intention to impose a non-Guidelines sentence for some previously unidentified reason. Notice, after all, would promote 'focused, adversarial' litigation at sentencing." (quoting Burns v. United States, 501 U.S. 129, 134 (1991))).

(154) See Lankford, 500 U.S. at 120-24.

(155) See Irizarry, 553 U.S. at 721-22 (Breyer, J., dissenting).

(156) U.S. Const, art. I, [section] 9, cl. 3; art. I., [section] 10, cl. 1.

(157) See Wayne A. Logan, The Ex Post Facto Clause and the Jurisprudence of Punishment, 35 Am. Crim. L. Rev. 1261, 1267 (1998).

(158) Calder v. Bull, 3 U.S. (3 Dali.) 386, 390 (1798). Calder identified four categories covered by the Ex Post Facto Clause:

   1st. Every law that makes an action done before the passing of the
   law, and which was innocent when done, criminal; and punishes such
   action. 2d. Every law that aggravates a crime, or makes it greater
   than it was, when committed. 3d. Every law that changes the
   punishment, and inflicts a greater punishment, than the law annexed
   to the crime, when committed. 4th. Every law that alters the legal
   rules of evidence, and receives less, or different, testimony, than
   the law required at the time of the commission of the offence, in
   order to convict the offender.

Id. at 390 (emphasis omitted). Subsequent cases have further refined these Calder categories, with the Supreme Court explicitly acknowledging that the meaning of the Clause is the product of the accretion of its own common law. Dobbert v. Florida, 432 U.S. 282, 292 (1977) ("Our cases have not attempted to precisely delimit the scope of this Latin phrase, but have instead given it substance by an accretion of case law.").

(159) Garner v. Jones, 529 U.S. 244, 251-52 (2000). That question is necessarily "a matter of 'degree,'" and the Court has acknowledged that it has never devised a "single 'formula' for identifying those legislative changes that have a sufficient effect on substantive crimes or punishments to fall within the constitutional prohibition." California Dep't of Corr. v. Morales, 514 U.S. 499, 509 (1995) (quoting Beazell v. Ohio, 269 U.S. 167, 171 (1925)).

(160) See, e.g., Jenny S. Martinez, International Courts and the U.S. Constitution: Reexamining the History, 159 U. Pa. L. Rev. 1069, 1130-31 (2011) (describing the "'prohibition of expost facto crimes'" as among the "procedural rights [given] to criminal defendants in federal court" (quoting Am. Soc'y of Int'l Law, U.S. Policy Toward the International Criminal Court: Furthering Positive Engagement 41-44 (2009))); Corey Rayburn Yung, The Disappearing Ex Post Facto Clause: From Substantive Bulwark to Procedural Nuisance, 61 Syracuse L. Rev. 447, 456-57 (2011) (describing the ex post facto prohibition as "procedural").

(161) See Colder, 3 U.S. (3 Dali.) at 388 ("The Legislature ... cannot change innocence into guilt; or punish innocence as a crime ...." (emphasis omitted)).

(162) See generally O.W. Holmes, The Path of the Law, 10 Harv. L. Rev. 457, 459 (1897) (noting that bad men obey law only to avoid consequences). For example, an individual who might embezzle when she faces two years of imprisonment might conclude that the benefits are not worth the possible consequences if the punishment were twenty years instead.

(163) Although providing notice and preventing vindictive legislation are the two main reasons for the prohibition on ex post facto laws, the Court has occasionally noted a third purpose: enforcing the separation of powers "by confining the legislature to penal decisions with prospective effect and the judiciary and executive to applications of existing penal law." Weaver v. Graham, 450 U.S. 24, 29 n.10 (1981). But there are reasons to doubt the strength of this "separation of powers" argument. The Ex Post Facto Clause does not prevent Congress from enacting retroactive civil laws. See Landgraf v. USI Film Prods., 511 U.S. 244, 263 & n.16 (1994). Moreover, the ex post facto prohibition also applies to the states, and states have not adopted this prospective/retrospective line distinguishing legislatures and courts. See, e.g., People v. Mitchell, 606 N.E.2d 1381, 1384 (N.Y. 1992) (allowing courts to create prospective law).

