Separation of powers and the criminal law.
INTRODUCTION I. THE CURRENT APPROACH TO SEPARATION OF POWERS A. The Separation of Powers in Criminal Cases 1. The independent counsel 2. The Sentencing Commission 3. Enemy combatants B. Criminal and Administrative Law Cases Compared II. CRIME IS DIFFERENT A. Constitutional Foundations B. Functional Differences 1. Lack of structural and procedural safeguards 2. Lack of political safeguards C. Individual Rights Protections Are Insufficient III. ENFORCING SEPARATION OF POWERS IN CRIMINAL MATTERS A. The Case for Greater Enforcement B. The Independent Counsel C. Sentencing D. Prosecutorial Discretion and Plea Bargaining E. Formalism and Federalism CONCLUSION
It is a familiar premise that the Constitution separates legislative, executive, and judicial power to prevent tyranny and protect liberty. (1) By preventing any one branch from accumulating too much authority, the separation of powers aims "not to promote efficiency but to preclude the exercise of arbitrary power." (2) The price of separation is that it makes it more difficult for the federal government to act--whether for good or bad purposes. (3)
The rise of the administrative state put a spotlight on this cost of the separation of powers. New Dealers in favor of a more efficient and active federal government argued for a relaxation of the division of powers to allow agencies to combine government functions in order to address social and economic ills. (4) Instead of relying on separated powers as the primary means of protection against government abuse, they proposed other checks on state power. For example, the Administrative Procedure Act (APA) requirements of notice and comment, (5) of separation between law enforcers and adjudicators, (6) and of judicial review (7) were designed to perform the same functions as the Constitution's separation of powers safeguards, but without hamstringing the government's ability to respond rapidly to the nation's problems.
The Supreme Court has accepted this compromise for administrative agencies. While the Court has rejected some institutional arrangements that strayed too far from the constitutional separation of powers, (8) it has allowed considerable blending of executive, judicial, and legislative power in regulatory agencies. (9) At the same time, the Court has taken an expansive reading of the APA to check government abuse. (10)
Scholars have filled volumes analyzing the relationship between the separation of powers and the administrative state. (11) Some have argued that the Court's allowance of blending promotes good government and accords with the Constitution. (12) Others have claimed that the existing administrative state flouts the basic structure of the Constitution and that the Court has been too permissive of government schemes that combine powers. (13)
What has been completely overlooked in both the scholarly literature and the Supreme Court's decisions is what the separation of powers requires when the government proceeds in a criminal action. Criminal cases could be viewed in one of three ways. One approach would be to treat separation of powers questions in criminal cases no differently than they are treated in administrative law cases. Just as in the administrative law context, some blending of powers would be permitted to allow the federal government to respond more readily to criminal matters. At the same time, and again following the administrative law model, other checks should take the place of the constitutional separation of powers to ensure that the government does not abuse its power.
A second alternative would be to distinguish criminal matters from administrative ones. Because state power is at its apex in the criminal context and the consequences of abuse are so high--an individual could lose his or her liberty or even life--this view would require strict adherence to separation of powers to make sure that the state acts appropriately against an individual. Under this approach, then, the need for government flexibility and expediency may justify blending when the government proceeds civilly but not when it proceeds in a criminal action.
Current law follows a third way. Criminal cases are not distinguished from administrative law cases, so the separation of powers is often relaxed to allow a blending of powers when the government claims it is necessary in the name of expediency. (14) Indeed, the Court has been even more permissive in the criminal context than it has in cases involving nonpenal laws. (15) But unlike the administrative law context, where agencies must adhere to the structural and process protections of the APA and their decisions are subject to judicial review, the government faces almost no institutional checks when it proceeds in criminal matters. The only safeguards come from the individual rights provisions of the Constitution, but those checks act as poor safeguards against structural abuses and inequities.
The current arrangement therefore takes the worst possible approach to separation of powers in the criminal law. The protection provided by the separation of powers is relaxed, but nothing takes its place. As a result, the potential for government abuse is, ironically, higher in the criminal context than in other regulatory spheres.
This perverse state of affairs has been overlooked in the literature because scholars have failed to treat criminal law as a separate category for analysis. Instead, questions involving the oversight of the administrative and regulatory state have tended to dominate the discussion of separation of powers. (16) So, the conventional wisdom has been that whatever theory works for the administrative state should work for anything else, too. And because most scholars have supported a flexible or functional approach that allows the concentration of different powers in one actor in the regulatory sphere, they have failed to see a problem with that same approach when it is applied to criminal matters. (17)
This Article breaks from that tradition and argues that the existing approach to separation of powers in criminal matters cannot be squared with constitutional theory or sound institutional design. Although the administrative state has structural and process protections that can justify some flexibility in the separation of powers, those checks are absent in the criminal context. And in their absence, it is critically important to maintain a strict division of powers.
The Article proceeds in three parts. Part I explores the Supreme Court's treatment of separation of powers claims, with particular emphasis on the criminal cases. As Part I explains, the Court does not employ a stricter test of separation of powers for criminal law cases than it does for administrative law cases. Just the opposite, the Court has allowed a greater relaxation of the separation of powers in its criminal cases.
Part II critiques the existing approach to separation of powers. As Part II explains, there are two key arguments for being more vigilant in protecting the separation of powers when the state proceeds in a criminal action. First, as a matter of traditional constitutional interpretation, a strict separation of powers in criminal law matters has a stronger textual and historical pedigree than in other contexts. The Constitution explicitly confronts the dangers of an abusive state in the context of criminal proceedings in several textual provisions that reflect a strict division of authority among the three branches and that give each branch, a strong check on the others in criminal proceedings. (18) Indeed, convictions require all three branches of government to agree, as well as the approval of a jury. In contrast, most other questions of separation of powers arise in administrative law contexts that the Constitution does not explicitly address. (19) Similarly, while the Framers did not confront the question of how to divide and balance government functions in light of the rapid expansion during the Industrial Revolution and the rise of the administrative state--the foundational premise for most functional theories of separation of powers that allow a blending of functions to create a more efficient government--they did have experience with the state's use and abuse of the criminal laws. Indeed, questions of state criminal power occupy a great deal of the Constitution's structure precisely because concentrated power in criminal matters was a danger of which the Framers were well aware. They feared the tyranny of majorities that would seek to oppress opponents through the use of criminal laws. They therefore established a constitutional structure that separates power among the branches and gives the judiciary (judges and juries) a particularly strong role in enforcing that separation. The individual rights protection provided by the separation of powers (20) has no greater purchase than in cases involving criminal defendants, for those were precisely the instances of abuse at the forefront of the minds of the Framing generation. So while there are strong arguments for accommodating the Constitution to changing circumstances in the case of unanticipated administrative law questions, those arguments are not as strong when it comes to matters of criminal law.
The second reason for maintaining a strict separation of powers when the federal government uses its criminal power, as Part II explains, rests on functional concerns. The state poses no greater threat to individual liberty than when it proceeds in a criminal action. Those proceedings, after all, are the means by which the state assumes the power to remove liberty and even life. Yet there are currently almost no institutional checks on federal criminal power. First, federal prosecutors face no restrictions on their powers that are comparable to the complex code of conduct and organizational design established by the APA. The federal agency responsible for setting federal sentences, the United States Sentencing Commission, likewise differs from virtually all other agencies because its rules cannot be challenged as arbitrary and capricious under the APA. In addition, the political controls over governmental crime policy are not as balanced as they are in the regulatory sphere. Those accused of crimes are among the most politically anemic groups in the legislative process. Criminal defendants do not coalesce into an organized group, and those individuals and organizations that represent their interests tend to be disorganized and weak political forces. In contrast, powerful interests often lobby for more punitive laws. The executive branch in particular has an incentive to push for tough laws to encourage plea bargaining and cooperation. The politics of crime definition and sentencing are therefore far more lopsided than the politics associated with the administrative state, where it is more common to have groups on both sides of the issue that act to check government abuse of power. (21) Thus, in the very area in which state power is most threatening--where it can lock away someone for years and impose the stigma of criminal punishment--institutional protections are currently at their weakest. While the numerous trial protections for criminal defendants aim to protect the interests of individuals, Part II explains why those rights-based protections do little to control the systemic abuse that the separation of powers regulates.
In light of these fundamental differences between criminal and administrative matters, Part III advocates a stricter test for policing separation of powers claims in criminal cases than the functional analysis currently employed by the Court. In the literature on separation of powers, this is typically referred to as a "formalist" approach to separation of powers, where legislative, executive, and judicial powers are to be separated and novel arrangements that allow a blending of functions or a weakening of one branch' s power are disallowed. This is typically contrasted with the "functional" approach, which allows a case-by-case inquiry to see if the particular relaxation of separation of powers in a given case will result in an inappropriate aggrandizement of one branch's power over another.
The view advanced here is not formalism for the sake of formalism, however. Rather, as Part III explains, the argument for strict enforcement of the separation of powers in criminal matters is grounded in functional reasons. (22) The problem with an approach to separation of powers that looks at the facts and circumstances of each case is that the long-term and systemic effects caused by blurring the lines of authority in criminal cases might not be immediately apparent, whereas the government's need for a more streamlined process is obvious, particularly to judges who directly bear the burden of more cases on their criminal docket. And, because they retain oversight of criminal trials, judges might not think structural protections are necessary. Moreover, because courts analyze separation of powers questions in regulatory contexts where efficiency arguments often trump other claims, that general attitude has spillover effects in criminal cases. Courts have become accustomed to blending arrangements and overlook the key differences between administrative and criminal matters described in Part II. Thus, a bright-line rule that disallows blending in criminal matters makes the most sense to ensure that the separation of powers and the liberty interests it protects are not undervalued.
Part III concludes with an overview of how this approach to separation of powers would change the existing view of criminal law. First, it would lead to different outcomes in the Court's criminal cases that raise separation of powers questions. It would disallow the independent counsel, and it would demonstrate the constitutional danger of having mandatory sentencing laws and guidelines applied by judges instead of juries. Second, greater enforcement of the separation of powers would call into question those laws and practices that undermine the judiciary's role in the separation of powers. Most significantly for day-to-day criminal justice practice, greater respect for the separation of powers would require a rethinking of the current approach to prosecutorial discretion over charging and plea decisions. Although a full analysis of the separation of powers problems posed by prosecutorial discretion and plea bargaining is beyond the scope of this Article, Part III explains the fundamental difficulty that this situation presents. Specifically, under current law, prosecutors are allowed to put a price on the defendant's exercise of the judicial check, which is a key element in the separation of powers, as Part II explains. Trials come at a high cost for individual defendants--prosecutors threaten longer sentences--and most individual defendants (even some who are innocent) are unwilling to take the chance of losing at trial. As a result, there is a systemic failing in which prosecutors make the key decisions in criminal matters without a judicial check and without any of the structural and procedural protections that govern other executive agencies. (23) It is precisely this kind of unchecked power that the separation of powers is designed to guard against. Finally, Part III concludes with some observations of how more stringent enforcement of separation of powers would serve the interests of federalism.
I. THE CURRENT APPROACH TO SEPARATION OF POWERS
The Supreme Court approaches separation of powers questions with little regard for the substantive contexts in which they arise. Instead, separation of powers cases can be categorized by the methodology that the Court employs to decide the case. Formalism and functionalism are shorthand labels for the two main interpretive strategies employed by the Court in its separation of powers cases. (24)
The formalist approach to separation of powers is characterized by the use of bright-line rules designed to keep each branch within its sphere of power. (25) Under this rationale, legislative power must rest with the legislative branch, executive power with the executive branch, and judicial power with the judicial branch. This methodology therefore requires the Court to characterize what type of power is at issue and to ensure that the power is being exercised by the correct branch of government and in compliance with any constitutional requirements for that type of power. If a governmental arrangement fails this test, it will not be upheld, even if it is an efficient or convenient solution to a public policy problem. (26)
The Court's decision in Chadha provides an illustration of formalist analysis. The law at issue in Chadha allowed one House of Congress to veto an agency decision that would allow a deportable alien to remain in the United States. The Court first concluded that the veto power constituted legislative power because it "had the purpose and effect of altering legal rights, duties, and relations of persons." (27) The Court then reasoned that, because the Constitution requires legislative power to be exercised according to the bicameralism and presentment requirements of Article I, (28) the one-House veto was unconstitutional. (29)
The Court used the same formalist methodology in several other cases. In Bowsher, the Court considered legislation that vested power in the Comptroller General to alter the federal budget in order to meet deficit-reduction quotas. (30) The Court determined that the law was unconstitutional because the Comptroller General was exercising executive power but was deemed to be an agent of Congress due to the fact that the position was subject to legislative removal. (31) The alleged benefits the legislation would bring in terms of a balanced budget were irrelevant to the Court's formalist analysis. The Court was no more sympathetic to these efficiency claims when they arose in the context of legislation that gave the President the power to exercise a line-item veto over spending and tax benefits, as the Court in Clinton v. New York once again applied a formalist analysis. (32) The Court similarly used a formalist analysis in concluding that a review board composed of members of Congress that had veto power over a local airport authority violated the separation of powers. (33) In that case, the Court reiterated that if a law violates the separation of powers, it does not matter that the law may nonetheless promote a "workable government." (34)
The Court's acceptance of modern administrative agencies poses the greatest challenge to the formalist approach to separation of powers. It is hard to square, for example, the Court's acceptance of agency rulemaking pursuant to a broad statutory delegation (35) with the pronouncement in Chadha that all legislative power must go through the procedures in Article I. (36) Agency rulemaking "alter[s] the legal rights, duties, and relations of persons," (37) but it does not need to pass both houses of Congress and obtain the President's approval. And because agencies exercise judicial and executive power as well, (38) they combine all three types of power in one actor (39)--the very danger the separation of powers aims to avoid. The Court has shown no indication that it intends to overthrow the administrative state, so it has sidestepped the agency challenge to the separation of powers even in cases employing formalist analysis. For example, it distinguished agency rulemaking from the legislative veto in Chadha by noting that agency rulemaking is merely "quasi-legislative" and is limited by the statute from which the agency derives its authority. (40) To take another example, in Northern Pipeline Construction Co., a case in which a plurality of the Court limited agency adjudication over private rights of action, the Court was nevertheless careful to recognize that agency adjudication of public rights remained permissible. (41) Thus, when formalist analysis is employed, the Court distinguishes prototypical federal administrative agencies without acknowledging that these agencies would fail under the same reasoning.
In further recognition and acceptance of the administrative state, the Court does not always employ a formalist methodology in its separation of powers cases. Instead, the Court will engage in a functional analysis that allows some mixing of power among the branches so long as one branch does not aggrandize its power at the expense of another (42) or otherwise impede a branch from performing its core responsibilities. (43) The Court's opinion in Schor provides an example of this methodology. Schor involved a separation of powers challenge to the Commodity Exchange Act, which allowed the Commodity Future Trading Commission to adjudicate customer claims that a broker had violated the terms of the Act and broker counterclaims (including claims under state law) arising out of the same transaction. (44) The Supreme Court rejected the formalist argument that jurisdiction over state law claims was judicial power that had to be exercised by an Article III court. The Court "declined to adopt formalistic and unbending rules" because "[a]lthough such rules might lend a greater degree of coherence to this area of the law, they might also unduly constrict Congress' ability to take needed and innovative action pursuant to its Article I powers." (45) Instead, the Court adopted a balancing test, in which it weighed a number of factors to determine "the practical effect that the congressional action will have on the constitutionally assigned role of the federal judiciary." (46) The Court upheld the legislation at issue in Schor because it concluded that the agency's jurisdiction over the counterclaims did not impermissibly intrude on the power of Article III courts.
Although the actual application of the functional test is context-specific, the test itself is invoked as if it applies to all settings. (47) When the Court uses a formalist approach in a separation of powers case, it similarly does not claim that it is employing formalism because a particular context or setting demands it. Instead, the Court simply employs that type of reasoning with little or no explanation. (48) The Court thus uses the methodologies interchangeably and does not claim to alter its approach depending on the substantive context.
Although the Court has not stated that a particular approach to separation of powers makes sense in the context of criminal cases, in the few cases in which such claims have come up, (49) the Court rejected formalist arguments and employed a functional analysis. Part I.A will discuss in more detail the Court's methodology in the major criminal law cases raising separation of powers claims, and Part I.B will consider why the Court employed a functional analysis in all of the criminal cases while applying a formalist methodology in most of its civil regulatory cases.
A. The Separation of Powers in Criminal Cases
Of the Supreme Court's major separation of powers cases, two involved criminal proceedings. (50) This Part will discuss both of those cases, Morrison v. Olsen (51) and Mistretta v. United States, (52) and also a recent case, Hamdi v. Rumsfeld, (53) which touched on related issues. In each of these cases, the Court has employed a flexible analysis to allow great blending of government power.
1. The independent counsel
In the wake of Watergate, Congress passed the Ethics in Government Act of 1978, which authorized the appointment of an impartial, independent counsel to investigate and prosecute criminal conduct by high-ranking government officials. (54) Pursuant to the Act, if the Attorney General received information that was "sufficient to constitute grounds to investigate" whether a person covered by the Act committed a federal crime, the Attorney General had to conduct a preliminary investigation. (55) The Attorney General was then obligated to apply to a special court, known as the Special Division, for the appointment of an independent counsel unless the Attorney General's investigation yielded "no reasonable grounds to believe that further investigation or prosecution is warranted." (56) The Act also gave congressional committee members the authority to "request in writing that the Attorney General apply for the appointment of an independent counsel." (57) Although the Attorney General was not obligated under the Act to file an application upon receiving such a request, the Attorney General was required to respond to the committee members' request within a specified time limit. (58)
Once appointed, the independent counsel remained in office until he or she completed the investigation or prosecution (59) unless the Attorney General could show that the counsel should be removed for "good cause, physical or mental disability ... or any other condition that substantially impairs the performance of such independent counsel's duties." (60) Although the good cause provision limited the executive branch's control over the independent counsel, the Act employed other mechanisms to create accountability for the independent counsel's actions. Congressional committees were given jurisdiction to oversee the conduct of the independent counsel, (61) and the counsel was required to inform the House of Representative of "substantial and credible information ... that may constitute grounds for impeachment." (62)
If the Court employed the same formalist methodology it used in Bowsher, it would have found the independent counsel law unconstitutional. Because investigating and prosecuting crimes are executive powers, (63) Congress's restriction of the President's removal authority would be unlawful under formalist reasoning. Indeed, that was the governing precedent at the time of Morrison. Under the Court's decisions in Myers (64) and Humphrey's Executor, (65) Congress could not place restrictions on the President's power to remove purely executive officials. The Court, however, rejected this approach and relaxed the separation of powers standard in the context of a criminal case. It essentially overruled the test established by Myers and Humphrey's Executor (66) and distinguished Bowsher as involving an attempt by Congress to aggrandize its own power by gaining a role in removing an executive official. (67) The Court opted instead for a balancing test that asked "whether the removal restrictions are of such a nature that they impede the President's ability to perform his constitutional duty." (68) Applying that test, the Court "simply [did] not see how the President's need to control the exercise of [the independent counsel's] discretion is so central to the functioning of the Executive Branch as to require as a matter of constitutional law that the counsel be terminable at will by the President." (69)
Justice Scalia alone dissented, largely on formalist grounds. He said the Act was unconstitutional because criminal prosecutions and investigations constituted the exercise of purely executive power, and the Act deprived the President "of exclusive control over the exercise of that power." (70) He rejected the Court's view that it was up to the Court to decide "how much of the purely executive powers of government must be within the full control of the President." (71)
But Justice Scalia also offered some functional reasons for his formalist conclusion. He believed that the threshold for requiring appointment of an independent counsel was inappropriately low. In his view, it would be a rare case in which there were no reasonable grounds to believe further investigation would be warranted, particularly if members of Congress assembled facts in favor of further investigation. (72) The Act created further undue political pressure for the appointment of an independent counsel because of the Act's provisions for congressional oversight. (73) "[H]ow easy it is," he observed, "for one of the President's political foes outside of Congress ... simply to trigger a debilitating criminal investigation of the Chief Executive under this law." (74) Without executive accountability, the Act removed the political check against prosecutorial abuse. (75) It also removed institutional checks against prosecutorial abuse, such as having a prosecutor with the perspective of other cases and targets to ensure that the laws are fairly applied. (76) Justice Scalia also offered a functional reason for rejecting functional tests for separation of powers claims as a general matter. He argued that the Court's balancing test was not a justiciable standard and that it lacked predictability. (77)
The Court dismissed the functional arguments raised by Justice Scalia, presumably because they may have seemed too remote and looked like nothing more than guesses about how bad political actors might use this new tool. In contrast, the value of the independent counsel statute was readily apparent--the statute was designed to make prosecuting corrupt government officials easier and to eliminate conflicts of interest. Weighing the removal of corrupt government officials against vague notions of political abuse may have made it seem like an easy case to most members of the Court.
2. The Sentencing Commission
In Mistretta v. United States, (78) the Court faced the question of whether Congress could delegate the authority to establish sentencing laws to an independent commission ostensibly housed in the judicial branch and containing federal judges as members. (79) The Court saw no trouble with the delegation, emphasizing that delegation jurisprudence "has been driven by a practical understanding that in our increasingly complex society, replete with ever changing and more technical problems, Congress simply cannot do its job absent an ability to delegate power under broad general directives." (80) It was, in the Court's view, "especially appropriate" that Congress delegated the "intricate, labor-intensive task" of establishing guidelines to the Commission. (81) Nor was the Court troubled by the fact that an agency placed in the judicial branch would engage in "significantly political" activity. (82) The Court emphasized that its "separation-of-powers analysis does not turn on the labeling of an activity" but instead focuses on the "practical consequences" of the government scheme at issue. (83) Thus, it did not matter to the Court that a judicial branch agency would be exercising rulemaking power. (84)
Justice Scalia again provided the lone dissent. Although he agreed with the majority that the Act provided an intelligible principle for purposes of the delegation doctrine, he objected to the delegation of lawmaking authority to an agency "created by Congress to exercise no governmental power other than the making of laws." (85) He relied on formalist reasoning, arguing that while agencies could engage in rulemaking as an incidental power to their enforcement responsibilities and courts could do so as part of their judicial functions, allowing an agency solely to pass rules would cross the constitutional line because there is no pretense that such an agency is doing anything other than exercising legislative power. He reasoned that the Court accepted the delegation of lawmaking power because it "inheres in most executive or judicial action," not because Congress is permitted to assign its responsibilities to someone else. (86) Because the Sentencing Commission had no other responsibilities that altered legal rights other than making rules, Justice Scalia thought it amounted to a "junior-varsity Congress" that did not fit anywhere within the three-branch structure of the Constitution. (87)
Aside from another attack on the Court's use of balancing tests, (88) Justice Scalia did not offer much in the way of functional arguments. And the parties' briefs likewise fell short on instrumental analysis. Faced with abstract notions of separated powers for the sake of separated powers, it is perhaps not surprising that most members of the Court sided with the government's claimed need for a commission to remedy the "shameful" disparity and uncertainty that plagued the federal system's prior sentencing scheme. (89)
3. Enemy combatants
The case of Hamdi v. Rumsfeld required the Court to decide what process is due to an American citizen who is captured on a foreign battlefield and who has been detained by the government as an enemy combatant. (90) It is not, strictly speaking, a criminal case because the government was asserting its war powers, not its criminal powers. But if the Court held that the President lacked the authority to label a citizen an enemy combatant and to detain Hamdi on that basis, the executive branch would have had to use its criminal powers or Congress would have had to suspend the writ of habeas corpus to detain him. Thus, the case explored the line between the President's authority during wartime and the Constitution's criminal protections, as well as the scope of executive and judicial power to determine that line. In particular, it raised the question of whether executive power includes the power to determine whether an individual is an enemy combatant. It therefore raised some of the same separation of powers issues that arise in traditional administrative and criminal law cases.
