The Quality of Mercy Strained: Wresting the Pardoning Power from the King
[69 Tex.L.Rev. 569 (1991); reprinted by permission in JURIST's Presidential Pardons guide]

Daniel T. Kobil*

Since ancient times the power of the executive to suspend the operation of the justice system by extending clemency to an accused or convicted criminal has been a fundamental part of criminal justice systems. It continues to figure prominently in the American system despite its discordance with the checks and procedures that characteristically attend other executive powers. In this Article, Professor Kobil examines the modern use of the clemency power in the United States in light of its historical origins and philosophical underpinnings. He concludes that exercises of the clemency power can usefully be divided into those that further principles of retributive fairness and those that are unrelated to principles of justice. Professor Kobil argues that the consistent, principled administration of clemency can best be realized by making this bifurcation explicit and removing the former category of clemency decisions from the unguided authority of the chief executive. He proposes a set of procedures and standards to divest clemency decisions from the individual discretion of the executive and insulate them from unwarranted political pressures.

I. Introduction
II. The Nature and Origins of Clemency: What's Justice Got to Do With It? A. Defining Clemency, and its Relationship to Notions of Justice B. Early Origins of the Clemency Power C. The Development of the Clemency Power in England D. The Clemency Power in the United States 1. Federal Clemency Power 2. The Clemency Power in the States III. A Proposal for Bifurcation of the Clemency Power: Replacing Politics with Justice A. Limitations on the Exercise of Executive Clemency Power 1. The Legislature 2. The Judiciary 3. The People 4. The Executive B. Proposed Clemency Commission C. Proposed Clemency Standards 1. Substantial Doubt of Guilt 2. Diminished Mental Capacity, Retardation, Intoxication, or Minority 3. Disparate Sentencing: Proportionality 4. Disparate Sentencing: Special Circumstances 5. Sentencing That is Unrelated to Deserts 6. Crimes Committed out of Necessity, Coercion, or Adherence to Moral Principles 7. Crimes in Which the Offender Has Suffered Enough D. Proposed Clemency Procedures E. Justice-Neutral Exercise of the Clemency Power by the Executive: What Remains? IV. Conclusion

"Use every man after his desert, and who should 'scape whipping? Use them after your own honour and dignity: the less they deserve, the more merit is in your bounty."

--Hamlet, Act II, scene ii.1

"[T]he longer I live, the larger loom those . . . decisions about justice and mercy that I had to make as governor. They didn't make me feel godlike then: far from it; I felt just the opposite. It was an awesome, ultimate power over the lives of others that no person or government should have, or crave. And looking back over their names and files now, despite the horrible crimes and the catalog of human weaknesses they comprise, I realize that each decision took something out of me that nothing--not family or work or hope for the future--has ever been able to replace."

--Edmund G. (Pat) Brown

Former Governor of California2

I. Introduction

In Act II, scene ii of Hamlet, Shakespeare's troubled hero urges Polonius to treat well the actors who have come to play for the king, for if every man is used "after his desert, . . . who should 'scape whipping?" Hamlet urges the Lord Chamberlain instead to use people after "your own honour and dignity: the less they deserve, the more merit is in your bounty." Hamlet displays a shrewd awareness of the advantage that can be taken of an act of clemency. His remark reflects a truth about human nature that has shaped the exercise of the clemency power by the monarchs of England, and by the various United States Presidents and governors who succeeded them in the New World. The clemency power traditionally has been used to entrench regimes by "endear[ing] the sovereign to his subjects,"3 rewarding political supporters,4 and even lining the executive's coffers.5

Yet there is also another aspect of the clemency power: clemency can be used to achieve justice, by individualizing sentences and remitting undeserved punishment. This use of the clemency power runs contrary to Hamlet's advice. It rests on a vision of human nature that is fundamentally less pessimistic, since it presumes that some people deserve to "'scape whipping." Clemency, exercised in this way, can properly be said to be a fundamental part of any system of justice.

Because of the political and societal benefits that the executive's power to grant clemency can impart, it is understandable that clemency has long occupied a significant place in Anglo-American jurisprudence. In the view of some, clemency has played as important a role in history as its better known counterpart, punishment.6 From ancient Athens to post-Vietnam America, timely extensions of mercy have often been used in moments of turmoil to bind together a social fabric in danger of rending. Controversy continues to flourish over the proper role of clemency in resolving our nation's most recent political crisis, as pundits debate whether the Iran-Contra conspirators deserve presidential pardons.7

The clemency power has proved philosophically significant as well. As the attention devoted to it by the most accomplished political theorists and philosophers attests,8 the notion of clemency lies at the heart of Judeo-Christian theology,9 and, ultimately, the philosophy of punishment and mercy underpinning modern criminal laws. Pilate's failure in the exercise of mercy10 stands in stark contrast to the divine extension of clemency brought about by Christ's sacrificial death.11 This biblical paradigm illustrates the centrality of mercy to the human condition, and the benefits it imparts both to the forgiver and to the object of the forgiveness. Shakespeare echoed these ideas in The Merchant of Venice: Portia's eloquent plea for mercy suggests that mercy "blesseth him that gives, and him that takes" because its source is not in "temporal power," but in the "hearts of kings," deposited there like the gentle rain of heaven by "God Himself."12

The attention clemency power has attracted is perhaps attributable to its inherently paradoxical nature: clemency is often thought of as an expression of mercy that enhances justice in a broader sense by suspending the operation of our justice system.13 Thus, consideration of clemency necessarily calls into question fundamental beliefs about the nature and purpose of punishment itself, and ultimately, about what constitutes justice.14 Given the relationship of clemency to conceptions of justice, it is surprising that clemency has not historically been exercised in any principled fashion. The legacy of arbitrary exercise of the clemency power, an atavistic remnant of the royal prerogative, is today firmly established in our jurisprudence. Some would assert that this can be explained by a typically American callousness toward matters of injustice;15 however, the exercise of clemency in many societies has been characterized by such arbitrariness.16

The lack of any standards or checks on the exercise of the clemency power has not stood the American system of justice in good stead. Commentators have noted that unbridled discretion in pardoning threatens to permit the President to shield himself and his subordinates from criminal prosecution17 and to undermine the essential functions of coordinate branches of government.18 Indeed, President Richard Nixon's advisors had such confidence in the scope of the presidential pardoning power that they seriously explored the possibility of the President pardoning himself.19 Governors (and, many would contend, Presidents) have regularly exercised the clemency power in ways that are clearly at odds with society's interests, including granting or denying pardons to convicted murderers solely because of campaign promises made to supporters.20 One governor was even impeached and removed for particularly blatant abuses of the pardoning power.21 Abuses of this nature are perhaps inevitable given the lack of standards governing exercise of the clemency power.

This Article examines the origins and theoretical underpinnings of the clemency power and considers the proper role of clemency in the American justice system. Although our notions of punishment have grown gradually more sophisticated since the days of lex talionis,22 the apparatus for dispensing clemency lurches along today in much the same way that it has for centuries. While a politically accountable representative or body is now substituted for the king as the dispenser of clemency, the pardoning power is still exercised in an ad hoc fashion, with little regard to principled decision making or, for that matter, consistency. Elected officials called upon to decide whether to remit punishment in a particular case can look for guidance to little more than their own personal sense of what is just or politically expedient.23 Consequently, clemency often operates as an arbitrary exception to our system of justice, rather than as a corrective.

Today's changing political climate underscores the need for principled exercise of the clemency power. Harsher sentencing standards and growing public sentiment in favor of capital punishment have resulted in an increasing number of death penalty cases finding their way into the clemency process.24 Additionally, prison overcrowding and the growing numbers of prisoners with AIDS make it imperative that the institution of clemency be reexamined to ensure that it augments, rather than undermines, our justice system.

It is the purpose of this Article to suggest that, like the imposition of punishment, the remission of punishment must be administered in a principled, consistent fashion. To identify those principles that should direct the exercise of the clemency power, I will first consider the philosophy of punishment that underlies our system of justice. It is then possible, I believe, to identify the circumstances under which the exercise of clemency is most appropriate, and to distinguish those uses of the clemency power that erode justice. In Part III of this Article, I suggest such a bifurcation and attempt to define its contours. Finally, in order to encourage principled clemency decisions, I propose in Part IV specific procedures and standards to govern the exercise of the clemency power.

II. The Nature and Origins of Clemency: What's Justice Got to Do With It?

A. Defining Clemency, and its Relationship to Notions of Justice

The clemency power is something of a living fossil, a relic from the days when an all-powerful monarch possessed the power to punish and to remit punishment as an act of mercy. It is the oldest form of release procedure, and it survives in some form in every state in the union25 and in every country of the world except China.26

Although the term "clemency" is sometimes treated as a synonym for all manifestations of mercy, I use it in this Article in its narrower sense to denote leniency or mercy in the exercise of authority or power.27 Thus, "clemency" will be used to refer to the forms of leniency extended by various branches of government, most often the executive, to remit the punishment of those who have transgressed society's laws.

Within the broad ambit of "clemency" are five specific varieties of leniency commonly recognized under American law: pardon, amnesty, commutation, remission of fines, and reprieve.28 Although the term "pardon" is sometimes used interchangeably with "clemency" to refer to the general power to remit punishment,29 in this Article I will discuss pardons as a discrete component of the clemency power. A pardon provides the most sweeping remission of the consequences that normally attend violation of the law. At common law, a pardon was an act of mercy whereby the king "forgiveth any crime, offence, punishment, execution, right, title, debt, or duty."30 According to an early pronouncement of the United States Supreme Court, a full or unconditional pardon not only releases the offender from any punishment for her31 crime, but also vitiates moral guilt for the offense, so that in the eyes of the law she is as innocent as if she had never been charged or convicted.32 Pardons may be granted either before or after conviction,33 but are most commonly used to restore the reputation and civil rights of an individual who has completed her designated punishment and demonstrated rehabilitation by leading an exemplary life upon release.34

Amnesty, derived from the Greek amnestia meaning forgetfulness,35 connotes that the offender's crime has been overlooked because that course of action benefits the public welfare more than punishment would.36 Amnesty is typically granted to groups of people, usually before conviction.37 Unlike a pardon, a grant of amnesty does not eradicate the infraction for which punishment is remitted.38 This distinction, however, has proved to be of more political than legal significance. For example, in justifying his grant of amnesty to those who evaded service in Vietnam, President Carter argued that their crimes were not forgiven, as they might have been through a pardon, only forgotten.39 Yet the practical effect of amnesty is virtually identical to that of a pardon, and American law recognizes little distinction between the two forms of clemency.40

A third, more limited form of clemency is commutation, which is the substitution of a milder punishment for the one imposed by the court.41 Unlike a pardon, commutation in no way relieves the offender of most of the legal consequences of an offense,42 nor, presumably, of moral guilt. Commutations are often granted to shorten the offender's sentence to time already served or to make her immediately eligible for parole.43

The clemency power also embraces remission of fines and forfeitures. The presidential power to pardon has been held to include the power to remit fines and forfeitures imposed by the United States.44 In a recent survey of state clemency procedures, eighteen of the thirty-six states responding reported that they likewise offer remission of fines as a form of clemency.45

Finally, the most limited form of clemency is reprieve, which is nothing more than the temporary postponement of punishment.46 A reprieve stays execution of the sentence for a specified time period. It typically is used to give the offender an opportunity to complete pending appeals, though as Blackstone recounts, it also has been used to allow a pregnant woman sentenced to death an opportunity to give birth to her child.47

Parole is not discussed in this Article because it is not properly a form of clemency.48 Rather, it is an "established variation on imprisonment of convicted criminals" characterized by the conditional extension of certain freedoms.49 Parole does not remit punishment, but instead conditionally softens the terms under which the punishment is carried out.50

Each type of clemency described above has its genesis in the sovereign's power to grant mercy to those who violate the law.51 Historically, the grounds for dispensing clemency have been limited only by the ingenuity of the human imagination.52 Clemency has long been considered an extraordinary remedy that can be extended for virtually any reason, whenever mercy, expediency, or personal whim dictated.53 However, examination of the evolution of the clemency power reveals two separate types of justifications that have typically been advanced in support of acts of clemency.

The various rationales used to justify the exercise of the clemency power may be categorized into those that are "justice-enhancing" and those that are unrelated to principles of justice, or "justice-neutral." Although a comprehensive discussion of what constitutes "justice" is beyond the scope of this Article,54 the use of "justice" as the dividing line between the two types of clemency justifications compels at least a short exposition of the theoretical underpinnings of criminal justice.

In analyzing the clemency power, I use the term "justice" in a retributive sense to denote fairness under the law so that each person is rendered her due.55 I am influenced in this choice of definitions by philosophers such as Kathleen Moore, a proponent of a retributivist theory of justice, who defines justice to mean that people "ought to get the punishment they deserve."56 Relying on principles of retributivist theory set out in Immanuel Kant's The Metaphysical Elements of Justice, Moore argues that both our system of punishment, and any system involving the remission of punishment, should be predicated on the notion of deserts.57 Each person should receive what is due her--either punishment or clemency--based on whether she has gained an unfair advantage over law-abiding citizens, or is morally blameworthy.58

Moore's approach to the granting of clemency presents two possible problems. The first is the difficulty in ascertaining deserts: how does one decide the punishment deserved by an offender for a particular crime, and concomitantly, if that punishment is properly remitted? Some philosophers have concluded that it is impossible to determine the amount of punishment deserved by an individual offender.59 We have no external, objective gauge that enables us to define with exactness the punishment deserved by a tax evader, as distinguished from that deserved by a burglar.60 The principle of equivalence that comes most readily to mind--an eye for an eye, a life for a life--disintegrates quickly when applied to modern notions of crime and punishment.61

Notwithstanding these philosophical difficulties, the criminal justice system is based on the assumption that persons who commit particular offenses are deserving of punishment, and that it is possible to establish the particular punishment deserved by each offender based on the nature of her crime.62 Legislatures act on this presupposition when they identify specific punishments for particular crimes. The system relies on the "guided discretion" of jurors and judges to do justice by rendering what is deserved in individual cases.63 Given that our system of justice is predicated on the assumptions that deserts for particular offenses are ascertainable and that we should punish only those who deserve to be punished, it is not only sensible, but also necessary, to base the remission of punishment on the same assumptions.

Retributive concerns alone, however, do not sufficiently describe the goals of punishment. The second problem with the strict approach to clemency that Moore and other Kantian philosophers advocate is its presumption that retributive principles are the only justification for punishment and must be the sole guideposts in clemency decisions.64 The better view, and one of the premises of this Article, is that deserts provide only a starting point, with utilitarian and other societal concerns establishing secondary limits on the remission of punishment generally, and in individual cases.65 Once a determination as to deserts has been made, other considerations such as general or specific deterrence of crime may limit both the imposition and remission of punishment.66 Even if desert-oriented evaluation of a particular crime required remission of punishment--for example, because brain damage rendered the offender undeserving of punishment--utilitarian concerns such as the need to deter this particular offender from engaging in the same behavior would limit the grant of clemency to something less than outright release. In this respect, although I agree that Moore correctly identifies the principle of deserts as the primary consideration in analyzing the clemency power,67 I disagree with her ultimate conclusion that the only justifiable acts of clemency are those which further justice by meting out what the particular individual deserves.

The definition of justice adopted in this Article emphasizes the centrality of retributive principles of fairness and proportionality to our system of punishment. While utilitarian (justice is what results in the most good to the greatest number of people), Christian (justice as love), or other theories of justice have proved useful in addressing some areas of social concern,68 the characterization of justice as fairness lies at the heart of American jurisprudence and is best suited to my task of evaluating the place of clemency in the legal system. The noble ambiguities of the Constitution's due process and equal protection clauses have regularly been interpreted as guaranteeing fairness and justice, with the two terms being used almost interchangeably.69 Moreover, lower courts have often equated justice with fairness and have relied on Justinian's famous formulation of justice as "the constant and perpetual desire to give each man his due right"70 in defining justice.71 Influential proponents of penological reform have advocated a retributivist theory of punishment which presupposes that doing justice means treating persons fairly by rendering the punishment that they deserve.72 These suggestions have recently been heeded, at least in the federal system's new determinative sentencing guidelines73 and abolition of parole.74

Once justice is defined as fairness under the law that renders each person her due, we can see that clemency has been exercised historically both to enhance justice and to further goals unrelated to justice. From classical times to the present, no differentiation has been made between these two very distinct uses of the clemency power. Bereft of standards for dispensing clemency in principled fashion, political leaders have frequently and flagrantly abused the clemency power, using it occasionally to render punishment more fair, but more commonly to further utilitarian goals such as the preservation of their regime or consolidation of their power.

Examining the evolution of clemency reveals why this conflation of justice-enhancing and justice-neutral aspects has occurred. Traditionally, all aspects of clemency were concentrated in the monarch, because she happened to possess power.75 A review of the history of clemency, particularly its justice-neutral manifestations, provides a perspective from which to discern ways of refining the oldest institution for remitting punishment.

B. Early Origins of the Clemency Power

Historically, the institution of clemency seems to have had more to do with power than justice. Justice Holmes accurately characterized the earliest pardons as "private act[s] of grace from an individual happening to possess power."76 "Mercy" was extended by those in power, typically because it was expedient to do so. Since the law was itself regarded as an adequate embodiment of justice, clemency, which circumvented strict operation of the law, was viewed as an exception to principles of justice made necessary by practical considerations.

In the democratic society of ancient Athens, the institution of clemency was not highly developed, largely because power rested with the people rather than with a monarch.77 By the time the Athenian Civil War ended in 403 B.C., the procedural difficulties that attended obtaining clemency in Athens were immense. Before anyone could receive clemency, she had to comply with the process of adeia, which required that at least 6000 citizens support a petition for clemency in a secret poll.78 Because the approval of this many people was difficult to obtain, clemency was seldom granted to individuals, at least those who were not celebrities.79 Thus, grants of clemency often hinged on popularity rather than concerns that a just result be reached.

Amnesties to groups or classes of persons, while more readily obtainable because of the broader base of popular support that they could engender, also were seldom justice-related. For instance, the general amnesty of 403 B.C. included all citizens who had participated in the Athenian Civil War, except for the leaders most directly involved in the insurrection.80 This pragmatic use of clemency by the moderates and democrats united the city-state of Athens and kept it from meeting a divisive end.81 Similarly, most other amnesties recorded in Greek history appear to have been granted to promote solidarity during emergencies such as the Peloponnesian and Persian wars.82

The Romans had a more refined system of clemency than the Greeks, but also frequently used it for purposes unrelated to justice or fairness. The most well known instance of the exercise of clemency under Roman law--Pilate's pardon of Barabas rather than Jesus83--exemplifies the Romans' propensity to use pardons "often and skillfully for their political ends."84 The judicious use of clemency was an effective means of quelling discord among the subjugated inhabitants of the Roman Empire. The Roman practice of disciplining mutinous troops through decimation--the killing of every tenth soldier--rather than executing an entire army of wrongdoers, is another example of using clemency in a politically expedient fashion, maintaining discipline while preserving resources that could prove useful to the state.85

The Romans were also willing to employ the pardon power to excuse crimes of which the people disapproved, but which furthered patriotic ends. Horatio, who killed his sister when she had the bad judgment to "bewail" the death of a foe of Rome to whom she was betrothed, was pardoned by the citizenry because of his patriotism.86 Finally, the Romans introduced an element of chance into the clemency process by automatically pardoning criminals sentenced to death if they encountered a Vestal Virgin on the way to the place of execution, so long as the encounter was an accidental one.87

As these examples illustrate, the clemency power of the ancients was often marked by considerations of expediency rather than justice. This is understandable, because the law itself was thought by many early cultures to originate in infallible sources such as God88 or the people,89 and thus to embody principles of justice adequately. If the law was itself synonymous with justice, there would have rarely been need to create exceptions to the law to further justice. This tendency to use the clemency power for reasons unrelated to justice eventually found its way into the law of England.

C. The Development of the Clemency Power in England

The Roman penchant for solidifying power through the shrewd use of clemency also manifested itself in England. Blackstone implicitly recognized the Roman roots of Britain's clemency power when he derided the cruelty of the inhabitants of the Isle of Gurnsey for failing to postpone the execution of a pregnant woman, a "barbarity which they never learned from the laws of antient Rome."90 In his Commentaries, Blackstone said that the Crown's use of the pardon power to ensure that justice was administered with mercy was one of the great advantages of monarchy over any other kind of government, because it softened the rigors of the general law.91 However, the king's use of the clemency power to enhance justice was apparently not an end in itself. Rather, the purpose of these "repeated acts of goodness" was to consolidate the monarch's power: Blackstone observed that acts of clemency "endear the sovereign to his subjects, and contribute more than any thing to root in their hearts that filial affection, and personal loyalty, which are the sure establishment of a prince."92 Coke likewise noted that "[m]ercy and truth preserve the king, and by clemency is his throne strengthened."93 Although the clemency power ultimately became a central component of the royal prerogative, the Crown at first had many competitors for the pardoning power, including the clergy, the great earls, and feudal courts.94 Gradually, however, the power to grant clemency was formally consolidated in the Crown, with Henry VIII finally seizing the pardoning power in 1535. In that year, Parliament was persuaded to pass an act committing to the king "the hole and sole power and auctoritie" to pardon or remit treasons, murders, manslaughters, felonies, or outlawries.95

The vesting of the clemency power in the Crown96 had survived in virtually absolute form97 for almost 165 years when a constitutional crisis of unprecedented proportions precipitated a dispute over whether the king could use the clemency power to frustrate Parliament's impeachment of the Lord High Treasurer of England. The impeachment of Thomas Osborne, Earl of Danby, in 167898 presented a scenario strikingly similar to the Iran-Contra scandal that shook the final years of the Reagan Administration. Osborne, in his capacity as Lord High Treasurer of England for Charles II, secretly followed the king's directive to extend an offer of neutrality to France in exchange for a substantial payment.99 This offer was made only five days after Parliament had passed an act to raise and appropriate funds for a war with France. Although Osborne was carrying out the express orders of Charles II, the king was beyond the reach of Parliament in this dispute over how foreign policy should be conducted. Consequently, Osborne was singled out by Parliament for impeachment, which in England included not only removal from office, but also more severe types of punishment.100

When Charles II, unlike Ronald Reagan in the Iran-Contra affair, frustrated the legislature by pardoning Osborne prior to the conclusion of the impeachment process, a crisis over the scope of the clemency power ensued. An outraged Parliament questioned the legality of the pardon and considered measures to limit the scope of the royal pardoning prerogative.101 A political compromise eventually resulted in Osborne's impeachment being aborted, although he remained imprisoned for five years in the Tower of London.102

In the aftermath of the Osborne impeachment, Parliament finally succeeded in limiting the royal pardoning prerogative, passing several acts which diminished the king's clemency power. The Habeas Corpus Act of 1679103 prohibited royal clemency in cases in which persons were convicted of causing others to be imprisoned outside of England, and thereby placing them beyond the reach of English habeas corpus protections.104 Although not affecting the clemency power directly, the 1689 Bill of Rights105 deprived the Crown of its former power to suspend the operation of a given law or disregard its execution.106 In the 1700 Act of Settlement,107 Parliament succeeded in permanently removing pardon as a bar to impeachment, although the Crown's power to pardon after sentencing was not correspondingly limited.108 Finally, in 1721, Parliament gained the power to pardon by legislative act.109

These limitations on executive clemency did not alter the established practice of exercising the power for reasons entirely unrelated to justice. Although Blackstone is undoubtedly correct in asserting that clemency in England often served the salutary purpose of mitigating a system of criminal justice which was harsh and inflexible,110 the lack of meaningful checks on the prerogative resulted in frequent abuses.111 In her study of early uses of clemency in homicide cases, Naomi Hurnard recounts that pardon was "scarcely ever available to those condemned to death in error, but in other respects it was grossly overemployed."112 Kings also used the clemency power to win the support of key nobles and clerics during times of strife113--even the outright sale of pardons was commonplace.114 Such questionable uses of clemency prompted widespread criticism of royal pardoning practices.115

In addition to enhancing the individual monarch's wealth and power, the justice-neutral use of pardons also furthered the needs of the state. For instance, the Crown employed pardons to provide cheap labor for the American colonies; felons were typically granted a pardon conditioned on their agreeing to travel to the colonies and work on the plantations.116 Similarly, the Crown used the conditional pardon to "man the navy" in the eighteenth century.117 The clemency power was also used to exact testimony from accomplices that would incriminate codefendants,118 a practice that became a "mainstay" of the English criminal justice system in the eighteenth and nineteenth centuries.119

Thus, while the clemency power in England had limits, they were not related to whether the institution of clemency was employed to serve the interests of justice. It was against this background of English clemency practices that the drafters of the United States Constitution vested in the President the power to grant reprieves and pardons.