(164) See Fletcher v. Peck, 10 U.S. (6 Cranch) 87, 137-38 (1810) ("[I]t is not to be disguised that the framers of the constitution viewed, with some apprehension, the violent acts which might grow out of the feelings of the moment...."); see also City of Richmond v. J.A. Croson Co., 488 U.S. 469, 513 (1989) (Stevens, J., concurring) (noting that the Ex Post Facto Clauses "reflect a valid concern about the use of the political process to punish or characterize past conduct of private citizens"); Logan, supra note 159, at 1267 (retroactive laws could be used to target "maligned individuals and groups 'of the moment' " (quoting Fletcher, 10 U.S. (6 Cranch) at 138)); R. Brian Tanner, Note, A Legislative Miracle: Revival Prosecutions and the Ex Post Facto Clauses, 50 Emory L.J. 397, 429 (2001) (similar).

(165) Indeed, Congress and the U.S. Sentencing Commission have increased penalties for white collar offenders, and those penalties are often retroactive under the Commission's "one book rule." See William P. Ferranti, Revised Sentencing Guidelines and the Ex Post Facto Clause, 70 U. Chi. L. Rev. 1011, 1026-30 (2003).

(166) Peugh v. United States, 133 S. Ct. 2072, 2085 (2013) (plurality opinion).

(167) 482 U.S. 423, 429-35 (1987).

(168) Garner v. Jones, 529 U.S. 244, 251-52 (2000).

(169) Peugh, 133 S. Ct. at 2088 (2013) (majority opinion). Before Booker, every federal circuit held that retroactive application of amendments to the Federal Sentencing Guidelines that increased sentences above the sentence in place at the time the crime was committed violated the Ex Post Facto Clause. See United States v. Seacott, 15 F.3d 1380, 1384 (7th Cir. 1994); see also James R. Dillon, Doubting Demaree: The Application of Ex Post Facto Principles to the United States Sentencing Guidelines After United States v. Booker, 110W. Va. L. Rev. 1033, 1047-49 (2008); Daniel M. Levy, Note, Defending Demaree: The Ex Post Facto Clause's Lack of Control over the Federal Sentencing Guidelines After Booker, 77 Fordham L. Rev. 2623, 2634 (2009).

(170) Peugh, 133 S. Ct. at 2083.

(171) Id. The Court also observed that the system of appellate review "puts in place procedural 1 hurdle [s]' that, in practice, make the imposition of a non-Guidelines sentence less likely." Id. at 2083-84.

(172) Peugh seems inconsistent with the Court's holding in Irizarry. One of the reasons for the Ex Post Facto Clause is to provide notice, and the Court in Irizarry held that, because the federal guidelines are now advisory, due process does not entitle a defendant to notice of the reasons a sentencing judge may increase his or her sentence. See supra Section III.C. Peugh distinguished Irizarry on the ground that the Ex Post Facto Clause is animated by "basic principles of fairness," not simply principles of notice. Peugh, 133 S. Ct. 2084-85 (plurality opinion). That rationale is doubly suspect--first because principles of fairness apply to due process as well, Landon v. Plasencia, 459 U.S. 21, 34-35 (1982) (noting "the essential standard of fairness" imposed by the Due Process Clause), and second because notice is not independent from fairness, but is instead one way of ensuring fairness.

(173) Peugh, 133 S. Ct. at 2088 (majority opinion); see also Garner v. Jones, 529 U.S. 244, 255 (2000) ("When the rule does not by its own terms show a significant risk, the respondent must demonstrate, by evidence drawn from the rule's practical implementation by the agency charged with exercising discretion, that its retroactive application will result in a longer period of incarceration than under the earlier rule."). Because Peugh held that the Ex Post Facto Clause applies to at least one discretionary system, one might conclude that the Court is already engaging in the type of analysis that we are advocating in this Article--namely, assessing the enforceability of different procedural rights according to their own terms rather than simply according to whether a sentencing system is mandatory or discretionary. But that is not so. Even in Peugh, the Court's analysis turned almost exclusively on the amount of discretion afforded to sentencing judges under the post-Booker federal system, rather than focusing on whether the degree of discretion should be relevant to ex post facto claims.