The government alleged that Hamdi was affiliated with the Taliban in Afghanistan and that his Taliban unit surrendered in a zone of active combat. (91) Because the United States was engaged in armed conflict with the Taliban, the government argued that individuals associated with them "were and continue to be enemy combatants" who could be lawfully detained by the President pursuant to his war powers. (92) The government contended that separation of powers principles prevented federal courts from inquiring further into Hamdi's status as long as the government provided "some evidence" that he was an enemy combatant. (93)
A majority of a fractured Court rejected the government's "some evidence" standard, (94) though the Court was still highly deferential to the government. Applying the Mathews v. Eldridge (95) balancing test from the administrative law context, (96) a majority of the Court concluded that a citizen-detainee was entitled to "notice of the factual basis of his classification, and a fair opportunity, to rebut the Government's factual assertions before a neutral decisionmaker." (97) A plurality of the Court cautioned, however, that "the exigencies of the circumstances may demand that, aside from these core elements, enemy combatant proceedings may be tailored to alleviate their uncommon potential to burden the Executive at a time of ongoing military conflict." (98) The plurality then noted that it would be acceptable for the government to use hearsay evidence or to adopt a presumption in favor of the government's evidence. (99) The plurality also noted "the possibility that the standards we have articulated could be met by an appropriately authorized and properly constituted military tribunal." (100) Thus, a majority of the Court--the plurality plus Justice Thomas--placed heavy weight on the government's claimed need for a streamlined process as against Hamdi's liberty interest.
Justice Scalia again employed a formalist analysis. In a dissent joined by Justice Stevens, Justice Scalia noted that "freedom from indefinite imprisonment at the will of the Executive" formed the core liberty interest protected by the separation of powers. (101) Because "[c]itizens aiding the enemy have been treated [historically] as traitors subject to the criminal process," in Justice Scalia's view it was "unthinkable that the Executive could render otherwise criminal grounds for detention noncriminal merely by disclaiming an intent to prosecute, or by asserting that it was incapacitating dangerous offenders rather than punishing wrongdoing." (102) After looking at the Constitution's text, the history of habeas, and writings from the founding generation, he concluded that the government had only one of two options if it wanted to detain Hamdi: either charge him with a crime or, if the exigencies of war required it, suspend the writ of habeas corpus. (103) Justice Scalia concluded by criticizing the plurality's willingness to craft a different solution in the face its conclusion that the government's existing procedures were inadequate. He said the plurality's approach reflects a "Mr. Fix It Mentality" that puts the unelected Court in the business of deciding what works best instead of "decree[ing] the consequences, as far as individual rights are concerned, of the other two branches' actions and omissions." (104)
Here, too, it seems relatively clear what animated the rest of the Court to reject Justice Scalia's formalist argument. The Court did not want to put the government to the choice of suspending the writ or pursuing criminal charges because the Court thought more flexibility was necessary for the government to respond to the threat of terrorism. This was particularly true when the individual being detained was captured on a battlefield, a situation in which the Court must have thought the risk of innocence was likely to be low and the government need for expediency particularly high. (105) A bright-line rule would not allow the same flexibility to adjust to the facts; by adopting a functional approach that balances the interests at stake, the Court allowed itself room to adjust the outcomes based on factual variations it deemed significant.
B. Criminal and Administrative Law Cases Compared
Although the Court has not decided many criminal law cases raising separation of powers claims, it has followed the same pattern: the Court has been highly receptive to functional arguments about the need for government flexibility. This contrasts with the Court's treatment of separation of powers arguments raised outside the criminal law context. Although occasional cases, like Schor, employed functional analysis, most other cases used formalist reasoning. (106)
Given the paucity of separation of powers cases generally and of criminal cases raising such issues in particular, it is unwise to draw any firm conclusions from the patterns in the Court's cases. But the Court's rejection of formalist arguments in each of the criminal cases when it has employed that methodology frequently in administrative law cases at least merits further exploration. After all, one might expect the pattern to be just the opposite. Because the threat to liberty when the state proceeds criminally is greater than when it proceeds civilly, one might think the Court would be particularly vigilant about enforcing the separation of powers when cases involve criminal matters. (107) What explains the opposite result, where the Court embraces functionalism more readily in criminal cases?
The explanation does not seem to rest on the Court's belief that criminal cases somehow deserve less protection. (108) In none of the Court's criminal cases did the Court indicate that separation of powers concerns were less pressing or that they were otherwise addressed by other means. And it would be difficult to imagine a line of reasoning that would lead the Court to believe that criminal proceedings deserve less protection than civil proceedings.
A more likely explanation for the differential pattern is that the Court simply failed to view Morrison, Mistretta, or Hamdi as criminal cases in the classic sense. Morrison was not about run-of-the-mill prosecutions but rather those involving crimes by government officials. (109) Mistretta was not about rules of criminal liability, but about sentencing, (110) and the Sentencing Commission on the surface seemed to mirror traditional regulatory agencies. Hamdi, too, was unique because it was about the executive's power to respond to terrorism. (111) Each case, at least on the surface, seemed to be much more about the scope of executive and agency power than about criminal power.
Thus, the Court seemingly viewed these cases through the lens of the administrative state and its need for efficiency and flexibility. (112) Indeed, the opinions themselves suggest as much. In upholding the independent counsel law, for example, the Court drew an explicit comparison to "various federal agencies whose officers are covered by 'good cause' removal restrictions" and noted that they "exercise civil enforcement powers that are analogous to the prosecutorial powers." (113) Although Hamdi was a case about executive power in wartime, it is telling that the Court employed the due process paradigm from an administrative law case, Mathews v. Eldridge, (114) which was designed to take into account government claims for efficiency.
The Court's failure to focus on the criminal aspects of these cases suggests that the Court did not view them as distinguishable from other cases involving executive power. If true, that would mean there is no clean split between administrative cases on the one hand and criminal cases on the other but just different treatment of different administrative law cases.
Because the purpose of this Article is to challenge the view that criminal and administrative law cases are indistinguishable for separation of powers purposes, Part II will explore the critical distinctions between these two areas. Once the differences between criminal and administrative power are explored, it will become clear that the Court's core separation of powers jurisprudence is almost completely backwards: a functional analysis that allows greater blending of powers is most troublesome in the context of criminal matters.
II. CRIME IS DIFFERENT
Perhaps because the Court does not divide its separation of powers decisions along substantive lines, (115) scholars likewise have failed to disaggregate separation of powers questions based on subject matter. (116) The unfortunate result is that the existing doctrine and literature neglect important differences between criminal matters and civil regulatory actions. This Part will explore these key distinctions to show the particular importance of separation of powers protections in criminal matters.
Part II.A begins by exploring how the exercise of government criminal power rests on a different constitutional foundation than administrative power. The argument for strict separation of powers when the state uses its criminal powers has strong support in the Constitution's text, structure, and history. Part II.B provides functional reasons for greater enforcement of separated powers in the criminal context. The agencies and individuals responsible for enforcing criminal laws are not subject to the same structural and institutional protections that exist in the typical regulatory context. The APA does not apply, and there is weak political oversight of government overreaching. Part II.C rebuts the view that the individual rights protections in the Bill of Rights provide an adequate replacement for the separation of powers.
A. Constitutional Foundations
The case for what might be called criminal law exceptionalism starts with the text and structure of the Constitution itself. One of the animating features of the Constitution is its preoccupation with the regulation of the government's criminal powers. Even before the adoption of the Bill of Rights, the Constitution provided protection for the rights of those accused of crimes through its structural provisions.
"The Framers were well acquainted with the danger of subjecting the determination of the rights of one person to the 'tyranny of shifting majorities,'" (117) and they were particularly focused on the dangers of legislative encroachment on the judicial power over crime. Because the state could potentially go after any citizen in a criminal proceeding, the normal course of politics should act as a threshold check on the passage of laws that criminalize too much ordinary conduct. (118) But that protection will not work if political actors can single out the conduct of disfavored minority groups and criminalize their conduct specifically.
The Framers recognized that risk--and concomitantly, the temptation for the legislature to engage in such behavior--and thus Article I establishes express limits on the legislative exercise of judicial power. First, it prohibits bills of attainder, which would allow Congress to identify those specific individuals affected by any given piece of legislation before passing it. (119) As the Court noted in United States v. Brown, "the Bill of Attainder Clause was intended not as a narrow, technical ... prohibition, but rather as an implementation of the separation of powers, a general safeguard against legislative exercise of the judicial function, or more simply--trial by legislature." (120) The legislature lacks the impartiality of the judiciary, and if the legislature had the power to single out individuals, it "might [be tempted] to act as a judge in its own cause." (121) Second, Article I also aims to prevent the legislature's ability to target individuals for criminal punishment through the prohibition on ex post facto laws. (122) Alexander Hamilton observed that "[t]he creation of crimes after the commission of the fact ... and the practice of arbitrary imprisonments, have been, in all ages, the favorite and most formidable instruments of tyranny." (123) The prohibition against ex post facto laws is therefore "one aspect of the broader constitutional protection against arbitrary changes in the law." (124) Third, Article I limits Congress's authority to suspend the writ of habeas corpus, which is a key protection for individuals against unlawful detention. (125)
Article I expressly addresses legislative interference with the judicial function, but what if the legislature works with the executive to single out disfavored minorities for prosecution? (126) That is, suppose the laws are generally applicable, but they are enforced only against unpopular groups or political enemies of the party in power. (127) The prohibition on bills of attainder will not suffice because the law itself does not target anyone specifically. And ex post facto prohibitions will not serve as a check if the laws are forward-looking. Yet the dangers would be the same as the ones that the ex post facto and bill of attainder provisions are designed to combat.
The separation of powers recognizes and addresses this threat of discriminatory enforcement. It requires not only that the executive and legislative branches agree to criminalize conduct but also includes the judiciary as a key check on the political branches. (128) Before a person can be convicted, he or she is entitled to judicial process. Federal judges, with life tenure and salary protections, have the independence that enables them to check both the legislature and the executive and to assure fair and impartial decisionmaking in a given case. (129) Judges are therefore tasked with enforcing provisions like the Bill of Attainder Clause and the Ex Post Facto Clause to prevent the legislature and the executive from obtaining judicial power. (130) Justice Scalia has argued that the "most significant role" for judges is "to protect the individual criminal defendant against the occasional excesses of th[e] popular will, and to preserve the checks and balances within our constitutional system that are precisely designed to inhibit swift and complete accomplishment of that popular will." (131)
Even these protections were inadequate to the Framers, however. Although Article III judges are relatively more independent than Congress and the executive branch, they are still part of the government. Because separation of powers is concerned, among other things, with conflicts of interest, (132) judges were not deemed sufficient protection against the possibility of state abuse in criminal cases because of their potential partiality toward the government. (133) The Constitution therefore provides in Article III--the Article establishing the judicial role in government--that the trial of all crimes must be by jury. The jury's unreviewable power to acquit gives it the ability to check both the legislative and executive branches. (134) And because federal juries must be unanimous, all representative members of the community must agree before political actors can impose criminal punishment. (135) The jury, then, is a key component of the separation of powers in the criminal law. (136)
Other constitutional provisions reflect a similar concern with the danger to liberty associated with the criminal process. (137) Although the executive branch can sometimes pose a threat to liberty, it can also act as a check on the other branches. Thus, Article II vests the President with the power to grant pardons for all federal offenses except in cases of impeachment. (138) As the Supreme Court has explained, this power "exists to afford relief from undue harshness or evident mistake in the operation or enforcement of the criminal law." (139)
The Framers' concern with expansive state criminal powers became even more evident with the adoption of the Bill of Rights. Four of the first ten Amendments deal explicitly with criminal process. The Fourth Amendment regulates the state's policing and investigative powers. (140) The Fifth Amendment acts as a check on the state's executive powers by providing for the right to a grand jury and prohibiting the state from prosecuting individuals twice for the same offense. (141) And, of course, the Fifth Amendment's Due Process Clause (142) (and later, the Fourteenth Amendment's Due Process Clause (143)) ensures that the government follows proper process before depriving an individual of life, liberty, or property. The Sixth Amendment reiterates the centrality of the jury's role in adjudicating criminal cases, making clear that the jury will be drawn from the community in which a crime occurs. (144) In addition, the Sixth Amendment provides a host of additional rights to defendants: the right to a speedy and public trial, the right to notice of criminal charges, the right to confrontation, and the right to assistance of counsel. (145) And the Eighth Amendment regulates the government's legislative judgments by putting a cap on punishment. (146) There is thus abundant constitutional regulation of all aspects of the government's criminal power, from investigation to prosecution, from adjudication to the legislation defining punishment. These powers are strictly defined and divided, just as they are in Articles I, II, and III.
Akhil Amar argues that the Bill of Rights protections were not originally conceived as a litany of rights but as structural limits to protect local majorities from national power. (147) At the heart of this scheme, according to Amar, is the jury. Thus, many of these protections were designed to preserve the power of the jury in a criminal trial to act as a kind of local government check on corrupt or abusive national power. (148) Regardless of whether Amar is correct in the totality of his analysis, there is no denying his claim that the Bill of Rights contains structural protections that serve to protect rights. (149)
When it comes to criminal justice, then, the separation of powers is divined not just from the fact that Articles I, II, and III separate legislative, executive, and judicial power. Rather, there are many additional textual indications that the separation of functions is critically important when the federal government uses its criminal powers. Under the scheme established by the Constitution, each branch must agree before criminal power can be exercised against an individual. Congress must criminalize the conduct, (150) the executive must decide to prosecute, (151) and the judiciary (judges and juries) must agree to convict.
This scheme provides ample evidence that the potential growth and abuse of federal criminal power was anticipated by the Framers (152) and that they intended to place limits on it through the separation of powers. Put another way, the Constitution's provisions addressing crime and the separation of powers reflect the fact that the Framers weighed the need for federal government efficiency against the potential for abuse and came out heavily in favor of limiting federal government power over crime. And, of course, one of their preferred methods for limiting government power was the separation of that power into strict categories.
It could be argued, however, that the extent of federal criminal law expansion, like the extent of the growth of the administrative state, was unexpected. In fact, federal criminal jurisdiction expanded alongside the growth of the administrative state. (153) The first major increase in federal criminal legislation occurred after the Civil War as Congress passed criminal statutes covering mail fraud and various crimes dealing with interstate commerce. (154) The passage of the Eighteenth Amendment and Prohibition led to another increase in federal criminal jurisdiction, (155) and the number of federal cases spiked as well. The average annual number of criminal cases terminated in federal court went from approximately 26,500 in 1920 to almost 96,000 in 1932. (156) After Prohibition, the number of criminal cases declined from its peak, but the annual average in the late 1930s and early 1940s still hovered roughly around 40,000 federal criminal cases. (157) That is due, in part, to the fact that even after the Eighteenth Amendment was repealed, federal criminal jurisdiction did not retreat. The New Deal-era Congress did not just expand federal administrative power; it passed a host of new federal criminal laws that covered everything from bank robbery to firearms to criminal penalties for the commission of regulatory offenses. (158) The number of federal cases dropped a small amount in the period from the late 1940s through the 1960s, (159) but just as the administrative state experienced a resurgence in the late 1960s and 1970s, (160) federal criminal law also expanded, as Congress turned its attention to drugs, violence, and organized crime. (161) The paths of federal criminal and regulatory law have diverged since the 1980s. There has been a strong push for deregulatory reforms and cost-benefit analysis on the civil side, (162) but federal criminal jurisdiction continues to expand at a feverish pace. (163) In the 1990s, the number of criminal cases terminated in federal court increased to roughly 60,000, and the annual average from 2000 to 2002 was more than 76,000 criminal cases terminated in federal court. (164)
If this federal criminal law expansion reflects the needs of a growing nation and the inadequacy of the states, perhaps the same arguments in favor of greater flexibility that have prevailed in the context of the administrative state should apply in the context of greater federal criminal law powers as well. (165) Perhaps the federal government should be allowed not merely to pass more substantive criminal laws, but also to use some of the same institutional tools that it uses in the context of the administrative state, including agencies that blend powers and more streamlined procedures.
One reason to hesitate before reaching this conclusion is that the presence of more federal criminal legislation may not reflect a greater need for federal action. First, federal criminal jurisdiction remains quite small as compared to state jurisdiction over crime. (166) Thus, unlike the administrative sphere, where federal regulatory authority often dominates the field in areas like environmental law and securities regulation, (167) states still bear most of the responsibility for the regulation of crime. Second, although the number of substantive federal offenses and prosecutions has increased, the number of federal enforcement officers remains relatively small. Many federal criminal laws are largely symbolic gestures that win members of Congress political points but result in few practical changes because they are not enforced. It is thus not always clear that a federal criminal law fills a void left by the states. On the contrary, federal law often duplicates state law without any showing that federal intervention is necessary. (168)
But even if it could be shown that, as in the context of the administrative state, there has developed over time a greater need for the federal government to take an active role on crime, it still does not necessarily follow that separation of powers restrictions should be relaxed when it comes to crime. That is because there are critical functional differences between the two settings, which are explored in the next Part.
B. Functional Differences
The greatest historical development when it comes to the separation of powers is, as previously noted, the rise of the administrative state. That development provides the main rationale for adopting a functional approach to separation of powers questions. (169) In response to the Depression and a government structure that seemingly failed to prevent it, reformers increasingly turned to expert agencies that would combine functions and address important social and economic problems. James Landis, one of the leading architects of the New Deal, explained that the government, like a business, needed to operate efficiently and that no business operated under the confines of anything like a separation of powers requirement. (170) On the contrary, successful managers combined functions for increased efficiency, and Landis and his fellow reformers wanted government to follow the same model to deal with the complexity of society after the Industrial Revolution. (171) They reasoned that the Framers could not have anticipated the problems of modern government and would not have intended the separation of powers to straitjacket the government into failure. The underlying idea is that an active, unimpeded government can produce beneficial results for society, so the government structure should be built around the agency concept and allow it to flourish.
The Court ultimately accepted these arguments and has long allowed administrative agencies to flout the separation of powers by combining executive, legislative, and judicial powers. (172) The acceptance of the administrative state is the ultimate example of a functional approach to separation of powers. But, as has been repeatedly emphasized by courts and commentators in justifying the acceptance of administrative agencies, regulatory agencies do not operate unchecked. (173) Many structural, procedural, and political safeguards exist to keep them from abusing power. Indeed, the courts' acceptance of the administrative state and its blending of powers was conditioned on the availability of judicial review and these institutional checks. (174) As this Part explains, these critical protections are lacking in the criminal arena.
1. Lack of structural and procedural safeguards
Although administrative agencies combine powers under one roof, they do not operate unchecked. On the contrary, their policy decisions are subject to judicial review and a host of procedural and structural requirements in the APA.
Agencies conducting formal adjudications must obey various structural rules designed to ensure impartiality. The individual at the agency who presides at the hearing must be impartial (175) and must be separated from individuals at the agency who perform investigative and prosecutorial functions. (176) Administrative law judges (Aids) also cannot be removed by the agency for which they adjudicate, but instead can be removed only after notice and hearing by the Merit Systems Protection Board. (177) Aids and anyone else at the agency involved in the decisionmaking process are prohibited from having ex parte communications related to the merits of the proceeding, (178) and the agency's decision must be based on the evidence in the record. (179) The APA also imposes various process requirements, including notice to interested parties, an opportunity for interested parties to submit evidence and arguments, and a chance for interested parties to submit proposed findings and to make exceptions to tentative agency decisions. (180) In all formal proceedings-rulemakings and adjudications--the agency must issue a decision on the record with a statement of its findings and conclusions. (181)
When agencies proceed through informal rulemaking instead of formal rulemaking or formal adjudication, they must issue a notice of their proposed rules and give the public an opportunity to comment. (182) As in formal proceedings, the agency's decision must be based on the facts in the record, and the agency must disclose the evidence on which it relied in reaching a decision. (183) The agency must consider the comments received and explain why it rejected arguments made in the comments. (184)
All agency proceedings--formal or informal, rulemaking or adjudication--are subject to extensive judicial review. Decisions made in even the most informal of adjudications are subject to judicial oversight to ensure that the agency's action is not arbitrary and capricious. (185) Agencies must give reasons if they change course from case to case, and there must be support for the agency's decision in the administrative record. If there is evidence in the record that undermines the agency's position or if a party or commenter raises a serious objection to the agency's proposal, the agency must offer reasons why those arguments do not hold sway. Thus, unlike the judicial rubber-stamping associated with rational basis review, courts take a "hard look" at the agency's explanation to provide a check against arbitrary implementation. (186)
Oversight laws--such as the Freedom of Information Act (FOIA) (187) and the Federal Advisory Committee Act (FACA) (188)--grant the public additional access to information about the agency decisionmaking process, which provides further protection against arbitrary agency action or agency decisions based on improper influences. And, to the extent that the Office of Management and Budget reviews agency rules under executive orders (189) for consistency across agencies, (190) this provides an additional check on agency policies.
This oversight regime of judicial review and structural constraints has been crucial to the Court's acceptance of broad delegations of legislative and judicial power to executive agencies. (191) As Rebecca Brown has noted, it is precisely because agencies combine all three types of government power that these other measures are necessary to protect individual rights. (192)
Notably, these protections do not apply to the actions of key governmental officials and agencies exercising criminal power, particularly prosecutors. Although Kenneth Culp Davis argued almost four decades ago that the discretion exercised by police officers and prosecutors should be subject to the same procedural and structural checks as other administrative officials, (193) his calls for reform were not heeded. Political actors did not impose police regulation modeled along the lines of the APA, (194) though that failure might be explained in part by the fact that the Supreme Court's Fourth and Fifth Amendment decisions imposed constitutional regulation on the police that might have made administrative regulation appear unnecessary. (195) But even if Court decisions explain the lack of administrative oversight mechanisms on the police, it cannot explain the continued lack of oversight for prosecutors.
