A Fiduciary Theory of Prosecution

A Fiduciary Theory of Prosecution

69 Am. U. L. Rev. 805 (2020).

* Louis Stein Chair, Fordham University School of Law.

** Professor of Law, New York Law School.

We would like to thank the participants in the Legal Ethics and Fiduciary Theory Workshop held at the Kylemore Abbey in Ireland in June 2019 for their insights and for the opportunity to bring these two fields of inquiry together. We would also like to thank the Criminal Justice Schmooze participants for their thoughts on an earlier draft.

Scholars have failed to arrive at a unifying theory of prosecution, one that explains the complex role that prosecutors play in our democratic system. This Article draws on a developing body of legal scholarship on fiduciary theory to offer a new paradigm that grounds prosecutors’ obligations in their historical role as fiduciaries. Casting prosecutors as fiduciaries clarifies the prosecutor’s obligation to seek justice, focuses attention on the duties of care and loyalty, and prioritizes criminal justice considerations over other public policy interests in prosecutorial charging and plea-bargaining decisions. As fiduciaries, prosecutors are required to engage in an explicit deliberative process for making these discretionary decisions. Finally, fiduciary theory offers some insight into prosecutorial regulation by clarifying that both accountability and independence are aimed at aligning prosecutors’ interest with that of the public. This, in turn, leads to the conclusion that proper regulation should aim to maximize both and helps identify when one might be more beneficial than the other.


Generations of scholars have failed to arrive at a unifying theory of prosecution, one that explains the complex role that prosecutors play in our democratic system. This Article draws on a developing body of legal scholarship on fiduciary theory to offer a new understanding that grounds prosecutors’ obligations in their historical role as fiduciaries.1See, e.g., Daniel Markovits, Sharing Ex Ante and Sharing Ex Post: The Non-Contractual Basis of Fiduciary Relations, inPhilosophical Foundations of Fiduciary Law 209, 209–10 (Andrew S. Gold & Paul B. Miller eds., 2014) (contrasting fiduciary law with contract law and arguing that fiduciary law “cannot be understood on the contractarian model”); Paul B. Miller, The Fiduciary Relationship, in Philosophical Foundations of Fiduciary Law, supra, at 63 (defining the fiduciary relationship through a powers-based theory); Deborah A. DeMott, Beyond Metaphor: An Analysis of Fiduciary Obligation, 1988 Duke L.J. 879 (providing an overview of fiduciary principles while arguing that analogizing fiduciary obligation with contract principles is a faulty approach); Frank H. Easterbrook & Daniel R. Fischel, Contract and Fiduciary Duty, 36 J.L. & Econ. 425, 425–27 (1993) (explaining fiduciary duty through a contractual perspective, noting that the duty of loyalty common in fiduciary relationships “replaces detailed contractual terms”); Tamar Frankel, Fiduciary Law, 71 Calif. L. Rev. 795 (1983) (analyzing the history and nature of fiduciary relations as a group rather than by type of fiduciary); Stephen R. Galoob & Ethan J. Leib, Intentions, Compliance, and Fiduciary Obligations, 20 Legal Theory 106 (2014) (investigating the intentional obligation of loyalty in fiduciary relations); D. Gordon Smith, The Critical Resource Theory of Fiduciary Duty, 55 Vand. L. Rev. 1399, 1401 (2002) (providing a uniform theory of fiduciary duty by differentiating between fiduciary and nonfiduciary relations and “rationaliz[ing] the content of fiduciary obligations”). This effort in turn contributes to the scholarship on fiduciary theory. While scholars have used fiduciary theory to analyze the role of public officials, including judges, they have not applied it to prosecutors, who serve a fiduciary role not only as public officials but also as lawyers.2E.g., Evan J. Criddle, Fiduciary Foundations of Administrative Law, 54 UCLA L. Rev. 117, 117 (2006) (applying fiduciary principles to administrative law to minimize abuse of discretion); Ethan J. Leib et al., A Fiduciary Theory of Judging, 101 Calif. L. Rev. 699, 705 (2013) (applying fiduciary principles to the public law context and to the judiciary, specifically); Paul B. Miller & Andrew S. Gold, Fiduciary Governance, 57 Wm. & Mary L. Rev. 513 (2015) (discussing a kind of fiduciary relationship in which the fiduciary is charged with pursuing abstract interests instead of the interests of a person).

We bring the theory to bear on two related but intransigent problems that preoccupy scholars of prosecutorial ethics in the United States. The first of these problems is how prosecutors should make discretionary decisions, especially regarding charging and plea bargaining.3See Rachel E. Barkow, Institutional Design and the Policing of Prosecutors: Lessons from Administrative Law, 61 Stan. L. Rev. 869, 876–78 (2009) (arguing that prosecutors’ offices should take their cue from administrative law by separating functions and increasing supervision); David Alan Sklansky, The Nature and Function of Prosecutorial Power, 106 J. Crim. L. & Criminology 473, 480–83 (2016) (noting that prosecutors have vast discretion and that prosecutorial power has only increased over time). The second is how prosecutors should be held accountable for making these discretionary decisions without compromising professional independence.4See Stephanos Bibas, Prosecutorial Regulation Versus Prosecutorial Accountability, 157 U. Pa. L. Rev. 959, 959–60 (2009) (arguing that prosecutors’ offices should change internal structure and management to regulate prosecutorial discretion); Angela J. Davis, The American Prosecutor: Independence, Power, and the Threat of Tyranny, 86 Iowa L. Rev.393, 408–12 (2001) (arguing that prosecutors utilize unrestrained discretion—that has no historical or constitutional justification—to engage in misconduct that leaves victims with little to no remedy); Daniel C. Richman, Old Chief v. United States: Stipulating Away Prosecutorial Accountability?, 83 Va. L. Rev. 939, 963–64 (1997). In making discretionary decisions, it is unclear how prosecutors ought to identify relevant considerations and balance competing public concerns. Unless we understand how prosecutors should balance these various interests, it is difficult to determine how to hold prosecutors accountable for failing to do so properly. Even if we could agree about what constitutes a failure of discretionary decision making, preserving prosecutorial independence requires some sacrifice in monitoring and accountability.

Recently, scholars have theorized unique qualities about fiduciary relationships. They explain that all fiduciaries have discretionary power over the beneficiary, who is inherently vulnerable.5E.g., Miller, supra note 1, at 69–75. Beneficiaries are asked to trust their important interests to the fiduciary, in part because monitoring costs are usually high.6Leib et al., supra note 2, at 706. Scholars have drawn on this work to develop a theory of fiduciary governance, in which public officials, a different but related brand of fiduciary, often serve an abstract interest on behalf of the public.7See Miller & Gold, supra note 2, at 565–78 (illustrating that while distinct from private fiduciary law, public fiduciary law is applicable to the judiciary, the executive branch, and the legislative branch). Prosecutors fit this mold because they pursue the public’s abstract interest in justice. Scholars and critics have pointed out both prosecutors’ vast discretion and the very real potential for abuse at the public’s expense.8See Davis, supra note 4, at 436–39 (arguing that the “breadth of prosecutorial discretion” available in the charging power leads to selective prosecution and other forms of prosecutorial misconduct that have yet to be adequately addressed by the Supreme Court or through other accountability mechanisms). While scholars have advanced different theories of prosecution in this country,9See Jeffrey Bellin, Theories of Prosecution, 108 Calif. L. Rev. (forthcoming 2020) (arguing that prosecutors owe an obligation to the law); Eric S. Fish, Prosecutorial Constitutionalism, 90 S. Calif. L. Rev. 237, 253–54 (2017) (explaining that prosecutors should enforce constitutional protections for defendants when the adversarial system fails to do so); Bruce A. Green, Why Should Prosecutors “Seek Justice”?, 26 Fordham Urb. L.J. 607, 612 (1999) (arguing from both a historical and contemporary perspective that prosecutors have a duty to seek justice). viewing prosecutors as fiduciaries is more consistent with historical understandings and has a greater practical value in shaping our understandings of what prosecutors should do and how we ought to hold them accountable.

Most agree that prosecutorial discretion is an inevitable aspect of the criminal justice system, but there is little consensus on how prosecutors should prioritize competing concerns. Prosecutors tend to make decisions in an impressionistic way, weighing multiple interests that may be in tension, such as the interests in truth-seeking, legality, deterrence, retribution, proportionality, equality, efficiency, and economy. Respect for the legislature’s judgment in defining particular conduct as a crime may suggest enforcing the criminal law whenever a crime can be proven, but the legislature assumes that prosecutors will not prosecute every guilty person because a punishment in a given case may be unnecessarily harsh or costly. Prosecuting a particular individual may deter that individual and others, but this abstract benefit may not be worth the literal costs of the prosecution, not only to taxpayers but to witnesses, jurors, and others. A conviction followed by imprisonment would achieve deterrence but at a greater cost to both the accused and the public. When multiple interests are in tension, the challenge is to identify some decision-making criteria and processes for prosecutors to employ, beyond simply “taking everything into account,” both to give prosecutors guidance and to give the public a basis for judging how well prosecutors are exercising discretion.

Courts and scholars of prosecution in the United States agree that prosecutors have a duty to seek justice. Fiduciary theory helps make that obligation less amorphous. Prosecutors, this Article argues, are fiduciaries who represent the public but are appointed or elected to pursue a particular abstract public interest, the interest in justice.10See generally Miller & Gold, supra note 2, at 523–24 (explaining that the role of public fiduciaries often involves pursuit of an abstract interest on behalf of the public). Viewing prosecutors as fiduciaries, against the background of fiduciary theory, makes three principal contributions.

First, this analysis focuses attention on the fiduciary duties of care and loyalty that a prosecutor owes to the public as a beneficiary or client. These duties are not incorporated in the duty to seek justice, which speaks to the public’s principal objective in a criminal case. These are further (but underexamined) duties that address the manner in which prosecutors should pursue the public’s objectives. Analyzing duties of loyalty and care shows that ordinary regulatory processes are less robust for prosecutors than for other lawyers and other fiduciaries generally, in that they fail both to define the nature of these duties and to enforce them. From a normative perspective, this leaves a host of unanswered, and potentially controversial, questions about how prosecutors should conduct their work as competent and disinterested public officials and professionals. From a regulatory perspective, the implication is that, for prosecutors, a premium is placed on alternative modes of accountability.

Second, fiduciary theory helps to narrow the appropriate considerations for discretionary decisions. It does so by reminding us that there may be relevant considerations that are not central to the prosecutor’s fiduciary obligation to pursue the public’s interest in justice. There are considerations that are intrinsic to determinations of justice, meaning that they bear directly on the justness of a particular prosecution. These intrinsic considerations include avoiding wrongful convictions; treating people proportionally and equally; and using the process to incapacitate dangerous individuals, deter future offenses, and secure retribution and restitution for victims. Extrinsic concerns, on the other hand, may be relevant to a particular prosecutorial decision but are not central to the justness of the case. These extrinsic concerns include foreign policy implications of a particular prosecution or its intersection with immigration policy. Prosecutors must balance intrinsic concerns in light of the law, traditions, and facts that are necessarily inaccessible to the public. They can also consider public concerns that are extrinsic to the justness of a particular case so long as doing so would not result in injustice. The fiduciary obligation to pursue the public’s interest in justice makes that abstract goal primary and renders all other extrinsic public interests subordinate. That prosecutors must mediate among a constellation of interests,11Id. and give priority to criminal justice interests, limits the extent to which prosecutors with different values can diverge in their approach to decision making.12Cf. Note, The Paradox of “Progressive Prosecution”, 132 Harv. L. Rev. 748, 770 (2018) (“The paradox of ‘progressive prosecution’ is that the criminal legal system is an oppressive institution. Attempting to make the ‘most powerful’ actor in such an institution more progressive seems to miss the point.” (footnote omitted)).

Third, fiduciary theory helps narrow the range of proper prosecutorial regulation. A related challenge in regulating prosecutors is achieving a proper balance between prosecutorial accountability and independence. It is essential to hold prosecutors accountable when they fail to fulfill their obligations because the potential for harm is so grave. For example, prosecutors at times fail to comply with the constitutional duty to provide exculpatory evidence to the defense or abuse their authority in deciding whether to institute criminal charges or to plea bargain.13Bruce Green & Ellen Yaroshefsky, Prosecutorial Accountability 2.0, 92 Notre Dame L. Rev. 51, 59–60 (2016). Although critics call for greater accountability to address these abuses of power,14See Barkow, supra note 3, at 874–84 (outlining prosecutors’ adjudicative and enforcement powers and explaining how the accumulation of these powers is problematic); Davis, supra note 4, at 395–448; Bruce A. Green & Fred C. Zacharias, Regulating Federal Prosecutors’ Ethics, 55 Vand. L. Rev. 381, 469–78 (2002) (discussing the pros and cons of various methods for regulating prosecutorial conduct); Ronald F. Wright, Reinventing American Prosecution Systems, 46 Crime & Just. 395, 396–99 (2017) (explaining prosecutors’ role, along with the complicity of the legislature, in expanding the incarceration rate). both the structure of American government and the rule of law itself require prosecutors, like judges, to be independent of those who might otherwise hold them accountable.15See generally Bruce A. Green & Rebecca Roiphe, Can the President Control the Department of Justice?, 70 Ala. L. Rev. 1 (2018) (arguing that federal prosecutors are—and must be—independent of the President). It is more than just a theoretical concern that if prosecutors were subject to direct oversight and control by other government actors, politicians might seek to use prosecutors to do their partisan or personal bidding, which would undermine prosecutors’ fiduciary responsibility to pursue the public interest in achieving justice.16See id. at 55 (describing how the professionalization of the DOJ emphasized independence of prosecutors from partisan influence); Bruce A. Green & Rebecca Roiphe, May Federal Prosecutors Take Direction from the President?, 87 Fordham L. Rev. 1817 (2019) (discussing the “consequences for prosecutors who receive the president’s orders”). Independence also requires some distance from factions of citizens with well-articulated interests, and it even requires that prosecutors be insulated from (although perhaps not oblivious to) a public consensus in favor of a particular act or outcome.17This is essentially, on a more explicit and grander scale, the kind of independent role that all lawyers are supposed to play. Robert W. Gordon, The Independence of Lawyers, 68 B.U. L. Rev. 1, 9–30 (1988). For a discussion of a judge’s failure to maintain this sort of independence from the public, see Bruce A. Green & Rebecca Roiphe, Punishment Without Process: “Victim Impact” Proceedings for Dead Defendants, FLR Online (2019), https://fordhamlawreview.org/wp-content/uploads/2019/11/Green-Roiphe_November_FLRO_4.pdf [https://perma.cc/QE9N-V5WX].

Striking a balance between prosecutorial accountability and independence is particularly difficult when it comes to charging and plea bargaining because these decisions are both momentous and by nature discretionary. When prosecutors, as trial lawyers, act unlawfully or abusively in the manner in which they conduct criminal investigations and proceedings, courts have constitutional and inherent authority to hold them accountable.18See, e.g., Massameno v. Statewide Grievance Comm., 663 A.2d 317, 322 (Conn. 1995) (rejecting the claim that disciplining the state prosecutor would violate separation of powers); State v. Davis, 972 P.2d 1099, 1105 (Kan. 1999) (affirming the trial court’s order holding the prosecutor in criminal contempt for failing to comply with discovery order). See generally Green & Zacharias, supra note 14, at 405–12 (describing federal courts’ authority to regulate federal prosecutors). As a practical matter, courts tend to be circumspect in their oversight of prosecutors’ investigative and trial conduct, but at least as a legal matter, courts have considerable authority both to define and remedy prosecutors’ trial misconduct and to sanction prosecutors for wrongdoing in their role as advocates.19Green & Zacharias, supra note 14, at 403–05 (explaining how federal courts can directly and indirectly sanction prosecutors by reprimanding them off the record, instituting fines, or negatively interpreting a particular prosecutor’s arguments). In contrast, courts do not regulate prosecutors’ charging and plea bargaining decisions except in the most extreme situations.20See Davis, supra note 4, at 435–37 (describing how the Supreme Court has promoted expansive prosecutorial discretion and discouraged challenges to prosecutors’ abuse of the charging power). These discretionary decisions, which implicate prosecutors’ role as public officials more than as trial lawyers, are free from judicial review largely because judicial interference threatens to undermine prosecutors’ effectiveness and inordinately entangle courts in executive branch decision making. And yet, in many ways, decisions about whether to initiate and dismiss criminal charges are prosecutors’ most significant ones both for individual defendants and for the community and, therefore, are the decisions for which prosecutors most need to be publicly accountable.