D. The Clemency Power in the United States

1. Federal Clemency Power.-- By choosing to repose the clemency power in the chief executive alone, the Framers of the Constitution aligned themselves with a vision of the power that was decidedly British in nature. When America was colonized, the king in most instances delegated the pardoning power to his counterpart and direct representative in the New World, the royal governor. The colonial charters of Virginia, Massachusetts Bay, Maryland, Maine, the Carolinas, New Jersey, Pennsylvania, and Georgia committed the clemency power to the executive.120 Although the charters of Connecticut and the Rhode Island and Providence Plantations provided that the general assembly could pardon or release criminals, it could only do so if the governor and six of the assistant governors were present.121

However, the Revolution ushered in a period of distrust of strong executive authority and temporarily brought to an end the executive's clemency monopoly. By the time the Constitution was drafted in 1787, most state governments placed the power to remit punishment for crimes in the legislative council and the governor jointly,122 or in the legislature alone.123 The pardoning power was exercised solely by the governor only in New York, Delaware, Maryland, North Carolina, and South Carolina.124

Thus, it was the British clemency model, not the model prevailing in most states, that Alexander Hamilton was following when he objected to the Virginia and New Jersey plans125 and proposed that a supreme executive "have the power of pardoning all offences except Treason; which he shall not pardon without the approbation or rejection of the Senate."126 The Report of the Committee of Detail retained the President as the sole repository of the power to grant reprieves and pardons, but instead of excepting treason, the report used language very similar to the English Act of Settlement127 in providing that presidential pardon "shall not be pleadable in Bar of an Impeachment."128 This language was shortened to its present form--"except in cases of impeachment"--without reported debate.129

The few reported exchanges at the Convention concerning the clemency power pertained to attempts to restrict the scope of the President's power. A motion to limit the President's power by allowing the President to grant reprieves until the ensuing session of the Senate and to grant pardons only with the consent of the Senate was soundly rejected.130 Concern that granting the President unlimited power to pardon in cases of treason was too dangerous prompted Edmund Randolph to move that the power be withdrawn in such cases.131 While a number of the delegates, including James Madison, agreed that the power to pardon treason should not be vested in the President alone, Randolph was unwilling to accede to those who would grant the power to the President and Senate jointly, and, thus, the motion failed.132

Following the federal Convention, critics of the proposed constitution voiced objections similar to those of George Mason of Virginia, who argued that the President's unrestrained power to pardon might lend itself to abuse, particularly in cases of treason.133 In their responses to such arguments, the two most often cited defenders of the clemency power, Alexander Hamilton134 and James Iredell,135 pursued nearly identical lines of reasoning. First, both noted the justice-enhancing aspects of clemency, arguing that the power is necessary to assure that people are treated fairly. Hamilton argued that the "benign prerogative of pardoning" should be as little fettered as possible so that exceptions in favor of "unfortunate guilt" could be made; otherwise, "justice would wear a countenance too sanguinary and cruel."136 Iredell was even more explicit:

[T]here may be many instances where, though a man offends against the letter of the law, yet peculiar circumstances in his case may entitle him to mercy. It is impossible for any general law to foresee and provide for all possible cases that may arise; and therefore an inflexible adherence to it, in every instance, might frequently be the cause of very great injustice. For this reason, such a power ought to exist somewhere; and where could it be more properly vested, than in a man who had received such strong proofs of his possessing the highest confidence of the people?137

Despite their initial observations on clemency's justice-enhancing aspects, however, both Hamilton and Iredell devoted the bulk of their responses to defending the clemency power on justice-neutral grounds. Hamilton contended that the principal argument for vesting the power to pardon in the President alone was that in "seasons of insurrection," a well-timed offer of pardon to the rebels could be essential to the preservation of the government.138 Iredell also argued that shrewd use of the pardoning power might prevent a civil war, but cited additional justice-neutral reasons for giving the power to the President. Iredell contemplated that the clemency power could be used to procure the testimony of the accomplices of great criminal offenders and to protect that "set of wretches whom all nations despise, but whom all employ"--namely, spies who have proved useful to the government.139 Such arguments, with little examination,140 carried the day. Thus, article II of the Constitution vested the "Power to Grant Reprieves and Pardons for offenses against the United States, except in cases of Impeachment" solely in the President.

Once the Constitution was ratified, Presidents did not hesitate to use clemency in justice-neutral ways that benefitted the republic, and, some contend, themselves. In 1795, President Washington granted an unconditional pardon to many of the participants in the Pennsylvania Whiskey Rebellion.141 John Adams, in order to serve "the public good," likewise issued a presidential pardon to all persons involved in an insurrection in Pennsylvania.142 After the Federalists were soundly defeated in the election of 1800, President Jefferson utilized the clemency power to pardon all those convicted and sentenced under the Alien and Sedition Act, which the Federalists had used, in clear violation of the First Amendment, to silence the Jeffersonian Republicans.143 However, Leonard Levy suggests that this use of the pardon power may have been motivated by politics rather than ideology, since Jefferson later failed to protest when Federalist publishers, editors, and judges were prosecuted for seditiously libeling the United States and Jefferson.144 President Jefferson also attempted to utilize the pardon power in pragmatic fashion to procure the testimony of an alleged "accomplice" of Aaron Burr during Burr's treason trial.145 This attempt failed when the witness indignantly refused the pardon and declared his and Burr's innocence on the stand, resulting in the acquittal of Jefferson's long-time political rival.146 The executive clemency power was also used in justice-neutral fashion, as it had been in England, to man the navy.147

The use of clemency to restore tranquility to the nation became especially pronounced following the Civil War, when both Lincoln and Andrew Johnson repeatedly issued amnesties to persons who had fought against the Union, conditioned on their voluntarily taking an oath to uphold the Constitution.148 Although Lincoln's amnesty proclamation was authorized by statute,149 Congress eventually became disenchanted with what were widely perceived as lenient presidential pardoning policies150 and, during Johnson's administration, sought to curtail the clemency power through legislation.151 However, unlike the British Parliament, which had succeeded in placing some limits on the royal pardoning prerogative nearly two centuries before,152 Congress's efforts to restrict the President's power were unavailing, largely because of the expansive interpretation given to the presidential pardoning power by the branch whose duty it is to say what the law is:153 the judiciary.

In United States v. Wilson,154 the first Supreme Court case to discuss the presidential clemency power, Chief Justice John Marshall defined a pardon as "an act of grace, proceeding from the power entrusted with the execution of the laws, which exempts the individual, on whom it is bestowed, from the punishment the law inflicts for a crime he has committed."155 Implicit in Marshall's definition is the notion that the executive can dispense such "grace" in any form, at any time, and for any reason, in all cases except impeachment. Although the Court later eschewed Marshall's vision of a pardon as something akin to divine forgiveness,156 it nevertheless has consistently interpreted the President's clemency power broadly.157

The President has been held to possess the power to grant not only reprieves and pardons as expressly authorized in the Constitution, but every other form of clemency as well. In Ex parte Wells,158 the Court interpreted the presidential pardoning power as commensurate with the English Crown's and held that the chief executive could condition the grant of pardons on the recipient's acquiescence to virtually any terms.159 During the Reconstruction era following the Civil War, when Congress was for the first time questioning the presidential power to grant general amnesties,160 the Court confirmed that the executive not only could dispense this form of clemency,161 but could also grant conditional amnesties.162 In a similar vein, the Court held in Osborn v. United States163 that the presidential clemency power includes the power to remit fines and forfeitures.164 More recently, the Court in Biddle v. Perovich165 limited an earlier holding that a presidential pardon could be refused166 and decided that the clemency power extended to commutations of sentences regardless of whether the recipient accepted the proffered commutation.167

The Court has also apparently rejected the notion that the coordinate branches of government can limit the executive's clemency power. In Ex parte Garland,168 the Court gave its broadest characterization of the scope of executive clemency:

The [clemency] power thus conferred is unlimited, with the exception [in cases of impeachment]. It extends to every offence known to the law, and may be exercised at any time after its commission, either before legal proceedings are taken, or during their pendency, or after conviction and judgment. This power of the President is not subject to legislative control. Congress can neither limit the effect of his pardon, nor exclude from its exercise any class of offenders. The benign prerogative of mercy reposed in him cannot be fettered by any legislative restrictions.169

In Schick v. Reed,170 the court reaffirmed this view by holding that in commuting a criminal sentence, the President was not limited to replacing it with another legislatively authorized sentence for the particular crime.171 Chief Justice Burger reasoned that, because the power to pardon flows from the Constitution alone, "it cannot be modified, abridged, or diminished by the Congress."172 Unfortunately, the majority opinion, insofar as it relies on the history of English pardoning practices to support this proposition, is not analytically sound. As discussed above, the English Parliament eventually did impose limits on the pardoning power,173 and thus the English experience would seem to support the opposite conclusion. The substantial differences between the English system of a supreme Parliament and the American constitutional system of enumerated congressional powers further undermine the usefulness of such comparisons in illuminating the proper relationship among the American branches of government.174 Still, the Constitution's commitment of the clemency power to the executive branch, coupled with the drafters' rejection of a legislative check on the executive clemency power,175 makes it reasonable to conclude that Congress cannot limit the President's clemency power.176

A more difficult issue implicit in Schick concerns the authority of the judiciary to limit the President's clemency power, an issue which pits the notion that the presidential clemency power should be unfettered against the principle that the judiciary is responsible for reviewing the constitutionality of executive actions. In Ex parte Grossman,177 the Court, in typical fashion, construed the clemency power expansively, this time at the judiciary's expense, by holding that the power to pardon extends to the offense of criminal contempt of court. In Grossman, the Department of Justice argued on behalf of the trial court that the President cannot pardon criminal contempt of court, because to do so would impair the power and functions of the judiciary.178 In rejecting this separation-of-powers argument, Chief Justice Taft reasoned that the possibility that the pardoning power could be perverted so as to destroy the deterrent effect of judicial punishment is not a sufficient basis for limiting the President's discretion to grant clemency.179 "Our Constitution," wrote Taft for a unanimous Court, "confers [full] discretion [to pardon] on the highest officer in the nation in confidence that he will not abuse it."180 If the power were abused, the remedy, according to the Grossman Court, would be "resort to impeachment rather than to a narrow and strained construction of the [clemency power] of the President."181

The holding in Grossman has been criticized as threatening to subvert our system of institutional restraints and divided powers by permitting the executive to undermine the ability of coordinate branches to carry out their respective functions.182 The courts, however, have not entirely refused to limit the exercise of the clemency power via the judiciary. Instead, courts have gradually developed a bifurcated approach to evaluating exercises of clemency that treats the President's reasons for using the power as sacrosanct, but recognizes that courts may review and invalidate some pardons because of their impermissible effect.

Hoffa v. Saxbe183 most clearly illustrates this bifurcated approach to analyzing presidential exercises of the clemency power. In Hoffa, the district court considered the validity of a commutation extended by Richard Nixon to Jimmy Hoffa that was conditioned on Hoffa's not engaging in union activities, either directly or indirectly, before his original sentence expired.184 Hoffa challenged the validity of the conditional commutation on two grounds. First, Hoffa attacked Nixon's reason for issuing the pardon. He contended that the condition imposed by the President was invalid because it was formulated as the result of a conspiracy involving Nixon, Charles Colson, and Teamster's union president Fred Fitzsimmons.185 Second, Hoffa argued that the condition was unconstitutional because it infringed on his first-amendment rights.186

The court refused to inquire into the President's reasons for issuing the pardon to Hoffa. The court observed even if Hoffa were correct that Nixon, in order to gain political advantage, had conspired with Teamsters officials to keep Hoffa out of the union, these improper motives could not invalidate the exercise of the clemency power.187 The court reasoned that, just as an act of Congress may not be attacked on the ground that the legislators who voted for its passage did so for improper reasons, so too is even the corrupt use of the presidential pardoning power insulated from judicial review.188

The Hoffa court's analogy is not a good one, however. Legislative motives are, in some instances, a proper subject for judicial inquiry, as the Supreme Court has made clear in its analysis of establishment clause189 and equal protection190 issues. Moreover, at least one district court has evaluated the merits of the exercise of the presidential pardon power.191 Still, the Hoffa court's conclusion that the President's motives in granting pardons may not be examined finds support in Grossman's suggestion that the remedy of impeachment expressly provided in the Constitution offers the sole vehicle for curing blatant abuses of the clemency power.

Notwithstanding this deferential approach concerning the exercise of the executive clemency power, the Hoffa court did show a willingness to review the effect that the judiciary should give to a presidential pardon. The court held that the "framework of our constitutional system" establishes "limits beyond which the President may not go in imposing and subsequently enforcing . . . conditions" on pardons.192 Thus, while the executive evidently may grant a pardon for any reason, she may not use a conditional pardon in a way that is not "directly related to the public interest" or in a way that will "unreasonably infringe on the individual commutee's constitutional freedoms."193 For example, if a President conditioned the granting of a pardon on the recipient's agreement to forego supporting any other candidate for political office, the condition would not be enforceable because it violates both prongs of the test promulgated by the Hoffa court.194 In applying the test to Hoffa's conditional commutation, the court substantially deferred to the executive, holding that the condition was related to the public interest when tested under a "reasonableness" standard.195 The court also held that the condition did not violate Hoffa's first-amendment rights because of the substantial governmental interest in preserving the integrity of labor organizations.196

The two-pronged test articulated in Hoffa remains the most explicit statement by a federal court of the limitations that may constrain the exercise of the executive clemency power.197 Although the Supreme Court has never spoken to the issue as directly, the Court has recognized that certain limitations on the power may be imposed through the process of judicial review. In Knote v. United States,198 the Court observed that a presidential pardon, like the king's pardon at common law, may not affect the vested rights of third parties.199 By similar reasoning, once proceeds from the sale of an offender's property have been paid into the United States treasury, a presidential pardon cannot secure the return of such proceeds to the offender, for to do so would contravene the appropriations power reserved to Congress in article I of the Constitution.200 In Burdick v. United States,201 the Court upheld an offender's right to refuse a presidential pardon granted in order to impel him to testify in a case in which he had asserted his right against self-incrimination.202 Implicit in the Court's reasoning is the notion that the clemency power must be balanced against the rights of a witness under the Fifth Amendment: "[b]oth have sanction in the Constitution, and it should, therefore, be the anxiety of the law to preserve both,--to leave to each its proper place."203 The Court in Schick v. Reed204 likewise recognized that the executive is free to attach any condition on the pardon "which does not otherwise offend the Constitution," presumably as determined by the judiciary.205 Thus, there is support in principle for the notion that the judiciary can review the constitutionality of certain aspects of the executive clemency power. However, only in Burdick, which was limited in a later case,206 has the Court actually used the power of judicial review to curtail the President's clemency power.

As a practical matter, then, the exercise of the clemency power has been subject only to the constraints that each chief executive has chosen to impose upon himself. Today, lawyers in the Office of the United States Pardon Attorney review all clemency applications and make nonbinding recommendations to the President. The President, with the help of advisors, then makes the final decision based on whatever factors he considers important, including political ones.207 "Clemency," as one deputy pardoning attorney puts it, is after all "unavoidably in some ways a political act."208 Over time, the politics inherent in clemency decisions have resulted in a patchwork of presidential pardoning practices that at best can be described as idiosyncratic. In his seminal work on the presidential pardoning power, Dr. W.H. Humbert observed:

Reasons for clemency from 1860 to 1932 perhaps varied with changes in conditions in the country, with the advancement in thought on penal questions, and with the periodic changes in the administrative personnel of the government. An adequate reason for clemency in normal times became an inadequate reason for clemency in times of trouble. Confinement for protection and reformation replaced confinement for protection and punishment. Circumstances which one President considered mitigating, another President considered unconvincing.209

Examination of Humbert's catalogue of reasons given by Presidents for granting clemency reveals that the rationales have ranged from those which are obviously justice-enhancing (e.g., doubt as to guilt, absence of criminal intent, insanity) to those which are so unrelated to justice as to border on abuse of the power (e.g., respectability of the prisoner's family, sympathy,210 recommendations by influential citizens, ability to earn a living after release).211

In recent times, however, the lack of consistent, principled standards governing the exercise of executive clemency has led not to the expansive use of the pardoning power, but to its atrophy. In the past twenty years, the use of the federal clemency power has precipitously declined, particularly its justice-enhancing manifestations. President Nixon granted thirty-six percent of all clemency requests (pardons and commutations combined), President Ford granted twenty-seven percent, and President Carter slightly more than twenty-one percent.212 The administration of Ronald Reagan represented the presidential pardoning nadir, in which only about twelve percent of the applications for pardons and commutations were granted.213 President Bush has shown signs of continuing the parsimonious pardoning practices of his predecessor, approving only four percent of the clemency applications received in 1989 and no requests for clemency in 1990.214

The atrophy of the clemency power is even more pronounced than these statistics indicate. Even when clemency is granted today, it rarely results in the remission of punishment. The executive clemency power has been used as far back as the 1950s primarily as a means of rehabilitating the criminal record of those who have long since been released from prisons, rather than as a way of assuring that justice is served.215 The rarity of granting clemency on grounds of innocence216 is due at least in part to a conservative philosophy about clemency's proper role. As Raymond Theim, deputy United States pardon attorney during the Carter, Reagan, and Bush Administrations, explained:

"The feeling is that we should do as little as possible to grant relief. . . . It's a dangerous trend for the executive to override the function of the courts and the parole system too much, both from the point of view of balance of power and of possible corruption. . . . Clemency is bestowed as an act of grace, not as a matter of right."217

The regulations governing applications for clemency also contribute to its infrequent use. For instance, the regulations adopted during the Kennedy Administration, which continued in effect until 1983, provided that no petition for pardon could be filed until at least three years after the offender's release from confinement, and in the case of "serious" crimes, such as those involving violence or narcotics, not until five years after release.218 In 1983, the Reagan Administration extended the waiting period for application for presidential pardon to five years after release from confinement and, for serious crimes, seven years.219 Applications for commutation, the form of clemency that would actually remit ongoing punishment in cases of unfortunate guilt, may be made "only if no other form of relief is available . . . or if unusual circumstances exist, such as critical illness, severity of sentence, ineligibility for parole, or meritorious service rendered by the petitioner."220

As one might predict, these regulations have resulted in presidential clemency only rarely being granted when it would affect the punishment inflicted.221 Since 1969, Presidents have granted sixteen times more pardons than commutations.222 Moreover, the percentage of commutation applications granted has steadily declined over the past twenty years: Nixon granted about seven percent of all requested commutations, Ford about five percent, Carter about three percent, Reagan, and, thus far, Bush, less than one percent.223

This dissipation of the clemency power is largely attributable to a lack of standards governing its exercise. Because there is no principled basis for remitting punishment, each clemency decision becomes an ad hoc one that hinges on the vagaries of the individuals weighing the factors they believe are significant in clemency determinations. A United States pardon attorney may believe that the disparateness of the applicant's sentence warrants clemency, while a White House advisor reviewing the application may change the recommendation because of the nature of the crime or potential political ramifications. In discussing the granting of clemency applications, pardon attorney John Stanish observed in 1979 that "`we have preferred to err on the side of too little rather than too much, because we haven't had any objective standards to work from.'"224

Thus, presidential clemency has been so trivialized that it is now used almost exclusively to cleanse the records of federal criminals after they have managed to stay out of trouble for the requisite five or seven years. The clemency power has become little more than a certification that, "with the advantage of FBI information and extensive study, the President has judged that the petitioner is clean."225 Justice-enhancement, in the sense that I have used that term to denote "justice as fairness," evidently has little to do with presidential clemency.226 Were Hamilton alive today, he might well not recognize the pardoning power in its present form as the same power he first proposed as being essential to provide "easy access to exceptions in favor of unfortunate guilt."227

2. The Clemency Power in the States.--Immediately following the American Revolution, most state governments rejected the British legacy of complete executive control of the clemency power. Eight of the thirteen states vested the authority to remit punishment in an executive legislative council and the governor jointly, or in the legislature alone.228 However, this "postindependence republican faith in legislative bodies" soon waned.229 The development of state constitutions and, presumably, the influence of the newly adopted federal constitution led to the abolition of the legislative council and an increase in the governor's clemency powers in a number of states.230 The idea that the executive branch was the proper repository of the clemency power rapidly gained popularity, and most of the new states admitted to the Union allocated the power to the governor alone.231

Today the governor retains control of the clemency power in most jurisdictions, typically pursuant to the state constitution. Twenty-nine states place the clemency power in the governor alone, although most of those states have established an advisory body that makes nonbinding recommendations to the chief executive.232 In sixteen states, the governor shares the power to make clemency decisions with an administrative board or panel.233 In the five remaining states, an administrative panel, usually appointed by the governor, has the principal authority to make clemency decisions.234

Apart from some limitations on the types of offenses for which clemency can be granted (treason and impeachable crimes are commonly excluded), only a handful of states have promulgated statutory or administrative standards governing use of the power.235 Thus, the findings of several studies of state clemency decisions are hardly surprising: each governor has different ideas about the function of executive clemency and, as with the administrations of various Presidents, the rate of granting clemency varies dramatically from administration to administration,236 as well as from state to state.237 Clemency is often seen by its purveyors as coming from their personal bounty and grace, and hence there is no need to justify its exercise.238 Consequently, it has often been the practice of the executive to grant "seasonal commutations" around holidays such as Christmas and Thanksgiving.239 Other governors take the position that pardons should be issued for reasons of justice, regardless of the season.240 Another frequently cited ground for granting clemency is to mitigate disproportionate sentencing, although some governors believe that to use clemency in this manner usurps the discretion of the sentencing judge.241

Although the variety of pardoning practices makes it difficult to generalize, it is evident that the executive's exercise of the clemency power is a more controversial matter at the state level than at the federal. Dr. Humbert speculates that this is because, unlike governors, the President "is ordinarily more remote from the scenes of the crimes and does not, therefore, have to reach a decision amid intense feelings which some cases arouse."242 It could also be attributable to the increasing reluctance of Presidents to exercise the power in any meaningful fashion.243 Whatever the reason, with the exceptions of Andrew Johnson244 and Gerald Ford,245 the clemency power has not been a significant factor in the political lives of our Presidents. The same has not been true of our governors.

Abuse of the pardoning power has led to the removal from office of one governor and indirectly, to the indictment of another for improprieties committed while in office. In 1923, Oklahoma Governor J.C. Walton was impeached and removed from office for selling pardons to hundreds of individuals.246 In the early 1980s, Tennessee Governor Ray Blanton's administration was similarly embroiled in a scandal involving the widespread sale of pardons and commutations, which ultimately led to Blanton's federal prosecution on charges of conspiring to take kickbacks for liquor store licenses.247

More common than outright removal from office is the toll that the exercise of the clemency power has taken on the political careers of various governors, which in turn has affected the way in which the clemency power has been used. A firestorm of criticism followed Illinois Governor John Peter Altgeld's unconditional pardon in 1893 of the three surviving anarchists who had been wrongfully convicted in 1887 of a bombing in Chicago's Haymarket Square during a labor uprising.248 Overnight, Altgeld became one of the most reviled men in the country. Newspapers across the nation accused him of being an "apologist for murder," a "fomenter of lawlessness," and of having encouraged "anarchy, rapine and the overthrow of civilization."249 Not surprisingly, he was not reelected to a second term as governor.