(174) For more on voluntary guidelines, including which American jurisdictions have such systems, see John F. Pfaff, The Continued Vitality of Structured Sentencing Following Blakely: The Effectiveness of Voluntary Guidelines, 54 UCLA L. Rev. 235, 244-46 & n.23 (2006). In contrast to purely voluntary guideline systems, judges in the federal system have less authority to sentence outside of the now-advisory guidelines. Their authority is limited by appellate review and other procedural mechanisms, which tend to funnel sentencing discretion towards within-guideline sentences. See Carissa Byrne Hessick, A Critical View of the Sentencing Commission's Recent Recommendations to "Strengthen the Guidelines System," 51 Hous. L. Rev. 1335, 1337-44 (2014).

(175) Jones, 529 U.S. at 255 ("When the rule does not by its own terms show a significant risk, the respondent must demonstrate, by evidence drawn from the rule's practical implementation by the agency charged with exercising discretion, that its retroactive application will result in a longer period of incarceration than under the earlier rule.").

(176) See Comment, Criminal Law--Sentencing Guidelines--Second Circuit Holds That Imposing Below-Guidelines Sentence Using Retroactive Guidelines Range Increase Does Not Violate Ex Post Facto Clause, 124 Harv. L. Rev. 2091, 2093-94 (2011).

(177) See Pfaff, supra note 176, at 268 (finding such an effect).

(178) See United States v. Demaree, 459 F.3d 791, 794 (7th Cir. 2006) (noting that the substantial risk test "interpreted literally, would encompass a change in even voluntary sentencing guidelines, for official guidelines even if purely advisory are bound to influence judges' sentencing decisions").

(179) See generally F. Andrew Hessick, Probabilistic Standing, 106 Nw. U. L. Rev. 55, 71-72 (2012) (explaining that even when each individual faces a low probability of experiencing a harm, the probability that at least one individual will experience the harm is high when the group facing the threatened harm is large enough).

(180) Calder v. Bull, 3 U.S. (3 Dali.) 386, 390 (1798).

(181) In Rogers v. Tennessee, the Supreme Court explained that the Ex Post Facto Clause does not apply to courts or common law judicial decisionmaking. 532 U.S. 451, 460-61 (2001). That conclusion seems questionable, given that judicial recognition of new sentencing factors presents a threat equal to legislative enactments which create new sentencing factors. Still, the Ex Post Facto Clause may not apply to situations where the legislature has given a judge complete discretion to increase a sentence based on previously unidentified factors by any amount, because in that case the law has specified ex ante that all considerations might bear on sentencing. Of course, other limitations, such as the prohibition on overly vague statutes, may limit the scope of discretion.

(182) See, e.g., United States ex rel. Greathouse v. Dern, 289 U.S. 352, 360-61 (1933).

(183) See Henry M. Hart, Jr. 8c Albert M. Sacks, The Legal Process 3-4 (William N. Eskridge, Jr. 8c Philip P. Frickey eds., 1994); see also Black's Law Dictionary 962 (9th ed. 2009) (defining "law" to include not only "rules" and "standards," but also "principles").

(184) See Miller v. Florida, 482 U.S. 423 (1987); Weaver v. Graham, 450 U.S. 24 (1981); Lindsey v. Washington, 301 U.S. 397 (1937); see also Cal. Dep't. of Corr. v. Morales, 514 U.S. 499, 505 (1995) (characterizing these cases as "holding that a legislature may not stiffen the 'standard of punishment' applicable to crimes that have already been committed.").

(185) 269 U.S. 167, 168-69 (1925).

(186) Id. at 171.

(187) Morales, 514 U.S. at 510.

(188) Cf. Garner v. Jones, 529 U.S. 244, 259 (2000) (Scalia, J., concurring) (explaining that any risk of increased punishment based on changes to parole procedures "is merely part of the uncertainty which was inherent in the discretionary parole system, and to which respondent subjected himself when he committed his crime").

(189) For more on the communicative nature of punishment, see, for example, R. A. Duff, The Intrusion of Mercy, 4 Ohio St. J. Crim. L. 361, 368-69 (2007); Dan M. Kahan, The Anatomy of Disgust in Criminal Law, 96 Mich. L. Rev. 1621 (1998) (book review); Dan Markel & Chad Flanders, Bentham on Stilts: The Bare Relevance of Subjectivity to Retributive Justice, 98 Calif. L. Rev. 907, 929-34 (2010).