Because of the operation of a broad federal criminal code and prosecutors' leverage over plea bargaining, (196) the only process--judicial or otherwise--that most defendants receive comes from prosecutors. In the course of reaching a negotiated disposition, "the prosecutor acts as the administrative decision maker who determines, in the first instance, whether an accused will be subject to social sanction, and if so, how much punishment will be imposed." (197) Despite the significance of prosecutorial power, prosecutors operate with little oversight or regulation. (198) The same prosecutor who investigates a case can make the final determination about what plea to accept. There is therefore no structural separation of adjudicative and executive power, and defendants have no right to a formal process or internal appeal within the agency. In addition, in the course of bargaining with a defendant over charges, the prosecutor can engage in ex parte contacts with the police and investigators, and the defendant need not be given access to the information on which the prosecutor relies--that is, the prosecutor's evidence of the defendant's guilt. (199)
The Supreme Court is of the view that a prosecutor's charging and plea bargaining decisions are largely off limits from judicial review. (200) "In the ordinary case, 'so long as the prosecutor has probable cause to believe that the accused committed an offense defined by statute, the decision whether or not to prosecute, and what charge to file or bring before a grand jury, generally rests entirely in his discretion.'" (201) Prosecutors cannot base their decisions on "an unjustifiable standard such as race, religion, or other arbitrary classification," (202) but the Supreme Court has "taken great pains to explain that the standard [for showing discriminatory prosecution] is a demanding one." (203) Prosecutors enjoy a "presumption of regularity," and "in the absence of clear evidence to the contrary, courts presume that they have properly discharged their official duties." (204) Indeed, for a defendant to obtain discovery on a discrimination claim, much less prevail on the claim itself, he or she must show that the government failed to prosecute similarly situated defendants. (205) The Court has emphasized that there is a "'background presumption' that the showing necessary to obtain discovery should itself be a significant barrier to the litigation of insubstantial claims." (206) Given the secrecy with which prosecutors' offices operate, it should come as no surprise that these claims rarely succeed.
Furthermore, while judges oversee prosecutors to make sure that pleas are knowing and voluntary (207) and have a factual basis, (208) their inquiries are cursory (209) Judge Gerard Lynch aptly describes the judicial part of this process as follows:
In a substantial number of cases, the judicial "process" consists of the simultaneous filing of a criminal charge by a prosecutor (often by means of a prosecutor's "information" rather than an indictment, with the defendant waiving the submission of the evidence and charge to a grand jury) and admission of guilt by the defendant. The charging document may be quite skeletal, the defendant's account of his guilty actions brief, and the judicial inquiry concerned more with whether the defendant is of sound mind and understands the consequences of what he is doing than with the accuracy of the facts to which he is attesting. (210)
Thus, instead of being subject to the hard-look review that other agencies face when they seek to impose fines or penalties or to require action of some kind, prosecutors enjoy a presumption of regularity and face only a cursory judicial inquiry of their discretionary decisions to enter plea agreements. (211)
Moreover, though a defendant has the right to reject a plea and take his or her case to trial, (212) that option fails to police structural abuses of power. (213) Indeed, a defendant takes a big gamble in exercising even this limited power of review, because if the defendant is found guilty, he or she is typically subject to harsher punishment.
Without judicial oversight to speak of or any internal constraints, the potential for arbitrary enforcement is high. (214) Prosecutors need not treat similar cases similarly for purposes of plea bargaining, and they need not explain why they agreed to reach a deal with one defendant but refused to do so with another defendant guilty of the same crime. Indeed, because prosecutors need not make the terms of their plea bargains available to the public through publication and because prosecutorial law enforcement is largely exempt from open government laws like FOIA, (215) a defendant might not even know that another similarly situated defendant received a particular deal. (216) Nor may defendants be aware that a prosecutor is diverging from office policy. Judge Lynch notes that prosecutors' offices will often change their enforcement policies and likens these shifts to an administrative agency's decision to issue a new set of regulations. (217) But unlike an administrative agency's policies, the prosecutor's policies are not openly disclosed to the public and are not subject to arbitrary and capricious review for reasoned consistency. (218)
Nor is the lack of structural oversight in the federal criminal process limited to prosecutors. Even when Congress created the United States Sentencing Commission, a federal criminal agency modeled in crucial respects after traditional administrative agencies, it failed to subject the Commission to key APA requirements. While the Sentencing Reform Act requires that the Sentencing Commission's rulemaking proceedings comply with the notice and comment requirements of the APA, (219) the rules themselves are not subject to judicial review to make sure that they are not arbitrary and capricious. (220)
There is, then, a sharp incongruity between the treatment of discretion in the administrative context and the criminal context. The Court accepted the constitutionality of the administrative state against the backdrop of structural and procedural protections that protect against the dangers of combining powers under one roof and of allowing one branch of government to exercise disproportionate authority. Because those protections are lacking in the context of criminal matters, the same arguments in favor of flexibility do not apply.
2. Lack of political safeguards
The structural and procedural checks supplied by the APA are not the only mechanisms for checking government abuse that are absent in the criminal context. Political oversight mechanisms provide a key check on administrative agencies, but they do not work as effectively when it comes to criminal enforcement. (221)
When the government regulates noncriminal as opposed to criminal conduct, many interested groups have the incentives and the power to police the government's policies. Typically, public interest and consumer groups seek more regulation while corporations and industry groups advocate less regulation. The targets of regulation are often the more powerful force, as they have the incentives and the means to fight government interference and to get procedural and substantive protections through the normal course of politics. (222) Regulated entities can make credible threats that they will obtain favorable legislation that overrules the agency or that they will challenge the agency in time-consuming court proceedings. This gives them leverage with the agency that acts as a check on government overreaching. (223) Thus, in the typical regulatory context, groups line up on both sides of the issue, and, if anything, there are more powerful forces operating to check too much government intervention.
The political process is more skewed when it comes to crime, particularly federal legislation aimed at substantive crime definition and sentencing. (224) Neither criminal defendants nor judges--the two main targets of criminal punishment legislation--have much sway in the political process. Those who have not been caught committing a crime are rarely going to self-identify in order to lobby for lesser punishments or more narrow crime definitions. (225) And individuals already convicted of a crime are perhaps the weakest of all groups in the political arena. By virtue of their conviction, criminals often lose the right to vote, and they are a weak lobbying force even when they retain that right. (226) And while their families and communities have an interest in advocating on their behalf, these groups currently lack political power. (227) Judges, the objects of legislation aimed at curbing sentencing discretion, are more effective lobbyists than criminal defendants, but they have been largely unsuccessful in stopping legislation that has limited their discretion. (228)
While the targets of regulation are weak, proponents of more expansive criminal laws are not. (229) Prosecutors have an incentive to request broader criminal laws and longer, mandatory sentences because those laws make it easier for them to obtain defendants' cooperation in plea bargaining. (230) Groups with a stake in the expansion of prisons--including rural communities, corrections officer unions, and private prison companies--are powerful forces in favor of longer sentences. Victims' rights groups similarly endorse longer sentences and more expansive criminal laws. (231) The public is also generally supportive of harsher criminal laws and is easily mobilized by politicians or interest groups to get behind "tough on crime" initiatives. (232)
Because the targets of regulation are weak and the voices in favor of broader laws and longer punishments are powerful, the political system is biased in favor of more severe punishments. There are few forces that can counter the government when it overreaches on crime. As Jeremy Bentham observed, "legislators and men in general are naturally inclined" in that direction because "antipathy, or a want of compassion for individuals who are represented as dangerous and vile, pushes them onward to an undue severity." (233) Bentham therefore advocated that "[i]t is on this side [towards severity], therefore, that we should take the most precautions, as on this side there has been shown the greatest disposition to err." (234)
The scheme of separated powers is designed to do just that. (235) The Constitution makes it difficult for the state to act in criminal cases against individuals and members of groups disfavored by the majority. (236) All three branches must agree to allow a criminal conviction, and the judiciary plays a particularly significant role because of its relative insulation from the political imbalance described above.
The impediments to action provided by the separation of powers check state abuse and preserve the interests of individuals and local and political minorities. This argument for separation of powers is therefore the classic representation-reinforcing theory for judicial review. Thus, arguments for dismantling this scheme on the basis of efficiency grounds--that the state is hamstrung in its ability to proceed in criminal cases--disrupt the very core of why we have separation of powers in the first place. The inefficiency associated with the separation of powers serves a valuable function, and, in the context of criminal law, no other mechanism provides a substitute.
C. Individual Rights Protections Are Insufficient
The separation of powers is not the only means by which the Constitution protects the interests of criminal defendants. The Bill of Rights, as noted, provides additional protections to prevent the political process from targeting individuals. It is important to consider whether the protections offered in the Bill of Rights render the importance of separation of powers protections less significant in criminal matters. Put another way, if the individual rights protections serve the same function as the various procedural and structural protections of the APA and check the political process failures, then perhaps there would not be the same need for greater enforcement of the separation of powers in the criminal arena.
Indeed, the view that the Bill of Rights acts as a sufficient check may explain the general lack of separation of powers arguments in criminal cases and the Court's relaxed treatment of them when they do arise. Most constitutional challenges in criminal cases are based on the rights protections in the Fourth, Fifth, Sixth, and Eighth Amendments. Because the Supreme Court was receptive to many of these claims in the 1960s and early 1970s, (237) litigants grew reliant on arguments couched in these terms. This tradition of regulating criminal process through rights has continued even as the Court has grown more conservative. It is quite possible that the Court's oversight of the criminal process through rights protections is one of the reasons why it was not concerned in Mistretta and Morrison with the novel governmental arrangements being used in criminal matters. Perhaps the Court's willingness to allow a melding of powers reflected the Court's confidence that it retained enough judicial oversight through the rights provisions to correct egregious misalignments of power in any given case. (238) In administrative law contexts where the Court applied a formalist separation of powers analysis, in contrast, the courts did not have the ability to oversee the procedures in the same way. (239)
If this does, in fact, explain the Court's reluctance to police separation of powers in criminal cases, the Court has a false sense of security. "It would be a grave mistake ... to think a Bill of Rights in Madison's scheme [at the Founding] or in sound constitutional theory now renders separation of powers of lesser importance." (240) The fights-based procedural measures are valuable, but they are incomplete protections against government overreaching in criminal matters. Although the rights protections police government abuse of power to an extent, they do not guard against the same structural abuses as the separation of powers. (241)
The separation of powers acts as a direct check against the accumulation of too much power in one branch and against the evasion of the process required of that branch, something that individual fights protections do not guard against. (242) For example, if the legislature were permitted to adjudicate criminal matters, none of the protections that apply to Article III courts would apply, nor would the legislature be subject to the rules of judicial process. The legislature is designed to pass general rules, not to decide matters affecting the liberty of individuals. (243) Thus, if Congress were allowed to have judicial powers, the protections associated with judicial process could be bypassed. Similarly, if the executive branch were permitted not merely to bring enforcement actions but to adjudicate them as well, the judiciary and all of its processes would be rendered a nullity. (244)
That is, in fact, what has happened. With the rise of plea bargaining, trials are anomalies, not the norm. And the current individual fights approach to plea bargaining has done nothing to prevent the executive's accumulation of judicial power. A court merely asks whether a plea in a given case is knowing and voluntary. For most defendants, the deal offered by the government will likely be in their interests, so they are accepting the plea knowingly and voluntarily. While an individual defendant might find it in his or her interest to waive a constitutional fight to get a better deal, plea bargaining is not in defendants' interests as a group. (245) It curtails the ability of the judicial branch to check abuses in the political process, and once plea bargaining is deemed acceptable and becomes the normal mode of case disposition, it encourages Congress to draft its criminal laws and sentences with plea bargaining in mind. (246) That is, it makes sense for Congress to draft criminal statutes broadly and with high mandatory penalties to give prosecutors the leverage they need to induce guilty pleas. (247) Indeed, plea bargaining pressures even innocent defendants to plead guilty to avoid the risk of high statutory sentences. And those who do take their case to trial and lose receive longer sentences than even Congress or the prosecutor might think appropriate, because the longer sentences exist on the books largely for bargaining purposes. (248) This often results in individuals who accept a plea bargain receiving shorter sentences than other individuals who are less morally culpable but take a chance and go to trial. (249) Plea bargaining therefore fails to serve the interests of the public, as it tends to undermine the legitimacy and accuracy of the criminal justice system. (250)
The individual rights perspective misses these structural concerns and cannot bear the weight of policing the inequalities that the separation of powers is designed to address. (251) Without other institutional or political checks to fill the void, the criminal process is susceptible to abuses associated with unchecked power unless careful attention is paid to the constitutional protections provided by the separation of powers.
III. ENFORCING SEPARATION OF POWERS IN CRIMINAL MATTERS
To conclude that criminal law matters merit greater separation of powers protection than administrative law matters still does not answer the question of how best to enforce those protections. (252) Should the courts employ an analysis along the lines of Chadha or Bowsher, in which the Court uses a bright-line rule approach that requires each branch to exercise only a certain type of power and to follow all of the constitutional procedures associated with the exercise of that power? Or should a more flexible analysis be employed, albeit with a thumb on the scale for maintaining strict separation in criminal matters, which would allow more blending of powers than the formalist approach? (253)
Although either of these approaches would be an improvement over the current functional methodology that has allowed much relaxation in the criminal sphere, there are reasons to believe that the bright-line rule variant of formalism, (254) employed in many of the Court's separation of powers cases, is the better option. In determining what type of power is being exercised and what checks the Constitution requires, the Court could continue to use conventional methods of interpretation, which would allow the Court to consider text, history, precedent, and evolved practices. (255)
The remainder of this Part will defend this approach to separation of powers in criminal cases. (256) Part III.A discusses the merits of maintaining a strict division of powers instead of employing a flexible balancing test in criminal matters. Because the application of this approach is best understood through concrete settings, Part III.B will explain how this approach would change the outcome in Morrison, and Part III.C will explore the relationship between separation of powers and federal sentencing practices. (257) Part III.D will then explore how this analysis would apply to prosecutorial discretion. Finally, Part III.E discusses the relationship between formalism and federalism.
A. The Case for Greater Enforcement
Although scholars have long touted the general benefits associated with bright-line formalism, such as its predictability and reduced decision costs, (258) the argument for formalism advocated here rests on more specific grounds. Because the judiciary (judges and juries alike) serves as the critical safety valve against the political abuse of the criminal process, the separation of powers threat in this context will likely result from subtle shifts in authority that have the result of stripping the judiciary of some of its authority.
One might think that this is therefore an area in which the judiciary will be particularly sensitive to the dangers of separation of powers because it will protect its own interests with vigilance. (259) But this has not been borne out by the case law. It has been in those cases where judicial power has been lessened that the Court has been least protective of the separation of powers. In Schor, the Court accepted agency adjudication of private law claims even though they had traditionally rested with Article III courts. (260) Similarly, in Mistretta, the Court permitted a sentencing scheme that shifted power from the judiciary to the executive branch and the legislature. (261) As Peter Strauss has noted, "an observer might conclude that the Court seems much more clearly committed to functionalism in examining its own place in the constitutional scheme than in dealing with issues concerning the other two heads of government." (262) Strauss posits two possible explanations for this trend. Either the Court is "appropriate[ly] modest in declaring constitutional principles that might appear to enshrine the Court's own place, or it may reflect self-interested relief at being freed of the need to decide matters of routine in litigation-rich times." (263)
Given the Court's general lack of modesty when it comes to deciding constitutional questions and the ever-expansive view it takes of its own power, (264) it would seem that the more likely cause is the Court's receptivity to claims that a proposed change in government will yield efficiency gains for the judiciary. This is particularly true if the judiciary retains some residual authority over the subject matter through appellate oversight. Thus, in Schor, the Court relinquished authority to agencies to adjudicate state private law claims in the first instance because the judiciary could review the agencies' decisions. (265) Similarly, in Mistretta, although trial judges would lose a great deal of sentencing discretion, appellate judges gained the authority to review trial judge departure decisions. (266) The Court's decision in Morrison also supports this hypothesis. In Morrison, it appeared that the judiciary's power increased vis-a-vis the executive, for the judges of the Special Division obtained authority to appoint a prosecutor. (267) If the Court were being "modest," presumably it would find this transfer of power disconcerting. That it did not suggests that the Court saw this apparent enlargement of authority acceptable, particularly given that it would not increase the workload of the courts. And if the scheme posed a threat to an individual criminal defendant, the courts retained oversight through their role in criminal trials.
The argument for formalism, then, essentially boils down to distrust that judges will be able to give sufficient weight to the long-term, systemic interests the separation of powers protects when faced with a reasonable claim of governmental need for flexibility in the criminal context. (268) This is, of course, a common justification for formalist analysis. (269) But it has special purchase in the area of criminal law, where resource pressures on the courts and the judicial system make claims of convenience particularly attractive to judges (270) and where the dangers of relaxing the separation of powers may seem remote because judges retain oversight over individual rights in the form of criminal trials. The federal court system is notoriously overburdened with criminal cases, (271) so any proposal that streamlines the criminal process and results in less criminal work for the judiciary is bound to have at least some appeal. And the extensive protections of the Bill of Rights mean that defendants still receive judicial oversight. On the surface, then, it might be difficult for judges to see why that oversight is insufficient and why some relaxation of the separation of powers is not sensible. (272)
The judiciary's lack of perspective is particularly likely at this historical moment, when judges have become desensitized to a criminal justice system in which prosecutors exercise extensive judicial power. (273) Moreover, because courts typically analyze separation of powers questions in regulatory contexts, courts have become accustomed to blending arrangements as a general matter and neglect the key differences between administrative and criminal matters described in Part II. Thus, even a functional test with bite is likely, at this point in time, to yield the same underprotection of the judiciary's role in criminal proceedings.
There is no denying that a formalist approach has shortcomings of its own. In particular, it might lead to the rejection of useful and productive government arrangements. (274) These same costs have been recognized--and accepted--even in the context of the administrative state, an area in which efficiency claims have had great sway. (275) Although the Court accepted the New Dealers' proposals for a more streamlined government, it did not allow efficiency values to trump all others. The procedural and structural safeguards of the APA often impede government action. (276) And, of course, whenever the Court interprets the separation of powers to prevent a particular government action, that serves as an obstacle to more streamlined government. (277)
Another potential shortcoming of a formalist approach to separation of powers is the opposite one: it might not prove to be that effective. If the Court strikes down a particular practice, Congress and the executive branch will likely seek to achieve the same goals through different--though perhaps more costly or less effective--means. (278) While undoubtedly courts may accept some concentration of power even under a formalist analysis, more stringent review should make a difference in at least some cases where the constitutional deficiency is too great to ignore, as Parts III.B, III.C, and III.D will explore. (279)
And because this review will make government action more costly or time-consuming--at the very least, the government needs to make an extra effort to work around the Court's decision--that, too, should deter some concentration of power. And, again, this criticism applies in the administrative context as well, where there is a similar risk that the government will find a way around the process to achieve the same ends.
If these costs and shortcomings of formalism have been deemed acceptable in important cases in the realm of administrative law--where the values of convenience and efficiency hold even more sway than in the criminal context--they should also be accepted when it comes to criminal justice. At least in the absence of other protections, the costs associated with separation of powers enforcement are the price of keeping government abuse in check and ensuring that no one is labeled a criminal without adequate process and the agreement of all the relevant constitutional actors.
B. The Independent Counsel
The Court's decision in Morrison, as noted, was the product of a functional analysis that permitted government flexibility. If the Court had employed a formalist methodology in that case, required a strict division of powers, and protected each branch's sphere of authority, the outcome and analysis would have changed.
In Morrison, the more obvious formalist objection to the regime was the one raised by Justice Scalia in dissent, namely that once prosecutorial power was deemed executive, Congress could not impose restrictions on the President's power of removal. But there is another objection to the independent counsel regime that also goes to the core of the separation of powers and the criminal law. In the criminal context, as noted, the Constitution requires that each branch must exercise independent judgment to convict. The Ethics in Government Act, however, allowed Congress to impose pressure on the executive branch to bring an indictment and limited prosecutorial discretion not to bring charges. (280) The Act imposed reporting requirements on prosecutors and set up a scheme where it would be difficult, if not impossible, for the Attorney General not to appoint an independent counsel once Congress established evidence showing hints of criminal activity. (281) And once appointed, the independent counsel retained the final decision over whether to bring charges. (282)
Thus, although a criminal conviction usually requires affirmative approval by all branches of government, the independent counsel law diluted the prosecutor's freedom not to bring charges. (283) It instead gave Congress more tools to exert political pressure on prosecutors to bring charges in particular cases. This shift in power, in turn, eliminated one of the Constitution's protections of individual liberty. A formalist analysis could therefore strike the law either based on Justice Scalia's argument that those exercising executive power must be subject to removal by the President or based on the claim that the law stripped the prosecutor of his or her discretion not to bring charges and gave Congress too much oversight of that decision. (284)
A formalist approach to separation of powers would also change the result in Mistretta and the analysis in the Court's recent cases addressing the relationship between juries and mandatory sentencing laws. (285)
Justice Scalia's formalist analysis in Mistretta uncovered one deficiency with the Sentencing Commission--namely the fact that the Sentencing Commission possessed legislative power outright and not as an incident to some other executive or judicial function. But, again, there was another separation of powers shortcoming with the Sentencing Reform Act that could have been uncovered if the Court strictly policed the authority of each constitutional actor. As I have explained in greater detail elsewhere, the Sentencing Reform Act operated to transfer significant discretionary power from the judicial branch, and particularly the jury, to the executive branch and to Congress. (286) The Sentencing Guidelines established under the Sentencing Reform Act, like other mandatory sentencing laws, dictate a given punishment on the basis of particular fact-findings. These laws operate no differently from general criminal laws. They define facts that yield punishment.