Similarly, Ohio Governor Michael DiSalle's political career was shortened, at least in part, because of his commutation of the death sentences of six individuals. Following the defeat of his bid in 1962 for a second term as governor, DiSalle recounted a conversation with members of the Statehouse press corps, who complained that he had let them down in losing the election:

"You left yourself wide open when you could have ducked. . . .

. . . .

[By commuting the death sentences, y]ou practically begged my editor to clobber you. He called you a bleeding heart and a fuzzy-thinking do-gooder, and you got licked."

"So you saved the lives of six nonentities," said the other man. "And who cares? If you'd kept your mouth shut, the world would be no poorer, and you'd be around for another four years to fight for the underdog."250

In his recent book, Public Justice, Private Mercy,251 former California Governor Edmund (Pat) Brown poignantly recalls the way political pressures affected his clemency decisions. Although Brown was morally opposed to the death penalty, he sometimes refused to commute capital sentences because of political considerations. In one case involving the killer of a six-year-old girl, Brown was convinced that the condemned murderer was mentally defective because of an injury he had suffered as a child.252 Brown believed that executing this criminal would amount to an act of societal vengeance rather than justice. However, while he was deciding whether to commute the sentence to life imprisonment, Brown learned that a legislator with the swing vote on an important piece of legislation for migrant workers was strongly in favor of the execution and would withhold his support if Brown granted clemency. Brown wrestled with his angel:

Rose Marie Riddle was dead, and nothing I could do would bring her back. By letting Richard Lindsey go to the gas chamber, I was giving her parents [who were farm laborers] and people like them a chance at a living wage. The scales tipped. I picked up my pen and on the first page of the clemency file wrote these words: "I will take no action." . . . Four days later, Lindsey was dead. That same week, the farm labor bill passed through committee and a few months later was signed into law.253

The decision has haunted Brown ever since.

Because of most officials' reluctance to reveal why they used, or failed to use, the clemency power, it is hard to ascertain precisely which factors influence their decisions. However, it is evident that political considerations continue to erode the justice-enhancing function of clemency in the states, as they have throughout history. A condemned murderer in Virginia has garnered widespread support for commutation of his sentence because of doubt about his guilt, yet the governor and attorney general have thus far withheld both clemency and a new trial,254 apparently because of fear of appearing weak on the law-and-order issue. A former governor of Louisiana, because of a campaign promise he had made to one of the criminal's victims, refused to commute the sentence of a black murderer who had served a longer sentence than any murderer in the state and who was by all reports rehabilitated.255 Current Louisiana governor Roemer recently disregarded a politically unpopular recommendation of the Pardon Board and refused to commute the death sentence of a convicted murderer who was seventeen years old at the time of the killing, and suffered from brain damage caused by physical abuse as a child.256 In 1988, the Governor of Ohio denied the commutation request of Larry Brown, a minor property offender who had served nearly eight years of a 22- to 55-year sentence for credit card fraud.257 The denial of clemency was unexpected in light of the harshness of the sentence and the doubts that had been raised about Brown's guilt.258 The previous commutation of the sentence of a judge who had arranged to have his wife killed259 had exposed the governor to harsh criticism, and it was widely thought that he denied Larry Brown's commutation request because of fear of further political fallout.260 Wariness of political repercussions may also be why the number of commutations has declined since 1982261 even as the number of persons slated for execution has increased.262 As Edward Hammock, the former chairman of the New York State Board of Parole, which reviews clemency applications for the governor, notes: "[W]ith the governor, it's not the individual case he's looking at. He picks a few from that pool of eligible individuals on the basis of a political statement he wants to make."263

Such anecdotal evidence supports the view often voiced by critics that clemency decisions in many states continue to be made for inappropriate reasons--reasons that are unrelated to justice. In practice, the process in many states is reminiscent of the ancient Greek procedure of adeia discussed above:264 the most important factor in successful clemency applications appears to be the widespread support of influential individuals in the community.265 Hammock likened the clemency process to a political campaign:

"To get your application looked at, you need a groundswell of support. You need mail, petitions to the governor, rallies. If you're John Inmate sitting in Auburn state prison cooling your heels, and the only friend you've got is a correctional officer who writes the nicest letter in the world for you . . . well, what kind of chance have you got? . . . I agree it's unfair. . . . But it's like trying to become president. You can be the finest candidate in the country, but you have to be able to get the people to vote for you."266

Recent commutations in Ohio appear to have been granted in response to precisely the sort of public campaigns described above.267

That political and other justice-neutral considerations should influence the clemency process is hardly surprising given the lack of objective criteria governing the exercise of the power. However, in view of the importance of the executive clemency power to our system of punishment and its potential to frustrate the functioning of coordinate branches of government, such haphazardness should no longer be tolerated. The remainder of this Article will be devoted to identifying principled standards and effective procedures that would help transform the clemency power from a boon benevolently bestowed on the blessed at random, into the "constituent part of the judicial system"268 that it has long been touted to be.

III. A Proposal for Bifurcation of the Clemency Power: Replacing Politics with Justice

The foregoing discussion suggests that the clemency power has failed to evolve with the rest of the judicial system. United States pardon attorney John Stanish's observation in 1979 that "`[t]here has never really been much rhyme or reason to clemencies in the past'"269 is still true with respect to both federal and state executive clemency decisions.

Today, however, our changing attitudes about crime and a trend toward removing flexibility and discretion in the sentencing of those convicted of crimes have created an increasing need for clemency as a means of remitting punishment in a principled and consistent fashion. Congress's recent enactment of the Federal Sentencing Guidelines represents "the most pervasive revolution in federal criminal procedure since the adoption of the exclusionary rule."270 The Guidelines, adopted in 1984, abolish parole and in its stead create a determinate sentencing scheme that virtually eliminates judicial discretion in sentencing.271 State legislatures have also taken steps to limit discretionary releases by parole authorities.272 The Supreme Court has upheld the constitutionality of state laws mandating the death sentence in certain cases, further limiting individualized sentencing in capital cases.273

At the same time, no doubt because of our changing views about the need for inflexible punishment, overcrowding is reaching crisis proportions in prison systems throughout the country.274 Unfortunately, a significant number of those serving time in our prisons are wrongfully incarcerated.275 In the wake of Randall Dale Adams's release from a Texas prison after serving twelve years on death row for a crime he did not commit,276 much attention has been focused on the injustices perpetrated by our penal system.277 One study by criminologists, conservatively assuming a wrongful conviction rate of one-half of one percent, estimates that about 6000 Americans are convicted every year of serious crimes that they did not commit.278 Another study of convictions in cases in which the death penalty was or could have been imposed reveals at least 350 individuals who have been erroneously convicted since 1900.279 A recent study of nearly one hundred capital cases in Florida, Alabama, Georgia, Louisiana, Texas, and Mississippi, where eighty percent of all executions have taken place since capital punishment was resumed in 1976, reveals that the imposition of the death penalty is related largely to whether defendants are able to afford experienced counsel or are forced to accept the representation of "ill-trained, unprepared, . . . grossly underpaid" court-appointed lawyers.280

Clemency, properly exercised and freed of political pressures, represents an ideal vehicle for remedying many of the problems inherent in an imperfect, overloaded, and increasingly rigid system of criminal justice. The extrajudicial corrective of clemency provides a safety valve for our criminal justice system, another opportunity for an offender to tell her story more thoroughly, or at least differently, than she could at trial.281 Moreover, with proper safeguards to assure access to sufficient information, clemency can provide a sense of perspective on how fairly a particular justice system is operating.282

Yet, unless our views about the nature of clemency are themselves changed, the history of clemency suggests that it will continue to be a remedy marred by randomness and outright abuse. Although the principal value of clemency today lies in its capacity to enhance justice in cases in which the infliction of the prescribed punishment would be manifestly unfair, some instances could arise in which justice-neutral uses of clemency would also have legitimacy.283 While we are unlikely to encounter again circumstances like those which followed the Civil War, when the amnesties of Lincoln and Johnson facilitated a national reunification that probably could not have been accomplished by the politically volatile Congress,284 there may nevertheless be times in which an act of clemency is justified by public welfare considerations.

Unfortunately, both justice-enhancing and justice-neutral aspects of clemency suffer when the executive has recourse only to her own moral and political sensibilities in making clemency decisions. This has led to the clemency power working in a way that resembles what scientists tell us of evolution: it is characterized by long periods of dormancy, punctuated by random bursts of fecundity.285 Clemency may be used for years in a relatively trivial way to rehabilitate the records of former offenders, until a governor goes on a pardoning binge at the end of her term,286 or political considerations prompt the pardon of a Nixon, a Hoffa, or perhaps a North. While it may be neither possible nor desirable to tinker with the mechanism of evolution, the institution of clemency would benefit from the adoption of consistent, principled standards to govern the use of clemency.287 I will discuss first the possible sources of such standards, and then consider what those standards should be.

A. Limitations on the Exercise of Executive Clemency Power

There are four possible sources of standards governing the exercise of the clemency power: the legislature, the judiciary, the people, and the executive branch itself. Each of these will be discussed seriatim. In considering the possible sources of clemency standards, this Article focuses primarily on the federal clemency power; however, inasmuch as many states have patterned their clemency systems on the federal model, with most vesting the power in the chief executive alone, many of the following observations are also relevant to the clemency process currently existing in a number of states.288

1. The Legislature.--Because the source of the clemency power under the federal system is an express constitutional grant to the executive, the legislature is generally thought to have effectively been removed from any role in the exercise of the executive clemency power.289 The constitutional grant of the clemency power to the executive branch is not exclusive, however. Although Congress may not restrict the President's use of the power, the power to grant clemency has been held to be vested concurrently in the Congress.290 This joint power to grant clemency is also found in some states.291

Consequently, while the legislature may not promulgate standards restricting the manner in which the executive can exercise the clemency power, Congress, as well as some state legislatures, could create a system of legislative clemency to augment existing executive clemency procedures.292 If the executive were unable or unwilling to exercise the clemency power to further the interests of justice, as seems to be true today, the legislature could create a parallel system for the remission of punishment. Presumably, the legislature would promulgate justice-enhancing standards governing the circumstances under which use of the power would be appropriate. The legislature could thereby have a ratchet-like effect on the institution of clemency: it could increase the opportunities for offenders to be granted clemency without restricting the executive's exercise of the clemency power.

2. The Judiciary.--Traditionally, the judiciary has refused to limit the executive's power to grant clemency.293 However, the Constitution presents no insurmountable obstacle to the judiciary's use of judicial review to impose standards on the exercise of the executive clemency power. If a court can refuse to give effect to a presidential pardon that attaches an unconstitutional condition294 or that affects the vested rights of third parties,295 to what extent can the judiciary invalidate a presidential pardon because it was issued for reasons that are contrary to public policy or that violate express provisions of the Constitution? Since it is the province of the judiciary to say what the law is, the courts must be willing to review the executive's exercise of the clemency power to assure that it comports with the Constitution. Moreover, the federal judiciary's relative independence from political pressure renders it an apt check on the far more politicized executive branch.296

Some commentators have gone so far as to urge the federal courts to impose limitations on the power to pardon, in order to "correct the flawed, self-serving component of the presidential pardon power and strengthen our constitutional governing system."297 Whether the judiciary could properly review "self-serving" assertions of the clemency power simply because they happened to be blatantly "self-serving," however, is problematic. What standards would the courts use in evaluating whether the executive's grant or denial of clemency was self-serving or contrary to public policy? In Murphy v. Ford,298 the district court considered whether President Ford's pardon of Richard Nixon was void because it was outside the scope of the President's pardoning power. Without questioning the appropriateness of its inquiry, the court evaluated Ford's reasons for pardoning Nixon and concluded that, because the country was "foundering in the wreckage of Watergate" and was in the grips of uncontrollable inflation and an unprecedented energy crisis, the pardon "was a prudent public policy judgment."299

As Murphy makes apparent, the courts in such cases can do little more than second-guess the executive's decisions based on the reviewing judge's own personal philosophy and standards. Notwithstanding the desirability of the judiciary's righting egregious abuses of the clemency power, the court is not usually equipped to do so. The judiciary is not the branch of government assigned the task of deciding what is good public policy; if a presidential pardon is not in the public interest, it is for the public to say so at the polls (as they did in the 1976 presidential election, when the Nixon pardon undoubtedly played a role in Gerald Ford's defeat). In addition, as the Court indicated in Ex parte Grossman,300 Congress can remedy unethical abuses of the clemency power through the impeachment process.301

The judiciary, however, should play a role in the clemency process. Courts would seem to have the responsibility, as well as the expertise and power, to decide when use of the pardoning power violates express provisions of the Constitution. For example, if Nixon had been impeached and then pardoned by Ford, the courts would surely have been obliged to declare the pardon invalid under the express language of article II: "except in cases of impeachment."302 By the same token, dispensing clemency in a way that conflicts with other provisions of the Constitution would also be within the power of the judiciary to invalidate. The Court fulfilled this responsibility in Burdick v. United States303 when it held that clemency could not be used to deprive a witness of his fifth-amendment rights, though the archaic rationale used to support its holding was that a pardon must be accepted to be effective, as had been true at common law.304

In addition, in some instances, inquiry into the President's motives for granting or denying clemency may be necessary to effectuate constitutional protections. If the President acted for racially discriminatory reasons and granted clemency to all white applicants for pardons, while denying clemency to all black applicants, the judiciary could review and presumably invalidate such use of the clemency power on equal protection grounds.305 Unlike instances in which the clemency power was challenged as being contrary to public policy, the judiciary would not be lacking standards against which to measure the propriety of the exercise of the clemency power, because the body of case law interpreting and applying the equal protection clause supplies a constitutional standard applicable to all governmental action.306

One possible argument against the assertion of such a power by the judiciary is that in eighteenth-century England, executive pardons were not subject to judicial review,307 and the Court, in interpreting the scope of the clemency power, has traditionally considered the manner in which clemency was exercised at common law in England.308 Unlike our own constitutional system, however, the English system of government contemplates a very different role for the judiciary, and historically has not countenanced judicial review of the actions of coordinate branches of government.309 Thus, the English tradition has little relevance to what our own system would require of the judiciary in evaluating the constitutionality of executive exercise of the clemency power.

The principal obstacles to judicial review of the clemency power stem from the Supreme Court's self-imposed rules of justiciability and standing. Strict adherence by the Court to the political question doctrine enunciated in Baker v. Carr310 could cause the Court to defer to the executive branch in clemency matters. A persuasive argument can be made that the article II power to grant reprieves and pardons is a textually demonstrable constitutional commitment of the issue to a coordinate political department. Similarly, the Court might decide that the lack of judicially discoverable and manageable standards precludes effective evaluation of the exercise of the clemency power.311 Although these arguments are persuasive with respect to public-policy-based challenges to the President's use of the clemency power,312 they should not apply when the basis of the challenge is an alleged violation of an express constitutional provision, which the Court has a responsibility to review. Furthermore, the Court itself has implicitly recognized that judicially discernible limitations on the clemency power do exist.313

Justiciability is not the only roadblock to judicial review of the executive clemency power; the issue of standing is also a potential problem. Assuming that the executive has acted improperly in granting clemency, for example by pardoning an official who has been impeached, who would have standing to challenge such an action? The recipient of the clemency would have nothing to complain of, and thus could not demonstrate the requisite injury in fact. A federal taxpayer and citizen who wished to challenge the action would merely have suffered a generalized grievance common to all and thus would also lack standing to bring the suit.314 Perhaps the best plaintiff to challenge such an action under the Court's current standing doctrine would be a member of Congress who voted for the impeachment that was pardoned by the President. However, such an action would squarely raise a political question issue. In light of the manner in which the federal courts have skillfully evaded resolving the dispute between executive and legislative branches over the application and constitutionality of the War Powers Resolution,315 it seems unlikely that the judiciary would adjudicate a dispute between members of Congress and the President over the exercise of the clemency power.

Seeking judicial review of the executive's decision to deny clemency because of racial discrimination would present less of a standing problem, but more of a practical one. The standing issues would not exist because the prisoner would have suffered the requisite injury in fact. One would hope that the judiciary would not shirk its responsibility to give effect to the anti-majoritarian provisions of the Constitution such as the equal protection clause of the Fifth Amendment. However, unless the Court was willing to sit as a review board for each clemency decision of the executive, it might decline to adjudicate such a dispute for institutional reasons similar to those set forth by Justice Harlan in his dissenting opinion in Flast v. Cohen.316

Finally, even assuming that justiciability problems are surmounted, separation of powers principles could preclude judicial review of executive clemency decisions. The Court has noted that "[t]he very structure of the Articles delegating and separating powers under Arts. I, II, and III exemplifies the concept of separation of powers."317 Although the Framers did not consider whether the judiciary could limit the executive power to pardon, they did explicitly reject a proposal to vest the clemency power jointly in the executive and the Senate,318 indicating a desire to place the power to pardon solely in the executive branch. Thus, where the judiciary would in effect be substituting its judgment as to the propriety of an act of clemency for that of the executive, substantial separation-of-powers issues would be presented. However, the use of judicial review to ensure that no other provision of the Constitution is violated would not offend the principle of separation of powers.319 In conclusion, while the judiciary probably cannot invalidate the President's grant or denial of clemency on public policy grounds, it can and should use judicial review to ensure that the executive is using the clemency power constitutionally.320

3. The People.--It is of course possible that a constitutional crisis of unprecedented magnitude could result in the limitation of the executive clemency power by the people through amendment of the Constitution. Just such a crisis in England led to Parliament's imposition of constraints on the royal pardoning prerogative at the end of the seventeenth century.321 Such an eventuality seems unlikely, however. The most controversial pardon in our history--President Ford's pardon of Richard Nixon--failed to lead to structural changes in the federal pardoning power.322 In addition, politically savvy Presidents apparently realize that the failure to grant clemency, even to the most deserving individuals, represents far less of a problem than does exercise of the power. Ronald Reagan exemplified this trend when he declined to pardon Iran-Contra figure Oliver North, notwithstanding Reagan's stated belief that North was a national hero.323 With the executive clemency power rapidly falling into disuse, an upswell of popular support for a constitutional amendment limiting the clemency power cannot be anticipated in the foreseeable future.

4. The Executive.--@If past is prologue, the executive branch offers little hope for reform of the clemency power. Most of our recent Presidents have been content to continue the clemency practices of their predecessors and use the power on an ad hoc basis to cleanse the records of offenders long after their release from prison.324 President Ford, in order to facilitate a partial amnesty of those who had committed crimes related to the Vietnam War, did modify the clemency process somewhat in 1974 by creating a civilian Presidential Clemency Board charged with making nonbinding recommendations to the President.325 However, this Board was little more than an ideologically diverse group that served the same advisory function as the Office of the United States Pardon Attorney in the special case of the Vietnam amnesties. It does not appear to have been governed by the sort of principled standards, consistently applied, that are advocated in this Article. Consequently, it was criticized as "inequitable," and created the perception that the "government was not giving the same bargain to all who were similarly situated."326

Still, the executive branch remains the most logical choice to pursue refinement of the clemency power. Article II's delegation of the clemency power to the President, combined with various Supreme Court decisions, have given the executive broad discretion in the exercise of the power. A chief executive who wished to dispense clemency in a principled fashion could do so by implementing by executive order the sort of standards described below.

B. Proposed Clemency Commission

Any refinement of the clemency power should take into account the justice-enhancing and justice-neutral aspects of remitting punishment. Different considerations underlie these two types of clemency and they must be applied discretely, with a view to the function each serves. Justice-enhancing acts of clemency have a single purpose: they ensure that our penal system operates fairly, so that each person is rendered her due. Justice-neutral acts of clemency can serve a variety of ends, ranging from preserving the unity of the state to advancing the political or financial aims of those granting clemency.

It is neither necessary nor desirable to follow blindly in the tradition of the royal prerogative and vest both of these aspects of clemency entirely in the chief executive. While some form of political check on the judicial and legislative branches by way of justice-neutral clemency decisions may be appropriate, justice-enhancing clemency decisions should not be made in response to political pressures and other factors unrelated to the fairness of the punishment imposed. The Framers' hope that Presidents would readily remit punishment in cases of unfortunate guilt has not been realized.

Therefore, the clemency process should be bifurcated. While the executive alone would remain responsible for making occasional justice-neutral clemency decisions, justice-enhancing clemency decisions would be made by a professional board that is independent of the political pressures which inevitably distort the decisions of elected officials.327 Such a board could, like the federal judiciary, be appointed during good behavior,328 and be selected based on expertise in various areas relevant to assessing the fairness of the punishment imposed.329 Michael DiSalle, following his term as governor of Ohio, advocated the creation of such a body to make clemency decisions:

I would like to give the [clemency] commission complete independence from political considerations by making appointments for life or good behavior, like the justices of the Supreme Court. I should also like to see the membership composed entirely of professional men--a psychiatrist, a jurist, a physician, a sociologist, an educator, perhaps a barrister, and a criminologist. Such a body, thus assured of independence and expertise, could be entrusted with decisions of life and death and with lesser matters of clemency and commutation with more freedom from extraneous pressures than an elected governor has.330

The patrician, sexually discriminatory membership suggested by DiSalle requires obvious modification. The commission should be democratized, particularly to ensure that minorities, who traditionally have made up a large percentage of those incarcerated or sentenced to death, are involved in the clemency decision. Representatives of the citizenry, victims of crime, and perhaps philosophers or clerics should also be part of the board.331

Moreover, the process for selecting members of the commission would need to be insulated from politicization as much as possible. In many states, the governor directly or indirectly appoints members of state boards. The clemency commission could instead be selected by the executive with the advice and consent of the legislature or the judiciary to ensure that the appointees do not reflect a single political persuasion and are not beholden to a particular individual or faction.

C. Proposed Clemency Standards

Once a clemency commission has been created, it should be guided in its decisions by explicit, internally consistent standards governing the mitigation of punishment. The modified retributive philosophy described above offers an attractive starting point for developing these standards because it represents one of the principal theories underlying our system of punishment. Beginning with the premise that "justice" means that each person should be rendered her due, the commission would use the clemency power to ensure that the criminal justice system serves that end.332 Clemency would be appropriate in any circumstances that called into question the fairness of the punishment imposed. The applicant would be entitled to clemency as a matter of right once the commission found that any of the justice-enhancing standards set forth below had been satisfied.333

1. Substantial Doubt of Guilt.--When there is substantial doubt of guilt, the offender's punishment is not deserved and should be remitted. This finding may be the result of new evidence, information suppressed at trial or withheld by the prosecution, incompetence of defense counsel, coercion of a guilty plea, or any other reason that seriously undermines the commission's confidence in the integrity of the judicial determination.

Because of the rigidity of various standards and rules that reviewing courts use in measuring legal error, the judiciary cannot always achieve a just result. For example, in 1987 the Ohio Supreme Court affirmed the death sentence in a case in which even the majority conceded that there was substantial doubt as to the offender's guilt.334 Although the court concluded that "reasonable minds could reach different conclusions as to whether each material element was proved beyond a reasonable doubt," it affirmed the death sentence by a four-to-three vote because the evidence was sufficient to withstand a motion for acquittal.335 The court was in effect saying that it was unsure about guilt, but that under its traditional rules of review, it had no choice but to affirm. In such a case, when the question of guilt is so problematic and rules of appellate review so rigid, only the extrajudicial remedy of clemency can ensure that the offender is not punished unfairly. One option that could be given to the commission in such cases is that of recommending that the offender be given a new trial.336

2. Diminished Mental Capacity, Retardation, Intoxication, or Minority.--@Retributive philosophy presumes that punishment is defensible insofar as it is related to blameworthiness.337 Thus, to the extent that the offender is less worthy of blame than another offender because of brain damage, retardation, intoxication, or age,338 punishment should be correspondingly diminished.339 Of course, the insanity defense takes this principle into account, but generally applies it in an "all or nothing" fashion. Clemency permits refinement of this notion: an individual who is sane enough to stand trial may still be so impaired that it would be unjust to impose the full measure of punishment on her.