(190) Nor does the fact that the guidelines might have little effect on sentences because of judicial discretion mean that they are consistent with the Ex Post Facto Clause. The Ex Post Facto Clause prohibits laws that retroactively increase punishment. A law enacted for that purpose, even if it fails to accomplish that goal, violates the Clause. See Paul Brest, The Conscientious Legislator's Guide to Constitutional Interpretation, 27 Stan. L. Rev. 585, 594 (1975) (explaining that, for a legislator to act consistent with the Constitution, he must not support a law that violates the Constitution).

(191) Michael J. Klarman, How Great Were the "Great" Marshall Court Decisions?, 87 Va. L. Rev. 1111, 1150 (2001) ("Courts are not immune from the pull of public opinion, either because judges generally share the sentiments of the public or because judges who remain oblivious to such opinion are susceptible to political discipline.").

(192) Cass R. Sunstein, If People Would Be Outraged by Their Rulings, Should Judges Care?, 60 Stan. L. Rev. 155, 170-75 (2007).

(193) The anchoring effect is a cognitive bias. It occurs when an initial judgment or number "serves as a perceptual 'anchor' " in an analytic exercise; that is, despite new information, the initial number continues to exert influence over--if not distort--the ultimate judgment reached. Timur Kuran & Cass R. Sunstein, Availability Cascades and Risk Regulation, 51 Stan. L. Rev. 683, 705-706 (1999); see also Bibas & Klein, supra note 26, at 779-80 (noting that the advisory federal guidelines "provide mental anchors, starring points that influence how judges think about cases" and the sentences they ultimately impose). For an excellent overview of the anchoring effect, see Ward Farnsworth, The Legal Analyst 230-36 (2007).

(194) The Supreme Court adopted this argument in Williams v. New York, 337 U.S. 241, 247 (1949), and a number of commentators have supported it, e.g., Huigens, supra note 71, at 1068-79.

(195) See Pepper v. United States, 131 S. Ct. 1229, 1240 (2011) ("Permitting sentencing courts to consider the widest possible breadth of information about a defendant 'ensures that the punishment will suit not merely the offense but the individual defendant.'" (quoting Wasman v. United States, 468 U.S. 559, 564 (1984))); United States v. Grayson, 438 U.S. 41, 53 (1978) (" [I] t is proper--indeed, even necessary for the rational exercise of discretion--to consider the defendant's whole person and personality, as manifested by his conduct at trial and his testimony under oath, for whatever light those may shed on the sentencing decision. The 'parlous' effort to appraise 'character,' degenerates into a game of chance to the extent that a sentencing judge is deprived of relevant information concerning 'every aspect of a defendant's life.'" (quoting United States v. Hendrix, 505 F.2d 1233, 1236 (2d Cir. 1974))); Williams, 337 U.S. at 247 ("Highly relevant--if not essential--to [the judge's] selection of an appropriate sentence is the possession of the fullest information possible concerning the defendant's life and characteristics.").

(196) See, e.g., Fennell & Hall, supra note 131, at 1634 (noting the tension between interests of reliability and completeness in the context of a right to disclosure of presentence reports). Some procedural protections do impair accuracy. The requirement of proof beyond a reasonable doubt, for example, often undermines the accuracy of factual findings by tilting the scales in the defendant's favor. See Michaels, supra note 1, at 1835. But the refusal to observe procedures at sentencing in discretionary schemes has not been limited to those sorts of procedures.

(197) Kg., Williams, 337 U.S. at 249-50.

(198) See State v. Hurt, 702 S.E.2d 82, 94 (N.C. Ct. App. 2010), rev'd on other grounds, 743 S.E.2d 173 (N.C. 2013); State v. Rodriguez, 754 N.W.2d 672, 681 (Minn. 2008); see also Vankirk v. State, 385 S.W.3d 144, 149 (Ark. 2011) (holding that, when sentencing is by jury, the right to confrontation applies).

(199) See Williams, 337 U.S. at 249-50.

(200) Crawford v. Washington, 541 U.S. 36, 61-62 (2004).

(201) See supra Section II.B.

(202) See supra text accompanying notes 150-52.

(203) For examples of this argument, see United States v. Clark, 792 F. Supp. 637, 644-47 (E.D. Ark. 1992); Mitchell, supra note 76, at 346; Young, supra note 32, at 302, 314-15.

(204) See Mitchell, supra note 76, at 346; Young, supra note 32, at 314-15.

(205) Mitchell, supra note 76, at 346 ("A sentencer might take into account a 10 percent likelihood that a defendant carried a gun during the crime....").