The Constitution has a carefully calibrated scheme dictating how laws that impose criminal punishment are to operate. Specifically, under the Constitution, juries must apply those laws because juries act as a critical check against government overreaching. With their power to nullify, juries have the discretion to check overbroad laws and ensure that the laws are properly applied to a given set of facts. (287) The jury provides a critical check on the legislature and the executive that judges cannot replicate in the context of mandatory sentencing laws. (288) But unlike other mandatory criminal laws that impose punishment, the Guidelines were to be applied by judges, not juries, and judges were not given discretion to check those laws. The Guidelines therefore take constitutional power away from the judiciary, thereby increasing the power of Congress and the executive. (289)
While the Supreme Court has recently identified the threat to the jury posed by the Sentencing Guidelines in United States v. Booker, (290) its failure to focus on the separation of powers and the structural check provided by the jury led it to miss the real problem with the Guidelines and other mandatory sentencing, laws that require judges to find facts which increase a defendant's sentence. (291) Instead, the Court analyzed these laws as interfering only with the defendant's individual rights. (292) This led the Court to overlook completely the constitutional problems with mandatory minimum sentences (293) and to produce a line of cases that lacks much in the way of coherence or analysis. (294) If the Court had instead viewed the relationship between the jury and the Sentencing Guidelines through the lens of the separation of powers, it would have seen that the danger of mandatory sentencing laws is that they allow the expansion of legislative and executive power without a sufficient judicial check. That is, the Court would have seen that the key problem with these laws is their mandatory nature, not whether they set a floor or ceiling on punishment. Thus, under an analysis that looked to the criminal jury's role in the separation of powers, the Court would reject not only those laws that require judges (not juries) to increase a defendant's maximum sentence but also those laws that require judges (not juries) to set a minimum sentence. (295)
There is, of course, no guarantee that any separation of powers analysis would unearth all the relevant objections, and the fact that the formalist dissents in Morrison and Mistretta did not highlight all of the failings of the respective laws at issue shows that a formalist, bright-line approach is far from a perfect theory. The Court is likely to miss deficiencies even using that methodology. But even an incomplete formalist analysis is more likely to correct government overreaching than a functional analysis. After all, even without seeing all the consequences to the separation of powers arising from the laws at issue in Morrison and Mistretta, the approach taken by the dissent would have addressed those concerns because the dissent struck down the laws at issue. And that is in a very real sense the point of formalism. It errs on the side of caution and prevents even those dangers that might not be foreseen. Without other checks serving the same purpose in the context of crime, that extra protection is worth the cost of letting some innovations slip away.
D. Prosecutorial Discretion and Plea Bargaining
Government innovations such as those at issue in Morrison and Mistretta pose a noticeable challenge to the separation of powers because of their structural novelty. Though perhaps not as obvious, the virtually unreviewable exercise of prosecutorial discretion over charging and bargaining also stands in sharp tension with the separation of powers. Though a full analysis of the implications and possible resolution of the issue requires separate study, merely highlighting the issues demonstrates that important questions about prosecutorial discretion and plea bargaining have been unasked and unanswered by the Court's criminal law jurisprudence because of an inattention to structural separation of powers concerns.
The Court allows prosecutors almost unlimited discretion to make charging decisions. (296) And, although plea bargaining existed as an underground and unapproved practice for most of the nation's history, (297) the Supreme Court explicitly acknowledged and accepted plea bargaining as legitimate in
Santobello v. New York in 1971. (298) The Court approved plea bargaining largely on the grounds of convenience. (299) The Court reasoned that plea bargaining had become "an essential component of the administration of justice" and noted that "[i]f every criminal charge were subjected to a full-scale trial, the States and the Federal Government would need to multiply by many times the number of judges and court facilities." (300) To avoid burdening the system, the Court stated that plea bargaining should be "encouraged." (301) The Court later acknowledged "that, by tolerating and encouraging the negotiation of pleas, this Court has necessarily accepted as constitutionally legitimate the simple reality that the prosecutor's interest at the bargaining table is to persuade the defendant to forgo his right to plead not guilty. (302)
Thus, in a departure from its unconstitutional conditions jurisprudence, (303) the Court allows prosecutors to condition sentencing or charging deals on the waiver of constitutional trial rights. (304) Prosecutors can obtain plea agreements by threatening criminal defendants with longer sentences or additional charges if they exercise their right to trial. (305) The plea must be knowing and voluntary, so courts review pleas to check that they are not the result of threats of force or promises or threats that are outside the plea agreement itself. (306) But otherwise, prosecutors are free to condition significant sentence and charge reductions on the waiver of judicial process as long as there is a factual basis for the plea. (307) For example, the Supreme Court concluded that it was lawful for a prosecutor to offer to recommend a five-year sentence if a defendant pleaded guilty but to threaten to bring charges subjecting the defendant to a mandatory life sentence if he did not. (308)
Many scholars have criticized plea bargaining on a number of grounds, (309) but they have largely ignored the separation of powers analysis. The dangers of plea bargaining come into full relief when approached in this manner. Prosecutors use the judicial process--the very means of checking the prosecutor and Congress--as the key bargaining chip in negotiations. In the classic plea bargaining scenario, if a defendant elects to go to trial, he or she faces a longer sentence and more charges. If the prosecutor charged a defendant $20,000 for going to trial but there was no charge if the defendant pleaded guilty, it would seem obvious that the bargain was unconstitutional. Yet when prosecutors put a different--in some cases, far more costly--price on going to trial, it is currently considered acceptable.
With this bargaining chip in hand, it is not surprising that almost all convictions are the result of pleas. (310) Defendants who do not want to risk facing more charges and longer sentences by exercising their trial rights instead opt to argue the merits of their cases before prosecutors, who decide the charges of which defendants are guilty. (311) As Ronald Wright and Marc Miller observe, "[w]e now have not only an administrative criminal justice system, but one so dominant that trials take place in the shadow of guilty pleas." (312) It is an administrative system where the prosecutor combines both executive and judicial power--posing the very danger the Framers tried to prevent. (313)
If there were institutional and procedural checks on the prosecutor as there are for other administrative actors, perhaps this would not be so troubling. (314) But as Part II explained, these protections are absent in the plea bargaining context. The prosecutor acts with discretion that is almost unmatched anywhere in law. The closest analog to the prosecutor's vast discretion is an agency's decision not to bring an enforcement action, (315) which enjoys a presumption against judicial review and has itself been criticized for allowing arbitrary behavior. (316) But an agency's decision not to enforce is akin to a prosecutor's decision not to bring charges. (317) The constitutional system of criminal justice protects the discretionary judgments of all key actors not to proceed criminally. That is why the executive has discretionary pardon and charging power, why the legislature has the freedom not to criminalize conduct, and the jury has the unreviewable power to acquit. Put another way, the law distinguishes between the imposition of punishment and the withholding of punishment. What is different between criminal and civil law is that a prosecutor's decision to bring charges--to affirmatively seek punishment--faces little oversight, whereas affirmative agency action in the civil context to impose fines or mandate a particular action is subject to extensive oversight. In the absence of the institutional and political protections that apply to other agencies, there is a risk of abuse of discretion and arbitrary and capricious behavior in criminal matters.
The real question in cases where defendants plead guilty, then, should not be whether the plea of any individual defendant is voluntary or knowing, but whether there is a sufficient check on prosecutors' use of the bargaining power. If the Court focused on the structural relationship among branches instead of on individual defendants, it would see that there is currently no check at all. Prosecutors have almost unbridled discretion to make or not make these deals in any given case. But this is the kind of unbridled discretionary power that the separation of powers is supposed to prevent. (318) If prosecutors can put a higher price on judicial oversight, the state can selectively target groups and individuals for prosecution in a manner that avoids both political and judicial oversight. (319) The political process will not work because the vast majority of people will be unaffected and will not mobilize to fight against the practice. And the judicial process will not work if the only question in a given case is whether the individual defendant before the Court made the deal knowingly and voluntarily. (320) The Framers recognized dangers such as this and required a strong judicial role in criminal cases to prevent it. A system where upwards of ninety-five percent of all convictions result from pleas and where prosecutors make all the key judgments does not fit comfortably with the separation of powers. (321)
A full analysis of what the separation of powers requires of plea bargaining and prosecutorial discretion is beyond the scope of this Article, for it raises many difficult and complicated questions. How do you define plea bargaining in a manner that distinguishes it from the prosecutor's ability to choose from among a range of charges? How much of a price, if any, can a prosecutor place on going to trial without violating the checking role that the judiciary is designed to play in the separation of powers? Can the legislature, consistent with the separation of powers, offer a sentencing discount for pleas or the acceptance of responsibility? Can judges exercise discretion to give sentencing breaks to those who plead guilty? If one gets past these questions of scope, there is then the issue of whether plea bargaining, however defined, should be prohibited outright or whether other checks serve the separation of powers interests. For example, if plea bargaining is, to some extent, inevitable, perhaps institutional or structural checks akin to those in the APA would better serve the separation of powers concerns. (322)
The point of this Article is not to answer these questions but to shift the focus to them. For too long the emphasis has been on whether plea bargaining interferes with individual rights. But the structural concerns are just as great and merit close attention. Plea bargaining causes a systemic imbalance of power by allowing prosecutors to bypass the check of the judicial process. A focus on structure is necessary to identify the scope of the problem and to identify the appropriate solution.
E. Formalism and Federalism
There is a significant limit to the separation of powers argument advanced here that deserves mention: it applies only to the federal government. It is a matter of state, not federal, constitutional law whether the same infirmities exist in a state system. (323) So, to the extent that similar structural abuses exist at the state level, this argument might not address them. While the states themselves have also shown a commitment to the separation of powers, (324) the arguments for using a bright-line approach to the separation of powers might not apply at the state level. (325)
Although that is a limit to the claim here, it is also a potential advantage. As noted above, strict enforcement of the separation of powers will limit the ability of the government to experiment with novel institutional arrangements to address increasing caseloads and other perceived problems with the criminal justice system. Because this strict enforcement scheme does not necessarily apply to the states, it leaves them free, within their own constitutional limits, to experiment with these mechanisms. They can thus serve as the laboratories of experimentation that Justice Brandeis heralded. (326)
At the same time, strict enforcement of the separation of powers should help to limit the expansion of federal criminal law. Maintaining a division among the branches and requiring compliance with the Constitution's procedures for each branch makes federal criminal procedure more costly than a streamlined process that allows shortcuts. For example, a system dominated by jury trials is undoubtedly more costly than a system dominated by plea bargaining with mandatory sentencing laws applied by judges. Limiting the use of mandatory sentencing guidelines and plea bargaining will therefore impose greater costs on the federal government when it seeks to proceed in a criminal action.
Consequently, if the federal government must internalize the costs of constitutional procedures, it might be less likely to enforce federal criminal laws. It is a common criticism that there are too many federal criminal laws. (327) Although the modern politics of crime make it unlikely that Congress will stop passing federal criminal laws or will eliminate some of those already in existence, increasing the cost of processing a criminal case should change enforcement practices. If federal prosecution becomes more costly, it would create incentives for prosecutors to be more selective in the use of federal resources. (328) The current system of plea bargaining makes federal criminal enforcement relatively cheap---at least relative to jury trials--so prosecutors need not pay as much attention to resource allocation and need not focus on those instances when federal involvement in criminal matters is most advantageous. If federal enforcement of criminal laws becomes more expensive through stricter adherence to the separation of powers, executive branch officials and prosecutors should pay more attention to how federal resources should be used.
One need look no further than the states to see how limited resources can help foster a more reasoned approach to criminal justice issues. Because the states have more limited budgets than the federal government and cannot carry deficits, they pay more attention to the soundness of their criminal justice policies. (329) Making federal criminal power more costly should have, at least to some extent, a similar effect. To be sure, the political value of a symbolic gesture might lead Congress to continue to pass more federal laws in the face of some high-profile local crime or to enact laws with harsher punishments to appear tough. But making federal intervention more costly should lead prosecutors to prioritize those areas where federal involvement fills a gap left by the states and where interested groups will specifically target the federal government for assistance. There should be fewer resources left over for intervention in areas that the states already address.
Enforcing separation of powers could therefore serve federalism values (330) without requiring the kind of intrusive substantive review that takes place when the Court enforces federalism through the Commerce Clause. Put another when increased costs are in a very real sense the point of separation of powers. (331) Federal criminal law enforcement should be expensive enough that the government has to think about where and when to use it. If it comes cheaply, it will be used too often, and the political process will be unable to stop it. One need look no further than the current incarceration rates for evidence of this phenomenon. As plea bargaining has increased, so have the incarceration rates. It seems to be more than a coincidence that this rate correlates with a plea bargaining process that combines the efficiency of the administrative model with none of the checks.
And while enforcing the separation of powers may make federal criminal power more costly, it does not make it impossible. It simply places a greater emphasis on prioritization. Moreover, if the federal government is sufficiently concerned that the system the Framers established has become too dangerous and too costly, then it could fix it by amending the Constitution to allow plea bargains, bench trials, or some other streamlined system. But the Framers had the foresight to set the default rules to protect minority interests that could be subject to abuse by political majorities. (332) At the very least, we should have to think long and hard before we abolish that system in the name of convenience. (333)
The Supreme Court's separation of powers jurisprudence has been criticized on any number of grounds. But what has been overlooked is its blanket approach to these questions without attention to the differences in substantive categories. Crime, in particular, raises concerns distinct from those present in matters associated with the oversight of the administrative state. The government does not face the same structural, institutional, and political checks when it proceeds criminally as when it proceeds in a civil regulatory action, so the Constitution's separation of powers takes on greater significance in the criminal context because it provides the only effective check on systemic government overreaching.
Unfortunately, the Supreme Court and scholars have overlooked the importance of separation of powers in the criminal context. The result has been a flexible approach to governmental blending of powers that has allowed innovations like the independent counsel law and the Sentencing Guidelines, as well as a pervasive system of plea bargaining in which prosecutors operate virtually unchecked. Without strong enforcement of the separation of powers or other institutional checks to take its place, the government thus faces far less oversight when it proceeds in a criminal matter than in a regulatory one.
Greater enforcement of the Constitution's separation of powers would prevent this perverse state of affairs. It would require Congress and the executive branch to internalize the costs of the Constitution's judicial protections in making their decisions, thus replacing the current system in which they are free to pressure defendants to forego judicial process. It would restore the checking function of judges and juries and would require that all the key players--Congress, the executive, judges, and juries--agree before an individual is convicted of a federal offense. At the same time, greater enforcement of the separation of powers would still give Congress the freedom to adapt and adjust substantive criminal laws and sentences as it sees fit. (334) In addition, the approach to the separation of powers advocated here has the virtue of having proven itself to be a viable methodology. The Court has already enforced a strict division of power in many administrative contexts other than crime. There is all the more reason to use it in the criminal context, where the stakes are higher and the potential for abuse is so much greater.
(1.) As James Madison declared in the Federalist Papers, "The accumulation of all powers legislative, executive and judiciary in the same hands, whether of one, a few or many, and whether hereditary, self appointed, or elective, may justly be pronounced the very definition of tyranny." THE FEDERALIST NO. 47, at 324 (James Madison) (J. Cooke ed., 1961).
(2.) Myers v. United States, 272 U.S. 52, 293 (1926) (Brandeis, J., dissenting). Burr Neuborne colorfully puts it this way:
[T]he principle of "negative separation" views government power as a bomb so potent that no single organ can be trusted with the formula.... Negative separation argues that the functions of government be carefully labeled and rigorously parceled out to distinct governmental bodies as a prophylactic against threats to individual liberty.
Burt Neuborne, Judicial Review and Separation of Powers in France and the United States, 57 N.Y.U.L. REV. 363, 372 (1982).
(3.) See GORDON S. WOOD, THE CREATION OF THE AMERICAN REPUBLIC, 1776-1787, at 604 (2d ed. 1998) (observing that the separation of powers "was designed to provide for the safety and ease of the people, since 'there will be more obstructions interposed' against errors and frauds in government") (quoting John Dickinson); Philip B. Kurland, The Rise and Fall of the "Doctrine" of Separation of Powers, 85 MICH. L. REV. 592, 601 (1986) ("[T]he underlying, if unstated, premise of all theories of separation seems to have been a minimalist government.").
(4.) As James Landis observed,
when government concerns itself with the stability of an industry it is only intelligent realism for it to follow the industrial rather than the political analogue. It vests the necessary powers with the administrative authority it creates, not too greatly concerned with the extent to which such action does violence to the traditional tripartite theory of government organization.
JAMES M. LANDIS, THE ADMINISTRATIVE PROCESS 11-12 (1938).
(5.) 5 U.S.C. [section] 553 (2006).
(6.) [subsection] 554, 556-557.
(7.) [section] 706.
(8.) See infra text accompanying notes 27-34.
(9.) See, e.g., Commodity Futures Trading Comm'n v. Schor, 478 U.S. 833 (1986) (allowing agency adjudication of state law counterclaims); Humphrey's Ex'r v. United States, 295 U.S. 602, 627-28 (1935) (approving of the FTC's "quasi-legislative" and "quasi-judicial" powers, in addition to its executive powers); J.W. Hampton, Jr. & Co. v. United States, 276 U.S. 394, 409 (1928) (allowing legislative delegations as long as there is an intelligible principle for the delegatee to follow).
(10.) See, e.g., Motor Vehicle Mfrs. Ass'n, Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 40-42 (1983) (providing hard-look review of agency rules); Universal Camera Corp. v. NLRB, 340 U.S. 474 (1951) (requiring that agency findings be supported by substantial evidence); see also Morgan v. United States, 298 U.S. 468 (1936) (requiring agency decisionmaker to consider evidence presented).
(11.) For a small sampling of this vast literature, see DAVID SCHOENBROD, POWER WITHOUT RESPONSIBILITY: HOW CONGRESS ABUSES THE PEOPLE THROUGH DELEGATION (1993); Peter H. Aranson et al., A Theory of Legislative Delegation, 68 CORNELL L. REV. 1 (1982); Steven G. Calabresi & Joan L. Larsen, One Person, One Office: Separation of Powers or Separation of Personnel?, 79 CORNELL L. REV. 1045 (1994); Elena Kagan, Presidential Administration, 114 HARV. L. REV. 2245 (2001); Lawrence Lessig & Cass R. Sunstein, The President and the Administration, 94 COLUM. L. REV. 1 (1994); John F. Manning, Constitutional Structure and Judicial Deference to Agency Interpretations of Agency Rules, 96 COLUM. L. REV. 612 (1996); Peter H. Schuck, Delegation and Democracy: Comments on David Schoenbrod, 20 CARDOZO L. REV. 775 (1999); Peter L. Strauss, The Place of Agencies in Government: Separation of Powers and the Fourth Branch, 84 COLUM. L. REV. 573 (1984).
(12.) See, e.g., Harold H. Bruff, Presidential Powers and Administrative Rulemaking, 88 YALE L.J. 451 (1979); Lessig & Sunstein, supra note 11 ; Strauss, supra note 11.
(13.) See generally Steven G. Calabresi & Saikrishna B. Prakash, The President's Power To Execute the Laws, 104 YALE L.J. 541 (1994); Stephen L. Carter, Constitutional Improprieties: Reflections on Mistretta, Morrison, and Administrative Government, 57 U. CHI. L. REV. 357 (1990); Stephen L. Carter, The Independent Counsel Mess, 102 HARV. L. REV. 105 (1988); Gary Lawson, The Rise and Rise of the Administrative State, 107 HARV. L. REV. 1231 (1994); Lee S. Liberman, Morrison v. Olson: A Formalistic Perspective on Why the Court Was Wrong, 38 AM. U. L. REV. 313 (1989).
(14.) See infra Part I.
(15.) See id.
(16.) Indeed, because administrative state questions have been the prototype, most scholars have addressed the question of separation of powers as a critical inquiry for administrative law. Thus, constitutional separation of powers is standard fare in administrative law textbooks and classes.
(17.) See, e.g., Stephanie A.J. Dangel, Is Prosecution a Core Executive Function? Morrison v. Olson and the Framers' Intent, 99 YALE L.J. 1069 (1990); Neal Devins, Congressional Factfinding and the Scope of Judicial Review, A Preliminary Analysis, 50 DUKE L.J. 1169 (2001); Lessig & Sunstein, supra note 11; William V. Luneburg, Civic Republicanism, The First Amendment, and Executive Branch Policymaking, 43 ADMIN. L. REV. 367 (1991).
(18.) See, e.g., infra Part II.A.
(19.) See JERRY L. MASHAW, GREED, CHAOS, AND GOVERNANCE: USING PUBLIC CHOICE To IMPROVE PUBLIC LAW 107 (1997) (discussing the separation of powers challenges posed by the administrative agency); Jody Freeman, The Private Role in Public Governance, 75 N.Y.U.L. REV. 543, 545 (2000) (noting that "[a]gencies can claim ... only a dubious constitutional lineage" because "the framers made no explicit provision for them, but instead divided power among the legislative and judicial branches and a unitary executive"); Strauss, supra note 11, at 604 (stating that "the actual text of the Constitution says little about the structure of the federal government beneath the apex").
(20.) As G. Edward White has observed, "The general justification for judicial monitoring in separation of powers cases ... is to prevent the evils of tyranny and arbitrariness associated with unconstrained invasions of individual liberty by government." G. EDWARD WHITE, THE CONSTITUTION AND THE NEW DEAL 96-97 (2000). For an excellent discussion of how the separation of powers aims to serve individual liberty, see Rebecca L. Brown, Separated Powers and Ordered Liberty, 139 U.PA. L. REV. 1513 (1991).
(21.) See Rachel E. Barkow, Administering Crime, 52 UCLA L. REV. 715, 723-30 (2005) (comparing interest-group dynamics of criminal sentencing with other regulatory contexts).
(22.) As Bill Eskridge observes, "you can have a functionalist argument for a formalist punch line." William N. Eskridge, Jr., Relationships Between Formalism and Functionalism in Separation of Powers Cases, 22 HARV. J.L. & PUB. POL'Y 21, 26 (1998).
(23.) See Gerard E. Lynch, Our Administrative System of Criminal Justice, 66 FORDHAM L. REV. 2117, 2120 (1998) ("[F]or most defendants the primary adjudication they receive is, in fact, an administrative decision by a state functionary, the prosecutor, who acts essentially in an inquisitorial mode.").
(24.) Of course, these terms can have many meanings. For a description of "formalism" and "functionalism" as those terms are used in the context of separation of powers, see generally Peter L. Strauss, Formal and Functional Approaches to Separation-of-Powers Questions--A Foolish Inconsistency ?, 72 CORNELL L. REV. 488 (1987).
(25.) See Eskridge, supra note 22, at 21 ; Strauss, supra note 24, at 489.
(26.) See, e.g., INS v. Chadha, 462 U.S. 919, 959 (1983) ("[T]he Framers ranked other values higher than efficiency.").
(27.) Id. at 952.
(28.) U.S. CONST. art I, [subsection] 1, 7.