Former California governor Pat Brown used this concept as a cornerstone of his clemency philosophy and routinely ordered electroencephalogram (EEG) tests in all capital cases.340 In contrast, Louisiana governor Roemer recently denied clemency in a capital case in which it would be merited under retributive principles. Dalton Prejean was convicted of killing a police officer at the age of 17, shortly after having been released from reform school for an earlier killing. A psychiatrist who examined Prejean concluded that severe abuse in childhood and resulting brain damage prevented Prejean from thinking about alternatives and moderating his behavior at the time of the killing.341 Assuming the psychiatric testimony was credible, the death penalty recently inflicted on Prejean342 was inappropriate because his mental state at the time of the killing rendered him less blameworthy than one who killed with complete appreciation of her actions.

The clemency commission would be justified in remitting punishment in a case of diminished blameworthiness, but would also be required to consider other interests in deciding how the original sentence should be modified, including protection of society and other utilitarian interests. Thus, while death or incarceration for life would not be retributively defensible punishments, detention and psychiatric treatment until the individual no longer posed a danger to herself or others could be appropriate.

3. Disparate Sentencing: Proportionality.--Clemency is also appropriate to remit punishment that is disproportionate when compared with that meted out to (1) other participants in the same crime, or (2) those convicted of similar crimes within the jurisdiction. It is fundamentally unfair to sentence one offender to death, while another offender judged to be equally culpable receives a lesser sentence.343 This corollary of a retributive approach to punishment has also been a fundamental part of the clemency philosophy of various governors.344 As former Ohio governor Thomas Herbert said when commuting the capital sentence of a convicted murderer whose accomplice had not received the death penalty, "`I am impelled to commute the sentence of Ames [to the same sentence received by his cohort] not for any sympathy for Ames, but in order that it may not be said that Ohio failed in comparative justice.'"345 Similarly, President Carter commuted the twenty-year sentence of Watergate conspirator G. Gordon Liddy after four years and three months because Liddy had served much more time than more significant figures like John Mitchell.346 According to Carter's associate White House counsel, "`[i]t was a clear case of unfair disparity.'"347

Lack of sentencing uniformity within the jurisdiction for similar crimes is also a retributively justifiable basis for clemency, though the Supreme Court has held that such proportionality in imposing the death penalty is not constitutionally mandated.348 If an individual's sentence is substantially harsher because of the particular area in which she was tried, punishment is not related to deserts, but to an accident of venue. Likewise, if a sentence greatly exceeds that which is normally imposed for similar crimes, so that it appears that the sentence is an aberration unrelated to deserts, retributive principles require that the punishment be remitted. Former California governor Brown considered sentencing disparity to be one of the most important factors in making clemency decisions, although his method of implementing equality was informal, relying on the intuitions of his clemency secretary rather than on any systematic comparison of sentences.349

Practical difficulties inhere in making uniform sentencing determinations. In order to implement this standard, the commission would need access to detailed data concerning the sentences handed down throughout the jurisdiction for virtually all offenses. Compilation of this sort of "universe" of sentences would be difficult, because it would inevitably entail gathering detailed information concerning the reasons underlying individual sentences. A further complication is that persons who committed similar crimes in different venues might be indicted on different charges, depending on the philosophy of the respective prosecuting attorney. However, inasmuch as clemency decisions based on disproportionate sentencing are presently being made on an ad hoc basis by pardoning authorities, it would be better to attempt to make such comparisons more systematically.

4. Disparate Sentencing: Special Circumstances.--A related reason for granting clemency based on disparate sentencing exists when a sentence is disproportionate in light of the offender's age or terminal illness.350 A seventeen-year-old murderer and a ninety-year-old murderer are not affected in the same way by the passage of time: life imprisonment would result in differing degrees of disadvantage for each.351 Similarly, a five-year sentence given to an offender in the waning stages of a battle with AIDS amounts to life imprisonment.352 General sentencing standards are not refined enough to ensure that prisoners in such special circumstances receive comparable degrees of punishment. Clemency can enhance justice by reducing sentences that, although fair in most instances, punish a particular offender disproportionately.

5. Sentencing That is Unrelated to Deserts.--Clemency also is properly granted where there is a substantial likelihood that the sentence was based on factors such as the race or gender353 of the criminal or victim, rather than on the punishment the individual deserved. A recent study of the imposition of the death penalty under post-Furman v. Georgia354 statutes in eight states concludes that capital punishment is undoubtedly imposed in a racially discriminatory fashion: "[T]he discrimination is based on the race of the victim, and it is a remarkably stable and consistent phenomenon."355 Allowing the discriminatory punishment imposed to stand would be unjust, because the penalty is not related to what the individual deserves, but to the immutable vagaries of race. Recently, Ohio governor Richard Celeste commuted to life imprisonment the death sentences of eight convicted killers, six of whom are black; he based his decision in part on the race of the offenders, noting the disproportionate number of blacks on death row.356

A sentencing court, which cannot compare the manner in which the death penalty is administered throughout the jurisdiction, has no means of assessing whether discrimination is occurring. Moreover, in jurisdictions in which judges are elected, the judiciary, like governors, are under strong political pressure to mete out the death penalty without regard to such niceties as nondiscrimination.357 A politically independent clemency commission with access to adequate information could commute sentences that are based on factors other than the crime committed and thereby assure that punishment is imposed fairly.

6. Crimes Committed out of Necessity, Coercion, or Adherence to Moral Principles.--Remission of punishment for certain crimes committed out of physical, psychological,358 or moral necessity can be justified retributively. One who is coerced into committing a crime is not as deserving of punishment as one who voluntarily chose to violate the law, and the punishment imposed on each should reflect this difference.359 By the same reasoning, a mother who steals to feed and clothe her children is not as morally blameworthy as one who steals to buy a new television. Yet, notwithstanding the discretion often vested in sentencing judges to consider these justifications,360 both thieves may receive the same sentence. The new federal determinative sentencing guidelines, although intended to serve the laudable end of reducing patent disparities in sentencing, also increase the likelihood that distinctions in blameworthiness will be ignored in sentencing.361 Clemency provides a means of refining the punishment imposed to reflect distinctions in moral blameworthiness, even though such distinctions will not always be easy to draw.

Particularly difficult issues arise when the moral principles acted upon by the accused have no widespread acceptance. The mother who steals to feed and clothe her children presents a relatively easy clemency determination because of our consensus as a society that there is no higher moral responsibility for parents than caring for their children. Clemency is appropriate in the case of the mother because few would consider her actions to be morally wrong, though she has clearly committed a crime. However, when the offender commits a crime out of moral beliefs on matters about which we have no societal unanimity, it becomes much harder to assess moral blameworthiness. What of the individual who commits treason out of a profound moral belief that her country is wrong? To consider a more contemporary problem, how should a clemency board treat the case of an anti-abortion activist who breaks the law by bombing an abortion clinic? Assuming the activist was motivated by an unwavering, sincerely held religious belief that she was saving innocent lives by so acting, it is difficult to say that she is as morally blameworthy as one who destroys a bank in order to rob it. Yet, those who view abortion as a woman's constitutionally guaranteed right would no doubt view the action as morally reprehensible and argue that mitigation of punishment is inappropriate on retributive grounds.

Kathleen Moore and other Kantian philosophers draw a distinction between conscientious, moral actions (e.g., the mother stealing to feed her children) and conscientious, immoral actions (e.g., someone who murders out of a sincerely held belief that God required such action).362 Under this dichotomy, clemency is always appropriate in the former cases because the person has done nothing wrong. In the latter cases, retributivists would engage in a delicate balancing act, weighing the value of behaving conscientiously against the offender's blameworthiness in "accepting false views about what is right, which may be the result of self-righteousness or zealousness or carelessness or even perversity."363 This refinement of the blameworthiness inquiry is helpful insofar as it demonstrates that acting on moral beliefs need not be an unqualified defense against societal punishment. However, such a scheme does not resolve the underlying problem raised in the abortion clinic bombing example: what is or is not a moral action? In the end, these determinations may well turn on whether the clemency decision maker would have resolved the moral problem in the same way as the offender or can understand how a reasonable person could have conscientiously acted in the same way as the offender. Consequently, it is essential that the clemency commission be composed of individuals who have diverse philosophical, social, and professional backgrounds.364

7. Crimes in Which the Offender Has Suffered Enough.--In a few cases, punishment may be unnecessary because the punishment that the offender deserves has already been inflicted as an unintended result of the crime. Thus, if an offender who assaulted someone was blinded as a result of injuries received in the fight, and the damage can be said to fairly approximate what she deserved for her crime, further punishment would be excessive and should be remitted. A more contemporary example is that given by philosopher Alwynne Smart of the hit-and-run driver who discovers that the person she has recklessly killed was her beloved only child.365 This principle of autrefois puni, which means "already punished," has long been part of criminal codes,366 and is consistent with a retributive approach to clemency. In these cases, "poetic justice" can be said to make further punishment by society superfluous.

A related reason for granting clemency exists where legal disabilities that are no longer deserved are imposed on the criminal. In many states, persons with criminal convictions are deprived of civil rights and other privileges, and may also have difficulty securing employment. Assuming that the punishment originally imposed was what the offender deserved, the conviction's ongoing effects constitute excessive punishment and should be remitted.367

D. Proposed Clemency Procedures

Implementing the standards set forth above would require that the clemency commission adopt certain procedures, and might trigger procedural due process protections. In the past, because clemency has been considered a matter of grace rather than of right, no procedure has been thought to be due to applicants.368 If, however, applicants are entitled to clemency as a matter of right under certain circumstances, then some due process protections would inhere in the clemency process. At a minimum, applicants should be given the procedural protections that the Supreme Court has said are necessary in the analogous release procedure of the parole process:369 an opportunity to be heard,370 notice of the hearing adequate to permit the applicant to secure letters,371 allowance of the testimony or statements of witnesses at the hearing,372 and, if clemency is denied, some indication of the reasons for the commission's decision.373

In addition to these minimal constitutionally mandated procedures, additional safeguards should be considered, especially in cases where the sentence is death. Counsel should be appointed to represent the applicants before the commission in order to facilitate informed decisions in capital cases, where the stakes are exceedingly high both for the offender and the state.374

It would also be necessary for the clemency commission to investigate clemency applications independently, to ensure that "as the state's last participant in the . . . punishment process, [it] will act deliberatively and with care."375 The importance of the commission's gathering information to protect the accuracy of the clemency process cannot be overestimated. The investigation should include: (1) interviewing the applicant, her family and friends, the sentencing judge, and the prosecuting attorney, as well as the victim and her family; (2) requiring psychological testing of the applicant and examination of her medical records; (3) reviewing statements from friends and members of the community, as well as any other information that would shed light on the applicant's character; (4) examining all information pertaining to the applicant's crime and trial, including evidence that was excluded, newspaper coverage, and the sentences of codefendants; and (5) reviewing judicial opinions pertaining to the applicant's case, particularly dissenting opinions. The federal clemency procedure already has an effective information-gathering process in place, with the FBI conducting investigations of clemency applicants.376 At the state level, however, only Georgia and Arkansas appear to have meaningful investigatory mechanisms.377

The clemency commission would also have to create a mechanism through which it could review the proportionality of various sentences throughout the jurisdiction, in order to evaluate disparate sentences and detect patterns of discrimination in sentencing.

Clearly, creating the clemency procedures described above would entail substantial costs. However, incarcerating prisoners is costly as well, running between twelve thousand dollars and thirty-five thousand dollars per prisoner, per year, depending on where the prisoner is imprisoned and the degree of security.378 Assuming a cost of twenty thousand dollars per year to keep offenders in prison, the release of 130 deserving prisoners one year would save enough money to fund the entire Georgia clemency budget for 1987.379 Finding 130 prisoners worthy of release is hardly unrealistic, as evidenced by the experience of the Georgia Board of Pardons and Paroles. In 1973, the Board hired a group of university professors to find prisoners serving lengthy sentences under obsolete laws. The investigators found "dozens" of these prisoners, and the Board commuted their sentences.380

Even if implementing comprehensive clemency procedures would ultimately cost the government more than maintaining the status quo, such expenditures are a necessary incident to our system of justice. Cut-rate clemency is no bargain, for without a meaningful clemency power, our government would, as the Supreme Court has remarked, "be most imperfect and deficient in its political morality."381

E. Justice-Neutral Exercise of the Clemency Power by the Executive: What Remains?

The Constitution's Framers had two goals in mind when they vested the pardoning power in the President:382 first, they sought to create a mechanism to oversee the judiciary by making justice-enhancing exceptions in favor of unfortunate guilt; second, they clearly intended that the clemency power be used to further the public welfare in ways that are entirely unrelated to justice, such as by preserving order in seasons of insurrection or by procuring the testimony of accomplices of criminals. The clemency procedure and standards proposed above serve the justice-enhancing function envisioned by the Framers. However, what is to become of utilitarian uses of clemency that advance the public welfare? Even if, as seems likely, Richard Nixon was not deserving of clemency based on principles of justice, might not his pardon have been proper purely as a matter of public policy, as President Ford argued?

Strict retributive philosophers contend that clemency should never be used solely to promote the public welfare at the expense of deserts. Retributive philosophers believe that such a "forward-looking, utilitarian" approach denigrates the individual's right to be treated as she deserves and works "the comparative injustice of treating similar offenders in different ways."383 They argue that the executive should shun all justice-neutral acts of clemency unless some retributive end is also served by the remission of punishment. Hamilton's suggestion in The Federalist No. 74 that the President may be able to make a well-timed offer of pardon to quell an insurrection is, to the strict retributivist, as unjust as outright abuses of the clemency power like granting pardons to promote the private welfare of the pardoner. Retributivists believe that neither should be granted unless the executive determines that the pardon will also serve a justice-enhancing function. Presumably, the executive could likewise pardon to promote her own private welfare, as long as it would also render the offender her just deserts.

In contrast, I do not believe that it is necessary to find a justice-enhancing element in every clemency decision. Notwithstanding the philosophical inconsistency that results, some justice-neutral acts of clemency are desirable elements of our justice system. Yet, it is essential to distinguish between justice-neutral exercises of the clemency power that truly further the public welfare, and those that are given for personal, political, or idiosyncratic reasons. At the close of the Civil War, even mercenaries who had fought for the Confederacy strictly to line their pockets and who were thus clearly deserving of punishment should have been pardoned if the President determined that such clemency was necessary to reunite a nation on the verge of disintegration. Although such pardons cannot be justified retributively, the Framers evidently were convinced that the benefit to the nation as a whole outweighed the potential harm which would result from these pragmatic exceptions to principles of justice. I consider this judgment a reasonable one and contend that the chief executive, who has been elected by the people to make political policy decisions, is the proper person to exercise the power.

However, as the pardoning power's history reveals,384 these justice-neutral acts of clemency in the name of the public good present the greatest potential for abuse. A President or governor might well couch wholly self-serving acts of clemency in the guise of acting in the public welfare. Consequently, I propose that the executive's power to grant clemency for justice-neutral reasons be limited in the following ways, largely to assure accountability in the exercise of the power.

First, the executive's power to make justice-neutral clemency decisions should be exercised only after the clemency commission has considered whether there are justice-enhancing reasons for granting clemency. If the offender does not deserve punishment, punishment is properly remitted for that reason, by the body best suited to making these determinations--the clemency commission. Such a grant of clemency revives our system of justice, while a justice-neutral pardon tends to undermine it. If the commission determines that punishment is indeed deserved, but the executive still believes there are sound public policy reasons for granting clemency, she may do so with the clear understanding that she is making a political determination which the people may either endorse or criticize.

Second, in order for the executive to be accountable for her decision, she should set forth her reasons for granting clemency. Since justice-enhancing reasons have already been considered and rejected by the clemency commission, she will be forced to confront the policy reasons for her decision and share them with her constituents. Inasmuch as the public is continuously evaluating the political determinations of the executive, it is better suited to critiquing these sorts of decisions rather than quasi-judicial determinations about guilt, innocence, and blameworthiness.

Finally, the executive should report all acts of clemency to the legislature immediately, and permit a super-majority to override acts of executive clemency.385 Since the portion of the clemency power that remains to be exercised by the executive is political in nature, a political check, short of impeachment, on the most controversial exercises of the power is appropriate. William Duker has argued cogently in support of such a limit on the federal executive clemency power:

The proposal [of Senator Mondale that no presidential pardon be effective if disapproved by two-thirds of the members of each house within 180 days of the proposal's issuance] would provide a sufficient check on the prerogative of the executive which need be invoked only in controversial cases. "The dilatory process of convening the legislature, or one of its branches . . . [and] letting slip the golden opportunity" would be avoided. The effect of the pardon need not await the sanction of Congress, for the Congress is given the burden of going forward with the process.386

A legislative veto, made suitably burdensome to wield, would presumably only be exercised in cases in which there were widespread, lingering doubts about the propriety of the executive's action.

IV. Conclusion

The institution of clemency represents our oldest existing procedure for the remission of punishment. The universality of clemency and the failure of this hoary relic to become obsolete in any culture in which it has existed, notwithstanding a long history of abuses, testifies to its continuing usefulness. As Beccaria suggested, so long as systems of justice are imperfect, there will be a need for clemency.387

Insofar as we have never achieved a perfect system of justice, clemency occupies an important place in our judicial scheme. However, because clemency has remained largely unchanged from the days when it was vested in the monarch, its exercise has been fraught with problems.

The purpose of this Article has been to suggest ways in which the clemency power can be refined to operate as a principled means of correcting some of the flaws extant in our penal system. This, I believe, can best be accomplished by establishing an independent commission with the requisite expertise which is directed to focus on justice-enhancing reasons for remitting punishment. However, the executive should still retain and exercise that portion of the clemency power that is suited to her skills, as recognized by the Framers of the Constitution: the granting of clemency for reasons related to the public welfare. By delineating these two largely unrelated aspects of the clemency power, and assuring that they will be exercised by the appropriate bodies, the clemency power can alleviate many of the shortcomings of our judicial system. Then, perhaps, the quality of justice will be strained no longer, but will filter into our judicial system, as ubiquitous as the rain. The quality of mercy is not strained. It droppeth as the gentle rain from heaven Upon the place beneath. It is twice blest: It blesseth him that gives, and him that takes. `Tis mightiest in the mightiest. It becomes The throned monarch better than his crown. His sceptre shows the force of temporal power, The attribute to awe and majesty, Wherein doth sit the dread and fear of kings; But mercy is above this sceptred sway. It is enthroned in the hearts of kings; It is an attribute to God himself, And earthly power doth then show likest God's When mercy seasons justice. --The Merchant of Venice, Act IV, scene i.388

Footnotes

* Assistant Professor of Law, Capital University Law School. B.A. 1978, J.D. 1983, University of Toledo. I would like to thank Emily Dargatz for her invaluable contributions to the researching and writing of this Article, and for becoming as interested in the clemency power as I am. I am also grateful to Carl Beckman, Arthur Collier, Jeff Erickson, Mike Gerhardt, Katherine Korda, Susan Looper-Friedman, Thomas Tucker, and Dean Rod Smith for their insightful comments on an earlier draft of this Article. Finally, I would like to thank my wife, Janet Lewis, for all of her support and encouragement. Funding for the writing of this Article was provided by a Faculty Research Stipend from Capital University.

1. W. Shakespeare, The Tragedy of Hamlet, Prince of Denmark, Act II, scene ii, in The Complete Works of William Shakespeare 733, 751 (W. Wright ed. 1936) [hereinafter Complete Works].

2. E. Brown & D. Adler, Public Justice, Private Mercy 163 (1989) (quoting Pat Brown).

3. 4 W. Blackstone, Commentaries *398.

4. See infra text accompanying notes 98-102 (describing Charles II's pardon of his Lord High Treasurer Thomas Osborne).

5. See K. Moore, Pardons: Justice, Mercy, and the Public Interest 18-19 (1989).

6. See id. at 22.

7. Compare Schorr, Pardon North, Poindexter and the Rest, N.Y. Times, Apr. 23, 1990, at A19, col. 2 (asserting that President Bush, "out of concern for the country rather than the culprits," should pardon all those associated with the Iran-Contra conspiracy) with Broder, The Leaders We Deserve, Wash. Post, Apr. 29, 1990, at D7, col. 1 (arguing that "crimes of such magnitude call out for severe punishment" and moral sanctions from society, but asserting that Americans lack "the moral courage to shun those who betray us").

8. Montesquieu, Kant, Hegel, and Bentham, among others, have wrestled with the complicated questions raised by remission of punishment through acts of clemency. See generally K. Moore, supra note 5, at 15-54 (discussing the history of philosophical thought regarding the clemency power).

9. Divine mercy and the command for humans to emulate it are recurrent biblical themes. See, e.g., Psalms 103:8-10 (King James) ("The Lord is merciful and gracious, slow to anger, and plenteous in mercy. He will not always chide; nor will he keep his anger forever. He hath not dealt with us after our sins, nor rewarded us according to our iniquities."); Ephesians 4:32 (King James) ("And be ye kind to one another, tenderhearted, forgiving one another, even as God, for Christ's sake, hath forgiven you."). Perhaps the most famous call for human mercy is in the Sermon on the Mount, which uses divine mercy as the exemplar: "Blessed are the merciful: for they shall obtain mercy." Matthew 5:7 (King James).

10. See Luke 23:14-25 (recounting Pontius Pilate's capitulation to the mob demanding Jesus's crucifixion and his admission that he was condemning an innocent man).

11. See John 15:25-27; 16:7 (showing that Jesus's death at the hands of those who hated him was a necessary precondition of God's redemption of mankind).

12. W. Shakespeare, The Merchant of Venice, Act IV, scene i, in Complete Works, supra note 1, at 467, 499.

13. Cf. C. Beccaria, An Essay on Crimes and Punishments 158-59 (W.C. Little & Co. new ed. 1872) (asserting that the power of pardon is not necessary under a perfect system of justice where punishment is mild and "proceedings in criminal cases regular and expeditious").

14. See I. Kant, The Metaphysical Elements of Justice 108 (J. Ladd trans. 1985) ("By exercising [clemency], [the sovereign] can demonstrate the splendor of his majesty and yet thereby wreak injustice to a high degree. With respect to a crime of one subject against another, he absolutely cannot exercise his right, for in such cases exemption from punishment constitutes the greatest injustice towards his subjects.").

15. William James identified such a tendency:

Exactly that callousness to abstract justice is the sinister feature and, to me, as well as to you, the incomprehensible feature of our U.S. civilization. When the ordinary American hears of cases of injustice he begins to pooh-pooh and minimize them and tone down the thing, and breed excuses from his general fund of optimism and respect for expediency.

Letter from William James to William Dean Howells (1906), quoted in G. Vidal, At Home: Essays 1982-1988, at viii (1988).

16. See Sebba, Clemency in Perspective, in Criminology in Perspective: Essays in Honor of Israel Drapkin 221, 225-28 (S. Landau & L. Sebba eds. 1977) (reporting criticisms of the clemency power in various societies throughout history).

17. See Smith & Johnson, Presidential Pardons and Accountability in the Executive Branch, 35 Wayne L. Rev. 1113, 1115 (1989).

18. See Duker, The President's Power to Pardon: A Constitutional History, 18 Wm. & Mary L. Rev. 475, 525 (1977).

19. Pardon of Richard M. Nixon and Related Matters: Hearings Before the Subcomm. on Criminal Justice of the House Comm. on the Judiciary, 93d Cong., 2d Sess. 154 (1974) (statement of Gerald Ford, President of the United States).