(206) See, e.g., Blackston v. Rapelje, No. 12-2668, 2014 WL 4977419, at *8 (6th Cir. Oct. 7, 2014) (limiting impeachment for collateral facts).

(207) To be fair to Mitchell, he does not necessarily endorse the prospect of different procedural rights based on this distinction; he merely says that it "provides a possible means by which to distinguish the 'sentencing factors' that must be found by juries from the fully discretionary sentencing regimes that have always been thought constitutional." Mitchell, supra note 76, at 346. Young, however, does appear to endorse different procedural rights on this basis. Young, supra note 32, at 314-15.

(208) C.M.A. McCauliff, Burdens of Proof: Degrees of Belief, Quanta of Evidence, or Constitutional Guarantees?, 35 Vand. L. Rev. 1293, 1332 (1982).

(209) See, e.g, Bowman, supra note 70, at 373 (noting this view among others); see also Beale, supra note 34, at 157-58 (noting the view that, in discretionary systems "sentencing was not a truly legal decision," and also noting that, "[w]hile the reality [of sentencing] did not live up to this vision," it almost certainly affected the provision of procedural protections).

(210) E.g., Beale, supra note 34, at 157 ("In a discretionary sentencing scheme dominated by at least a rhetoric of rehabilitation, the sentence was not a product of any findings of fact about the nature of the offense, but rather a product of the judge's intuition about the defendant's prospects for rehabilitation.").

(211) See supra Section II.B. To be sure, as a practical matter, sentencing judges traditionally could impose on any offender basically any sentence within the statutorily authorized range. And it has often been stated that differences in sentences imposed were attributable, not to factual differences between offenders, but rather to the identity or mood of a particular judge. E.g., Blakely v. Washington, 542 U.S. 296, 332 (2004) (Breyer, J., dissenting) (stating that a person's prison sentence appeared to depend on what judge was preceding over the case, what the judge ate that day, or other extraneous factors). But that is not because the law authorized judges to impose sentences based on what he or she had eaten for breakfast. Rather, judges had the ability to impose different sentences based on such reasons only because their decisions were subject to extremely limited appellate review. In other words, those judges had de facto discretion to impose sentences based on such trivial reasons. But the absence of appellate review is a reason for more procedures in the trial court, not fewer.

(212) See United States v. Main, 598 F.2d 1086, 1094 (7th Cir. 1979) ("'[O]nce it is determined that a sentence is within the limitations set forth in the statute under which it is imposed, appellate review is at an end,' unless the sentencing judge relied on improper or unreliable information in exercising his or her discretion, or failed to exercise any discretion at all, in imposing sentence." (quoting Dorszynski v. United States, 418 U.S. 424, 431 (1974))); see also Hessick & Hessick, Constitutional Rights, supra note 11, at 54-56 (collecting cases identifying improper sentencing factors).

(213) Hessick & Hessick, Constitutional Rights, supra note 11, at 54-56.

(214) Albemarle Paper Co. v. Moody, 422 U.S. 405, 416 (1975) (quoting United States v. Burr, 25 F. Cas. 30, 35 (C.C.D. Va. 1807) (No. 14,692)).

(215) Friendly, supra note 53, at 765.

(216) To be sure, some procedures may increase the difficulty of making a value judgment. For example, requiring a judge to notify a defendant of the possible considerations at sentencing or to articulate the reasons for his sentence may pose obstacles because of the difficulty in identifying the basis for a value judgment. But that is not a valid reason to dispense with those procedures. Discretionary sentencing does not authorize judges to base sentences on general impressions of the defendant instead of on precise reasons. If discretionary sentencing is "a standardless enterprise, based on no more than a gut feeling," Hessick & Hessick, Constitutional Rights, supra note 11, at 86, then it is incompatible with even the most basic notions of due process. See Foucha v. Louisiana, 504 U.S. 71, 80 (1992) (describing "[f]reedom from bodily restraint" based on "arbitrary governmental action" as "the core of ... the Due Process Clause").

(217) See Hart & Sacks, supra note 185.

(218) See generally Tom R. Tyler, Governing Amid Diversity: The Effect of Fair Decisionmaking Procedures on the Legitimacy of Government, 28 Law & Soc'y Rev. 809 (1994) (concluding that "procedural justice judgments about the fairness of the process used in formulating national policies shape the legitimacy of such governmental authorities").