In addition to shedding light on how prosecutors should approach decision making, fiduciary theory offers insight into how to achieve the proper balance between accountability and independence in the context of criminal prosecution. While at times in tension, prosecutorial accountability and independence are not contradictory aspirations. Both accountability and independence are mechanisms to promote and protect prosecutors’ core fiduciary duties of care and loyalty. Each addresses risks that threaten to distort prosecutorial adherence to the needs and interests of the public in achieving justice. While fiduciary theory does not provide an algorithm for determining when to emphasize one over the other, it may help to craft an institutional design for prosecutors’ offices that can maximize both.

A key and often controversial question in determining the proper balance between accountability and independence is how much control the public ought to have over prosecutors either directly or through other public monitors. Fiduciary theory may help solve this puzzle in two ways. First, fiduciary theory helps clarify that both aims are designed to ensure loyalty to the client, and second, it helps determine which sorts of decisions need to be insulated from popular input and control. The fiduciary obligation for prosecutors is to pursue the public’s interest in justice. The public should not have direct input in determining and weighing considerations that directly bear on criminal justice in particular cases. These traditional criminal justice questions are embodied in decisional law and the Constitution and developed by the traditions and practices of prosecutors over time. The discretionary power of prosecutors at the core of their fiduciary mission derives from making these sorts of calculations in the best interest of the public rather than at its behest. Extrinsic considerations that might also affect prosecutors’ decisions, such as foreign policy questions or the intersection between prosecution and immigration policy, are less central to prosecutors’ fiduciary mission, and there is no reason why the public either directly or through other elected officials could not have more input into these secondary public policy concerns.

Part I of this Article offers both historical and theoretical bases for drawing on fiduciary theory to explain the U.S. prosecutor’s role. From an historical perspective, prosecutors were, at times throughout early American history, regarded as fiduciaries.21See Miller & Gold, supra note 2, at 565–78 (explaining that public officials often served as fiduciary on behalf of the public). Thought of as repositories of a public trust, prosecutors are, like public officials generally, fiduciaries. But most of these historical references are used as a rhetorical flourish. They do not offer much in terms of content for the unique relationship that prosecutors have to the public. For many of the same reasons that theorists have cast public officials, including judges, as fiduciaries, prosecutors too can be characterized as fiduciaries—that is, as professionals who hold a trust and wield considerable discretion on behalf of a vulnerable beneficiary. Drawing on the fiduciary theory of governance, this Part concludes that while prosecutors’ beneficiary is the public, prosecutors serve the public not by satisfying the preferences of an amalgam of citizens at a particular moment in time but by pursuing the abstract public interest in justice that is, and ought to be, elaborated within prosecutors’ offices over time.

In Part II, we examine prosecutors’ role as a fiduciary, focusing on the core fiduciary duties of care and loyalty. This exercise illuminates the complexity of prosecutors’ role, particularly in making discretionary charging and plea-bargaining decisions. Rather than simply relying on intuition, prosecutors should explicitly and consciously consider particular relevant factors. For the idea of justice to develop more particular meaning over time, the policies and practices that surround the duty of loyalty and care must be articulated, reviewed, and revised in the context of individual investigations and prosecutions. By exploring the complex nature of discretionary decision making, this section highlights the importance of both accountability and independence.

Finally, in Part III, we consider whether fiduciary theory helps determine how to enhance prosecutorial accountability and independence. We conclude with cautious optimism. Fiduciary governance mandates a balance between insulation and responsiveness. Prosecutors should be insulated from direct popular control but only insofar as that allows them to develop stable norms and principles governing decision making. To that end, oversight should include a mechanism for ensuring that prosecutors are explicitly weighing proper concerns in making discretionary decisions and developing more concrete goals and priorities that give substance to the mandate to do justice. That oversight requires not only some degree of transparency but also reciprocity. In other words, prosecutors ought to serve as educators, explaining the value of the norms and traditions that govern their work.

I.   Prosecutors as Fiduciaries—History and Theory

A.   The Historical Background to U.S. Prosecutors’ Fiduciary Role

The shift from private to public prosecutions in America could be seen as a shift from a private service model to a public fiduciary model.22Nicholas R. Parrillo, Fiduciary Government and Public Officers’ Incentives, in Fiduciary Government 146, 156 (Evan J. Criddle et al. eds., 2018) (arguing that the shift in public officials’ compensation from private to public funding demonstrates a shift from a service model toward a fiduciary model of government). In the Middle Ages, crime was originally seen as a personal offense, a wrong inflicted on the victim.23Juan Cardenas, The Crime Victim in the Prosecutorial Process, 9 Harv. J.L. & Pub. Pol’y 357, 359 (1986). This view persisted through the nineteenth century in England and America. Private prosecution in England was justified, in part, as a restriction on the power of the Crown.24Douglas Hay, Controlling the English Prosecutor, 21 Osgoode Hall L.J. 165, 167 (1983). While a limited system of public prosecution developed over time, England did not officially recognize public prosecution until 1879 when the Office of Public Prosecutions was created.25Prosecution of Offences Act, 1879, 42 & 43 Vict. c. 22 (Eng.). In nineteenth-century England, however, even private prosecutions were thought to be brought in the public interest: the prosecutor “has as much a public duty to discharge as the sovereign himself, and has a public trust to exercise.”26Would the Ends of Justice Be Promoted by the Appointment of a Public Prosecutor?, 4 Philomathic J. & Literary Rev. 309, 345 (1826).

The American colonies borrowed significantly from the British system, and early prosecutions were primarily brought by individual victims. Colonists similarly resisted the public prosecution model as fraught with the potential for abuse.27Cardenas, supra note 23, at 368; William F. McDonald, Towards a Bicentennial Revolution in Criminal Justice: The Return of the Victim, 13 Am. Crim. L. Rev. 649, 653 (1976). However, a growing population, increased crime, and the chaos created by laws giving rewards for successful prosecution led to criticism of private prosecution and calls for a public alternative.28Cardenas, supra note 23, at 368–69. Critics of private prosecution emphasized that the system introduced problems, as crime victims and their paid advocates sought unfair financial benefit.29Id. at 369. By the American Revolution, many colonies had moved toward partial public prosecution.30Id. at 371.

Proponents of public prosecution in nineteenth-century America emphasized how the private model vindicated the private interests of parties at the public expense. Increasingly, the prosecutor came to be viewed as a fiduciary of the public at large or of the public interest in the abstract, not as an agent of an individual victim. There was a practical significance to the shift. When the defendant’s guilt was uncertain, a private lawyer might vigorously prosecute out of fidelity to the victim-client, if not as a matter of self-interest, whereas a public prosecutor would be expected to refrain from doing so to avoid convicting an innocent person. In 1888, recognizing this concern, a Wisconsin court declared private prosecution unconstitutional, explaining, “[t]he prosecutingofficer represents the public interests, which can never be promoted by the conviction of the innocent. His object, like that of the court, should be simply justice; and he has no right to sacrifice this to the pride of professional success.”31Biemel v. State, 37 N.W. 244, 247 (Wis. 1888); see also Meister v. People, 31 Mich. 99, 103–04 (1875) (arguing that prosecutors owe a “[d]uty of impartiality” that is inconsistent with the privately funded prosecution). Another court explicitly described the prosecutor’s role as a “public trust, committed by the public to an individual.”32State ex rel. Gibson v. Friedley, 34 N.E. 872, 875 (Ind. 1893); see also State ex rel. Black v. Taylor, 106 S.W. 1023, 1027 (Mo. 1907) (refusing to allow the Attorney General to “farm out” his obligations to private parties because prosecution is a “public trust”); People ex rel. Peabody v. Attorney General, 13 How. Pr. 179, 183 (N.Y. Sup. Ct. 1856) (refusing to grant a writ of mandamus forcing the Attorney General to bring an action on the application of a private party and stating: “Our legislature have seen fit to invest the attorney general with this discretion. His office is a public trust. It is a legal presumption that he will do his duty—that he will act with strict impartiality. In this confidence he has been endowed with a large discretion, not only in cases like this, but in other matters of public concern. The exercise of such discretion is, in its nature, a judicial act, from which there is no appeal, and over which courts have no control”); Commonwealth v. Burrell, 7 Pa. 34, 39 (Sup. Ct. 1847) (“[T]he office of the attorney-general is a public trust . . . .”).

Likewise, one of the founders of modern legal ethics, George Sharswood, conceptualized prosecutors as public trustees—that is, as fiduciaries of the general public. In 1869, he wrote:

There is no obligation on an attorney to minister to the bad passions of his client; it is but rarely that a criminal prosecution is pursued for a valuable private end, the restoration of goods, the maintenance of the good name of the prosecutor, or closing the mouth of a man who has perjured himself in a court of justice. The office of the Attorney-General is a public trust, which involves in the discharge of it, the exertion of an almost boundless discretion, by an officer who stands as impartial as a judge.33George Sharswood, An Essay on Professional Ethics 95 (3d ed. 1869).

While analogizing prosecutors to judges in their obligation of impartiality, Sharswood did not otherwise specify prosecutors’ fiduciary obligations as holders of the public trust.

It would be an overstatement to suggest that the private service model of prosecution was entirely supplanted. Private prosecution persisted for quite some time in the United States, and remnants of it still exist.34Roger A. Fairfax, Jr., Delegation of the Criminal Prosecution Function to Private Actors, 43 U.C. Davis L. Rev. 411, 413–15 (2009). Meanwhile, early concerns about the dangers of political control of prosecution expressed by proponents of private prosecution subsided but did not abate.

At the end of the nineteenth century, professionalism emerged as a way to address concerns that prosecutors would use the State’s criminal justice authority to promote narrow parochial or partisan interests.35Green & Roiphe, supra note 15, at 49–55. Prosecutors, as professionals, would be constrained by ethical norms, experience, and training not only in their actions but also in the process by which they make decisions.36Stephen R. Galoob & Ethan J. Leib, Fiduciary Political Theory and Legitimacy, in Fiduciary Government, supra note 22, at 163, 165–66 (arguing that what distinguishes legitimate fiduciary relationships is limits on the fiduciary’s cognitive function). Just as judges draw on judicial norms, such as those governing the interpretation of statutes and application of precedent, prosecutors too look to evolving written and unwritten prosecutorial traditions.37See Bruce A. Green & Fred C. Zacharias, Prosecutorial Accountability, 2004 Wis. L. Rev. 837, 870–83 (describing prosecutors’ decisions making based on principles derived from legislation, the purposes of criminal law, and elsewhere). These contemporary understandings are consistent with the idea of the prosecutor as a fiduciary who exercises discretion on behalf of others who are vulnerable and dependent. This understanding also offers a rationale for subjecting prosecutors to normative constraints intended to minimize the risk of self-dealing and other abuses of discretion.

B.   Theoretical Background

The theory of fiduciary governance has drawn on the private law and theory of fiduciaries to develop an understanding of the role and responsibilities of public officials, including judges.38See Miller & Gold, supra note 2, at 567–70. As lawyers appearing in court on behalf of a client, prosecutors are fiduciaries in the most classical sense. But like other public officials, prosecutors are also a different sort of fiduciary. Instead of serving a defined interest of a particular beneficiary, this fiduciary administers the law on the public’s behalf in furtherance of an abstract public purpose.39In this role prosecutors are involved in administration for abstract purposes rather than for an individual or set of individuals. Id. at 523. Miller and Gold put it this way: “A fiduciary relationship is one in which one person (the fiduciary) enjoys discretionary power to pursue an abstract other-regarding purpose . . . of another person (an individual beneficiary or ascertained set of beneficiaries).” Id. at 549. In deciding whether to initiate an investigation or criminal charge, to plea bargain or to dismiss a prosecution, prosecutors do not take direction from clients or defer to clients’ objectives. Nor do they engage in ministerial acts. They exercise discretion as other public officials, particularly judges, do.

The fiduciary relationship is defined as one involving discretionary power and structural vulnerability.40Miller, supra note 1, at 69–75. Trustees, for instance, exercise a great deal of discretionary power over the beneficiary, who, as a result, is vulnerable to abuse. A hallmark of the relationship is trust, and monitoring costs are usually high.41Leib et al., supra note 2, at 706. Prosecutors fit this mold. Scholars and critics have pointed out both prosecutors’ vast discretion and the very real potential for abuse at the expense of the public.42See, e.g., Bibas, supra note 4, at 961 (“Prosecutors have great leeway to abuse their powers and indulge their self-interests, biases, or arbitrariness.”); Davis, supra note 4, at 408–16 (surveying “prosecutorial discretion and how it is abused”); Bruce A. Green, Prosecutorial Discretion: The Difficulty and Necessity of Public Inquiry, 123 Dick. L. Rev. 589, 606–18 (2019) (discussing why prosecutorial abuse of discretion is hard to define and detect).

Arguments that judges are fiduciaries are largely applicable to prosecutors as well.43See Leib et al., supra note 2 (applying fiduciary principles in public law to the judiciary). Prosecutors have been described as quasi-judicial officials—“minister[s] of justice.”44See, e.g., Model Rules of Prof’l Conduct r. 3.8 cmt. 1 (Am. Bar Ass’n 2016) (“A prosecutor has the responsibility of a minister of justice and not simply that of an advocate.”). See generally Daniel Epps, Adversarial Asymmetry in the Criminal Process, 91 N.Y.U. L. Rev. 762, 833 (2016) (contrasting “punishment-maximizing” prosecution with “an idealized vision of the prosecutor as minister of justice”); Eric S. Fish, Against Adversary Prosecution, 103 Iowa L. Rev. 1419, 1463–68 (2018) (describing the significance of conceptualizing prosecutors as “minister[s] of justice”). In exercising discretion in the criminal context, prosecutors both determine the relevant public interests and decide how to balance and pursue them in particular factual circumstances. In other words, their role is constitutive in defining the abstract interest they are supposed to serve and instrumental in furthering it in particular cases. Prosecutors do not have a traditional beneficiary who defines the objectives of the fiduciary relationship. They discern, and contribute to developing, the collective understanding of justice as they implement it in any given case.

It is almost universally recognized that U.S. prosecutors’ duty is to “seek justice.”45For a description and explanation for why prosecutors have a duty to seek justice, see Green, supra note 9, at 633–37. But in concrete cases, it is a challenge to give substance and meaning to this vague mandate.46See, e.g., Bellin, supra note 9, at 11 (arguing that the duty to “advocate for justice” is too vague to give prosecutors meaningful guidance, and that greater guidance is provided by conceptualizing prosecutors as “servant[s] of the law”);Bibas, supra note 4, at 961 (“In theory, prosecutors are beholden to the public interest or justice. These concepts, however, are so diffuse and elastic that they do not constrain prosecutors much, certainly not in the way that an identifiable client would.”). Others see prosecutors’ “duty to seek justice” as a useful starting point for ascertaining prosecutors’ particular obligations. See, e.g., Fish, supra note 9, at 305 (“In the many situations where judges are unable to fully implement constitutional protections, prosecutors should step in and perform the task themselves. The theoretical resources for this role can be found in commonplace maxims about prosecutors: they have a duty to ‘seek justice,’ not just obtain convictions, and they are obligated to uphold the Constitution through their oaths of office.”); Fred C. Zacharias,Structuring the Ethics of Prosecutorial Trial Practice: Can Prosecutors Do Justice?, 44 Vand. L. Rev. 45, 46 (1991) (arguing that prosecutors’ duty to do justice implies specific obligations to compensate for the inadequacies of the adversarial process); see also Lissa Griffin & Ellen Yaroshefsky, Ministers of Justice and Mass Incarceration, 30 Geo. J. Legal Ethics 301 (2017) (arguing for changes within the prosecutorial system to better balance prosecutors’ roles as ministers of justice and advocates in light of prosecutors’ contribution to mass incarceration); K. Babe Howell, Prosecutorial Discretion and the Duty to Seek Justice in an Overburdened Criminal Justice System, 27 Geo. J. Legal Ethics 285 (2014). To some extent, the law establishes the outer limits of this obligation by, among other things, restricting how prosecutors acquire evidence, prescribing the minimum amount of information they must disclose to the defense, and precluding certain jury arguments.47See Fish, supra note 9, at 275–78 (describing ABA Model Rule 3.8 and state model rule equivalents that impose requirements on prosecutors). See generally Green & Zacharias, supra note 14 (detailing the various ethical rules implemented on the state and federal level, along with other prosecutorial accountability mechanisms). But within the confines of the law, prosecutors have vast discretion. Some guidance may be implicit in the law.48See Bellin, supra note 9, at 49–51. Professional tradition or consensus and office policies may also work to fill in the gaps. But prosecutors interpret this guidance differently and take vastly divergent approaches to discretionary decision making. Identifying the prosecutor as a public fiduciary suggests that the diversity of approaches may not be a problem as long as prosecutors follow a set process for deliberately and consciously discerning and pursuing the interests of the beneficiary.