(29.) Justice Powell's concurrence followed formalist reasoning to a different conclusion. He determined that the one-House veto exercised in Chadha was an exercise of judicial power, and as such it could not be exercised by Congress. Chadha, 462 U.S. at 960 (Powell, J., concurring).
(30.) Bowsher v. Synar, 478 U.S. 714 (1986).
(31.) Id. at 733-34.
(32.) 524 U.S. 417, 438 (1998) (concluding that the President's cancellation power under the law amounted to the power to repeal the law, which the Court held must conform to the procedures in Article I). As Justice Kennedy observed in his concurring opinion, "It]he Constitution's structure requires a stability which transcends the convenience of the moment." Id. at 449 (Kennedy, J., concurring). The Court's leading formalist, Justice Scalia, dissented because he believed that the Line Item Veto Act did comply with the procedures of Article I and that the President's power to cancel items raised a delegation question, not an Article I question. Id. at 463-64 (Scalia, J., concurring in part and dissenting in part). Justice Scalia thought the delegation doctrine was flexible enough to permit the President's authority under the Act. Id. at 472-73.
(33.) Metro. Wash. Airports Auth. v. Citizens for Abatement of Aircraft Noise, Inc., 501 U.S. 252 (1991). In classic formalist analysis, the Court reasoned that if the Board's power was executive, "the Constitution does not permit an agent of Congress to exercise it." Id. at 276. And if the Board's veto power was deemed legislative, "Congress must exercise it in conformity with the bicameralism and presentment requirements of Art. I, [section] 7." Id.
(34.) Id. (quoting Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 635 (1952) (Jackson, J., concurring)). In addition to these cases, a plurality of the Court also embraced formalist reasoning to strike down the Bankruptcy Reform Act's broad grant of jurisdiction to non-Article III judges. See N. Pipeline Constr. Co. v. Marathon Pipe Line Co., 458 U.S. 50 (1982). But a majority of the Court ultimately rejected the approach of Northern Pipeline in Commodity Futures Trading Commission v. Schor, 478 U.S 833 (1986). Myers v. United States, 272 U.S. 52 (1926), which struck down a congressional restriction on the President's power to remove a postmaster because he was an executive official, is another example of formalism, but it was dramatically limited by three functionalist decisions: Humphrey's Executor v. United States, 295 U.S. 602 (1935), Wiener v. United States, 357 U.S. 349 (1958), and Morrison v. Olson, 487 U.S. 654 (1988).
(35.) See, e.g., Am. Trucking Ass'ns, Inc. v. United States, 344 U.S. 298, 310-13 (1953).
(36.) Justice White observed in Chadha that the majority's conclusion "that all legislative-type action must be enacted through the lawmaking process ignores that legislative authority is routinely delegated to the Executive Branch, to the independent regulatory agencies, and to private individuals and groups." INS v. Chadha, 462 U.S. 919, 984 (1983) (White, J., dissenting).
(37.) Id. at 952.
(38.) See, e.g., Crowell v. Benson, 285 U.S. 22, 50-51 (1932) (allowing agency adjudication).
(39.) See Strauss, supra note 24, at 492 (noting that agencies "exercise all three of the governmental functions the Constitution so carefully allocates among Congress, President, and Court").
(40.) Chadha, 462 U.S. at 953 n.16 (distinguishing legislative veto from agency rulemaking because the latter is "quasi-legislative"). The quasi-legislative language comes from Humphrey's Executor, 295 U.S. at 628, which refers to the powers of the Federal Trade Commission as "quasi legislative or quasi judicial."
(41.) N. Pipeline Constr. Co. v. Marathon Pipe Line Co., 458 U.S. 50, 67-70 (1982) (distinguishing prior cases upholding agency adjudications as involving public rights and then rejecting bankruptcy law that vested authority in non-Article III tribunal to decide adjudications involving private rights). Commodity Future Trading Commission v. Schor, 478 U.S. 833 (1986), later expanded the scope of agency adjudication to include some private rights.
(42.) Schor, 478 U.S. at 856 ("[T]his case raises no question of the aggrandizement of congressional power at the expense of a coordinate branch.").
(43.) Mistretta v. United States, 488 U.S. 361, 382 (1989) (noting that the Court will strike down laws that "either accrete to a single Branch powers more appropriately diffused among separate Branches or that undermine the authority and independence of one or another coordinate Branch"). Functionalists interpret the Constitution's vesting clauses as speaking only to the heads of each branch---Congress, the President, and the Supreme Court--and granting flexibility in Congress to vest other governmental actors with combined powers as long as the result does not interfere with the core functions of Congress, the President, or the Supreme Court. See, e.g., Strauss, supra note 24, at 510-14.
(44.) 478 U.S. at 836-37.
(45.) Id. at 851.
(46.) Id. As the majority noted:
Among the factors upon which we have focused are the extent to which the "essential attributes of judicial power" are reserved to Article III courts, and, conversely, the extent to which the non-Article III forum exercises the range of jurisdiction and powers normally vested only in Article III courts, the origins and importance of the right to be adjudicated, and the concerns that drove Congress to depart from the requirements of Article III.
Id. (citations omitted).
(47.) In addition to Schor, the Court has applied a functional analysis in a case involving the United States Sentencing Commission, the independent counsel statute, and even the treatment of those labeled "enemy combatants." See infra Part I.A; see also Nixon v. Adm'r of Gen. Servs., 433 U.S. 425,443 (1977) (rejecting a separation of powers challenge to law vesting custody of presidential papers in the Administrator of General Services after analyzing "the extent to which [the law] prevents the Executive Branch from accomplishing its constitutionally assigned functions").
(48.) That is not to say that the Court has not provided arguments in support of the use of formalism as a general matter. See, e.g., INS v. Chadha, 462 U.S. 919, 959 (1983) ("[W]e have not yet found a better way to preserve freedom than by making the exercise of power subject to the carefully crafted restraints spelled out in the Constitution."). Scholars have been more expansive in their functional reasons for taking a formalist approach. See, e.g., MARTIN H. REDISH, THE CONSTITUTION AS POLITICAL STRUCTURE 100-02 (1995) (criticizing a functional approach because it "inherently guts the prophylactic nature of the separation-of-powers protections" and offering pragmatic reasons for formalism as a better alternative); Larry Alexander, "With Me, It's All er Nuthin'": Formalism in Law and Morality, 66 U. CHI. L. REV. 530, 534-36 (1999) (arguing that formalism "solves the problems of coordination, expertise, and efficiency" and reduces error and decisionmaking costs); Thomas W. Merrill, Toward a Principled Interpretation of the Commerce Clause, 22 HARV. J.L. & PUB. POL'Y 31, 31 (1998) (endorsing formalism for functionalist reasons). Rick Pildes "suspect[s] that formalists of all stripes, including the classical legal formalists, have always defended formalism, surely at the level of ultimate justification, in terms of purported desirable consequences." Richard H. Pildes, Forms of Formalism, 66 U. CHI. L. REV. 607, 612-13 (1999).
(49.) More often than not, separation of powers questions are not raised at all in criminal law cases, even among the lower courts. Instead, when constitutional issues arise in criminal matters, they tend to be grounded in protections emanating from the Bill of Rights. This is likely an outgrowth of the Warren Court's jurisprudential reforms in criminal procedure. Even as the Court has become more conservative, it has continued to regulate criminal procedure by taking a rights-based approach, so claims are couched in those terms. In turn, constitutional criminal law scholarship is also dominated by a rights-based focus. Bill Stuntz has been the most persuasive voice arguing against the doctrine and the scholarship. He argues that the Court's procedural rights cases have been counterproductive and urges the Court to turn its attention instead to substantive review of criminal laws and sentences under the Eighth and the Fourteenth Amendments. See William J. Stuntz, The Political Constitution of Criminal Justice, 119 HARV. L. REV. 780, 818-46 (2006); William J. Stuntz, The Uneasy Relationship Between Criminal Procedure and Criminal Justice, 107 YALE L.J. 1, 3-6 (1997). Although this Article looks to constitutional structure instead of substantive review, I share Stuntz's view that the rights-based focus of constitutional law is incomplete.
(50.) Although United States v. Nixon, 418 U.S. 683 (1974), involved a presidential challenge to a subpoena issued as part of a criminal case, the separation of powers claim in that case rested on the scope of the executive privilege more than it rested on the relationship among branches. The Court's discussion of separation of powers was also quite brief. See id. at 706-07 (concluding that the President's claimed absolute privilege "as against a subpoena essential to enforcement of criminal statutes on no more than a generalized claim of the public interest in confidentiality of nonmilitary and nondiplomatic discussions would upset the constitutional balance of 'a workable government' and gravely impair the role of the courts under Art. III").
(51.) 487 U.S. 654 (1988).
(52.) 488 U.S. 361 (1989).
(53.) 542 U.S. 507 (2004).
(54.) Pub. L. No. 95-521, 92 Stat. 1867 (1978) (codified as amended at 28 U.S.C. [subsection] 591-599 (2006)).
(55.) 28 U.S.C. [section] 592(c)(2) (2006).
(56.) [section] 592(b)(1).
(57.) [section] 592(g)(1).
(58.) [section] 592 (g)(2).
(59.) The statute gave the Special Division the authority to find that the office of the independent counsel should be terminated if "the investigation of all matters within the prosecutorial jurisdiction of such independent counsel ... have been completed or so substantially completed that it would be appropriate for the Department of Justice to complete such investigations and prosecutions." [section] 596(b)(2).
(60.) [section] 596(a)(1).
(61.) [section] 595(a)(1).
(62.) [section] 595(c).
(63.) There is some debate in the literature over the history of federal prosecution and whether it was an exclusive province of the executive branch. For example, some scholars point out that at the time of the Framing and for some time thereafter, state and private prosecutors initiated prosecutions, and prosecutors were often associated with the judicial branch. See, e.g., William B. Gwyn, The Indeterminacy of the Separation of Powers and the Federal Courts, 57 GEO. WASH. L. REV. 474, 502 (1989) (arguing that the history suggests that prosecution should be seen as both executive and judicial in nature); Joan E. Jacoby, The American Prosecutor in Historical Context, 39 PROSECUTOR 28, 30 (2005) (arguing that early prosecutors were "minor figure[s]" and "primarily judicial and only quasi-executive"); Harold J. Krent, Addressing the Constitutionality of the Independent Counsel Statute: Executive Control over Criminal Law Enforcement: Some Lessons from History, 38 AM. U. L. REV. 275, 286-96 (1989) (describing the role of private and state prosecutors). Saikrishna B. Prakash disputes this historical claim and argues that the historical evidence shows that the President retained control over federal prosecutions. Saikrishna B. Prakash, The Chief Prosecutor, 73 GEO. WASH. L. REV. 521, 546-65 (2005); see also Krent, supra, at 296 (discussing evidence that nonfederal prosecutors could start prosecutions, but noting that federal prosecutors retained the discretion to drop the cases). Regardless of the historical background, the Supreme Court in Morrison conceded that prosecution was an executive power. Morrison v. Olson, 487 U.S. 654, 691 (1988). Moreover, as discussed infra Part II.A, it is critical to maintain separation between judicial and executive power because the judiciary supplies a critical check on prosecutions. See also Jacoby, supra, at 32 (noting that, by the Civil War, the public "began to recognize and ask for a clear and distinct separation between the duties and powers of the prosecutor and those of the courts"); Prakash, supra, at 569 ("To regard prosecution as part of the judicial power in any way, shape, or form, is to nullify one of the Constitution's central features--its judicial safeguard against prosecutorial overreach.").
(64.) Myers v. United States, 272 U.S. 52 (1926).
(65.) Humphrey's Ex'r v. United States, 295 U.S. 602, 631-32 (1935).
(66.) Elizabeth Magill argues that, although Morrison created "an important change in the doctrine," in some sense the outcome was already accepted by Humphrey's Executor because independent agencies often perform some purely executive functions. Thus, Professor Magill contends that "Morrison admitted what had been true in practice." M. Elizabeth Magill, The Rehnquist Court: The Revolution that Wasn't, 99 NW. U. L. REV. 47, 52 (2004). It remains noteworthy that the Court chose a criminal case as the setting in which to abandon the stated test, even if the standard itself was somewhat fictional.
(67.) Morrison, 487 U.S. at 686.
(68.) Id. at 691.
(69.) Id. at 691-92.
(70.) Id. at 705 (Scalia, J., dissenting).
(71.) Id. at 709.
(72.) Id. at 702.
(73.) Id. ("The context of this statute is acrid with the smell of threatened impeachment.").
(74.) Id. at 713.
(75.) Id. at 728.
(76.) Justice Scalia quoted from an amicus brief filed by former Attorneys General who noted that the "'institutional environment of the Independent Counsel'" is worrisome-"'specifically, her isolation from the Executive Branch and the internal checks and balances it supplies.'" Id. at 731 (quoting Brief for Edward H. Levi, Griffin B. Bell, and William French Smith as Amici Curiae 11). The brief listed the dangers of "'too narrow a focus, of the loss of perspective, of preoccupation with the pursuit of one alleged suspect to the exclusion of other interests.'" Id. (quoting Brief for Edward H. Levi, Griffin B. Bell, and William French Smith as Amici Curiae 11). For a discussion of how escalation theory would have predicted that the independent counsel law would create incentives for prosecutors to continue on failing courses of action, see Jerry Ross, Avoiding Captain Ahabs: Lessons from the Office of the Independent Counsel, 35 ADMIN. & SOC. 334 (2003).
(77.) Morrison, 487 U.S. at 711-12, 733 (Scalia, J., dissenting).
(78.) 488 U.S. 361 (1989).
(79.) Although Congress labeled the Sentencing Commission a judicial branch agency, its relationship to the judiciary is tenuous. Although judges serve as members, the agency is not under the control of a judicial body nor does it decide cases or controversies.
(80.) Mistretta, 488 U.S. at 372.
(81.) Id. at 379.
(82.) Id. at 393.
(84.) The Court also rejected the petitioner's claim that the Act's requirement that judges serve on the Commission compromised the integrity of the judiciary. See id. at 397-404.
(85.) Id. at 413 (Scalia, J., dissenting).
(86.) Id. at 417.
(87.) Id. at 427.
(88.) Id. at 426-27.
(89.) Id. at 366 (quoting legislative history of the Sentencing Reform Act).
(90.) 542 U.S. 507 (2004). The government defined an enemy combatant as an individual who was "part of or supporting forces hostile to the United States or coalition partners in Afghanistan and who engaged in an armed conflict against the United States there." Id. at 516 (internal quotations omitted).
Hamdi also raised the question of whether Congress gave the President the authority to detain citizen enemy combatants pursuant to the Authorization for Use of Military Force (AUMF), 115 Stat. 224 (2001), or whether the detention violated 18 U.S.C. [section] 4001 (2006), which provides that "[n]o citizen shall be imprisoned or otherwise detained by the United States except pursuant to an Act of Congress." The plurality concluded that the AUMF gave the President the necessary authority to detain Hamdi. Hamdi, 542 U.S. at 517-18. Justice Thomas, in dissent, concluded that the President had detention authority as part of his war powers. Id. at 579 (Thomas, J., dissenting). Justice Souter, in an opinion joined by Justice Ginsburg, disagreed that the detention was authorized under the AUMF and concluded that the detention was forbidden by [section] 4001(a). Id. at 554-54 (Souter, J., concurring in part, dissenting in part, and concurring in the judgment). Justice Scalia, joined by Justice Stevens, concluded that the detention was unlawful because the AUMF did not suspend the writ of habeas corpus. Id. at 574-75 (Scalia, J., dissenting).
(91.) Id. at 512-14.
(92.) Id. at 513.
(93.) Id. at 527. The government clarified that the "some evidence" standard did not allow the court to weigh evidence, but merely to see if there was any evidence in the record to support the government's conclusion. Id.
(94.) Id. at 535-36 ("reject[ing] the Government's assertion that separation of powers principles mandate a heavily circumscribed role for the courts" and noting that the "some evidence" test is "inadequate"); id. at 553-54 (Souter, J., concurring in part, dissenting in part, and concurring in the judgment) (agreeing with plurality that Hamdi deserved notice, the opportunity on remand to offer evidence that he is not an enemy combatant, and the right to counsel).
(95.) 424 U.S. 319 (1976).
(96.) The Court employed the Mathews test on its own, as the parties did not couch their arguments in those terms. Hamdi, 542 U.S. at 574 n.5 (Thomas, J., dissenting) (observing that the parties did not "cite Mathews even once").
(97.) Id. at 533; id. at 553-54 (Souter, J., concurring in part, dissenting in part, and concurring in the judgment).
(98.) Id. at 533.
(99.) Id. at 533-34.
(100.) Id. at 538. Justice Sourer made clear that be did not agree with the plurality's resolution of these issues. Id. at 553-54 (Souter, J., concurring in part, dissenting in part, and concurring in the judgment).
(101.) Id. at 554-55 (Scalia, J., dissenting).
(102.) Id. at 559 (Scalia, J., dissenting).
(103.) Id. at 554, 562-70 (Scalia, J., dissenting).
(104.) Id. at 577-78 (Scalia, J., dissenting).
(105.) In contrast, a majority of the Court seems to reach a different conclusion when faced with an alleged enemy combatant who is an American citizen captured in the United States. In Rumsfeld v. Padilla, where the alleged "enemy combatant" was an American citizen taken into custody at O'Hare Airport, four Justices made clear that they would not allow the executive to proceed in the same way. 542 U.S. 426, 454-55 (2004) (Stevens, J., dissenting, joined by Justices Souter, Ginsburg, and Breyer). Those votes, coupled with Justice Scalia's in Hamdi, 542 U.S. at 554 (Scalia, J., dissenting), seem to make five votes in favor of Padilla's argument that the executive cannot detain a citizen captured in the United States on the claim that he is an enemy combatant.
(106.) See supra notes 27-34.
(107.) That is, if the purpose of separation of powers is, as Rebecca Brown has argued, to protect individual due process interests, there is a greater need for enforcement in the context of criminal actions. See Brown, supra note 20, at 1516 (stating that "government action that jeopardizes government process poses a concomitant danger to individual rights and that the potential for such danger should be a significant factor in separation-of-powers analysis").
(108.) That would, in fact, be contrary to the Court's demand for greater procedures in criminal cases than in other contexts. Compare Goldberg v. Kelly, 397 U.S. 254, 270-71 (1970) (not requiting the state to provide counsel in welfare termination proceedings), with Gideon v. Wainwright, 372 U.S. 355 (1963) (requiting the appointment of counsel in criminal proceedings). See generally KENNETH CULP DAVIS, DISCRETIONARY JUSTICE 211-12 (1969) (arguing that there is a greater need for a judicial check over prosecutors than of other administrative or executive functions).
(109.) Morrison v. Olson, 487 U.S. 654 (1988).
(110.) Mistretta v. United States, 488 U.S. 361, 395 (1989).
(111) Hamdi v. Rumsfeld, 542 U.S. 507 (2004).
(112.) It is also interesting that Justice Breyer, perhaps the foremost defender of the administrative state on the Court, tends to treat questions of criminal law structure and design like all other agency questions and tends to side with views that give the executive branch much flexibility. See, e.g., United States v. Booker, 543 U.S. 220 (2005) (Breyer, J., delivering the opinion of the Court in part and dissenting in part); Hamdi v. Rumsfeld, 542 U.S. 507 (2004) (Breyer, J., joining the majority); Blakely v. Washington, 542 U.S. 296, 328 (2004) (Breyer, J., dissenting); Harris v. United States, 536 U.S. 545, 569 (2002) (Breyer, J., concurring in part and concurring in the judgment); Apprendi v. New Jersey, 530 U.S. 466, 555 (2000) (Breyer, J., dissenting).
Parts II.C and III.A, infra, explore the reasons why the Court as a whole has been so receptive to government claims for flexibility in criminal cases.
(113.) Morrison, 487 U.S. at 692 n.31.
(114.) 424 U.S. 319 (1976).
(115.) Indeed, as many scholars have noted, the Court does not divide its cases along any discernible lines. Instead, the Court proceeds on an ad hoc basis. See Brown, supra note 20, at 1517-18 & n.10 (citing those who criticize the lack of coherence in separation of powers jurisprudence).
(116.) For example, scholars critical of the formalist approach do not differentiate among categories and instead assume that a functional analysis is appropriate for all types of cases. Similarly, scholars who support a formalist approach reject all of the functional cases, not just a subset. See Brown, supra note 20, at 1522 ("Most writers in the field have proceeded by selecting the interpretive theory they consider superior [as between formalism and functionalism], then evaluating each of the Court's separation-of-powers decisions against the template of that theory."). An exception seems to be foreign affairs, which is often treated as a separate category for purposes of separation of powers and constitutional analysis more generally. See, e.g., LAURENCE H. TRIBE, AMERICAN CONSTITUTIONAL LAW [section] 4-2, at 634; Laurence H. Tribe, Taking Text and Structure Seriously: Reflections on Free-Form Method in Constitutional Interpretation, 108 HARV. L. REV. 1221, 1249 (1995).
(117.) INS v. Chadha, 462 U.S. 919, 961 (1983) (Powell, J., concurring in the judgment).
(118.) William J. Stuntz, The Pathological Politics of Criminal Law, 100 MICH. L. REV. 505, 547-48 (2001).
(119.) U.S. CONST. art. I, [section] 9, cl. 3; id. [section] 10, cl. 1; see Chadha, 462 U.S. at 962 (Powell, J., concurring in the judgment) (observing that the prohibition on bills of attainder reflects the Framers' "concern that a legislature should not be able unilaterally to impose a substantial deprivation on one person"); Cummings v. Missouri, 71 U.S. 277, 323 (1866) (observing that "[a] bill of attainder is a legislative act which inflicts punishment without a judicial trial" and noting that "[t]hese bills are generally directed against individuals by name; but they may be directed against a whole class").
(120.) 381 U.S. 437, 442 (1965).
(121.) Paul R. Verkuil, Separation of Powers, the Rule of Law and the Idea of Independence, 30 WM. & MARY L. REV. 301,309 (1989).
(122.) As the Court has explained, the prohibition on ex post facto laws "restricts governmental power by restraining arbitrary and potentially vindictive legislation." Weaver v. Graham, 450 U.S. 24, 29 (1981) (citations omitted); see also Calder v. Bull, 3 U.S. (3 Dall.) 386, 390 (1798) (calling the Ex Post Facto Clause an "additional bulwark in favour of the personal security of the subject"). This prohibition "upholds the separation of powers by confining the legislature to penal decisions with prospective effect and the judiciary and executive to applications of existing penal law." Weaver, 450 U.S. at 29 n. 10.