(219) See Blakely v. Washington, 542 U.S. 296, 309 (2004).

(220) Id.

(221) Id.

(222) Harris v. United States, 536 U.S. 545, 566 (2002) (quoting Apprendi v. New Jersey, 530 U.S. 466, 498 (2000) (Scalia, J., concurring)).

(223) Cf. Charles Fried, Constitutional Doctrine, 107 Harv. L. Rev. 1140, 1143 (1994) (arguing that the expectation created by past decisions is not a sound basis for stare decisis, because no one could reasonably have those expectations without stare decisis).

(224) 133 S. Ct. 2151 (2013).

(225) Alleyne involved a defendant who had been convicted of "using or carrying a Firearm in relation to a crime of violence." Id. at 2155. The relevant federal statute provides for a five-year mandatory minimum for anyone who "uses or carries a firearm," a seven-year mandatory minimum if "the firearm is brandished," and a ten-year mandatory minimum if "the firearm is discharged." 18 U.S.C. [section] 924(c)(1)(A) (2012). The jury in Alleyne's case indicated on a verdict form that Alleyne had used or carried a firearm, "but did not indicate a finding that the firearm was '[b]randished.'" Alleyne, 133 S. Ct. at 2156. The trial judge, however, made a finding that Alleyne had brandished a firearm, and sentenced Alleyne to the applicable seven-year mandatory minimum sentence. The Supreme Court reversed, holding that, because whether an individual brandishes a firearm "alter [s] the prescribed range of sentences to which a defendant is exposed and do[es] so in a manner that aggravates the punishment," then whether a defendant brandishes a firearm is an issue that "must be submitted to the jury and found beyond a reasonable doubt." Id. at 2158-61.

(226) Alleyne, 133 S. Ct. at 2158-63.

(227) See supra text accompanying notes 196-97.

(228) Courts have held that imposing the maximum sentence for reasons unrelated to a particular defendant or his crime is an abuse of discretion even under broadly discretionary schemes. See United States v. Hartford, 489 F.2d 652 (5th Cir. 1974); United States v. Thompson, 483 F.2d 527 (3d Cir. 1973); Woosley v. United States, 478 F.2d 139 (8th Cir. 1973); United States v Daniels, 446 F.2d 967 (6th Cir. 1971); United States v. McCoy, 429 F.2d 739 (D.C. Cir. 1970).

(229) One might think that a judge in a discretionary system may impose the maximum sentence without a factual finding. After all, while findings of fact are the only decisions left to judges in mandatory systems, judges in discretionary systems are also free to make policy and application decisions. See supra Section II.B. But because judges are required to individualize sentences, the decision must be based on some fact unique to the offender or her crime--that is the basic premise of individualized sentencing. Bowman, supra note 70, at 371 ("[E]ven in a purely discretionary sentencing system, the only way for a judge rationally to distinguish one defendant from others who have committed the same statutory crime is to ascertain facts other than the fact of conviction that suggest a sentence at, above, or below the norm for that crime."). That different judges might make different policy or application decisions when faced with the same facts does not suggest that the decision to impose the maximum sentence is not a decision based on a finding of fact. It simply means that different facts might trigger that sentence for different judges.

(230) Alleyne, 133 S. Ct. at 2155; Apprendi v. New Jersey, 530 U.S. 466, 494 n.19 (2000).

(231) See Apprendi, 530 U.S. at 478-79.

(232) Indeed, we ourselves have previously stated such a rigid view of early American sentencing. Hessick & Hessick, Constitutional Rights, supra note 11, at 51. In more recent articles our description of sentencing history has been more nuanced. Carissa Byrne Hessick & F. Andrew Hessick, Double Jeopardy As a Limit on Punishment, 97 Cornell L. Rev. 45, 70 n.120 (2011).

(233) Alleyne, 133 S. Ct. at 2158 (noting that "some early American statutes provided ranges of permissible sentences").

(234) Id. at 2158-59 (stating that sentencing ranges were "linked to particular facts constituting the elements of the crime," and that historically, a crime was understood to treat "every fact which 'is in law essential to the punishment sought to be inflicted'" as an element (quoting 1 Joel Prentiss Bishop, Commentaries on the Law of Criminal Procedure 50 (2d ed. 1872)).