This raises the question of who or what is the prosecutor’s beneficiary? Fiduciary theorists have sought to define the beneficiary of judges,49E.g., Leib et al., supra note 2 (applying fiduciary principles to the judiciary to analyze the role of judges in democracy). the legislature,50E.g., D. Theodore Rave, Politicians as Fiduciaries, 126 Harv. L. Rev. 671, 676 (2013) (arguing that courts should apply a fiduciary duty of loyalty to political representatives to hold them accountable for political gerrymandering). and administrative agencies.51Evan J. Criddle, Fiduciary Administration: Rethinking Popular Representation in Agency Rulemaking, 88 Tex. L. Rev.441, 466–68 (2010). We know who the prosecutor’s client is—the public entity named in the caption of the criminal indictment (the United States, the State, the Commonwealth, the People of the State, etc.). Is the beneficiary the entire public in the abstract sense or some segment of the public? If so, is there a tension between the prosecutor’s fiduciary duties as a lawyer and the prosecutor’s fiduciary duties as a public official? Does the prosecutor owe particularly strong obligations to some subgroup of the public, like the victim, or even the accused? While the prosecutor’s beneficiary is the public, the public’s objective in the criminal context is to render or achieve justice; hence, the prosecutor’s duty as a fiduciary is to pursue the public’s abstract interest in justice, which entails a constellation of interests and values.

Prosecutors and other public officials balance competing concerns of a complex group, but they also balance the interests of the public with the needs of the State and the proper functioning of the criminal justice system. As Evan Criddle and Evan Fox-Decent have argued, both public fiduciaries and private fiduciaries are required to engage in a careful balancing of competing interests. Tensions arise both among a group of beneficiaries and between the beneficiary and the state.52Evan J. Criddle & Evan Fox-Decent, Guardians of Legal Order: The Dual Commissions of Public Fiduciaries, inFiduciary Government, supra note 22, at 67. A prosecutor, like other fiduciaries, could be seen as owing first order obligations to the public and a separate second order fiduciary obligation to the courts and criminal justice system. Alternatively, the prosecutor’s obligation to the public, courts, and the criminal justice system could all derive from the fiduciary obligation to the public. In either iteration, fiduciary obligation shifts from loyalty to a duty of fairness and reasonableness in this context. The prosecutor must be fair in assessing the interest of the beneficiary and in balancing competing priorities and values. Values embraced by some members of the public might overlap with the interests of the State in a properly functioning criminal justice system, while others may not. Prosecutors’ duties to fairly consider the interests of the public as a whole may involve at least offering reasons for prioritizing some criminal justice ends over others. Again, prosecutorial independence from powerful majorities is crucial to preserve this discretionary balancing.

Drawing on fiduciary theory, this Article argues that the prosecutor’s obligation is to pursue the public’s abstract and evolving interest in justice. Other public officials may have a more defined mandate. Administrative agencies, for instance, usually have a particular mission. For example, the Federal Election Commission is tasked with enforcing and administering the federal election laws.53Mission and History, Fed. Election Comm’n, https://www.fec.gov/about/ mission-and-history [https://perma.cc/EQ67-JVT3]. The prosecutor’s mandate, in contrast, is vague and subject to multiple conflicting interpretations. Even so, as fiduciaries, prosecutors are required to continually take part in an ongoing process of exposition. This exposition ought to take place according to the traditions and policies of the prosecutor’s office. Like the limits of permissible judicial reasoning, these practices will limit discretion and confine the process in a way that ought to reassure us that we are not simply subject to any one prosecutor’s idiosyncratic view of justice. At any given moment, there will be competing understandings, but if prosecutors engage in a deliberate and transparent process of seeking to define justice in a consistent and rational way within the context of the traditions, policies, and practices of the office, that abstract ideal will gain meaning, and hopefully some consensus, through practice over time.

One difference between prosecutors’ objective and the abstract objectives pursued by most other fiduciaries is that justice is exceedingly vague. In a donative trust or a corporation, the trustee’s mission may be a group goal, like the goal of profit maximization in the corporate context, but it is defined. Without such specifically designed purpose, one may wonder whether prosecutors are doing anything other than indulging their own personal views and priorities. In the context of prosecution, what gives justice meaning beyond the personal view of the prosecutor is developed traditions and practices of prosecutors’ offices. Monitoring and accountability, therefore, must focus on the effectiveness of these norms and practices.

If we define prosecutors’ objectives in this way, what then are the duties of care and loyalty in the context of a criminal prosecution? In the private fiduciary context, where there is a discernable beneficiary or group of beneficiaries, loyalty can denote obedience to the beneficiary, or it can mean pursuing the best interest of the beneficiary even if the beneficiary prefers a different course,54Miller & Gold, supra note 2, at 558–59. but, at the very least, it means avoiding opportunism and eschewing the interests of third parties. Unlike an ordinary fiduciary for a private party, but like other public officials, the prosecutor does not exercise discretion for the benefit of a person or even a group of people who can give guidance or direction.

If we conceive of the prosecutor’s objective as the abstract public interest in justice, it becomes clear that prosecutors ought not operate with conflicts of interest that threaten to warp disinterested decision making on behalf of the public. We have argued previously that the absence of such conflicts is the key to proper prosecutorial conduct.55See generally Bruce A. Green & Rebecca Roiphe, Rethinking Prosecutors’ Conflicts of Interest, 58 B.C. L. Rev. 463 (2017). But even if they avoid conflicts of interest, how do prosecutors give content to these duties when the law is not explicit and decisions are entrusted to their discretion? Do the duties of care and loyalty add anything to the duties identified with prosecutors’ role as ministers of justice, and, if so, are these duties in tension?

This concept of the fiduciary obligation of prosecutors helps to clarify the complex relationship between independence and accountability. As Paul Miller and Andrew Gold explain, a public official is not policed by the beneficiary (the public), but, if at all, by individuals and institutions who, as co-fiduciaries, are assigned a monitoring role.56Miller & Gold, supra note 2, at 555. In fact, there is reason to maintain distance from direct popular control because prosecutors ought not define justice by the interest of the collectivity or any subset of the collectivity at any given time. Instead, they ought to build on past understandings. In this way, the abstract purpose, like the common law, is evolving and changing with time.57Id. at 526. Some have criticized the idea that judges are fiduciaries for abstract interests like the law. See Leib et al., supranote 2, at 702–03 & n.16. But this seems more apt when it comes to prosecutors. The public interest in justice has to be defined over time and with reference to prosecutors’ work, including the errors and misconduct that has been exposed.

For prosecutors, however, these cofiduciary monitors are few and weak. The gap in monitoring is necessary because prosecutors must be afforded independence from both political influence and popular control to pursue even-handed justice. But given this absence of external control, prosecutors must be held accountable in a different way. Rather than imagining public officials entering into a kind of contract with citizens, fiduciary theory helps highlight the need for disinterested decision making by public officials and the attendant need to avoid conflicts of interest that might compromise that neutral approach.58Miller & Gold, supra note 2, at 557–58 & n.132. It further pushes us to recognize the need for mechanisms of accountability that acknowledge the complexity of a fiduciary relationship when the beneficiary is an abstraction (the public), not a defined individual or group of individuals, and the beneficiary’s principal objective is an idea (justice) that is a distillation of interests shared by the public over time. The best form of accountability in this context may be institutional in the following sense: prosecutors’ offices must maintain required, transparent, and deliberate processes and procedures for decision making. Prosecutors need not be transparent about individual deliberations. But they must be transparent about prosecutors’ compliance with these procedures and how the various mechanisms within their offices work to align prosecutors’ deliberations with the abstract public interest in justice.59These mechanisms should work to ensure that the goal of justice and the definition of that goal persists over time despite the change in the makeup of the citizenry. How we understand justice may evolve, but it may not be radically displaced because one set of citizens so chooses. Id. at 525.

The virtually unbridled power in prosecutors’ offices is troubling. The internal structure of these offices ought to be altered to improve monitoring without compromising independence too seriously. That said, imperfect monitoring is an inevitable condition of public fiduciaries.60Leib et al., supra note 2, at 708. The character of the fiduciary, the integrity of the public official, is also critical to a well-functioning relationship.61Id. at 712. Therefore, reforms should focus not only on transparency and processes, but also on improving culture and education to ensure that those who wield this sort of power use it prudently.

II.   Prosecutors’ Fiduciary Duties

Prosecutors’ fiduciary role is complex. As a public official, the prosecutor, on behalf of the public, has authority to identify general policy objectives in criminal justice and to decide how to pursue these objectives in any individual case, including when, as is typically true, these objectives are at cross-purposes.62Green & Roiphe, supra note 55, at 471. As a lawyer in the adversary process, a prosecutor serves a more conventional fiduciary role. Within the bounds of judicial procedure and other law, advocates ordinarily strive to accomplish the objectives identified by their clients. In doing so, advocates exercise some discretion to decide how best to accomplish the client’s objectives while acting within the law. But even as an advocate, the prosecutor’s fiduciary obligations are complex because the objectives to be pursued on behalf of the public—the varying objectives that together comprise “justice”—are often themselves amorphous, and because, while the prosecutor is advocating for the public’s objectives in a criminal case, the prosecutor as a public official is trying to ascertain those objectives, which may shift or evolve.

In both roles, prosecutors owe the public fiduciary duties of care and loyalty—the two core duties that fiduciaries owe their principals. In general, these concepts have been underemphasized and underdeveloped in the literature on prosecutors’ role and regulation. The literature addresses both prosecutors’ adversarial role, including prosecutors’ legal obligations to the defendant and to the court and prosecutors’ exercise of discretion on behalf of the public writ large.63See, e.g., Fish, supra note 9, at 244–48 (describing prosecutors’ position in the criminal justice system both as adversarial and quasi-judicial); Sklansky, supra note 3 (arguing that prosecutors are “mediating figures” who must balance between “law and discretion”). But scholars rarely identify care and loyalty as touchstones for the exercise of prosecutorial discretion. Intuitively, prosecutors’ duty of care or loyalty to the public may seem insignificant since prosecutors have no identifiable client complaining of being disserved or betrayed. Nonetheless, the fiduciary duties of care and loyalty help explain how prosecutors ought to exercise discretion, particularly with regard to the crucial questions of whether to bring charges and which charges to pursue. Emphasizing the prosecutor’s fiduciary role may not provide concrete answers in individual cases, but, as this Part shows, it has implications for how prosecutors exercise discretion from both normative and procedural perspectives. The complexities of prosecutors’ fiduciary role add to the importance of developing mechanisms of accountability.

A.   Prosecutors’ Duty of Care and Competence

Prosecutors have a duty of care both as public officials and as advocates. As public officials defining the objectives of an investigation or prosecution, prosecutors have broad discretion like that of other executive branch officials in higher office, but they must exercise that discretion in light of the public interest.64Green & Roiphe, supra note 55, at 470. Prosecutors must also exercise care as advocates—for example, in selecting investigative techniques, in preparing for trial, in selecting legal theories and making legal arguments, in negotiating pleas, in complying with discovery obligations and other legal obligations, and so on—in light of the public objectives they have identified.65Bruce A. Green, Urban Policing and Public Policy—The Prosecutor’s Role, 51 Ga. L. Rev. 1179 , 1189–90 (2017). One might expect that, at least in the advocacy role, prosecutors would be subject to the same accountability mechanisms as lawyers for private clients, who may be disciplined or civilly liable when their professional work is so substandard that it violates the duty of care.66Susan R. Martyn, Lawyer Competence and Lawyer Discipline: Beyond the Bar?, 69 Geo. L.J. 705, 708–11 (1981). But even here, conventional accountability mechanisms are likely less effective. With an eye toward preserving prosecutorial independence, both the law and legal institutions (i.e., disciplinary authorities and courts) largely insulate prosecutors from external monitoring.67Bruce A. Green, Prosecutors and Professional Regulation, 25 Geo. J. Legal Ethics 873, 874–75 (2012).

For prosecutors, one potential monitor is the lawyer disciplinary agency of the jurisdiction where the prosecutor is licensed. But discipline has historically been ineffective in enforcing prosecutors’ duty of care.68See generally Fred C. Zacharias & Bruce A. Green, The Duty to Avoid Wrongful Convictions: A Thought Experiment in the Regulation of Prosecutors, 89 B.U. L. Rev. 1, 16 (2009) (“In theory, prosecutors are subject to the . . . obligation to ‘provide competent representation to a client’” under state disciplinary rules based on Model Rule 1.1, but “[a]s a practical matter, disciplinary regulators have not implemented rules like Model Rule 1.1 against prosecutors. Prosecutorial neglect has been regulated almost exclusively through internal administrative sanctions or informally by courts.”). All lawyers are subject to a disciplinary duty of competence, which is enforced selectively within the lawyer disciplinary process, typically in response to complaints by disgruntled clients who can show extreme or systematic neglect.69See generally Martyn, supra note 66, at 712. Prosecutors, however, have no clients who can lodge complaints with the disciplinary authorities, and complaints by criminal defendants and their lawyers asserting that prosecutors were careless may not be taken seriously. Disciplinary authorities will take judges’ complaints more seriously. In other words, discipline of prosecutors for incompetence—which, in practice, is exceedingly rare—will be limited to situations where courts are aggrieved by prosecutors’ carelessness.70For a rare example, see Livingston v. Va. State Bar, 744 S.E.2d 220, 224–25 (Va. 2013) (upholding discipline of prosecutor for negligently filing a succession of defective indictments).

Nor does civil litigation provide a meaningful oversight role for prosecutors. Unlike lawyers representing private clients, prosecutors have no aggrieved clients who might bring a malpractice or breach of fiduciary duty claim when a prosecutor performs carelessly. Criminal defendants who are injured by prosecutors’ legal violations have a very limited right to bring civil claims,71Prosecutors have absolute immunity from civil liability for conduct in their advocacy role, and they have qualified immunity for investigative or administrative work. See Imbler v. Pachtman, 424 U.S. 409, 427–31 (1976). For critiques of prosecutors’ absolute immunity, see Randall Grometstein & Jennifer M. Balboni, Backing Out of a Constitutional Ditch: Constitutional Remedies for Gross Prosecutorial Misconduct Post Thompson, 75 Alb. L. Rev. 1243, 1268–69 (2012); Margaret Z. Johns, Reconsidering Absolute Prosecutorial Immunity, 2005 BYU L. Rev. 53, 106–07. but not claims predicated on mere negligence.72See, e.g., McGhee v. Pottawattamee Cty., 475 F. Supp. 2d 862, 909 (S.D. Iowa 2007) (stating that “mere negligence” is never sufficient to overcome qualified immunity).

Judges generally have authority to remedy and sanction lawyers’ wrongdoing in the cases over which the judges preside,73See United States v. Modica, 663 F.2d 1173, 1182–86 (2d Cir. 1981) (reviewing remedies and sanctions available to trial and appellate courts for prosecutors’ improper closing statements). but when prosecutors violate the law through carelessness, courts are limited in their ability to hold prosecutors accountable. In some cases, they can provide juridical redress, which may be accompanied by public criticism of the prosecutor.74See, e.g., People v. Velasco-Palacios, 235 Cal. App. 4th 439, 447 (Ct. App. 2015) (upholding dismissal of indictment as sanction for prosecutor’s fabrication of evidence during plea negotiations). This sort of shaming may serve as an accountability mechanism.75See generally Lara Bazelon, For Shame: The Public Humiliation of Prosecutors by Judges to Correct Wrongful Convictions, 29 Geo. J. Legal Ethics 305, 313 (2016). But courts’ remedial role is an indirect and weak means of accountability and one that many consider inadequate.76See, e.g., H. Mitchell Caldwell, The Prosecutor Prince: Misconduct, Accountability, and a Modest Proposal, 63 Cath. U. L. Rev. 51, 83–84 (2013) (maintaining that there is “little incentive for offending prosecutors to refrain from future misconduct” when a judicial remedy is afforded but no personal sanction is imposed for prosecutorial misconduct beyond perhaps “a verbal reprimand”). Courts also have authority to sanction individual prosecutors who violate the law, but they are unlikely to do so where the violation is unintentional.77See, e.g., United States v. Jones, 686 F. Supp. 2d 147, 148, 152–53 (D. Mass. 2010) (declining to sanction prosecutor for an “egregious error” in failing to disclose “plainly important exculpatory information” because she was contrite and subsequently educated herself regarding her disclosure obligations). Ultimately, courts play a limited role in elaborating and enforcing a standard of prosecutorial care, and virtually no role where the lack of care relates to prosecutors’ exercise of discretion rather than to their compliance with the law.78See Bruce A. Green & Samuel J. Levine, Disciplinary Regulation of Prosecutors as a Remedy for Abuses of Prosecutorial Discretion: A Descriptive and Normative Analysis, 14 Ohio St. J. Crim. L. 143, 166–77 (2016) (reviewing small number of cases in which prosecutors have been disciplined for misconduct relating to their charging and plea bargaining decisions).