(123.) THE FEDERALIST NO. 84, at 511-12 (Alexander Hamilton) (J. Cooke ed., 1961). Indeed, this was so important that the original Constitution prohibited not only Congress, but the states, from passing ex post facto laws and bills of attainder. U.S. CONST. art. I, [section] 10, cl. 1 ("No state shall ... pass any ... Bill of Attainder, ex post facto Law.... ").
(124.) Lynce v. Mathis, 519 U.S. 433,440 (1997).
(125.) U.S. CONST. art. I, [section] 9, cl. 2 ("The privilege of the Writ of Habeas Corpus shall not be suspended, unless when in Cases of Rebellion or Invasion the public Safety may require it."). Alexander Hamilton praised the writ of habeas in the Federalist Papers because it checks "the practice of arbitrary imprisonments." THE FEDERALIST NO. 84, at 444 (Alexander Hamilton) (J. Cooke ed., 1961).
(126.) Montesquieu warned against combining executive and legislative powers because the executive in that instance would not be able to check the legislature and it would increase the likelihood of tyrannical law execution. BARON DE MONTESQUIEU, THE SPIRIT OF THE LAWS 70 (Frank Neuman ed., Encyclopedia Britannica 1952) (1748). But it does not take a formal alliance for this to occur. As Daryl Levinson points out, the executive and legislative branches often ally with one another because they are members of a common party. Daryl J. Levinson, Empire-Building Government in Constitutional Law, 118 HARV. L. REV. 915, 952-58 (2005).
(127.) M. Elizabeth Magill, The Real Separation in Separation of Powers Law, 86 VA. L. REV. 1127, 1193 (2000) ("[D]ecisions by prosecutors about how to enforce a statute are indistinguishable from lawmaking. That is, given that the range of permissible enforcement actions under criminal laws (and many other laws) is extremely broad, it is the prosecutors' pattern of decisions that shape the meaning of law, not the underlying statute itself.").
(128.) Prakash, supra note 63, at 568 ("Americans viewed judging, in part, as a check on the executive's law enforcement.").
(129.) "The Framers also understood that a principal benefit of the separation of judicial power from the legislative and executive powers would be the protection of individual litigants from decisionmakers susceptible to majoritarian pressures." Commodity Future Trading Comm'n v. Schor, 478 U.S. 833, 860 (1986) (Brennan, J., dissenting); see also Neuborne, supra note 2, at 399 (noting that separation of powers "calls for an independent particularizer with power to resolve disagreements over the meaning of the preexisting general rule and to ascertain the precise facts of the particular case").
(130.) As Philip Kurland notes, the "judiciary was at the cornerstone of the concept of a 'limited constitution' for which separation of powers was to be a guarantee." Kurland, supra note 3, at 599; see also N. Pipeline Constr. Co. v. Marathon Pipe Line Co., 458 U.S. 50, 58 (1982) ("The Federal Judiciary was therefore designed by the Framers to stand independent of the Executive and Legislature--to maintain the checks and balances of the constitutional structure, and also to guarantee that the process of adjudication itself remained impartial.").
(131.) Antonin Scalia, The Rule of Law as a Law of Rules, 56 U. CHI. L. REV. 1175, 1180 (1989).
(132.) Verkuil, supra note 121, at 304-07 (describing the historical roots of the conflict-of-interest concerns motivating the separation of powers and noting that "the notion that no man can be a judge in his own cause was among the earliest expressions of the rule of law in Anglo-American jurisprudence").
(133.) Rachel E. Barkow, Recharging the Jury: The Criminal Jury's Constitutional Role in an Era of Mandatory Sentencing, 152 U. PA. L. REV. 33, 49-50, 56-58 (2003).
(134.) See id. at 48-51 (explaining how this checking function operates); see also id. at 63-64 ("Injecting the jury into the affairs of the judiciary and giving it a nullification power that the judge does not possess gives the people a greater say on how criminal laws are applied.... Not only does this curb the authority of the judges themselves, but it also provides a check on the legislature and executive, which both serve broader constituencies that may not have the same interests as the jury drawn from the community.").
(135.) Kim Taylor-Thompson, Empty Votes in Jury Deliberations, 113 HARV. L. REV. 1261, 1271-72 (2000).
(136.) Barkow, supra note 133, at 64-65 ("The criminal jury provides yet an additional check--one from outside the government itself.").
(137.) The Constitution's concern is with the deprivation of individual liberty, not the granting of privileges to individuals. INS v. Chadha, 462 U.S. 919, 966 n.9 (1983) ("When Congress grants particular individuals relief or benefits under its spending power, the danger of oppressive action that the separation of powers was designed to avoid is not implicated.").
(138.) U.S. CONST. art. II, [section] 2, cl. 1.
(139.) Ex parte Grossman, 267 U.S. 87, 120-21 (1925).
(140.) U.S. CONST. amend. IV.
(141.) U.S. CONST. amend. V.
(143.) U.S. CONST. amend. XIV.
(144.) U.S. CONST. amend. VI.
(146.) U.S. CONST. amend. VIII.
(147.) AKHIL AMAR, THE BILL OF RIGHTS: CREATION AND RECONSTRUCTION (1998). For example, the Fourth and Eighth Amendments, in Amar's view, were designed to place limits on state power in those instances in which the jury could not provide a check. That is, because courts issue arrest warrants, set bail, and sentence without juries, additional protections were needed. Id. at 87.
(148.) For the Framers, the worry was overreaching by the national government. The Fourteenth Amendment expands the protections to cover state abuse of power as well and to place more emphasis on the rights-protecting function of the amendments.
(149.) See Roderick M. Hills, Back to the Future? How the Bill of Rights Might Be About Structure After All, 93 NW. U. L. REV. 977, 996 (1999) (reviewing AMAR, supra note 147) ("[R]ights depend on a set of complex institutions for their defense and definition and without these institutions rights became mere 'parchment barriers.'" (citing JACK N. RAKOVE, DECLARING RIGHTS: A BRIEF HISTORY WITH DOCUMENTS 22-23 (1998)).
(150.) The Court long ago rejected the judiciary's power to recognize a federal common law of crimes. United States v. Hudson & Goodwin, 11 U.S. (7 Cranch) 32, 34 (1812) ("The legislative authority of the Union must first make an act a crime, affix a punishment to it, and declare the Court that shall have jurisdiction of the offence."); see also United States v. Coolidge, 14 U.S. (1 Wheat.) 415 (1816) (reaffirming Hudson & Goodwin).
(151.) The executive must also refrain from exercising the pardon power.
(152.) Further support for the notion of limited federal jurisdiction over crime can be gleaned from the fact that the Constitution vested the federal government with relatively little express jurisdiction over crime. See U.S. CONST. art. I, [section] 8 (giving Congress the power to punish counterfeiting, piracies, and offenses committed on federal territory); id. [section] 8, cls. 6, 10, 17; id. art. III, [section] 3, cl. 1 (defining treason and granting Congress the power to set the punishment for treason). Today, Congress relies on the Commerce Clause as its jurisdictional authority for most federal crimes. Whether Congress's view of its authority under the Commerce Clause is consistent with the Constitution has, of course, been subject to renewed debate in the past decade. See Gonzales v. Raich, 125 S. Ct. 2195 (2005); United States v. Lopez, 514 U.S. 549 (1995).
(153.) Sara Sun Beale, Federalizing Crime: Assessing the Impact on the Federal Courts, 543 ANNALS AM. ACAD. POL. & Soc. SCI. 39, 42 (1996) ("As the scope of federal regulatory activities expanded, the scope of federal criminal laws extended as well.").
(154.) Id. at 41. In the period from 1871 to 1879, the average number of annual criminal cases terminated in federal court was just under 7000. Ronald F. Wright, Trial Distortion and the End of Innocence in Federal Criminal Justice, 154 U. PA. L. REV. 79, 89 tbl.1 (2005). Around this same time, the first cases of plea bargaining began to appear in appellate reports. Jason Mazzone, The Waiver Paradox, 97 NW. U. L. REV. 801, 852 (2003).
(155.) Beale, supra note 153, at 41.
(156.) Wright, supra note 154, at 89.
(157.) Id. at 89 tbl. 89; Ronald F. Wright, Federal Criminal Workload, Guilty Pleas, and Acquittals: Statistical Background (2005), http://ssrn.com/abstract=809124.
(158.) Beale, supra note 153, at 41-42. This same time period witnessed the Court's acceptance of waivers of jury trials for bench trials. See Patton v. United States, 281 U.S. 276 (1930).
(159.) Wright, supra note 154, at 89 tbl.1. Ronald Wright attributes the decline in the number of defendants in the late 1940s and 1950s in part to the decline in the number of immigration cases, Id. at 90.
(160.) CASS R. SUNSTEIN, AFTER THE RIGHTS REVOLUTION 24-31 (1990).
(161.) Beale, supra note 153, at 42-43; see also Gerald G. Ashdown, Federalism, Federalization, and the Politics of Crime, 98 W. VA. L. REV. 789, 792-93 (1996) (noting that, in the two-and-a-half decades since 1971, "[f]ederal criminal jurisdiction thus became virtually limitless"). The average annual number of cases terminated in federal court went from just under 33,000 in the 1960s to just under 47,000 in the 1970s. Wright, supra note 154, at 89 tbl.1. This also coincided with the Court's acceptance of plea bargaining in 1971. See infra note 298 and accompanying text.
(162.) See Exec. Order No. 12,866, 58 C.F.R. 51735 (1983), as amended by Exec. Order 13,258, 67 C.F.R. 9385 (2002).
(163.) See TASK FORCE ON FEDERALIZATION OF CRIMINAL LAW, AM. BAR ASS'N, THE FEDERALIZATION OF CRIMINAL LAW 7 & n.9 (1998) (noting that "[m]ore than 40% of the federal [criminal] provisions enacted since the Civil War have been enacted since 1970" and that "more than a quarter of the federal criminal provisions enacted since the Civil War have been enacted within a sixteen year period since 1980").
(164.) Wright, supra note 154, at 89 tbl.1.
(165.) Cf. Strauss, supra note 11, at 582 (arguing that "the size alone of contemporary American administrative government" was unanticipated by the Framers and puts a strain on the formalist view of separation of powers).
(166.) In 2000, less than seven percent of felony convictions were in federal court. BUREAU OF JUSTICE STATISTICS, U.S. DEP'T OF JUSTICE, SOURCEBOOK OF CRIMINAL JUSTICE STATISTICS--2002, at 421 tbl.5.22, 477 tbl.5.44. Almost ninety-nine percent of violent crime prosecutions occur in state court. Id. at 416 tbl.5.17,447 tbl.5.44.
(167.) Strauss, supra note 24, at 492-93 (noting "[v]irtually every part of the government Congress has created--the Department of Agriculture as well as the Securities and Exchange Commission--exercises all three of the governmental functions" and arguing that the use of such agencies "is unavoidable given Congress's need to delegate at some level the making of policy for a complex and interdependent economy, and the equal incapacity (and undesirability) of the courts to resolve all matters appropriately characterized as involving 'adjudication'").
(168.) See Rachel E. Barkow, Our Federal System of Sentencing, 58 STAN. L. REV. 119, 122-23 (2005).
(169.) Justice White, the leading separation of powers functionalist on the Court, explicitly linked the administrative state and the need for a flexible approach. See, e.g., INS v. Chadha, 462 U.S. 919, 999 (1983) (White, J., dissenting) (noting that a flexible approach to separation of powers is necessary to have an effective government and noting that the "Court, recognizing that modern government must address a formidable agenda of complex policy issues, countenanced the delegation of extensive legislative authority to Executive and independent agencies" (citations omitted)); N. Pipeline Constr. Co. v. Marathon Pipe Line Co., 458 U.S. 50, 113 (1982) (White, J., dissenting) (arguing that it is "too late" to turn to the original intent of Article III given the development of Article I courts and administrative agencies).
(170.) LANDIS, supra note 4, at 11-12.
(171.) Id. at 24 (discussing need for efficiency).
(172.) See, e.g., Bowsher v. Synar, 478 U.S. 714, 761 (1986) (White, J., dissenting) ("[W]ith the advent and triumph of the administrative state and the accompanying multiplication of the tasks undertaken by the Federal Government, the Court has been virtually compelled to recognize that Congress may reasonably deem it 'necessary and proper' to vest some among the broad new array of governmental functions in officers who are free from the partisanship that may be expected of agents wholly dependent upon the President."); Freeman, supra note 19, at 545 ("The combination of executive, legislative, and adjudicative functions in administrative agencies appears to violate the separation of powers principles embodied in the Constitution."); Strauss, supra note 11, at 579 (noting the challenge agencies' combined functions pose for a formalist analysis).
(173.) Indeed, many novel government arrangements raising separation of powers concerns are themselves often developed to serve as checks on agency power. For example, the legislative veto ultimately found unconstitutional in Chadha was designed to check agency power. Chadha, 462 U.S. at 968-69 ("The legislative veto developed initially in response to the problems of reorganizing the sprawling Government structure created in response to the Depression. ... [and] offered the means by which Congress could confer additional authority while preserving its own constitutional role.").
(174.) WHITE, supra note 20, at 94-127 (2000) (explaining how the availability of judicial review was critical to the acceptance of administrative agencies). For example, G. Edward White notes that William Howard Taft observed in 1921 that "judicial review of the actions of administrative agencies could be anticipated as an ongoing feature of the new field." Id. at 100. He also observes that, throughout the 1920s and 1930s, "in each case the eventual creation of a new agency was accompanied by a procedure for judicial review of its decisions that afforded the agencies substantive discretion but allowed due process challenges to its decisionmaking processes." Id. at 102. And, of course, the APA made judicial review a statutory right and imposed additional structural and procedural checks. See also LOUIS L. JAFFE, JUDICIAL CONTROL OF ADMINISTRATIVE ACTION 320 (1965) ("The availability of judicial review is the necessary condition, psychologically if not logically, of a system of administrative power which purports to be legitimate, or legally valid."); Rebecca L. Brown, Caging the Wolf: Seeking a Constitutional Home for the Independent Counsel, 83 MINN. L. REV. 1269, 1277 (1999) (noting that "the existence of judicial review of an independent agency's exercise of delegated power has always been critical to its legitimacy").
(175.) 5 U.S.C. [section] 556(b) (2006).
(176.) [section] 554(d); see also Brown, supra note 20, at 1557 (noting that the separation of executive and judicial functions in agencies has been necessary to "compensate for departures from the structural constitutional norms"); Verkuil, supra note 121, at 316 (noting that the APA's "separation of functions and the use of administrative law judges has done much to allay due process concerns in that setting"). Agency heads, however, are exempt from the structural separation requirement. 5 U.S.C. [section] 554(d)(2)(C) (2006).
(177.) [section] 7521 ; see also [subsections] 3105, 5372 (establishing that administrative law judges "may not perform duties inconsistent with their duties and responsibilities as administrative law judges" and providing salary protections, respectively).
(178.) [section] 557(d)(1).
(179.) [section] 556(d), (e); [section] 557(c). Formal rulemakings, like formal adjudications, must also be based on the evidence in the record.
(180.) [section] 554(b), (c); [section] 557(c).
(181.) [section] 557(c).
(182.) [section] 553(c).
(183.) See Nova Scotia Food Prods. Corp. v. United States, 568 F.2d 240 (2d Cir. 1977) (holding that if an agency relies on scientific data it must make that material available on the record).
(184.) Motor Vehicle Mfrs. Ass'n. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29 (1983).
(185.) 5 U.S.C. [section] 706(2)(A) (2006). The substantial evidence test for formal proceedings serves the same function. See Universal Camera Corp. v. NLRB, 340 U.S. 474 (1951).
(186.) Motor Vehicle Manufacturers Ass'n., 463 U.S. at 40-42; Citizens To Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402 (1971).
(187.) 5 U.S.C. [section] 552 (2006) (mandating the release of government records).
(188.) 5 U.S.C. app. [section] 2 (imposing open meeting requirements on advisory committees and requiring additional disclosures).
(189.) See, e.g., Exec. Order 12,291, 46 Fed. Reg. 13,193 (1981); Exec. Order 12,866, 58 Fed. Reg. 51,735 (1993).
(190.) Richard Revesz and Nicholas Bagley point out that the harmonizing function of OMB has been minimal. Nicholas Bagley & Richard L. Revesz, OMB and the Centralized Review of Regulation 2 (N.Y.U.L. & Econ. Res. Paper No. 05-16, 2005) (draft on file with author).
(191.) See supra notes 173-74; Clinton v. New York, 524 U.S. 417, 489 (1998) (Breyer, J., dissenting) (noting the importance of judicial review and agency rulemaking to the Court's acceptance of broad delegation because they "diminish the risk that the agency will use the breadth of a grant of authority as a cloak for unreasonable or unfair implementation") (citations omitted); McGautha v. California, 402 U.S. 183, 278-79 (1971) (Brennan, J., dissenting) (explaining that delegations of rulemaking and adjudicatory authority "have invariably provided substantial protections to insure against arbitrary action and to guarantee that underlying questions of policy are considered and resolved" and noting the "importance of administrative or judicial review in providing a check on the exercise of arbitrary power") (citations omitted). Scholars embracing a functional approach to separation of powers have also relied on these checks to justify their position. See, e.g., Strauss, supra note 11, at 577 (noting that "a web of other controls--judicial review and legislative and executive oversight ... give reasonable assurance against systemic lawlessness").
(192.) Brown, supra note 20, at 1555 ("[T]o avoid an unconstitutional delegation of legislative power to an administrative agency, Congress 'must enjoin upon the agency a certain course of procedure and certain rules of decision in the performance of its function.'") (citing Wichita R.R. & Light Co. v. Pub. Utils. Comm'n, 260 U.S. 48, 59 (1922)).
(193.) See DAVIS, supra note 108, at 80-96, 188-214; see also Anthony G. Amsterdam, Perspectives on the Fourth Amendment, 55 MINN. L. REV. 349, 380 (1974) (arguing for more rulemaking by the police); Malcolm M. Feeley, Two Models of the Criminal Justice System: An Organizational Perspective, 7 LAW & SOC'Y REV. 407, 423 (1973) (arguing that "the system of criminal justice is a highly formalized and defined set of rules, norms, and goals, but also an organization which possesses no corresponding set of incentives and sanctions which act to systematically enforce them" and suggesting "a solution requiring more bureaucracy, not less"). Even before Davis, Sanford Kadish argued for oversight of the discretion exercised by officials in the administration of criminal law because the administration of criminal law "is not sui generis, but another administrative agency which requires its own administrative law." Sanford H. Kadish, Legal Norm and Discretion in the Police and Sentencing Processes, 75 HARV. L. REV. 904, 931 (1962).
(194.) Scholars from Jerome Hall to Jerome Skolnick have highlighted the problems associated with police discretion. For an overview of this scholarship and a discussion of how theories of democracy and accountability intersect with policing, see David Alan Sklansky, Police and Democracy, 103 MICH. L. REV. 1699, 1736-41 (2005).
(195.) See Stuntz, supra note 49, at 17 (noting that the Court's Fourth and Fifth Amendment decisions "do not just set outer boundaries for police conduct" but instead "constitutional criminal procedure occupies the field").
(196.) Lynch, supra note 23, at 2136-37 ("So long as our criminal codes contain too many prohibitions, the contents of which are left to be defined by their implementation, or which cover conduct that is clearly not intended to be punished in every instance, or which provide for the punishment of those who act without wrongful intent, prosecutors must exercise judgment about which of the many cases that are technically covered by the criminal law are really worthy of criminal punishment.").
(197.) Id. at 2135.
(198.) Id. at 2124 ("Because our governing ideology does not admit that prosecutors adjudicate guilt and set punishments, the procedures by which they do so are neither formally regulated nor invariably followed.").
(199.) Id. at 2128-29.
(200.) Wayte v. United States, 470 U.S. 598, 608 (1985) (noting that courts are "properly hesitant to examine the decision whether to prosecute").
(201.) United States v. Armstrong, 517 U.S. 456, 464 (1996) (quoting Bordenkircher v. Hayes, 434 U.S. 357, 364 (1978)).
(202.) Id. at 464. Vindictive prosecutions are also impermissible. Blackledge v. Perry, 417 U.S. 21, 27 (1974).
(203.) Armstrong, 517 U.S. at 463.
(204.) Id. at 464 (citation omitted).
(205.) Id. at 458.
(206.) Id. at 463-64.
(207.) Hill v. Lockhart, 474 U.S. 52, 56 (1985) (noting that the "longstanding test for determining the validity of a guilty plea is 'whether the plea represents a voluntary and intelligent choice. ... '") (citations omitted).
(208.) FED. R. CRIM. P. 11 (a)(1).
(209.) See Lynch, supra note 23, at 2122; see also Wright, supra note 154, at 82 (describing the "knowing and voluntary" standard as "anemic, since the facts supporting guilty pleas can be remarkably thin, and many 'knowing' and 'voluntary' guilty pleas are nevertheless coercive and unjust").
(210.) Lynch, supra note 23, at 2122. As Lynch further notes, judges in this system typically lack sufficient information to make an informed decision about the defendant's guilt, and the federal rules do not require that a judge determine that the defendant is guilty, "let alone guilty beyond a reasonable doubt." Id. at 2212 & n.5 (citing FED. R. CRIM. P. 11 (f)).
(211.) To be sure, civil regulatory agencies can also engage in something akin to plea bargaining in some limited contexts, for example, by placing conditions on the issuance of a license or by agreeing to a lesser fine or penalty for an enforcement violation if the regulated entity makes certain concessions. Lars Noah, Administrative Arm-Twisting in the Shadow of Congressional Delegations of Authority, 1997 WIS. L. REV. 873, 876. But because the circumstances in which agencies can proceed this way are limited, the agency is checked by the vast majority of instances in which it must proceed in a rulemaking or adjudication subject to arbitrary and capricious review. That is, the agency will almost invariably have announced its policies in many rulemakings or adjudications, which will act as a check on its attempts to engage in so-called arm-twisting behavior with regulated entities. Prosecutors do not have the same check because they do not have to announce their policies in any proceeding governed by court review. Even in the limited cases that go to trial, prosecutors' policies are not challenged. The jury either acquits or convicts on the facts. There is thus no judicial review of how the prosecutor's office proceeds as a general matter.