(235) Compare Pepper v. United States, 131 S. Ct. 1229, 1240 (2011) ("[B]oth before and since the American colonies became a nation, courts in this country and in England practiced a policy under which a sentencing judge could exercise a wide discretion in the sources and types of evidence used to assist him in determining the kind and extent of punishment to be imposed within limits fixed by law." (quoting Williams v. New York, 337 U.S. 241, 246 (1949)) (internal quotation marks omitted)), with Apprendi, 530 U.S. at 478 ("Any possible distinction between an 'element' of a felony offense and a 'sentencing factor' was unknown to the practice of criminal indictment, trial by jury, and judgment by court as it existed during the years surrounding our Nation's founding." (footnote omitted)); Apprendi, 530 U.S. at 478 (noting a "defendant's ability to predict with certainty the [sentence] from the face of the felony indictment" based on "the invariable linkage of punishment with crime").

(236) Pepper, 131 S. Ct. at 1240.

(237) E.g., Williams v. New York, 337 U.S. 241, 246 (1949).

(238) Indeed, the Court relies largely on mid-nineteenth century sources when making its historical arguments. See, e.g., Alleyne, 133 S. Ct. at 2158-60; Apprendi, 530 U.S. at 501-18 (Thomas, J., concurring).

(239) That felonies had determinate penalties does not mean that defendants always received that penalty. Judges could reduce the penalty by benefit of clergy, and juries would often convict the defendant of a lesser charge solely to avoid imposing the harsh penalty--usually death though sometimes transportation--for committing a felony. See Stephanos Bibas, Judicial Fact-Finding and Sentence Enhancements in a World of Guilty Pleas, 110 Yale L.J. 1097, 1124 n.204 (2001).

(240) See, e.g., 9 Geo. I., c. 8, [section] 4 (1722) (conferring discretion on judges "to punish [an] offender or offenders corporally, by causing him, her[,] or them, to be publickly whipped, or committed to some publick workhouse, there to be kept to hard labour, for the space of six months, or a less time"); King v. Bland, (1793) 168 Eng. Rep. 400 (discussing the discretion conferred by statute to choose punishments).

(241) 1 James Fitzjames Stephen, A History of the Criminal Law of England 481-82 (1883). In theory, that discretion was unlimited: law did not impose an upper limit on the possible term of imprisonment. But in practice, judges did not impose terms of imprisonment longer than five years. See Bibas, supra note 241, at 1125-26 n.206.

(242) See, e.g., Nancy J. King & Susan R. Klein, Essential Elements, 54 Vand. L. Rev. 1467, 1474-76 (2001); see generally Mitchell, supra note 76, at 330-42 (surveying sentencing in nineteenth-century cases).

(243) Id. at 339-42. Nancy King and Susan Klein's pout-Apprendi analysis of the relevant historical cases concludes that the historical account from Justice Stevens's majority opinion "most accurately reflects past practice." King & Klein, supra note 244, at 1472. In contrast, they note that the "historical basis for the broader rule advocated by Justice Thomas" in his Apprendi concurrence--that is, the rule ultimately adopted by the Court in Alleyne--"is much more ambiguous than the clear support for the limited rule advanced in the majority opinion of Justice Stevens." Id. at 1474. In particular they note that Justice Thomas's rule, though not inconsistent with the language from various historical sources, is not supported by the holdings of any relevant cases; while aggravating facts sometimes did raise minimum sentences, those same facts also raised the maximum sentence. In other words,

   courts of this earlier era were not presented with the necessity of
   deciding whether a fact, other than prior conviction, that triggers
   a mandatory minimum sentence but not a higher maximum sentence, was
   an essential ingredient of an offense that must be pled in the
   indictment and proven to a jury beyond a reasonable doubt.


(244) See Michaels, supra note 1, at 1775 (noting that courts do not offer the "historical practice" justification when it does not support the courts' desired result). The departure from history is not limited to procedural rights at sentencing. Courts have similarly expanded the substantive rights of individuals at sentencing. Historically, courts could consider matters such as race and gender in imposing sentence, but today those considerations are forbidden. Hessick & Hessick, Constitutional Rights, supra note 11, at 76; Carissa Byrne Hessick, Race and Gender as Explicit Sentencing Factors, 14 J. Gender Race & Just. 127 (2010).

(245) See Hessick & Hessick, Constitutional Rights, supra note 11, at 74-77.