The lack of meaningful legal accountability inhibits the development of understandings about what it means for prosecutors to perform their work with care. Courts do not help define the objectives that prosecutors are supposed to serve, whether in general or in any given case; nor do they determine how prosecutors are supposed to achieve those objectives other than to hold prosecutors accountable when they break the law. Other lawyers look to their peers to determine the standard of care because they are potentially subject to civil liability for negligence when their work falls below expectations of other lawyers in the legal community performing similar work.79See, e.g., Gibbons v. Price, 514 N.E.2d 127, 136 (Ohio Ct. App. 1986) (“[A] claim of legal malpractice [is] based on an alleged failure to exercise the knowledge, skill and ability ordinarily possessed and exercised by members of the legal profession similarly situated . . . .”). But no analogous legal mechanism encourages prosecutors to compare their work with that of other lawyers.

This may not seem to be a serious problem when it comes to prosecutors’ work as advocates. In a criminal case, one might think that the prosecution’s objectives are obvious—for example, to secure a guilty verdict at trial or to secure a guilty plea—and one might measure the quality of prosecutors’ work by its likelihood of achieving these objectives. Further, prosecutors’ advocacy appears relatively similar to that of other lawyers’ advocacy, and therefore, the analogous work of other advocates may establish a relevant standard. But prosecutors’ advocacy role is in fact distinctive, and it poses challenges because, at the same time that prosecutors may be striving to secure a conviction, they may have other, potentially countervailing, objectives, such as to avoid a wrongful conviction and to ensure the defense a fair (not merely lawful) process. Given these further objectives, some argue that prosecutors fail to exercise adequate care when, although acting lawfully, they use unreliable evidence, ignore or exploit defense lawyers’ substandard work, or withhold information that would be useful to the defense.80See, e.g., Bruce A. Green, Access to Criminal Justice: Where Are the Prosecutors?, 3 Tex. A&M L. Rev. 515, 526–27 (2016) (discussing a prosecutor who, in exploiting unreliable evidence, failed to “take reasonable steps to avoid convicting [an] innocent [person]”).

The challenge of giving meaning to prosecutors’ duty of care is even more complicated because prosecutors serve as public officials, making discretionary judgments such as whether to investigate, whether to bring or drop charges, or whether to plea bargain. One cannot measure the competence of these decisions by assessing how well they achieve the prosecution’s objectives because this is the process by which prosecutors define their objectives. Discretionary decisions involve multiple and complex considerations. Further, these are the kinds of decisions that clients ordinarily make in a lawyer-client relationship. Therefore, one cannot look to the ordinary work of trial advocates representing private clients to measure whether prosecutors’ discretionary judgments are or are not competent.

Nor, for the most part, can one look to judicial decisions and pronouncements for a standard governing the competence of prosecutors’ discretionary decision making. The very premise of prosecutorial discretion is that, when exercised within the law, it is not susceptible to judicial review despite the “potential for both individual and institutional abuse.”81Bordenkircher v. Hayes, 434 U.S. 357, 365 (1978). Indeed, even when exercised in an unconstitutionally arbitrary fashion, prosecutors’ discretionary decisions may escape judicial review. United States v. Redondo-Lemos, 955 F.2d 1296, 1300 (9th Cir. 1992) (“Our only available course is to deny the defendant a judicial remedy for what may be a violation of a constitutional right—not to have charging or plea bargaining decisions made in an arbitrary or capricious manner.”). Even if ninety-nine out of 100 prosecutors would decline to bring charges in a case because the defendant’s guilt is too doubtful and the likelihood of a conviction is too low, as long as the prosecutor meets the minimum legal standard of “probable cause,” a court will rarely interfere with the prosecutor’s judgment either through the adjudicative process or through the disciplinary process.82See Green & Levine, supra note 78, at 164–65 (discussing the rarity of judicial review of prosecutorial charging discretion). Recently, courts have been scrutinizing decisions by progressive prosecutors not to charge in entire categories of cases. See e.g., Justin Jouvenal & Rachel Weiner, Prosecutors Won’t Pursue Marijuana Possession Charges in Two Northern Virginia Counties, Wash. Post (Jan. 2, 2020). To say that the court cannot overturn the prosecutor’s decision is not to say that the prosecutor exercised prosecutorial authority competently. On the contrary, ninety-nine out of 100 prosecutors, and 100% of judges, may privately regard the decision as an egregious abuse of discretion. But the point is that there is no legal process for developing and accumulating legal decisions, as there is for private clients, to express the prevailing normative understanding.

Because the quality of prosecutors’ discretionary decision making is not subject to meaningful legal review, other mechanisms are needed to establish and enforce norms of prosecutorial competence. Prosecutors might look to their peers within their office, or to other prosecutors’ offices, to ascertain expectations. In general, the former is more likely than the latter.83See Ellen Yaroshefsky & Bruce A. Green, Prosecutors’ Ethics in Context: Influences on Prosecutorial Disclosure, in Lawyers in Practice: Ethical Decision Making in Context 269, 279 (Leslie C. Levin & Lynn Mather eds., 2012) (“Prosecutors’ offices are something of an exception [to offices’ tendency to converge on a single way of working], partly because of their highly localized character. Since they are not directly in competition for business, the homogenizing pressures are much weaker. They do not adopt similar policies [regarding disclosures to the defense].”). The organized bar has developed norms of prosecutorial conduct, including those governing discretionary decision making.84See, e.g., Criminal Justice Standards for the Prosecution Function Standard 3–4.4 (Am. Bar Ass’n, 4th ed. 2018) [hereinafter Standards for the Prosecution Function] (addressing “Discretion in Filing, Declining, Maintaining, and Dismissing Criminal Charges”). But, in part because of prosecutors’ influence, the bar’s writings provide limited guidance. And because prosecutors regard the organized bar as subject to capture by defense lawyers, they may ignore these writings.85See Green, supra note 67, at 898 (“Even if professional conduct rules might legitimately regulate them, prosecutors suggest, the existing process for developing professional conduct rules is illegitimate because of the influence of bar associations, which are subject to capture by criminal defense interests.”).

As for prosecutors’ accountability for upholding the duty of care, in the absence of legal enforcement mechanisms, a heavy burden is placed on prosecutors to regulate themselves, which is to say that we rely on their professional commitment to standards of care as they have come to understand them. Internal self-regulation by a prosecutor’s office may be the most meaningful mechanism to ensure the accountability of subordinate prosecutors.86See generally Norman Abrams, Internal Policy: Guiding the Exercise of Prosecutorial Discretion, 19 UCLA L. Rev. 1, 7–8 (1971); Ronald F. Wright, Sentencing Commissions as Provocateurs of Prosecutorial Self-Regulation, 105 Colum. L. Rev. 1010, 1028 (2005). The elected prosecutor, or chief appointed prosecutor, has a stake in the quality of the office’s work and can therefore be expected to police both advocacy and discretionary decision making by subordinate lawyers. But at the same time, a chief prosecutor who makes discretionary decisions badly cannot be expected to enforce a high standard of care but, on the contrary, is likely to influence subordinates to make their own decisions just as badly.

B.   Prosecutors’ Duty of Loyalty

Like other fiduciaries, prosecutors have a fiduciary duty of loyalty. As lawyers for the public, their loyalty duty may seem comparable to that of a lawyer for a private entity. Unlike many fiduciaries for private beneficiaries, such as lawyers for private entities who can take direction from duly authorized representatives of the entity-client, prosecutors are the public officials who make decisions for the public entity-client. In this role, too, prosecutors are fiduciaries.

In criminal cases, there may be several threats to disinterested prosecutorial decision making. One risk is that prosecutors will act in their own self-interest at the expense of the public interest—a variation of what Thomas Rave calls the principal-agent problem.87D. Theodore Rave, Two Problems of Fiduciary Governance, in Fiduciary Government, supra note 22, at 49, 49–50. Another is that prosecutors will serve the private interests or preferences of identifiable third parties.88Id. Although these concerns apply equally to prosecutors’ role as lawyers and as public officials, a third risk relates uniquely to prosecutors’ governance role. In the context of governance, loyalty generally means serving the broader purpose or goal of a community.89See Leib et al., supra note 2, at 731. The risk is that the prosecutor will be influenced by a subgroup’s well-articulated interest at the expense of the broader public interest.90Id. at 712. Rave calls this risk—that a powerful subgroup within the group that comprises the beneficiary can dominate at the expense of the others—the tyranny-of-the-majority problem.91Rave, supra note 87, at 49–66. The ordinary problem of subgroup dominance is compounded in the case of prosecutors once we conceive of the beneficiary’s objective as the public interest in justice, the meaning of which develops over time by accretion in common law fashion. Given that this objective is so vague and subject to interpretation, there is a heightened risk that a subgroup’s exercise of undue influence will be indiscernible.92

In general, it is hard to prove and defend against improper influence on prosecutorial decision making because the process is so opaque. For example, the Manhattan District Attorney recently disputed claims that criminal defense lawyers’ campaign contributions influenced the prosecutors’ decision not to pursue charges against their clients, leading to a prosecutors’ reexamination of campaign finance practices. See Elizabeth Holtzman & David Yassky, The Lessons of Cyrus Vance’s Campaign Contributions, N.Y. Times (Nov. 6, 2017), https://www.nytimes.com/2017/11/ 06/opinion/cyrus-vance-contributions-weinstein.html; see also Ctr. for Advancement of Pub. Integrity, Raising the Bar: Reducing Conflicts of Interest and Increasing Transparency in District Attorney Campaign Fundraising 18 (2018), https://www.law.columbia.edu/sites/default/files/microsites/public-integrity/raising_the_bar.pdf [https://perma.cc/UXG8-4FHE]. Questions might also be raised about whether, and when, subgroup influence may be appropriate. For example, prosecutors are obviously subject to the influence of the police or other investigative agents. See generally Jeffrey Bellin, The Power of Prosecutors, 94 N.Y.U. L. Rev. 171, 191–94 (2019) (describing police influence on criminal justice decisions); Daniel Richman, Prosecutors and Their Agents, Agents and Their Prosecutors, 103 Colum. L. Rev. 749, 771–72 (2003). Ordinarily, investigators’ influence over charging and plea bargaining decisions may seem unremarkable. But their influence may be problematic, for example, in cases where the propriety of police or investigative conduct is in issue.

Scholars and courts assume that prosecutors must serve the public and, at a minimum, avoid conflicts of interest.93See, e.g., In re Cole, 738 N.E.2d 1035, 1037 (Ind. 2000) (finding that the prosecutor “served a public trust to enforce the law” and “[t]he state is entitled to a prosecutor’s undivided loyalty”); see also Standards for the Prosecution Function Standards, supra note 84, at Standard 3–1.3 (“The prosecutor generally serves the public and not any particular government agency, law enforcement officer or unit, witness or victim.”); id. Standard 3–1.7 (addressing prosecutors’ conflicts of interest). To some extent, this expectation is codified in the law and enforced by legal institutions.94See generally Green & Roiphe, supra note 55, at 469, 471. Unlike lawyers for private clients, prosecutors are not subject to civil lawsuits for breach of loyalty. But courts require prosecutors to adhere to the conflict of interest rules that govern lawyers generally.95Id. at 484–88 & n.104 (citing examples). Constitutional decisions recognize prosecutors’ obligation to be disinterested, which is another way of enforcing a duty of loyalty.96Id. at 488–91. Judges exercise statutory or inherent authority to disqualify prosecutors, and prosecutors recuse themselves, in some situations where there is a serious risk that prosecutors will inappropriately serve private interests.97Id. at 491–99. Interestingly, the conflict of interest law seeks to preserve disinterestedness in prosecutors’ role as public officials exercising prosecutorial discretion no less than in their role as courtroom advocates.98See Young v. United States ex rel. Vuitton Et Fils S.A., 481 U.S. 787, 807 (1987) (“[T]he fact that the judge makes the initial decision that a contempt prosecution should proceed is not sufficient to quell concern that prosecution by an interested party may be influenced by improper motives. A prosecutor exercises considerable discretion in matters such as the determination of which persons should be targets of investigation, what methods of investigation should be used, what information will be sought as evidence, which persons should be charged with what offenses, which persons should be utilized as witnesses, whether to enter into plea bargains and the terms on which they will be established, and whether any individuals should be granted immunity. These decisions, critical to the conduct of a prosecution, are all made outside the supervision of the court.”).

While incomplete and imperfect, conflicts of interest standards help to ensure against the danger that prosecutors will seek to advance their own personal interests instead of that of the public. Conflict of interest rules are most likely to apply when a particular prosecutor is at risk of engaging in self-dealing because the particular prosecutor has a unique and tangible self-interest.99See Green & Roiphe, supra note 55, at 472 (discussing “personal-interest conflicts [that] relate to a particular prosecutor in an idiosyncratic way”). For example, a prosecutor who has a financial stake in a corporation would not be expected to make prosecutorial decisions regarding the corporation. And a prosecutor who is a victim of a crime would not be expected to prosecute the perpetrator.

Prosecutors’ obligation to refrain from acting self-interestedly poses some interesting challenges, however, when the prosecutor’s interests are intangible, particularly when it comes to the pervasive and unavoidable self-interest in one’s own reputation and career advancement.100See id. at 480–81 (“Even prosecutors who do not seek professional advancement are jealous of their professional reputation. This broad self-interest can come into play in every criminal case in ways that are inconsistent with the expectations of disinterested prosecution.”). Prosecutors almost always stand to benefit professionally if they conduct their work in a manner that seems adept or successful from the view of their supervisors or others. This could theoretically provide motivation to carry out the fiduciary duties of loyalty and care, but is often charged that prosecutors are simply looking to put notches on their belts, at the expense of the public interest in “doing justice.”101See Daniel S. Medwed, The Zeal Deal: Prosecutorial Resistance to Post-Conviction Claims of Innocence, 84 B.U. L. Rev. 125, 134–35 (2004) (“Prosecutors with the highest conviction rates (and, thus, reputations as the best performers) stand the greatest chance for advancement internally.”). Prosecutors cannot, however, recuse themselves from cases where their professional self-interests are implicated. Even if they could, it might be impossible, given how vague the pursuit of justice is, to tell when a prosecutor’s assessment of the purpose or goal of the public is tainted.

If self-interested prosecutors cannot simply be replaced by disinterested ones, then how can prosecutors avoid or at least minimize the effect of self-interest in decision making, and how can the public hold prosecutors accountable when they fail to do so? It may be hard for prosecutors themselves to tell whether they are acting in their own self-interest since they can rationalize self-interested behavior.102There is an abundant literature on how unconscious thought processes—e.g., cognitive biases—influence prosecutors’ decision making. See, e.g., Susan Bandes, Loyalty to One’s Convictions: The Prosecutor and Tunnel Vision, 49 How. L.J. 475, 491 (2006); Barkow, supra note 3, at 883 (“Prosecutors may feel the need to be able to point to a record of convictions and long sentences if they want to be promoted or to land high-powered jobs outside the government, and that will affect their assessment of a defendant’s case.”); Alafair S. Burke, Improving Prosecutorial Decision Making: Some Lessons of Cognitive Science, 47 Wm. & Mary L. Rev. 1587, 1603–04 (2006); Aviva Orenstein, Facing the Unfaceable: Dealing with Prosecutorial Denial in Postconviction Cases of Actual Innocence, 48 San Diego L. Rev. 401, 425–26 (2011). It may be harder still for the public to determine whether prosecutors are acting disloyally. Much of what prosecutors do is not publicly visible, and the public can at best infer the motivations behind prosecutors’ visible conduct.103See, e.g., Brandon L. Garrett, Globalized Corporate Prosecutions, 97 Va. L. Rev. 1775, 1784 (2011) (referring to “the opaque machinery of prosecutorial discretion”). Both from the inside and from the outside, it may be impossible to disentangle prosecutors’ professional self-interest from the public interest. For example, when prosecutors publicize their successes, their intent may be to serve the public interest by keeping the public informed and instilling public confidence in their work. But it is also possible that prosecutors are indifferent to the public interest and are motivated simply to promote their own career ambitions. Not only is it impossible to assess prosecutors’ motivation to hold them accountable for disloyalty, but the ability of prosecutors to justify their self-promotion as an act of public accountability suggests a tension between the duties of loyalty and accountability.

The problem of averting conflicts of interest is even more intractable with respect to what we have called institutional conflicts of interest—that is, situations where prosecutors have an incentive to serve the institutional interest of the prosecutor’s office at the expense of the broader public interest.104Green & Roiphe, supra note 55, at 477–79. Prosecutors’ offices have reputational interests that may compromise their lawyers’ disinterested judgment—for example, when the office has convicted an innocent person, its interest in avoiding embarrassment may lead it to defend the conviction, even though the public interest is in correcting wrongful convictions.105Medwed, supra note 101, at 136. Prosecutors’ offices may also have compromising financial interests—for example, an office’s interest in benefitting financially from a civil forfeiture may influence prosecutors to pursue a prosecution that might be undeserved.106Id. at 135. These institutional interests may not be wholly illegitimate, and in any event, prosecutors cannot recuse themselves to avoid their influence.