In addition, unlike plea bargaining, agency "bargaining" is further limited by the fact that the regulated entity, if it chooses not to accept the agency's enforcement offer or licensing conditions, can proceed through the regular agency process and then challenge that outcome in court. The agency is often capped in terms of what it can threaten--a license denial is the worst that it can do in a licensing proceeding, for example. Even in the enforcement context, agencies do not have the same wealth of enforcement provisions from which to choose that prosecutors have. The options stemming from regulatory violations are more limited than the options available in the federal criminal code. Consequently, agency discretion is also limited. Furthermore, even if an agency does have broad discretion to threaten a greater fine if a regulated entity opts to challenge its alleged wrongdoing, the regulated entity is likely to have more resources and political power than a criminal defendant to make sure that agency enforcement is not abusive. See Noah, supra, at 921 (observing that regulated entities are more sophisticated and noting that the stakes are lower in regulatory proceedings); see also infra Part II.B.2 (discussing political process differences between civil regulation and crime). And because the agency's imposition of fines and enforcement practices will be a matter of public record, it will be easier for a regulated entity to make a decision about whether to accept the agency's offer and to challenge arbitrary enforcement practices in court.
(212.) Lynch, supra note 23, at 2135 (noting that in plea bargaining, "the formal adversarial jury trial serves as a kind of judicial review, in which a defendant who is not content with the administrative adjudication by the prosecutor has a right to de novo review of the decision in another forum").
(213.) See infra Parts II.C, III.D.
(214.) Cf. INS v. Chadha, 462 U.S. 919, 966 (1983) (Powell, J., concurring) (concluding that the problem with the legislative veto is that "Congress is not subject to any internal constraints that prevent it from arbitrarily depriving [Chadha] of the right to remain in this country" and "[u]nlike the judiciary or an administrative agency, Congress is not bound by established substantive rules" "[n]or is it subject to the procedural safeguards ... that are present when a court or an agency adjudicates individual rights") (citations omitted). For an insightful discussion of how the Court's formalist opinions protect against arbitrary behavior, see Lisa Schultz Bressman, Beyond Accountability: Arbitrariness and Legitimacy in the Administrative State, 78 N.Y.U.L. REV. 461,519-23, 525-27 (2003) (explaining how the formalist approach taken in Chadha, Bowsher, and Clinton v. City of New York (the lineitem veto case) can be justified as a means for preventing arbitrary action).
(215.) 5 U.S.C. [section] 552b(c)(5) (2006) (exempting information "involv[ing] accusing any person of a crime"); [section] 552b(c)(7) (exempting "investigatory records compiled for law enforcement purposes" but noting that the exemption only applies under certain circumstances, such as if release of the information would interfere with the proceedings, would disclose investigative techniques or procedures, or would deprive a person of a right to an impartial adjudication).
(216.) Lynch, supra note 23, at 2132 ("[U]nlike the opinions of courts, [prosecutorial plea] decisions are not published, permitting discriminatory advantage to defendants represented by 'insider' counsel who are well informed about local prosecutorial practice, and leaving the precise 'holdings' of prior cases subject to reinterpretation, shifting memory, and policy change.").
(217.) Id. at 2141.
(218.) See Norman Abrams, Internal Policy: Guiding the Exercise of Prosecutorial Discretion, 19 UCLA L. REV. 1, 25 (1971) (noting that the general practice of prosecutors is not to publish their policies).
(219.) 28 U.S.C. [section] 994(x) (2006).
(220.) Barkow, supra note 21, at 761.
(221.) Cf. Clinton v. New York, 524 U.S. 417, 490 (1998) (Breyer, J., dissenting) (accepting broad delegation to the President under the Line Item Veto Act because the President is an elected official subject to oversight by the voters).
(222.) See Bagley & Revesz, supra note 190, at 24-25 (citing studies that show regulated industries typically spend more money and file more comments than their public-interest counterparts); Barkow, supra note 21, at 723-24 (describing the typical interest-group scenario); Steven P. Croley, Theories of Regulation: Incorporating the Administrative Process, 98 COLUM. L. REV. 1, 126-41 (1998) (citing studies that show regulated parties have greater access to agencies than public-interest groups).
(223.) Some argue that regulated interests have too much power over agencies and that mechanisms should be put in place to combat agency capture, such as more stringent judicial review. Einer R. Elhauge, Does Interest Group Theory Justify More Intrusive Judicial Review?, 101 YALE L.J. 31, 44-46 (1991) (discussing scholarship advocating more review as a check on agency capture and interest-group influence). But see id. at 66-68, 84 (arguing that interest-group theory does not justify more intrusive judicial review because of a lack of independent normative standards for evaluating the influence of interest groups and because the litigation process is no less defective than the political process). Whether regulated interests have too much power is beyond the scope of this Article. The argument here is simply that the political process is more balanced outside the criminal sphere than within it.
(224.) I have explored the politics of sentencing in greater detail in Barkow, supra note 21, at 723-35; Rachel E. Barkow, Federalism and the Politics of Sentencing, 105 COLUM. L. REV. 1276 (2005). Ronald Wright has explained that the interest-group dynamics vary depending on whether the government is regulating substantive criminal law, policing, adjudication of criminal cases, or punishment. Ronald F. Wright, Parity of Resources for Defense Counsel and the Reach of Public Choice Theory, 90 IOWA L. REV. 219, 255-56 (2004). In particular, he asserts that the process is more balanced when it comes to legislation addressing adjudication. Id. at 259-60.
(225.) Individuals and corporations engaged in legal enterprises who anticipate the danger that their conduct might be seen as crossing the line over to criminality are the ones most likely to pay attention to crime definition and sanctions to protect themselves. Thus, the political process is likely to be most balanced when it comes to regulatory and white-collar crimes.
(226.) As Harold Krent notes, criminal offenders are not only weak because of their status as convicted individuals, but also because they tend to be "poor and disproportionately comprised of minorities," which further weakens the group's political clout. Harold J. Krent, Should Bouie Be Buoyed?: Judicial Retroactive Lawmaking and the Ex Post Facto Clause, 3 ROGER WILLIAMS U. L. REV. 35, 85-86 (1997).
(227.) Barkow, supra note 21, at 725-27 (explaining that those advocating for criminal defendants have little power now but that circumstances may eventually change).
(228.) Federal judges have lobbied for sentencing and substantive criminal law reforms largely through the Judicial Conference. See Judith Resnik, The Programmatic Judiciary: Lobbying, Judging, and Invalidating the Violence Against Women Act, 74 S. CAL. L. REV. 269, 281-83 (2000) (describing Judicial Conference reform efforts on sentencing and criminal justice policy). While those efforts yielded some success in the first half of the twentieth century, id. at 281-82 (describing judicial influence over the 1950 Federal Youth Corrections Act), judges have had little success in recent decades curbing congressional efforts to limit their discretion through mandatory minimum sentences and more stringent Federal Sentencing Guidelines. See id. at 283 ("Between the late 1950s and the 1990s, the Judicial Conference had many times opposed legislation that, despite its commentary, became law."); see also Barkow, supra note 21, at 724-25 (noting that judges do not engage in concerted lobbying in the same way as traditional regulated entities).
(229.) As Bill Stuntz points out, because "it is much cheaper for interest groups to lobby for criminal legislation than against it," the usual political dynamic--"where legislation is easier to block than to generate"--is inverted. William J. Stuntz, Reply: Criminal Law's Pathology, 101 MICH. L. REV. 828, 836 (2002).
(230.) Barkow, supra note 21, at 728 & n.25 (giving examples of prosecutor lobbying).
(231.) Id. at 729.
(232.) Id. at 729-30.
(233.) JEREMY BENTHAM, PRINCIPLES OF PENAL LAW, pt. II, bk. 1, ch. 6 in THE WORKS OF JEREMY BENTHAM 401 (John Bowring ed., 1843) (1962).
(234.) Id. Schumpeter also argued that the content of criminal laws should not be left purely to politics because crime "is a complex phenomenon" that leads to "fits of vindictiveness and ... sentimentality" and where "[p]opular slogans about it are almost invariably wrong." JOSEPH A. SCHUMPETER, CAPITALISM, SOCIALISM AND DEMOCRACY 292 (Harper Perennial 1976) (1942).
(235.) For example, scholars such as Harold Krent and Dan Kahan have noted that this rationale best explains the prohibition on ex post facto laws. Dan M. Kahan, Some Realism About Retroactive Criminal Lawmaking, 3 ROGER WILLIAMS U. L. REV. 95, 112-17 (1997); Krent, supra note 226, at 88-92.
(236.) "[T]he processes embedded in the structure of our institutions [should] be respected whenever the government seeks to act in derogation of values which are vulnerable to majoritarian overreaching." Neuborne, supra note 2, at 437-38.
(237.) See, e.g., Miranda v. Arizona, 384 U.S. 436 (1966) (requiring police to warn suspects of their right to counsel and their right to remain silent); Mapp v. Ohio, 367 U.S. 643 (1961) (applying the Fourth Amendment exclusionary rule to the states).
(238.) Interestingly, the measures required in Hamdi were designed to give the courts some degree of oversight over the process. Hamdi v. Rumsfeld, 542 U.S. 507 (2004).
(239.) For example, the Court may have rejected the budget reforms in Bowsher and the line-item veto in Clinton because the budget process is largely immune from judicial review. Clinton v. New York, 524 U.S. 417, 490 (1998) (Scalia, J., concurring in part and dissenting in part); Bowsher v. Synar, 478 U.S. 714, 761 (1986) (White, J., dissenting). And in one of the other cases in which the Court employed a functional analysis, the continuing role of judicial review was critical to the Court's decision. Commodity Future Trading Comm'n v. Schor, 478 U.S. 833,860 (1986) (allowing agencies to adjudicate private rights claims).
(240.) Clinton, 524 U.S. at 450 (Kennedy, J., concurring).
(241.) Akhil Amar, the leading proponent of the view that the Bill of Rights protects the same structural values as the original Constitution, does not argue that the Bill of Rights renders the original constitutional protections irrelevant. Rather, he argues that they are designed to complement those protections. See Akhil Reed Amar, The Bill of Rights as a Constitution, 100 YALE L.J. 1131, 1132 (1991).
(242.) "[L]iberty demands limits on the ability of any one branch to influence basic political decisions." Clinton, 524 U.S. at 450-51 (Kennedy, J., concurring).
(243.) That is why "Montesquieu emphasized the importance of judicial procedures, even when costly or cumbersome, as a protection for the individual from this type of harm-as a guarantor of 'liberty.'" Brown, supra note 20, at 1536.
(244.) To be sure, to the extent the Court interprets individual rights protections to require the adjudication by the right constitutional actor, it will be serving the same interests as the separation of powers. So, for example, if the Court insists that the defendants must have a jury trial, that will have the effect of precluding the legislature or executive from usurping the judicial function. But not all individual rights protections work in this way, and the danger is that Congress or the executive will find ways around trial protections by evading trial itself.
(245.) As Richard Epstein has pointed out, this logic explains the unconstitutional conditions doctrine. Although an individual might find it in his or her interest to waive a constitutional right, the danger is that as a group, they are worse off. Because of structural concerns of monopoly government power, collective action problems, and externalities, the unconstitutional conditions doctrine does not allow most rights to be the subject of bargains. Richard A. Epstein, Unconstitutional Conditions, State Power, and the Limits of Consent, 102 HAZY. L. REV. 4 (1988). This doctrine has not been applied to criminal trial rights, see infra Part III.D, so the structural dangers that the unconstitutional conditions doctrine polices are prevalent.
(246.) Assuming that criminal justice expenditures will be relatively constant in a regime with or without plea bargaining, plea bargaining frees up resources that would otherwise be spent on trial process and allows these resources to be used for incarceration instead. Conversely, the elimination of plea bargaining would require the shortening of sentences to shift funds currently spent on incarceration to the trial process. Stephen J. Schulhofer, Plea Bargaining as Disaster, 101 YALE L.J. 1979, 1993 (1992). At the state level, budget issues are complicated by the fact that law enforcement is funded largely through local budgets while the states themselves fund the prisons. But at the federal level, Congress is the funding source for all criminal justice expenditures.
(247.) See Barkow, supra note 21, at 728 n.25 (citing examples of Department of Justice requests for more stringent sentences because it would make defendants more likely to cooperate with prosecutors); see also Stuntz, supra note 118, at 529-31.
(248.) See Barkow, supra note 21, at 728.
(249.) See, e.g., Douglas A. Berman, Federal Judge Bemoans 90-Year Minimum Sentence, Aug. 24, 2005, http://sentencing.typepad.com/sentencing_law_and_policy/2005/ week34/index.html (noting that Judge Thomas McAvoy lamented the ninety-year sentence meted out to a bank robbery getaway driver while another participant who threatened bank employees with a handgun received no more than forty-three years because of a guilty plea).
(250.) Schulhofer, supra note 246, at 1985 (noting that the conviction of innocents imposes "serious negative externalities" on the public); id. at 2001 (arguing that there is a "social interest in not punishing defendants who are factually innocent ... even if individual defendants would prefer to have that option").
(251.) The argument here is not that the procedural rights must therefore be eliminated. Rather, the point is that they alone cannot protect the interests of defendants and the overall functioning of the criminal justice system.
(252.) See Adrian Vermeule, The Judicial Power in the State (and Federal) Courts, 2000 SUP. CT. REV. 357, 364 (pointing out that "the disagreement between formalists and functionalists is not about constitutional meaning, but rather about a choice between alternative decision-making strategies").
(253.) "Functional analysis ... favors complex arrangements that blend the powers of the branches, which formalism is likely to condemn." Harold H. Bruff, Independent Counsel and the Constitution, 24 WILLAMETTE L. REV. 539, 546 (1988).
(254.) For a helpful typology of forms of formalism, see generally Pildes, supra note 48.
(255.) Being a formalist does not mean being an originalist. See Merrill, supra note 48, at 32 (distinguishing between formalism that is linked with originalism and what he calls a "'conventionalist' approach to constitutional interpretation" that "draw[s] upon a variety of sources that our legal community regards as authoritative" including text, history, precedent, and evolved practices).
But the evolved practices should be viewed with caution to the extent that they resulted from the Court's laissez-faire attitude to separation of powers in the criminal context without concern for proper checks. That is, because many practices in criminal law developed because the Court accepted innovations in the criminal sphere on the assumption that they mirrored administrative law developments, special attention must be paid to ensure that the protections that exist in the administrative context are present in criminal prosecutions. If sufficient protections have not evolved to check those practices, the Court should not permit them to continue until safeguards are in place.
(256.) The Article takes no position on the proper approach to separation of powers questions outside the criminal law context. It is therefore a variant of what my colleague Burt Neuborne has called a "limited" theory of separation of powers--it identifies an area of "fundamental values" that "trigger[s] strict separation of powers review." Neuborne, supra note 2, at 368.
(257.) The outcome in Hamdi would change as well. But because Hamdi does not involve a classic criminal law issue in the same way that Morrison does, it is less helpful for purposes of illustrating how formalism operates in criminal law. Thus, I have omitted further discussion of Hamdi.
(258.) See, e.g., Alexander, supra note 48, at 534-36. Of course, the debate over bright-line rules versus flexible standards is much broader than the scope of this Article. For a general overview, see Louis Kaplow, Rules Versus Standards: An Economic Analysis, 42 DUKE L.J. 557 (1992); Duncan Kennedy, Form and Substance in Private Law Adjudication, 89 HARV. L. REV. 1685 (1976); Scalia, supra note 131; Kathleen M. Sullivan, The Supreme Court, 1991 Term--Foreword: The Justices of Rules and Standards, 106 HARV. L. REV. 22 (1992). This Article takes no position on this larger debate because the argument for formalism presented here rests on a set of virtues unique to the context of criminal law.
(259.) Adrian Vermeule noticed precisely this dynamic in examining state cases with separation of powers claims involving "freestanding claims of judicial power." The state courts employed a functional analysis that, in Vermeule's description, overvalued judicial power and prerogatives. Vermeule, supra note 252, at 357, 360-61, 390. Vermeule distinguishes the claims he analyzed from those involving "specific constitutional provisions that protect or regulate the judiciary's authority and jurisdiction," such as the fight of jury trial. Id. at 357 n.l. Because some of the separation of powers arguments in criminal cases discussed here rest on the jury requirement in the Sixth Amendment and Article III, see infra Part III.C, they fall outside the scope of Vermeule's argument. Additionally, there is little empirical or theoretical reason for believing that, even in the context of more generalized judicial power claims, the same move toward increased power identified by Vermeule in the state cases would apply in federal criminal cases. While Vermeule is right to point out that the "judiciary has better information about, and greater solicitude for, its own interests than about competing social interests," Vermeule, supra note 252, at 402, that fact is likely to mean that the judiciary will pay more attention to the effect criminal cases have on the functioning of the judicial system and less attention to how a system of plea bargaining and prosecutorial sentencing power acts systematically to disadvantage criminal defendants.
(260.) Indeed, the Court expressly mentioned in Schor that "bright-line rules cannot effectively be employed to yield broad principles applicable in all Article III inquiries." Commodity Future Trading Comm'n v. Schor, 478 U.S. 833, 857 (1986).
(261.) Indeed, the Court itself observed that "the power of the Judicial Branch is, if anything, somewhat diminished by the Act." Mistretta v. United States, 488 U.S. 361,395 (1989).
(262.) Strauss, supra note 24, at 515.
(264.) See Rachel E. Barkow, More Supreme than Court? The Fall of the Political Question Doctrine and the Rise of Judicial Supremacy, 102 COLUM. L. REV. 237, 300-19 (2002) (discussing the Court's expansion of its own powers and the increasing lack of deference it gives the judgments of political actors).
(265.) Schor, 478 U.S. at 858-59.
(266.) Mistretta, 488 U.S. at 417-18.
(267.) Morrison v. Olson, 487 U.S. 654 (1989); see also supra Part I.A.1.
(268.) See FREDERICK SCHAUER, PLAYING BY THE RULES 149-55, 158 (1991).
(269.) See Schor, 478 U.S. at 863 (Brennan, J., dissenting) (criticizing the Court's functional approach because it "requires that the legislative interest in convenience and efficiency be weighed against the competing interest in judicial independence" and thus "pits an interest the benefits of which are immediate, concrete, and easily understood against one, the benefits of which are almost entirely prophylactic, and thus often seem remote and not worth the cost in any single case"); Strauss, supra note 24, at 508 (describing the formalist worry that "[t]he balance between concrete gains in efficiency and convenience promised by legislative assignments of matters to agency decision, and the more remote and theoretical benefits of separation of powers, will never appear to favor the latter in any particular case").
(270.) See, e.g., Jackie Gardina, Compromising Liberty: A Structural Critique of the Sentencing Guidelines, 38 U. MICH. J.L. REFORM 345, 373 (2005) (noting that the "federal judiciary is as addicted to plea bargaining as the Department of Justice").
(271.) See Kathleen F. Brickey, Crime Control and the Commerce Clause: Life After Lopez, 46 CASE W. RES. L. REV. 801, 840 (1996) (citing Chief Justice Rehnquist's year-end reports on the judiciary, the Federal Courts Study Committee, the Judicial Conference, and the Judicial Conference's Committee on Long Range Planning as all expressing concern with the burden criminal cases place on the federal courts).
(272.) Functionalism is criticized as a general matter because it "calls for a prediction that cannot accurately be made." M. Elizabeth Magill, The Real Separation in Separation of Powers Law, 86 VA. L. REV. 1127, 1145 (2000). These concerns are exacerbated in the criminal context for the reasons stated above.
(273.) Vermeule, supra note 252, at 391 ("Judges, like other people, become habituated to and invested in the tasks, activities, and procedures they customarily and repetitively perform.").
(274.) Brown, supra note 20, at 1526 (stating that formalism "tends to straitjacket the government's ability to respond to new needs in creative ways"); Strauss, supra note 11, at 620 (noting that a functional inquiry has the advantage of "toler[ating] periodic changes in relative political effectiveness as between President and Congress, Congress and Court, Nation and States").
(275.) Indeed, as Sanford Kadish pointed out, "the common demand [in the early 1940s] for freedom of the administrator to get on with his job free of the harassment of legal imperatives is the same demand made today by those who administer the new penology." Kadish, supra note 193, at 930-31.
(276.) Thomas O. McGarity, The Courts and the Ossification of Rulemaking: A Response to Professor Seidenfeld, 75 TEX. L. REV. 525, 528 (1997) (arguing that judicial and political review mechanisms have caused the agency "rulemaking process to 'ossify' to a disturbing degree"); Thomas O. McGarity, Some Thoughts on "Deossifying" the Rulemaking Process, 41 DUKE L.J. 1385, 1385-86 (1992) (noting that rulemaking has become burdensome because of judicial oversight).
(277.) As the Court observed in Chadha, "[w]ith all the obvious flaws of delay [and] untidiness ... we have not yet found a better way to preserve freedom than by making the exercise of power subject to the carefully crafted restraints spelled out in the Constitution." INS v. Chadha, 462 U.S. 919, 959 (1983); see also id. at 944 ("Convenience and efficiency are not the primary objectives--or the hallmarks--of democratic government.... ").
(278.) Michael A. Fitts, The Foibles of Formalism: Applying a Political "Transaction Cost" Analysis to Separation of Powers, 47 CASE W. RES. L. REV. 1643, 1648 (1997) (arguing that political actors can get around formal requirements through informal means). By way of comparison, Congress responded to Chadha by using alternative measures of control of the executive, including its power to conduct oversight hearings and its authority over appropriations. Id. at 1652.
(279.) Einer Elhauge has pointed out that interest-group dynamics that exist in the political process are likely to replicate themselves in the judicial process. Elhauge, supra note 223, at 70-71. This would suggest that the imbalance that disfavors defendants in the legislative process would repeat itself in the judicial process as well. While an imbalance of power between the government and defendants is a cause for concern in the judicial branch as well, there are reasons to believe the imbalance will not be as great in the judicial sphere. First, defendants are entitled to counsel as a constitutional right in the litigation process, whereas there is no guarantee of representation in the political process. While the prosecution will likely still maintain a resource advantage, it will not be as pronounced an advantage as exists in the legislative sphere.
Second, the judicial process focuses judges on the facts of concrete cases instead of abstract notions of crime, which corrects some of the informational shortcomings and cognitive biases that exist in the political process. See Barkow, supra note 133, at 61-62 (describing the differences that result from viewing criminal law as a legislative matter as opposed to viewing it in the context of a concrete case).
Third, federal judges do not face the same political pressures because they have life tenure and salary protections, so they have a degree of insulation from political pressure to be "tough on crime." This is not to say that political pressure will not exist--it certainly will through, for example, the appointment process. But that is a more indirect form of pressure, in which criminal justice policy will be one of many factors that will affect a judge's selection.
(280.) Ethics in Government Act of 1978, Pub. L. No. 95-521, 92 Stat. 1824 (codified as amended at scattered sections of 2, 5 & 28 U.S.C.)
(281.) 28 U.S.C. [section] 592 (2006).
(282.) [section] 594.
(283.) Heckler v. Chaney, 470 U.S. 821, 832 (1985) (noting that decision not to indict "has long been regarded as the special province of the Executive Branch").