20. See Barnett, The Grounds of Pardon, 17 J. Crim. L. & Criminology 490, 507 n.91 (1926) (discussing former South Carolina governor Coleman Blease's pardon of William Mills based on a campaign promise in 1912); 20/20: Why Not Wilbert Rideau? (ABC television broadcast, Apr. 14, 1989) (discussing former Louisiana governor Edwin Edwards's refusal to pardon rehabilitated murderer Wilbert Rideau because of a campaign promise that Edwards made to a woman who had been shot by Rideau).

21. See 3 U.S. Dep't of Justice, The Attorney General's Survey of Release Procedures 150-53 (1939) [hereinafter Survey] (describing the impeachment of Oklahoma Governor J.C. Walton for selling pardons).

22. Lex talionis is defined as "[t]he law of retaliation; which requires the infliction upon a wrongdoer of the same injury which he has caused to another." Black's Law Dictionary 822 (5th ed. 1979). The notion of lex talionis is strongly evident in the Old Testament:

And he that killeth any man shall surely be put to death. And he that killeth a beast shall make it good; beast for beast. And if a man cause a blemish in his neighbour; as he hath done, so shall it be done to him; Breach for breach, eye for eye, tooth for tooth: as he hath caused a blemish in a man, so shall it be done to him again. And he that killeth a beast, he shall restore it: and he that killeth a man, he shall be put to death. Ye shall have one manner of law, as well for the stranger, as for one of your own country: for I am the LORD your God.

Leviticus 24:17-22 (King James).

23. It is not unusual that in evaluating clemency, the executive may find that what is "just" directly conflicts with what is "politically expedient." The case of a Washington D.C. teenager recently sentenced under federal drug-trafficking laws presents such a potential conflict for President Bush.

In September of 1989, Keith Jackson, a 19-year-old high school student, was arrested after selling cocaine in Lafayette Square, across the street from the White House. Wash. Post, Nov. 1, 1990, at B11, col. 5. Jackson had been lured to the Square by DEA agents, allegedly to sell the crack that President Bush used as a prop in a televised speech kicking off the Administration's war on drugs. Id. Jackson was convicted of three counts of distributing drugs in locations other than the Square, including one count of selling crack within 1000 feet of a school, a crime that carries a mandatory sentence of 10 years, without possibility of parole.

United States District Judge Stanley Sporkin sentenced Jackson, who had no prior record, to the 10-year term, but urged Jackson to seek commutation from President Bush because he believed that the sentence was overly harsh. Judge Sporkin told Jackson, "[Bush] used you, in the sense of making a big drug speech. But he's a decent man, a man of great compassion. Maybe he can find a way to reduce at least some of that sentence." Id.

If Jackson does request that his sentence be commuted, he will be seeking justice from President Bush, the man who "used" Jackson to make a political point. Bush will be forced to choose between considerations of political expediency and justice.

24. See Howell, Deaths and a Dearth of Pardons, Newsday (New York), Dec. 25, 1988, at 9 (reporting that prisoners' advocates feel that "[t]he need for a more liberal use of executive pardon is particularly great now").

25. See Nat'l Center for State Courts, Clemency: Legal Authority, Procedure, and Structure passim (1977).

26. See Sebba, Criminology: The Pardoning Power--A World Survey, 68 J. Crim. L. & Criminology 83, 85-110 table (1977) (Tabular Comparison of Provisions Relating to Pardoning Powers) (comparing the laws authorizing pardons in approximately 90 countries).

27. Clemency is defined as "[m]ildness or gentleness of temper, as shown in the exercise of authority or power; mercy, leniency." 3 Oxford English Dictionary 309 (2d ed. 1989).

28. See Nat'l Governors' Ass'n Center for Policy Research, Guide to Executive Clemency Among the American States 164 (1988) [hereinafter Guide to State Clemency]; id. at 165 table 2 (Types of Clemency Available in Each State According to Survey Results). In addition, some states recognize local varieties of clemency. Florida, for example, offers special forms of clemency such as the restoration of Florida civil rights, restoration of resident rights, and restoration of the right to own and possess a firearm. Minnesota offers "pardon extraordinaire" and the expungement of records as forms of clemency. See id. at 164.

29. See, e.g., U.S. Const. art. II, 2 (giving the President the power to grant pardons and reprieves, which the United States Supreme Court has construed to include all forms of clemency); E. Coke, The Third Part of the Institutes of the Laws of England 233 (4th ed. London 1669) (noting that "[a] pardon is a work of mercy, whereby the king either before attainder, sentence, or conviction, or after, forgiveth any crime, offence, punishment, execution, right, title, debt or duty, temporal or ecclesiastical").

30. E. Coke, supra note 29, at 233.

31. Throughout this Article, I will use feminine pronouns to refer to people in a generic sense or in hypothetical situations, for two reasons. First, I believe that because criminals, and particularly violent offenders, are commonly stereotyped as male, use of the female pronoun in this context will prompt readers to think about the offender as an individual, and thereby to focus more clearly on what that person deserves as the result of her actions. Second, because a gender-neutral pronoun that I find satisfactory has not yet found its way into common, or legal, parlance, I have decided that it is less distracting to alternate between masculine and feminine pronouns from essay to essay than from sentence to sentence within a particular piece.

32. See Ex parte Garland, 71 U.S. (4 Wall.) 333, 380 (1866); see also Schick v. Reed, 419 U.S. 256, 273 n.8 (1974) (Marshall, J., dissenting) (distinguishing between the legal effects of pardons and commutations). But see United States v. Noonan, 906 F.2d 952, 959-60 (3d Cir. 1990) (relying on British and British Commonwealth court decisions to hold that a presidential pardon does not eliminate a criminal conviction or create a fiction that the conviction did not occur).

33. Garland, 71 U.S. (4 Wall.) at 380.

34. See generally Cozart, Clemency Under the Federal System, 23 Fed. Probation 3 (1959) (discussing the nature and types of clemency and clemency procedures in the federal system).

35. 1 Oxford English Dictionary 406 (2d ed. 1989).

36. See Burdick v. United States, 236 U.S. 79, 95 (1915) (explaining the difference between a pardon and amnesty).

37. See R. Goldfarb & L. Singer, After Conviction 350-51 (1973) (observing that amnesty is most often granted for political offenses such as treason, rebellion, or civil disorder); see also Migliore, Amnesty: An Historical Justification for its Continuing Viability, 12 J. Fam. L. 63 (1973) (discussing the history and future of amnesty).

38. See R. Goldfarb & L. Singer, supra note 37, at 351.

39. See K. Moore, supra note 5, at 81-82 (describing the rationale and effect of President Carter's grant of amnesty to those who evaded the Vietnam draft); see also Exec. Order No. 11967, 3 C.F.R. 91 (1978) (directing the Attorney General to dismiss with prejudice indictments against draft evaders and to terminate investigations of draft evasion).



40. See Knote v. United States, 95 U.S. 149, 152-53 (1877).

It is sometimes said that [amnesty] operates as an extinction of the offence of which it is the object, causing it to be forgotten, so far as the public interests are concerned, whilst [pardon] only operates to remove the penalties of the offence. This distinction is not, however, recognized in our law.

Id. at 152.

41. See Schick v. Reed, 419 U.S. 256, 273 n.8 (1974) (Marshall, J., dissenting) (comparing commutations and pardons as legal forms of forgiveness). Occasionally, the term "remission" is used to refer to reduction of a sentence; however, this term is generally used in the United States to denote the reduction of a fine. See Boudin, The Presidential Pardons of James R. Hoffa and Richard M. Nixon: Have the Limitations on the Pardon Power Been Exceeded?, 48 U. Colo. L. Rev. 1, 3 n.14 (1976). In Great Britain, however, "remission" is used as a synonym for "pardon." Gane, The Effect of a Pardon in Scots Law, 25 Jud. Rev. 18, 18 (1980).

42. See Schick, 419 U.S. at 273 n.8 (Marshall, J., dissenting); see also W. Humbert, The Pardoning Power of the President 27 (1941) (noting that commutation does not restore the privilege to vote or the competency to testify).

43. See R. Goldfarb & L. Singer, supra note 37, at 344.

44. The Laura, 114 U.S. 411, 413 (1885); Osborn v. United States, 91 U.S. 474, 479 (1875); see also Illinois Cent. R.R. v. Bosworth, 133 U.S. 92, 103 (1890) (noting that a pardon after conviction removes the "penalties and disabilities" imposed).

45. Guide to State Clemency, supra note 28, at 163-64 & table 2 (Types of Clemency Available in Each State According to Survey Results).

46. See Ex parte United States, 242 U.S 27, 43-44 (1916); 4 W. Blackstone, supra note 3, at *394.

47. 4 W. Blackstone, supra note 3, at *394. Blackstone asserted that granting such a reprieve to a pregnant mother was "dictated by the law of nature." Id. at *395. Thus, he condemned the cruelty exercised on the Isle of Gurnsey of "burning a woman big with child: and when, through the violence of the flames, the infant sprang forth at the stake, and was preserved by the bystanders, after some deliberation of the priests who assisted at the sacrifice, they cast it again into the fire as a young heretic." Id. (footnote omitted).

48. See generally R. Goldfarb & L. Singer, supra note 37, at 257-315, 319-20 (describing the origin of parole, summarizing the procedures used in administering the parole system, and contrasting parole with pardon).

49. Morrissey v. Brewer, 408 U.S. 471, 477 (1972). The Supreme Court has explicitly distinguished parole from "an ad hoc exercise of clemency." Id.

50. See id. at 478 (describing the various restrictions that accompany parole, such as forbidding the use of liquor and prohibiting association with certain categories of undesirable persons); W. Humbert, supra note 42, at 27-28 (comparing parole and pardon).

51. See generally Schick v. Reed, 419 U.S. 256, 260-61 (1974) (tracing the President's pardon power to the English sovereign); 6 W. Holdsworth, History of English Law 203 (1938) (describing the evolution of the pardoning power of the king).

52. For instance, pardons have been granted to condemned criminals where they have held fast to the bridle, saddle, or other part of the horse carrying certain clerics or royalty into a particular city for the first time; where a woman has run naked three times through the town or around the prison where the prisoner awaited execution; and where the prisoner has earned clemency by executing her fellow prisoners. See Survey, supra note 21, at 17, 20, 295.

53. See Weihofen, Pardon: An Extraordinary Remedy, 12 Rocky Mtn. L. Rev. 112, 114 (1940).

54. Indeed, philosophers and theologians disagree considerably about the proper definition of justice. See, e.g., K. Lebacqz, Six Theories of Justice: Perspectives from Philosophical and Theological Ethics 118-20 (1986). Dr. Lebacqz's book provides a succinct introduction to several of the most commonly advanced theories of justice.

55. See generally 1 E. Brunner, Justice and Social Order 17 (M. Hottinger trans. 1945) ("From time immemorial the principle of justice has been defined as the suum cuique--the rendering to each man of his due."); A. Heller, Beyond Justice 156-79 (1987) (arguing that retribution--the rendering to each offender her due--is the sole principle of punishment that can be called a principle of justice); J. Rawls, A Theory of Justice 11 (1971) (establishing a concept of justice as fairness that is based on "the principles that free and rational persons concerned to further their own interests would accept in an initial position of equality as defining the fundamental terms of their association").

56. K. Moore, supra note 5, at 94. But see Bedau, Retribution and the Theory of Punishment, 75 J. Phil. 601, 608-11 (1978) (discussing various meanings of the term "desert" and arguing that it is impossible to ascertain what punishment is deserved).

57. See K. Moore, supra note 5, at 89-178.

58. See id. at 94-95. For a more sophisticated discussion of the philosophical justifications underlying a desert-based theory of punishment, see G. Sher, Desert 69-90 (1987). Sher argues that "fairness" provides the moral foundation for punishment based on deserts; because a wrongdoer "has had an extra measure of freedom from moral restraints, his punishment restores a fair overall balance of benefits and burdens." Id. at 91.

59. See, e.g., Bedau, supra note 56; Gale, Retribution, Punishment, and Death, 18 U.C. Davis L. Rev. 973, 1004-05 (1985).

60. Mary Ellen Gale notes the difficulties inherent in arriving at a concept of proportional punishment:

[E]ven if we manage to construct acceptable scales of moral gravity for offenses, offenders, and punishments, there is no principle that correlates one with the others. There is no fixed starting point to match offenses and offenders with appropriate punishments, no fixed interval between offenses comparable to that between punishments of the same type (moral gravity, unlike time and money, cannot be measured in standard units), and no fixed standard for converting punishments of one type into punishments of another.

Gale, supra note 59, at 1015 (footnote omitted). Gale also points out that even retributive proponents have not been successful in identifying criteria for determining what punishments are deserved. See id. at 1013.

61. Id. at 1012 (arguing that the "progressive unintelligibility" of such notions as a rape for a rape, or a tax evasion for a tax evasion "destroys any general theory of punishment that seeks to incorporate equivalent punishments").

62. The truth of this proposition is illustrated by considering the converse situation: punishment imposed without regard to whether it is deserved. Consider a system that imposed preventive sanctions, not in regard to specific crimes committed, but to crimes predicted, for example, interring potential political enemies and hostile ethnic groups in order to deter the populace from obstructing the aims of the government. It is clear that such a use of punishment, divorced from principles of desert, would offend our intuitive notions of justice. See A. Heller, supra note 55, at 164 (arguing that sanctions implemented with the intent to deter actions violate individual and group rights). Similarly, a system in which the punishment for a particular crime was selected at random, rather than related to the nature of the offense committed, would strike most of us as fundamentally unjust.

63. See Gale, supra note 59, at 1004-05 (stating that the private standards of both judge and jury are accommodated by and incorporated into the punishments prescribed by our legal system).

64. See K. Moore, supra note 5, at 129-30 (arguing that, because a person should always receive the punishment which she deserves, pardons should only be granted where the punishment exceeds the crime).

65. See Gale, supra note 59, at 1005-11 (recognizing the contrasting goals of utilitarian and retributive theories of punishment).

66. See id. at 1008-09.

67. See K. Moore, supra note 5, at 122-23.

68. See K. Lebacqz, supra note 54, at 119, 120.

69. See, e.g., International Shoe Co. v. Washington, 326 U.S. 310, 316 (1945) (stating that due process requires that notice procedures not offend "`traditional notions of fair play and substantial justice'") (quoting Milliken v. Meyer, 311 U.S. 457, 463 (1940)); Fleming v. Kemp, 794 F.2d 1478, 1482 (11th Cir. 1986) (per curiam) (observing that "`the "ends of justice" [in habeas proceedings] are defined by objective factors, such as whether there was a full and fair hearing'") (quoting Witt v. Wainwright, 755 F.2d 1396, 1397 (11th Cir. 1985)); Sotelo v. State, 273 Ind. 694, 700, 408 N.E.2d 1215, 1218 (1980) ("It may be well to observe that justice has been defined as the end result of the application of fair procedures.").

70. J. Inst. 1.1 (J.A.C. Thomas trans. 1975).

71. See Larson v. American Wheel & Brake, Inc., 610 F.2d 506, 510 (8th Cir. 1979) (noting that the equitable aspects of justice, as Justinian defines the term, require one to "`live honestly, to harm nobody, to render every man his due'" (quoting J. Inst., supra note 70, at 1.1.3)); Berkowitz v. United States, 340 F.2d 168, 172 (1st Cir. 1965) ("To give each man his due (the test which Justinian made so famous a juristic maxim and which in an earlier form is found in Romans 13:7) is only to give him what society says is owing,--nothing more."); Castro v. Beecher, 365 F. Supp. 655, 660 (D. Mass. 1973) (characterizing a provision to give certain groups hiring priority as "not an example of Orwellian justice which treats some as more equal than others [, but instead] nothing more than, in Justinian's celebrated phrase, giving each man his due"); In re Pellegrino, 42 Bankr. 129, 138 (Bankr. D. Conn. 1984) (stating that "justice is the end result when the rule of law is interpreted with common sense, reason and simple fairness"); Gay St. Corp. v. United States, 130 Ct. Cl. 341, 350 n.1 (1955) (stating that "equity denotes the spirit and habit of fairness, justness, and right dealing . . . the rule of doing to all others as we desire them to do to us; or as it is expressed by Justinian--`to live honestly, to harm nobody, to render to every man his due'"); Collier v. Lindley, 203 Cal. 641, 654, 266 P. 526, 530 (1928) (interpreting the word "justice" in a trust agreement according to the definition that Justinian provided, that is, "`[t]he constant and perpetual disposition to render every man his due'"); Livingston Oil Corp. v. Henson, 90 Okla. 76, 77, 215 P. 1057, 1059 (1923) (reiterating that "justice in its common acceptance means rendering to every man his due so that neither party may gain by the other's loss").

72. E.g., A. Von Hirsch, Doing Justice: The Choice of Punishments (Report of the Committee for the Study of Incarceration) xxxviii-xxxix (1976) ("[The Committee] embraced the seemingly harsh principle of just deserts. . . . There is the feeling of a Kantian imperative behind the word `deserts.' Certain things are simply wrong and ought to be punished. And this we do believe.").

73. 18 U.S.C.A. 3551-3580 (West Supp. 1990) (setting forth the rules for sentencing, ordering punishment, and executing judgment in federal criminal proceedings); 28 U.S.C.A. 991-998 (West Supp. 1990) (establishing a commission to promulgate federal sentencing policies and practices).

74. 18 U.S.C.A. 3624(a)-(b) (West Supp. 1990).

75. See Survey, supra note 21, at 20.

76. Biddle v. Perovich, 274 U.S. 480, 486 (1927).

77. See Survey, supra note 21, at 8. The Survey also suggests that pardons were not prevalent in Athens because of the substantial role played by private suits in criminal law, whereby the victim received monetary damages from the offender. Consequently, pardons were limited to public crimes such as treason. Id.

78. See D. MacDowell, The Law in Classical Athens 258-59 (1978); Survey, supra note 21, at 9.

79. See Survey, supra note 21, at 9. Because they were known to many people, athletes and orators were the offenders most likely to be granted clemency through adeia. Only a few pardons of individuals are recorded in Greek history: the recall of Alcibiades in 408 B.C., the pardon of Demosthenes in 323 B.C., and the recall of Thucydides, the historian. Id. Professor MacDowell also recounts the pardon of Dorieus, a Rhodian athlete who had been condemned to death, because it was thought to be a pity that a man of such athletic prowess should be brought so low. See D. MacDowell, supra note 78, at 258. The relative rarity of clemency being granted to individuals is also reflected by the Greek language's lack of a specific term for individual pardon. See Survey, supra note 21, at 9.

80. See Aristotle, Constitution of Athens, ch. 39 (K. Von Fritz & E. Kapp trans. 2d printing 1961).

81. See N. Hammond, A History of Greece to 322 B.C., at 446-47 (1967).

82. See Survey, supra note 21, at 9.

83. Pilate saith unto him, "What is truth?" And when he had said this, he went out again unto the Jews, and saith unto them, "I find in him no fault at all. But ye have a custom, that I should release unto you one at the passover: will ye therefore that I release unto you the King of the Jews?" Then cried they all again, saying, "Not this man, but Barabas."

John 18:38-40 (King James).

84. K. Moore, supra note 5, at 17. Moore points out that clemency was often dispensed in conjunction with local holidays, when the Romans were most interested in quieting crowds. Id. at 16-17.

85. See id. at 16.

86. Survey, supra note 21, at 12 & n.54. Cicero provides the following account of Horatio's crime and defense of the homicide:

A retort of the charge occurs when the defendant admits the act of which he is accused but shows that he was justified in doing it because he was influenced by an offence committed by the other party. The following is an example: Horatius after killing the Curiatii and losing his two brothers returned home in triumph. He noticed his sister not distressed by the death of her brothers, but repeatedly calling on the name of Curiatius, her betrothed, with groans and tears. Filled with rage he killed the girl. He is brought to trial. The charge is: "You killed your sister without warrant." The answer is: "I was justified in killing her." The question is:  "Was he justified in killing her?" The defendant's reason is: "For she was distressed at the death of our enemies; she was unmoved by the fall of her brothers; she was grieved that I and the Roman people were victorious." The prosecutor's answer is: "Nevertheless she ought not to have been killed by her brother uncondemned." From this the point for decision arises: "Granted that Horatia was unmoved by the death of her brothers, and was distressed at the death of our enemies and did not rejoice over the victory of her brother and the Roman people, ought she to have been killed by her brother without condemnation?"

Cicero, de Inventione, bk. II, at 243, 245 (H. Hubbell trans. 2d printing 1960).

87. See Plutarch, Numa Pompilius, in The Lives of the Noble Grecians and Romans 74, 83 (J. Dryden trans., A. Clough rev., Modern Library ed. n.d.) ("[I]f in their walks they chance to meet a criminal on his way to execution, it saves his life, upon oath made that the meeting was an accidental one, and not concerted or of set purpose.").

88. Survey, supra note 21, at 7 (noting that it is not surprising that no clear-cut institution of clemency emerged under Mosaic law because "God was the supreme leader and he transferred his commands by the voice of priests and prophets and by laws"; thus one who violated the laws "had offended God and no one but God could pardon him").

89. Id. at 12 (noting that older Roman law considered repeal of the sentence of a jury by the Comitiae to be an abuse of the people's sovereignty).

90. 4 W. Blackstone, supra note 3, at *388.

91. Id. at *390-91.

92. Id. at *398.

93. E. Coke, supra note 29, at 233.

94. See Grupp, Some Historical Aspects of the Pardon in England, 7 Am. J. Legal Hist. 51, 55 (1963).

95. An Acte for Recontynuyng of +mctayne li+mbties and francheses heretofore taken frome the Crowne, 1535-36, 27 Hen. 8, ch. 24, I; see also G. Elton, Reform & Reformation: England, 1509-1558, at 201-02 (1977) (describing the consolidation of the clemency power, as well as the enforcement of all writs, in the king, which brought additional profits to Henry VIII).

The year in which Henry VIII seized pardoning power, 1535, was also the year in which Sir Thomas More was executed, making it clear that Henry VIII would tolerate no opposition. G. Smith, A History of England 222 (2d ed. 1957). Ironically, Henry VIII actually employed the pardoning power on More's behalf, but only to change his former Lord Chancellor's sentence from being hanged and disemboweled while alive to hanging. Upon learning of this, More is reported to have "merrily said, `God forbid the king should use any more such mercy to any of my friends, and God bless all my posterity from such pardons.'" T. Howell, Trial of Sir Thomas More, in State Trials 385, 394 (1535).

96. There were some limitations placed on the clemency power of the Crown at common law. Traditionally, the Crown could not impair the rights of third parties in recovering reparations from the wrongdoer. See Duker, supra note 18, at 486. In addition, the pardon had to specify the particular crime for which the offender was being forgiven. See 4 W. Blackstone, supra note 3, at *400. A few statutory limitations on the power also existed, including a requirement that no pardon for treason, murder, or rape should issue unless the Crown specified the offense with particularity. See Survey, supra note 21, at 135.

97. Some encroachment on the royal clemency power by the church continued well into the nineteenth century. The "benefit of the clergy" exemption, which exempted clerics and their associates from criminal responsibility, was not abolished until 1827. See Grupp, supra note 94, at 57.

98. For an excellent account of the impeachment and circumstances surrounding Osborne's pardon, see Duker, supra note 18, at 487-95.

99. Id. at 488.

100. See Survey, supra note 21, at 137 (noting that punishment for impeachment might also include forfeiture, exile, imprisonment, fine, or even death).