The law also addresses conflicts of interest arising out of a prosecutor’s personal relationship with a third person, such as a defendant or a victim.107Green & Roiphe, supra note 55, at 472–73 (“[A] prosecutor’s familial relationship to a defendant or victim may undermine the prosecutor’s disinterestedness, leading the prosecutor to be unusually lenient where the defendant is a relative and unusually harsh where the victim is one.”). Prosecutors are expected to avoid situations where they are strongly tempted to subordinate the public interest to the private interests of victims, suspects, defendants or others, although the law is not necessarily coextensive with one’s intuitions about prosecutors’ duty to eschew private interests. For example, prosecutors are expected to recuse themselves, or to be disqualified by a court, when they have a close familial or economic relationship with a crime victim, defendant, or other interested third party.108See, e.g., People v. Vasquez, 137 P.3d 199, 205–06 (Cal. 2006) (finding that court erroneously failed to disqualify prosecutor where member of prosecutor’s staff was defendant’s parent); State v. Mantooth, 788 S.E.2d 584, 587–88 (Ga. Ct. App. 2016) (upholding prosecutor’s recusal where member of prosecutor’s staff was related to the defendant).

But prosecutors’ responsibility to avoid serving the private interests of third parties is not so simple because fidelity to the public interest, to some extent, presupposes care for interested private parties. For example, it is assumed that prosecutors’ public obligation to pursue justice includes concern for victims’ interests.109See, e.g., People v. Superior Court, 182 P.3d 600, 611–13 (Cal. 2008) (finding that it was proper for the prosecutor to argue in favor of minor victim’s privacy interests, which were aligned with those of the State, and that doing so was not tantamount to representing the victim). Indeed, the ends of criminal justice (which prosecutors are charged with serving) often include restitution for the victim and retributivism, and procedural laws establish victims’ rights which prosecutors must respect and, to some extent, implement.110See, e.g., People v. Subramanyan, 201 Cal. Rptr. 3d 443, 447–48 (App. Dep’t Super. Ct. 2016) (holding that under Marsy’s Law, only the prosecutor, not the victim, has authority to seek restitution or appeal a restitution order). What does it mean to say that a prosecutor owes undivided loyalty to the public’s interest in justice if, at the same time, the prosecutor has an obligation (though not a fiduciary duty) of care to the victim? Even prosecutors may have difficulty resolving this riddle.111See, e.g., In re Flatt-Moore, 959 N.E.2d 241, 245–46 (Ind. 2012) (sanctioning prosecutor who, as a condition of a plea bargain, insisted that the defendant comply with the victim’s demand for an excessive amount in restitution). Similarly, and perhaps more controversially, prosecutors are supposed to have concern for defendants’ rights and, perhaps, defendants’ interests; it is sometimes said that, as public representatives, prosecutors speak for defendants among others.112People v. Dehle, 83 Cal. Rptr. 3d 461, 466 (Ct. App. 2008) (“The prosecutor speaks not solely for the victim, or the police, or those who support them, but for all the People. That body of ‘The People’ includes the defendant and his family and those who care about him.”). Intuitively, one can understand that prosecutors, in determining what is in the public’s best interest, should give some weight to both victims’ interests and defendants’ rights without owing loyalty to either. But, at the very least, this reflects a complication in prosecutors’ decision making that most lawyers for private clients do not encounter.

Finally, insofar as there is a risk that prosecutors will favor the interests of a powerful or vocal subgroup, the law is essentially silent. As public officials, prosecutors might be said to serve a purpose that transcends any person or group of people: they serve justice, an abstract principle distilled from the objectives of an abstract public over time, not the objectives of any particular individual or even any group of individuals at the moment.113Miller & Gold, supra note 2, at 571–72. The fiduciary obligation to pursue the public’s interest in justice offers some guidance for prosecutorial discretionary decision making. If prosecutors owed a general obligation to the public, there would be no clear restriction on how they should balance various public interests. Deterrence could be weighed equally with promoting good immigration policy. Because, as fiduciary theory helps clarify, the prosecutor’s duty is to guard justice, not the public’s interest in general, however, it follows that considerations intrinsic to the justness of a case ought to take precedence over any other public value. Thus, in making discretionary decisions, prosecutors must primarily consider concerns such as protecting the community, deterring future offenses, rehabilitating offenders who can be reformed, incapacitating dangerous offenders, and seeking retribution and restitution. Other interests extrinsic to the justness of the prosecution, such as foreign policy implications of a particular case or its intersection with immigration policy, are not central to a prosecutor’s job. Of course, prosecutors can give those weight in making discretionary decisions, but only if they can be advanced consistently with the central fiduciary mandate to ensure justice.114See Green, supra note 65, at 1204–05 (arguing that, as public officials, “[p]rosecutors can and should take account of a broad range of social policy considerations that bear on their work” and not just those implicating traditional criminal justice concerns such as proportionality).

This insight can help explain why it has been acceptable to engage in “spy swaps,” where the U.S. government exchanges a foreign individual arrested or charged with espionage for a U.S. citizen or valued non-citizen held abroad.115See Green & Roiphe, supra note 15, at 1834. The prosecutor who dismisses the charges against a foreign defendant has not violated the fiduciary duty to do justice because the deal, which furthers the extrinsic interest in foreign relations, does not result in an injustice. After all, many guilty individuals go free for a myriad of reasons in our system. On the other hand, if a prosecutor were to pursue an innocent foreign individual in order to gain a foreign policy advantage for the United States, that would be a breach of fiduciary duty because the prosecutor will have prioritized an extrinsic consideration that resulted in the prosecution of an innocent individual.116Id.

Of course, there are some considerations that might fall on the margin. For instance, are reducing mass incarceration and racial injustice in prosecution intrinsic or extrinsic concerns? It is possible that they are both. Overincarceration burdens taxpayers and potentially disrupts communities, which makes it seem an important extrinsic concern. But the length of incarceration also informs questions of deterrence and incapacitation, which are at the core of criminal justice. Similarly, racial injustice and bias may be questions that bear on communities as a whole and race relations, which sound like extrinsic concerns. But they may also bear on fairness and proportionality, which are central to pursuing justice in an individual case.

In sum, enforcing loyalty is difficult where the beneficiary to whom prosecutors owe loyalty is so amorphous; the objectives that prosecutors must pursue on their beneficiary’s behalf are so vague; and prosecutors have so much discretion to discern the relevant objectives, balance them, and decide how best to pursue them. Like other public officials, prosecutors typically make discretionary decisions by engaging in a complicated balancing of competing public interests. The interests of particular segments of the public may not reflect the best interest of the public in general. Because prosecutors’ decision-making process is not transparent, it may be impossible to discern whether prosecutors are privileging the interests of a vocal majority or even narrower parochial interests. And, particularly in the case of elected prosecutors, prosecutorial self-interest may dovetail with the interests of powerful political factions. The law neither prescribes prosecutors’ discretionary decision-making criteria in general nor excludes considering or even privileging particular factions’ interests and preferences; nor does the law prescribe a process for making discretionary decisions that reduces the risk that particular public interests will be overvalued. The limited ability of existing law and legal institutions to ensure prosecutors’ duty of loyalty, especially in the face of public pressure, underscores the importance of alternative means of prosecutorial accountability.

C.   Developing Affirmative Theories of Care and Loyalty

To a large extent, fiduciary theory is concerned with avoiding opportunism on the part of the fiduciary. It identifies abusive conduct—violations of the fiduciary duties of care and loyalty—and mechanisms to identify, redress, and deter such bad conduct.117Galoob & Leib, supra note 1, at 117–18. But does the theory offer anything positive—an affirmative vision of good conduct? We assume that the duties of care and loyalty entail more than avoiding negligence and betrayal.118See Leib et al., supra note 2, at 736 (“Although the duty [of care] seemingly requires little more than avoiding negligence, most concede that it entails affirmative obligations (unlike the mostly prohibitive duty of loyalty) . . . .” (footnote omitted)). Any illumination would be valuable because the law and legal processes are so deficient. Rules of professional conduct offer virtually no guidance to prosecutors about what it means to make good decisions.119On the limited reach of prosecutorial ethics rules, particularly with regard to prosecutors’ discretionary decisions, see Bruce A. Green, Prosecutorial Ethics as Usual, 2003 U. Ill. L. Rev. 1573, 1587–91 (explaining that the only exercise of discretion pretrial or during trial that the prosecutorial ethics rules addresses is whether to initiate a criminal charge, with respect to which the prosecutorial ethics rule requires only “probable cause”). Existing professional accountability mechanisms, such as professional discipline and judicial sanctions, are directed at punishing the very worst conduct, not defining good conduct, and, in any event, they have little to say about prosecutorial discretion.

In the case of private clients, lawyers largely accept their clients’ stated objectives, try to accomplish them, and defer to at least certain decisions regarding how to do so. But prosecutors do not take direction. No one would suggest that prosecutors should take a public referendum on whether particular individuals should be prosecuted or on the terms of a proposed plea bargain. Fiduciary theory of governance helps focus the question not on whether the prosecutor has been loyal to a particular set of people but rather to a public purpose, here, the duty to seek justice in criminal cases. But the problem remains how to give such a vague notion a concrete meaning, other than what any individual prosecutor believes is just.

In determining the meaning of the prosecutor’s mandate, should the public—the beneficiary of a prosecutor’s work as fiduciary—have any voice at all in prosecutors’ decisions beyond, in the case of elected prosecutors, deciding which prosecutor makes those decisions? Obviously, the public has no direct voice; the prosecutor acts as its decision-making surrogate. But in making decisions, must prosecutors look for a way to discern, and give weight to, popular demands?

Some have suggested that prosecutors should defer to, or at least take account of, public preferences regarding decisions in individual cases.120See, e.g., Russell M. Gold, Promoting Democracy in Prosecution, 86 Wash. L. Rev. 69, 71 (2011) (“Because prosecutors act on the public’s behalf, their decisions should reflect their constituents’ preferences.”). This, however, misconstrues prosecutors’ job. While prosecutors’ duty to “seek justice” is a vague concept, it decidedly does not mean carrying out the public’s will in each individual case. On the contrary, seeking justice presupposes that prosecutors stand as a bulwark against mob justice and avoid making decisions based on the public’s (or their own) passions and prejudices.121See Green & Zacharias, supra note 37, at 869–70 (“[W]hatever else prosecutors do, they should act nonpolitically. This encompasses both avoiding obligations to the political parties with which they are affiliated . . . and holding themselves above public outcry and frenzy about particular cases. The latter principle derives from society’s aversion to mob justice.” (footnotes omitted)); David Alan Sklansky, The Changing Political Landscape for Elected Prosecutors, 14 Ohio St. J. Crim. L. 647, 673 (2017) (“The danger of politicizing the handling of particular cases is, in fact, a worrisome aspect of the growing attention voters seem to be paying to prosecutorial elections.”). Prosecutors are supposed to make discretionary decisions in individual cases in accordance with the law and with norms and traditions that are relatively constant over time and that reflect generally applicable law enforcement considerations and principles (such as proportionality and equal treatment).122See Green & Zacharias, supra note 37, at 869–70 (“A nonpolitical prosecutor arguably can ignore the public’s desires concerning a specific case at a heated moment of time while remaining true to the public will in a more general sense. In other words, the nonpolitical prosecutor will ignore a momentary hue and cry but continue to heed public expectations as they are expressed over time in the law and popular culture.” (footnote omitted)). For example, prosecutors have a duty to avoid convicting innocent people, which may require declining to bring charges in light of their own professional doubts about an individual’s guilt, even if the public is clamoring for a prosecution.123See generally Alafair S. Burke, Prosecutorial Agnosticism, 8 Ohio St. J. Crim. L. 79, 79 (2010) (the prevailing view is “that an ethical prosecutor should pursue criminal charges against a defendant only if the prosecutor personally believes that the defendant is guilty”); Bennett L. Gershman, A Moral Standard for the Prosecutor’s Exercise of the Charging Discretion, 20 Fordham Urb. L.J. 513, 522 (1993) (arguing that prosecutors must be “morally certain” of the defendant’s guilt before pursuing charges). This is not to say that prosecutors may disregard public preferences; it is simply to say that the public preferences that prosecutors implement are principally those discernable in the Constitution, other laws, and legal traditions, including the norms and traditions governing criminal prosecution.124Here, the beneficiary’s objective is a public purpose, but one that is not expressed in a particular law that a public official is entrusted with implementing, but one that acquires meaning over time. It is embodied in criminal law and procedure, combined with decades of spoken, written, and accepted norms of practice that help define the beneficial purpose. See Miller & Gold, supra note 2, at 570 (explaining that fiduciaries can serve an abstract public purpose).

Particularly hard questions of prosecutorial discretion are presented when the cross-cutting principles governing criminal prosecution intersect with questions of social policy. The public may have a reasonable claim that, at least as to social policy, its preferences should carry the day. Take, for example, the question of whether an urban prosecutor should prosecute individuals for quality-of-life offenses, such as defacing buildings with graffiti, smoking marijuana in public, or fare-beating.125This is a question on which New York City prosecutors have disagreed and taken different approaches over time. See Shawn Cohen et al., Manhattan DA Won’t Prosecute Quality-of-Life Offenses Anymore, N.Y. Post (Mar. 1, 2016), https://nypost.com/2016/03/01/manhattan-da-wont-prosecute-quality-of-life-offenses-anymore [https://perma.cc/8XGW-5MZ7]. A prosecutor has a range of options. At one extreme, the prosecutor might almost entirely ignore certain offenses, making a judgment, for example, never to prosecute fare-beating cases, as Manhattan’s District Attorney did recently.126Brendan Cheney, Manhattan DA Will No Longer Prosecute Turnstile Jumping, Politico (Feb. 1, 2018, 1:55 PM), https://www.politico.com/states/new-york/city-hall/story/2018/02/01/manhattan-da-will-no-longer-prosecute-turnstile-jumping-229568 [https://perma.cc/PG7P-NXKH]. For a discussion of the separation of powers issues involved in this sort of decision, see Jeffrey Bellin, Defending Progressive Prosecution, at 9 (2019), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3479165 (arguing that declining to prosecute entire classes of cases is a proper check on legislatures). At the other extreme, the prosecutor might decide to prosecute these cases aggressively based, for example, on the empirically questionable “broken windows” theory that ignoring quality-of-life offenses inevitably leads to more serious offenses.127See Michelle Chen, Want to See How Biased Broken Windows Policing Is? Spend a Day in Court, The Nation (May 17, 2018), https://www.thenation.com/article/want-to-see-how-biased-broken-windows-policing-is-spend-a-day-in-court [https://perma.cc /M5HP-SAUS]; see also William J. Bratton, N.Y. Police Dep’t, Broken Windows and Quality-of-Life Policing in New York City 3, 5 (2015), http://www.nyc.gov/html/nypd/downloads/pdf/analysis_and_planning/qol.pdf [https://perma.cc/SZB6-DJNP]. Or, taking an intermediate approach, a prosecutor might bring quality-of-life offenders to a community court or other problem-solving court in which some alternative to incarceration is available.128See Andrew Denney, After 25 Years, NYC’s Midtown Community Court Still Takes ‘Problem-Solving’ Approach to Low-Level Crime, N.Y. L.J. (Dec. 7, 2018), https://www.law.com/newyorklawjournal/2018/12/07/after-25-years-nycs-midtown-community-court-still-takes-problem-solving-approach-to-low-level-crime [https://perma.cc/P5FL-MGGP]. To the extent that the choice turns on social policy judgments, do public preferences carry weight?

No one would expect the public to have the last word on whether any particular arrested individual should be prosecuted for a quality-of-life offense (or any other offense) because that determination rests in part on matters of evidentiary fact and criminal justice principles that are uniquely within the prosecutor’s expertise, such as whether there is sufficient evidence of guilt and whether the particular defendant is so culpable or dangerous relative to others who commit the offense as to deserve or necessitate punishment.129See Green, supra note 65, at 1196–97. These are criminal justice questions that prosecutors conventionally resolve based on the relevant evidence and their understanding of the principles governing prosecutors’ work. Likewise, the general question of how to allocate resources as between, say, graffiti cases and arson cases, is one that prosecutors would conventionally resolve based on considerations intrinsic to the law.130See Bratton, supra note 127, at 6–7. The conventional understanding that arson is the more serious wrong, as reflected in the legislative sentencing scheme, would lead a prosecutor to prioritize arson cases. And questions of how to allocate limited prosecutorial resources also reflect an administrative judgment that is ordinarily entrusted to prosecutors.

But, at the core, there is also a criminal justice policy question on which prosecutors might disagree and as to which the public might have a view. The criminalization of quality-of-life offenses reflects a legislative judgment that the public interest is at least sometimes served by prosecuting low-level offenders.131Id. at 2. Members of the community may take the view that, to promote public safety, these offenders should be prosecuted as a matter of course. Conversely, the public may take the view that prosecuting these offenders undermines relationships between the law enforcement officials and communities and is excessively disruptive.132Id. at 6. Underlying the policy question may be empirical questions, such as whether the prosecution of quality-of-life offenses leads to the discovery of, or deters, more serious crimes.

Does prosecutors’ overarching duty to pursue justice in the abstract mean that prosecutors should resolve public policy questions themselves without regard to public preferences? If not, to what extent should prosecutors defer to the public’s preferences regarding policy questions such as this? Suppose, for example, that there is a clear public demand to prosecute quality-of-life offenses aggressively in order to deter more serious crimes. Does loyalty to the public presuppose that the prosecutor defer to that demand, even if the prosecutor’s own judgment is that the public’s preference is founded on an empirical misunderstanding or that the public interest is better served by a less aggressive prosecutorial approach?

An elected prosecutor may run on a criminal justice policy platform and fairly claim that, once elected, it is within the prosecutor’s authority to determine and implement criminal justice policy within the area of discretion afforded by the legislature. In other words, the point of the election is to decide who has the better judgment regarding policy and to elect the lawyer who can be trusted with responsibility for best resolving questions of policy. But that would mean that, once elected, the prosecutor can essentially ignore public preferences (except to the extent useful to win reelection). It is unclear why an elected prosecutor would have a stronger claim than other elected officials to set policy without regard to the public will.

The problem is not entirely avoided by appointed prosecutors. In federal cases, it may be assumed that United States Attorneys defer to the policy preferences of the appointed Attorney General and, indirectly, those of the President.133See generally Rachel E. Barkow & Mark Osler, Designed to Fail: The President’s Deference to the Department of Justice in Advancing Criminal Justice Reform, 59 Wm. & Mary L. Rev. 387, 467 (2017); Barack Obama, The President’s Role in Advancing Criminal Justice Reform, 130 Harv. L. Rev. 811, 823 (2017). But many policy questions may be unanswered or unresolved at higher levels, leaving the question whether United States Attorneys may interpose their own policy preferences or must discern and implement those of the public.134See generally Leslie B. Arffa, Note, Separation of Prosecutors, 128 Yale L.J. 1078, 1103 (2019) (describing and justifying the decentralization of federal criminal prosecution).

If elected prosecutors may not decide policy questions entirely on the basis of their own best professional judgment, but must defer, or give weight, to public preferences regarding broad questions of criminal justice policy, how is the relevant policy to be discerned? The public cannot be polled. Public preferences cannot necessarily be inferred from existing legislation or even vocal social movements. Arguably, some other elected representative—a governor or mayor—may speak for the public. But it is hard to see why another elected official would be better able than the elected prosecutor to discern the public will regarding criminal justice policy questions.

If the prosecutor must discern the public will, another question remains, namely, which public? If, as we assume, the prosecutor’s client is the public, what do we mean by that? Does a United States Attorney for the Southern District of New York have a special obligation to the population of New York or owe obligations to citizens of the United States?135For example, in deciding whether to prosecute an obscenity charge, should a federal prosecutor be concerned about the social norms of the federal district (which would be the relevant ones for determining whether material is obscene under the federal criminal law) or the social norms of the broader national population? Cf. Freedberg v. U.S. Dep’t of Justice, 703 F. Supp. 107, 111 (D.D.C. 1988) (enjoining federal obscenity prosecutions in multiple federal districts, notwithstanding that states may apply different obscenity standards). In general, the popular norms that restrain and influence prosecutors are likely those of the population served, from which jurors are chosen, if only for the instrumental reason that those are the norms on which a jury is likely to draw. See generally Anna Offit, Prosecuting in the Shadow of the Jury, 113 Nw. U. L. Rev. 1071, 1105–06 (2019) (describing empirical findings that federal prosecutors make discretionary decisions with an eye toward how they believe a jury would perceive the case). Would a local prosecutor owe obligations to all citizens in the State? Or only to those citizens in the local prosecutor’s district? Assuming the answer is the former, should the prosecutor assess the interests of the State with a special eye toward the needs of the local community?

Further, efforts to discern the will of the public may be in tension with the previously discussed principle that prosecutors may not serve the interests of private subgroups. If the prosecutor implements the policy judgments of the particular political party of which the prosecutor is a member, one might consider that the prosecutor is acting in an impermissibly partisan fashion, promoting the interests of a political party—a private group—at the expense of the general public. Or one might say that the prosecutor is giving impermissible priority to the preferences of a subgroup, not necessarily carrying out the will of the public at large. Of course, no resolution of a policy question will ever reflect the preference of every member of the public. But arguably, the general preference to which a prosecutor must defer is not that of the prosecutor’s party.136See Green & Zacharias, supra note 37, at 869–70 (discussing ideal of prosecutorial nonpartisanship). Needless to say, prosecutors have not always comported with this understanding. Both elected and appointed prosecutors may be heavily engaged in partisan politics and beholden to political parties for their positions. See Michael J. Ellis, Note, The Origins of the Elected Prosecutor, 121 Yale L.J. 1528, 1565–67 (2012); cf. Scott Ingram, George Washington’s Attorneys: The Political Selection of United States Attorneys at the Founding, 39 Pace L. Rev. 163, 164–65 (2018). At least before the prevalence of contemporary understandings regarding prosecutors’ obligation to be impartial and not politically partisan, prosecutors who were beholden to a political party were more likely to accede to its influence when making discretionary decisions. See Carolyn B. Ramsey, The Discretionary Power of “Public” Prosecutors in Historical Perspective, 39 Am. Crim. L. Rev. 1309, 1338–39 (2002) (describing influence of the Tammany Hall political machine on late nineteenth century prosecutors in New York City).

Finally, assuming that loyalty to the public requires deference to the public’s discernable policy preferences, the question remains how to make discretionary decisions that give appropriate weight to those policy preferences along with the considerations that are properly within the prosecutor’s bailiwick. For example, a prosecutor adopting a “tough on crime” approach in deference to public demand would still be expected to weed out cases where the evidence is too weak to justify prosecution. The prosecutor might still be expected to treat certain defendants leniently in light of mitigating considerations that suggest that the offense was aberrant or understandable. The prosecutor might still prioritize the prosecution of more serious crimes, given resource limitations.

All of these questions pose a challenge to the idea of prosecutorial accountability, to which we will return. If prosecutors owe the public duties of care and loyalty, there presumably must be some process to hold prosecutors accountable. But the ambiguity regarding the scope of prosecutors’ obligations poses an obstacle. What does it mean for a prosecutor to make careful discretionary decisions or to be loyal to the public, in the affirmative sense; what does it mean to be faithful to the public interest or to carry out the public purpose competently in a criminal case? The complexity and opacity of prosecutors’ decision making as well as the amount of independence that prosecutors must necessarily exercise make it difficult to determine whether prosecutors are acting carefully and faithfully or even to know what faithful execution of the criminal law entails in any given situation.

III.    Accountability vs. Independence

This Part argues that while accountability and independence seem at odds, they are, in truth, two alternate ways to best define the beneficiary’s interest and ensure that the prosecutor remains loyal to that interest.137For an argument that prosecutorial independence is at odds with accountability, see Davis, supra note 4, at 438–48. Prosecutors need to give meaning to the vague mandate to do justice. Other public fiduciaries like agency heads may be able to look to statutes or bylaws to assess the purpose or goal they serve and give it more concrete meaning.138See Criddle, supra note 2, at 151–52 (explaining that “administrative agencies owe fiduciary duties to their statutory beneficiaries” in executing their “statutorily defined missions”). Prosecutors have no such framework. The structure of the office, created in part by the system of accountability and independence itself, must over time serve that role.

Accountability demands consequences when prosecutors fail to pursue the public interest. It assumes an institutional mechanism for monitoring prosecutors. Independence, on the other hand, ensures that prosecutors can use their experience and expertise to assess the public interest and determine how best to achieve it without caving in to political influence or mob pressure.139See Davis, supra note 4, at 438 (“[P]rosecutors require a certain level of independence to make their decisions without inappropriate and extraneous political pressures.”); Richman, supra note 4, at 957–59 (discussing the view that “insulation from narrow interest groups and corrupt influences” allows prosecutors to effectively “divine the public interest”). Independence requires some insulation from elected officials and the public. The challenge is to maximize accountability without jeopardizing independence.

Accountability can have any of several different, if inter-related, meanings. At a minimum, accountability suggests meeting one’s responsibilities—that is, doing what one is supposed to do in the way in which one is supposed to do it. Accountability may also mean justifying what one does,140See Barbara O’Brien, A Recipe for Bias: An Empirical Look at the Interplay Between Institutional Incentives and Bounded Rationality in Prosecutorial Decision Making, 74 Mo. L. Rev. 999, 1018 (2009) (“Psychologists who study accountability define it broadly as the experience of feeling pressure to justify judgments or decisions to others. Under the right conditions, imposing accountability on decision makers can make them more thorough and objective.” (footnote omitted)). and facing consequences when failing to fulfil one’s responsibilities. Prosecutors can be held accountable in different ways. The lay public can exercise direct control over prosecutorial decisions or could have a more robust role in how those decisions are made. A prosecutor can be subject either to direct political consequences for his or her work, or the political official who appointed the prosecutor can face repercussions for the prosecutor’s conduct. Alternatively, accountability can require processes for reaching decisions. Prosecutors can be subject to internal and external mechanisms of review and resulting sanctions. Finally, all these forms of accountability may require varying degrees of transparency. Transparency in prosecutors’ decision making is complicated because much of what prosecutors do requires secrecy.

Prosecutorial independence, on the other hand, assumes that experience and expertise are the best guarantee that prosecutors will seek the public interest. Professional reputation, legacy, and self-image will ensure that prosecutors adhere to their obligations of duty and care. Any intrusion into their work threatens the purity of the exercise. Lawyers’ independence was originally conceived in republican terms. Lawyers, as part of an aristocracy, were uniquely suited to guard liberty by ensuring the good of all citizens.141See Rebecca Roiphe, Redefining Professionalism, 26 U. Fla. J.L. & Pub. Pol’y 193, 203–04 (2015) (describing the historical view that lawyers—who, according to Alexis de Tocqueville, “constituted an American brand of aristocracy”—had “sufficient means to think beyond [their] own self-interest”). Their wealth made them independent of any faction, immune to the kind of domination that might distract them from this duty, or in Rave’s terms, trigger the “tyranny of the majority” problem.142Rave, supra note 87, at 54–61; see Roiphe, supra note 141, at 203–05 (summarizing the theory that only those with means to “provide for [their] own basic material needs” could exercise the independence required to be a lawyer).” Initially, only a small class of aristocrats were thought to possess the necessary virtue to fulfill this role.143Id. at 205. As the nineteenth century drew to a close, a burgeoning middle class based its claim to this special status on merit, skill, and training, as opposed to aristocratic pedigree. Lawyers were seen as a separate estate that stood poised to protect against an arbitrary exercise of power by the State and the dominion of one man or one group over another.144See id. at 206 (describing the belief that lawyers’ education and training uniquely positioned them to act for the good of all). Fiduciary theorists have also traced fiduciary law to republican theory.145See Evan J. Criddle, Liberty in Loyalty: A Republican Theory of Fiduciary Law, 95 Tex. L. Rev. 993, 994–1000 (2017) (arguing that republican theory is the best way to interpret fiduciary law).

While many have been justifiably skeptical about this rhetoric, it can be useful.146See Rebecca Roiphe, The Decline of Professionalism, 29 Geo. J. Legal Ethics 649, 650–53 (2016) (suggesting that “aspects of the older understanding of professionalism,” which have generated skepticism among lawyers and scholars, “can and should be relevant and vital today”). Attorney General William Barr, when testifying before the Senate as a nominee for the position, used this notion of professionalism and accountability to the American people to explain why he was suited for the job. Hearing on the Nomination of the Honorable William Pelham Barr to be Attorney General of the United States Before the S. Comm. on the Judiciary, 116 Cong. 1–2 (2019) (written testimony of William P. Barr), https://judiciary.senate.gov/imo/media/doc/Barr%20Testimony.pdf [https://perma.cc/YHE2-KFDS]. At least some prosecutors, like many other lawyers and public officials, take pride in what they do and want to do it well. Additionally, they are trained in certain practices that are, at least theoretically, designed to ensure justice in particular cases.147See Timothy Fry, Prosecutorial Training Wheels: Ginsburg’s Connick v. Thompson Dissent and the Training Imperative, 102 J. Crim. L. & Criminology 1275, 1277–78 (2012) (explaining that while the Court noted in Connick v. Thompson that “[i]ndividual prosecutors have received ‘professional training and have ethical obligations’” related to the fair administration of justice, the dissent viewed this training as more theoretical than practically effective). This training allows them to draw on their experience, knowledge, and skill to exercise judgment in ways that avoid miscarriages of justice, distortions, and abuse. Even when an individual prosecutor may lack the judgment we would like and expect, that prosecutor is constrained by, and often invested in, the procedures and traditions of the office that create a real check on the prosecutor’s conduct.

The dangers to the beneficiary are significant when prosecutors are not held accountable for their acts. Social movements, like the innocence movement and Black Lives Matter, have highlighted just how vulnerable communities are to prosecutorial abuse.148See Green & Yaroshefsky, supra note 13, at 89–90, 93–95. Some have even pointed to independence as the problem.149See Davis, supra note 4, at 408–15 (describing prosecutors’ wide discretion to make outcome-determinative decisions, like charging, and the limited mechanisms for monitoring potential abuses of such discretion). But fiduciary theory helps explain why eroding prosecutorial independence has its own potential perils. If we are not vigilant, partisan political groups or powerful private individuals or groups might seek to use prosecutors to advance their own personal or political interests at the expense of the broader public mandate. Organized factions within the public might try to influence or even take over prosecutors’ offices for their own ends. Those without an understanding of the criminal justice system might seek results that are not in keeping with all the rules and procedures designed to ensure justice. Vulnerable groups or unpopular defendants might suffer at the hands of a majority that does not fully grasp the need for processes and protections. There is a risk not only to individuals who may be wrongly pursued for a political pay-off but also to unpopular defendants, who may face a kind of mob justice.150Green & Roiphe, Punishment Without Process, supra note 17. The rule of law requires not only accountability but independence.

Fiduciary theory offers a way to balance these competing concerns. If a prosecutor’s fiduciary obligation is to pursue the public’s abstract interest in justice, then prosecutors need some degree of insulation to weigh those central concerns. The public should not have a direct or even indirect ability to control how prosecutors weigh concerns like fairness, proportionality, deterrence, and retribution. Those concerns, which are central to the prosecutors’ fiduciary mission, ought to be evaluated both in light of the office’s prior decisions in similar cases and in the context of the facts of that particular case. The public is ill-equipped to make this sort of evaluation. On the other hand, when a prosecutor considers extrinsic concerns that are not central to the core criminal justice mission, there is more room for public input, either directly or through another elected official. The extent to which prosecutors should serve social justice or take into account the economic disadvantage of the accused, potential foreign policy, or immigration repercussions, are all questions of broad social policy that may be relevant in any given prosecution but are not central to the prosecutor’s fiduciary mandate. Other actors may well be better situated than the prosecutor to assess the public’s interest with regard to those extrinsic concerns.

This Part will proceed by identifying each mechanism of accountability and analyzing the risks they pose to prosecutorial independence. This, in turn, provides a framework for assessing which mechanisms of accountability will work best to align the prosecutor’s interest with that of the public without making too great a sacrifice to prosecutorial independence. This Part then returns to fiduciary theory, arguing that the unique nature of prosecutors’ role offers new insight into the fiduciary theory of governance.

A.   Mechanisms of Accountability

1.      External control over prosecutors

Some scholars, particularly those who lament the decline in the role of the jury trial, have suggested greater lay control over prosecutorial decision making. Stephanos Bibas has argued that juries should review plea bargain sentence recommendations.151Stephanos Bibas, Observers as Participants: Letting the Public Monitor the Criminal Justice Bureaucracy, 127 Harv. L. Rev. F. 342, 342–44 (2014) (advocating for public participation in plea hearings); Stephanos Bibas, Essay, Transparency and Participation in Criminal Procedure, 81 N.Y.U. L. Rev. 911, 961 (2006) (“[P]lea and sentencing juries would serve many of the functions that grand and petit juries once did, checking executive and judicial conduct.”); see also Roger A. Fairfax, Jr.,Grand Jury Innovation: Toward a Functional Makeover of the Ancient Bulwark of Liberty, 19 Wm. & Mary Bill Rts. J. 339, 356, 358 (2010) (arguing for grand jury supervision over plea bargaining and sentencing decisions). Josh Bowers and Jocelyn Simonson have similarly suggested greater lay participation in the criminal justice system.152See Josh Bowers, Legal Guilt, Normative Innocence, and the Equitable Decision Not to Prosecute, 110 Colum. L. Rev. 1655, 1657–62 (2010) (arguing for lay input into charging decisions); Jocelyn Simonson, The Criminal Court Audience in a Post-Trial World, 127 Harv. L. Rev. 2173, 2176–77 (2014) (arguing that the Constitution embodies “the idea that the function of the public in the criminal courtroom goes beyond the protection of individuals to implicate the ability of citizens to . . . hold the criminal justice system accountable”). These scholars argue that the rise in plea bargaining and the decline in jury trials account for some of the harshness of the current system.153Cf. Adriaan Lanni, The Future of Community Justice, 40 Harv. C.R.-C.L. L. Rev. 359, 387, 399 (2005) (arguing that “[i]nvolving local laypeople in charging and sentencing decisions would . . . reverse the current trend toward ever-harsher policies,” and specifically that “in a world of guilty pleas, the grand jury as focus group or judicial body may be the only mechanism to ensure that charging policies do not deviate too much from local community opinion”).

Some of these proposals may be wise, particularly those that seek to educate members of the public about the way the criminal justice system works so that they can exercise their rights to alter and reform the system when it has gone awry.154See Simonson, supra note 152, at 2174–78 (emphasizing the importance of the public audience in courtrooms for “hold[ing] the criminal justice system accountable”). Involving the lay public directly in prosecutorial decision making is more problematic, at least where those decisions involve the core interests intrinsic to justice. Doing so risks undermining prosecutorial independence by allowing the influence of those who may lack a commitment to the processes and traditions of the office and who certainly lack the information to make appropriate decisions. While lay input might arguably mitigate an unjustifiably harsh system, public influence may also threaten greater harshness toward unpopular defendants.

There may be a greater role for the public to play either directly or indirectly through elected officials in determining which extrinsic values ought to bear on prosecutors’ decisions. So, prosecutors ought to be more responsive to public concerns and public opinion about immigration or foreign policy, both of which are at least to some degree extrinsic to a determination of what is just in any individual case.

Holding public fiduciaries responsible for their actions is necessarily difficult and imperfect. The imperfect mechanisms for holding fiduciaries responsible are justified by trust. Fiduciary models rely, in part, on the idea that the fiduciary has the character to resist temptation and abide by the beneficiary’s interest.155See Leib et al., supra note 2, at 706–07 (identifying the beneficiary’s trust in the fiduciary as one of three indicia of the fiduciary relationship). Theorists suggest that when the beneficiary is particularly vulnerable, the law should recognize residual control rights to ensure that the agent does not betray the principal.156Cf. id. at 707–08 (“[W]here residual control rights are particularly weak, the beneficiary’s vulnerability to predation is greater and, therefore, the fiduciary must meet a higher standard of conduct.” (footnote omitted)). The public is quite vulnerable to discretionary decisions made by prosecutors, but it is difficult to give control over prosecutors directly to the public in part because the beneficiary’s interest is an abstraction rather than a particular individual or group with a defined interest.157Cf. Miller & Gold, supra note 2, at 523–24 (explaining that some fiduciary mandates “are not identified with determinate persons and their practical interests; they are, in this sense, abstract”). Any control exercised indirectly by the legislature or executive risks undermining prosecutorial independence. That leaves us with the need to devise creative ways to hold prosecutors accountable that will not simultaneously undermine independence, thereby risking other distortions of their duty of loyalty.

Others suggest that legislatures, courts, or disciplinary authorities ought to exercise control over prosecutors’ decisions.158See Angela J. Davis, Arbitrary Justice: The Power of the American Prosecutor 180–82, 189 (2007) (arguing for more rigorous bar disciplinary processes for prosecutors and prosecutorial reform legislation as necessary); Richard A. Rosen,Disciplinary Sanctions Against Prosecutors for Brady Violations: A Paper Tiger, 65 N.C. L. Rev. 693, 733–36 (1987) (proposing changes in bar disciplinary proceedings); James Vorenberg, Decent Restraint of Prosecutorial Power, 94 Harv. L. Rev. 1521, 1568–72 (1981) (calling for more robust judicial review of prosecutors’ discretionary decisions). These mechanisms have proved ineffective. Legislatures are incapable of taking into account the kind of specific facts and circumstances necessary to constrain prosecutors’ decisions, and courts are limited both by separation of powers concerns and by their capacity to review the fact-specific decisions in individual cases. Disciplinary authorities are limited because the directly punitive structure is best suited to intentional misconduct.

2.      Political accountability

Some argue that the best way to ensure that prosecutors pursue socially beneficial ends is to hold them politically accountable for their acts.159See, e.g., Jeremy Travis et al., Prosecutors, Democracy, and Justice: Holding Prosecutors Accountable 6–10 (2019), https://static1.squarespace.com/static/5c4fbee5697a9849dae88a23/t/5d6d8d224f45fb00014076d5/1567460643414/Prosecutors%2C+Democracy%2C+Justice_FORMATTED+9.2.19.pdf [https://perma.cc/Q28B-JS9W] (suggesting that the public can hold prosecutors accountable by electing prosecutors who are truthful, transparent, and equitable); Sara Sun Beale, Rethinking the Identity and Role of United States Attorneys, 6 Ohio St. J. Crim. L. 369, 391–413 (2009) (arguing that United States Attorneys lack the political accountability of local prosecutors). But see David Alan Sklansky, Unpacking the Relationship Between Prosecutors and Democracy in the United States, inProsecutors and Democracy: A Cross-National Study 276, 277 (Máximo Langer & David Alan Sklansky eds., 2017) (discussing the widely-shared view that “[t]he last thing we should want from prosecutors is ‘democratic accountability’”). Perhaps the most prominent proponent of political accountability for prosecutors is Justice Scalia in his dissent in Morrison v. Olson.160487 U.S. 654 (1988). In arguing for the unconstitutionality of the special prosecutor under the Ethics in Government Act, Scalia complained that by insulating the role from presidential control, the legislature had essentially ensured that no political actor could be held accountable for the special prosecutor’s acts.161See id. at 728–31 (Scalia, J., dissenting) (“[E]ven if it were entirely evident that unfairness was in fact the result [of the work of an independent counsel] . . . there would be no one accountable to the public to whom the blame could be assigned.”).

As Professor Dan Epps has asked, “why is the right goal letting politically accountable prosecutors follow the political winds?”162See Epps, supra note 44, at 848; see also Richman, supra note 4, at 973–74 (arguing that electoral control is far less likely to force prosecutors to be responsive to the community than if prosecutors are encouraged to decide cases in the shadow of jury decisions). A prosecutor attuned to the political majority might well do the right thing. The prosecutor might, for instance, decline to prosecute a technically guilty individual who is not morally blameworthy. But at the same time, a political majority might be particularly bloodthirsty, especially in a case that has had an immediate impact on the community. A political majority might unfairly target a particular group or demand a conviction when a prosecutor thinks the evidence is lacking. In addition, we have long abandoned the notion that political actors always pursue socially useful ends. Powerful interests can capture the political process: even if following the majority’s will would lead the prosecutor to do the right thing, political calculations often lead elected officials to follow a powerful minority’s interest rather than that of the majority.

The public is often only interested in a select group of cases—the most gruesome crimes or those involving celebrities or other popular causes. This, too, threatens to warp a prosecutor’s ability to respond to the more abstract public purpose. Prosecutors, who are directly responsive to the public, may have little incentive to follow the will of the public if no one cares. Even if citizens do care, the public would have a hard time assessing a prosecutor’s job, at least when it comes to the obligation to weigh intrinsic criminal justice concerns, given that cases hinge on facts and law that are hard to assess.163Cf. Richman, supra note 4, at 965 (“[T]he bulk of the discretionary decisions that prosecutors make turn . . . [on] case-specific factors. Electoral or appointive politics are, at best, a poor way of holding prosecutors accountable for this myriad of low-visibility enforcement decisions.”). A few cases then could work as cover for the rest. Even if the public could be made to care about all cases equally, voters are not in a good position to assess the prosecutors’ discretionary choices in more complex, factually or legally complicated cases. As we discussed above, it is not clear whether a local prosecutor owes loyalty to members of the district in which the prosecutor works, to the state as a whole, or to the United States.164See supra Section II.C. If it is the latter, elections of local or state prosecutors will not work to align her interest with that of the beneficiary.165Others have noted the difficulty in identifying the relevant principal. See Ethan J. Leib et al., Translating Fiduciary Principles into Public Law, 126 Harv. L. Rev. F. 91, 94 (2013).

Despite Justice Scalia’s assertions in Morrison, federal prosecutors are even less accountable than state prosecutors. United States Attorneys are appointed by the President,166Beale, supra note 159, at 370. often with input from local legislators. They have little incentive to serve local communities, and it is unlikely that the President would be held responsible for something that happened in one of the ninety-four federal districts. Theoretically, the Attorney General would be held responsible, and this, in turn, would reflect on the President, but that seems unlikely at best.167Cf. William J. Stuntz, The Pathological Politics of Criminal Law, 100 Mich. L. Rev. 505, 543 (2001) (“United States Attorneys’ offices . . . have the power to set their own agendas, to decide what cases they wish to spend time on and what cases they wish to ignore.”). Local prosecutors may be more responsive to the public, but pathologies in local politics render this solution problematic. Most district attorneys run uncontested, and the voting public pays little attention to the campaigns.

Not only is political accountability ineffective, it can also be dangerous in its threat to prosecutorial independence. The greater the hierarchical control of prosecutors either by a political actor or by the voting public, the greater the danger that decisions that ought to be characterized by the disinterested application of law to facts will instead reflect the partisan preference of certain groups.168Cf. Green & Roiphe, supra note 15, at 72 (“[C]onflicts of interest threaten to undermine [prosecutors’] impartial decision-making . . . . Allocating responsibility for decisions in individual cases to career prosecutors who are lower down in the hierarchy helps achieve the fair and disinterested administration of criminal justice by making these sorts of conflicts less likely.”). Political accountability may have a role to play in aligning the interest of the public with its principal, but it cannot serve as a panacea. Political accountability works better where the question is prosecutorial policy as opposed to effectiveness. It works better when it concerns policies extrinsic to the central fiduciary mission of the prosecutor. A local prosecutor and a federal prosecutor may well be held responsible for implementing policy promises. As we discussed above, the public may prefer or disfavor a broken windows approach to prosecution and is in a fairly good position to assess the prosecutor’s position.169Supra notes 125–28 and accompanying text. As we have argued elsewhere, and in part because of the question of accountability, the President has a role to play in setting federal prosecutors’ policy agendas but not in controlling individual cases.170Green & Roiphe, supra note 15, at 6.

3.      Internal processes and structures for decision making

A different mechanism to ensure accountability requires internal processes for decision making. Many scholars have suggested reforming prosecutors’ offices to better align their actions with the public interest. Even if the substance of public interest is elusive, it is easier to agree on processes designed to approximate it. If public officers are fiduciaries, then their actions must be made on behalf of the public. Processes are important and at times critical to ensure that this is the case.171Cf. Evan J. Criddle & Evan Fox-Decent, Fiduciaries of Humanity: How International Law Constitutes Authority103–04 (2016) (exploring how non-domination and a fiduciary theory of international norms protect individuals subject to public power against abuse).

We have argued elsewhere that deliberation among differently situated prosecutors, the recording of decision-making processes, and internal review of these processes can help reduce prosecutorial conflicts of interest.172See Green & Roiphe, supra note 55, at 525–27. Others have, in different contexts, proposed changes in institutional design to counter implicit bias and other distortions in prosecutors’ judgment.173See Barkow, supra note 3, at 883, 887–88. Prosecutors should, at the very least, consciously weigh different factors in making important decisions about the public interest and the meaning of the broad mandate to serve justice. While people may not agree about which factors ought to take precedence over others in any given decision, all can agree that prosecutors should consider only proper factors and make a deliberate decision about how to weigh them given the facts of the case.

Extrapolating from the private law theory, Ethan Leib and Stephen Galoob argue that all fiduciary relationships are characterized by a respect for the principal and therefore involve a commitment to deliberation, conscientiousness, and robustness. Deliberation involves not just a particular outcome but a process for arriving at that outcome. Leib and Galoob claim that fiduciaries must maintain a continued commitment to a decision-making process. To be conscientious, a fiduciary must act for the right reasons, namely, to help realize and pursue the interest of the beneficiary. Finally, robustness requires that this process of conscientious deliberation be continuous and account for new information.174See Ethan Leib & Stephen Galoob, Fiduciary Political Theory: A Critique, 125 Yale L.J. 1820, 1839 (2016). While Leib and Galoob’s theory is controversial, we do not plan to wade into this argument because we are using their contribution as well as that of others as a means of exploring prosecutors’ obligations. Since we are doing so in the spirit of analogy as a pragmatic endeavor rather than applying it in a literal fashion, all aspects of the theory can be useful. Some critics of Leib and Galoob argue that this elaboration of fiduciary duty mistakes a moral conception of loyalty for a fiduciary one. Paul B. Miller, Dimensions of Fiduciary Loyalty, in Research Handbook on Fiduciary Law 180–81 (Andrew S. Gold & D. Gordon Smith eds., 2018). While we do not take a side in the debate about with whether or not the deliberative norms Leib and Galoob identify are a necessary feature of all fiduciary relationships or a common feature of some, we do draw on the observation that for prosecutorial decision making, the duty of loyalty requires certain deliberative processes. Prosecutors could be required to engage in this sort of conscientious deliberative process. They could be encouraged through structural change within offices to account for new information and consciously and explicitly engage with prior practices within the office.175Green & Roiphe, supra note 55, at 525–33.

These sorts of reforms are promising in part because they pose little threat to, and possibly foster, prosecutorial independence. Encouraging internal processes and structural change offers a way to help ensure that prosecutors adhere to certain traditions and thought processes that work to align their discretionary decisions with the public interest. Among other things, the problem with these sorts of internal controls is that there are few guarantees that prosecutors will adhere to them and that if they do, they will do so seriously. Those who are concerned that prosecutors, left to their own devices, are not good at monitoring their own behavior may be skeptical that these sorts of changes will really help align prosecutors’ decisions with the public interest.

4.      Transparency

Transparency can work in conjunction with these other forms of accountability to ensure a more direct kind of control for the beneficiary. Linked to progressive politics, transparency is thought to be necessary to promote democratic goals, efficiency, and egalitarianism.176See David E. Pozen, Transparency’s Ideological Drift, 128 Yale L.J. 100, 107–08 (2018) (discussing the aims of transparency policies during the Progressive Era and in the decade between the mid-1960s and the mid-1970s, which included more “efficient,” “egalitarian,” and “democratically accountable regulation”). An electorate cannot hope to hold a public official accountable without a clear understanding of how well that official’s job has been done. If political accountability has any function in reining in prosecutors, their work must be, in some sense, transparent. Direct control over the prosecutor’s decisions similarly requires greater transparency.177See Simonson, supra note 152, at 2205, 2216 (explaining that the public’s ability to hold government officials accountable depends on the transparency of “the routine appearances that make up criminal justice in [the public’s] neighborhoods”). If the public is going to enjoy greater participation, then it needs more information about the decisions prosecutors are making, prosecutors’ reasons for making these decisions, and the processes by which decisions are made. Perhaps greater public engagement in prosecutorial decision making, in turn, would educate the public about the nature of the criminal justice system.178For a discussion of transparency and prosecutors, see generally Jessica A. Roth, Prosecutorial Declination Statements, 110 J. Crim. L. & Criminology (forthcoming Apr. 2020).

Transparency, which in other contexts can be key to accountability, is more problematic for prosecutors, however. Much of what prosecutors do is necessarily secret. Grand jury secrecy, enforced by statute, ensures the safety of witnesses and the dignity of uncharged suspects.179See JoEllen Lotvedt, Availability of Civil Remedies Under the Grand Jury Secrecy Rule, 47 Cath. U. L. Rev. 237, 241–42 (1997) (“[Grand jury] [s]ecrecy . . . protects the anonymity of the witness pool and . . . prevents the release of derogatory information about an unindicted individual.”). Secrecy preserves the integrity of future or ongoing investigations and, at times, protects national security.180See id. at 241 (“Maintaining secrecy decreases the possibility that a suspect may escape, destroy evidence, or harass adverse witnesses.”). As discussed above, the public is not particularly well suited to review discretionary decisions, and there is a risk to prosecutorial independence when it does.

Even reforms in processes within prosecutors’ offices could work together with transparency to create a closer alignment between prosecutors’ interests and those of the public. While it is unrealistic and undesirable to require prosecutors to reveal the details of each decision, they could be required to reveal the mechanisms by which they came to that decision and the evolving standards governing those decisions. This would give the public an opportunity to supervise prosecutors’ work without involving them in factual and legal inquiries for which they are untrained and ill suited.

In addition, when the integrity of the institutions is in doubt, transparency may trump independence. When the public ceases to believe that the processes for ensuring loyalty and care are working, it may be worth compromising the independence of prosecutors for the sake of transparency and accountability.

B.   Assessing Mechanisms for Prosecutorial Accountability

In a private fiduciary relationship, accountability can mean that the beneficiary enjoys direct control over some or all of the fiduciary’s work. This model of accountability does not and should not translate to prosecutors’ work. As we discuss above and elsewhere, direct and plenary control by political actors does not work to align prosecutors’ work with the public interest.181See Green & Roiphe, supra note 15, at 71 (“The discretion of . . . prosecutors promotes the fair and even-handed administration of justice . . . . American prosecutors sometimes fail; but presidential influence over individual cases would only make matters worse.” (footnotes omitted)). Once we conceive of the prosecutor’s objective as seeking justice, not carrying out the contemporary public’s will in a more general sense, itfollows that direct control would be unwise.

Because prosecutors’ work is specialized in nature, it might be better to hold them accountable to other prosecutors or to other knowledgeable public officials or official bodies rather than to the public directly.182See O’Brien, supra note 140, at 1046–47 (“Research on accountability demonstrates that decision makers come closest to this ideal when they know that they will be judged primarily for the process of their decision making, as opposed to the outcome. This sort of accountability could come through internal procedures, by way of review within a prosecutor’s office, or through an outside agency’s supervision.” (footnotes omitted)). When it comes to federal and other appointed prosecutors, the public lacks removal authority in any event. At best, it can influence the official who possesses the authority to remove the prosecutor. In federal cases, public influence is especially attenuated because the President has direct authority to fire only the Attorney General and United States Attorneys,18328 USC § 541(c) (2012). not subordinate prosecutors.184§ 542(b) (Assistant United States Attorneys are subject to removal by the Attorney General). Further, the termination of a subordinate prosecutor is subject to review and reversal by the Merit Systems Protection Board. See, e.g., Goekev. Dep’t of Justice, 2015 M.S.P.B. 1, 2 (2015). And it is unlikely given the many public concerns that voters are motivated by federal prosecutorial policy when they cast their vote for president. With respect to elected prosecutors, the public does not presently have enough information to assess whether prosecutors are faithful to professional norms.185Russell M. Gold, “Clientless” Lawyers, 92 Wash. L. Rev. 87, 117 (2017) (“[T]he political check on elected prosecutors does not work well because voters lack sufficient information about their prosecutors’ enforcement priorities.”); Richman, supra note 4, at 963 (“[E]ven direct elections are not likely to prove an effective means of giving prosecutors guidance as to a community’s enforcement priorities or of holding them accountable for the discretionary decisions that they have already made.”); Ronald F. Wright, How Prosecutor Elections Fail Us, 6 Ohio St. J. Crim. L. 581, 591 (2009) (“[P]rosecutor elections . . . do not assure that the public knows and approves of the basic policy priorities and implementation of policy in the prosecutor’s office.”). But even if prosecutors were substantially more transparent, it is doubtful whether voters’ criterion would be whether prosecutors were “seeking justice” as that concept is understood in the legal profession.

Other public officials or official bodies might be particularly good at offering input or monitoring the extrinsic factors that go into prosecutorial decision making. They too would be removed from the facts of individual cases, making it hard for them to monitor discretionary decisions in that regard. But they might serve as good proxies for public opinion on other factors, even those that straddle the line between concerns intrinsic and extrinsic to criminal justice, like how much prosecutors ought to consider social or racial justice or mass incarceration.

Additionally, accountability can mean being subject to some procedural mechanism to ensure that one meets one’s responsibilities and can be removed from one’s position when one does not live up to the applicable standard. For public officials, accountability in this sense may or may not imply some public transparency, depending on whether the removal process involves a public election or an act by a supervisory official or body. Even in an electoral process, it is not obvious what must be publicly disclosed and when. The answer presumably turns, at least in part, on the nature of the official’s responsibilities—on what it means to do the job well or poorly.

A fiduciary theory of prosecutors suggests that monitoring the processes of decision making would be the optimal way to hold prosecutors accountable. Public officials giving input into extrinsic considerations that ought to drive discretionary decisions might also prove useful in aligning the interest of the prosecutor with that of the beneficiary.

C.   Fiduciary Theory of Governance Revisited

Analyzing prosecutors as fiduciaries contributes to the developing fiduciary theory of governance. The question of accountability and independence is not as central for most other public officials. But the general tension between the two values is relevant to some extent for all public fiduciaries. Are leaders picked to use their judgment, knowledge, and skill on behalf of the public or are they expected to be more directly accountable to the electorate’s will?186See generally Sophia Rosenfeld, Democracy and Truth: A Short History 15, 20, 26 (2019) (arguing that truth has been contested throughout American history and ought to be a collaboration between experts in government and the lay public). The answer, it seems, is both. The proper balance will depend on the role and responsibilities of different public officials, but to some degree, expertise—which is fostered by a kind of exclusivity—and popular responsiveness must go hand in hand.

Our examination of prosecutors’ fiduciary role does suggest a critique of fiduciary theory: it tends to exalt the value of discretion. As Paul Miller argues, the fiduciary by definition exercises discretionary power over the interests of the beneficiary.187Paul B. Miller, A Theory of Fiduciary Liability, 56 McGill L.J. 235, 262 (2011). In the private law context, scholars have argued that this definition of fiduciary obligation does a disservice to the principal, diminishing the principal’s power and rationalizing a paternalistic relationship. The agent should serve the principal. The fiduciary should take direction from the beneficiary.188See W. Bradley Wendel, Should Lawyers Be Loyal to Clients, the Law, or Both? (forthcoming) (on file with authors) (describing the fiduciary theory that the agent should “interpret the instructions of the principal . . . in accordance with the agent’s understanding of the principal’s wishes”). Of course, when the beneficiary has a vast and complex amalgam of abstract interests as the fiduciary governance model points out, minimizing discretion is difficult, if not impossible. And even if it were possible, it is not necessarily desirable.

In the context of governance, Seth Davis has expressed an analogous concern. He argues that the fiduciary model generally minimizes and at times can mask domination and hegemony.189Seth Davis, Pluralism and the Public Trust, in Fiduciary Government 281, 288–99 (Evan J. Criddle et al. eds, 2018). There is reason to be concerned about this dynamic in prosecution in particular.190See Davis, supra note 4, at 408–15 (describing prosecutors’ discretion and abuses of such discretion that result in a level of control over the accused). In fact, some have argued that domination under the mantel of fiduciary service characterizes the criminal justice system in general.191See Michelle Alexander, The New Jim Crow: Mass Incarceration in the Age of Colorblindness 1–15 (2010) (likening the domination over people of color and the poor that resulted from the “government’s zealous . . . efforts to address rampant drug crime” during the War on Drugs beginning in the 1980s to the Jim Crow-era racial caste system). It is no surprise that scholars of the criminal justice system who are concerned with minority rights and protecting the less powerful seek to restrict rather than expand or justify discretion.192See, e.g., Davis, supra note 4, at 400, 409–10 & n.63 (suggesting that prosecutors’ discretionary decisions may be the result of bias, like unconscious racism, and proposing reforms to hold prosecutors accountable to their constituents).

Broad discretionary power is dangerous. It risks not only abuse but also a more insidious form of power in which expert dominance takes on the guise of disinterestedness. If we assume that professionals are particularly good at assessing what is in the public interest and particularly well suited to avoid this kind of dynamic, then perhaps the fiduciary model of prosecutorial power could avoid this critique. But ever since the 1970s, scholars and critics have shown just how central professions and other experts have been to just this sort of social control.193See Roiphe, supra note 146, at 675, 677 (explaining scholars’ critique of lawyers who “substitut[ed] their own political and ideological agenda[s] for those of the . . . communit[ies] for whom they were purportedly fighting” and scholars’ increasing focus on a theory of lawyers as “zealous advocate[s]” and “minimiz[ing of] the professional obligation to society as a whole”).

That said, discretion in the exercise of government power is inevitable.194H. L. A. Hart, Discretion, 127 Harv. L. Rev. 652, 663–64 (2013). And discretion in the application of the criminal laws is no exception. As we explained above, allowing the public direct control over intrinsic factors that go into decisions in individual cases is not only impracticable but also potentially dangerous.195See supra Section III.A.2. Thus, we are stuck with Paul Miller’s definition of a fiduciary.196See Miller, supra note 187, at 262 (defining a fiduciary relationship as “one in which one party (the fiduciary) enjoys discretionary power over the significant practical interests of another (the beneficiary)” (emphasis added)). A more direct control over prosecution would risk partisan influence or a powerful faction controlling prosecutors’ decisions. It could also involve those without experience or knowledge who will likely act in an arbitrary way inconsistent with the public interest in criminal justice.

Independence should be protected when it can foster expertise, experience, and professional pride. While they should never go unchecked, these aspects can be harnessed to promote processes and norms that will tend toward good social outcomes. In order to do that, prosecutors and public officials in general need to be insulated from the public. But, expertise is not infallible; experience is not incorruptible. To make sure that these processes do not get coopted or ignored for the private ends of individual officials or corrupted by implicit bias and other distortions in decision making, the public has a right to monitor their public officials. The proper degree of input and control from the public will depend on the particular official and the nature of his or her role. Some functions are better performed in isolation, protected from public clamor; others are better aired in public, performed in collaboration with the beneficiary.

With regard to prosecutors, the best way to ensure this balance is to insulate prosecutors in making discretionary decisions in individual cases but require them to be more deliberate, rational, and transparent in developing processes to make these decisions. In addition to implementing regular procedures, prosecutors’ offices should be transparent both about the policies motivating their decisions and how the decisions are made.

Bringing this insight to bear on the fiduciary theory of governance in general leads to the conclusion that the nature of the fiduciary relationship between public officials and the electorate is complex and depends on the nature of the office. The best way to ensure faithful principals is by employing all possible mechanisms to ensure that the official is accountable to the public in a way that respects and preserves the unique nature of his or her expertise and experience.197Evan Criddle has essentially made this argument about federal administrative agencies. Under the fiduciary model, administrative agencies are not anti-democratic but rather fiduciaries. The key in the relationship is to foster accountability while preserving an arena for expert deliberation. Criddle, supra note 51, at 447–49. While the mechanism of doing so will vary depending on the official’s role and responsibilities, the goal remains constant.

Political officials ought to educate the public about the role of different officials and institution and avoid spreading misinformation that might undermine the norms that have developed over the years.198Many have accused the Trump Administration of failing to live up to this responsibility. See, e.g., Aaron Blake, The Trump Team is Running a Disinformation Campaign About Russian Interference, Wash. Post: The Fix (Aug. 1, 2018), https://www.washingtonpost.com/news/the-fix/wp/2018/08/01/the-trump-teams-vast-disinformation-campaign-about-russian-interference; Max Boot, Donald Trump’s Biggest Disinformation Campaign Yet, Foreign Pol’y: Voice (Oct. 31, 2017), http://www.foreignpolicy.com/2017/10/31/donald-trumps-biggest-disinformation-campaign-yet [https://perma.cc/VDJ8-DENF]; David Rhode, How Disinformation Reaches Donald Trump, New Yorker: News Desk(Oct. 3, 2019), http://www.newyorker.com/news/news-desk/how-disinformation-reaches-donald-trump [https://perma.cc/N3VY-NLY3]. In order to enhance accountability, public officials who are in the right position to do so should clarify what the norms and processes that govern their role are and why they are necessary for the proper functioning of government.199Some officials seem to intuit this role. In giving a press conference after charging Michael Cohen—President Trump’s personal lawyer—in federal court, Deputy United States Attorney Rob Khuzami not only announced the charges but also explained the importance of the laws and the equal application of those laws. Robert Khuzami Statement on Michael Cohen Case, CSPAN (Aug. 21, 2018), https://www.c-span.org/video/?450331-1/lead-prosecutor-speaks-reporters-michael-cohen-guilty-plea [https://perma.cc/MKP2-P4L3] (“The rule of law applies . . . . It is our commitment [as law enforcement] that we will pursue . . . those who choose to break the law and vindicate the majority of people who lead law-abiding lives . . . . The message is that we are here, prosecutors are here . . . we are a nation of laws and the essence [of] . . . this case . . . is justice[:] . . . that is[,] an equal playing field for all persons in the eyes of the law.”). A faithful public servant should be responsible not just for making decisions according to these norms and traditions but for educating the public about how important they are.200See Bruce A. Green & Russell G. Pearce, “Public Service Must Begin at Home”: The Lawyer as Civics Teacher in Everyday Practice, 50 Wm. & Mary L. Rev. 1207, 1211–13 (2009) (describing lawyers as civics teachers who have “an obligation to convey to clients [their] understanding of proper civic conduct”).


This Article introduces the fiduciary theory of prosecution not as a fully developed proposal but as an invitation for further inquiry. In addition to reframing old debates in a new and potentially helpful way, the theory raises novel questions about how prosecutors ought to function and offers broad outlines to be filled in. We invite others to join us in developing this theory, which, unlike other theories of prosecution, is firmly rooted in history and tradition.

The principle value of the fiduciary theory of prosecution is not prescriptive in the following sense. It does not tell prosecutors what to do, except perhaps when it comes to extreme conduct that prosecutors should avoid under virtually any understanding of their function. Because it places discretion at the center of prosecutors’ work and asks prosecutors to act in service of an amorphous beneficiary with a vague objective, it cannot dictate particular decisions in concrete situations.

The theory does prescribe, in broad outline, how prosecutors should reach decisions, but not the decisions they should ultimately reach. In other words, a fiduciary theory of prosecution requires a certain process for decision making but not particular outcomes. Observers cannot use the theory to evaluate or critique charging or plea-bargaining decisions because the relevant facts about prosecutors’ decision-making processes will rarely be available. Nor does the theory, in itself, help answer whether traditional, progressive, or other styles of prosecuting are preferable.

The theory’s value is primarily explanatory. To begin with, the fiduciary theory makes sense of, and legitimates, conventional understandings of the prosecutor’s role, including the idea of prosecution as a public trust, the requirement of prosecutorial independence, and most importantly, the prosecutorial duty to seek justice. The theory offers insight into the meaning and significance of the vague duty to seek justice and underscores that, as fiduciaries, prosecutors have further duties—in particular, duties of care and loyalty. These other duties are not themselves elements of “justice” but rather are legal imperatives governing how prosecutors should pursue justice. Further, the theory offers a new distinction between prosecutors’ pursuit of justice as opposed to other relevant social policy objectives (while acknowledging that the distinctions are not always clear and that there is sometimes overlap), and it gives priority to the pursuit of conventional criminal justice interests. Consequently, the theory both gives greater clarity to the defining concept of “seeking justice” and shows how and why there is more to prosecutors’ work than this pursuit.

Fiduciary theory also contributes to our understanding of how to regulate prosecutors. By demonstrating that accountability and independence are two mechanisms designed to align prosecutors’ interests with those of the public, fiduciary theory suggests regulatory reforms that maximize both. Scholars and critics of the criminal justice system often point to the insulation of prosecutors from outside regulation as cause for concern. This may be true, but any effort at reform must be careful not to sacrifice too much in prosecutorial independence for the sake of transparency or direct public accountability because independence, too, is essential in aligning prosecutors’ work with the interest of the beneficiary.

And perhaps most importantly, fiduciary theory helps justify some features of prosecuting that many find frustrating, including that different prosecutors evidently take different approaches to discretionary decision making, resulting in disparate outcomes on similar facts; that prosecutors often seem to act undemocratically, ignoring public preferences; and that prosecutors often give no explanations for the controversial decisions they make. These frustrations are understandable and, to some extent, can be addressed. As we suggest, both internal and external processes can strike a better balance between prosecutors’ accountability and independence. Further, the processes for training, electing, appointing, and hiring prosecutors can better identify lawyers who will exercise good judgment in their fiduciary role. But, in the end, the fiduciary theory reminds us that the essential features of prosecutorial decision making and regulation, which may give one pause, are neither arbitrary nor the product of a political process in which prosecutors have accumulated power for its own sake. These features grow out of a long legal tradition, undergirded by a theory, that casts prosecutors as fiduciaries, a professional role with significant substantive and procedural implications. With prosecutors’ power, comes fiduciary responsibilities. And that should be a source of some comfort.