(284.) Note that functionalism with a thumb on the scale for the interests of defendants probably would not have changed the outcome. This is evidenced by the fact that even Rebecca Brown, who explicitly considered whether the independent counsel law threatened the interests of individuals, concluded that the law should survive a separation of powers challenge. Brown, supra note 20, at 1559.
(285.) A focus on separation of powers and legislative interference with the judicial function should also govern the constitutional analysis of recent congressional efforts to limit the discretion of judges and to exercise greater oversight of their decisions. See, e.g., Prosecutorial Remedies and Other Tools To End the Exploitation of Children Today (PROTECT) Act of 2003, Pub. L. No. 108-21, 117 Stat. 650 (limiting downward departures and requiring that the names of judges who issue departures be reported to the House and Senate Judiciary Committees); Defending America's Most Vulnerable: Safe Access to Drug Treatment and Child Protection Act of 2005, H.R. 1528, 109th Cong. (2005) (proposing the elimination of most downward departures).
(286.) Barkow, supra note 133, at 84-102; see also Brown, supra note 20, at 1560 (arguing that the Act "placed the bulk of sentencing decisionmaking in the hands of the prosecutor through a combination of the charging choices available and the mandatory sentencing laws," having the effect of "consolidating in the Executive Branch the power both to prosecute and to sentence").
(287.) Barkow, supra note 133, at 50-65, 77-84 (describing the constitutional and historical basis for the jury's checking function).
(288.) While the Supreme Court has allowed judges to apply laws that allow for discretionary sentencing, the same separation of powers threat is not raised by those laws because judges have the requisite discretion to check the executive and legislative branches. Id. at 70-74. Mandatory laws, in contrast, can be checked only by the nullification power of the jury. Id. at 85-86; see also Bruce A. Antkowiak, The Ascent of an Ancient Palladium: The Resurgent Importance of Trial by Jury and the Coming Revolution in Pennsylvania Sentencing, 13 WIDENER L.J. 11, 24 (2003) (arguing that judges are incapable of checking state abuse in the context of crime and punishment because the court's role is "largely ministerial, relating facts to elements, and without imposing any further judgment as to the necessity of such finding").
(289.) For a fuller description of the constitutional defects of these laws, see Barkow, supra note 133, at 84-102.
(290.) 543 U.S. 220 (2005).
(291.) Blakely v. Washington, 542 U.S. 296 (2004); Apprendi v. New Jersey, 530 U.S. 466 (2000).
(292.) See, e.g., Apprendi, 530 U.S. at 476 (noting that the case turned on the Sixth and Fourteenth Amendments). In Patton v. United States, 281 U.S. 276, 296 (1930), the Court determined that the right to trial by jury was a right of the accused and not part of the structure of government. But the Court has also recognized the public interest in a jury trial, so it does not allow a defendant to waive a jury without the consent of the prosecutor and the judge. Id. at 312; see also Singer v. United States, 380 U.S. 24, 37-38 (1965).
(293.) See Harris v. United States, 536 U.S. 545 (2002) (allowing judges to find facts that trigger mandatory minimum sentences); see also Booker, 543 U.S. at 237. Four of the Justices (Justices Stevens, Souter, Thomas, and Ginsburg) who found the Sentencing Guidelines unconstitutional in Booker would also require juries to find facts that trigger mandatory minimum sentences. Justice Scalia was the fifth vote to strike down the Guidelines in Booker, but he has upheld mandatory minimum laws applied by judges without writing an opinion explaining why a law imposing a mandatory minimum is different from a law that raises the statutory maximum.
(294.) For a critique of the Court's approach, see Barkow, supra note 133, at 38-44; Rachel E. Barkow, The Devil You Know: Federal Sentencing After Blakely, 16 FED. SENT'G REP. 312, 312-13 (2004).
(295.) As it stands now, the only check on state power when laws mandate minimum sentences is the political process itself. Under the Court's current approach, for example, a legislature could make selling crack cocaine a crime punishable by up to life in prison. The jury would have to find that the defendant did, in fact, sell crack cocaine. But then the legislature could pass additional laws that set the defendant's sentencing floor. It could, for example, pass a law that provides that, if a judge finds by a preponderance of the evidence that an individual uses or carries a gun while dealing crack, he gets a mandatory minimum sentence of sixty years. The potential for state abuse here should be apparent. Who gets charged with this law will be entirely at the discretion of prosecutors. And as long as there is enough evidence to pass the preponderance standard, the judge must give this mandatory sentence. No judicial actor has the discretion to ignore the law in a given case if justice would require it. Indeed, the jury check is so anemic that the jury could acquit the defendant of possessing a gun but the defendant would still be subject to the mandatory minimum sentence so long as the jury convicted him on the crack charge and the judge made the requisite finding about the gun.
To be sure, even if the jury were to apply these laws, the check is imperfect because the jury is not told of the mandatory sentence. Whether a prohibition on the defendant's ability to instruct the jury on a mandatory sentence also violates the separation of powers and jury guarantee is beyond the scope of this Article. But for an argument along those lines, see Kristen K. Sauer, Note, Informed Conviction: Instructing the Jury About Mandatory Sentencing Consequences, 95 COLUM. L. REV. 1232, 1249 (1995).
(296.) See Wayte v. United States, 470 U.S. 598, 607 (1985) (acknowledging the prosecutor's discretion to decide what charges to bring).
(297.) For excellent historical overviews of plea bargaining, see GEORGE FISHER, PLEA BARGAINING'S TRIUMPH (2003); Albert W. Alschuler, Plea Bargaining and Its History, 79 COLUM. L. REV. 1 (1979); Lawrence M. Friedman, Plea Bargaining in Historical Perspective, 13 LAW & SOC'Y REV. 247 (1979); John H. Langbein, Understanding the Short History of Plea Bargaining, 13 LAW & SOC'Y REV. 261 (1979).
(298.) 404 u.s. 257 (1971); see also Blackledge v. Allison, 431 U.S. 63, 76 (1977) (noting that "[f]or decades [plea bargaining] was a sub rosa process shrouded in secrecy and deliberately concealed by participating defendants, defense lawyers, prosecutors, and even judges" and that it "was not until [the Court's] decision in Santobello ... that lingering doubts about the legitimacy of the practice were finally dispelled"). The Court foreshadowed Santobello in Brady v. United States, 397 U.S. 742 (1970), in which the Court rejected the claim that it violates the Fifth Amendment for a prosecutor "to influence or encourage a guilty plea by opportunity or promise of leniency and that a guilty plea is coerced and invalid if influenced by the fear of a possibly higher penalty for the crime charged if a conviction is obtained after the State is put to its proof." Id. at 750.
(299.) In other cases, the Court has highlighted other benefits of plea bargaining, but they also boil down to convenience claims. According to the Court, defendants get the benefit of less pretrial incarceration, a speedier disposition, avoidance of the uncertainty of the trial outcome, and "a prompt start in realizing whatever potential there may be for rehabilitation." Blackledge, 431 U.S. at 71. "The public is protected from the risks posed by those charged with criminal offenses who are at large on bail while awaiting completion of criminal proceedings." Id.
(300.) Santobello, 404 U.S. at 260; see also id. at 264 (Douglas, J., concurring) (noting that plea bargains "serve an important role in the disposition of today's heavy calendars").
(301.) Id. at 260.
(302.) Corbitt v. New Jersey, 439 U.S. 212, 221 (1978) (quoting Bordenkircher v. Hayes, 434 U.S. 357, 364 (1978)). This is codified in the Sentencing Guidelines, which allow up to a three-level deduction for defendants who plead guilty, U.S. SENTENCING GUIDELINES MANUAL [section] 3El.l(b) (2004), and up to a four-level deduction for defendants who plead in fast-track jurisdictions, id. [section] 5K3.1.
(303.) The Supreme Court has rejected government attempts to condition the receipt of other government benefits on the relinquishment of constitutional rights on the theory that
[i]t would be a palpable incongruity to strike down an act of state legislation which, by words of express divestment, seeks to strip the citizen of rights guaranteed by the federal Constitution, but to uphold an act by which the same result is accomplished under the guise of a surrender of a right in exchange for a valuable privilege which the state threatens otherwise to withhold.
Frost & Frost Trucking Co. v. R.R. Comm'n, 271 U.S. 583, 593 (1926). For insightful discussions of how the waiver of criminal trial rights has been treated differently from the waiver of other constitutional rights, see Loftus E. Becker, Jr., Plea Bargaining and the Supreme Court, 21 LOY. L.A.L. REV. 757, 776-94, 829-32 (1988) (noting that plea bargaining is treated sui generis by the Court and is inconsistent with the treatment of compelled confessions and conditions on other constitutional rights); Mazzone, supra note 154, at 804-45 (explaining that the there is a general presumption against the government's ability to condition benefits on the waiver of constitutional rights other than criminal process rights, where the presumption is that those rights are subject to bargaining). For a sampling of unconstitutional conditions cases, see, for example, Nollan v. California Coastal Commission, 483 U.S. 825 (1987) (disallowing the conditioning of building permits on the granting of an easement to the public); Elrod v. Burns, 427 U.S. 347 (1976) (rejecting as unconstitutional the discharge of state employees on the basis of party affiliation); Shapiro v. Thompson, 394 U.S. 618 (1969) (holding unconstitutional the conditioning of welfare benefits on a residency requirement).
(304.) United States v. Mezzanatto, 513 U.S. 196, 201 (1995) (noting that defendants can waive in plea agreements, among other things, a double jeopardy defense, the privilege against self-incrimination, the right to a jury trial, and the right to confrontation). Under "ancient doctrine ... the accused could waive nothing." Patton v. United States, 281 U.S. 276, 307 (1930) (internal quotations and citation omitted). According to the Court, this "ancient doctrine" was based on the fear that innocent defendants might be convicted because of process deficiencies. Id. While the Court concluded in Patton that the fears were no longer justified in light of trial protections, a system that is dominated by plea bargaining runs the same risk of innocent defendants pleading guilty. See Wright, supra note 154, at 84 (observing that acquittal rates have dropped over the last thirty years as guilty pleas have increased).
(305.) The Court has also accepted statutory schemes where a defendant faces a greater penalty after a jury trial. See Brady v. United States, 397 U.S. 742 (1970) (holding that a defendant's plea was not involuntary where a defendant avoided the risk of the death penalty under the statute by pleading guilty and avoiding a jury trial).
(306.) FED. R. CRIM. P. 11(b)(2). The judge also sees if there is a factual basis for the plea. Id. 11(b)(3); see also Miller v. Angliker, 848 F.2d 1312, 1320 (2d Cir. 1988) (noting that a plea is valid "if it is not the product of actual or threatened physical harm, mental coercion overbearing the defendant's will, or the defendant's sheer inability to weigh his options rationally").
(307.) Defendants in federal court who waive their right to a jury trial receive, on average, a 300% reduction in their sentence. Gardina, supra note 270, at 348.
(308.) Bordenkircher v. Hayes, 434 U.S. 357, 358 (1978). The Court reasoned that "in the 'give-and-take' of plea bargaining, there is no ... element of punishment or retaliation so long as the accused is free to accept or reject the prosecution's offer." Id. at 363. This varies from the Court's treatment of increased sentences following a conviction after a retrial. In that context, a judge can give a longer sentence after a new trial only based upon "objective information concerning identifiable conduct on the part of the defendant occurring after the time of the original sentencing proceeding." Chaffin v. Stynchcombe, 412 U.S. 17, 25 (1973) (internal quotations and citations omitted). The Court adopted this latter rule so that defendants would not be deterred from raising their rights on appeal because of fear of retaliation. Id.
(309.) See, e.g., John H. Langbein, Torture and Plea Bargaining, 46 U. CHI. L. REV. 3, 12-19 (comparing plea bargaining to medieval European torture); id. at 12 (noting that the "sentencing differential is what makes plea bargaining coercive"); Schulhofer, supra note 246, at 1985-91 (highlighting numerous flaws with plea bargaining, including the conflicts of interest of counsel and the pressure it puts on innocent defendants to plead, which causes negative extemalities on society). Other scholars defend plea bargaining on the basis that individuals--even those who are innocent of the charges against them--should have the right to waive trial and get a discounted sentence. See Frank H. Easterbrook, Plea Bargaining as Compromise, 101 YALE L.J. 1969, 1978 (1992) (arguing that plea bargaining is efficient and based on defendant's autonomy); Robert E. Scott & William J. Stuntz, Plea Bargaining as Contract, 101 YALE L.J. 1909, 1968 (1992) (arguing that plea bargaining should be treated like other contractual arrangements and regulated accordingly).
One scholar who has noted the separation of powers concerns is Donald Dripps, who recently observed that
the actual practice of plea bargaining poses a still worse separation-of-powers problem. For if the prosecutor dominates plea bargaining, and plea bargaining simply is the criminal justice process, the real trial is the one, quite informal and necessarily based mostly on hearsay, at which the prosecutor decides what charges to file and what plea to accept.
Donald A. Dripps, Overcriminalization, Discretion, Waiver: A Survey of Possible Exit Strategies, 109 PENN. ST. L. REV. 1155, 1159 (2005). For an analysis of the jury's role in the system of separated powers, see Barkow, supra note 133, at 46-65. See also Gardina, supra note 270, at 386-88.
(310.) In 2003, 95.7% of cases in the federal system were settled by guilty pleas. U.S. SENTENCING COMM'N, 2003 SOURCEBOOK OF FEDERAL SENTENCING STATISTICS fig.C (2004), http://www.ussc.gov/ANNRPT/2003/Fig-c.PDF (last visited Jan. 22, 2006).
(311.) Gerard E. Lynch, Screening Versus Plea Bargaining: Exactly What Are We Trading Off?, 55 STAN. L. REV. 1399, 1403-04 (2003) ("Most plea negotiations, in fact, are primarily discussions of the merits of the case, in which defense attorneys point out legal, evidentiary, or practical weaknesses in the prosecutor's case, or mitigating circumstances that merit mercy ... [made] to a prosecutor, who assesses their factual accuracy and likely persuasiveness to a hypothetical judge or jury, and then decides the charge of which the defendant should be adjudged guilty."); The Supreme Court 1969 Term, 84 HARV. L. REV. 30, 151 (1970) (discussing plea bargaining and noting that there are "few safeguards" in this process, as "[t]he prosecutor's job does not require him to be assured of guilt beyond a reasonable doubt").
(312.) Ronald Wright & Marc Miller, Honesty and Opacity in Charge Bargains, 55 STAN. L. REV. 1409, 1415 (2003).
(313.) THE FEDERALIST No. 47, at 326 (James Madison) (J. Cooke ed., 1961) ("Were [the power of judging] joined to the executive power, the judge might behave with all the violence of an oppressor.") (internal quotations omitted); THE FEDERALIST No. 78, at 523 (Alexander Hamilton) (J. Cooke ed., 1961) ("[L]iberty ... would have everything to fear from [the judiciary's] union with either of the other departments."). As a plurality of the Court recently noted in Hamdi, "[t]hat even purportedly fair adjudicators 'are disqualified by their interest in the controversy to be decided is, of course, the general rule.'" Hamdi v. Rumsfeld, 542 U.S. 507, 538 (2004) (citing Tumey v. Ohio, 273 U.S. 510, 522 (1927)); see also id. at 545 ("In a government of separated powers, deciding finally on what is a reasonable degree of guaranteed liberty whether in peace or war (or some condition in between) is not well entrusted to the Executive Branch of Government, whose particular responsibility is to maintain security.") (Souter, J., concurring in part, dissenting in part, and concurring in the judgment).
(314.) Judge Lynch notes that imposing formal administrative law rules on prosecutors' offices "would vastly increase the complexity and expense of the prosecutorial agency." Lynch, supra note 23, at 2145. He therefore proposes some modest changes to the current process, namely greater discovery rights for defendants and the right of a defendant to present his or her case to a supervising prosecutor. Id. at 2147-49. While these changes would be an improvement over the current process and are well worth further inquiry, it is not clear why a more complete range of administrative law rules should not apply to prosecutors. Additional processes always require some costs. The real question is whether the benefits outweigh the costs, and there is no reason for believing that the calculus is different for prosecutors than for any other agency. On the contrary, because of the liberty interests at stake in criminal proceedings, one would think the price of additional process is well worth it. I am exploring these topics in a companion article.
(315.) See, e.g., Heckler v. Chaney, 470 U.S. 821 (1985).
(316.) See, e.g., Lisa Schultz Bressman, Judicial Review of Agency Inaction: An Arbitrariness Approach, 79 N.Y.U.L. REV. 1657 (2004); Cass R. Sunstein, Reviewing Agency Inaction After Heckler v. Chaney, 52 U. CHI. L. REV. 653 (1985).
(317.) "[F]ederal courts have traditionally and, to our knowledge, uniformly refrained from overturning, at the instance of a private person, discretionary decisions of federal prosecuting authorities not to prosecute persons regarding whom a complaint of criminal conduct is made." Inmates of Attica Corr. Facility v. Rockefeller, 477 F.2d 375, 379 (2d Cir. 1973).
(318.) "When it comes to law execution, the genius of the separation of powers is that, typically, two branches must independently conclude that some party has violated the law before anyone is punished. The benefit is clearly absent when the executive and judiciary are one and the same." Prakash, supra note 63, at 545 n. 147.
(319.) Cf. Adrian Vermeule, Veil of Ignorance Rules in Constitutional Law, 111 YALE L.J. 399, 426-27 (2001) (pointing out that selective enforcement of the laws by the executive branch poses the same threat that the Ex Post Facto Clause is designed to prevent when it is done through legislative action).
(320.) Unless, of course, "voluntary" is given more bite and forms the basis for the analysis suggested here.
(321.) The threat prosecutors pose to individual liberty would be magnified even further if one were to give the Department of Justice Chevron deference to its interpretations of criminal law. See Dan M. Kahan, Is Chevron Relevant to Federal Criminal Law, 110 HARV. L. REV. 469, 496 (1996) (arguing that "Chevronizing" federal criminal law would "transfer lawmaking authority from the judiciary to the executive" and would "reallocate it within the executive branch itself'). This would erode the judicial role still further, and it would expand prosecutorial power without creating any checks on its exercise.
(322.) As noted, supra note 314, I am pursuing these questions in a separate article.
(323.) See Dreyer v. Illinois, 187 U.S. 71, 84 (1902) ("Whether the legislative, executive, and judicial powers of a state shall be kept altogether distinct and separate, or whether persons or collections of persons belonging to one department may, in respect to some matters, exert powers which, strictly speaking, pertain to another department of government, is for the determination of the state."); Jim Rossi, Institutional Design and the Lingering Legacy of Antifederalist Separation of Powers Ideals in the States, 52 VAND. L. REV. 1167, 1188 (1999) ("[T]he U.S. Constitution fails to dictate a specific form of separation of powers for state governments.").
(324.) See Rossi, supra note 323, at 1190-91 (noting that the "[s]eparation of powers is a bedrock principle to the constitutions of each of the fifty states" and that "[t]he overwhelming majority of modern state constitutions contain a strict separation of powers clause").
(325.) See id. at 1218-22 (noting similarities and differences in state and federal interpretive approaches).
(326.) New State Ice Co. v. Liebmann, 285 U.S. 262, 311 (1932) (Brandeis, J., dissenting).
(327.) Barkow, supra note 133, at 104-05; see also Ashdown, supra note 161, at 802 (criticizing the fact that many of the more than 3000 federal crimes cover conduct prosecutable under state law).
(328.) Although it is also possible that Congress and prosecutors will try to save resources in other ways--such as by streamlining the trial process, see Scott & Stuntz, supra note 309, at 1950 (arguing that abolition of plea bargaining would lead to a truncated trial process)--Article III and the Bill of Rights will set outer limits on what can be done.
(329.) See generally Barkow, supra note 224.
(330.) This argument has parallels to Brad Clark's claim that the separation of powers, by making it more difficult to enact federal laws, serves the values of federalism. Bradford R. Clark, Separation of Powers as a Safeguard of Federalism, 79 TEX. L. REV. 1321, 1324, 1339-41 (2001). The Constitution imposes certain costs on federal criminal enforcement through the separation of powers and the judicial process that protect not only the interests of individuals but the values of federalism as well. For arguments in favor of limited federal jurisdiction over crime, see Barkow, supra note 168, at 103-06.
(331.) Note that the costs of constitutional rights have been accepted outside the area of criminal process rights, where the unconstitutional conditions doctrine does not allow arguments of efficiency and convenience to trump the interests of the constitutional rights in question. Mazzone, supra note 154, at 849.
(332.) The separation of powers works in this regard like a strong canon of construction. It forces political actors to overcome a large hurdle before dismantling the existing rightsprotecting framework.
(333.) It could be argued that a middle ground is possible that would allow some plea bargaining as long as there is sufficient oversight by the judiciary. See, e.g., Scott & Stuntz, supra note 309, at 1930-31, 1959-60 (arguing for more intense judicial scrutiny of plea bargaining outcomes). Just as the administrative state satisfies the separation of powers analysis with its various mechanisms for judicial and political oversight, plea bargaining could also coexist with the Constitution's requirements as long as it was sufficiently regulated. It is beyond the scope of this Article to determine what arrangements might suffice for that purpose, though that is certainly an avenue worth pursuing and one that I am considering in another paper. For present purposes, what is clear is that the unregulated system of plea bargaining that exists today cannot be squared with the separation of powers.
(334.) Congress's power is subject, of course, to constitutional limits, such as the Eighth Amendment and jurisdictional requirements.
Rachel E. Barkow, Associate Professor of Law, New York University School of Law. For helpful conversations and comments on earlier drafts, I owe thanks to Barry Adler, Robert Ahdieh, Anthony Barkow, A.J. Bellia, Lisa Bressman, Darryl Brown, Brad Clark, Jennifer Collins, Peggy Cooper Davis, Norman Dorsen, Lee Fennell, Victor Fleischer, Barry Friedman, David Gans, David Garland, David Golove, Orin Kerr, Daryl Levinson, Wayne Logan, Burt Neuborne, Rick Pildes, Steve Schulhofer, Robert Tsai, Ron Wright, David Zaring, and Kathy Zeiler, as well as to participants in the Emory Law School Faculty Workshop, the UCLA Law School Faculty Workshop, the May Gathering Junior Faculty Conference at the Georgetown University Law Center, the Constitutional Law Workshop at the University of Chicago Law School, and the NYU Law School Summer Workshop. Leslie Dubeck and Julia Stahl provided excellent research assistance. An earlier version of this Article won the 2006 Outstanding Paper Competition sponsored by the Criminal Justice Section of the American Association of Law Schools.