101. See Grupp, supra note 94, at 491-94.

102. See Duker, supra note 18, at 495.

103. Habeus Corpus Act, 1679, 31 Car. 2, ch. 2, 11.

104. See id.; Grupp, supra note 94, at 57.

105. Bill of Rights, 1689, W. & M., ch. 2, 2.

106. See id.; Grupp, supra note 94, at 57.

107. See 13 H.C. Jour. 624 (1701); 16 H.L. Jour. 737 (1701).

108. See 2 W. Blackstone, supra note 3, at *399-400.

109. See Act of Settlement, 1721, 7 Geo. 1, ch. 29; Grupp, supra note 94, at 57.

110. See supra note 91 and accompanying text.

111. See Survey, supra note 21, at 29-30 ("It is true, a power with no or few limits laid in the hands of a personality of supreme political power could not escape improper use. A glance at the facts of medieval legal life and criminal practice in England exceeds by far the most pessimistic expectations.").

112. N. Hurnard, The King's Pardon for Homicide, Before A.D. 1307, at vii (1969); see also Grupp, supra note 94, at 58-59 (suggesting that pardons were granted too generously in times of war, at the beginning of new reigns, and during the Black Plague).

113. See Survey, supra note 21, at 30. To obtain a pardon, criminals "had only to secure the good offices of some" lord or bishop, of whose services the king was in need in times of war. Id. "An entry would then appear upon the rolls to the effect that, at the special request of some nobleman named, who testified to the innocence of the accused, a pardon had been granted to him." Id.

114. See id. ("There was apparently no disgrace in the offer of a sum of money to the King or to other interested persons [for a pardon]."); Grupp, supra note 94, at 59 (noting that "the purchase of pardon was once a common practice and seems to have been generally associated with the procurement of a pardon in early times").

115. See Sebba, supra note 16, at 225-28 (citing C. Beccaria, On Crimes and Punishments (H. Paolucci ed. 1963), H. Fielding, An enquiry into the Causes of the Late Increase of Robbers (1751), and J. Monteil, La Grace en Droit Francais Moderne (1959), as critics of the vesting of the pardon power in the executive).

116. See id. at 224.

117. 11 W. Holdsworth, supra note 51, at 573.

118. This use of the clemency power may in one sense be viewed as actually enhancing "justice" because it enables the state to punish more readily someone who has transgressed society's laws. However, a conception of justice as the rendering to each person of her due under the law, see supra notes 55-74 and accompanying text, would require that both the principal wrongdoer and the accomplice receive punishment commensurate with their culpability. A decision by the state to spare a smaller fish the barbed hook of punishment in an effort to land a more substantial one may be a pragmatically sound decision, but it does not result in the fair allocation of deserts. See generally K. Moore, supra note 5, at 199-200 (classifying such uses of the clemency power as in the public welfare rather than in the interests of justice). Consequently, I characterize conditional pardons predicated on the recipient's turning "king's evidence" as justice-neutral rather than justice-enhancing.

119. See Sebba, supra note 16, at 225.

120. See 7 F. Thorpe, The Federal and State Constitutions, Colonial Charters, and Other Organic Laws of the States, Territories, and Colonies Now or Heretofore Forming the United States of America 3801 (1909) (Virginia Charter of 1609); 3 id. at 1858 (Charter of Massachusetts Bay, 1629); id. at 1679-80 (Charter of Maryland, 1632); id. at 1629 (Charter of Maine, 1639); 5 id. at 2791 (Charter of Carolina, 1665); id. at 2551 (New Jersey Charter, 1676); id. at 3038 (Charter for the Province of Pennsylvania, 1681); 2 id. at 770 (Charter of Georgia, 1732); C. Jensen, The Pardoning Power in the American States 3-8 (1922).

121. See 1 F. Thorpe, supra note 120, at 534 (Charter of Connecticut, 1662); 6 id. at 3215 (Charter of Rhode Island and Providence Plantations); C. Jensen, supra note 120, at 5.

122. Indeed, in arguing that strict separation of powers was neither desirable nor practiced in most states, Madison used as examples the sharing of the pardoning power by the executive and legislative branches of several states. The Federalist No. 47, at 306-08 (J. Madison) (C. Rossiter ed. 1961).

123. See C. Jensen, supra note 120, at 10 (stating that in Georgia the assembly made final pardon decisions).

124. See Del. Const. of 1776, art. VII; Md. Const. of 1776, art. XXXIII; N.C. Const. of 1776, art. XIX; N.Y. Const. of 1777, art. XVIII; S.C. Const. of 1790, art. II, 7; C. Jensen, supra note 120, at 10.

125. Neither the Virginia Plan, proposed by Edmund Randolph on May 29, 1787, nor the New Jersey Plan, proposed by William Paterson on June 15, 1787, made any provision for the granting of clemency. See 1 The Records of the Federal Convention of 1787, at 20-23 (M. Farrand ed. 1911) [hereinafter M. Farrand] (Virginia Plan); id. at 243-45 (New Jersey Plan).

126. Id. at 292.

127. See 13 H.C. Jour. 624, 627 (1701); 16 H.L. Jour. 737, 737 (1701).

128. 2 M. Farrand, supra note 125, at 171-72.

129. See id. at 411, 419-20.

130. Id. at 419. "Mr. Sherman moved to amend the `power to grant reprieves & pardons' so as to read `to grant reprieves until the ensuing session of the Senate, and pardons with consent of the Senate.'" Id. The motion was voted down eight to one. See id.

131. Randolph argued: "The prerogative of pardon in these cases was too great a trust. The President may himself be guilty. The Traytors may be his own instruments." Id. at 626.

132. The Records of the Federal Convention contain the following exchanges:

Col: Mason supported [Mr. Randolph's] motion.

Mr Govr Morris had rather there should be no pardon for treason, than let the power devolve on the Legislature.

Mr Wilson. Pardon is necessary for cases of treason, and is best placed in the hands of the Executive. If he be himself a party to the guilt he can be impeached and prosecuted.

Mr. King thought it would be inconsistent with the Constitutional separation of the Executive & Legislative powers to let the prerogative be exercised by the latter--A Legislative body is utterly unfit for the purpose. They are governed too much by the passions of the moment. In Massachusetts, one assembly would have hung all the insurgents in that State: the next was equally disposed to pardon them all. He suggested the expedient of requiring the concurrence of the Senate in Acts of Pardon.

Mr. Madison admitted the force of objections to the Legislature, but the pardon of treasons was so peculiarly improper for the President that he should acquiesce in the transfer of it to the former, rather than leave it altogether in the hands of the latter. He would prefer to either an association of the Senate as a Council of advice, with the President.

Mr. Randolph could not admit the Senate into a share of the Power. The great danger to liberty lay in a combination between the President & that body--

Col: Mason. The Senate has already too much power--There can be no danger of too much lenity in legislative pardons, as the Senate must concur, & the President moreover can require two-thirds of both Houses.

Id. at 626-27. The motion failed by a vote of eight to two, with one divided delegation. See id at 627.

133. See id. at 639.

134. See The Federalist No. 74, supra note 122, at 447-49.

135. See Address by James Iredell, North Carolina Ratifying Convention (July 28, 1788) [hereinafter Iredell Address], reprinted in 4 The Founders' Constitution 17 (P. Kurland & R. Lerner ed. 1987).

136. The Federalist No. 74, supra note 122, at 447.

137. Iredell Address, supra note 135, at 17-18.

138. The Federalist No. 74, supra note 122, at 449.

139. Iredell Address, supra note 135, at 18.

140. See Duker, supra note 18, at 504-06 (describing how the few arguments presented at state ratification conventions against granting pardon powers to the President were quelled by proponents of the power, such as Hamilton).

141. See Presidential Proclamation (July 10, 1795), reprinted in 1 A Compilation of the Messages and Papers of the Presidents, 1789-1895, at 181 (J. Richardson ed. 1896) [hereinafter Messages and Papers].

142. See Presidential Proclamation (May 21, 1800), reprinted in 1 Messages and Papers, supra note 141, at 303-04.

143. 11 The Writings of Thomas Jefferson 43-44 (A. Bergh ed. 1907) (1853) (letter to Mrs. John Adams, July 22, 1804). In New York Times Co. v. Sullivan, 376 U.S. 254 (1964), the Court noted that although the constitutionality of the Sedition Act was never tested, "the attack upon its validity has carried the day in the court of history." Id. at 276. Indeed, Jefferson considered informing Congress of his intention to render the Alien and Sedition Act nugatory through the use of the pardon power. However, because he decided that the unlimited nature of the power was clear without such a statement, he deleted it from the final draft of his first address to Congress. Duker, supra note 18, at 530.

144. L. Levy, Jefferson and Civil Liberties 163 (1963).

145. Id. at 72-73.

146. Id. at 73.

147. See 1 Op. Att'y Gen. 341 (1820). Attorney General Wirt wrote: "I understand from Judge Duvall that convicts have been pardoned by the President of the United States on condition of joining the navy of the United States: evidence of which, if it be so, may, I presume, be found in the Department of State." Id. at 343. Wirt was of the opinion that the President could not only pardon conditionally, but could also pardon before conviction, because a pardon "pre-supposes an offence and nothing more." Id.

148. See Presidential Proclamation (May 29, 1865), reprinted in 6 Messages and Papers, supra note 141, at 310; Presidential Proclamation (Sept. 7, 1867), reprinted in id. at 547; Presidential Proclamation (July 4, 1868), reprinted in id. at 655; Presidential Proclamation (Dec. 25, 1868), reprinted in id. at 708. Andrew Jackson in 1830 also issued a full pardon by executive order to all army deserters. See Exec. Order No. 29 (June 12, 1830), reprinted in 2 Messages and Papers, supra note 141, at 499.

149. See Act of July 17, 1862, ch. 195, 13, 12 Stat. 589, 592, repealed by Act of Jan. 21, 1867, ch. 8, 14 Stat. 377 ("[T]he President is hereby authorized, at any time hereafter, by proclamation, to extend to persons who may have participated in the existing rebellion in any State or part thereof, pardon and amnesty, with such exceptions and at such time and on such conditions as he may deem expedient for the public welfare.").

150. Lincoln had a reputation for being "a pardoner, softhearted rather than hardhearted," even though it interfered with army discipline. 3 C. Sandburg, Abraham Lincoln: The War Years 512 (1939).

151. See Act of Jan. 21, 1867, ch. 8, 14 Stat. 377. The Act attempted to limit the power of the President to restore the property of secessionists via executive pardon by repealing 13 of the Act of July 17, 1862, discussed at supra note 151. The 1867 Act became law without Johnson's signature. For an account of the dispute between the executive and legislative branches over the pardoning of secessionists, see Duker, supra note 18, at 509-19.

152. See supra text accompanying notes 103-09.

153. See Marbury v. Madison, 5 U.S. (1 Cranch) 137, 166-67 (1803).

154. 32 U.S. (7 Pet.) 150 (1833).

155. Id. at 160.

156. See Biddle v. Perovich, 274 U.S. 480, 486 (1927). Justice Holmes stated that "[a] pardon in our days is not a private act of grace from an individual happening to possess power. It is part of the Constitutional scheme. When granted it is the determination of the ultimate authority that the public welfare will be better served by inflicting less than what the judgment fixed." Id.

157. See Schick v. Reed, 419 U.S. 256, 266 (1974) (stating that throughout American history, attacks on the President's right to grant conditional pardons "have been firmly rejected by the courts" and that "[t]he plain purpose of the broad power conferred by 2, cl. 1, was to allow plenary authority in the President to `forgive' the convicted person in part or entirely, to reduce a penalty in terms of a specified number of years, or to alter it with conditions which are in themselves constitutionally unobjectionable"). See generally W. Humbert, supra note 42, at 33-81 (reviewing Supreme Court decisions that have interpreted the President's pardoning power broadly).

158. 59 U.S. (18 How.) 307 (1855).

159. Id. at 310-15.

160. See W. Humbert, supra note 42, at 36-41.

161. See Armstrong v. United States, 80 U.S. (13 Wall.) 154, 155-56 (1872) (upholding the validity of Andrew Johnson's proclamation of pardon and amnesty of December 25, 1868).

162. See United States v. Klein, 80 U.S. (13 Wall.) 128, 142 (1872) (upholding the President's power to "annex to his offer of pardon any conditions or qualifications he should see fit").

163. 91 U.S. 474 (1875).

164. Id. at 478. Although the statement in Osborn concerning the remission of penalties may be characterized as dicta, there is little doubt, in light of the expansive tone of the Court's decisions regarding the clemency power, that the President possesses the power. See, e.g., Illinois Cent. R.R. v. Bosworth, 133 U.S. 92, 103 (1890) (holding that property, previously forfeited, is restored upon grant of pardon); see also W. Humbert, supra note 42, at 50 n.63, 96-99 (noting that Osborn cannot, strictly speaking, be used as authority for saying that the President may remit fines and forfeitures when such remission is not accompanied by a pardon, even though there is no doubt that, in practice, the President may remit without pardoning).

165. 274 U.S. 480 (1927).

166. See Burdick v. United States, 236 U.S. 79 (1915).

167. Biddle, 274 U.S. at 487-88.

168. 71 U.S. (4 Wall.) 333 (1866).

169. Id. at 380 (emphasis added). The only extraconstitutional limitations on the operation of the clemency power recognized by the Garland Court were the English common law limitations that the pardon power cannot restore offices forfeited or affect the vested property rights of third parties. See id. at 381 (citing 4 W. Blackstone, supra note 3, at *402).

170. 419 U.S. 256 (1974).

171. Id. at 268. President Eisenhower had commuted the death sentence of Maurice Schick to life imprisonment, subject to the condition that Schick would never be eligible for parole. The no-parole condition did not coincide with the punishment prescribed by Congress for Schick's crime, and thus Schick sought to have the condition set aside.

172. Id. at 266.

173. See supra text accompanying notes 103-09; see also Schick, 419 U.S. at 277-78 (Marshall, J., dissenting) (disagreeing with the majority's reliance on the English pardoning power by pointing out that even the king did not have the authority to define extralegislative punishments); Hoffa v. Saxbe, 378 F. Supp. 1221, 1227 (D.D.C. 1974) (listing three important limitations on the king's pardoning power: the Habeas Corpus Act, the Bill of Rights, and the Act of Settlement).

174. See Duker, supra note 18, at 521-24. Duker points out a similar flaw in the Court's reasoning in Brown v. Walker, 161 U.S. 591 (1896), in which the Court, relying on the English pardoning system, held that Congress also has the power to pass acts of amnesty. See Duker, supra note 18, at 523.

175. See supra notes 126-40 and accompanying text.

176. The Schick majority's conclusion that the President's clemency is not limited to imposing those lesser forms of punishment that have been prescribed by Congress for a particular offense is not as vulnerable to attack on separation-of-powers grounds as the dissenters suggest. Justices Marshall, Douglas, and Brennan argued in dissent that the attachment to an act of clemency of a punitive condition that has not been authorized by Congress permits the executive to engage in the lawmaking function of prescribing punishment. Schick, 419 U.S. at 275.

However, this argument seems to ignore the nature of the clemency power as a fundamental exception to separation-of-powers principles that operates in favor of the President: clemency was intended to give the executive a check on the legislative and judicial functions of proscribing conduct and assigning punishments for crimes. The reasoning advanced by the dissenting justices would turn this constitutional check on its head by rendering the President's clemency options merely a subset of the possible punishments prescribed by Congress for a particular offense. This crabbed interpretation of the clemency power would undoubtedly limit the ability of the President to remedy injustices or further the interests of the government through clemency. See id. at 266-67; see also supra notes 137-40 and accompanying text (describing the justice-neutral and justice-enhancing interests furthered through clemency). Moreover, recognition of what is in effect a legislative check on the clemency power would itself appear to offend traditional notions of separation of powers. See generally Morrison v. Olson, 487 U.S. 654, 697-715 (1988) (Scalia, J., dissenting) (discussing the importance of keeping separate the constitutional powers respectively allocated in articles I, II, and III, to the Congress, President, and judiciary).

177. 267 U.S. 87 (1925).

178. Id. at 98.

179. Id. at 121.

180. Id.

181. Id.

182. In his insightful historical treatment of the President's power to pardon, William Duker argues that the Court in Grossman effectively "gave away the store" by ratifying a power of the President that has been, and probably will continue to be, used by the President to shield subordinates from judicial and congressional scrutiny. See Duker, supra note 18, at 526-30. Duker points out that the rationale of Grossman would permit the President to pardon contempts of Congress as well as judicial contempts, a result that is antithetical to the principle of separation of powers. See id. at 529. Duker's concern echoes that of Justice Story, who warned that recognition of such a broad pardoning power would render Congress wholly dependent on the good will of the President for the exercise of its powers. See J. Story, Commentaries on the Constitution of the United States 774 (J. Nowak & R. Rotunda reprint 1987) (abr. ed. 1833); cf. W. Humbert, supra note 42, at 63 (asserting that the exercise of the clemency power "lies in the absolute and uncontrolled discretion of the President"). The Grossman decision, read in concert with the Court's blanket rejection in Garland and Schick of any sort of legislative limitation on the executive clemency power, indicates that a presidential pardon could be used to frustrate the legislature in just such a fashion. Of course, by employing the clemency power in such an obstructive fashion, the President would run the risk of having Congress commence impeachment proceedings. For a thoughtful discussion of the grounds for impeachment, see Gerhardt, The Constitutional Limits to Impeachment and Its Alternatives, 68 Texas L. Rev. 1, 82-88 (1989) (arguing that Congress properly may define those "political crimes" which constitute impeachable offenses).

183. 378 F. Supp. 1221 (D.D.C. 1974). While appeal to the United States Court of Appeals for the District of Columbia was pending, Hoffa disappeared, prompting the court to remove the case from its calendar. The appeal was never considered, because Hoffa was never found. See Boudin, supra note 41, at 22.

184. Hoffa, 378 F. Supp. at 1224.

185. Id.

186. Id.

187. Id. at 1225.

188. Id.

189. The test enunciated in Lemon v. Kurtzman, 403 U.S. 602 (1971), requires that a statute challenged under the establishment clause have a secular legislative purpose, id. at 612-13, which has resulted in the Court invalidating any act of the Legislature that is enacted with an improper purpose. See, e.g., Meeks v. Pittenger, 421 U.S. 349, 365 (1975) (holding that Pennsylvania acted with an impermissible religious purpose in loaning instructional materials and providing guidance to nonpublic schools). Justice Scalia has criticized this prong of the Lemon test because it involves the Court in the task of ascertaining the motives of individual legislators. See Edwards v. Aguillard, 482 U.S. 578, 636-638 (1987) (Scalia, J., dissenting). Yet the test is still followed by a majority of the Court. See id. at 583.

190. See, e.g., Village of Arlington Heights v. Metropolitan Hous. Dev. Corp., 429 U.S. 252, 265 (1977) (holding that, to establish an equal protection violation, a plaintiff must show that the legislative or administrative body was motivated by racially discriminatory intent or purposes).

191. The propriety of President Ford's pardon of Richard Nixon was reviewed in Murphy v. Ford, 390 F. Supp. 1372 (W.D. Mich. 1975), a case that has largely been ignored by courts and commentators. In Murphy, the plaintiff sought a declaratory judgment that Nixon's pardon was void because Ford had exceeded the scope of the clemency power set forth in article II and because the pardon was contrary to public policy. Id. at 1372-73. In an acerbic opinion that characterized Nixon as "the putative rebel leader," the district court addressed the merits and held that the pardon was "a prudent public policy judgment" that was within both the letter and spirit of the pardoning power. Id. at 1374. Unfortunately, the court did not address the standing and separation-of-powers issues raised by its decision to review whether the pardon was in the public interest. See infra text accompanying notes 298-321.

192. Hoffa, 378 F. Supp. at 1234-35.

193. Id. at 1236. Judge Pratt fashioned this standard without reference to any Supreme Court decisions, relying on a handful of decisions from lower federal courts and state courts which discussed the validity of conditional executive pardons. Id. at 1236-37 (citing, e.g., Bishop v. United States, 233 F.2d 582 (D.C. Cir. 1955); Lupo v. Zerbst, 92 F.2d 362 (5th Cir. 1937); and Kavalin v. White, 44 F.2d 49, 51 (10th Cir. 1930)).

194. Id. at 1235 n.48.

195. Id. at 1237-38.

196. Id. at 1239-40. In United States v. O'Brien, 391 U.S. 367 (1968), Justice Warren set forth a test for determining when certain types of speech can be regulated: "[A] government regulation is sufficiently justified if it is within the constitutional power of the Government; if it furthers an important or substantial governmental interest; if the governmental interest is unrelated to the suppression of free expression; and if the incidental restriction on alleged First Amendment freedoms is no greater than is essential to the furtherance of that interest." Id. at 377. In evaluating Hoffa's first-amendment claims, the court used the test set forth in O'Brien, because it considered the commutation condition to be directed primarily at Hoffa's conduct, rather than at "pure speech."

197. No other court, however, has had occasion to adopt the Hoffa standard.

198. 95 U.S. 149 (1877).

199. Id. at 154.

200. Id. at 154-55.

201. 236 U.S. 79 (1915).

202. Id. at 94-95.

203. Id. at 93-94.

204. 419 U.S. 256 (1974).

205. Id. at 266.

206. See Biddle v. Perovich, 274 U.S. 480, 488 (1927) (holding that a prisoner need not consent to commutation of his sentence from death to life imprisonment; the reasoning of Burdick "is not to be extended to the present case").

207. The written recommendation made by the pardon attorneys on each application for clemency are reviewed by two officials at the Department of Justice, and then by two officials at the White House, each of whom generally winnows out the favorable recommendations, so that very few eventually reach the President. See Krajick, The Quality of Mercy, 5 Corrections Magazine, June 1979, at 46, 51.

208. Id. at 52.

209. W. Humbert, supra note 42, at 126 (footnotes omitted).

210. For a fascinating instance of a presidential pardon being granted because of sympathy with the plight of the offender, see former U.S. Attorney General Harry M. Daugherty's account of the pardoning of socialist activist Eugene Debs. See H. Daugherty & T. Dixon, The Inside Story of the Harding Tragedy 115-21 (1932). Daugherty wrote that he had never met "a man of more appealing personality than Eugene V. Debs" and recommended that his sentence be commuted despite the fairness of his trial and sentence, and despite the fact that hundreds of other people were in prison for the same crime. See id. at 120. President Harding followed Daugherty's recommendation, but moved up the effective date of commutation to December 24, 1921, because, as Daugherty relates, the President smiled and said "`I want him to eat his Christmas dinner with his wife.'" Id. at 121.

211. See W. Humbert, supra note 42, at 124-25 tables V & VI (cataloging reasons given by Presidents for granting clemency from 1885 to 1931).

212. See Office of the United States Pardon Attorney, Executive Clemency Case Statistics (accompanying this Article as an Appendix) [hereinafter Clemency Case Statistics].

213. See id.; see also Clark, Reagan Parsimonious in Use of Pardon Power, 42 Cong. Q. 2878, 2878 (1984) (comparing the percentage of grantings of pardons and noting that Reagan's rate was the lowest of any President in the past 50 years).

214. See Clemency Case Statistics, supra note 212.

215. See Cozart, supra note 34, at 3. Cozart, who was a United States pardon attorney during the Eisenhower Administration, noted that presidential commutation of an ongoing sentence is "infrequent," id., and that it is the position of the Department of Justice that rigidities of the law should not be corrected through clemency, but through changes in the law itself, see id. at 4.

216. United States pardon attorney Cozart candidly acknowledged that the power of the President to pardon on grounds of innocence "is rarely exercised." Id.

217. See Krajick, supra note 207, at 48 (quoting Raymond Theim).

218. 28 C.F.R. 1.3 (1963).

219. 28 C.F.R. 1.2 (1983).

220. 28 C.F.R. 1.3 (1989). This section is virtually unchanged from the form in which it was originally adopted by the Kennedy Administration in 1962. See 28 C.F.R. 1.4 (1963).

221. The most famous presidential pardon in our history, that of Richard Nixon by President Ford, is a notable exception. However, Ford's White House counsel, rather than the Office of the Pardon Attorney, handled the Nixon pardon. If Nixon had sought clemency pursuant to the regulatory provisions applicable to other applicants, the pardon attorney "`would have had to say "no" under our rules.'" Clark, supra note 213, at 2880 (quoting United States pardon attorney David C. Stephenson).

222. From 1969 through 1989, 2181 presidential pardons were granted, while only 136 commutations were granted. See Clemency Case Statistics, supra note 212.

223. See id.

224. Krajick, supra note 207, at 53 (quoting John Stanish).

225. K. Moore, supra note 5, at 83.

226. Even President Carter's 1977 grant of amnesty to draft resisters from the Vietnam era, Proclamation No. 4483, 3 C.F.R. 4 (1978), which, except for the pardon of Richard Nixon, was probably the most significant use of the clemency power in recent times, was essentially justice-neutral. In the tradition of most presidential amnesties, Carter's amnesty was issued to put the Vietnam War behind the country and enhance national unity. See Pardon for Draft Evaders: Carter's First Act Touches Off a Storm, U.S. News & World Rep., Jan. 31, 1977, at 22, 22; see also Saliterman, Reflections on the Presidential Clemency Power, 38 Okla. L. Rev. 257, 264 (1985) (noting that President Ford's Clemency Program for those who had committed crimes related to the Vietnam War contributed to the alleviation of national stress and the restoration of national tranquility).

227. The Federalist No. 74, supra note 122, at 447.

228. C. Jensen, supra note 120, at 10.

229. K. Hall, The Magic Mirror: Law in American History 104 (1989).

230. See id.; C. Jensen, supra note 120, at 10.

231. See C. Jensen, supra note 120, at 10. The constitutions of 26 of the first 35 states to join the Union vested the pardoning power in the executive alone. Id.

232. The twenty-nine states are Alaska, Arkansas, California, Colorado, Hawaii, Illinois, Iowa, Kansas, Kentucky, Louisiana, Maryland, Michigan, Mississippi, Missouri, New Hampshire, New Jersey, New Mexico, New York, North Carolina, Ohio, Oregon, South Dakota, Tennessee, Vermont, Virginia, Washington, West Virginia, Wisconsin, and Wyoming. Guide to State Clemency, supra note 28, at 15; id. at 17-20 table 1 (Comparative View of Clemency in the United States and Territories).

233. See id. at 15. The sixteen states are Arizona, Delaware, Florida, Indiana, Maine, Massachusetts, Minnesota, Montana, Nebraska, Nevada, North Dakota, Oklahoma, Pennsylvania, Rhode Island, Texas, and Utah. Id. at 17-20 table 1 (Comparative View of Clemency in the United States and Territories).

234. See id. at 15. The five states are Alabama, Connecticut, Georgia, Idaho, and South Carolina. Id. at 17-20 table 1 (Comparative View of Clemency in the United States and Territories).

235. See, e.g., Colo. Rev. Stat. 16-17-102 (1986) (providing that the governor may give weight to a petitioner's "[g]ood character previous to conviction, good conduct during confinement in the correctional facility, the statements of the sentencing judge and the district attorneys . . . and any other material concerning the merits of the application" when making the determination of whether a pardon should be granted); Minn. Stat. 638.02 (1990) (providing that the Board of Pardons has the discretion to grant a pardon if a petitioner has "been convicted of no criminal acts other than the act upon which such conviction was founded and is of good character and reputation"); Wash. Rev. Code Ann. 9.94A.150(4) (West 1988 & Supp. 1990) (providing that the governor, upon recommendation from the clemency and pardons board, may grant a pardon "for reasons of serious health problems, senility, advanced age, extraordinary meritorious acts, or other extraordinary circumstances"). Missouri requires by statute that the governor be informed of any inmate suffering from an incurable disease or life-endangering condition, though the decision whether to commute the prisoner's sentence is left entirely to the executive's discretion. Mo. Rev. Stat. 217.250 (1986).

236. See Gillin, Executive Clemency in Wisconsin, 42 J. Crim. L., Criminology & Pol. Sci. 755, 757-58 (1951) (describing variations in the clemency practices of several governors of Wisconsin).

237. A few states, including Georgia and South Carolina, grant clemency applications relatively freely, while at least one state, Rhode Island, currently refuses to use the clemency power. See Guide to State Clemency, supra note 28, at 164.

238. Cf. Barnett, supra note 20, at 491 (reviewing the prevailing theories of executive pardon and asserting that the strict legal theory of pardon as mercy requiring no justification is dangerous to the public and unjust to those who are pardoned on the basis of merit).

239. Id. at 492-93. For instance, the number of commutations granted in Maryland in 1978 appears quite high relative to other states until one realizes that nearly all of those were given out at Christmas "and almost none had a real impact on the prisoners' sentences." Krajick, supra note 207, at 50.

240. See Barnett, supra note 20, at 493.

241. See id. at 501-04; Gillin, supra note 236, at 755.

242. W. Humbert, supra note 42, at 5.

243. See supra text accompanying notes 212-27. The nature of the 1988 presidential campaign, which focused in large part on the prison furlough policies of Massachusetts Governor Michael Dukakis, indicates that a President's extensive use of the clemency power could also be politically dangerous. See Love, Bush Backers Have Horton Victims Speak, L.A. Times, Oct. 8, 1988, at 1-23, col. 4 (reporting on a press conference in which relatives of Willie Horton's victims described the crimes he committed while free on a weekend furlough); Cohen, William Horton's Furlough, Wash. Post, July 8, 1988, at A23, col. 3 (stating that "[t]he Horton episode is a blemish that cannot be dismissed by citing how well Dukakis has handled crime in Massachusetts").

244. See Boudin, supra note 41, at 15-16. The author notes that prior to impeaching President Johnson, Congress investigated his bank accounts in a search for money he might have received in exchange for granting pardons. However, abuse of the pardon power did not appear as a charge in the eventual indictment for the impeachment of Johnson. Id. at 16.

245. See id. at 34.

246. See Survey, supra note 21, at 150-53.

247. See P. Maas, Marie: A True Story 415 (1983). Blanton's conviction was ultimately reversed by the United States Court of Appeals for the Sixth Circuit, though other members of the Blanton Administration pleaded guilty of conspiring to sell pardons and commutations to inmates. See id. at 414-15. See generally Blackmore, Tennessee's Clemency-Selling Scheme: Could Blanton Not Have Known?, 5 Corrections Magazine, June 1979, at 55 (describing the circumstances and aftermath of Governor Blanton's eleventh-hour grants of clemency and his slightly premature ousting from office).

248. See P. Avrich, The Haymarket Tragedy 415-27 (1984).

249. Id. at 424.

250. M. DiSalle, The Power of Life or Death 204 (1965).

251. E. Brown & D. Adler, supra note 2.

252. See id. at 78-79.

253. Id. at 84.

254. See Margolick, Legal Scholar on Death Row Fights to Halt Own Execution, N.Y. Times, Mar. 5, 1990, at A1, col. 5, B6, col. 3 (discussing the attempts of jailhouse lawyer Joseph Giarratano to halt his own execution and noting that Virginia governor Douglas Wilder abandoned his opposition to the death penalty during his gubernatorial campaign); Kilpatrick, Why Kill When There's Reasonable Doubt?, Columbus Dispatch, Jan. 31, 1991, at 11A, col. 1 (discussing the doubts about the reliability of Giarratano's confessions and urging Governor Wilder to commute his death sentence).

255. See 20/20, supra note 20 (reporting on the refusal of Governor Edwin Edwards to commute the sentence of Wilber Rideau). More recently, Governor Buddy Roemer has also refused to commute Rideau's sentence. See Nightline: Why Won't Louisiana Parole Wilber Rideau? (ABC television broadcast, June 7, 1990).

256. See 60 Minutes: Life or Death (CBS television broadcast, Jan. 21, 1990) (reporting on Roemer's denial of clemency to Dalton Prejean). On May 18, 1990, Prejean was executed in the electric chair. Louisiana Executes Man Who Killed at Age 17, N.Y. Times, May 19, 1990, 1, at 9, col. 1.

257. I represented Mr. Brown in his request for clemency, the denial of which sparked my academic interest in the topic of clemency and the development of clemency standards.

258. See Larkin, Please, Governor, Let Man Go Free, Cleveland Plain Dealer, Dec. 4, 1988, at B1, col. 1 (editorial).

259. See infra note 267 for a discussion of this case.

260. See Larkin, supra note 258; see also Sharkey & Webb, Celeste Sought Review for Steele, Cleveland Plain Dealer, Sept. 12, 1987, at A1, col. 1 (reporting that Governor Celeste's office petitioned for the release of former Municipal Court Judge Steele); Gov. Celeste Owes Ohioans Explanation, Canton Repository, May 11, 1988, at A4, col. 1 (editorial) (attributing Celeste's refusal to commute Larry Brown's sentence to the small-time thief's lack of political clout).

261. For example, current New York Governor Mario Cuomo, characterized as the "stingiest" governor regarding the use of clemency, has granted clemency on only a handful of occasions since 1982 and did not grant a single clemency petition in 1989. See Marks, Cuomo Grants Few NY Inmates a Holiday Break, Newsday (New York), Dec. 23, 1990, at 8. Some have speculated that Cuomo's reluctance to grant clemency stems from his desire not to give his opponents an opportunity to attack him as soft on crime. Id.

262. See S. Lichten, Executive Clemency in Capital Cases 5 (unpublished article obtained from the Ohio Public Defender's office, reviewing executive clemency practices) (copy on file with the Texas Law Review); see also Goldberg, Introduction, 23 Loyola L.A.L. Rev. 1, 2-4 (1989) (noting that the number of persons on death row presently exceeds 2000, as compared to the approximately 600 who were scheduled for execution under mandatory sentencing laws when Furman v. Georgia, 408 U.S. 238 (1972), was decided).

263. Krajick, supra note 207, at 50 (quoting Edward Hammock) (emphasis added).

264. See supra notes 78-79 and accompanying text.

265. See Barnett, supra note 20, at 506-08. This approach to clemency is illustrated by the outrageous comment of former Governor Blease of South Carolina: "`I took the position that I was the servant of the people . . . and when a community where a crime had been committed, with the best people, the white people, signing the petition, said that the criminal had been punished enough, I turned him out without regard to criticism.'" Id. at 507 (quoting Governor Blease).

266. Krajick, supra note 207, at 50 (quoting Edward Hammock).

267. Despite a statewide campaign against drugs, marked by the stiffening of laws pertaining to drug use and sale, Governor Richard Celeste recently commuted the drug-trafficking sentence of Michael Badinovac from five years to six months. Badinovac, who had evaded incarceration for 10 years and had become a successful bank executive in Hawaii, was the subject of a massive letter-writing campaign from family and friends in Ohio and from politicians in Hawaii. See Snell, Governor Chops Hawaii Banker's Drug Sentence, Columbus Dispatch, July 27, 1989, at 1A, col. 4. Because there was evidently no question as to Badinovac's guilt, it appears that his release was based entirely on justice-neutral reasons pertaining to his behavior while evading incarceration and the influence of his supporters.

Celeste was also harshly criticized for commuting the sentence of former Municipal Judge Robert Steele, who was serving a life sentence for the murder of his wife. The Cleveland Plain Dealer reported that Celeste was petitioned by more than a hundred people, including the Governor's own father, to release Steele. See Sharkey & Webb, supra note 260. Following the parole board's unanimous recommendation that clemency be denied, Celeste took the extraordinary action of requesting that the parole board reconsider its decision, despite a board policy that would have precluded review. Id. Upon reconsideration, the board unanimously recommended that clemency be granted, and Celeste commuted the sentence, creating a storm of controversy. Id.

268. United States v. Wilson, 32 U.S. (7 Pet.) 150, 161 (1833).

269. Krajick, supra note 207, at 53 (quoting John Stanish).

270. Williams, Practice Under the Federal Sentencing Guidelines, 4 Ohio Law., Mar.-Apr. 1990, at 6, 7.

271. See Ogletree, The Death of Discretion? Reflections on the Federal Sentencing Guidelines, 101 Harv. L. Rev. 1938, 1956 (1988) (advocating amendments to the Guidelines designed to increase judges' discretion to consider a broad range of individual offender characteristics in sentencing). But cf. Williams, supra note 270, at 6-7 (asserting that even though the Guidelines invite mechanical and unjust application, they offer enough flexibility that judicial discretion need not necessarily be foreclosed).

272. See, e.g., Ohio Senate Bill 18, passed in May, 1990, which, among other things, eliminates the current threshold points when persons serving time for violent felonies become eligible for parole hearings. Columbus Dispatch, Mar. 11, 1990, at 2C, col. 1.

273. See Blystone v. Pennsylvania, 110 S. Ct. 1078 (1990). In the words of dissenting Justice William Brennan, "for the first time, the Court upholds a statute containing a mandatory provision that gives the legislature rather than the jury the ultimate decision whether the death penalty is appropriate in a particular set of circumstances." Id. at 1085.

274. See Malcolm, Florida's Jammed Prisons: More In Means More Out, N.Y. Times, July 3, 1989, at 1, col. 1 (city ed.). The Times reported in an article about Florida's overcrowded prison system that an influx of new prisoners means that more prisoners will be let out "the back door." Id. Many states currently operate over capacity. "Forty-six states now have historically high rates of imprisonment," while "[i]n 37 states some prisons or entire penal systems are under court orders to remedy overcrowding" or problems directly related to overcrowding. Id. Indeed, in a recent survey of state clemency practices, prison crowding was mentioned most commonly as an emerging issue pertaining to clemency. One state office went so far as to predict that "increased numbers at the State institutions will probably make clemency decisions more liberal in the future." Guide to State Clemency, supra note 28, at 175.

275. See Huff, Rattner & Sagarin, Guilty Until Proved Innocent: Wrongful Conviction and Public Policy, 32 Crime & Delinq. 518, 520-23 (1986); see also M. Yant, Presumed Guilty (forthcoming from Prometheus Books, 1991) (describing numerous instances of wrongful conviction and proposing that a special court of last resort be created to address such cases).

276. See Fricker, Crime and Punishment in Dallas, A.B.A. J., July 1989, at 52, 52-53. Adams's odyssey of injustice is movingly recounted in the film documentary, The Thin Blue Line (Minemax Films 1988).

277. See, e.g., Behind the Overturning of a '76 Murder Verdict, N.Y. Times, Aug. 3, 1989, at B1, col. 3 (describing how pressure for conviction wrongfully put a New York man in prison for 13 years); Juror Laments his Vote to Execute Innocent Man, Columbus Dispatch, Apr. 9, 1989, at 1C, col. 1 (describing the remorse of a juror who served on a Florida jury that convicted Joseph Brown of a murder of which he was cleared after 14 years on death row).

278. See Huff, Rattner & Sagarin, supra note 275, at 523; see also E. Borchard, Convicting the Innocent (1932) (reporting a study of 65 documented cases in which innocent men and women were convicted of various crimes).

279. See Bedau & Radelet, Miscarriages of Justice in Potentially Capital Cases, 40 Stan. L. Rev. 21, 23-24 (1987).

280. Coyle, Strasser & Lavelle, Fatal Defense: Trial and Error in the Nation's Death Belt, Nat'l L.J., June 11, 1990, at 30.

281. That the one attempt to abolish the institution of clemency and leave matters of justice entirely to the courts was a failure attests to the usefulness of clemency. Following the French Revolution, clemency in France was temporarily abolished in favor of reliance solely on the judicial system. However, the need for clemency or an analogous extrajudicial form of mitigating punishment soon became apparent. See Sebba, supra note 16, at 227-28.

282. For example, a single court that imposed the death penalty from time to time in individual cases might not detect a racially discriminatory pattern in the imposition of capital punishment. In contrast, a clemency commission that possessed a universe of all death sentences imposed within the jurisdiction could more easily determine if capital punishment was being imposed unfairly because of the race of the offender or the victim.

283. The legitimacy of such justice-neutral uses of clemency seems clear at least from a political perspective, if not a philosophical one. See K. Moore, supra note 5, at 201-02. Moore asserts that the pardons of Confederate soldiers solely to promote national harmony are indefensible under strict retributivist philosophy. Id. However, Moore does seem to waver in her argument for complete philosophical consistency in this context:

It is very difficult to tell a government that it is unjust to take the steps necessary to save itself. Would even Kant expect a just society to forbear from the one unjust act that would keep an unjust revolution from sweeping the country? Fortunately, what cannot be justified on one ground can often be justified on others.

Id. at 202.

284. See Duker, supra note 18, at 509-19 (discussing attempts by Republican members of Congress to limit the pardoning policies of President Johnson, which were considered too lenient toward former secessionists).

285. For a fascinating and elegantly written exposition of current scientific thinking about evolution, see S. Gould, Wonderful Life 13-323 (1989).

286. For example, the former governor of New Mexico, Toney Anaya, commuted the sentences of all inmates on death row as he concluded his final term in office. See Anaya Spares All Inmates on New Mexico Death Row, L.A. Times, Nov. 27, 1986, at A1, col. 1.

More recently, outgoing Ohio governor Richard Celeste commuted the sentences of 25 female prisoners who had been convicted of assaulting or killing their husbands or companions at trials in which they had been unable to offer evidence of battered-woman syndrome. 25 Women Granted State's Clemency, Columbus Dispatch, Dec. 22, 1990, at 1A, col. 1. Celeste issued a final wave of clemencies three days before leaving office when he commuted the death sentences of eight convicted killers, including all four women on death row. Columbus Dispatch, Jan. 11, 1991, at 1A, col. 5. On the same day, Celeste also commuted the prison term of country singer Johnny Paycheck, who had served nearly 2 years of a 7- to 9 1/2-year sentence for shooting a man in a bar. Columbus Dispatch, Jan. 11, 1991, at 5B, col. 1.

287. Other commentators have noted the need to develop standards to govern the exercise of the clemency power. See, e.g., Feerick, The Pardoning Power of Article II of the Constitution, 47 N.Y. St. B.J. 7, 44 (1975). Developing these standards has been left largely to philosophers such as Kathleen Moore, supra note 5, and largely ignored by those whose responsibility it is to exercise the power.

288. Moreover, those states that do not have constitutional provisions dictating the implementation of clemency may more easily refine the clemency process to ensure that it promotes the ends of justice.

289. See supra notes 168-76 and accompanying text.

290. See Brown v. Walker, 161 U.S. 591, 601 (1896) (stating that although the Constitution vests the power to pardon in the President, this does not take from Congress the power "to pass acts of general amnesty"); see also Duker, supra note 18, at 521-23 (discussing Brown and other Supreme Court cases holding that Congress has the power to pardon).

291. See, e.g., In re Briggs, 135 N.C. 118, 120, 47 S.E. 403, 404 (1904) (applying a North Carolina statute that provides a witness with absolute immunity from punishment for an offense in which the witness participated and is required to testify); State v. Morris, 55 Ohio St. 2d 101, 113, 378 N.E.2d 708, 715 (1978) (upholding the power of the Ohio General Assembly to create new sentencing guidelines under which previously convicted criminals could apply for reductions in sentences). See generally Survey, supra note 21, at 102-22 (discussing theoretical justifications for the legislature's power to grant clemency).

292. Possible problems could result from such a system, including questions as to what should happen when the two clemency procedures reach conflicting results in the same case.

293. See supra notes 177-206 and accompanying text.

294. See Schick v. Reed, 419 U.S. 256, 266 (1974).

295. See Knote v. United States, 95 U.S. 149, 154 (1877).

296. See Amar, A Neo-Federalist View of Article III: Separating the Two Tiers of Federal Jurisdiction, 65 B.U.L. Rev. 205, 226 n.81 (1985) (discussing the "structural independence" of the federal judiciary from political constraints).

297. Smith & Johnson, supra note 17, at 1129-30.

298. 390 F. Supp. 1372 (W.D. Mich. 1975).

299. Id. at 1374.

300. 267 U.S. 87 (1925).

301. Id. at 121.

302. This scenario would be analogous to the situation presented in Powell v. McCormack, 395 U.S. 486 (1969), in which the Court held that it could properly review whether Congress had followed the appropriate constitutional procedure in making what was otherwise an unreviewable political decision. Id. at 574-79. Similarly, if the executive were to exercise the clemency power in a manner that violated a provision of the Constitution, Powell supports the proposition that the Court could review and invalidate such action.

303. 236 U.S. 79 (1915).

304. Id. at 90-92.

305. The Supreme Court first applied the principle of equal protection to the federal government, through the due process clause of the Fifth Amendment, in Bolling v. Sharpe, 347 U.S. 497, 500 (1954). Although Bolling concerned the actions of Congress, the general language of the fifth-amendment due process clause indicates that all branches of the federal government, including the executive, must comply with equal protection principles in the exercise of their constitutional powers. Indeed, the Court has held that equal protection principles constrain the executive in the discharge of its duty to prosecute violators of the law. Wayte v. United States, 470 U.S. 598, 607-09 & n.9 (1985) (holding that although the executive has broad discretion as to whom to prosecute, a decision to selectively prosecute may not violate "ordinary equal protection standards"); United States v. Batchelder, 442 U.S. 114, 125 & n.9 (1978) (holding that selectivity in the enforcement of federal criminal laws is subject to equal protection constraints).

306. See supra note 306.

307. See generally Smith, The Prerogative of Mercy, the Power of Pardon and Criminal Justice, 1983 Pub. L. 398, 432.

308. See, e.g., United States v. Wilson, 32 U.S. (7 Pet.) 150, 160 (1833) ("As this power had been exercised from time immemorial by the executive of that nation whose language is our language, and to whose judicial institutions ours bear a close resemblance; we adopt their principles respecting the operation and effect of a pardon.").

309. See S.A. de Smith's Judicial Review of Administrative Action 286-89 (J.M. Evans ed. 1980) (noting that where executive action derives from the royal prerogative, judicial review has traditionally been very limited); Jones, Should Judges Be Politicians?: The English Experience, 57 Ind. L.J. 211, 211-27 (1982) (discussing the English courts' limited judicial review over other branches of government and comparing it with the broader power of judicial review exercised in American courts).

310. 369 U.S. 186 (1962). The Court noted that:

Prominent on the surface of any case held to involve a political question is found a textually demonstrable constitutional commitment of the issue to a coordinate political department; or a lack of judicially discoverable and manageable standards of resolving it; or the impossibility of deciding without an initial policy determination of a kind clearly for nonjudicial discretion; or the impossibility of a court's undertaking independent resolution without expressing lack of the respect due coordinate branches of government; or an unusual need for unquestioning adherence to a political decision already made; or the potentiality of embarrassment from multifarious pronouncements by various departments on one question.

Id. at 217.

311. See Heckler v. Chaney, 470 U.S. 821, 830 (1985) (indicating that an agency decision would be nonjusticiable if the statute "`committed' the decisionmaking to the agency's exercise of discretion" by failing to specify judicially manageable standards for review).

312. Interestingly, the district court in Murphy v. Ford, 390 F. Supp. 1372 (W.D. Mich. 1975), did not consider any of these justiciability arguments in addressing the legality of President Ford's pardon of Nixon, but instead proceeded straight to the merits.

313. See supra text accompanying notes 198-206.

314. See United States v. Richardson, 418 U.S. 166, 179-80 (1974) (holding that a claimant must have a "particular, concrete injury" and more than a generalized grievance in order to have standing); Schlesinger v. Reservists Committee to Stop the War, 418 U.S. 208, 224-28 (1974) (holding that a taxpayer must have a necessary personal stake in order to have standing).

315. H.R.J. Res. 542, 87 Stat. 555 (1973); see, e.g., Lowry v. Reagan, 676 F. Supp. 333 (D.D.C. 1987) (refusing to exercise jurisdiction over a complaint by 110 members of the House of Representatives seeking to enforce the War Powers Resolution).

316. 392 U.S. 83, 130-33 (1968) (Harlan, J., dissenting) (opining that review by the judiciary of congressional spending decisions under the establishment clause strains the judicial function and presses to the limit judicial authority).

317. INS v. Chadha, 462 U.S. 919, 946 (1983).

318. See supra notes 130-32 and accompanying text.

319. See Powell v. McCormack, 395 U.S. 486, 516-50 (1969) (reaffirming the responsibility of the Court to determine whether the action of a coordinate branch exceeds whatever authority has been committed to that branch by the Constitution); supra notes 294-306 and accompanying text.

320. Judicial review could also result in the judiciary's attempting to curtail the exercise of the executive clemency power. With regard to the federal system, however, the "structural judicial independence" of the courts, see Amar, supra note 296, at 226 n.80, would presumably render them less liable to usurp the clemency power for political reasons. In contrast, many state judiciaries lack life tenure, salary guarantees, and other safeguards against politicization found in the federal system and hence are no better suited than the executive to the task of evaluating clemency decisions.

321. For a discussion of the controversial pardon by Charles II of Thomas Osborne, the Earl of Danby, see supra notes 98-102 and accompanying text.

322. In the wake of the Nixon pardon, Senator Mondale proposed a constitutional amendment that would have created a legislative check on the presidential pardoning power when two-thirds of the members of each house disapproved the granting of the pardon within 180 days of its issuance. See S.J. Res. 241, 93d Cong., 2d Sess. (1974). The amendment was referred to the Senate Judiciary Committee, 120 Cong. Rec. 31551 (1974), from which it apparently never emerged.

323. See Smith & Johnson, supra note 17, at 1122.

324. See supra text accompanying notes 215-27.

325. See generally Saliterman, supra note 226 (containing the musings of an attorney who advised the Presidential Clemency Board on the Board's operation).

326. Id. at 261.

327. Of course, if this proposed commission were created by order of the executive, the executive would still retain the ultimate authority over clemency decisions, thereby obviating potential separation-of-powers objections to this proposal. Presumably, the commission's charge to decide whether clemency is deserved based on specific, justice-enhancing standards would make apparent that a contrary decision by the executive was justice-neutral in nature. See infra text accompanying notes 382-86 for a discussion of the appropriateness of the executive making justice-neutral clemency decisions.

However, if the commission were created by the Congress pursuant to its power to grant clemency, possible separation-of-powers issues arise. The superficial similarity of the commission's review to that of an appellate court, coupled with a formalistic notion of the limits imposed by article III, could render the proposed clemency commission vulnerable to attack on grounds similar to those employed by the Court in Northern Pipeline Constr. Co. v. Marathon Pipe Line Co., 458 U.S. 50 (1982). In Northern Pipeline, the Court held that the creation of bankruptcy courts with jurisdiction over all bankruptcy cases was unconstitutional because it removed the essential characteristics of the judicial power from an article III court to a non-article III adjunct. Id. at 76-87. Although consideration of these issues is beyond the scope of this Article, I believe that such an interpretation of article III is unwarranted, and as Professor Laurence Tribe suggests, "would threaten to hamstring, in the name of separation of powers, politically innovative schemes of the sort that may well be essential to the functioning of an ambitious government." L. Tribe, American Constitutional Law 3-5, at 54 (2d ed. 1988).

328. To prevent abuses, "good behavior" could be defined as a limitation on tenure to include the good faith application of the standards that the commission is charged with following. Thus, a particular commissioner who voted either for or against clemency with extreme regularity could be subject to removal for apparent adherence to a personal philosophy of clemency rather than the promulgated standards.

329. If other goals such as rehabilitation, as opposed to retribution alone, are determined by the legislature to be ends of the criminal justice system, the clemency board could be selected with a view to making determinations about the extent to which those goals have been achieved by the punishment inflicted. This Article analyzes the manner in which clemency should be refined to further the goal of retributive punishment. There is no reason, however, why a clemency board could not be constituted so as to evaluate other ends that punishment is intended to further.

330. M. DiSalle, supra note 250, at 212.

331. Such a commission would differ from the Office of the United States Pardon Attorney in several ways. Whereas the commission would bear the primary responsibility for justice-enhancing clemency decisions, the pardon attorneys act largely in an advisory capacity. The proposed commission would also comprise individuals with backgrounds and skills more diverse than those typically possessed by attorneys. Finally, unlike the Office of the United States Pardon Attorney, the commission would be governed by specific, defined standards regarding whether to remit punishment, as is discussed below.

332. Retributivism is of course not the only theory of punishment underlying our criminal justice system. To the extent that rehabilitation or other goals are also served by incarceration, it would be possible for a commission with the requisite expertise to consider these goals as well. Whichever body attempts to overhaul the clemency process would need to examine the ends properly served by punishment and promulgate clemency standards accordingly.

333. Conversely, considerations of "mercy" would be antithetical to this philosophy of justice and hence would be beyond the scope of the commission's charge. Thus, sympathy for the offender or the desire to reward service to the state, because they are irrelevant to what an individual deserves as punishment for the crime she committed, would not be reasons for granting clemency. See K. Moore, supra note 5, at 74-75. These considerations would continue to reside in the executive. See infra notes 382-86 and accompanying text.

334. State v. Apanovitch, 33 Ohio St. 3d 19, 23, 514 N.E.2d 394, 399 (1987) (per curiam).

335. Id.

336. Obviously, the commission could not order that there be a new trial because such an order would effectively usurp the judicial function. However, providing courts with the discretion to order a new trial whenever there has been a favorable recommendation from the clemency commission would provide an officially sanctioned means for deserving prisoners to obtain a new trial.

337. See A. Heller, supra note 55, at 170 (noting that if we are not free and rational actors who are the sole authors of our deeds, punishment is inherently unjust); see also K. Moore, supra note 5, at 138 (arguing that "anyone who cannot be blamed cannot be punished, and if blameworthiness is reduced, so must punishment be").

Obviously, notions of blameworthiness are not clear-cut. Retributive theorists suggest that we evaluate blameworthiness through a delicate weighing of the harm intended, risked, and achieved by the actor. See H. Hart, Punishment and Responsibility 234 (1968); A. Von Hirsch, supra note 72, at 69; Gale, supra note 59, at 1013-14. As Hart asks, however, "if the subjective wickedness of the criminal act is relevant, can human judges discover and make comparisons between the motives, temptations, opportunities and wickedness of different individuals?" H. Hart, supra, at 162. Yet that is precisely what judges and juries do when they pronounce judgment on and sentence offenders, despite the fact that "[d]eserving punishment is . . . inevitably indeterminate." Gale, supra note 59, at 1004. Thus, it is fitting that a clemency commission, consisting of conscientious individuals skilled in matters of human behavior, should likewise be trusted to assess the relative blameworthiness of individual conduct.

338. In addition to the factors relating to mental capacity, it may also be appropriate for clemency decisions to take into account sociological and cultural factors that pertain to blameworthiness. Retributive punishment presumes that society is structured relatively equitably with shared values and rules to which each person in the community, as an equal beneficiary of these rules, owes obedience. A criminal who rationally chooses to disobey these rules must in effect compensate society for the benefits he has derived from such a society.

However, as Marxist commentators have noted, to think that many criminals from the underclasses fit this model is to distort the true nature of our society--"to live in a world of social and political fantasy." Murphy, Marxism and Retribution, 2 Phil. & Pub. Aff. 217, 240 (1973). The vast majority of criminals come from the lowest 15% income level, suggesting that crimes are "normal, and not psychopathological, acts" which "grow out of need, greed, indifference to others, and sometimes even a sense of indignation--all, alas, perfectly typical human motives." Id. at 235. Thus, any system of punishment or punishment remission must arguably consider the sociological background of offenders when attempting to assess blameworthiness.

339. Inasmuch as one of the goals of incarceration is to protect society from dangerous individuals, the remission of punishment imposed on some mentally deranged individuals would not necessarily entail release. The commission could recommend that these individuals be given psychological help within the confines of a hospital or other facility.

340. See E. Brown & D. Adler, supra note 2, at 83.

341. See 60 Minutes, supra note 256, at 5.

342. See Louisiana Executes Man Who Killed at Age 17, supra note 256.

343. Songwriter Elvis Costello distilled his outrage at this form of injustice into the song Let Him Dangle, Spike, track 2 (Warner Brothers Records, 1989) (copyright 1988 Plangent Visions Inc.). The song recounts the story of Derek Bentley, who was hung in England for the crime of murder, while Chris Craig, his accomplice who actually pulled the trigger, received a prison term:

Bentley said to Craig, "Let him have it Chris."
They still don't know today just what he meant by this.
Craig fired the pistol but was too young to swing
So the police took Bentley and the very next thing,
Let him dangle
Let him dangle . . .
Not many people thought that Bentley would hang,
But the word never came, the phone never rang.
Outside Wandsworth Prison there was horror and hate,
As the hangman shook Bentley's hand to calculate his weight.
Let him dangle.

According to Costello, Bentley's sister is still seeking a postmortem pardon of her brother in recognition that a mistake was made. See Rowland, Elvis Costello in Love & War, Musician, Mar. 1989, at 62, 76.

344. For instance, former Indiana Governor James Goodrich wrote that the pardoning power should be used to dispel the notion that

the law does not apply with equal weight to all persons. . . . If the law fell with equal weight upon everyone, if high and low, rich and poor who are guilty of its infraction, alike suffered its penalty, there might be some reason for insisting that the guilty should in all cases pay the full penalty of the law; but too often in this state I have seen the law applied with unequal weight. . . .

Goodrich, 11 J. Crim. L., Criminology & Pol. Sci. 334, 342 (1920).

345. M. DiSalle, supra note 250, at 58 (quoting former Ohio governor Thomas Herbert).

346. John Dean spent four months in prison, Jeb Magruder and Charles Colson seven months, and John Mitchell, H.R. Haldeman, and John Ehrlichman one and a half years. N.Y. Times, Feb. 18, 1983, at A30, col. 1.

347. Krajick, supra note 207, at 53 (quoting Patrick Apodaca).

348. Pulley v. Harris, 465 U.S. 37, 45 (1984); see also Liebman, Appellate Review of Death Sentences: A Critique of Proportionality Review, 18 U.C. Davis L. Rev. 1433, 1433 (1985) (interpreting Pulley as a message to states that "as long as the proper [capital sentencing] procedures are in place, the Court will not scrutinize those procedures to determine if they actually work").

349. Governor Brown and his clemency secretary had agreed at the outset:

[T]he most important part of his job was to make sure that the death penalty was administered equally in all fifty-eight California counties. "What I plan to do is look at each case from the perspective of a Los Angeles district attorney in a busy office and ask myself, `Would this case merit a first-degree murder charge and a death penalty in L.A. County?'"

E. Brown & D. Adler, supra note 2, at 81-82.

350. Some states already take such factors into account. In South Carolina, consideration for clemency is to be given to any inmate with a terminal illness expected to live less than one year. See S.C. Code Ann. 24-21-970 (Law. Co-op. 1989).

351. See K. Moore, supra note 5, at 173-75 (noting that an elderly inmate's life sentence is brief when compared to that of a teenage murderer).

352. In a recent survey of state clemency practices, one of the most commonly reported emerging issues concerning the exercise of the clemency power involved the remission of punishment for prisoners with AIDS. See Guide to State Clemency, supra note 28, at 175.

New York governor Mario Cuomo, for the first time in his tenure as governor, recently granted clemency to Miguel Lopez, a prisoner who is dying of AIDS. N.Y. Times, Jan. 1, 1991, at A22, col. 4. The grant of clemency followed the New York Legislature's rejection of Cuomo's request for the creation of a "medical parole" program whereby dying prisoners would be eligible for parole. Columbus Dispatch, Jan. 2, 1991, at 10A, col. 2. Lopez is one of about 1000 inmates in New York's prison system who are suffering from AIDS, prompting questions from prisoners' rights advocates about why Cuomo hasn't released other dying prisoners. Id.

353. For an example of disparate sentencing based on gender, see Hutton, Pommersheim & Feimer, "I Fought the Law and the Law Won": A Report on Women and Disparate Sentencing in South Dakota, 15 New Eng. J. on Crim. & Civ. Confinement 177 (1989).

354. 408 U.S. 238 (1972).

355. S. Gross & R. Mauro, Death and Discrimination 109 (1989). The General Accounting Office has recently corroborated this phenomenon. In a study of more than two dozen research projects on capital sentencing, the GAO concluded that killers of whites are far more likely to receive the death sentence than killers of blacks. Columbus Dispatch, Mar. 30, 1990, at 8A, col. 1; see also Civiletti, Race and Punishment: Death by the Numbers, Texas Lawyer, June 11, 1990, at 37 (citing several studies on disparate capital sentencing based on the race of the victim).

356. Columbus Dispatch, Jan. 11, 1991, at 1A, col. 5. All 4 women and 54 of the 101 men on death row prior to the commutations are black. Id. at 1A, col. 6. Celeste said that while these statistics alone do not "prove racism," they indicate the need "for further study of Ohio's system of rendering decisions in capital cases." Id. at 2A, col. 1.

357. A recent survey reports that public support for the death penalty remains at 72%, close to the record highs reached at the end of the 1988 presidential campaign. See Oreskes, The Political Stampede on Execution, N.Y. Times, Apr. 4, 1990, at A16, col. 1. In light of this popular support for capital punishment, one political commentator has noted that "[y]ou cannot be against the death penalty and survive a campaign for major office" in some states. Id. at A16, col. 2. Consistent with this observation, former Ohio governor Celeste has been harshly criticized for the death penalty commutations discussed above, which were issued at the close of his term. For example, the Columbus Dispatch editorialized that by commuting the death sentences to life imprisonment without possibility of parole,

Celeste, in effect is thumbing his nose at the majority, who support capital punishment for heinous crimes. As an avowed opponent of capital punishment, the departing governor has abused flagrantly the authority of his public office in order to impose his private vision of justice. It is a sad and sorry performance, an exercise of arrogance, if not outright contempt.

Killers Spared, Columbus Dispatch, Jan. 14, 1991, at 8A, col. 1.

Celeste's eleventh-hour commutations have spawned not only harsh commentary, but also attempts by Ohio politicians who favor the death penalty to invalidate the acts of clemency. Incoming Ohio governor George Voinovich, a supporter of capital punishment, has vowed to nullify the commutations. Columbus Dispatch, Jan. 19, 1991, at 1B, col. 1.

Surprisingly, Ohio Attorney General Lee Fisher, a liberal Cleveland Democrat like Celeste, has sided with the Republican Voinovich in attempting to have seven of the eight commutations set aside and the death sentences reinstated. Fisher, who until recently opposed capital punishment, has sought a judicial declaration that the commutations are invalid because Celeste failed to file applications for clemency with the Ohio Adult Parole Authority, which is charged under Ohio law with investigating all clemency applications and making a nonbinding recommendation to the governor. Columbus Dispatch, Jan. 30, 1991, at 1A, col. 4. Fisher's position that the commutations are invalid is based on a provision in the Ohio Constitution making the governor's clemency power "subject, however, to such regulations, as to the manner of applying for pardons, as may be prescribed by law." Ohio Const. art. III, 11. Ohio political insiders described his decision to challenge the commutations as "smart" in light of Celeste's lingering unpopularity and the fact that Fisher's own recent election as attorney general is being challenged in the Ohio Supreme Court by his Republican opponent. Johnson, Fisher Starts with `Smart' Decisions, Columbus Dispatch, Feb. 1, 1991, at 9A, col. 1. One Republican official is reported to have remarked that the decision shows that Fisher is "going to be on A-plus good behavior as long as this election [which was decided by only 1234 votes out of 3.3 million cast] is in doubt." Id. (quoting "a top GOP official").

358. On the eve of his departure from office, Ohio governor Richard Celeste commuted the sentences of 25 female prisoners who had killed or assaulted their husbands or male companions because of the psychological condition known as the battered-woman syndrome. According to Celeste, the women whose sentences he commuted had suffered the "`profound and terrifying experience' of being victims of `physical and psychological violence and intimidation.'" 25 Women Granted State's Clemency, supra note 286 (quoting Governor Celeste).

Celeste's clemency decision can be justified retributively both because of the diminished blameworthiness of the offenders and as a means of rectifying disparities in punishment. Recent changes in Ohio law now permit the introduction of evidence of battered-woman syndrome in proving self-defense. Id. Many of the commutees may well have been acquitted had they been able to present a battered-woman syndrome defense. Clemency Decision Stirs Controversy, Columbus Dispatch, Dec. 23, 1990, at 1A, col. 1, 2A, col. 2. However, recent investigations into the cases of the 25 women given clemency by Celeste raise questions about whether each of the women actually had been battered. Yocum, Women's Clemency Angers Prosecutors, Columbus Dispatch, Jan. 27, 1991, at 5F, col. 1.

359. See K. Moore, supra note 5, at 153-54; Smart, Mercy, 43 Philosophy 345, 348 (1968).

360. For instance, under Ohio Rev. Code Ann. 2951.02 (Anderson Supp. 1989), a court is permitted, but not required, to consider "substantial grounds tending to excuse or justify the offense, though failing to establish a defense."

361. There are recent indications that the federal courts are interpreting the sentencing guidelines inconsistently, with several circuits allowing sentencing judges wide discretion to ignore the guidelines, particularly to increase the sentences of offenders, while other circuits limit departures from the guidelines more stringently. See generally Jackson, Departure from the Guidelines: The Frolic and Detour of the Circuits--How the Circuit Courts Are Undermining the Purposes of the Federal Sentencing Guidelines, 94 Dick. L. Rev. 605, 613-31 (1990). Thus, while the guidelines may not be eliminating disparities in sentencing, they may result in some courts giving less weight to mitigating sentencing factors.

362. See K. Moore, supra note 5, at 155-65.

363. Id. at 161.

364. This diversity in the composition of the commission also has a potential disadvantage. If all interests and philosophies are represented equally, it could conceivably lead to the commission's acceptance of only those notions of blameworthiness that appeal to the majority of the members--a sort of least common denominator approach that might limit the cases in which clemency would be granted. This is a risk that seems reasonable, however, in light of the apolitical nature of the commission and the need to have offenders' blameworthiness assessed by persons who share some of the same values or beliefs.

365. See Smart, supra note 359.

366. See Survey, supra note 21, at 85.

367. See K. Moore, supra note 5, at 224 (noting the injustice of continuing to deprive individuals of their civil rights after they have served their time). However, even a presidential pardon may not be recognized in some states as restoring civil rights. See Houle, Forgive and Forget: Honoring Full and Unconditional Pardons, 41 Me. L. Rev. 273, 274, 283 (1989).

368. See Dunn v. Maggio, 712 F.2d 998, 1001-02 (5th Cir. 1983) (per curiam) (holding that a statute which provided that the "governor may commute the sentence" of a prisoner did not "provide a prisoner with a constitutional right to a pardon or early release from a valid sentence"), cert. denied, 465 U.S. 1031 (1984); Williams v. Briscoe, 641 F.2d 274, 275-77 (5th Cir. 1981) (holding that the language of an early release statute "cannot reasonably be taken to encourage the expectancy of the right to release" so as to create "a constitutionally protectible entitlement" of release), cert. denied, 454 U.S. 854 (1981); Dumschat v. Board of Pardons, 618 F.2d 216, 218-22 (2d Cir. 1980) (per curiam) (arguing that because pardon is discretionary, inmates must show a legitimate expectation of pardon in order for procedural due process protections to attach).

369. See Greenholtz v. Inmates of the Nebraska Penal and Correctional Complex, 442 U.S. 1, 13-16 (1979).

370. Id. at 15.

371. Id. at 5; id. at 21 (Powell, J., concurring & dissenting).

372. Id. at 4.

373. Dumschat v. Board of Pardons, 618 F.2d 216, 222 (2d Cir. 1980) (per curiam) (requiring that inmates who have a legitimate expectation of pardon and are denied relief be provided with "short statements of reasons explaining why they have been denied relief").

Arizona provides due process protections to consideration of clemency applications in the form of notice and an opportunity to be heard, which may be compelled by an action in mandamus. See McGee v. Arizona State Bd. of Pardons and Paroles, 92 Ariz. 317, 320, 376 P.2d 779, 781 (1962). However, a requirement that applicants be provided with reasons for denial of commutations has been held to be outside the scope of due process. Banks v. Arizona State Bd. of Pardons and Paroles, 129 Ariz. 199, 202-03, 629 P.2d 1035, 1038-39 (Ct. App. 1981); see also Ill. Ann. Stat. ch. 38, para. 1003-3-13(c) (Smith-Hurd Supp. 1990) (requiring the Prisoner Review Board, if requested and upon due notice, to give each application a hearing at which representation by counsel is allowed).

374. The importance of capable counsel participating in all phases of capital cases is underscored by the recent National Law Journal study which concluded that ineffective representation occurs in the majority of trials resulting in the imposition of the death penalty. See Coyle, Strasser & Lavelle, supra note 280, at 30-32.

375. Note, A Matter of Life and Death: Due Process Protection in Capital Clemency Proceedings, 90 Yale L.J. 889, 901 (1981).

376. See Kaplan, Not an Innocent; Hammer's Pardon Fuels his Illusions, Legal Times, Aug. 28, 1989, at 1 (noting that the President often uses FBI background checks and generally follows the FBI's recommendations).

377. A recent survey of state clemency practices found that Georgia and Arkansas had 262 and 55 investigatory employees working full-time in their respective clemency departments. The next highest number of full-time investigatory employees was five, reported by North Dakota. California had only six part-time investigatory employees, and most of the states that responded to the survey had no more than one. See Guide to State Clemency, supra note 28, at 174 table 10 (Clemency Staffing Arrangements and Salary Levels).

378. See Greene, America's Prisons: Who's Punishing Whom?, Forbes, Mar. 21, 1988, at 132.

379. Georgia reported by far the highest budget for clemency of any state. In 1987, Georgia allocated $2.6 million for its Board of Pardons and Paroles. Guide to State Clemency, supra note 28, at 173 table 9 (States' Average Annual Clemency Budgets).

380. See Krajick, supra note 207, at 49. In a similar effort, law students at Tulane University are using clemency and parole to get elderly Louisiana prisoners who are serving inordinately harsh sentences released. The Chronicle of Higher Education, Apr. 4, 1990, at A37, col. 1.

381. Ex parte Wells, 59 U.S. (18 How.) 307, 310 (1855).

382. See generally supra notes 125-40 and accompanying text.

383. K. Moore, supra note 5, at 200-01.

384. See id. at 217 (noting that historically, "[p]ardons are potentially too dangerous, too destructive of trust and justice, to be left to a President's whim").

385. This aspect of my proposal is aimed primarily at state systems of clemency, because the federal constitution would have to be amended before a legislative veto provision could be implemented. See INS v. Chadha, 462 U.S. 919, 944-59 (1983). Notwithstanding my conviction that federal clemency procedures tend to ignore the justice-enhancing goals of the Framers, recent efforts to amend the First Amendment and weaken its guarantees of free expression make me reluctant to urge constitutional amendments until all other solutions have been tried. See Ginsberg, On Amending the Constitution: A Plea for Patience, 12 U. Ark. Little Rock L.J. 677 (1990). Thus, I would urge the President to adopt by executive order the clemency procedures discussed above.

386. Duker, supra note 18, at 537 (quoting The Federalist No. 47, at 449 (A. Hamilton) (J. Cooke ed. 1961)).

387. Clemency, which has often been deemed a sufficient substitute for every other virtue in sovereigns, should be excluded in a perfect legislation, where punishments are mild, and the proceedings in criminal cases regular and expeditious. This truth will seem cruel to those who live in countries where, from the absurdity of the laws and the severity of punishments, pardons and the clemency of the prince are necessary.

C. Beccaria, supra note 13, at 158

388. W. Shakespeare, The Merchant of Venice, supra note 12, at 501.