(246) See 4 William Blackstone, Commentaries *379 (2d ed. 1799); see also Apprendi v. New Jersey, 530 U.S. 466, 478 (2000) ("Any possible distinction between an 'element' of a felony offense and a 'sentencing factor' was unknown to the practice of criminal indictment, trial by jury, and judgment by court as it existed during the years surrounding our Nation's founding." (footnote omitted))

(247) See Mempa v. Rhay, 389 U.S. 128 (1967).

(248) See Glover v. United States, 531 U.S. 198 (2001).

(249) See, e.g., Sumner v. Shuman, 483 U.S. 66 (1987); Roberts v. Louisiana, 431 U.S. 633 (1977) (per curiam); Roberts v. Louisiana, 428 U.S. 325 (1976); Woodson v. North Carolina, 428 U.S. 280 (1976).

(250) At early common law, death was--at least theoretically--the penalty for all felonies. John H. Langbein, Shaping the Eighteenth-Century Criminal Trial: A View from the Ryder Sources, 50 U. Chi. L. Rev. 1, 36 (1983). By the middle of the eighteenth century, however, some felonies carried lesser penalties such as transportation. Stephen, supra note 243, at 481-82.

(251) Jones v. United States, 526 U.S. 227, 244-45 (1999) (noting uncertainty over whether the Framers' would have understood the Sixth Amendment as tolerating "exclusively judicial factfinding to peg penalty limits" because "the scholarship of which [the Court is] aware does not show that a question exactly like this one was ever raised and resolved in the period before the framing"). This trend of recognizing more rights at sentencing also refutes the argument that the longevity of the practice of not recognizing procedural rights in discretionary schemes establishes the constitutionality of that practice. See Hessick & Hessick, Constitutional Rights, supra note 11, at 75-77. Although a "'universal and long-established' tradition" of allowing certain conduct may be reason to presume that the conduct is constitutional, Republican Party of Minn. v. White, 536 U.S. 765, 785 (2002) (quoting McIntyre v. Ohio Elections Comm'n, 514 U.S. 334, 377 (1995) (Scalia, J., dissenting)), the repeated breaks from tradition in this area suggest that previous decisions not to provide procedural rights at sentencing are not entitled to a presumption of constitutionality.

(252) For some rights the case is stronger than for others. For example, there is a strong case for applying the right of notice, because notice would likely enhance the overall quality of the sentencing decision, increase the perceived fairness of sentencing, and impose only a small burden on the government. See supra text accompanying notes 147-50. By contrast, there is arguably less need for the proof beyond a reasonable doubt requirement at sentencing, because one justification for the right--to avoid stigmatizing a defendant through a conviction--does not apply given that the defendant has already been convicted. See Bibas, supra note 241, at 1179.

(253) See supra notes 35-36 and accompanying text.

(254) See Bibas & Klein, supra note 26, at 785-86; John F. Pfaff, The Future of Appellate Sentencing Review: Booker in the States, 93 Marq. L. Rev. 683, 700 (2009).

Carissa Byrne Hessick, Professor of Law, S.J. Quinney College of Law, University of Utah. J.D., Yale Law School. B.A., Columbia University.

F. Andrew Hessick, Professor of Law, S.J. Quinney College of Law, University of Utah. J.D., Yale Law School. B.A., Dartmouth College. We want to thank Albert Alschuler, Shima Baradaran, William Berry, Jennifer Chacon, Jack Chin, Chad Flanders, Gerry Leonard, Justin Marceau, Dan Markel, John Pfaff, Amelia Rinehart, Meghan Ryan, Jeff Schwartz, John Stinneford, and Michael Teter for their helpful comments on this project. Thanks also to the participants at the 2013 Law and Society Annual Meeting, the Southwest Criminal Law Workshop at UC Irvine, the SMU Criminal Justice Colloquium, and the Utah Junior Law Faculty Workshop, as well as to participants at faculty workshops at the University of North Carolina School of Law and the Washington University School of Law. Beth DeFelice provided valuable research assistance.
COPYRIGHT 2014 University of Notre Dame Law School
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2014 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Hessick, Carissa Byrne; Hessick, F. Andrew
Publication:Notre Dame Law Review
Date:Nov 1, 2014
Previous Article:Mirrored externalities.
Next Article:A house divided: when state and lower federal courts disagree on federal constitutional rights.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters